JSSH Vol. 10 (2) Sep. 2002

ISSN: 0128-7702

P e r t a n i k a J o u r n a l o f

socialscienceHumanities

VOLUME 10 NO. 2SEPTEMBER 2002

A scientific journal published by Universiti Putra Malaysia Press

Pertanika Journal of Social Science and Humanities

About the JournalPertanika, the pioneer journal of UPM, beganpublication in 1978. Since then, it has establisheditself as one of the leading multidisciplinary journalsin the tropics. In 1992, a decision was made tostreamline Pertanika into three journals to meetthe need for specialised journals in areas of studyaligned with the strengths of the university. Theseare (i) Pertanika Journal of Tropical AgriculturalScience (ii) Pertanika Journal of Science &Technology (iii) Pertanika Journal of Social Science& Humanities.

Aims and ScopePertanika Journal of Social Science and Humanitiesaims to develop as a flagship journal for the SocialSciences with a focus on emerging issues pertainingto the social and behavioral sciences as well as thehumanities, particularly in the Asia Pacific region.It is published twice a year in March and September

The objective of the jou rna l is to p romoteadvancements in the fields of anthropology,business studies, communications, economics,education, extension studies, psychology, sociologyand the humanities. Previously unpublished

| original, theoretical or empirical papers, analyticalreviews, book reviews and readers critical reactionsmay be submitted for consideration. Articles maybe in English or Bahasa Melayu.

Submission of ManuscriptThree complete clear copies of the manuscript areto be submitted to

The Chief EditorPertanika Journal of Social Science and HumanitiesUniversiti Putra Malaysia13400 UPM, Serdang, Selangor Darul EhsanMALAYSIATel: 0S-89468854 Fax: 05^9416172

| Proofs and OffprintsPage proofs, illustration proofs and the copy-editedmanuscript will be sent to the author. Proofs mustbe checked very carefully within the specified timeas they will not be proofread by the Press editors.

Authors will receive 20 offprints of each article anda copy of the journal. Additional copies can beordered from the Secretarv of the Editorial Board.

1 EDITORIAL BOARD

Prof. Dr. Abdul Rahman Md Aroff - Chief EditorFaculty of Human Ecology

Prof. Dr. Annuar Md. NasirFaculty of Economics & Management

Prof. Dr. Mohd. Ghazali MohayidinFaculty of Economics & Management

Prof. Dr. Hjh. Aminah Hj. AhmadFaculty of Education

Assoc. Prof. Dr. Rozumah BaharudinFaculty of Human Ecology

Assoc. Prof. Dr. Abdul Halin HamidFaculty of Human Ecology

Assoc. Prof. Dr. Rosli TalifFaculty of Modern Language Studies

Sumangala Pillai - SecretaryUniversiti Putra Malaysia Press

INTERNATIONAL PANEL MEMBERS

Published by Universiti Putra Malaysia PressISSN No: 0128-7702

Prof. Jean Louis FloriotInternational Graduate Institute of Agribusiness

Prof. Bina AgarwralUniversity Enclave India

Prof. V.T KingUniversity of Hull

Prof. Royal D. ColleCornell Unix)ersity, Ithaca

Prof. Dr. Linda J. NelsonMichigan State University

Prof. Dr. Yoshiro HatanoTokyo Cakugei University

Prof. Max LanghamUniversity of Florida

Prof. Mohamed AriffMonash I niversity Australia

Prof. Fred LuthansUniversity of Nebraska

Prof. D.H. RichieUniversity of Toledo

Prof. Gavin W. JonesAustralian National University

Prof. Dr. Lehman B. FlectherIowa State University

Prof. Ranee PL Lee(Ihinese University. Hotig Kong

Prof. Stephen H.K. YehUniversity ofHaiuaii at Manoa

Prof. Graham W. ThurgoodCalifornia State University

ARCHIVE COPY>/pncP r>n Not Remove)

PERTANIKA EDITORIAL OFFICEResearch Management Centre (RMC)

1 st Floor, IDEA Tower liUPM-MTDC, Technology Centre

Universiti Putra Malaysia43400 Serdang, Selangor, Malaysia

Tel: +603 8947 1622, 8947 1619, 8947 1616

Pertanika Journal of Social Science & Humanities

Volume 10 Number 2 (September) 2002

Contents

Modelling the Volatility of Currency Exchange Rate Using GARCH 85Model - Choo Wei Chong, Loo Sin Chun and Muhammad Idrees Ahmad

The Influence of Value Orientations on Service Quality Perceptions in 97a Mono-Cultural Context: An Empirical Study of Malay UniversityStudents - Hazman Shah Abdullah and Razmi Chik

The Sociolinguistics of Banking: Language Use in Enhancing 109Capacities and Opportunities - Ain Nadzimah Abdullah and RosliTalif

Faktor-faktor Mempengaruhi Agihan Pendapatan di Malaysia 1970- 1172000 - Rahmah Ismail dan Poo Bee Tin

Performances of Non-linear Smooth Transition Autoregressive and 131Linear Autoregressive Models in Forecasting the Ringgit-Yen Rate- Liew Khim Sen and Ahmad Zubaidi Baharumshah

Interpretation of Gender in a Malaysian Novel: The Case of Salina 143- Jariah Mohd. Jan

Penterjemahan Pragmatik dalam Konsep Masa Arab-Melayu: Satu 153Analisis Teori Relevan - Muhammad Fauzi bin Jumingan

Tingkah Laku Keibubapaan dan Penyesuaian Tingkah Laku Anak 165dalam Keluarga Berisiko di Luar Bandar - Zarinah Arshat, RozumahBaharudin, Rumaya Juhari dan Rojanah Kahar

Pertanika J. Soc. Sci. & Hum. 10(2): 85-95 (2002) ISSN: 0128-7702© Universiti Putra Malaysia Press

Modelling the Volatility of Currency Exchange Rate Using GARCH Model

CHOO WEI CHONG, LOO SIN CHUN & MUHAMMAD IDREES AHMADFaculty of Economics & Management

Universiti Putra Malaysia43400 UPM, Serdang, Selangor

E-mail: [email protected]

Keywords: Exchange rates, volatility, forecasting, GARCH, random walk

ABSTRAKKertas ini mengkaji model GARCH dan modifikasinya dalam menguasai kemeruapan kadarpertukaran mata wang. Parameter model tersebut dianggar dengan menggunakan kaedahkebolehjadian maksimum. Prestasi bagi penganggaran dalam sampel didiagnosis denganmenggunakan beberapa statistik kebagusan penyuaian dan kejituan telahan satu langkah kedepan dan luar sampel dinilai dengan menggunakan min ralat kuasa dua. Keputusan kajianmenunjukkan kegigihan kemeruapan kadar pertukaran mata wang RM/Sterling. Keputusandaripada penganggaran dalam sampel menyokong kebergunaan model GARCH dan modelvariasi malar pula ditolak, sekurang-kurangnya dalam sampel. Statistik Q dan ujian pendarabLangrange (LM) mencadangkan penggunaan model GARCH yang beringatan panjangmenggantikan model ARCH yang beringatan pendek dan berperingkat lebih tinggi. ModelGARCH-M pegun berprestasi lebih tinggi daripada model GARCH lain yang digunakan dalamkajian ini, dalam telahan satu langkah ke depan dan luar sampel. Apabila menggunakan modelperjalanan rawak sebagai tanda aras, semua model GARCH berprestasi lebih baik daripada modeltanda aras ini dalam meramal kemeruapan kadar pertukaran mata wang RM/Sterling.

ABSTRACT

This paper attempts to study GARCH models with their modifications, in capturing the volatilityof the exchange rates. The parameters of these models are estimated using the maximumlikelihood method. The performance of the within-sample estimation is diagnosed using severalgoodness-of-fit statistics and the accuracy of the outof-sample and one-step-ahead forecasts isevaluated using mean square error. The results indicate that the volatility of the RM/Sterlingexchange rate is persistent. The within sample estimation results support the usefulness of theGARCH models and reject the constant variance model, at least within-sample. The Q-statistic andLM tests suggest that long memory GARCH models should be used instead of the short-termmemory and high order ARCH model. The stationary GARCH-M outperforms other GARCHmodels in out-of-sample and one-step-ahead forecasting. When using random walk model as thenaive benchmark, all GARCH models outperform this model in forecasting the volatility of theRM/Sterling exchange rates.

INTRODUCTIONIssues related to foreign exchange rate havealways been the interest of researchers inmodern financial theory. Exchange rate, whichis the price of one currency in terms of anothercurrency, has a great impact on the volume offoreign trade and investment. Its volatility hasincreased during the last decade and is harmfulto economic welfare (Laopodis 1997). The

exchange rate fluctuated according to demandand supply of currencies. The exchange ratevolatility will reduce the volume of internationaltrade and the foreign investment.

Modelling and forecasting the exchange ratevolatility is a crucial area for research, as it hasimplications for many issues in the arena offinance and economics. The foreign exchangevolatility is an important determinant for pricing

Choo Wei Chong, Loo Sin Chun & Muhammad Idrees Ahmad

of currency derivative. Currency options andforward contracts constitute approximately halfof the U.S. 880bn per day global foreignexchange market (Isard 1995). In view of this,knowledge of currency volatility should assistone to formulate investment and hedgingstrategies.

The implication of foreign exchange ratevolatility for hedging strategies is also a recentissue. These strategies are essential for anyinvestment in a foreign asset, which is acombination of an investment in theperformance of the foreign asset and aninvestment in the performance of the domesticcurrency relative to the foreign currency. Hence,investing in foreign markets that are exposed tothis foreign currency exchange rate risk shouldhedge for any source of risk that is notcompensated in terms of expected returns (Santisetal 1998).

Foreign exchange rate volatility may alsoimpact on global trade patterns that will affect acountry's balance of payments position and thusinfluence the government's national policy-making decisions. For instance, Malaysia fixedthe exchange rate at RM3.80/US$ in September,1998, due to the economic turmoil and currencycrisis in 1997. This turmoil has spread todeveloped countries such as USA, Hong Kong,Europe and other developing South Americancountries such as Brazil and Mexico. Due to thiscurrency crisis, various governments haveresorted to different national policies so as tomitigate the effect of this crisis.

In international capital budgeting ofmultinational companies, the knowledge offoreign exchange volatility will help them inestimating the future cash flows of projects andthus the viability of the projects.

Consequently, forecasting the futuremovement and volatility of the foreign exchangerate is crucially important and of interest tomany diverse groups including marketparticipants and decision makers.

Beginning with the seminal works ofMandelbrot (1963a, 1963b, 1967) and Fama(1965), many researchers have found that thestylized characteristics of the foreign currencyexchange returns are non-linear temporaldependence and the distribution of exchangerate returns are leptokurtic, such as Friedmanand Vandersteel (1982), Bollerslev (1987),Diebold (1988), Hsieh (1988, 1989a, 1989b),

Diebold and Nerlove (1989), Baillie andBollerslev (1989). Their studies have found thatlarge and small changes in returns are ' clustered'together over time, and that their distribution isbell-shaped, symmetric and fat-tailed.

These features of data are normally thoughtto be captured by using the AutoregressiveConditional Heteroskedasticity (ARCH) modelintroduced by Engle (1982) and the GeneralisedARCH (GARCH) model developed by Bollerslev(1986), which is an extension of the ARCHmodel to allow for a more flexible lag structure.The use of ARCH/GARCH models and itsextensions and modifications in modeling andforecasting stock market volatility is now verycommon in finance and economics, such asFrench et al (1987), Akgiray (1989), Lau et al(1990), Pagan and Schwert (1990), Day andLewis (1992), Kim and Kon (1994), Frames andVan Dijk (1996) and Choo et al (1999).

On the other hand, the ARCH model wasfirst applied in modeling the currency exchangerate by Hsieh only in 1988. In a study done byHsieh (1989a) to investigate whether dailychanges in five major foreign exchange ratescontain any nonlinearities, he found thatalthough the data contain no linear correlation,evidence indicates the presence of substantialnonlinearity in a multiplicative rather thanadditive form. He further concludes that ageneralized ARCH (GARCH) model can explaina large part of the nonlinearities for all fiveexchange rates.

Since then, applications of these models tocurrency exchange rates have increasedtremendously, such as Hsieh (1989b), Bollerslev,T. (1990), Pesaran and Robinson (1993),Copeland et al (1994), Takezawa (1995),Episcopos and Davies (1995), Brooks (1997),Hopper (1997), Cheung et al. (1997), Laopodis(1997), Lobo etal (1998) and Duan etal (1999).

In many of the applications, it was foundthat a very high-order ARCH model is requiredto model the changing variance. The alternativeand more flexible lag structure is the GeneralisedARCH (GARCH) introduced by Bollerslev(1986). Bollerslev et al (1992) indicated thatthe squared returns of not only exchange ratedata, but all speculative price series, typicallyexhibit autocorrelation in that large and smallerrors tend to cluster together in contiguoustime periods in what has come to be known asvolatility clustering. It is also proven that small

86 PertanikaJ. Soc. Sci. & Hum. Vol. 10 No. 2 2002


lag such as GARCH (1,1) is sufficient to modelthe variance changing over long sample periods(French et al 1987; Franses and Van Dijk 1996;Choo et al. 1999).

Even though the GARCH model caneffectively remove the excess kurtosis in returns,it cannot cope with the skewness of thedistribution of returns, especially the financialtime series which are commonly skewed. Hence,the forecasts and forecast error variances from aGARCH model can be expected to be biased forskewed time series. Recently, a few modificationsto the GARCH model have been proposed, whichexplicitly take into account skewed distributions.One of the alternatives of non-linear modelsthat can cope with skewness is the ExponentialGARCH or EGARCH model introduced byNelson (1990). For stock indices, Nelson'sexponential GARCH is proven to be the bestmodel of the conditional heteroskedasticity.

In 1987, Engle et aL developed the GARCH-Mto formulate the conditional mean as functionof the conditional variance as well as anautoregressive function of the past values of theunderlying variable. This GARCH in the mean(GARCH-M) model is the natural extension dueto the suggestion of the financial theory that anincrease in variance (risk proxy) will result in ahigher expected return.

Choo et al. (1999) studies the performanceof GARCH models in forecasting the stockmarket volatility and they found that i) thehypotheses of constant variance models couldbe rejected since almost all the parameterestimates of the non-constant variance (GARCH)models are significant at the 5% level; ii) theEGARCH model has no restrictions andconstraints on the parameters; iii) the long-memory GARCH model is more suitable thanthe short-memory and high-order ARCH modelin modelling the heteroscedasticity of thefinancial time series; iv) the GARCH-M is best infitting the historical data whereas the EGARCHmodel is best in outof-sample (one-step-ahead)forecasting; v) the IGARCH is the poorest modelin both aspects.

Since Choo et aL (1999) have indicated thatthe GARCH-M model performs well in within-sample estimation and the EGARCH modelperforms best in outof-sample forecasting, thecombination of both models, EGARCH-M shouldbe able to enhance the performance in bothaspects.

In order to know the out-of-sampleforecasting performance of EGARCH-M, wecompare the performance of EGARCH-M andthe other modifications of the GARCH model tothe simple random walk forecasting scheme.

The models are presented in the followingsection. The third section is the background ofcurrency exchange rate data and themethodology used in this study. All the resultswill be discussed in the fourth section. Theconclusion will be in the final section.

MODELThe conditional distribution of the series ofdisturbances which follows the GARCH processcan be written as

s/^ ~ N(0,A,)

where iptl denotes all available informationat time f- 1. The conditional variance h is

Hence, the GARCH regression model forthe series of rt can be written as

-(p]B-K -<f>sBs

et~N{0,l)

iy-i

where B is the backward shift operatordefined by &yt = yt- k. The parameter pi reflectsa constant term, which in practice is typicallyestimated to be close or equal to zero. Theorder of s is usually 0 or small, indicating thatthere are usually no opportunities to forecast r(

from its own past. In other words, there is alwaysno auto-regressive process in r(.

1) ARCH

The GARCH (p,q) model is reduced to theARCH(q) model when p = 0 and at least one ofthe ARCH parameters must be nonzero (q > 0).

PertanikaJ. Soc. Sci. 8c Hum. Vol. 10 No. 2 2002 87


2) Stationary GARCH, SG(p,q)

If the parameters are constrained such that

Z ai + X &j < 1 > they imply the weakly stationaryi-\ j-\

GARCH (SG(p,q)) model since the mean,variance and autocovariance are finite andconstant over time.

3) Unconstrained GARCH, UG(p,q)The parameter of w, a. and ft can beunconstrained, thus yielding the unconstrainedGARCH (UG(p,q)) model.

4) Non-negative GARCH\ NG(p,q)

If p * 0 , q > 0 and w > 0, a :> 0, p. * 0, yieldsthe non-negative GARCH (NG{p,q)) model.

5) Integrated GARCH, IG(p,q)

Sometimes, the multistep forecasts of the variancedo not approach the unconditional variancewhen the model is integrated in variance; that is

,• + 2 0 ; " 1 * Th e unconditional variance for

the IGARCH model does not exist. However, itis interesting that the integrated GARCH orIGARCH (IG(p,q)) model can be stronglystationary even though it is not weakly stationary(Nelson 1990a, b).

6) Exponential GARCH, EG(p,q)

The exponential GARCH or EGARCH (EG(p,q))model was proposed by Nelson (1991). Nelsonand Cao (1992) argue that the nonnegativityconstraints in the linear GARCH model are toorestrictive. The GARCH model imposes thenonnegative constraints on the parameters, a.and /3, while there is no restriction on theseparameters in the EGARCH model. In theEGARCH model, the conditional variance, ht, isan asymmetric function of lagged disturbances,

The coefficient of the second term in g(Z;)is set to be 1 (y = 1) in this formulation. Notethat E/Z/= (2/JT)1 / 2 if Z{ ~ N(0,l).

7) GARCH-in-Mean, G(p,q)-M

The GARCH-in-Mean, G(p,q)-M model has theadded regressor that is the conditional standarddeviation

r{ =

et =et

where kt follows the GARCH process.

8) Stationary GARCH-in-Mean, SG(p,q)-M

This model has the added regressor that is theconditional standard deviation

r, -

where h( follows the stationary GARCH,process.

9) Unconstrained GARCH-in-Mean, UG(p,q)-M


r, =

et =

where ht follows the unconstrained GARCH,UG(p,q) process.

10) Non-negative GARCH-in-Mean, NG(p,q)-M


ln{h,) - w +

where where k( follows the non-negative GARCH,NG(p,q) process.

11) Integrated GARCH-in-Mean, IG(p,q)-M


88 PertanikaJ. Soc. Sci. 8c Hum. Vol. 10 No. 2 2002


where ht follows the integrated GARCH,\G{p,q) process.

12) Exponential GARCH-in-Mean, EGip,q)-M


where h( follows the exponential GARCH,EG(pyq) process.

Since a small lag of the GARCH model issufficient to model the long-memory process ofchanging variance (French et al. 1987; Fransesand Van Dijk 1996; Choo et al 1999), theperformance of GARCH models in forecastingRM-Sterling exchange rate volatility is evaluatedby using SG(1,1), UG(1,1), NG(1,1) IG(1,1),

NG(1,1)-M, IG(1,1)-M, and EG(1,1)-M.

DATA AND METHODOLOGYIn this study, simple rate of returns is employedto model the currency exchange rate volatility ofRM-Sterling. Consider a foreign exchange rateEt, its rate of return r(, is constructed as

E - Ert = —- —. The exchange rate t denotes daily

exchange rate observations.The foreign exchange rate used in this study

is focused on the Malaysian Ringgit (RM) to thePound Sterling. This exchange rate is chosenbecause in addition to the US dollar, the PoundSterling is also one of the major currenciestraded in the foreign exchange markets.Traditionally and historically, the UK has always

been one of the important trading partners ofMalaysia. The data was collected from 2 January1990 to 13 March 1997, from 1810 observations.The daily closing exchange rates were used asthe daily observations. The first 1760 observationsare used for parameters estimation and the last50 observations reserved for forecastingevaluation.

Fig. 1 shows nearly 1810 daily observer crossrates of the Malaysian Ringgit to the PoundSterling, covering the seven years from 2 January1990 to 13 March 1997. Some characteristics ofthe rate of returns, r, are given in Table 1. Themeans and variances'are quite small. The excesskurtosis indicates the necessity of fat-taileddistribution to describe these variables. Theskewness of-0.200 indicates that the distributionof rate of returns for RM-Sterling is negativelyskewed.

The family of GARCH models is estimatedusing the maximum likelihood method. Thismethod enables the rate of return and varianceprocesses being estimated jointly. The log-likelihood function is computed from theproduct of all conditional densities of theprediction errors.

where et = r - fi and h( is the conditionalvariance. When the GARCH{p,q)-M model is

estimated, et = rt - \i ~ b^ht . When there are no

regressors (trend or constant, [i the residuals et

are denoted as r or r( - rt - dfy. The likelihood

function is maximized via the dual quasi-Newtonand trust region algorithm. The starting valuesfor the regression parameters \x are obtainedfrom the OLS estimates. When there areautoregressive parameters in the model, theinitial values are obtained from the Yule-Walker

TABLE 1Summary statistics of currency exchange rate data on rate of returns from 2 January 1990 to

13 March 1997

Currency ExchangeRate

RM/Sterling

n

1809

Mean( x 10-5 )

-3.183

Variance( x 10-5 )

4.076

Skewness

-0.200

ExcessKurtosis

2.370

Source of data: The Federal Reserve, the Central Bank of the United States



t 183 365 547 729 911 1093 1275 1457 1639

92 274 456 638 820 1002 1184 1366 ! 548 ! 730

Time in trading day units

Fig. 1: RM/Sterling, daily from 2 January 1990 to 13 March 1997

estimates. The starting value IE - 6 is used forthe GARCH process parameters. The variance-covariance matrix is computed using the Hessianmatrix. The dual quasi-Newton methodapproximated the Hessian matrix while the quasi-Newton method gets an approximation of theinverse of Hessian. The trust region methoduses the Hessian matrix obtained using numericaldifferentiation. This algorithm is numericallystable, though computation is expensive.

In order to test for the independence of theindices series, the portmanteau test statistic basedon squared residual is used (McLeod and Li1983). This Q statistic is used to test the non-linear effects, such as GARCH effects, present inthe residuals. The GARCH {p,q) process can beconsidered as an ARMA (m3x(p,q),p) process.Therefore, the Q statistic calculated from thesquared residuals can be used to identify theorder of the GARCH process. The Lagrangemultiplier test for ARCH disturbances is proposedby Engle (1982). The test statistic is asymptoticallyequivalent to the test used by Breusch and Pagan(1979).

The LM and Q statistics are computed fromthe OLS residuals assuming that disturbance iswhite noise. The Q and LM statistics have anapproximate (^ ) distribution under the whitenoise null hypotheses.

Various goodnessof-fit statistics are used tocompare the six models in this study. Thediagnostics are the mean of square error (MSE),the loglikelihood (Log L), Schwarz's Bayesianinformation criterion (SBC) by Schwarz (1978)and Akaike's information criterion (AIC) (Judgeet ai 1985).

The 'true volatility' is measured to evaluatethe performance of the six GARCH models inforecasting the volatility in stock returns. As inthe studies by Pagan et ai (1990) and Day et ai(1992), the volatility is measured by

where ris the average return. The measureof the one-step-ahead forecast error is

where hM is generated using the h( equationsof the GARCH models being studied. Theestimated parameters of the GARCH modelssuch as w, or, jS, 6 and 6 are substituted during

the generation of ht+l. In order to show theperformance of GARCH models over a naive no-change forecast, the forecast errors of therandom walk (RW) are calculated as follows:

~ vc

This is a very important naive benchmark inthe comparison of the forecasts from the GARCHmodels (Brooks 1997).

RESULTS AND DISCUSSIONParameter Estimations

The parameter estimates for eleven variations ofGARCH models of the rate of returns series arepresented in Table 2 (a) and Table 2 (b). Thesewithin-sample estimation results enable us toknow the possible usefulness of the GARCH



models in modeling the currency exchange rateseries.

It can be seen from Table 2 (a) that exceptfor \i, all the parameter estimates of the RM/Sterling (w> a and p) are significant at 5% level.However, in Table 2(b), all the two additionalparameter estimates (5 and 6) of the EGARCHand all the GARCH models with means are notsignificant. It appears that for the within-sampleestimations, all the family GARCH modelsperform well in modeling the exchange rate ofRM/Sterling.

In general, it can be concluded that almostall a and fi (ARCH and GARCH terms) of theRM/Sterling series examined are significant.Hence, the constant variance model can berejected, at least for the within-sample estimation.

For the linear GARCH models such as SG(1,1),the sum of a and p is close to unity. Theproperties of + = 1 of IG(1,1) also hold for theseries.

Diagnostics Checking

The basic ARCH (q) model is a short memoryprocess in that only the most recent q squaredresiduals are used to estimate the changingvariance. The results for Q statistic and LagrangeMultiplier (LM) test are shown in Table 3. Thesecan help to determine the order of the ARCHprocess in modeling the RM/Sterling series.

The tests are significant at less then 1%level though order 12. These indicate that theheteroscedasticity terms of the daily RM/Sterlingexchange rate series needed to be modeled by a

TABLE 2 (a)Estimation results of rate of returns for the currency exchange rate

CurrencyExchange Rate Model

Parameter estimates

t Ratio t Ratio

RM/Sterling

-0.125-0.125-0.125-0.125-0.104-0.056

-1.305-1.308-1.306-1.306-1.229-0.622

-0.047

-0.093

-0.518

TABLE 2(b)Estimation results of rate of returns for the currency exchange rate

CurrencyExchangeRate

Model

Parameter estimates

t Ratio co(xlO^) t Ratio a t Ratio(xlO-4)

t Ratio

RM/Sterling 1.61.61.6

1.421.068.368.388.378.377.116.03

1.2461.2431.2431.1120.7871.5591.561.561.561.528

0.7720.7640.7640.350

-291080.00.7580.7650.7560.7560.347

-294460.0

4.7494.6994.6994.842-2.5044.7544.8024.7444.7444.904-2.709

0.0720.0720.0720.0760.1620.0710.0700.0710.0710.0760.163

9.0879.0619.06

10.2116.5889.0069.0349.0059.00510.166.664

0.9100.9100.9100.9240.9710.9110.9110.9110.9110.9240.970

93.61893.69793.678123.44285.15394.63

94.66794.65394.646124.18891.004

PertanikaJ. Soc. Sci. & Hum. Vol. 10 No. 2 2002 91


Diagnostics for

CurrencyExchange Rate

rm/pound

currency exchange

Q(12)

273.447

TABLE 3rate using Q statistic and Lagrange

Diagnostics

Prob>Q(12) LM(12)

0.0001 147.373

Multiplier test

Prob>LM(12)

0.0001

very high order of ARCH model. These resultssupport the use of GARCH model, which allowslong memory processes to estimate the currentvariance of the daily RM/Sterling series insteadof the ARCH model.

Goodness of Fit Tests

The result of the goodness-of-fk statistics for theRM/Sterling series is presented in Table 4. Table5 shows the rankings of various GARCH models.

From Table 5, the ranking of the MSE valueindicates that all the family of GARCH in meanmodels outperform the GARCH models with aslight value of 0.000001. The Log L valueshowever, suggest EG(1,1)-M to be the best modelfor modeling the volatility of RM/Sterling,followed by UG(1,1)-M, NG(1,1)-M andG(l,l)-M. The SBC values in contrast, rankedindifferently SG(1,1), UG(1,1) and NG(1,1) tobe the best model followed by IG(1,1). The AICvalues on the other hand, proposed UG(1,1)and NG(1,1) to be the best two models, followedby SG(U).

From the goodnessof-fit test, it appears thatfor within-sample estimations, almost all theGARCH models outperform the GARCH in mean

models in the SBC and AIC test while in theMSE and Log L test, all the GARCH in meanmodels perform well to model the daily exchangerate compared to their ordinary GARCH modelcounterparts.

One Step Ahead Forecasting

The good performance in the parameterestimation and goodness-of-fit statistics do notguarantee the good performance in forecasting(Choo et ai 1999). The performance of theGARCH models is evaluated through the one-step-ahead forecasting. 50 one-step-aheadforecasts are generated and the mean squareerror (MSE) is calculated to evaluate theforecasting performance. The results of theforecasting for the GARCH models and therandom walk model are shown in Table 6. Therankings of the models based on the performanceof the one-step-ahead forecasting are presentedin Table 7.

In Table 7, the ranking results of MSEsuggest that SG(1,1)-M is the best model forone-step-ahead forecasts, followed by SG(1,1)and G(l,l)-M. It is also noted that, SG(1,1)-M,UG(1,1)-M and NG(1,1)-M clearly outperform

TABLE 4Goodness-of-fit statistics on rate of returns for the currency exchange rates

CurrencyExchangeRate

Rm/pound

Model

SG(1,1)UG(1,1)NG(1,1)IG(1,1)EG(1,1)G(l,l)-MSG(1,1)-MUG(1,1)-MNG(ltl)-MIG(1,1)-MEG(1(1)-M

MSE(xlO-4)

0.410.410.410.410.410.400.400.400.400.400.40

Goodness-of-Fit Statistics

LogL

6525.3716525.4146525.4146521.1516525.7296526.2716526.2326526.2836526.2836521.9926526.674

SBC

-13020.9-13020.9-13020.9-13019.9-13014.1-13015.2-13015.1-13015.2-13015.2-13014.1-13008.5

AIC

-13042.7-13042.8-13042.8-13036.3-13041.5-13042.5-13042.5-13042.6-13042.6-13036

-13041.3



TABLE 5Rankings of the models averaged across the currency exchange based on the performance of

various goodness-of-fit statistics

RM/pound

Model MSE Log L SBC AIC

7 9 1 3

) 7 7 1 1) 7 7 1 1

7 11 4 107 6 9 8

VI 1 4 5 6- M l 5 8 6)-M 1 2 5 4)-M 1 2 5 4- M l 10 9 11i-M 1 1 11 9

TABLE 6Out-of-sample forecasting performance of various GARCH models and random

walk models for the volatility of the currency exchange rates

MSE (xlO-9) of one-step-ahead forecast (forecast period = 50)

Model RM/pound

3.0803.0893.0893.6073.1493.0853.0753.0873.0873.6253.150

RW 6.849

TABLE 7Rankings of the models averaged across the currency exchange rates

based on the performance of one-step-ahead forecasting

Model MSE of one-step-ahead forecast for RM/pound

SG(U) 2) 7) 6

108

VI 3- M 1)-M 4I-M 5•M 11-M 9

RW 12



their ordinary GARCH models counterparts whileEG(1,1) and IG(1,1), in contrast, outperformtheir with mean GARCH counterparts.

In general, almost all the GARCH in meanmodels outperform the ordinary GARCH modelswith the exception of EG(1,1) and IG(1,1).However, the family of GARCH models is clearlybeing proposed instead of their naive benchmark,the random walk model.

CONCLUSIONUsing seven years of daily observed RM/Sterlingexchange rate, the performance of GARCHmodels, including the family of GARCH in meanmodels to explain the commonly observedcharacteristics of the unconditional distributionof daily rate of returns series, were examined.

The results indicate that the hypotheses ofconstant variance model could be rejected, atleast within-sample, since almost all the parameterestimates of the ARCH and GARCH models aresignificant at 5% level.

The Q statistics and the Lagrange Multipliertest reveal that the use of the long memoryGARCH model is preferable to the short memoryand high-order ARCH model.

The results from various goodness-of-fitstatistics are not consistent for RM/Sterlingexchange rates. It appears that the SBC and AICtest proposed GARCH models to be the best forwithin-sample modeling while the MSE and LogL test, suggest the GARCH in mean models tobe best to model the heteroscedasticity of dailyexchange rates.

The forecasting results show that SG(1,1)-Mis the best model for forecasting purpose,followed by SG(1,1) and G(l,l)-M. Almost allthe GARCH in mean models outperform theordinary GARCH models. On the other hand,the family of GARCH models has clearly shownthat they perform better than the naivebenchmark, the random walk model.

ACKNOWLEDGEMENTSWe would like to thank the anonymous refereesand reviewers of this paper who have providedus with many useful comments and suggestions.This research was supported by the short termresearch grant funded by the Ministry of Science,Technology and the Environment, Malaysia,through the Faculty of Economics andManagement, Universiti Putra Malaysia.

REFERENCESAKGIRAY, V. 1989. Conditional heteroskedasticity in

time series of stock returns: Evidence andforecasts. Journal of Business 62: 55-80.

BAILLIE, R. T. and T. BOLLERSLEV. 1989. The messagein daily exchange rates : A conditional variancetele. Journal of Business and Economic Statistics7: 297-305.

BOLLERSLEV, T. 1986. Generalized autoregressiveconditional heteroskedasticity. Journal ofEconometrics 31: 307-327.

BOLLERSLEV, T. 1987. A conditional heteroskedastictime series model for speculative prices andrates of return. Review of Economics and Statistics69: 542-547.

BOLLERSLEV, T. 1990. Modelling the coherence inshort-run nominal exchange rates: Amultivariate generalized ARCH model. TheReview of Economics and Statistics:. 498-504.

BOLLERSLEV, T., R. Y. CHOU and K. F. KRONER. 1992.Arch modelling in finance. A review of thetheory and empirical evidence. Journal ofEconometrics 52: 5-59.

BRCX)KS, C. and S. P. BURKE. 1998. Forecastingexchange rate volatility using conditionalvariance models selected by informationcriteria. Economics Letters 61: 273-278.

CAPORALE, G. M. and N. Prrris. 1996. Modellingsterling-deutschmark exchange rate: Non-lineardependence and thick tails. Economic Modelling13: 1-14.

CHEN, A. S. 1997. Forecasting the S & P 500 indexvolatility. International Review of Economics &Financed: 391-404.

CHEUNG, Y. W. and C Y. P. WONG. 1997. Theperformance of trading rules on four Asiancurrency exchange rates. Multinational FinanceJournal 1(1): 1-22.

CHOO, W. C, M. I. AHMAD and M. Y. ABDULLAH.1999. Performance of GARCH models inforecasting stock market volatility. Journal ofForecasting 18: 333-343.

COPKIAND, L. S. and P. WANG. 1994. Estimating dailyseasonality in foreign exchange rate changes.Journal of Forecasting 13(6): 519-528.

DAY, T. E. and C. M. LEWIS. 1992. Stock marketvolatility and information content of stockindex options. Journal of Econometrics 52: 267-287.



DIEBOLD, F. X. and M. NERLOVE. 1989. The dynamicsof exchange rate volatility : A multivariatelatent variable factor ARCH model. Journal ofApplied Econometrics 4(1): 1-21.

DIEBOLD, F. X. 1988. Empirical Modeling of ExchangeRate Dynamics. Berlin: Springer-Verlag.

DUAN , J. C. and Z. W.JASON. 1999. Pricing foreigncurrency and cross-currency options underGARCH. The Journal of Derivatives (Fall).

ENGLE, R. F. 1982. Autoregressive conditionalheteroskedasticity with estimates of the varianceof U.K. inflation. Econometrica 50: 987-1008.

ENGLE, R. F., M. L. DAVID and P. R. RUSSEL. 1987.Estimating time varying risk premia in theterm structure: The ARCH-M model.Econometrica 55: 391-407.

EPISC;OPOS, A. and J. DAVIKS. 1995. Predicting returnson Canadian exchange rates with artificialneural networks and EGARCH-M models.Neural Computing and Applications (forthcoming).

FRENCH, K. R., G. W. SCHWERT and R. F. STAMBAUGH.1987. Expected stock returns and volatility.Journal of Financial Economics 19: 3-30.

FRIEDMAN, D. and S. VANDERSTEEL. 1982. Short-runfluctuations in foreign exchange rates :Evidence from the data 1973 - 1979. Journal ofInternational Eonomics 13: 171-186.

HOPPER, G. P. 1997. What determines the exchangerate: Economics factors or market sentiment.Business Review: 17-29. Federal Reserve Bank ofPhiladelphia.

HSIEH, D. A. 1988. The statistical properties of dailyforeign exchange rates: 1974 - 1983. Journal ofInternational Economics: 129-145.

HSIEH, D. A. 1989a. Modeling hetereskedasticity indaily foreign exchange rates. Journal of Businessand Economic Statistics 7: 306-317.

HSIEH D. A. 1989b. Testing for non-lineardependence in daily foreign exchange rates.Journal of Business 62: 339-368.

LAOPODIS, N. T. 1997. U.S. dollar asymmetry andexchange rate volatility. Journal of AppliedBusiness Research 13(2) Spring: 1-8.

LAU, A., H. LAU and J. WINGENDER. 1990. Thedistribution of stock returns : New evidenceagainst the stable model. Journal of Businessand Economic Statistics 8 (April): 217-223.

LOBO, B. J. and D. TUFTE. 1998. Exchange ratevolatility: Does politics matter? Journal ofMacroeconomics 20(2) Spring: 351-365.

MCKENZIE, M. D. 1999. Power transformation andforecasting the magnitude of exchange ratechanges. International Journal of Forecasting 15:49 - 55.

PESARAN, B. and G. ROBINSON. 1993. The Europeanexchange rate mechanism and the volatility ofthe sterling-deutschmark exchange rate.Economic Journal 103: 1418-1431.

SANTIS, G. D. and B. GERARD. 1998. How big is thepremium for currency risk. Journal of FinancialEconomics 49: 375-412.

TAKEZAWA, N. 1995. A note on intraday foreignexchange volatility and the informational roleof quote arrivals. Economics Letters 48: 399-404.

TAYLOR, S. J. 1986. Modelling Financial Time Series.New York: Wiley.

(Received: 11 June 2001)



The Influence of Value Orientations on Service Quality Perceptions in aMono-Cultural Context: An Empirical Study of Malay University Students

HAZMAN SHAH ABDULLAH 8c RAZMI CHIKFaculty of Administrator & Law

Universiti Teknologi MARA40450 UiTM Shah Alam

Selangor, Malaysia

Keywords: Service quality, culture studies, value variations, variance analysis

ABSTRAK

Kajian hubungan mutu perkhidmatan dan nilai-budaya telah secara tradisi menggunakan kaedahantara budaya untuk membuktikan kesannya. Tujuan kajian ini ialah untuk menunjukkan bahawaterdapat perbezaan yang penting dalam sesuatu budaya dan perbezaan ini mempunyai implikasiterdapat reka bentuk, penyediaan dan mutu perkhidmatan. Selaras dengan pendirian ini, kajianini menyelidiki nilai-budaya dalam konteks suatu kumpulan budaya tertentu. Kajian melibatkanseramai 712 siswa-siswi Bumiputera sebuah universiti tempatan telah menghasilkan 2 gugusannilai yang signifikan. Analisis varian menunjukkan bahawa gugusan 'True Traditionalists'dan 'Transitory Traditionalists' memperlihatkan kesan yang berbeza kepada dimensi mutuperkhidmatan. Hasil kajian menyokong pendirian bahawa kajian hubungan budaya-mutuperkhidmatan mestilah juga fokus kepada perbezaan dalam sesuatu budaya.

ABSTRACT

Service quality and culture studies have traditionally used polar opposite cultures to make theircase. This paper argues that these cultural extremes conceal significant variations in culture andhas implications for service design, delivery and quality. The study explores the existence of valuevariations within ostensibly homogenous groups. It is posited that the knowledge of thesespectrum of value orientations will enhance the service marketers' ability to 'situate' services asthey enter new markets or introduce service innovations. A study conducted among 712 Malayuniversity students produced 2 significantly different value clusters. Variance analysis showed thatthese clusters labeled as True Traditionalist and Transitory Traditionalist have significantlydifferent impact on service quality dimensions. The findings support the argument that servicequality and culture studies must examine between as well as within culture variations.

INTRODUCTIONThe rapid extension of service products to globalmarkets has invoked the question of situatingthe services in the local cultural and socialenvironment (Matilla 1999; Stauss and Mang1999). Besides the highly general and stylizedcharacterisation of national groups that followedHofstede's seminal work, service managers havelittle to go on in designing and localizing theirservices. The globalisation of service productsand inherent intangibility and humaninteractivity that marks most services has sharplyraised the potential for service product failures.

There is, therefore, a growing interest inunderstanding the interaction between thenational and sub-national cultural influences andservice products. Concomitantly, there is anoticeable burst of research examining theculture-service nexus in regions other thanEurope and North America (Winstedt 1997;Stauss and Mang 1999). Traditionally, this meantthe need to understand the cultures of theEuropean and Asians.

Reflecting this need Anderson and Fornell(1994) in their 'consumer satisfaction researchprospectus' called for more systematic investigation

Hazman Shah Abdullah & Razmi Chik

into the variations in satisfaction across nations.Due to the interactive and intangible nature ofservices, cultural expectations play an importantrole in predisposing the customers towards theconsumption experience and their attention andreaction to cues in the service environment. Itstrongly influences the values that customers arelikely to assign to specific service attributes, theperception of the characteristics of the serviceproviders and the strength of their reaction tothe presence or absence of the attributes (Matilla1999).

Since their appeal, there have been severalstudies to examine the influence of culture oncustomer satisfaction (Winsted 1997; Donthuand Yoo 1998; Matilla 1999). Despite the obviousrole of culture in service quality, the understandingis still rather nascent. As services become moreglobal, there is need to develop betterunderstanding of the influence of differentcultures on different dimensions of servicequality.

The research thus far has exclusively focusedon national and ethnic groups/cultural groups.Because these groups are distinct and commonlybecome the basis of marketing decisions, theyare selected as the natural units of observation.While broad cultural categories still form thebasis of global market segmentation, the culturalstereotyping often conceal significant variationswithin groups that allow for finer segmentation(Matilla 1999). Yet, much of the culture andservice quality research relies on the most notablecultural denomination, the national culture.Additionally, the focus of culture-service qualityinteraction study has been on polar oppositecultures. The national cultural classifications anddistinctions conceal much of the distinct culturalsub-groups. These subgroups evince variations,which range from shades of the main culture tovastly different cultural preferences withinsupposedly homogenous cultures. As moreevidence of culture-service quality nexusbecomes available, the question is no longer ofthe connection between the two but rather theexpansion of research to even ostensiblymonocultural environments. There is need,therefore, to look for cultural variations withinsupposedly homogenous cultural groups.

REVIEW OF LITERATURE

Of particular interest within the service qualityresearch stream has been the interaction between

the service provider and the customer. Thedominant service quality model places thecustomer expectation as the subjective standardby which a customer evaluates the serviceperformance (Zeithaml, Parasuraman and Berry1993). Although the explanatory role of customerexpectations in service quality assessment hasbeen questioned, it still is accepted as providingvaluable means to judge performance assessmentsby the customer (Cronin and Taylor 1992). Theexpectation itself is a product of a complexnumber of factors. The values or culturalorientations of the customer are believed toprovide the broadest framework to understandexpectations. Consequently, a new stream ofresearch has begun to explore the role of culturein expectation formation and how differentcultural orientations impact their evaluation ofthe various elements in the service performance.Winsted (1997) succinctly brings out theconceptual link between service encounters andsocial encounters through the followingobservation;

"Because service encounters are social encounters,rules and expectations related to services encountersshould vary considerably according to culture, yetvery little guidance has been provided regardingthe influence of culture on perceptions of serviceprovision" (p.106).

Many writers have argued for the need forgoods and services to be adapted to the differentlocal cultures. Alden, Hoyer and Lee (1993)showed how the use of humour in advertisingmust be carefully vetted for offensive elementswhen applied cross-culturally. Generally, thecultural comparisons have been between culturesthat can be characterised as polar opposites likethe Japanese and the Americans. Winsted (1997)studied the influence of the cultural values onthe service quality expectations and evaluations.She found that the Americans expectedegalitarianism in service and higher degree ofpersonalisation while their Japanese counterpartpreferred more formality in treatment. Malhotra,Ugaldo, Agarwal and Baalbaki's (1994) study isamong the few studies on the cultural dimensionsof developing and developed countries and theireffect on the service quality dimensions. Theyfound that the value orientations as measuredvia Hofstede's 5 dimensional continua had asignificant bearing on the service qualityevaluations of the respondents. The findings


The Influence of Value on Service Quality Perceptions in a Mono-Cultural Context

point toward the need to localise in internationalmarketing. Donthu and Yoo (1998) examinedthe effect of cultural values captured viaHofstede's five dimensional scales and theSERQUAL dimensions of reliability, assurance,empathy, responsiveness and tangibles. On mostof the service quality dimensions there werestatistically significant differences in theirevaluations of the retail banking services.Although people processing services are positedas most susceptible to cultural effects (Furrer,Ben and Sudharshan 2000), Matilla (1999)explored the impact of culture on hedonicservices. The experience rich services permit thecultural nuances to play a greater role than inother forms of services. Accordingly, it wasreported that Western and Non-western businesstravelers responded to different service cues.Generally it was found that Asian travelers paidmore attention to non-tangible and nonverbalcues more that their western counterpart.However, the study also highlighted and alertedattention to the variations possible withinotherwise monolithic cultures. Stauss and Mang(1999) tested the hypothesis that inter-culturalservice encounters are more problematic thanintra-cultural encounters using critical incidentmethod. Interestingly, the results confirmed thereverse. Intra-cultural encounters were moreproblematic than the inter-cultural ones. Thestudy also used somewhat polar cultures in testingthis hypothesis.

Furrer et ai (2000) provide a recent study ofthe impact of culture on service quality. Theystudied the cultural orientations of American,Asian and European students. They developed acultural service quality index that captured theinteraction between the service qualitydimensions and the cultural dimensions. Theyshowed how the groups of students could besegmented on the basis of their culturalproclivities and the service quality dimensions ofvalue.

It is apparent that most of the above-mentioned studies have sought to explore theculture-service question using polar oppositescultures. The use of these 'maximally' differentcultures is understandable as they enhance thepower of the design to test the postulations. TheWestern vs. Non-Western or American vs. Non-American designs have shown that the culture-service effect is real and must be addressed byglobal service producers.

What is of significance is that thesepostulations can be more stringently tested ifthey are subjected to less extreme culturalvarieties. The exploration of the conceptuallyviable thesis of finer cultural variations and theireffect of services evaluations has been put forthby Matilla (1999) who observed that ".. .consumerexperiences do not remain stable across culturesbut instead are open to influences of specificcultures". Indeed, the study of this postulationwithin what is known as homogenous cultures,can open the same advantages to marketers ashas been suggested about inter-cultural studiesin international marketing. Niche marketing canimmensely benefit from the understanding ofthe differences in what is otherwise believed tobe mono-cultural societies, by exploiting theinteraction between specific cultural nuancesand the sensitivity to specific service dimensions.Where the service attributes can be easilymodified, the within culture value orientationscan be a basis to customise services for the nichemarkets.

Problem Statement

From the review of the literature, it is evidentthat there is a dire need for culture-servicestudies to examine the role of value orientationswithin a culturally homogenous context (Winsted1997; Matilla 1999). The focus on within culturevariations in values will add greater credence tothe culture-service studies and allow for finerdistinctions and their a t tendant serviceimplications. This study explores this new andpotentially fruitful focus question. The centralresearch question is whether there are distinctvalue orientations within a cultural group.

The Conceptual Framework

The relationship between the service and cultureemanates from the basic characteristic of servicesas intangible and interactive (Shostack 1977;Lovelock and Wright 2002). Because serviceencounters are essentially social exchanges, thevalues undoubtedly colour the perception ofboth parties. Though not directly apparent,values underpin the expectations, biases,preferences, self-confidence etc. of the customers.Though not directly observable, values have beenconceptualized along several key dimensions.One such framework is advanced by Hofstede(1980). Hofstede captured the 'collective


Hazman Shah Abdullah 8c Razmi Chik

» Power distance• Uncertaintly Avoidance• Masculinity-Feminity• Collectivism-individualism

Service QualityPerception

ResponsivenessReliabilityTangibleAssuranceEmpathy

Fig. 1: Conceptual framework

programming of the mind' via four valuedimensions namely, Power Distance (PD),Uncertainty Avoidance, Masculinity-femininityand Collectivism-individualism (Fig, 1). Powerdistance refers to the acceptance of asymmetricalpower distributions by members of a group orcommunity. In high PD societies, hierarchy isaccepted and may even be revered. In servicecontexts, PD conditions the perceptions of thestatus of the service provider and the desiredbehaviour on the part of the customer.Masculinity-femininity relates to the extent towhich strong, aggressive and assertive behavioursare preferred or accepted or desired. Uncertaintyavoidance is the aversion to risks andunstructured behaviour situations. The clarity ofone's role is desired as opposed to self-development of the roles in any context. Finally,collectivism-individualism indicates the premiumplaced on self as opposed to the group, be it thesociety, community or the team. These fourdimensions are landmarks of value orientationof any group. The impact of the values onservices is eventually felt in the customers'assessment of the service quality itself. The valuesare expected to impact service quality throughthe customers' perceptions of the extent ofresponsiveness, reliability, empathy, assuranceand tangibles. These dimensions are susceptibleto the preferences and biases that the customerbrings into a service encounter. The values tendto affect the customer's position vis-a-vis theservice provider by creating mental zones ofcomfort and discomfort and culturallyappropriate roles and behaviours.

However, the interaction between the valuesand the services is not likely to be the same in alltypes of service encounter (Chase 1978; Lovelockand Wright 2002: 54). Some services involvehigh contact between the customer and the

service provider. In high contact services, theextended nature of the social exchange createsmore opportunity for values to affect servicequality perceptions. In low contact services, theinteraction may be short or even momentary.Consequently, the social expectations and valueorientations are unlikely to leave much impact.

Research Hypotheses

HI: There are significantly differing valueorientations.

H2: Value orientations correlate significantly withservice quality dimensions.

H3: The influence of value orientation on theservice quality perception is more evident inhigh contact rather than low contact services.

RESEARCH DESIGNA cross-sectional correlational study was carriedout involving 712 students of UiTM to determinethe influence of value orientation of Malayuniversity students on their service qualityexpectations and perceptions. The Malays haveexperienced dramatic socio-economic changesover the last two decades. This has introducedand amplified the cultural variations within theMalay community. The current concern aboutthe lack of unity among the Malays is arguablyengendered by greater variations in values andconsequently, different expectations andassessments. The Malay university students are aclose microcosm of the larger Malay society.Therefore, it offers a good setting to test theresearch question advanced in this study. Arepresentative, though not a random, sample ofthe student population was obtained for thisstudy from two of the 13 campuses of thisuniversity. The main campus represents an urbancentre while the East Coast campus captures amore rural background.

To examine the impact of value orientationon service quality expectations, 3 types ofuniversity services having different degrees ofcustomer contact were identified. These servicesrange from counseling (high contact) to medical(moderate contact) to library services (lowcontact) (Table 1). It is well established that notall services allow or require prolonged contactwith the customer/user. The influence of theuser's value orientation is most likely to mattera great deal in shaping his/her involvement andhis/her reaction to the behaviour of the service



TABLE 1Distribution of the sample

Nature of Service Sample Size Sample Size(actual) (planned)

High ContactCounseling (S. Alam 8cTerengganu) 149 200

Moderate ContactMedical Care (S. Alam &Terengganu) 273 200

Low ContactLibrary Service (S.Alam8c Terengganu) 290 200

Total Sample 712 600

provider in high contact services. Conversely,the influence of value orientation of the studentis less likely to impact service quality where thecustomer and service provider contact ismomentary, limited and tangible. The three typesof services were used to detect the moderatingrole of contact in examining the influence ofvalues on service quality perceptions of thestudents.

Development of the Measurement InstrumentsAlthough the two constructs involved in thisstudy have been defined and measured in manyprevious studies, a conscious decision was madeto review this definition and the performance ofthese instruments. Hofstede's measures of valueswere work organisation based. Their relevanceand performance in the context of the servicesand the sector under study in this research arequestionable. Therefore, several items weregenerated for each of the five value dimensions.The item development followed the processsuggested by Dunn et al (1994). The items werereviewed by peers familiar with the subject asrecommended by Dunn et al. (1994). Twoacademics were required to link the items to thedimension the item appears to measure. Throughthis process the items that were not identified bythe peers as linked to a dimension were dropped.This process of substantive validation is stated byDunn et al. (1994) as the most crucial step inconstruct validation because substantiveconvergence should precede statisticalconvergence.

The value orientations were measured usingHofstede's 5 dimensional instrument (Hofstede1980,1991). These dimensions are Power-

Distance, Individualism-collectivism, UncertaintyAvoidance, Masculinity-femininity and TimeOrientation. Although this instrument wasdeveloped and used to measure the nationalvalues, it has been successfully used to studyculture at an individual level (Matilla 1999).

The original measures of value orientationdeveloped by Hofstede were specifically focussedon work-related values. Since this study addressesa university context, the original items weredeemed inappropriate. Based on the 5 key valueorientation dimensions, 28 items were generated.Only items that passed the substantive validationprocess were finally accepted for use in the pilottest. The pilot test based on a sample of 30individuals was collected and the CronbachAlphas were determined. The measure attainedthe minimum threshold of 0.7 (Nunnally 1978).In the study however, the reliability coefficientswere slightly below the recommended thresholdof .7. Collectivity, masculinity, uncertaintyavoidance and power distance achieved aCronbach alpha of 0.66, 0.60, 0.63, and 0.60respectively. Since the Cronbach alphas wereonly marginally lower than the threshold andthe lower Cronbach alphas have been used inorganisational studies, we did not think that thiswould seriously affect the outcome.

The service quality perception was measuredusing the SERQUAL dimensions (Parasuramanet al. 1988). This instrument has five servicequality dimensions namely; tangibles,responsiveness, reliability, assurance andempathy. Parasuraman et al. (1989) viewed servicequality as the difference between the perceptionand the expectation. As Cronin and Taylor(1992) pointed out, the measure of perceptionitself is sufficient measure of the service withoutthe weighting by expectation attached by theclients. Because Cronin and Taylor's approachyields a simpler measure, we have adopted it forthis study. Just like Hofstede's measures, theSERQUAL items are generic items that may beinappropriate for the present educationalcontext. Therefore, the items were developedreflecting the dimensions and put through thesame substantive validation process as in thecase of value orientation measures. The itemswere measured on a 7-point Likert scale with 1denoting Strongly Disagree. All measures attaineda minimum Cronbach Alpha of .60 in the mainsurvey, slightly less than the values obtainedduring the piloting stage. Though the reliability

Pertanika J. Soc. Sci. 8c Hum. Vol. 10 No. 2 2002 101


coefficients were lower than Nunnally's .70, lowerreliabilities have been used in published studies(Hinkin 1995).

Data Collection and Data AnalysisData were collected from two campuses namely,the Shah Alam and the Terengganu campuses.Trained enumerators were positioned at theservice centres to identify and sample therespondents (based on the quota samplingmethod) as they left the service centre after aservice encounter. This method allowed foraccurate recollection of the experience than at alatter time. They were asked to complete aquestionnaire containing the instruments. Inthe case of the counseling service, the counselorsprovided the questionnaires to the respondentswhen the students came in for consultation.This deviation was unavoidable because of theunplanned and irregular nature of the serviceneed. Basic descriptive statistics was used toexplore the distributional and locationalcharacteristics of the variables and determinethe appropriateness of the statistical techniquesgiven the descriptive properties. Unlike otherstudies that have used demographic factors toexamine the existence and the influence ofvalue orientations on variables of interest, thisstudy follows a method suggested by Furrer et ai(2000). Value orientations are composed ofunique combinations of the value dimensions.Demographic, ethnic and other common a prioriclassifications may not necessarily correlate withvalue orientations. As such, using any one ofthese a priori groups may result in erroneousfindings. Therefore, Furrer etal (2000) suggestedthat the value groups must be empirically derivedthrough the use of grouping techniques likecluster analysis. Accordingly, cluster analysis wascarried out to examine the cluster properties ofthe respondents. Subsequent analyses of variance(ANOVA) used the value clusters (Traditionalistsand Transitory Traditionalists) to examine therelationship between the value clusters andservice quality dimensions.

Profile of the Respondents

There is a greater representation of studentsfrom the Terengganu campus than from theShah Alam campus in the sample. This reflectsmore the accessibility to respondents and theintensity of use of the selected services thananything else.

There is two to one ratio of female to malestudents. This skewed distribution is reflective ofthe overall student composition in UniversitiTeknologi MARA. From Table 2 it is clear thatthe respondents are preponderantly Diplomaholders. This reflects the general distribution ofstudents and also because these students aregiven priority for campus housing. They,therefore, are in campuses and presumably, alsouse the services more than others who areaccorded the same privileges.

In keeping with the university's socialcommitment, the bulk of the respondents fallunder the category of the lower income group.The distribution is also influenced by the greatershare of the Terengganu campus in the totalsample, which attracts students from the EastCoast which is a lower income region in Malaysia.The distribution of the sample is weighted slightlyin favour of the library services. This is, asexplained in the methods section, an outcomeof the nature of the use of the library services.Library services are more intensively used ascompared with medical and counseling services.The former are dictated by the nature of thecampus activity while the latter are peripheralservices.

FINDINGSIntra-Cultural Variations

The correlation matrix in Table 3 displays thespecific dynamics of the culture-service qualityrelationships. All correlation coefficients > .10are significant. The 4 dimensions of the culturalorientations are not strongly correlated,indicating that the dimensions are distinct andnot overlapping ones. The highest correlation isbetween uncertainty avoidance and collectiveorientation (.410). The correlation between theservice quality dimensions and culturaldimensions is of particular interest. Powerdistance is significantly correlated with all servicequality dimensions except responsiveness.However, the correlation values are small or low.This suggests that while the relationship issignificant, the impact of this orientation onservice quality is quite limited at best. Thecorrelation between the service quality and thecultural orientation dimensions is low. This is tobe expected given that this study is focussed onexamining relationship between thesedimensions within a mono-cultural context.Uncertainty avoidance also displays similar



TABLE 2Profile of respondents

No.

CampusShah AlamTerengganu

GenderMaleFemale

ProgrammeDegreeDiplomaCertificateOthers

Parents Income*<500501-10001001 -15001501 -20002001 - 25002501 - 30003001 - 35003501 - 4000>4000

Type of ServiceCounselingMedicalLibrary

* n = 622

297415

234478

246449

98

16722415563329272447

149273290

4258

3367

356311

27362955448

213841

TABLE 3Correlation between the service quality and culture variables

No Variables 1 4

123456789

EmpathyAssuranceResponsivenessTangibleReliablePower DistanceUncertainty AvoidanceMasculinity-FemininityCollectivism-Individualism

.585**

.696**

.608**

.530**

.106**

.121**.043

.217**

.529**

.523**

.465**

.108**

.187**.088*

.252**

.602**

.581**.030.076*-.051

.183**

.566**

.182**

.174**.061

.282**

.172**

.249**

.107**

.265**

.197**

.187**

.132**.295**.410** .285**

p<.005, **p<.001 (2-tailed test).

correlation with all service quality dimensionsbut appears to be relatively more correlated withReliability. Masculinity is least significantlycorrelated with the value orientations.

The value orientations of the students weremeasured via 5 dimensional continua providedby Hofstede (1990) but time orientation itemsfailed to show satisfactory convergence andtherefore, have been excluded from further

analysis. It has been argued that since culturalvalues condition the mind and behaviour in acollective fashion, it should be combined tocreate recognisable value groups for analysis(Furrer et al. 2000). Consequently, it is imperativethat the value orientations are understood as abundle or cluster rather than individual variables.Following in the footsteps of Furrer et al (2000),


Hazman Shah Abdullah 8c Razmi Chik

the data was cluster analysed to detect groupsthat have distinctive combinations of the 4 valuedimensions. Cluster analysis generated twodistinct groups. We have labeled the clusters asTrue Traditionalist (cluster 1) and TransitoryTraditionalist (cluster 2).

The value orientations were measured on a7-point scale with 1 denoting low and 7 high.Cluster 1, the True Traditionalists display agreater proclivity to the collective interest,appears to accept the appropriacy of greaterassertiveness, greater power distance in generalrelationship and greater aversion to uncertainty.Typical Malay society places a very high premiumon collective rather than individualistic interest,the culture can only be classified as feminine innature with its emphasis on gentleness, decorumand politeness in social encounters and even indisagreement, accepting and also veneratingpower distance and preferring to avoiduncertainty. The Transitory Traditionalists(cluster 2) scored lower on all dimensions of the

value orientations but in case of Masculinity, it isonly marginally lower than their TrueTraditionalists peers. The Transitory Traditionalistsare a group experiencing some dilution of thecultural values that typify the Malay communityat large. Based on the evidence from Table 4, wecan conclude that there are distinctly differinggroups within the Malay student community.Therefore, the 1st hypothesis that there aredistinct sub-cultural groups within presumablyhomogenous groups is supported.

Table 5 provides some answers to thequestion whether there is significant relationshipbetween the value clusters and the perceptionsof service quality. The one-way ANOVA showsthat the service quality dimensions differsignificantly between the 2 value clusters. Thisthen provides the support for the hypothesisthat value orientations influence or have someimpact on service quality perceptions. Thus, thesecond hypothesis is also supported.

TABLE 4Value clusters

Value Orientations Clusters

True TransitoryTraditionalists Traditionalist

(Means) (Means)

CollectivityMasculinityPower DistanceUncertainty Avoidance

6.14 (high)4.69(med)4.18(med)5.58(high)

4.72 (med)3.99(med)3.90(med)4.69(med)

N 417 295

TABLE 5Influence of cultural clusters on service quality

Service QualityDimensions

Empathy

Assurance

Responsiveness

Tangible

Reliable

ValueClusters

Between groupsWithin group





Sum ofSq.

33.596925.860

13.076407.631

18.063763.768

37.252742.772

42.572724.217

df

1710

1710

1710

1710

1710

MeanSq

33.5961.304

13.076.574

18.0631.076

37.2521.046

42.5721.020

&g

.000

.000

.000

.000

.000

104 Pertanika J. Soc. Sci. & Hum. Vol. 10 No. 2 2002


The last hypothesis advanced was whetherthe postulated relationship between valueorientation and service quality differs given thetype of service involved. Multivariate analysis ofvariance showed that that there are nointeractions between service type and valueorientations. Therefore, the last hypothesis thatthe relationship between value orientation andservice quality will be more distinct in highcontact service was not supported.

In summary, the findings of this studysupport two of the three hypotheses advanced.Significant variations in values can be observedeven within such homogenous groups as Malayuniversity students. These variations are notwithout influence on the service qualitydimensions.

DISCUSSION AND CONCLUSIONFrom a theoretical standpoint, this study expandson the current culture-service quality researchby seeking out finer distinctions and how theymay be pertinent to service providers. Pastresearches tended to use cultural extremes orpolar opposites to show the effect of culture ofservice perceptions (Furrer et al 2000; Winsted1997; Matilla 1999). While this is an admirableapproach and design, the design is intended toshow the postulated relationship. In fact, it couldbe argued that the design is too powerful andtherefore, the outcome is almost a certainty.This study by examining the same issue in anintra-cultural setting is actually putting thepostulation to a much more stringent test thanhas been the case thus far. By seeking out therelationship between service quality and valueorientations intra-culturally, the hypothesis is putto a more stringent test.

The study also provides some evidence ofthe existence of a spectrum of value orientations(though the range is limited) within an ostensiblyhomogenous group. Although polar oppositecultures dominate culture-service studies, valueorientations within homogenous cultures areequally valid and fruitful areas of scrutiny(Winsted 1997). This study established that therewere two cultural clusters. These groups, labeledas Transitory Traditionalists and TrueTraditionalists, provide a significantly differentvalue profile. While both groups display Malaycultural proclivities, the Transitory Traditionalistsis markedly less respectful of old values. Theuniversities have become grounds to question

the wisdom of the old ways. The value profileshowed here reflects the changing socio-psychological landscape within the universitystudent population and to some extent, withinthe society at large. The lack of strongly distinctvalue orientations among the groups is more afunction of the homogeneity of the sample thananything else. If a more heterogeneous samplehad been acquired, the value profile would havevaried much more than observed in this study.

From a managerial standpoint, the serviceproviders in the public sphere, especially thepublic tertiary institutions must be more awareof the value orientations and how they impactthe many quality initiatives that are currentlyinstituted (Cheong 2000). Students still haveand are therefore, likely to display values thatplace a high premium on collective interests.Therefore, services that explicitly or implicitlyrequire one to show individualistic tendenciesmay cause significant discomfort or dissonance.Students are likely to feel at ease when doingthings together and for the benefit of all ratherthan self only. Relatively high tolerance for powerdistance is expected to manifest itself in ratherpassive, meek and unassertive behaviour. Thisdisposition will prevent effective feedback fromthe service users. Users are likely to be verycognizant of the structure, hierarchy, order andauthority and thus less inclined to question orcomplaint or provide feedback which is notanonymous. Quite unexpectedly, the respondentshave expressed a more masculine interest. Thisproclivity for assertive behaviour does not quitefit with the other orientations especially powerdistance and collective interest. Given that theMalay community is in the midst of intense andoften acrimonious changes socially and culturally,conflicting behaviour especially among theimpressionable population may account for thecontradiction (Mastor, Jin and Cooper 2000).Making services changes without understandingthe cultural nuances of the users will result insupplier-orientated changes. The interactionbetween the value orientations and service qualitydimensions allows for changes that are alignedwith the cultural preferences of the students.Many service innovations have strong implicitvalue base. Most, if not all innovations wereborn in Caucasian cultures. The effectiveness ofthese innovations is implicitly a product of thevalue orientations of the co-producers orco-creators - the service recipients. Changes that



embody greater egalitarianism or empowermentmay actually have the effect of inconveniencingthe service users because of their value system.The total quality management initiatives willonly be effective if the user values are capturedand used to bring about service innovation.

As is customary, some caveats are in order.The measures did not display a high level ofreliability as was expected. Improvements andfurther purification of the measures will enhancethe strength the correlations and the clarity ofthe nomological network. The use of a fairlyhomogenous sample (Malay university students)probably did restrict the range of valueorientations. Perhaps, a more representative mixof the members of the Malay community willdisplay even greater variation in valueorientations than is the case here. Future studiesshould examine the interactions between valueorientations and the service quality dimensionsin specific service sectors to reveal profitablepossibilities in service adaptation.

In summary, the study has shown that thereis risk in stereotyping the cultural traits of Non-Western societies and in this particular case, theMalays. There exist significant value variationswithin ethnic and sub-cultural groups.Recognising this will open up new possibilitiesfor service adaptations, which is necessary tocompete in the highly competitive market placeas well as the non-competitive public sector. Aswe move beyond the first round of TQMinitiatives, finer distinctions will become a centerof focus in further service improvements andadaptations.

REFERENCESALDEN, D. L., W. D. HOYER and C. LEE. 1993.

Identifying global and culture specificdimensions of humor in advertising: Amultinational analysis. Journal of Marketing 57April: 64-75.

ANDERSON, E. W. and C. FORNELL. 1994. A customersatisfaction research prospectus. In ServiceQuality: New Directions in Theory and Practice, ed.Roland T. Rust and Richard L. Oliver, p. 241-268. CA Sage: Thousand Oaks.

CHASE, R. C. 1978. Where does the customer fit ina service operation? Harvard Business ReviewNov-Dec: 137-142.

CHEONG, C. Y. 2000. Cultural factors in educationaleffectiveness: A framework for comparative

research. School Leadership and Management20(2): 207-225.

CRONIN, J. J. and S. A. TAYLOR. 1992. Measuringservice quality: A reexamination and extension.Journal of Marketing Research 3: 55-68.

DONTHU, N. and B. Yoo. 1998. Cultural influenceson service quality expectation. Journal of ServiceResearch 1(2): 178-186.

DUNN, S. C, R. F. SEAKER and M. A. WALLER. 1994.

Latent variables in business logistics research:Scale development and validation. Journal ofBusiness Logistics 15(2): 145-172.

FURRER, O., B. S. C. Liu and D. SUDARSHAN. 2000.The relationships between culture and servicequality perceptions. Journal of Service Research2(4): 355-371.

GRONROOS. 1993. Services Management and Marketing.Singapore: Maxwell Macmillan.

GuMMESsoN, E. 1991. Marketing Organisations inservice businesses: The role of part-timemarketers. In Managing and Marketing Servicesin the 1990% ed. Teare, R., L. Muotinho andN. Morgan, Cassel, p. 35-48.

HINKIN, T. R. 1995, A review of the scaledevelopment practices in the study oforganizations. Journal of Management 21: 967-988.

HOFSTEDE, G. 1980. Cultures Consequences; InternationalDifferences in Work Related Values. Beverly Hills:Sage.

HOFSTEDE, G. 1991. Cultures and Organisations:Software of the Mind. London: McGraw-Hill.

KOTLER, P. and G. ZALTMAN. 1971. Social marketing:An approach to planned social change^/ourrza/of Marketing 35 July: 3-12.

LOVELOCK, C. H. 1995. Services Marketing. New York:Prentice-Hall.

LOVELOCK, C. H. and L. WRIGHT. 2002. Principles ofService Marketing and Management. 2nd ed. NewJersey: Prentice-Hall.

MALHOTRA, N. K., F. M. UGALDO, J. AGARWAL and I.

B. BAALBAKI. 1994. International servicesmarketing: A comparative evaluation of thedimensions of service quality betweendeveloped and developing countries.International Marketing Review 11(2): 5-15.

MASTOR, K. A., P. JIN and M. COOPER. 2000. Malayculture and personality. American BehavioralScientist 44(1): 95-111.



MATTILA, A. S. 1999. The role of culture in the

service evaluation. Journal of Service Research1(3): 250-261.

NUNNAIXY, J. C 1978. Psychometric Theory. 2nd ed.

New York: McGraw-Hill.

OSTROM, A. and D. IACOBUCCI. 1995. Consumer trade-

offs and the evaluation of services. Journal ofMarketing 59: 17-28.

PRASURAMAN, A., V. A. ZEITHAML and L. L. BERRY.

1988. SERVQUAL: A multiple item scale formeasuring customer perceptions of servicequality. Journal of Retailing 64: 12-40.

ROTH, S. M. 1995. The effects of culture and socio-economics on the performance on global brandimage strategies. Journal of Marketing Research32 May: 163-175.

RUST, R. 1998. What is the domain of serviceresearch? Journal of Service Research 1(2): 107.

STAUUS, B. and P. MANG. 1999. Cultural shocks in

inter-cultural service encounters? Journal ofServices Marketing 13(4/5): 329-346.

SHOS~TACK, L. 1977. Breaking free from productmarketing. The Journal of Marketing: 76-85.

WINSTED, K. F. 1997. The service experience in two

cultures: A behavioral perspective. Journal ofRetailing 73(3): 337-360.

ZEITHAML, V. A., L. L. BFRRY and A. PARASURAMAN.

1993. The nature and determinants of customerexpectations. Journal of Academy of MarketingScience 21(1): 1-12.

(Received: 14 August 2001)



The Sociolinguistics of Banking: Language Use in EnhancingCapacities and Opportunities

AIN NADZIMAH ABDULLAH 8c ROSLI TALIFDepartment of English Language

Faculty of Modern Languages & CommunicationUniversiti Putra Malaysia

43400 UPM, Serdang, Selangor, Malaysia

Keywords: Bilingualism, commercial development, banking, executive-level employees,multicultural, multilingual

ABSTRAK

Penggunaan bahasa Malaysia (bahasa kebangsaan) dan bahasa Inggeris (BI) menawarkankeuntungan ekonomi kepada para pekerja sektor swasta di Malaysia. Para majikan menganggapkedwibahasaan BM-BI sebagai satu cara penting untuk perkembangan perdagangan kerana iniakan membuka pintu kepada pasaran dan menyediakan wawasan negara terhadap senarioantarabangsa (Rosli Talif dan Ain Nadzimah 1998). Bagaimanapun, masalah yang dihadapi olehkedwibahasaan BM-BI di Malaysia adalah kurangnya kata sepakat dari segi tujuan dan matlamatnya.Walaupun setelah mengambil kira keuntungan daripada pengetahuan kedwibahasaan BM-BI dankonteks di mana ia telah dimajukan, dipertahankan dan dikembangkan, ramai lagi yang masihragu tentang kesannya terhadap bahasa kebangsaan. Perdebatan lebih kepada sama ada memajukanBI sebagai alat kemajuan akan menenggelamkan peranan BM. Kajian ini bertujuan untukmeneliti keperluan dan penggunaan BM dan BI di kalangan pekerja eksekutif dalam industriperbankan. Industri ini dipilih sebagai gambaran perniagaan yang global, pelbagai budaya danpelbagai bahasa. Satu soal selidik lapangan dan temu duga yang dikendalikan meliputi BankNegara, bank tempatan dan antarabangsa yang terdapat di Kuala Lumpur. Soal selidik tersebutjuga akan menentukan bahasa yang digunakan untuk komunikasi sama ada lisan atau tulisandalam industri perbankan.

ABSTRACT

The use of BM (the national language) and English offers economic profitability to private sectoremployees in Malaysia. Employers see BM-English bilingualism as an important tool for commercialdevelopment since it can open doors to new markets and provide for the country's visibility onthe international scene (Rosli Talif and Ain Nadzimah 1998). However, a problem facing BM-English bilingualism in Malaysia is the lack of agreement as to its purpose and aims. Even afterconsidering the advantages associated with BM-English bilingualism and the contexts in which itis promoted, sustained and developed, many are concerned about its effects on the nationallanguage. The debate seems to be about whether promoting English as a tool of developmentwould confine BM to the backwoods. The purpose of this study is to investigate the necessity anduse of bahasa Malaysia (BM) and English among executive-level employees in the bankingindustry. The industry was selected as it reflects a business that is global, multicultural andmultilingual. A survey questionnaire and participant interviews were conducted involving theCentral Bank, local and international banks in Kuala Lumpur, Malaysia. Contexts of use forlanguages used in communication, oral and written, within the banking industry were identified.

INTRODUCTION been studied from two major theoreticalSociolinguistics is the study of the relationship orientations: (1) linguistic, such as differencesbetween language and society, that is between in the realisation of structural featureslinguistic and social facts. This relationship has (phonological, morphological, syntactic, and

Ain Nadzimah Abdullah 8c Rosli Talif

lexical) by speakers, and (2) pragmatic, functionor purpose of language for a particularcommunity. This second theoretical orientationexamines language and the social forces whichinfluence its use. It is of particular concern tothis study. Research in this area focussing onthis functional view has been undertaken inMalaysia. (Ahmad Mohd Yusof et al 1992; Asmah1992; Shanta 2000).

In studying language choice and use,Fishman (1971), poses the question of "whospeaks what language to whom and when." Thefirst part of the question (who?) is answered inthis study by limiting the subjects of the study toexecutives working in local and internationalbanks in Kuala Lumpur. The other two parts ofFishman's question are left open, that is, whenand to whom BM and English are spoken andwill be discussed below.

Speakers of languages are constantlycomparing and evaluating language in terms ofits usefulness for a given speech act. Even thougha language should never be considered "better"or "worse" than another, attempts to favour onelanguage over another are always present andmaybe reflected in the language choice ofspeakers. These language choice decisions arenever made in a vacuum. They are, instead,influenced consciously or unconsciously by anumber of factors among which some are socialand others economic.

As a matter of fact, Gumperz has argued in1974 that,

Choice of styles or language is seen as a strategyon the part of speakers trying, for instance, topresent themselves as individuals with particularsocially defined qualities, or, as another example,trying to convey a particular attitude or impressionconcerning a topic of conversation (Gumperz inGal 1979: 91).

When choosing one language over another,a speaker makes a statement about hisinterpretation of the situation or to turn theargument around, certain social factors (whoyou are talking to, the social context of the talk,the function and topic of the discussion) becomeimportant in accounting for language choice(Holmes 2001).

BM AND ENGLISH: A SMART PARTNERSHIPAND NOT A LINGUISTIC RIVALRY

The current situation in Malaysia shows thatbahasa Malaysia (BM) is the national and officiallanguage and English, the second most importantlanguage. There are policies to maintain andpromote allegiance to BM. In view of thecountry's political stability and its expandingeconomic opportunities, and in line withmodernisation, international recognition, andthe desire for progress, English is the logicalchoice for such a second language.

Obviously, English is the language of widercommunication and the need for the generalpopulation to be proficient in English isapparent. In spite of this, the government doesnot plan to accord official language status toEnglish and the use of English can only becomplementary to the national language. BM isused consistently at all levels of governmenttransactions.

As stated in the Constitution, Article 152(1) The national language shall be the Malay languageand shall be in such script as Parliament may by lawprovide:

Provided that: (a) no person shall beprohibited or prevented from using(otherwise than for official purposes), orfrom teaching or learning, any otherlanguage; and (b) nothing in this clauseshall prejudice the right of the FederalGovernment or of any State Government topreserve and sustain the use and study ofthe language of any other community in theFederation.

Today, societal bilingualism (multilingualismin reality) exists in Malaysia with a large portionof the population speaking both BM and English.It is a consequence of a specific language planto promote linguistic duality withoutundermining the status, role and allegiance tothe national and official language. The linguisticsituation in Malaysia encourages the use of BMfor intra-national needs and English forinternational needs. This co-existence of BMand English is an indication of a smartpartnership between the two languages, thusavoiding any manifestation of conflict betweenthem.


The Sociolinguistics of Banking: Language Use in Enhancing Capacities and Opportunities

Nevertheless, the presence of English in thecountry has been construed by national languageloyalists to be disruptive to the linguistic ecology.BM and English are traditionally seen as rivals,and the elevation of one is seen to mean thesuppression of the other. There is a negativeconnotation associated with English coupled witha fervent defence of the national language. Thebelief is that English has become the languageof many crucial domains in Malaysia, while thenational language, although retained, finds itselfin a precarious state. There was a time not toolong ago when "using the English language wasregarded unpatriotic and frowned upon. Timehas proven that the national language, BM, hasachieved the status it deserves." (The Star,November 8, 2000, p. 2).

The partnership between BM and English isoften misunderstood as it is believed toincorporate the symbolic struggle over culturalor national identity. Language is too often seenas a political force that can unite or threaten tosplit a country. English is the language ofinternational communication and the gatewayto world citizenship. Nationalistic sentiments tendto impede the use of English when it is presentedas a threat to the national language. The reasonsfor preserving a national language may on manyoccasions be equated with maintaining thenational identity of a people. This linguisticrivalry must not exist as we should not beconfined by nationalistic sentiments.

Another argument against the use of Englishupholds the view that English is a coloniallanguage. Such an argument is outmoded. Atthe end of colonisation, it is the peoplethemselves who became co-processors of thelanguage. The English language is actually beingshared by 700 million of the world's population(TESI^EJ 1994). In a global sense, English is nolonger a language that belongs to or can beidentified with a single race or country.

The following polemics concern theintrusiveness of English and the effects it mayhave on the wider use of BM. At times thewidespread use of English meets with resistanceas it is still considered to be the language of thecolonialists. The adoption of a foreign (in thiscase colonial) language is thought to possess thepower of re-orientating the people of the countrytowards the culture and thinking of their formercolonial masters. The country will thus becomean ardent follower of the western powers and be

unable to develop her own independent mind,minda bahasa, or national intellect.

Malaysians must realise that English is thelanguage of wider communication and is noenemy to nationalism. BM is secure in its statusand role as official and national language. Weare living in an era of information explosion.The need for the general population to beproficient in English is even more apparent nowas "three-quarters of the world's mail and four-fifths of its electronic information is in English"(Asiaweek June 16 1995).

With the country's present aspiration ofbecoming industrially developed through aknowledge-based workforce, it seems that thesmart partnership of BM and English wouldprovide the platform for Malaysia to take thequantum leap to international success. There isa necessity in Malaysia to master both languagesif not more. Having the capacity to use twolanguages is beneficial to any individual.

LANGUAGE AND BANKINGThe banking system in Malaysia was introducedby the British in the mid-19th century. Theindependence of Malaya in 1957 and theestablishment of Bank Negara (Central Bank)two years later did not change banking practicesintroduced earlier; hence, the dependence onthe English language remained in place.Moreover, Acts related to the licensing offinancial institutions do not stipulate languagerequirements for the industry. In fact, not evenBank Negara is obliged to use BM. Theconstitution does not legislate the use of BMwithin the financial system. The NationalLanguage Act 1963/67 made it only compulsoryfor the national language (BM) to be used atofficial functions. However, this does not includethe private sector, and especially not the bankingand financial sectors.

The Banking System

Bank Negara Malaysia is Malaysia's central bankand is responsible for supervising the bankingsystem. It also issues the Malaysian currency, actsas banker and financial adviser to thegovernment, administers foreign exchangecontrol regulations, and is lender of last resortto the banking system.

The banking system in Malaysia alsocomprises the commercial banks, merchant


Ain Nadzimah Abdullah & Rosli Talif

banks, and finance companies. They are themajor institutional sources of credit. Thirty-fivelicensed commercial banks operate through atotal of 1888 branches. Representative offices by36 foreign banks have also been established. Awide range of merchant banking services areprovided by 12 merchant banks with a networkof 22 branches, many of which have affiliationswith merchant banks established overseas. AnIslamic bank provides all the conventionalbanking services, based on the Islamic conceptof banking and credit. Twenty-three financecompanies operate through 960 branches thataccept retail deposits and provide finance forhire purchase and leasing transactions.

Today, the banking industry in Malaysiareflects a business that is global, multiculturaland multilingual. Being proficient in thedominant languages (BM and English) isindispensable in this sector. Industrialization andmodernisation bring with them the need for theEnglish language and an English-speakingpopulation more adapted to the technologicalenvironment that the process generates. A societymay find that only through the learning of anadditional language will the people have accessto social mobility via the power and resourcesthat industrialisation brings.

Possessing linguistic skills in the nationallanguage (BM) and English is now important inthe banking industry. This becomes increasinglytrue with the international market becomingintegrated into the local scene. These emergingmarkets offer a tremendous potential for businessexpansion. Therefore, the possibility ofcommunication in just one language does notsuffice. Employees in the banking industry needto measure up to increasingly demanding globalexpectations for better proficiency in English.

Rosli and Ain Nadzimah show in their 1998study that English is of instrumental value in theMalaysian private sector workplace. The studyalso showed that opportunities for employmentand commerce tend to be open only to thosewho are proficient in the dominant languages.Ali Abul Hassan (2000), the former Governor ofBank Negara, states that human capital willcertainly be a key driver and critical determinantto the success of an organization.

The banking industry has expanded beyondnational boundaries and is now even takingplace over cyber space. Markets no longer needto be physical but could exist in virtual networks.

The Star, a national daily, states 'it goes withoutsaying that English is the language of informationand technology. Just flip through an internationalpublication, switch on the television, scour theairwaves, click on to the internet or open acomputer manual and you will face an Englishworld. Statistics from local IT players show thatonly 1% of websites use BM' (The Star, November8, 2000, p.2).

The strategic alliance between BM andEnglish should be mutually beneficial and highlyrelevant in the new economic order of today.This would be in line with meeting globalbanking standards, which impose increasedrequirements on banks. Greater importance isnow accorded to non-financial as well as financialmeasures of performance. As a result, proficiencyin the national language (BM) and English wouldbe an integral part of evaluating employeeperformance in the banking sector.

Therefore, it is pertinent that employees inthe banking industry communicate clearly andeffectively in both BM and English. The country'sfuture lies in a knowledge-based economy whereknowledge is an important component indetermining performance and management ofbusinesses. The need for the effective use ofappropriate language or languages is vital. Withthe technological revolution that is taking place,the need to communicate and function in thenew economy is becoming a major factor forgenerating growth and transforming businesses.Good, accurate, written and spoken English andBM are no longer a side issue. It is widelyaccepted as essential to the image of aninstitution.

The Study

This study investigates the necessity and use ofBM and English among executive-level employeesin the banking industry. It attempts to find out(1) with whom banking executives use BM andEnglish; and (2) what the location, setting,communicative intent, and mode are. The studyalso identified languages used for communication,oral and written within the banking industry.This industry was selected as it reflects a businessoperation that is global in nature. The bilingualtrend observed here will determine long-termcompetitiveness for those linked to the bankingsector.

A survey questionnaire was used to gatherdata from executives working in the central

112 Pertanika J. Soc. Sci. 8c Hum. Vol. 10 No. 2 2002


bank, local and international commercial banksin Kuala Lumpur, Malaysia. The questionnairecomprised three parts. Part one sought informationon the respondents' biodata (Table 1), workplaceand work experience (Table 2), and educationalbackground (Table 3). Part two aimed at gettingrespondents* perceptions on the necessity anduse of BM and English within the bankingindustry, while part three sought to identifyrespondents' code choices for oral and writtencommunication in a variety of situations withinthe banking industry. The data were reported interms of percentages. Interview sessions wereheld at the banks.

Descriptive statistics are presented in thisstudy. The figures shown are reported inpercentage because this study is concerned withthe tendency of language use rather thanabsolute frequencies of use. To describe patternsof code choice, the researchers have identified anumber of typical day-to-day interactionsinvolving bank executives. The term, domains oflanguage use, follows Fishman's criteria and seeksto show typical interactions between typicalparticipants in typical settings. Though a verygeneral concept, the notion of domain has beenfound to be useful in capturing broadgeneralisations of a speech community. It allowsone to summarise the norms of language use fora large group of people.

TABLE 1Biodata

TABLE 3Educational background

Respondents (n=37)

Respondents

1. Race

2. Gender

(n=37)

MalayChineseIndian

MaleFemale

TABLE 2Workplace and work experience

Respondents

Workplace

WorkExperience

(n-37)

Central BankCommercial Bank

0-5 yrs6-10 yrs11-15 yrs>15 years

19117

2314

829

318142

Education1. Received tertiary education 332. Did not receive education 43. Educated in the country 194. Educated abroad 14

DISCUSSION OF FINDINGSRespondents

The study involved 37 middle management tosenior management personnel from sevendifferent banking institutions. There were 11Chinese, seven Indian and 19 Malay respondents.Of these, eight were working with Bank NegaraMalaysia (Central Bank). The respondents' agesranged between 30 and 50. A large majority ofthe respondents (62%) were male. A majority ofthem (86%) have worked in banks for 6 to 15years. All but four of the respondents receivedtertiary education. A majority (57%) receivedtheir tertiary education in the country and theothers received it abroad.

A. Perception Towards the Necessity and Use ofBM and English in BankingProficiency in both English and BM is regardedas very important in the banking industry.English is used in six out of seven bankinginstitutions surveyed while BM is used in all. Therespondents noted that it would be difficult towork in a bank if one only knows BM. Conversely,knowing English would be most helpful whenworking in a bank. Respondents also stated thatif one had knowledge of only English, it wouldstill be possible to work in certain departmentsof the bank such as the legal department,corporate banking, corporate credit and loans,research, investment and training, as well asinsurance.

Thirty-one respondents agreed that Englishshould play a more important role in the bankingindustry. They also noted that English is used inthe following situations: at meetings, for bothinternal and external correspondence,communication with clients and amongemployees. However, they also indicated thatBM would be used when dealing with thegovernment as it is the national and officiallanguage of the country.

Pertanika J. Soc. Sci. & Hum. Vol. 10 No. 2 2002 113


Language competency or the lack of it is abarrier to employment and promotion withinthe banking system. Eighty-six percent of therespondents noted that it was necessary for anemployee from the middle management tomanagement level to be proficient in English inorder to be considered well qualified and eligiblefor further job promotion.

Proficiency in English prepares one to facethe increasing demand and challenges of today'sworld. Although not all jobs require fluency inboth BM and English, the ability is a definiteadvantage and often becomes the decisive factorfor employment. Effective use of English seemsto be necessary for employees to be marketableand it is essential to those who want to competeand succeed within the banking system.

In recognising that the future lies in aknowledge-based economy where knowledge isan important component in determiningperformance, it is important to put themanagement of language knowledge within thebanking industry into its right place and cultivateit. This management of language knowledgewould enhance the competitive position of thebank and allow for competitive advantage. It istherefore important for the banking industry toensure that their human resources are managedeffectively and optimised for future growth.

(i) English Proficiency

Proficiency in English is important in therespondents' work place for the purposes ofcommunication and business transactions. Whenasked about the percentage of staff that is fluentin English in their organization, the lowestpercentage reported is 60%, while three banksmentioned 90% of their staff being fluent in thelanguage. Interestingly, respondents from theCentral Bank noted a higher percentage of staffthat is fluent in English. Most respondentsindicated that all their departments are involvedin using English in their daily routines especiallythose departments that have corporate andinternational dealings. The banking industry,therefore, needs to encourage employees to usethe national language (BM) as well as English.

(ii) Languages Other Than BM and EnglishWhen asked whether knowledge of a languageother than BM and English, that is, a thirdlanguage, is important in the banking industry,only four respondents said that it is not necessary.

The remaining 33 respondents mentioned thata third language is important for the purposesof marketing their services, and communicatingwith prospective clients. The most popular thirdlanguage stated by the respondents is Mandarin,followed by Tamil, Japanese and French.

There seems to be the awareness thatknowledge of a third language could confer thebenefit of broadening one's scope beyond thelimits of one's own culture and country. It ishard to ignore the many benefits this extraknowledge can bring. One must realise thatadded knowledge is an added gift as it allowsone to possess a better means of communicationbecause it suggests an increase in individualpotential. It also enables one to have manyperspectives on life, foster open-mindednesswhile eliminating cultural ignorance.

(iii) Employment

All respondents said that it is important foremployees to be proficient in both, BM andEnglish. Thirty-four respondents stated that theywould not employ someone who is not fluent inEnglish while only three said they would. It isinteresting to note that when it comes toemploying prospective candidates who have thesame qualifications but differ in their languageproficiency, all respondents stated that thecandidates who are more proficient in Englishstand a better chance of being employed in thebanking industry. Twenty-seven respondentsmentioned that they would not considerprospective candidates who are more proficientin BM than in English. Eight respondents, inturn, stated that they would consider suchcandidates while two of them said that it dependson the job requirement.

Being able to use both BM and Englishshows that one has learned two languages. Thisdemonstrates discipline, spirit of initiative and apredisposition to learn new things. Thus, usingmore than one language makes the employeemore flexible and more valuable. Having thecapacity to adapt to two languages is thereforeprofitable to an individual.

Thirty-two respondents indicated that it wasdifficult to get prospective employees who areproficient in English, while five others said thatit was not difficult to find such candidates. Thefive respondents were all involved in thecorporate banking departments of theirrespective banks. It could be that this department



had higher expectations and more stringentlanguage requirements and therefore found itharder to obtain employees who could meetthose requirements.

B. Code Choices as Reflected in ActualLinguistic PracticesAreas of CommunicationThe researchers identified four different areasof communication in the banking situation.They are: (i) meetings, (ii) correspondence,(iii) communication between employees, and (iv)communication with clients.

English seems to be the code choice atmeetings, for correspondence (both internal andexternal) and with clients as reported by 35respondents. However, BM was used more oftenin conversation between employees as reportedby 32 respondents. It was also stated that asuperior would commonly use English whenspeaking to a subordinate officer while asubordinate officer would most likely use BMwhen speaking to a superior unless he wasaddressed in English first.

(i) MeetingsWhen asked to state which language was mostcommonly used during meetings within the bank,the respondents stated that the chairperson usesEnglish to introduce the meeting, present anidea, advance an argument, pose a question,report a problem, share progress, ask for andmake clarifications as well as conclude a meeting.Some of the common issues discussed in Englishare those related to policy, central banking affairs,electronic banking and foreign exchange. It wasalso reported that at management and middlemanagement meetings, both the chairpersonand participants of the meeting use Englishpredominandy. This was true in six of the sevenbanks surveyed. Instances of code switching werereported when the matters discussed were non-technical in nature. Whenever meetings wereheld with government agencies, the dominantlanguage used was BM.

(ii) CorrespondenceIn all the seven banks surveyed, respondentsstated that English was used for all corres-pondence between departments within the bank.This included correspondence with the HeadOffice, between high-level management, and withclients. BM, in turn, is used in correspondence

only with government agencies. As for the useof BM, Ali Abul Hassan (2000) reiterates theneed for banking institutions to promoteincreased use of BM in the financial sectorhereby catering to the needs of the largest ethnicgroup of population. A wider use of BM inbusiness correspondence would make bankinginstitutions friendlier and more approachable.At the very least, simple contracts includinghousing and car loan contracts as well asapplication forms for banking products andservices to the general public should alwaysinclude a BM version.

(iii) Communication Between Employees

Thirty-four respondents stated that English wasused amongst employees when discussingbanking-related matters. BM was used whendiscussing personal matters and when engagingin small talk. BM was also used for salutationpurposes.

The government has successfully imple-mented the use of BM. As a result, BM's primacyis well established. It is the medium of instructionfor all levels of education and used forcommunication, especially in the governmentsector. As reported by Ain and Chan (2000), BMis the dominant language used for inter-racialcommunication in Malaysia, especially amongpost-independence Malaysians.

(iv) Communication with Clients

The banking industry is a service-based industry.It is therefore important that frontline employeesdevelop the necessary language proficienciestoward achieving and depicting a professional,competent, honest, and responsible workforce.This helps to uphold the image of theorganisation.

In all except one of the banks surveyed,respondents reported that BM and not Englishis most often used at reception counters of thebanks. However, when meeting clients over loan,insurance and credit card matters, English isused. In the rapidly evolving banking industry,responsiveness to client needs is a crucial test ofthe industry's survival. With increasingsophistication of needs and higher expectationfrom clients, there is urgency for banks to set upcompetitive strategies leading to a highly skilledknowledgeable and language proficientworkforce.



CONCLUSION

Language conflicts generally incorporatesymbolic struggles over religious, ethnic, cultural,or national identity. The debate in Malaysiaseems to be about whether promoting Englishas a tool of development would confine BM tothe backwoods (Far Eastern Economic Review,October 10, 1992). There is also the belief thatpromoting English may result in the loss of theMalay identity as stated by Professor IsmailHussein, "Language is the soul of the race. Weshould master English but let it not master us"(Malaysian Business 1992).

There is a lack of agreement as to thepurpose and aims of using both BM and English.Even after considering the advantages associatedwith these two languages and the contexts inwhich they are promoted, sustained, anddeveloped, many are concerned about the effectsEnglish could have on the national language.Viewpoints on these issues are linked to generalpolitical philosophy and ideals of what it meansto be a Malaysian.

The Prime Minister of Malaysia has arguedthat BM has a role to play and the governmentis not de-emphasizing that role. No languageshould ever be seen as having the ability todebase another. In that respect, he argued that,"We should not be fanatical about it (BM)because that would make it difficult for us tobenefit our race in a competitive world. Whetherwe like it or not, English is an internationallanguage" (Malaysian Business 1992). Being thecommon international language, it is Englishthat allows us to overcome the influence oflinguistic separatism.

In the banking sector, the way business isconducted is rapidly changing. Thus, in such adynamic and ever changing environment, banksare under enormous pressure to measure up toincreasing global expectations especially in thefield of communication; hence, die knowledgeof both languages, BM and English, is animportant component of the growth equationfor the banking industry in this new economicorder.

Those in the banking sector face a new kindof challenge, that of being able to communicatein both the national language as well as English.This study shows that proficiency in these twolanguages contributes towards enhancing thecapacities and opportunities within the bankingsector. Using both these languages empowers

the banking sector to develop global alliancesand enhance market possibilities. The simpletruth lies in the fact that those who can functionin the country's predominant languages findthemselves at a considerable advantage over thosewho cannot.

REFERENCES

AHMAD MOHD YUSOF et al. 1992. Laporan Kaji Selidik

Penggunaan Bahasa Melayu dalam BidangKewangan dan Bank. Kuala Lumpur: DewanBahasa dan Pustaka.

AIN NADZIMAH ABDULLAH and S. H. CHAN. 2000.

Insights into Language Functioning: Exploringthe Connectedness of Its Multidimensionality.Paper presented at the Malaysia InternationalConference on English Language Teaching, 15-17

, May, Riviera Bay Resort, Melaka, Malaysia.

ALI ABUL HASSAN BIN SUIAIMAN. 2000. Towards aMore Efficient, Effective and Stable FinancialSystem. Keynote address 14th Conference withFinancial Institutions, 12 April. Bank NegaraMalaysia.

ASIAWEEK. 1995. The Language is No Enemy ofNationalism. 16 June.

ASMAH HAJI OMAR. 1992. The Linguistic Scenery in

Malaysia. Kuala Lumpur: Dewan Bahasa danPustaka.

Federal Constitution (as at 25th July 2000). KualaLumpur: International Law Book Series.

FISHMAN,J. A. 1972. The Sociology of Language. Rowley,MA.: Newbury House Publishers.

GAL, S. 1979. Language Shift: Social Determinants ofLinguistic Change in Bilingual Austria. NewYork: Academic Press.

HOLMES, J. 2001. An Introduction to Sociolinguistics.Essex: Pearson Education.

ROSLI TALIF and AIN NADZIMAH. 1998. Bilingualism is

Profitable. Paper presented at MalaysiaInternational Conference on English LanguageTeaching, 18-20 May. Shangri-La Rasa SayangResort, Penang, Malaysia.

THE STAR. 2000. Crucial to Master English Language.p.2. November 8.

VENUGOPAL, S. 2000. An Introduction to Sociolinguistics.Essex: Pearson Education.

TESl^EJ Forum (F-l)*. 1994. Whose English is it?April 1(1). (*Electronic journal from theinternet).

(Received: 11 September 2001)



Faktor-faktor Mempengaruhi Agihan Pendapatan di Malaysia 1970-2000

RAHMAH ISMAIL & POO BEE TINJabatan Ekonomi Pembangunan

Fakulti EkonomiUniversiti Kebangsaan Malaysia

43600 UKM Bangi, Selangor, Malaysia

Kata kunci: Agihan pendapatan, pekali Gini, Keluaran Dalam Negara Kasar, bayaran pindahan,keterbukaan ekonomi

ABSTRAK

Pembangunan ekonomi Malaysia telah menyaksikan trend turun naik dalam ketaksetaraan agihanpendapatan di samping penurunan yang berterusan dalam kadar kemiskinan. Antara tahun1957/58 sehinggalah 1999, berlaku peningkatan dan penurunan dalam ketaksetaraan agihanpendapatan. Pekali Gini telah mengalami peningkatan antara tahun 1957/58 hingga 1976, iaitumenunjukkan agihan pendapatan menjadi semakin tidak seimbang. Tetapi antara tahun 1976hingga 1990, pekali Gini mengalami penurunan yang menggambarkan agihan pendapatanmenjadi lebih seimbang. Malangnya antara tahun 1990 hingga 1997, pekali Gini sekali lagimeningkat, tetapi menurun sedikit pada tahun 1999. Objektif artikel ini ialah untuk menganalisisfaktor-faktor yang mempengaruhi agihan pendapatan di Malaysia melalui kaedah persamaanregresi dengan menggunakan data-data dari tahun 1970 hingga 2000. Di samping melihatkepada keseluruhan tempoh, analisis juga dilakukan dalam dua tempoh yang berbeza iaitu 1970-1984 dan 1985-2000. Keputusan kajian ini menunjukkan pengaruh faktor-faktor yang dipilihterhadap agihan pendapatan adalah berbeza antara dua tempoh. Antara tahun 1970-1984, kadarpertumbuhan Keluaran Dalam Negara Kasar (KDNK) per kapita, peratus guna tenaga dalamsektor industri, keterbukaan ekonomi, bayaran pindahan dan pelaburan langsung asing memberipengaruh yang positif terhadap pekali Gini. Ini bererti peningkatan dalam pemboleh ubah-pemboleh ubah ini meningkatkan lagi ketidakseimbangan agihan pendapatan. Dalam tempoh1985-2000, pelaburan langsung asing masih mempunyai pengaruh positif terhadap pekali Gini disamping kadar celik huruf, kadar pengangguran dan buruh asing. Sementara kadar pertumbuhanKDNK per kapita, guna tenaga sektor industri dan bayaran pindahan memberi kesan negatif yangmenggambarkan kenaikan dalam pemboleh ubah-pemboleh ubah ini akan menyebabkan agihanpendapatan menjadi lebih seimbang.

ABSTRACTIncome inequality in Malaysia has followed unstable trends despite a continually decreased inincidence of poverty. Between 1957/58 and 1976, the Gini coefficient had increased, whichreflected an increased in inequality. But between 1976 and 1990 the Gini coefficient descreased,which implied that income inequality became more equal. Unfortunately between 1990 and 1997the Gini coefficient again increased, but decreased slightly in 1999. The objective of this articleis to analyse determinants of income inequality in Malaysia using regression method. The datacovered for the analysis were between 1970 and 2000. Between 1970 and 1984 it was found thatthe Gross Domestic Products growth (GDP), percentage employment in the manufacturingsector, openness of the economy, transfer payment and foreign direct investment statistically havea significant positive relationship with the Gini coefficient. This implies that an increase in thesevariables will increase income inequality. Between 1985 and 2000, foreign direct investment,unemployment rate, literacy rate and foreign labour show a positive significant relationship withthe Gini coefficient, while GDP growth, manufacturing employment and transfer paymentnegatively affect income inequality.

Rahmah Ismail 8c Poo Bee Tin

PENGENALAN

Pembangunan ekonomi Malaysia telah memberiimpak yang besar terhadap agihan pendapatandan kemiskinan. Walaupun kadar kemiskinantelah mengalami penurunan, agihan pendapatanmasih menunjukkan trend yang tidak stabil. Datarasmi menunjukkan kadar kemiskinan di Malaysiatelah mengalami penurunan secara berterusantetapi agihan pendapatan menunjukkan trendyang turun naik. Berdasarkan nilai pekali Giniyang berada antara 0 hingga 1, semakin rendahnilai ini, semakin setara agihan pendapatan.Antara tahun 1957/58 sehingga tahun 1976agihan pendapatan di Malaysia telah menjadisemakin tidak seimbang, tetapi menjadi lebihseimbang antara tahun 1976 hingga 1990. Walaubagaimanapun dalam tahun-tahun 1990-anketidakseimbangan agihan pendapatan mulaimeningkat semula. Ini ditunjukkan oleh pekaliGini yang meningkat daripada 0.45 pada 1990kepada 0.46 pada 1993 dan 0.47 pada tahun1997. Namun demikian nilai pekali Ginimenurun sedikit pada tahun 1999 kepada 0.44.Apa yang lebih membimbangkan ialah pekaliGini yang dicapai sepanjang tempoh, walaupunmengalami penurunan, tidak pernah turunserendah nilainya pada permulaan pembangunantahun 1957/1958.

Jadual 1 menunjukkan trend agihanpendapatan dari tahun 1957/1958 hingga 1999.Umumnya trend agihan pendapatan di Malaysiaboleh dianalisis berdasarkan tiga fasa, iaitu 1957/1958 hingga 1976, selepas tahun 1976 hingga1990 dan tahun-tahun selepas 1990. Dalamtempoh 1957/1958 hingga 1976, berlaku

peningkatan yang berterusan dalam peratusbahagian pendapatan isi rumah 20% teratas.Sebaliknya bahagian pendapatan bagi 40%pertengahan dan 40% termiskin mengalamipenurunan. Implikasinya, pekali Gini telahmeningkat dari 0.41 pada tahun 1957/58 kepadamasing-masing 0.50 dan 0.53 pada tahun 1970dan 1976 yang menunjukkan agihan pendapatanmenjadi lebih tidak seimbang. Pelaksanaan DasarEkonomi Baru (DEB) pada 1971 masih tidakmampu mengatasi masalah ketidaksetaraanagihan pendapatan, malahan semakin melebarkerana pada tempoh tersebut prosespembangunan masih dalam era penyesuaian.

Dalam tempoh fasa kedua iaitu dari 1979hingga 1990 menampakkan penurunan dalamnilai pekali Gini dari 0.51 pada 1979 kepada0.45 pada 1990. Agihan pendapatan yangsemakin setara ini juga ditunjukkan olehpenurunan bahagian pendapatan penduduk 20%teratas, iaitu menurun dari 55.8% kepada 50.3%pada tempoh yang sama. Sementara bagikumpulan 40% pertengahan dan 40% termiskin,bahagian pendapatan mereka meningkat.Kesannya, pendapatan purata rakyat terusmeningkat dari RM693 sebulan pada 1979kepada RM1167 sebulan pada 1990. Secarakeseluruhannya, agihan pendapatan semakinpulih pada fasa ini. Fasa ketiga, iaitu selepas1990 menunjukkan ketidaksetaraan agihanpendapatan meningkat semula. Keadaan inidigambarkan oleh pekali Gini yang meningkatdaripada 0.45 pada 1990 kepada 0.46 pada 1993dan 0.47 pada 1997, tetapi menurun semulakepada 0.44 pada tahun 1999.

JADUAL 1Agihan pendapatan isi rumah Malaysia 1957-1999

Peratus 1957/ 1967/ 1970 1976 1979 1984 1987 1990 1993 1995 1997 1999Isi rumah 58 68

20% Teratas 48.6 51.3 56.1 57.7 55.8 53.2 51.2 50.3Pertengahan 35.5 34.4 32.5 31.2

40% Terbawah 15.9 14.3 11.6 11.1PendapatanPurata 215 240 267 514PendapatanPenengah 156 154 167 388

32.311.9

693

436

34.012.8

1095

723

35.013.8

1074

738

35.214.5

1167

--

-

2008

1346

-_

-

2607

1682

50.535.514.0

-

2472Pekali Gini 0.41 0.44 0.50 0.53 0.51 0.48 0.46 0.45 0.46 0.46 0.47 0.443

Nota: 1957/58-1987: Semenanjung Malaysia.Sumber. Snodgrass, 1980

Anand, 1983.Malaysia, 1981, 1984, 1989, 1991(b), 1993, 1996, 1999, 2001.



Terdapat banyak perbincangan yangmengaitkan faktor kenapa fenomena ini berlaku.Faktor yang sering diutarakan ialah dasarkerajaan, transformasi ekonomi, globalisasi danliberalisasi, kemasukan buruh asing, agihanpencapaian pendidikan dan sebagainya.Kebanyakan kajian sebelum ini mengutarakanatau menjawab isu di atas hanya melaluipemerhatian kasar. Belum ada satu kajian yangrnenganalisis faktor-faktor yang mempengaruhiagihan pendapatan di Malaysia melaluipendekatan ekonometrik. Artikel ini bertujuanmenganalisis faktor-faktor tersebut antara tahun1970 hingga 2000 dengan menggunakan kaedahpersamaan regresi. Penulisan artikel inidibahagikan kepada 6 bahagian. Bahagianseterusnya meliputi kajian lepas, kerangka teori,metodologi dan hipotesis, keputusan regresi,rumusan dan implikasi dasar serta kesimpulan.

Sorotan Kajian Lepas

Kebanyakan hasil kajian empirikal yangmenggunakan data keratan lintang data pelbagainegara meyokong hipotesis Kuznets yangmenghujahkan bahawa perhubungan antarapertumbuhan ekonomi dan agihan pendapatanadalah berbentuk U-terbalik (Paukert 1973;Ahluwalial976; Papanek dan Kyn 1986; Campanodan Salvatore 1988; Milanovic 1995; Jha 1996).Walau bagaimanapun, menurut Oshima (1994),bukti tentang kewujudan bentuk keluk U-terbalikKuznets di negara-negara sedang membangunadalah tidak jelas. Jika dilihat dalam tempohjangka masa panjang, beliau menyimpulkanhipotesis Kuznets tidak wujud di negara-negaraAsia pada hari ini. Sementara kajian yangdijalankan oleh Hassan dan Rahman (1997)menunjukkan hubungan antara pertumbuhanekonomi dengan agihan pendapatan bagi kesMalaysia (1957-1990) berbentuk U-terbaliksebagaimana yang dihipotesiskan oleh Kuznets.Walau bagaimanapun, selepas 1990 agihanpendapatan menjadi semakin tidak setara apabilaberlaku pertumbuhan ekonomi yang semakinpesat.

Agihan pendapatan yang seimbang bagi satugenerasi dapat menjamin kesetaraan agihanpendapatan generasi berikutnya melaluipengumpulan modal manusia (Chiu 1998). Jadiagihan pendapatan yang lebih seimbang bolehmembawa kepada penjanaan modal manusiayang lebih tinggi dan seterusnya pertumbuhanekonomi yang lebih tinggi (Romer 1986; Alesina

dan Podrik 1994; Persson dan Tabellini 1994;Perotti 1996). Namun demikian, pencapaianpendidikan dan keupayaan memperolehpendidikan adalah berbeza antara individu.Perbezaan inilah yang dikatakan meluaskanjurang perbezaan pendapatan. Kebanyakankajian telah menggunakan kadar celik hurufsebagai proksi kepada pencapaian modal manusiadalam melihat hubungan antara pendidikandengan agihan pendapatan (Rubinson 1976;Weede dan Tiefenbach 1981; Chase-Dunn 1985;Bollen dan Jackman 1985). Hasil kajian lepastelah membuktikan ketidakseimbangan dalamagihan pencapaian pendidikan memberi impakpositif dan signifikan terhadap ketidakseimbanganagihan pendapatan (Hammermesh dan Rees1984; Caniglia 1988).

Pada peringkat awal proses pembangunansesebuah negara akan menyaksikan berlakunyajurang perbezaan yang luas antara sektor. DiMalaysia, penekanan terhadap sektorperindustrian, terutamanya dalam dekad-dekad1970-an dan 1980-an, telah menyebabkan sektorini lebih terkehadapan dalam segala aspekdibandingkan dengan sektor-sektor lain.Produktiviti yang lebih tinggi dalam sektor inimenyebabkan pekerja mereka juga dibayar lebihtinggi. Maka perbezaan upah antara sektormenjadi lebih ketara dan ini selanjutnyamemburukkan jurang agihan pendapatan diMalaysia (Ishak Shari et ai 1999). Malah dalamsektor industri itu sendiri berlaku perbezaanpendapatan buruh antara kemahiran (Rahmah2000). Proses perindustrian juga telahmeningkatkan peratus guna tenaga sektor inidan ini selanjutnya boleh memberi kesan negatifterhadap keseimbangan agihan pendapatan.

Proses industrialisasi dan globalisasi yangpesat juga telah membawa peningkatankebergantungan kepada perdagangan bebas danpengaliran modal secara bebas. Wes (1996)mengatakan peningkatan pesat dalam pengaliranmodal juga merupakan salah satu faktor utamapenggerak globalisasi. Pengaliran modal bolehdibahagikan kepada pelaburan langsung asing,pelaburan portfolio dan pengaliran kewangan.Pengaliran modal merupakan faktor yangmempunyai pengaruh yang kuat terhadapintegrasi ekonomi secara global. Globalisasi pulamempunyai kesan ke atas agihan pendapatan,maka secara tidak langsung pengaliran modaljuga merupakan salah satu faktor yang akanmemberi kesan kepada agihan pendapatan.


Rahmah Ismail & Poo Bee Tin

Kebanyakan hasil kajian empirikal menunjukkanterdapat hubungan positif yang signifikan antarapekali Gini dengan pemboleh ubah pelaburanlangsung asing (Weede dan Tiefenbach 1981;Tsai Pan Long 1995; Beer 1999). Aliran masukpelaburan langsung asing dikatakan akanmemburukkan lagi agihan pendapatanterutamanya bagi negara yang kurangmembangun, walaupun ia membawa beberapakesan positif seperti meningkatkan pertumbuhanekonomi, menyediakan peluang pekerjaan,pemindahan teknologi dan sebagainya.

Ketidakseimbangan agihan pendapatanboleh juga dikaitkan dengan keterbukaanekonomi. Lazimnya, pemboleh ubah perdaganganantarabangsa iaitu jumlah eksport dan importdigunakan sebagai ukuran kepada keterbukaanekonomi( Rubinson 1976; Tsai Pan Long 1995;Carol 1998). Kajian Patrik(1998) menunjukkantujuh peratus peningkatan ketidakseimbanganagihan pendapatan adalah disebabkanperdagangan. Oleh sebab perdagangan bebasakan menguntungkan sistem ekonomi secarakeseluruhan, maka beliau mencadangkankeuntungan daripada perdagangan bebas harusdiagihkan kepada golongan yang rugi melaluisistem cukai dan sistem kebajikan.

Perbelanjaan kerajaan adalah antara faktorpenting yang mempengaruhi agihan pendapatan(Snodgrass 1974, 1980; Meerman 1975).Snodgrass (1974) telah mengelaskan perbelanjaankerajaan kepada empat kategori, iaitu pembelianbarang-barang dan perkhidmatan yangkemudiannya dikelaskan kepada penggunaanawam dan pelaburan awam, bayaran pindahan,bayaran akaun hutang dan pembelian aset. Hasilkajiannya menunjukkan wujudnya hubunganpositif antara perbelanjaan awam danketidakseimbangan agihan pendapatan. Walaubagaimanapun, hasil kajian Caniglia (1988) diAmerika Syarikat menunjukkan bayaranpindahan mempunyai hubungan negatif denganpekali Gini, iaitu agihan pendapatan menjadilebih seimbang dengan meningkatnya pendapatangolongan yang menerima bayaran pindahantersebut. Sebenarnya, kesan perbelanjaankerajaan terhadap pendapatan bergantungkepada siapa yang mendapat manfaatperbelanjaan tersebut, contohnya kemudahankesihatan, pendidikan dan kebajikan. Sekiranyaia lebih memanfaatkan golongan kaya danpertengahan berbanding golongan miskin,

agihan pendapatan akan menjadi lebih tidakseimbang (Xu dan Zou 2000).

Kitaran perniagaan juga dikatakanmerupakan antara faktor yang penting dalammempengaruhi agihan pendapatan. Duapemboleh ubah makroekonomi, iaitu kadarpengangguran dan kadar inflasi biasanyadigunakan sebagai proksi kepada kitaranperniagaan dalam menguji hubungan antarakitaran perniagaan dengan agihan pendapatan(Blinder dan Esaki 1978; Caniglia 1988). Hasilkajian Caniglia (1988) menunjukkanketidakseimbangan agihan pendapatan akanmeningkat jika kadar pengangguran adalahtinggi. Sementara pemboleh ubah kadar inflasipula mempunyai hubungan negatif dengankoefisien Gini.

Perkembangan ekonomi yang pesat danmasalah kekurangan tenaga buruh menyebabkankebanjiran buruh asing di negara ini. Kajianmenunjukkan kehadiran buruh asing,terutamanya buruh tidak mahir, akanmenyebabkan keluaran marginal buruh padakategori ini semakin menurun. Ini seterusnyaakan menyebabkan upah yang diterima olehgolongan ini juga akan menurun. Kebanjiranburuh asing juga telah menyekat kenaikan upahpekerja tempatan yang tidak mahir keranamereka sanggup menerima upah rendah akibatdesakan hidup yang teruk di negara asal mereka.Dalam jangka masa panjang, jurang perbezaanpendapatan antara buruh mahir dengan buruhtidak mahir semakin membesar. Keadaanini seterusnya akan menyebabkan ketidakseimbangan agihan pendapatan semakinmeningkat (Winegarden dan Lay Boon Khor1993).

KERANGKA TEORI, METODOLOGIDAN HIPOTESIS

Ahli-ahli sains sosial termasuklah ahli-ahliekonomi, telah lama mengutarakan perhubunganantara agihan pendapatan dan pembangunan,Marx (1952), seorang ahli ekonomi sosialismengkritik sistem ekonomi kapitalis yang hanyamenguntungkan golongan pemilik modal danmerugikan golongan majoriti buruh dalam prosespembangunan yang berorientasikan pertumbuhanpesat semata-mata.

Kajian terhadap agihan pendapatan padatahap-tahap pembangunan yang berlainandimulakan selepas Perang Dunia Kedua oleh



Kuznets(1955). Ukuran pertumbuhan ekonomiyang beliau gunakan adalah tahap atau kadarpertumbuhan Keluaran Negara Kasar (KNK) atauKeluaran Dalam Negara Kasar (KDNK). Beliaucuba mencari jawapan kepada persolan-persoalanseperti adakah ketidaksetaraan agihan pendapatanakan meningkat atau menurun dengan berlakunyapertumbuhan ekonomi? Apakah faktor-faktoryang mempengaruhi corak dan arah aliranagihan pendapatan? Kuznets kemudiannya telahmemperkenalkan Hipotesis U-terbalik yangmenjelaskan bahawa ketidaksetaraan agihanpendapatan akan bertambah buruk pada tahapawal proses pertumbuhan. Keadaan ini terjadikerana majoriti penduduk adalah miskinsementara minoriti penduduk sahaja berpeluangmenambahkan kekayaan mereka. Apabila prosespertumbuhan telah berlaku pada tingkatekonomi yang lebih tinggi, lebih ramai pendudukberpeluang meninggalkan sektor luar bandardan melibatkan diri dalam sektor bandar ataumoden. Pendapatan golongan ini akanbertambah dan menyumbang kepada corakagihan pendapatan yang lebih seimbang.

Berlandaskan kepada kerangka teori ini dankajian-kajian lepas dalam bidang ini, beberapapemboleh ubah selain daripada indikatorpertumbuhan ekonomi telah dimasukkan kedalam persamaan regresi. Pemboleh ubah-pemboleh ubah tersebut adalah kadar celikhuruf, peratus guna tenaga sektor industri, nisbaheksport-import kepada KDNK, nisbah bayaranpindahan kepada KDNK, nisbah pelaburanlangsung asing kepada KDNK, nisbahpenganggur kepada jumlah tenaga kerja dannisbah buruh asing kepada jumlah guna tenaga.Kesemua pemboleh ubah ini dirasakan bolehmempengaruhi agihan pendapatan seperti yangakan dinyatakan dalam bahagian hipotesis. Bagimenjawab objektif kajian ini, dua persamaanregresi dibentuk seperti di bawah:

PGINI = pQ + ^RKDNKNP + p2UT +^PGUNATEI + pJXMKDNK +05RBPINKDNK + ^RFDIKDNK +

}i0 ( 1 )

(2)

Dengan,PGINIRKDNKDNKP

LITPGUNATEI

JXMKDNK

RBPINKDNK

RFDIKDNK

RPANGGUR

RBAGUNAT

= pekali Gini= kadar pertumbuhan KDNK

nominal per kapitaa kadar celik huruf= peratus guna tenaga sektor

pembuatan daripada jumlahguna tenaga

= nisbah jumlah eksportdan import kepada KDNKnominal

= nisbah bayaran pindahankepada KDNK nominal

= nisbah pelaburan langsungasing kepada KDNK nominal

- nisbah penganggur kepadajumlah tenaga kerja

= nisbah buruh asing kepadajumlah guna tenaga

• pemboleh ubah rawak

PGINI = pQ + ^RKDNKNP + $JP^PGUNATEI + PJXMKDNK +^RBPINKDNK + ^RFDIKDNK +^RPANGGUR + PSRBAGUNAT+

Kedua-dua persamaan di atas mempunyaiperbezaan dengan adanya pemboleh ubah buruhasing dalam persamaan (2). Objektif memasukkanpemboleh ubah ini ke dalam persamaan (2)adalah untuk melihat sejauh mana pembolehubah buruh asing boleh mempengaruhi agihanpendapatan. Dalam hubungan ini dua persamaanperlu dianggarkan kerana data tentang buruhasing hanya tersedia bagi tempoh 1985-2000sahaja.

Persamaan di atas dianggarkan denganmenggunakan kuasa dua terkecil (ordinary leastsquare atau OLS) mengikut tiga tempoh masa,iaitu 1970-2000, 1970-1984 dan 1985-2000.Persamaan (1) akan dianggarkan bagikeseluruhan tempoh iaitu 1970-2000 dan 1970-1984 dan persamaan (2) akan dianggarkan untuktempoh 1985-2000.

Hipotesis

Sebelum menganalisis hasil penganggaran modelregresi, beberapa hipotesis telah dibentuk dalammenerangkan perhubungan antara pembolehubah bebas dengan pemboleh ubah bersandar.KDNK per kapita dihipotesiskan mempunyaihubungan yang negatif dengan pekali Gini. Inikerana apabila semakin meningkat pertumbuhanekonomi, diikuti oleh peningkatan dalam KDNKper kapita, maka lebih ramai masyarakatmendapat manfaat daripada ini, dan ini



mengurangkan perkadaran masyarakat kurangberada dan agihan pendapatan menjadi lebihseimbang.

Kadar celik huruf juga dihipotesiskanmempunyai hubungan negatif dengan pekaliGini kerana semakin meningkat kadar celikhuruf, maka semakin ramai anggota masyarakatyang berpendidikan. Selanjutnya, pendapatanmereka akan meningkat melalui peningkatanproduktiviti seperti yang diramalkan oleh teorimodal manusia, dan agihan pendapatan menjadilebih seimbang. Peratus guna tenaga sektorindustri pula dijangkakan mempunyai hubunganpositif dengan pekali Gini kerana produktivitidan keperluan kemahiran sektor ini lebih tinggidaripada sektor lain yang selanjutnyameningkatkan upah pekerja mereka.

Tahap keterbukaan ekonomi diandaikanmempunyai hubungan yang positif dengan pekaliGini. Ini kerana manfaat daripada keterbukaanini contohnya liberalisasi perdagangan antarabangsa dan pelaburan hanya mampudimanfaatkan oleh sebahagian masyarakatterutamanya mereka yang sedia memiliki sumber,pengetahuan dan hubung jalin yang luas danlebih terkehadapan daripada individu lain. Olehitu, peningkatan dalam tahap keterbukaan akanmeningkatkan jurang agihan pendapatan. Begitujuga dengan pelaburan langsung asing.Kemasukan pelabur asing ini hanya memberimanfaat kepada individu tertentu dan syarikatmultinasional, manakala syarikat-syarikat kecilakan menghadapi persaingan yang hebat.Selanjutnya agihan pendapatan antaramasyarakat akan menjadi lebih tidak seimbang.

Bayaran pindahan dihipotesiskan mempunyaihubungan negatif dengan pekali Gini. Ini keranaprogram bayaran pindahan adalah bertujuanmembantu golongan kurang berada. Manfaatyang diperoleh daripada program ini akanmeningkatkan keupayaan golongan kurangberada menjana pendapatan dan meningkatkantaraf hidup mereka. Maka agihan pendapatanakan menjadi lebih setara.

Kadar pengangguran diandaikan berhubungsecara positif dengan ketidakseimbangan agihanpendapatan. Apabila individu itu menganggur,maka ia kehilangan sumber pendapatan danpeningkatan bilangan mereka akan meningkatkanperkadaran masyarakat dengan pendapatanrendah. Begitu juga, kemasukan buruh asingdihipotesiskan memberi impak positif kepadaketidakseimbangan agihan pendapatan. Ini

kerana kebanyakan pekerja asing yang masukadalah tidak mahir dan sanggup menerima upahmurah. Ini selanjutnya menjejaskan keupayaanpekerja tempatan untuk mendapatkan upah tinggidan meningkatkan golongan berpendapatanrendah.

Sumber Data

Data bagi menganggarkan persamaan (1) dan(2) di atas diperoleh daripada pelbagai sumber.Pekali Gini diperoleh daripada sumber sekunderseperti Anand, Snodgrass dan buku-bukuRancangan Lima Tahun Malaysia. Disebabkandata ini tidak dilaporkan setiap tahun antara1970-2000, maka anggaran dilakukan untukmendapatkan data setiap tahun dengan andaiania meningkat atau menurun pada kadarpertumbuhan purata yang tetap mengikut trendselepasnya. Contohnya, nilai pekali Gini padatahun 1970 adalah 0.50 dan nilainya pada tahun1976 adalah 0.53. Maka kadar pertumbuhanpurata bagi tempoh ini adalah 0.01 (iaitu denganmenggunakan formula r = ((0.53/0.50)1/6- 1).Oleh itu, dalam tempoh ini pekali Ginimeningkat pada kadar 1.0 peratus. Proseduryang sama digunakan bagi mengira nilai-nilaipekali Gini pada tahun-tahun lain. KeluaranDalam Negara Kasar nominal, jumlah penduduk,jumlah eksport dan import serta pelaburanlangsung asing diperoleh daripada LaporanEkonomi, Kementerian Kewangan Malaysia.Sementara data bayaran pindahan kerajaandiperoleh daripada Buletin Statistik Bulanan,Bank Negara Malaysia. Data tenaga kerja, gunatenaga, penganggur, guna tenaga sektor industridikutip daripada Buletin Perangkaan Sosial,Jabatan Perangkaan Malaysia dan LaporanEkonomi. Data buruh asing pula dikutip dariJabatan Imigresen Malaysia. Akhir sekali, datatentang kadar celik huruf diperoleh daripadaWorld Development Report, World Bank danWorld Tables, United Nation, Yearbook.

Jadual 2 menunjukkan bagi keseluruhantempoh kajian, iaitu 1970-2000 nilai purata pekaliGini adalah 0.484. Bagi tempoh 1970-1984 nilaipekali Gini lebih tinggi daripada tempoh 1985-2000 iaitu 0.508 berbanding 0.461. Kadarpertumbuhan KDNK nominal purata adalah 8.9peratus dalam keseluruhan tempoh kajian.Sekiranya kita mengambil tempoh 1970-1984,kadar ini adalah 11.4 peratus dan menurunkepada 6,6 peratus bagi tempoh 1985-2000. Kadarcelik huruf purata yang menggambarkan



JADUAL 2Nilai purata dan sisihan piawaian pemboleh ubah dalam model

Pemboleh ubah

PGINIRKDNKPLITPGUNATEIJXMKDNKRBPINKDNKRFDIKDNKRPANGGURRBAGUNAT

1970-2000

Nilai purata

0.4840.08970.8290.1731.1040.8022.5930.072

Sisihanpiawaian

0.0270.0799.8640.0540.3773.07010.5310.094

1970-1984

Nilai purata

0.5080.11462.1670.1320.8360.0170.0180.098

Sisihanpiawaian

0.0140.0864.9340.0310.0890.0060.0090.131

1985-2000

Nilai purata

0.4610.06678.9500.2131.3551.5385.0070.0460.057

Sisihanpiawaian

0.0130.0655.1460.0400.3714.20114.4550.0220.051

pencapaian pendidikan penduduk adalah 70.8peratus bagi tempoh 1970-2000. Pembaikandalam sistem dan kemudahan pendidikan negaratelah meningkatkan kadar ini daripada 62.1peratus bagi tempoh 1970-1984 kepada 78.9peratus bagi tempoh 1985-2000. Kepesatanperkembangan sektor pembuatan telahmeningkatkan peratus guna tenaga sektor inisecara mendadak. Dalam tempoh 1970-1984peratus guna tenaga sektor ini daripadakeseluruhan guna tenaga adalah 15.2 peratusdan meningkat kepada 21.3 peratus dalamtempoh 1985-2000.

Penglibatan ekonomi Malaysia dalamperdagangan antarabangsa sentiasa meningkatdari tahun ke tahun. Ini dapat dilihat daripadavolume eksport dan import negara. Tahapketerbukaan sesebuah ekonomi diukur olehnisbah antara jumlah eksport dan import denganKDNK Dalam tempoh 1970-184 nisbah ini adalah1.104. Jika kita membandingkan kedua-duatempoh, iaitu 1970-1984 dan 1985-2000, didapatiberlaku peningkatan yang besar dalam nisbahini daripada 0.836 kepada 1.355. Inimenggambarkan tahap keterbukaan ekonomiMalaysia telah meningkat.

Satu indikator lain yang boleh mengukurtahap keterbukaan atau liberalisasi ekonomiMalaysia adalah nisbah pelaburan langsung asingkepada KDNK nominal. Dalam tempoh 1970-2000 nilai purata nisbah ini adalah 2.593. Secarapuratanya nisbah ini jauh lebih tinggi bagitempoh 1985-2000 dibandingkan dengan tempoh1970-1984, iaitu 5.007 berbanding 0.018.

Nilai bayaran pindahan sangat bergantungkepada prestasi ekonomi sesebuah negara danselalunya berhubung rapat dengan program

pembangunan kerajaan. Nisbah bayaranpindahan kerajaan kepada KDNK nominalmeningkat daripada 0.017 dalam tempoh 1970-1984 kepada 1.538 dalam tempoh 1985-2000.Pembaikan dalam pasaran buruh telah banyakmembantu kepada pengurangan dalam bilanganpenganggur dan kadar pengangguran negara.Secara puratanya, kadar pengangguran Malaysiatelah menurun daripada 9.8 peratus dalamtempoh 1970-1984 kepada hanya 4.6 peratusdalam tempoh 1985-2000. Bagi tempoh 1985-2000 bilangan buruh asing secara puratanyaadalah 5.7 peratus daripada keseluruhan gunatenaga negara.

KEPUTUSAN REGRESI

Penganggaran persaman (1) dan (2) di atastelah melalui dua proses. Percubaan awalmenunjukkan wujud masalah autokorelasi dalampenganggaran. Oleh itu bagi mengatasi masalahini penganggaran terpaksa menggunakanprosedur autoregresive sehingga kepada peringkatkedua. Hasil daripada penganggaran inidipaparkan dalam Jadual 3.

Penganggaran bagi keseluruhan tempohmenunjukkan nilai R2 adalah 0.884 dan bagikedua-dua tempoh nilai R2 adalah 0.985 yangmenggambarkan antara 88.4 peratus hingga 98.5peratus variasi dalam pekali Gini dapatditerangkan oleh pemboleh ubah bersandar yangdipilih. Hasil penganggaran bagi keseluruhantempoh 1970-2000 menunjukkan hanya satupemboleh ubah yang signifikan dalammenentukan pekali Gini, iaitu nisbah bayaranpindahan. Hasil kajian ini menunjukkan bayaranpindahan mempunyai hubungan yang negatif



JADUAL 3Keputusan anggaran penentu agihan pendapatan

Pemboleh ubah

Pemalar

RKDNKP

LIT

PGUNATEI

JXMKDNK

RBPINKDNK

RFDIKDNK

RPANGGUR

RBAGUNAT

R2

D.W.

1970-2000

0547(7.248)***

0.025(1.063)

-0.001(-0.371)

-0.159(-1.106)

-0.006(-0.536)

-0.002(-1.757)*

0.0001(0.520)

0.007(0.394)

0.8842.091

1970-1984

0.331(8.708)***

0.102(5.386)***

-0.001(-1.009)

0.242(5.580)***

0.117(6.088)***

2.619(6.653)***

1.954(7.060)***

-0.020(-1.333)

0.9712.078

1985-2000

0.228(5.972)***

-0.144(-7.653)***

0.003(8.486)***

-0.247(-4.560)**

0.007(1.333)

-0.002(-2.350)*

0.001(2.515)*

0.740(4.514)**

0.0003(8.379)***

0.9852.009

Nota: angka dalam kurungan adalah nilai t-statistik*** = signifikan pada aras keertian 1 peratus.** = signifikan pada aras keertian 5 peratus.* • signifikan pada aras keertian 10 peratus

dengan pekali Gini. Peningkatan 1 titik peratusandalam nisbah bayaran pindahan akanmenurunkan pekali Gini sebanyak 0.002. Inibermakna peningkatan dalam nisbah bayaranpindahan menurunkan ketakseimbangan agihanpendapatan di Malaysia yang menggambarkanprogram bayaran pindahan kerajaan mencapaimatlamat untuk membantu golongan yangkurang berada, dan memperbaiki agihanpendapatan.

Analisis mengikut tempoh yang berbeza telahmenghasilkan keputusan yang lebih baik danmenarik. Bagi tempoh 1970-1984 misalnya,kesemua pemboleh ubah, kecuali kadar celikhuruf dan kadar pengangguran, menunjukkannilai yang signifikan dan berhubung secara positifdengan pekali Gini. Ini bermakna peningkatandalam pemboleh ubah-pemboleh ubah ini akanmembawa kepada meningkatnya ketidakseimbangan agihan pendapatan. Peningkatan 1titik peratusan dalam kadar pertumbuhan KDNK

akan meningkatkan pekali Gini sebanyak 0.102.Pada peringkat awal pembangunan, peningkatandalam kadar pertumbuhan pendapatan perkapita meningkatkan lagi jurang perbezaanagihan pendapatan di Malaysia, dan dapatan iniselari dengan apa yang dihipotesiskan olehKuznets dan selari dengan hasil kajian Hassandan Rahmah (1997). Walaupun dalam tempohini DEB telah dilaksanakan, objektifnya belumtercapai. Proses pembangunan dalam tempohini lebih memanfaatkan golongan yang lebihberada kerana mereka merupakan pemilik modaldan lebih mengetahui peluang ekonomi.

Seperti yang dijangkakan, peningkatandalam guna tenaga sektor industri meningkatkanketidakseimbangan agihan pendapatan.Peningkatan 1 titik peratusan dalam peratusguna tenaga sektor industri akan meningkatkanpekali Gini sebanyak 0.242. Keputusan inimenggambarkan dalam tempoh 1970-1984,proses perindustrian yang berlaku di Malaysia



telah menyebabkan berlakunya perbezaan upahmengikut sektor, dengan sektor industrimendahului sektor-sektor lain. Pekerja dalamsektor ini dibayar upah lebih tinggi keranaproduktiviti mereka lebih tinggi. Dasar kerajaanpada awal pembangunan lebih menekankansektor perindustrian, contohnya melalui dasarpengembangan eksport.

Dalam tempoh ini juga, hasil kajianmenunjukkan tahap keterbukaan ekonomiMalaysia memainkan peranan yang signifikandalam meningkatkan jurang perbezaan agihanpendapatan dan ia selari dengan penemuanPatrik (1998). Hasil kajian menunjukkankenaikan 1 titik peratusan dalam nisbah eksport-import kepada KDNK akan meningkatkan pekaliGini sebanyak 0.117. Peningkatan dalam aktivitiperdagangan antarabangsa menyebabkansegelintir masyarakat mendapat manfaat yanglebih besar daripada golongan yang lain.Contohnya, golongan yang mempunyai dayasaing yang lebih tinggi dari segi teknologi, rakankontak dan pasaran akan mendapat keuntunganyang lebih besar kerana dapat meningkatkaneksport. Sehubungan dengan ini juga hasil kajianini menunjukkan pelaburan langsung asingmemberi impak yang positif terhadapketidakseimbangan agihan pendapatan diMalaysia. Ini sekali lagi menggambarkanliberalisasi pelaburan hanya memberi manfaatkepada golongan yang berada.

Hasrat kerajaan memberi bayaran pindahanadalah untuk membantu golongan kurangberada. Tetapi dalam tempoh 1970-1984,nampaknya objektif ini kurang berjaya keranakesan bayaran pindahan terhadap ketidakseimbangan agihan pendapatan adalah positif.Nilai koefisien ini agak besar, iaitu menunjukkankenaikan 1 titik peratusan dalam nisbah bayaranpindahan meningkatkan pekali Gini sebanyak2.619. Ini ada kaitannya dengan ketakberkesanandalam pelaksanaan bayaran pindahan tersebutyang menyebabkan ia tidak sampai kepadagolongan sasaran. Kemungkinan dalam tempohawal ini, jentera kerajaan masih kurang cekapuntuk mengendalikan program ini secaraberkesan. Justeru, keputusan kajian ini selaridengan penemuan Snodgrass (1974). Sementarakemasukan pelaburan langsung asing memberiimpak positif dan signifikan terhadapketidakseimbangan agihan pendapatan.Kenaikan 1 titik peratusan dalam nisbah

pelaburan langsung asing akan meningkatkannilai pekali Gini sebesar 1.954.

Bagi tempoh 1985-2000 pula, keputusananggaran menunjukkan kadar pertumbuhanKDNK per kapita, peratus guna tenaga industridan nisbah bayaran pindahan mempunyaihubungan negatif dengan pekali Gini.Peningkatan 1 titik peratusan dalam pembolehubah-pemboleh ubah ini masing-masing akanmenurunkan pekali Gini sebesar 0.144, 0.247dan 0.002. Ini menunjukkan proses pertumbuhanselanjutnya telah berjaya mengurangkanketidakseimbangan agihan pendapatan.Peningkatan dalam produktiviti sektor lain jugamembawa kepada perbezaan upah yang lebihtidak ketara dan pelaksanaan kerajaan terhadapprogram bayaran pindahan adalah lebihberkesan. Dalam tempoh ini pencapaianpendidikan masyarakat dan pekali Ginimempunyai hubungan yang positif yangmenggambarkan agihan pendapatan lebih tidakseimbang apabila meningkatnya pencapaianpendidikan. Kenaikan 1 titik peratusan dalamkadar celik huruf akan meningkatkan nilai pekaliGini sebesar 0.003. Ini menggambarkan tahappencapaian pendidikan masih tidak seimbang diMalaysia dan keputusan ini selari dengan teorimodal manusia serta penemuan kajianHammermesh dan Rees (1984) dan Caniglia(1988).

Pelaburan langsung asing masih mempunyaihubungan positif dengan pekali Gini dalamtempoh 1985-2000, tetapi kesannya terhadappekali Gini menjadi semakin kecil, iaitu hanya0.001 berbanding 1.954. Dalam tempoh ini jugakadar pengangguran memberi impak yang positifdan signifikan terhadap ketidakseimbanganagihan pendapatan. Kenaikan 1 titik peratusandalam kadar pengangguran akan menaikkan nilaipekali Gini sebesar 0.740. Namun demikian tahapketerbukaan ekonomi Malaysia tidak memberikesan yang signifikan. Seperti yang telahdihipotesiskan, kemasukan pekerja asingmemburukkan lagi jurang ketidakseimbanganagihan pendapatan di Malaysia. Kenaikan 1 titikperatusan dalam nisbah buruh asing akanmeningkatkan pekali Gini sebanyak 0.0003.Walaupun kesannya agak kecil, tetapi sangatsignifikan dan dikhuatiri meningkat dalam jangkapanjang.



RUMUSAN DAN IMPLIKASI DASAR

Sebagai kesimpulan, terdapat tiga rumusanpenting kajian ini, iaitu berkait dengan enampemboleh ubah yang signifikan dalammenganalisis faktor penentu agihan pendapatan.Pertama, hubungan antara kadar pertumbuhanKDNK per kapita, guna tenaga sektor industridan bayaran pindahan dengan agihanpendapatan adalah tidak stabil, iaitumeningkatkan jurang agihan pada awal tempohkajian dan menurunkan jurang ini pada tempohakhir. Kedua, pelaburan langsung asing dan kadarpengangguran sememangnya memburukkankeadaan agihan pendapatan di Malaysia. Iniditunjukkan oleh hubungan yang positifpemboleh ubah ini dengan pekali Gini dalamkedua-dua tempoh kajian. Ketiga, kadar celikhuruf dan buruh asing memainkan perananpenting dalam meningkatkan jurang agihanpendapatan dalam tempoh 1985-2000.

Berdasarkan kepada hasil kajian ini terdapatbeberapa implikasi dasar yang bolehdiketengahkan.• Pelaksanaan skim bayaran pindahan kerajaan

yang lebih berkesan. Terdapat pelbagai skimyang telah diperkenalkan kerajaanterutamanya untuk membantu golonganberpendapatan rendah seperti programpembangunan rakyat termiskin (PPRT).Kebanyakan skim adalah secara tidaklangsung dalam erti kata bukan dalambentuk pemberian tunai secara langsungkepada golongan tersebut kecuali subsidi.Maka proses pelaksanaan skim tersebutmelibatkan pelbagai peringkat sebelumsampai kepada golongan sasaran. Oleh itudalam proses ini boleh berlaku kesilapanyang boleh menyebabkan bantuan tersebuttidak sampai kepada golongan sasaran. Bagimengatasi masalah ini kerajaan seharusnyadapat mengawasi dengan lebih rapi aktivitipenyampaian skim bantuan mereka bagimemastikannya sampai kepada golongansasaran.

• Peluang pendidikan yang lebih sama rata.Di Malaysia peluang pendidikan telahmengalami pembaikan yang begitu pesatsekali hasil usaha kerajaan. Namun demikianmasih terdapat jurang yang besar pencapaianpendidikan antara rakyat dan ia mengalamipeningkatan. Keadaan ini juga dapatdiperhatikan dalam pasaran buruh di manasebahagian besar pekerja adalah

berpendidikan rendah. Walaupun peratusmereka yang berpendidikan menengah dantinggi telah meningkat, peratus pekerja yangberpendidikan rendah masih lagi tinggi. Bagimengatasi masalah ini, saluran pendidikanyang mencukupi perlu disediakanterutamanya bagi menampung pelajar yangtercicir di peringkat rendah lagi.

• Meningkatkan pembangunan sumbermanusia. Sering diperkatakan bahawa upahperlu dibayar mengikut tingkat produktiviti.Produktiviti pula berhubung secara positifdengan tingkat modal manusia seseorang.Untuk meningkatkan modal manusia, makaseseorang perlu melabur dalam modalmanusia tersebut. Satu bentuk pelaburanyang lazimnya dilakukan di organisasipekerjaan adalah latihan. Dalam soal inimajikan adalah bertanggungjawabmenyediakan kemudahan latihan untukpekerja mereka. Aspek latihan ini menjadisangat kritikal terutamanya bagi pekerja tidakmahir dan separuh mahir untukmeningkatkan kerjaya mereka menjadimahir dan sekali gus mampu dibayar upahyang lebih tinggi. Strategi ini bukan sahajadiperlukan dalam sektor industri tetapidalam sektor-sektor lain bagi mengatasiperbezaan produktiviti pekerja antara sektordan juga dalam sektor yang sama.

• Mengawal kemasukan pelaburan langsungasing. Banyak manfaat yang diperolehdaripada pelaburan langsung asing sepertimenyelesaikan masalah kekurangan modaldalam negara, kekurangan kepakaran,menyediakan pekerjaan, meningkatkanpenyelidikan dan pembangunan sertakemahiran teknologi. Walau bagaimanapunhasil kajian ini menunjukkan pelaburanlangsung asing memberi impak positifterhadap ketidakseimbangan agihanpendapatan. Oleh itu, kerajaan perlu berhati-hati dalam soal ini dengan mengehadkanpelaburan asing dalam aktiviti-aktiviti yangtidak dapat diterokai oleh rakyat tempatankerana kekurangan modal dan kepakaran.Begitu juga, kemasukan pelabur asingseharusnya mampu memindahkan teknologikepada rakyat tempatan supaya kebergantungankepada mereka ini tidaklah berpanjangan.

• Meningkatkan peluang pekerjaan. Hasilkajian ini menunjukkan peningkatan kadarpengangguran meningkatkan jurang agihan



pendapatan. Bagi mengurangkan kadarpengangguran, kita perlu melihat kepadaaspek permintaan dan penawaran buruh.Dari sudut permintaan, peluang-peluangpekerjaan perlu diwujudkan terutamanyaoleh pihak swasta. Dari sudut penawaranpula, pencari kerja perlu melengkapkan dirimereka dengan kemahiran yang diperlukanmajikan sama ada melalui sistem pendidikanyang lebih sesuai atau menghadiri latihanyang disediakan. Dalam soal ini sistempendidikan negara memainkan perananpenting dalam menawarkan tenaga kerjayang sesuai dengan keperluan majikan.

• Mengawal kemasukan buruh asing.Kebanjiran buruh asing ke negara ini sangatberkait dengan kekurangan pekerja dalamnegara. Namun demikian, dalam masa yangsama terutamanya akhir-akhir ini, negaramenghadapi masalah pengangguran yangagak membimbangkan. Oleh itu, soalkekurangan pekerja tempatan seperti yangsering dikemukakan oleh majikankemungkinannya tidak benar dan dirasakanada perkara-perkara lain mengapa majikanlebih suka mengambil pekerja asingdibandingkan pekerja tempatan. Walau apapun alasan majikan, kita merasakankemasukan buruh asing perlu dikawal untukmengurangkan masalah pengangguran danjuga memperbaiki agihan pendapatan diMalaysia. Kerajaan perlu meletakkan kuotapekerja asing untuk sektor-sektor tertentudan kuota ini perlu diperkecilkan melaluimasa.

KESIMPULAN

Trend agihan pendapatan sesebuah berubahmengikut masa. Trend yang dialami Malaysiapada peringkat awal pertumbuhannya adalahselari dengan apa yang dihipotesiskan olehKuznets. Tetapi pada masa sekarang Malaysiatelah memasuki pusingan kedua hipotesis U-terbalik Kuznets dengan mengalami peningkatandalam ketidakseimbangan agihan pendapatanselepas 1990. Sebenarnya banyak faktor yangboleh mempengaruhi trend agihan pendapatan.Kajian ini telah cuba mengenal pasti danmenentukan tahap kepentingan faktor-faktortersebut dan mendapati terdapat lapan faktorpenting iaitu pertumbuhan KDNK per kapita,

bayaran pindahan, pelaburan langsung asing,kadar celik huruf, guna tenaga sektor pembuatan,kadar pengangguran, keterbukaan ekonomi danpekerja asing. Oleh yang demikian bagimenjamin pertumbuhan yang tinggi diikuti olehagihan pendapatan yang sama rata beberapastrategi adalah dicadangkan dan keberkesananstrategi tersebut sangat bergantung kepadakerjasama semua pihak,

RUJUKAN

AHLUWALIA, M. D. 1976. Inequality, poverty anddevelopment. Journal of Development Economics3(4): 307-342.

ALESINA, A. dan D. RODRIK. 1994. Distributive politicsand economic growth. Quarterly Journal ofEconomics 109(2): 465-80.

ANAND, S. 1983. Inequality and Poverty in Malaysia:Measurement and Decomposition. OxfordUniversity Press.

BANK NEGARA MALAYSIA. Buletin Statistik Bulanan.

Pelbagai tahun.

BEER, L. 1999. Income inequality and transnationalcorporate penetration. Journal of World SystemResearch 5: 1.

BLINDER, A. S. dan H. Y. ESAKI. 1978.

Macroeconomic activity and incomedistribution in the postwar United States. Reviewof Economics and Statistics 60: 604-609.

BOLLEN, K. A. dan R. W. JACKMAN. 1985. Political

democracy and the size distribution of income.American Sosiobgical Review 50: 438-457.

CAMPANO, F. dan D. SALVATORE. 1988. Economic

development, income inequality and Kuznets'U-shaped hypothesis. Journal of Political Modeling10(2): 265-280.

CANIGLIA, A.S. 1988. Inequality in educationalattainment and the distribution of income.Review of Business 10(2): 5-10.

CAROL, L. 1998. Trade and income distribution indeveloping countries. Working Papers inEconomics. No 9, Department of Economics,Goteborg University.

CHASE-DUNN, C. 1975. The effects of internationaleconomic dependence on development andinequality: A cross-national study. AmericanSociological Review 40: 720-738.


Rahmah Ismail 8c Poo Bee Tin

CHTU, W. H. 1998. Income inequality, human capitalaccumulation and economic performance. TheEconomic Journal 108(January): 44-59.

HAMMERMESH, D.S. dan A. REES. 1984. The Economics

of Work and Pay. New York: Harper & RowPublishers.

HASSAN HJ . ALI dan RAHMAH ISMAIL. 1997.

Perhubungan antara pertumbuhan ekonomidengan agihan pendapatan : Kes Malaysia. SiriKertas Penyelidikan, Jabatan EkonomiPembangunan, Fakulti Ekonomi, UniversitiKebangsaan Malaysia, Bangi.

ISHAK SHARI, RAGAYAH HJ. MAT ZIN, RAHMAH ISMAIL,

NIK HAIRI NIK OMAR dan MOHD. SHUKRI HJ.

NOOR. 1999. Perindustrian dan AgihanPendapatan di Malaysia. Laporan AkhirPenyelidikan Kumpulan Wang PengajianPembangunan, Bank of Tokyo. FakultiEkonomi, UKM.

JABATAN IMIGRESEN MALAYSIA. 1996. Pendafiaran Pekerja

Asing. Bahagian Maklumat dan Rekod.

JABATAN PERANGKAAN MAIAYSIA. Buletin Perangkaan

SosiaL Pelbagai tahun.

JHA, S. 1996. The Kuznets curve, a reassessment.World Development 24(4): 773-780.

KEMENTERIAN KEWANGAN MALAYSIA. Laporan Ekonomi

Pelbagai tahun.

KUZNET, SIMON. 1955. Economic growth and incomeinequality. American Economic Review 45(1):1-28.

MALAYSIA. 1981. Rancangan Malaysia Keempat, 1981-85. Kuala Lumpur: Jabatan Percetakan Negara.

. 1984. Kajian Separuh PenggalRancangan Malaysia Keempat, 1981-85. KualaLumpur: Jabatan Percetakan Negara.

. 1986. Rancangan Malaysia Kelima,1986-90. Kuala Lumpur: Jabatan PercetakanNegara.

.. 1989. Kajian Separuh PenggalRancangan Malaysia Kelima, 1986-90.

. 1991 (a). Rancangan Malaysia Keenam,1991-95. Kuala Lumpur: Jabatan PercetakanNegara.

. 1991 (b). Rangka Rancangan JangkaPanjang Kedua, 1991-2000. Kuala Lumpur:Jabatan Percetakan Negara.

. 1993. Kajian Separuh PenggalRancangan Malaysia Keenam 1991-95. KualaLumpur: Jabatan Percetakan Negara.

. 1996. Rancangan Malaysia Ketujuh,1996-2000. Kuala Lumpur: Jabatan PercetakanNegara.

. 1999. Kajian Separuh PenggalRancangan Malaysia Ketujuh, 1996-2000. KualaLumpur: Jabatan Percetakan Negara.

MAIAYSIA. 2001. Rancangan Malaysia Ketujuh, 1996-2000. Kuala Lumpur: Jabatan PercetakanNegara.

MARX, KARL. 1952. Wage Labour and Capital Progress.Moscow: Progress.

MEERMAN, J. 1975. Public Expenditure in Malaysia:Who Benefits and Why? New York: OxfordUniversity Press for the World Bank.

MILLANOVIC, B. 1995. Poverty, inequality and socialpolicy in transition economies. TransitionEconomics Division, Research Paper 9. TheWorld Bank, Washington, DC.

OSHIMA, H.T. 1994. The impact of technologicaltransformation on historical trends in incomedistribution of Asia and the West. DevelopingEconomies 32(3): 237- 255.

PAPANEK, G. dan O. KYN. 1986. The effect of incomedistribution on development, the growth rateand economic strategy. Journal of DevelopmentEconomics 23: 55-65.

PATRICK, L. 1998. World trade survey: The wages offear. The Economist, 6 October, London.

PAUKERT, F. 1973. Income distribution at differentlevels of development. International LabourReview August-September: 81-95.

PEROTTI, R. 1996. Growth, income distribution anddemocracy: What the data say. Journal ofEconomic Growth 1: 149-87.

PERRSON, T. dan G. TABELUNI. 1994. Is inequality

harmful for growth? American Economic Review84(3): 600-21.

RAHMAH ISMAIL. 2000. Perbezaan pendapatan buruhmengikut kemahiran dalam sektor pembuatandi Malaysia. Kertas kerja yang dibentangkandalam Bengfal Hasil Penyelidikan Kumpulan WangPengajian Pembangunan, 28-29 Oktober, PortDickson, anjuran Fakulti Ekonomi, UKM.

128 Pertanika J. Soc Sci. & Hum. Vol. 10 No. 2 2002


ROMER, P. M. 1986. Increasing returns and long-run growth. Journal of Political Economy 94:1002-37.

RUBINSON, R. 1976. The world economy and thedistribution of income within states : A cross-national study. American Sociological Review 41:638-659.

SNODGRASS, D. R. 1974. The fiscal system as anincome redistributor in West Malaysia. PublicFinance 20: 1.

SNODGRASS, D. R- 1980. Inequality and Economic

Development in Malaysia. Kuala Lumpur: OxfordUniversity Press.

THE WORLD BANK. 2001. World Development Indicator

2001. New York: Oxford University Press.

THE WORLD BANK. 1996. World Development Report

1996/97. New York: Oxford University Press.

TSAI P. L. 1995. Foreign direct investment andincome inequity: Further evidence. Journal ofWorld Development 23(3): 469-483.

WATERS, M. 1995. Globalization. Routledge, London& New York.

WEEDE, E. dan H. TIEFENBACH. 1981. Some recentexplanations of income inequality: Anevaluation and critique. International StudiesQuarterly 25: 255-258.

WES, M. 1996. Globalization: Winner and Losers.Commission on Public Policy and BritishBusiness : Issue Paper No.3.

WINEGARDEN C. R. dan L. B. KHOR. 1993.

Undocumented immigrant and incomeinequality in the native-born population of theUS: Econometric evidence. Journal of AppliedEconomic 25(2): 157-163

Xu L. C. dan H. Zou. 2000. Explaining the changesof income distribution in China. China EconomicReview 11: 150-169.

(Diterima: 18 Januari 2002)


Pertanika J. Soc. Sci. & Hum. 10(2):131-141 (2002) ISSN: 0128-7702© Universiti Putra Malaysia Press

Performances of Non-linear Smooth Transition Autoregressive and LinearAutoregressive Models in Forecasting the Ringgit-Yen Rate

LIEW KHIM SEN 8c AHMAD ZUBAIDI BAHARUMSHAHDepartment of Economics,

Faculty of Economics and Management,Universiti Putra Malaysia,

Serdang, Selangor, Malaysia

Keywords: Autoregressive, smooth transition autoregressive, non-linear time series, forecastingaccuracy

ABSTRAK

Kajian ini membandingkan prestasi peramalan di antara model ketaklinearan AutoregresiBerperalihan Licin (Smooth Transition Autoregressive atau STAR) dengan model linear siri masaiaitu model Autoregresi (Autoregressive atau AR) sebagai peramal kadar pertukaran Ringgit-Yen.Berdasarkan kepada prosedur penentuan kelinearan piawai, kami memperoleh bukti empirikbahawa penyelarasan Ringgit-Yen ke arah keseimbangan pariti kuasa beli (Purchasing PowerParity) adalah secara tidak linear. Dari segi prestasi peramalan, keputusan empirik menunjukkanbahawa model STAR dan AR mempunyai purata mutlak ralat ramalan (mean absolute forecasterror atau MAFE), purata mutlak perarus ralat ramalan (mean absolute percentage forecast erroratau MAPFE) dan purata punca kuasa dua ralat ramalan (mean square forecast error atauRMSFE) yang lebih kecil jika dibanding dengan model SRW. Keputusan yang diperoleh jugamenunjukkan ramalan model STAR lebih baik jika dibandingkan dengan pesaing linearnya, iaitumodel AR. Hasil kajian ini adalah konsisten dengan penyelidikan yang memberi penekanankepentingan membenarkan pelarasan yang tidak linear bagi kadar pertukaran asing ke arahkeseimbangan jangka panjang.

ABSTRACT

This study compares the performance of Smooth Transition Autoregressive (STAR) non-linearmodel and the conventional linear Autoregressive (AR) time series model in forecasting theRinggit-Yen rate. Based on standard linearity test procedure, we find empirical evidence that theadjustment of the Ringgit-Yen rate towards its long-run Purchasing Power Parity equilibriumfollows a non-linearity path. In terms of forecasting ability, results of this study suggest that boththe STAR and AR models exceed or match the performance of SRW model based mean absoluteforecast error (MAFE) mean absolute percentage forecast error (MAPFE) and mean squareforecast error (RMSFE). The results also show that the STAR model outperforms the AR model,its linear competitor. Our finding is consistent with the emerging line of research that emphasisedthe importance of allowing non-linearity in the adjustment of exchange rate toward its long runequilibrium.

INTRODUCTIONIn 1926, Yule first formally introduced the timeseries model in the form of autoregressive (AR)model, which assumes that the future values ofa variable depend solely on its historical values.Since then, time series analysis has been viewedas a powerful forecasting tool. In the past twodecades or so the theory of time series

econometrics is progressing rapidly(Montogomery et al 1990). As the methodologyprogresses, the issue of non-linearity wasincorporated into the analysis of time series.Smooth Transition Autoregressive (STAR) modelis one of the most recent models developedunder this concept. The STAR model is a non-linear time series model that allows the variable

Uew Khim Sen 8c Ahmad Zubaidi Baharumshah

under investigation to move within two differentstate spaces with a smooth transition process.STAR offers an alternative to the modelling oftime series variables that exhibit non-linearities.

In our review of the literature, we foundthat the application of the STAR model inempirical works is still very limited, and itsforecasting performance particularly withreference to exchange rates has yet to bedetermined. Taylor and Peel (2000), Sarno(2000a) and Baum et al (2001) are among thefirst to demonstrate the usefulness of the STARin modelling exchange rate dynamics. However,all these authors did not evaluate the forecastingperformance of the model. Hence, the mainobjective of this study is to examine theapplicability of the STAR model to the MalaysianRinggit against the Japanese Yen (RM/YEN). Inaddition, the present article compares theforecasting performance of the STAR and ARmodels by using the simple random walk (SRW)model as the yardstick of comparison. This paperextends previous studies by evaluating theforecasting performance of the studied modelsusing mean absolute forecast error (MAFE),mean absolute percentage forecast error(MAPFE) and root mean square forecast error(RMSFE).

We chose to apply the model to theexchange rate for the following reasons: First,the bulk of the literature shows that structuralmodels failed to outperform a simple randomwalk (SRW) model, and attempts by analystsusing more elaborate models have also failed toimprove the forecast performances at short orlong run horizons significantly. Second, severalauthors have argued that the failure of existingexchange rate models to yield superior forecastis because these models ignore the non-linearityadjustment of exchange rates towards itsequilibrium value (Micheal et al. 1997; Taylorand Peel 1997; Sarno 2000a,b; Coakley andFuertes 2001). These studies also argued thatthe classical unit root tests may not be able todetect mean reverting behaviour of exchangerate if the variable is a stationary non-linearprocess. Finally, in applying this model to theRinggit, we intend to broaden our understandingon the appropriateness of the STAR model as aforecasting tool in the currency markets.

This study models the adjustment process ofthe deviations of Yen-based Ringgit movementfrom its fundamental equilibrium as determinedby the Purchasing Power Parity (PPP) hypothesis.Simply, the PPP hypothesis postulates that thenominal exchange rate is given by the ratio ofthe domestic and foreign price levels. It statesthat exchange rates should tend to equalizeprices for identical goods in different countries.Recent studies based on careful application oftime series econometrics methods, are moresupportive of the mean reverting behaviour ofexchange rates towards the long-run PPPequilibrium value (M. Azali et ai 2001; Baum etal. 2001)1.

To anticipate our results, we find that RM/YEN rate adjusts in a non-linear fashion towardsits long-run equilibrium path using the LagrangeMultilpier (LM)-type test developed by Terasvirta(1994). Unlike most of the earlier studies, wefind that both the STAR and AR modelsoutperform a random walk forecast for nominalRM/YEN rate. In addition, the empirical resultssuggest that the non-linear STAR model performsbetter than the linear AR model in the out-sample forecasts. The set out of the paper isorganised as follows. The first two sections offera brief review on the development of STARmodels and a discussion on the data used in theanalysis. The section that follows immediatelydescribes the linearity test and the test results.We report and interpret the results of forecastaccuracy comparison just before offering someconcluding remarks in the last section.

THE STAR MODELSThe origin of the non-linear Smooth TransitionThreshold Autoregressive or just SmoothTransition Autoregressive (STAR) model couldbe traced back to the Threshold Autoregressive(TAR) model first proposed by Tong in 1977(see Tong and Lim 1980). The TAR modelassumes that a variable has different behaviourwithin different regimes. The basic ideaunderlying the TAR model is piecewiselinearisation of non-linear models over the statespace by the introduction of the thresholds. Anexample of the TAR model is the Self-excitedTAR (or SETAR) model, which assumes that a

The stylized fact that emerged from this literature is that exchange rate adjusts non-linearly towards its long-run PPPequilibrium (Mahajan and Wagner 1999; Sarno 2000a; Baum et al 2001 and Coakley and Fuertes 2001).


Performances of Non-linear Smooth Transition Autoregressive and Linear Autoregressive Models

variable (say, exchange rate), y{ is a linearautoregression within regime, but may movebetween regimes depending on the value takenby a lag of yt, say y^ where d is known as delayparameter. For two-regime case {q-2) where yt

follows an AR (px) process in the one regimeand AR (£,) process in the other, the SETAR (2;pv p2) representation of can be written compactly

as:

' ft

M

(1)

where /(r) = 1 if yt > rand 0 otherwise is thethreshold. For h = 1, 2, e^ ~ G (0, c£ ) where G(•) may be a Gaussian distribution but this is notnecessarily the case. fi. for i = 0, .., px and /T fori = 0, .., p2 are parameters to be estimated.

The introduction of non-linear time seriesmodel such as SETAR model is motivated by thefact that linear time series model should giveplace to a much wider class of models if we wereto gain more understanding into the morecomplicated phenomena such as limit cycles,time irreversibility, amplitude-frequencydependency and jump resonance (Tong andLim 1980). Since its introduction, few attemptshave been made in applying and validating theSETAR mode, and hence the usefulness of themodel in empirical work is yet to be determined.For instance, Diebold and Nason (1990) pointout that there is no guarantee that SETAR modelwill perform better than linear AR model. Asimilar view is expressed in Clements and Smith(1997), where they note that neither in-sample,nor the rejection of null of linearity in a formaltest in favour of non-linearity guarantees thatSETAR predicts more accurately than AR models.

The deficiency in SETAR is deemed due tothe unrealistic fixed threshold in the model.The fixed threshold of SETAR model is laterreplaced with a smooth function and thus leadsto the formation of STAR model in the early1990s. STAR model allows the variable understudy to alternate between two different regimeswith a smooth transition function between theseregimes, so that there can be a continuum ofstates between extreme regimes. STAR

representation is given by (Terasvirta andAnderson 1993):

(2)

where yt is mean-corrected, /SQ, and /J*o areconstants, p. and )S*, i = 1, ..., p are autoregressiveparameters, F() is the transition functiondepending on the lagged level, y^ , where d isknown as the delay length or delay parameter,and et is a white noise with zero mean andconstant variance </.

For the application of STAR model, Grangerand Terasvirta (1993) have proposed exponentialfunction as one of the plausible transitionfunctions, thus resulting in the exponential STARor ESTAR model. The exponential function isdefined as:

(3)

where y2 is the unstandardized transitionparameter, u is the equilibrium or thresholdvalue of the mean corrected yt series and hence

E(u) - 0, and &2yt is the estimated variance of yt.

Note that f/ a2yt is the standardized

transition parameter and the speed of transitionbetween the two regimes is positively related tothe value of the transition parameter f. In otherwords, higher values of f- imply much fasterspeed of transition. Taylor and Peel (2000) use

a version of transition function, F() with a2yt •

1. Nevertheless, Granger and Terasvirta (1993,p. 124) argue that scaling the exponential termby the sample variance speeds the convergenceand improves the stability of the non-linear leastsquares estimation algorithm. It also makes itpossible to compare estimates of transitionparameter across equations.

The exponential transition function isbounded between zero and one. Judging fromEquation 3, when y^ equals its equilibrium value

u or when f/(J2yt goes to zero, F() = 0 and

Equation 2 reverts to a standard linear AR(p)representation:


Liew Khim Sen 8c Ahmad Zubaidi Baharumshah

(4)

In such case, the conventional restriction of

PE Pt < 1 applies so that yt is mean-reverting.

For extreme deviations from the fundamentalequilibrium, F() = 1 (when f/o2

yt approachesinfinity), and Equation 2 becomes a non-linearAR (p) model:

If the non-linear model as in Equation 5 is the

correct specification, it is expected that 12, A ^ 11

such that yt may exhibit unit root behaviour butthe requirement for global stability is that

< 1 must be met.

The exponential F( ) allows a smoothtransition between two regimes symmetryadjustment for deviations above and below thefundamental equilibrium. This function isconsidered suitable for the non-linear modellingof exchange rate as it has a number of attractiveproperties. For instance, it can capture thesymmetrical response to positive and negativedeviations from its fundamental equilibrium(Baum et al 2001) by its inverted bell-shapeddistribution around zero. Despite the potentialusefulness of the ESTAR model in modellingthe exchange rate specifically and other non-linearities in general, much more empirical workhas yet to be done to fill up the related literature.To this end, there are only few published articleson the STAR model. Moreover, most of themare theoretical in nature and the application isonly for illustration purposes. Earlier works, forinstance by Chan and Tong (1986), Luukkonnenet al. (1988), Saikkonen and Luukkonen (1988),Luukonnen (1990), Terasvirta (1994) andEirtheim and Terasvirta (1996) discuss thetheoretical issues on the linearity tests and modelspecification of the STAR models. One notable

exception is Terasvirta and Anderson (1993),which evaluates the forecast performance ofESTAR model, in the context of business cycles.

PRELIMINARY DATA ANALYSIS

The data used in this paper are end-of-periodnominal bilateral exchange rate for the MalaysianRinggit vis-a-vis the Japanese Yen (RM/YEN)and relative price (P(), which is constructed asthe ratio Consumer Price Index (CPI) of Malaysiato CPI of Japan. The data are mainly fromInternational Monetary Fund's InternationalFinancial Statistics (IMF/IFS), comprising ofseasonally unadjusted observations. Our sampleperiod ranges from 1980:1 to 2000:2. The fullsample period is divided into two portions. Thefirst sub-period, which starts from 1980:1 andends in 1997:2 is used for the model estimationpurpose while the remaining observations arekept for assessing the out-sample forecastperformance of the studied models2.

To test whether RM/YEN rate exhibits meanreverting behaviour to its long-run PPPequilibrium, we check for the cointegratingrelationship between the two price series.However, prior to any cointegration test, theseries involved should be tested for stationarityand order of integration beforehand. This isimportant as only variables of the same order ofintegration may provide a meaningfulrelationship. The commonly used AugmentedDickey-Fuller (ADF) and non-parametric Philips-Perron (PP) unit root tests are employed for thispurpose. The results of the unit root tests assummarised in Table 1 strongly suggest that thevariables are first difference stationary, whichimplies they are all integrated of the same order,that is, 1(1). These results hold whether trend orwithout trend.

Next, we proceed to investigate whether ornot the long-run PPP conditions hold using theJohansen and Juselius (1990) multivariatecointegration test. Results of the trace test aredepicted in Table 2. The test result providesstrong evidence that RM/YEN rate and relativeprice are cointegrated at standard significancelevels, thereby verifying that RM/YEN rateexhibits mean reverting behaviour to its long-run PPP equilibrium. The results so far obtained

2 The estimation period ends in 1997: 2 and the forecasting horizon includes the 1997/98 Asian financial crises. Thevolatility of the exchange rates during the currency turmoil allows us to evaluate the robustness of our forecastsduring the crisis and post-crisis periods.



Tests

ADFPP

UnitTABLE 1

root tests

Intercept Without Trend

X AX

-0.362 -4.958*-0.528 -8.653*

P

1.9893.726

AP

-5.537*-11.67*

results

Intercept with Trend

X

-2.696-2.987

AX

-4.953*.8.653*

P

^0.0400.233

AP

-6.260*-12.86*

Notes'. ADF and PP refer to Augmented Dickey-Fuller test and Phillip-Perrontest respectively. X and P denote exchange rate and relative price respectively.Variable with A in front means its first difference. Test-statistics with asterisk (*)imply reject null hypothesis of unit-root at 1% significance level.

TABLE 2Johansen andjuselius cointegration test result

Likelihood Ratio of Eigen Value

Optimal Lag = 0 r <; 1

10 24.369*

Critical Values5%1%

19.9024.60

5. 061

9.2412.97

Notes, r denotes the hypothesized number of cointegrating equation. Optimumlag-length is determined by the Akaike Information Criterion (AIC). The singleasterisk (*) denotes rejection of hypothesis at 5% significance level.

are consistent with those reported inBaharumshah and Ariff (1997) and M. Azali etaL (2001).

This finding enables us to estimate theequilibrium values of RM/YEN rate based onthe PPP hypothesis. Deviations of RM/YEN r a t e

from its equilibrium (^)can then be deduced bysubtracting its observed values from the estimatedequilibrium values. The nature of adjustmentprocess of these deviations towards theequilibrium position is not known yet. Todetermine the linearity (or non-linearity) of thisadjustment process, we employ the linearity testsagainst the STAR models as described below.

LINEARITY TESTS

The minimum requirement for the estimationof STAR models is to reject the linearity of thevariable under study (Tong and Lim 1980).Various linearity tests have been developed basedon the idea of testing the null hypothesis that allP* s in Equation 2 are simultaneously zero,against the alternative hypothesis that at leastone p* is not zero. Note that if the null hypothesiscannot be rejected, Equation 2 would simply be

reduced to the linear AR (p) model. By thesame token, rejection of null hypothesis impliesthe presence of non-linearity in favour of STAR(p) model. As the properties of the transitionparameter (y2), the coefficients of non-linearterms (jft* s) and the mean value (/1) of thevariable under estimation are not identifiedunder the null hypothesis, linearity is tested inthe context of auxiliary model instead of theoriginal STAR specification as in Equation 2.Theoretical issues on linearity tests against STARmodels are found in Luukkonen et aL (1988),Saikkonen and Luukkonen (1998), Terasvirtaand Anderson (1993), Terasvirta (1994) andEirtheim and Terasvirta (1996). Interestedreaders may refer to these articles for moredetailed discussion on the tests.

This study only highlights a specification oflinearity test with alternative hypothesis in favourof the ESTAR model, a variant of STAR modelrelevant to this study. This specification asproposed by Terasvirta (1994), is based on thefollowing auxiliary regression:

PertanikaJ. Soc. Sci. &: Hum. Vol. 10 No. 2 2002 135

Liew Khim Sen & Ahmad Zubaidi Baharumshah

The null hypothesis to be tested is uiat:

(7)

In practice, Terasvirta's Lagrange Multiplier(LM) linearity tests can be performed byfollowing these steps:(1) Regress yt on (1, ytj; j • 1, ..., p). Obtain the

estimated residuals it and compute the residual

sum of squares, SSRQ = 2 ^ > where T is the

sample size;

(2) Regress it on {1, j ^ , yj^ yj*J. Obtain theestimated residuals w{ and compute the residual

sum of squares SSR = 2* ^ ;

(3) Compute the test statistic:

LM =

estimated variance of et

where a2 is the

(8)

Under the null hypothesis die LM statistic isasymptotically distributed as a chi-squared (x2)with 2p degrees of freedom, given that the delayparameter d is known. For unknown d, thedegrees of freedom would be as large asO.5p(p±l)+2p. This LM linearity test is actuallysimilar to that of Luukkonen, Saikkonen andTerasvirta (LST) (1988), which is given as:

LST =T(SSR0 - SSR)

SSRn(9)

LST is also asymptotically distributed as a x2

with 0.5p(p*-l)+p degrees of freedom. Luukkonenet ai (1988) point out that if the delay parameterd in Equation 2 is assumed known, the numberof degrees of freedom for LST statistic wouldreduce largely to /H-1 only. This shows that priorknowledge about d is thus very useful in testinglinearity against ESTAR model

Briefly, the optimum lag length p in the firststep of the above auxiliary regression procedure

is usually unknown even if the true model islinear, and it has to be determined from thedata. Model selection criteria such as FinalPrediction Error (FPE), Schwarz InformationCriterion (SIC) and Akaike's InformationCriterion (AIC) are normally used for thispurpose. However, these criteria are of coursenot without any shortcomings (see for instance,Liew and Shitan (2002) for a brief review of theproperties of these selection criteria). In general,these information criteria tend to penalise high-order lags.

On the other hand, if the selected p is toolow, the estimated AR (p) model may sufferfrom autocorrelated residuals. Terasvirta andAnderson (1993) pointed out that neglectingthe autocorrelation structure of the residualsmay lead to false rejection of the linearityhypothesis in favour of the non-linearitiesalternative, because often the test also has lowpower against serially correlated errors. As such,one may think that over-parameterisation of thelinear AR (p) is preferable to under-parameterisation. However, selecting a maximumlag-length greater than the true order p may alsoweaken the power of the test compared to thecase where the maximum lag is known (Terasvirtaand Anderson 1993). Thus, it is important toselect order p sufficient enough to eliminateautocorrelation.

In this study, the optimal lag length p oflinear AR (p) model is selected based on theAkaike's biased Corrected Information Criterion(AICC). This criterion selects the minimum AICCmodel, among a class of models with no serialcorrelation. Liew and Shitan (2002) examinethe behaviour of AICC through a simulationstudy and find that it has little tendency tounderestimate the true order p. Thus the use ofAICC avoids the problem of too parsimoniousmodel being selected. Based on AICC, theoptimal p is determined as 2 in the present case.

Having selected p, d needs to be determined.In order to specify d, linearity test is carried outfor the range of values considered appropriate,in this case, Isdk5. If the linearity is rejected formore than one value of d, then d is determinedsuch that Z(d)= sup Z(d) for I<=«k5 where Z isthe selected test LM or LST. The argumentbehind this rule of maximising the test statisticis that the test has maximum power if d ischosen correctly, whereas an incorrect choice ofd weakens the power of the test. Ljung-Box



portmanteau Q test is also employed to confirmthe absence of serial correlation up to 20 lags.

Results of the linearity tests are summarisedin Table 3. It is clear from the table that linearityis rejected at 1% significance level for thedeviations of RM/YEN rate from the equilibriumvalue and hence in favour of the ESTAR model.The optimal values of p and d are determined as2 and 1 respectively. The Q statistic suggests thatthe combination of p and d selected for themodel yields residuals that are free fromautocorrelation problem up to 20 lags. Thisimplies ESTAR (2) process with a delayparameter, d=\ is the appropriate representationof the adjustment of deviations towards the long-run PPP equilibrium for the RM/YEN rate duringthe sample period.

ESTIMATED MODELS

In this study, two versions of ESTAR model,namely the unrestricted and restricted ESTARmodels are estimated for yt, the deviations ofRM/YEN rate. The unrestricted ESTAR model,is actually the model as specified in Equation 2with a transition function given in Equation 3.The word "unrestricted" is given to differentiateit from a special case whereby certain restrictions

( 2 Pi = * Pi * "Pi a n d ( T a v l o r and peel 2000))

are imposed on it, thus resulting in a so-calledrestricted ESTAR model. The results obtainedfrom these models are reported in Table 4.

Several features for the estimatedunrestricted model are noteworthy here: First,the non-linear parameters (/T| and y) of theunrestricted ESTAR (2) model are statistically

TABLE 3Linearity tests results

Test statistics

LM testLST testLjung-Box Q

Sup Z (d) Delay

19.62719.920

statistic 15.912

Notes. Null hypothesis, HQ: Linear modelnonlinearity in favour of ESTAR model.

Parameters

A,Aftr,A*2Y

* 2°ESTAR

® ESTAR f °:AR

Q(20)WHITEARCH (4)GARCH (1, 1)LR (3)Adjusted R2

TABLEEstimated ESTAR

parameter, d 1% Critical value

1 ^ (4 ) = 13.281 ^ (3 ) = 11.34

^(20)=37.57

is correct. Rejection of Ho implies

4(2) models

Estimated Values (Standard Errors)

Unrestricted model

-0.031(0.19)1.799 (0.33)**-0.340 (0.23)-1.170 (0.40)**0.349 (0.50)1.364 (0.25)**

0.171

Restricted model

-0.025 (0.02)1.442 (0.19)**-0.253 (0.16)--0.357 (0.05)**

0.171

Diagnostic Tests (Marginal Significance Values)

0.002

0.769

17.522 [0.619]5.276 [0.809]1.563 [0.816]0.546 [0.761]-0.882

0.002

0.819

20.492 [0.428]8.270 [0.507]1.595 [0.809]0.531 [0.767]8.952 [0.030]0.878


liew Khim Sen & Ahmad Zubaidi Baharumshah

significant at 1% level. Second, the residualvariance ratio of this unrestricted ESTAR (2)model to the linear AR (2) model is 0.769indicating that the former has a much smallervariance. This implies that the non-linear modelhas the ability to produce smaller forecast errorsthan the linear model. Third, the model passeda battery of diagnostic tests at conventionalsignificance levels. Fourth, the adjusted R2 value(0.882), the explanatory power of this non-linearmodel on the adjustment of deviations, is fairlyhigh. Fifth, the sum of linear parameters ft and02 equals 1.459 > 1, suggesting that yt exhibitsunit root behaviour and therefore linear AR (2)itself is inadequate representation of yr On the

other hand, | ( f t + P-) = 0.638 <1 implies that

the requirement for global stability is met. Thisconfirms that yt is mean-reverting in the non-linear specification (Baum et at. 2001; Taylorand Peel 2000). Lastly, our results seem to besupportive of die following relationships: ft =-ft and ft = -pr

The conclusion that can be drawn from thefirst five findings is that the above non-linearmodel is appropriate representation of 3?̂ thedeviations of RM/YEN rate; whereas the two lastresults enable us to estimate the restricted ESTARmodel, with the following restrictions imposedon the unrestricted model: f$x + p2 =1, ft = -px

and ft = -P2 (Taylor and Peel 2000). We utilised

the likelihood ratio (LR) test to determine thevalidity of these restrictions. The computed LRstatistic is compared with chi-squared criticalvalue with 3 degrees of freedom. The LR statisticof 8.952 suggests that the above restrictionscannot be rejected at the 5% significance level.The adequacy of die model is verified by theabsence of serial correlation (Ljung-Box Qstatistic) and heteroscedasticity (WHITE statistic).Moreover, neither ARCH (4) nor GARCH (1, 1)suggests the presence of ARCH effect. Thus, therestricted ESTAR (2) model passes a battery ofdiagnostic tests and thereby can be used as aforecasting model, as its unrestricted version.

The linear AR model is also estimated forthe purpose of forecast accuracy comparison.The estimation of AR model requires that thevariables must be stationary; otherwiseinterpretation from the outcome would bespurious. We employed classical ADF and PPstationarity tests to check whether the stationaryrequirement is met. The stationarity tests resultsare summarized in Table 5. The results in Table5 postulate that instead of yt, which is notstationary, we should estimate linear AR modelfor stationary series, A^ the first difference of yc

Strategically, we need to estimate AR (1) modelfor Ayt, in order to obtain the requiredbenchmark model, namely AR (2) model for yf

The estimated AR (1) model is reported inTable 6.

Tests

ADFPP

Intercept

y<

-0.858-1.244

TABLE 5Stationarity tests

Without Trend

-4.081*^8.657*

results

Intercept with

yt

-0.034-0.493

Trend

Ay(

4.081*-8.627*

Note. * denotes the variable is stationary at 1% significance level.

Series

Ay,

Coefficient of

Ay^ (t statistic)

0.024 (-0.96)

TABLE 6Estimated AR (1)

R2

0.857

model

Diagnostic Tests (Marginal Significance Values)

Q(20)

16.919(0.657)

WHITE

2.267(0.757)

ARCH (4)

1.149(0.362)

GARCH

0.037(0.982)



From Table 6, our AR (1) model for Ay is givenby:

= - 0.024 (10)

Diagnostic tests results show that Equation10 is free from all the autocorrelation,heteroscedasticity and ARCH effects. Theadjusted R2 values also suggest that the AR (1)model fits fairly well for the A^ series. To sumup, the estimated AR (1) model is an appropriaterepresentation of first difference series of yt. Theonly shortcoming of this model is that the linearautoregressive parameter is not significant basedon the standard t-test. Equation 10 can berewritten as:

* = 0 .976^ + 0 .024^ (11)

which is exactly the AR (2) specification of yf

the deviation of MYR/JPY rate from its PPPequilibrium position. If we take into account theinsignificance of the autoregressive parameter,Equation 10 would effectively be reduced to:

or its equivalent:

y. • y*

(12)

(13)

Equation 13 is simply the commonly usedbenchmark for the evaluation of exchange rateforecasting models, namely the simple randomwalk (SRW) models. With the availability of twobenchmarks model, in particular the linear AR(2) model (Equation 11) and SRW model(Equation 13), this study proceeds to comparethe forecast accuracy of the unrestricted andrestricted ESTAR (2) models with the benchmark(AR(2) and SRW) models.

FORECAST PERFORMANCE

The models are used to generate out-sampleforecasts and the forecasting performances ofthese models are evaluated. The out-sampleperformance of the estimated forecasting modelsover the forecast horizon of n = 4, 8 and 12quarters during the period 1997:3 to 2000:2 areevaluated based on mean absolute forecast error(MAFE), mean absolute percentage forecast error(MAPFE) and root mean square forecast error(RMSFE). The overall forecasting performancesare reported in Table 7. Generally, all accuracycriteria consistently suggest that all threeforecasting models, namely the unrestrictedESTAR (2), restricted ESTAR (2) and AR (2)models outperformed the SRW model at allhorizons. This implies that both the linear andnon-linear time series models under this studyimprove over the SRW model in the short- and

TABLE 7Comparison of out-sample forecast accuracy

Models

ForecastHorizon

Unrestricted ESTAR (2)

4

Simple Random WalkMAFERatioMAFPERatioRMSPERatio

Linear ARMAFERatioMAFPERatioRMSPERatio

0.721

0.587

0.731

(2) model0.590

0.824

0.886

8

model as0.919

0.838

0.846

12

benchmark0.954

0.890

0.851

as benchmark0.844

0.930

0.891

0.903

0.952

0.885

Restricted ESTAR

4

0.671

0.668

0.804

0.931

0.938

0.975

8

0.880

0.876

0.903

0.958

0.972

0.950

(2)

12

0.934

0.923

0.908

0.979

0.988

0.944

4

0.721

0.712

0.825

AR (2)

8

0.919

0.901

0.950

12

0.954

0.935

0.962


Dew Khim Sen & Ahmad Zubaidi Baharumshah

medium-run.It is interesting to know whether the non-

linear models yield more accurate forecasts thanthe linear model. As a matter of fact, it isrational for one to expect the former to improveupon the latter, since the former is moreinvolved. In view of the fact that the actualanswer to this issue is important to the applicationof time series analysis, this empirical studyproceeds further to contrast the forecastperformance between these two competingmodels directly. The conclusion from thisexercise is that both the unrestricted andrestricted ESTAR (2) models predict better thanlinear AR (2) model on the basis of all accuracycriteria, for forecast horizon equals 4, 8 and 12quarters. For instance, MAFE of unrestricted(restricted) ESTAR (2) model are 0.590 (0.931),0.844 (0.958) and 0.903 (0.979) times smallerthan the MAFE of AR (2) model, for n = 4, 8and 12, in that order. These results areoverwhelmingly supported by the MAFPE andRMSPE criteria. As such, our extra resourcespent on the modelling of ESTAR model is atleast paid off. Perhaps, more importantly, thisstudy has provided evidence that theperformance of ESTAR is superior to its linearcompetitor, the AR model.

CONCLUSION

The empirical performance of foreign exchangerate models has been frequently criticized inrecent years. These critiques come from studiesthat have found exchange rate models predictpoorly out of sample periods. In the foreignexchange market, central banks often intervene,in an effort either to attenuate or to amplifyvariations in the exchange rate. The Ringgit isno exception and this explained partly the poor-out-of sample prediction of the exchange ratemodels in previous studies.

In this study, we demonstrate that theadjustment of the RM/YEN rate is in factpredictable based on time series model. TheSTAR model and AR model both outperformthe random walk in the out sample forecastingat all horizons. The results also show that theSTAR model outperforms the AR model, itslinear competitor. Importantly, we demonstrateformally that the adjustment of the Ringgit to itslong run equilibrium follows a non-linearity path.This suggests that there is a systematic predictablecomponent in the movement of nominal RM/

YEN exchange rate.Two major implications of this finding are:

(1) non-linear ESTAR model should be givenpriority in modelling exchange rate time series;and (2) exchange rate forecasters could rely onthe non-linear model as a more reliableforecasting tool.

ACKNOWLEDGEMENT

This paper is part of the ongoing project ofIRPA 2000 (Project No: 05-02-04-0046) andFundamental Project 2002 (06-02-02-0098S). Theauthors are grateful to the referees for thecomments and suggestions on the earlier versionof the paper. All remaining errors are theresponsibility of the authors.

REFERENCES

AZAU, M., M. S. HABIBULIAH and A. Z. BAHARUMSHAH.

2001. Does PPP hold between Asian andJapanese economies? Evidence using panel unitroot and panel cointegration. Japan and theWorld Economy 13: 35-50.

BAHARUMSHAH, A. Z. and M. ARIFF. 1997. Purchasing

power parity in South East Asian countrieseconomies: A cointegration approach. AsianEconomic Journal 11(21): 141-153.

BAUM, C. F., J. T. BARKOULAS and M. CAGLAYAN. 2001.

Non-linear adjustment to purchasing powerparity in the post-Bretton Woods era. /International Money and Finance 20: 379-399.

CHAN, K. S. and H. TONG. 1986. On estimation

thresholds in autoregressive models. / . TimeSeries Analysis 7: 179-190.

CLEMENTS, M. P. and J. SMITH. 1998. Non-linearities

in exchange rates. Working Paper. Departmentof Economics, University of Warwick.

COAKLKY, J. and A. FUERTES. 2001. Non-parametric

cointegration analysis of real exchange rates.Applied Financial Economics 11: 1-8.

DIEBOLD, F. X. andj . A. NASON. 1990. Non-parametric

exchange rate predictions. / InternationalEconomics 28: 315-332.

EIRTHEIM, 0 . and T. TERASVIRTA. 1996. Testing the

adequacy of Smooth Transition Autoregressivemodels. / . Econometrics 74: 59-75.

GRANGER, C. W. J. and T. TERASVIRTA. 1993. Modelling

Non-linear Economic Relationships, London:University Press, Oxford.

JOHANSEN, S. and K. JUSELIUS. 1990. Maximum



likelihood estimation and inference oncointegration - with application to the demandfor money. Oxford Bulletin of Economics andStatistics 52: 169 - 210.

LIEW, K. S. and M. SHITAN. 2002. The performanceof AICC as order determination criterion inthe selection of ARMA time series models.PertanikaJ. Science (Forthcoming).

LUUKKONEN, R. 1990. Estimating Smooth TransitionAutoregressive models by conditional leastsquares. In ed. On Linearity Testing and ModelEstimation in Non-linear Time Series Analysis.Luukkonen, R. Helsinki: Finnish StatisticalSociety.

LUUKKONEN, R., P. SAIKKONEN and T. TERASVIRTA.1988. Testing linearity against SmoothTransition Autoregressive models. Biometrika75: 491- 499.

MAHAJAN, A. and A. J. WAGNER. 1999. Non-lineardynamics in foreign exchange rates. GlobalFinance Journal 10(1): 1 - 23.

MONTOGOMERY, D. C, L. A. JOHNSON and J. S.Gardiner. 1990. Autoregressive Intergrated MovingAverage Models. Forecasting and Time SeriesAnalysis. 2nd edition. US: McGraw-Hill.

SAIKKONEN, P. and R. LUUKKONEN. 1988. LagrangeMultiplier tests for testing non-linearities intime series models. Scandinavian]. Statist. 15:55-68.

SARNO, L. 2000a. Real exchange rate behaviour inhigh inflation countries: Empirical evidencefrom Turkey, 1980 - 1997. Applied EconomicsLetters 7: 285-291.

SARNO, L. 2000b. Real exchange rate behaviour inthe Middle East: A re-examination. EconomicUtters 66: 127-136.

TAYLOR, M. P. and D. A. PEEL. 2000. Non-linearadjustment, long-run equilibrium andexchange rate fundamentals. /. InternationalMoney and Finance 19: 33-53.

TONG, H. and JL S. LIM. 1980. Threshold Auto-regression, limit cycles and cyclical data. / .Royal Statist Soc. 42(3): 245-292.

TERASVIRTA, T. 1994. Specification, estimation, andevaluation of Smooth Transition Autoregressivemodels./ Amer. Statist. Assoc. 89(425): 208-218.

TERASVIRTA, T. and H. M. ANDERSON. 1993.Characterizing non-linearities in business cyclesusing Smooth Transition Autoregressivemodels. In Non-linear Dynamics, Choas andEconometrics, ed. M. H. Pesaran and S. M. Potter.US: John Wiley.

(Received: 7 March 2002)



Interpretation of Gender in a MalaysianNovel: The Case of Salina

JARIAH MOHD. JANEnglish Department

Faculty of Languages & LinguisticsUniversiti Malaya, 50603 Kuala Lumpur, Malaysia

Keywords: Gender differences, gender stereotype, literary text, speech styles

ABSTRAKKebanyakan karya bertulis yang dianggap kreatif, khususnya dari segi plot, latar dan watak terhasildengan berdasarkan pemerhatian penulis terhadap realiti sekitarannya. Daripada tinjauan,didapati bahawa anggota sesuatu budaya dan masyarakat mempunyai persepsi atau pandanganyang stereotaip tentang tingkah laku lelaki atau wanita dari segi gaya berpakaian, aktiviti, kerjayadan peranan sosialnya dalam masyarakat. Penulis yang berpegang kepada pandangan ini biasanyaturut membuat pemerhatian secara umum dan stereotaip, serta mencurahkannya ke dalam karyamereka khususnya melalui watak-watak ceritanya agar karya mereka boleh dianggap logik,realistik, dan berkesan. Kajian ini bertujuan untuk menganalisis gaya pertuturan watak lelaki danwanita yang bercorak stereotaip yang tergambar melalui novel Salina karya A. Sam ad Said. Disamping itu, penelitian juga dilakukan terhadap dampak daripada perbezaan gender yangdipaparkan dalam novel ini. Dari aspek fitur linguistik, penggunaan bentuk sopan dan bentukberkias turut diteliti untuk analisis ini. Analisis akan difokuskan kepada penggunaan fiturlinguistik ini oleh watak-watak dalam novel yang dikaji. Dapatan kajian ini menunjukkan bahawapenggunaan fitur linguistik ini berbeza mengikut gender dan unik sebagai ciri masyarakatMalaysia semula jadi.

ABSTRACT

Most written products that are considered creative, particularly in terms of plots, contexts andcharacters are created based on the writers* observations of the reality of their surroundings. Ithas been observed that members of most cultures and societies have certain perceptions orstereotype views of the behaviour of men and women regarding styles in dressing, activities, careerand social roles in society. Writers rely on these usually stereotyped and generalized observationsand portray these traits in their writing especially in their characters in order to make theirwritten products believable, real and effective. This study sets out to analyse the stereotypedcharacteristics of male and female speech styles as portrayed in a Malaysian novel, Salina, writtenby A. Samad Ismail. It also examines the effect of gender differences depicted in the novel. Thelinguistic features that are examined for the analysis are the use of super polite forms and hedges.The analysis focuses on the use of these features by the characters in the novel. The findings ofthe study suggest that there are gender differences in the use of these features and are uniquelyMalaysian in nature.

INTRODUCTION conforming to the typical behaviour of their sex

There have been numerous hypotheses formed &OUP- T h i s m i S h t involve characteristics ofon the relationship between people's general gender-biased costumes, activities, areas ofexpectations and beliefs about men and women expertise and social roles in society as well asand the behaviour of members of these groups differences in ways of communicating.(Worrell 1978; Taylor and Hall 1982). In most Our constructs of masculinity and femininitysocieties, men and women see themselves as influence our reactions to men and women.

Jariah Mohd. Jan

The moment we have categorized someone as amale or female, 'our gender constructs comeinto play, providing us with judgmental standardsagainst which to compare persons in formingimpressions about their masculinity andfemininity' (Smith 1985:28). Thus, in a particularsociety, men and women conform to the acceptedand desired characteristics, as society'sexpectations and perceptions shape theirbehaviour in that society.

Various researchers have pointed out thatthere are differences in ways of communicatingbetween males and females (Lakoff 1975;Spender 1975; Coates 1986; Tannen 1991; JariahMohd. Jan 1999). Most empirically based studiesof differences between male and female speechstyles have been dealt with in the naturallyoccurring or 'real life' conversations in the publicsphere (Fishman 1983; Coates 1986; Jariah Mohd.Jan 1999). There has been very little evidence ofstudies on gender conversational styles in thewritten text particularly in novels, short stories,plays or screenplays. This study intends toexamine male and female conversational styleswith a focus on speech manifested in thelanguage of the characters in the novel. Thefocus of attention is to explore whether thesemanifestations are indeed the writer's ownperceptions and experience.

A general observation indicates that mostwritten products that are considered creative,particularly in terms of plots, contexts andcharacters are created based on the writers'observations of the reality of their surroundings.It has been observed that members of mostcultures and societies have certain perceptionsor stereotype views of the behaviour of men andwomen regarding styles in dressing, activities,career and social roles in society. Writers rely onthese usually stereotyped and generalizedobservations and portray these traits in theirwriting especially in their characters in order tomake their written products believable andeffective. For instance, the stereotyped view ofgender characteristics in the society is portrayedby the differences in ways of communicatingbetween men and women. As such, the speechstyles that are associated with either male orfemale characters are often manifested by thewriters in their stories in order to make theircharacters real and acceptable to the readers.

OBJECTIVE

This paper attempts to analyse the stereotypedcharacteristics of male and female speech stylesand to examine the effect of gender differencesas depicted in a novel, Salina, written by A.Samad Ismail who is a renowned Malaysianliterary writer.

GENDERED BEHAVIOURAL CONSTRUCTSAND CONVERSATIONAL STYLES

In any society, men and women vary in thedegree to which they see themselves asconforming to gendered typical norms ofpersonality and behaviour. The cultural normsthat are regulated in a society according to thesocial desire of that society itself are referred toas behavioural construct. These norms includegender (sex) roles in society, occupation, andlanguage use.

Frank and Ashen (1983) note that languagedifferences are partly due to our socialexpectations. For instance, since mostbehavioural constructs in terms of cultural normsare regulated in accordance with the values ofmen, we tend to respond to boys and girlsdifferently. Our level of acceptance of theirbehaviour may differ. For example, it is generallyaccepted for men to be dominant, aggressive,independent and objective. Women, on theother hand, are more likely to be accepted assubmissive, dependent, passive and subjective.As such, the fact that women are expected to(act as ladies' and to 'respect' people aroundthem reflects their inferior status, and thus, areexpected to respect their superiors who are men.These behavioural constructs have empoweredmen while women are considered as one stepbelow men or rather they hold a secondarystatus in the society. This encoding behaviour ofmen and women in the society has significantramifications on the way they communicate andsocialize in inter-gender interactions.

Inevitably, our social expectations andexperiences may be a contributing factor to thelanguage differences or differing linguistic stylesbetween men and women at any time and in anysituation.

There have been numerous studies thatappear to support the idea of stylistic genderdifferences in conversation. LakofFs observations(1975 and 1977) of women's language are


Interpretation of Gender in a Malaysian Novel: The Case of Salina

considered one of the earliest regarding thedifferences in speech styles between men andwomen.

According to Lakoff, the style of languagewhich was typically used by women and thusdenied them 'the means to express (themselves)strongly', and instead make them sound 'trivial'and 'uncertain' were as follows:i. The use of words connected with women's

culture (colour terms, sewing terms) whichare not in men's vocabulary, or are useddisparagingly by men.

ii. So-called "empty" adjectives, such as "divine"and "charming."

iii. Tag questions and rising intonation usedwith grammatically declarative utterances,

iv. The use of hedges to avoid being tooassertive or direct,

v. The use of "so" as an intensifier, as in "Ilike him so much."

vi. Hypercorrect forms in pronunciation andgrammar.

vii. Being "superpolite."viii. Not telling jokes,ix. Speaking in italics (which presumably refers

to women's stress patterns).(LakofT 1975:53-56)

Lakoff was working on the principle thatwomen's speech patterns are worse than men's,in that they confirm women's subordinate socialstatus and prevent them from being treated asequals. Women are interested in maintainingharmony and goodwill in their communicationand sometimes can be misunderstood as beingtentative and full of uncertainties. The use ofmodal auxiliaries in women's speech is said toreflect this perception.

In 1977, LakofT further characterized herbasic assumptions of women's speech styles under

three (3) main categories, namely lexical traits,phonological traits and syntactic-pragmatic traits(see Table 1).

This study will only focus on two linguisticfeatures proposed by Lakoff (1977) in the analysiswhich are the use of super polite forms andhedges.

POLITE FORMSAccording to Holmes (1995), 'politeness' refersto an expression of concern for the feelings ofothers. People may express concern for others'feelings in many ways, both linguistic and non-linguistic. Being polite means expressing respecttowards the persons you are talking to and notoffending them. In other words, politeness maytake the form of an expression of goodwill orcamaraderie, establishing rapport, as well as themore familiar non-intrusive behaviour which islabelled 'polite' in everyday usage.

Malaysians observe a politeness system thatembodies specific codes of verbal and non-verbalbehaviour in their interactions with others(Jamaliah Mohd. Ali 1995b: 65). They aregenerally receptive to those who display goodmanners in face-to-face interaction. In the Malaycommunity, people are expected to be polite ordemonstrate finesse in behaviour whenever theyinteract with others. This is evident especially ininteractions that involve persons of a certaininstitutional or societal rank. Such behaviourindicates proper upbringing. In addition, thenotion of relational distance, respect for elders,hierarchical position as well as status andauthority are very much rooted in the Malayvalue system (Jariah Mohd. Jan 1999: 206).

Some aspects of behaviour that Malaysiansconsider finesse are as follows (Asmah Haji Omar1992: 23-24):

TABLE 1A summary of the characteristics of women's speech styles (Lakoff 1977)

No. Female Speech Styles Speech Characteristics

1. Lexical Traits

2. Phonological Traits

3. Syntactic-Pragmatic Traits

special lexiconimprecise intensifierssuper polite formshypercorrect grammarclear and precise pronunciationepistemic modality and hedgestag questionsdirect quotations


Jariah Mohd. Jan

• Not being forthright or assertive oraggressive;

• Not being blunt or direct in expressingone's views;

• Not causing interpersonal conflict oravoiding ioss of face'.

When someone is being polite, there aremany possible features of their use of thelanguage that are being referred to. It isgenerally observed that polite people tend tophrase their utterances considerately, respondencouragingly and positively to others' talk, andexpress appreciation generously. They avoidconfrontation, public displays of being tooforceful or direct because these are all examplesof behaviour which are regarded as impolite,coarse, rude, and may even suggest ill-breeding(Jamaliah Mohd. Ali 1991).

Lakoff (1975) observes that in women'sspeech, there is also a more frequent expressionof emotions such as love, endearment, and grief,and avoidance of angry and hostile expressions.In doing so, women tend to use super politeforms and euphemistic forms such as 'would youmind... \ Td aprreciate it if...\ ' ... if you don9tmind.', 'may' and 'could'.

Face

Jamaliah Mohd. Ali (1995b: 75) stipulates thatthe norms of society require that we abide bysocial rules in our daily interactions. 'Face' or'jaga air muka' is one of the most importantfactors that Malaysians should observe and adhereto in their daily interaction in order to maintainnot only the stability of the interactants but thatof the interaction as well.

In Malaysian communities, 'face-saving'means saving another person or oneself, fromembarrassment. In other words, the speakertries not to put the listener or himself in aposition in which he or she might beembarrassed. The 'face' that a person maintainsbecomes important especially when it is subjectedto risk and assumes significance particularly whenevents are being interpreted and evaluated.Therefore, it must be constandy attended to inthe interaction.

Ostman (1981: 4) states that there aredifferences between 'face-saving' and politeness.She explains that the 'face-saving' aspect of verbaland (non-verbal) behaviour is characterised byan egocentric, direct and efficient expression of

wants and needs whilst the politeness aspectworks at suppressing such potential egocentricityand prevents it from occurring overtly inbehaviour.

In general, people try to build up eachother's 'positive face' (the positive image thatpeople have and want to be appreciated andapproved by others) and avoid posing threats tothe 'negative face' (where one's actions shouldnot be impeded by others) in co-operativeinteraction.

Behaviour which avoids imposing on othersor avoids 'threatening their face' is described asevidence of negative politeness, while sociablebehaviour expressing warmth towards anaddressee is positive politeness behaviour (Brownand Levinson 1987). According to this approach,any utterance which could be interpreted asmaking a demand or intruding on anotherperson's autonomy can be regarded as a potentialface-threatening act (henceforth FTA). Politepeople avoid obvious FTAs, such as insults andorders. They usually attempt to reduce thethreat of unavoidable FTAs such as requests orwarnings by softening them, or expressing themindirectly; and they use positively politeutterances such as greetings and complimentswhere possible.

The mutual relationship between theparticipants in an interaction is very significantin deciding the degree of face threat. Forinstance, something that is fairly minimally face-threatening in private encounters, such as afriendly talk, becomes highly face-threatening inpublic encounters. Competent conversationparticipants often look for ways that can minimisethe threat to face in one way or another. Thisof course requires an ability to reach each other'ssignals and to accommodate each other's wants.Furthermore, preserving each other's face insocial interactions is very important inmaintaining harmony and stability. Failure todo so may result in undesirable communicativeconsequences. As such, this phenomenon attimes imposes further constraints on theconversational behaviour of the participants.

HEDGING - A STRATEGY

Hedging refers to the linguistic forms whichspeakers use to express their degree ofcommitment to the truth of a proposition. Theforms that realise speakers' commitment aretypically modal auxiliary verbs such as 'should',



'would', 'could) ' may 'and 'might'. Other linguisticforms that fulfil the same function are itemssuch as 'perhaps ', ' really \ and also hedges such as'sort of7, 'I think' and 'kind of.

According to Lakoff (1975: 53), hedges are"...words that convey the sense that the speakeris uncertain about what he or she is saying orcannot vouch for the accuracy of the statement."Speakers use hedges as a strategy to expresstheir personal attitude in the conversation. Onthe other hand, Coates defines hedges by thefunction of items, rather than their grammaticalcategory. According to Coates (1989: 113):

"Epistemic modal forms are definedsemantically as those linguistic forms whichare used to indicate the speaker's confidenceor lack of confidence in the truth of theproposition expressed in the utterance. Ifsomeone says, Perhaps she missed the train, theuse of the word perhaps indicates the lack ofconfidence in the proposition 'she missedthe train'. Lexical items such as perhaps, Ithink, sort of, probably, as well as certainprosodic and paralinguistic features, are usedin English to express epistemic modality."

Coates (Ibid: 113-14) also extends thefunctions usually associated with epistemic modalforms as those that are "used by speakers notjust to indicate their lack of commitment to thetruth of propositions, but also to hedge assertionsin order to protect both their own andaddressees' face."

She claims that women, especially in single-sex groups, exploit these forms more than menand they use them to mitigate the force of anutterance in order to respect the addressees'face needs. An illustration of this point is asfollows:

(speaker describes an old friend she'srecently bumped into)she looks very sort of urn (-) kind of matronly Ireally

(Coates 1989)

She claims that the italicised forms in theexample given "hedge the assertion she looksmatronly not because the speaker doubts itstruth but because she does not want to offendher addressees by assuming their agreement"because describing a friend in unflattering termsis controversial, politeness as well (Coates Ibid).

The suggestion here is that the instances ofepistemic modal forms above have one functionof protecting the face of the speaker's addresseesby permitting their dissent from the truth contentof her proposition. However, according to Coates,the polypragmatic nature of forms also protectsthe speaker's face. In this instance, the speakerin the example can retreat from the propositionexpressed if it turns out to be unacceptable.She emphasised that "Presumably such topics donot trigger the use of epistemic modal formsbecause they are not so face-threatening". Inaddition, hedges also function as politenessstrategies often used by women in theirinteraction.

METHODOLOGY

The Malaysian novel, Salina written by A. SamadSaid who is a renowned Malaysian literary writerwas selected for this study. For the purpose ofanalysis, a translated version of Salina by HawaAbdullah was used. Unlike many other translatedversions, this is close to the original Malay version.

Salina is regarded as the line of demarcationwhich separates the early writers from the presentones (Hanna Sheikh Mokhtar 1993). The styleused by the author differs from other writers inthat it is told by means of dialogues orconversations among the characters, reflectingthe social lifestyles and behaviour practiced bythe society at that time. The language use in thedialogues direcdy or indirectly reflects the socialexpectations as depicted in the cultural normsand practices of the society.

The story depicts the struggles faced bywomen in order to survive in the male-dominatedworld. It could be said that women in Salinarepresent a new breed of women who weremodernised by colonial rule and were toughenedby Japanese occupation.

Five excerpts from Salina that featured theconversations of male and female charactersportrayed by the writer were selected at random.The excerpts were randomly selected to avoidinclination towards the existence of the featuresof the speech styles if they were studied andanalysed beforehand.

Analysis of the male and female speechstyles in the selected excerpts is based on twolinguistic features stipulated by LakofF (1975)i.e. i) the use of super polite forms and ii)hedges. Examples were extracted and quoted


Jariah Mohd.Jan

from the excerpts as evidence of the use of thelinguistic features by the characters.

Aspects of the phonological traits of thediscourse that include intonation and voice pitchwere not included in the analysis. Table 2summarizes the speech characteristics in thestudy.

ANALYSIS AND FINDINGS

The Use of Super Polite FormsIn general, analyses of the data indicate thatmost of the characters in Salina are polite toeach other. The degree of politeness andimpoliteness is difficult to measure as this variesfrom culture to culture.

Although politeness in the characters' speechis not clearly illustrated in terms of the use ofspecial words, or phrase, it is found that boththe men's and women's speech styles areconsistent with Holmes' descriptions of politenessthat is "polite people tend to phrase theirutterances considerately, respond encouraginglyand positively to others' talk, pay compliments,and express appreciation generously" (1995, 24-

25). This is a typical feature of the Malaysianculture that "... when talking, we need tomoderate our voice, and not talk in a way thathurt others' feelings" (Shasel 1997). Except fortwo instances of the use of please, there is noother evidence of the use of other polite formsin speech among the female characters.

• Idah, Please sit down ... p 283• Please light it, my hands are wet. p 212

The Use of Expletives

Analyses of the data show that expletives areused among male characters in Salina. Thefindings show that there is only one instance ofstrong expletive attributing to male style. Theweak forms are those used by the male charactersand these forms are attributed to the femalestyle (Table 3).

The examples shown in Table 3 areutterances made by Abdul Fakar who is one ofthe male characters in Salina. His languagereflects his flirtatious behaviour. He speaks gently

TABLE 2Examples of linguistic features examined in Salina

No. Female Speech Styles Speech Characteristics Examples

1. Lexical Traits

Syntactic-PragmaticTraits

Super polite forms

Expletives• Strong expletives• Weak expletives

Epistemic modality and hedges• Modal auxiliary• Modal adverbs• Qualifying adjectives• Egocentric sequences

• Would you mind ...,• Could you please...,

• Damn, I'll be damn, Shit,Hell• My goodness,...• Oh, dear,...

• May, might, could• Perhaps, possibly, may be• So, very,• / believe, I think, I guess

TABLE 3Strong and weak expletives used by male characters in Salina

Strong Expletives Weak Expletives

That wretched tramp! He doesn't know his place . . .p 76 If you become dead wood in the house, goodness, the tramp willtake advantage... p 321

My! Oh my! p 324

Lo and behold after that, the clerk disappears for months.p 325

Very well! You say it is difficult ... p 79



and appears to understand women very well.The linguistic features used by Abdul Fakar

indicate that his speech lacks male characteristics.As such, except for the strong expletivementioned earlier, all the other expletives usedare weak and considered to have feminine style.

Other expletives found in the excerpts are:

Oh dear, you shouldn 't have troubled yourself, sis

... p 47

Oh! Such a pity! p 198

Praise be to God, p 198

Another male character in Salina who uses theseweak forms of expletives is Hilmi. He is portrayedas an educated person, as well as polite andrespectful to others. His religious backgroundand good upbringing influence his speech wheninteracting with others.

It appears that the female characters usemore weak forms of expletives compared to the

male characters in Salina. Forms of endearmentsuch as 'dear' and 'my sweet* are widely used inthe text as indicated in Table 4.

The Use of Hedges

There are seven types of hedges used by bothmale and female characters in Salina. The use ofthis feature is tabulated in Table 5. Contrary tothe assumptions that hedges are exclusively afemale speech style, there is considerableevidence of the use of this feature in malespeech as found in the selected excerpts fromthe novel, Salina.

The data in Table 5 shows that in contrastwith the general expectations based on theperceptions of the gender speech styles, themale characters have been portrayed to usemore hedges compared to the female characters.This finding on the use of hedges, contradictsthe claim made by LakofT (1975) and otherresearchers (Coates 1986; Githens 1991; Jariah

TABLE 4The use of polite forms and expletives in Salina

Female Speech Styles Male Speech Styles

Dear me! Exercising so early in the morning? p 36

Goodness! What a hopel p 39

How long is it since he passed away ? p 45

Goodness, it s been going on for so long, how could you notknow} p 75

Goodness! Now you say: Praise be to God. p i98p 9 0

/ am fond of you, my sweet, p 210

Idah, Please sit down ... p 283

Please light it, my hands are wet. p 212

Dear me! If it is as you say, the landlord s simply squeezingmoney from us. p 40

Oh dear you shouldn't have troubled yourself sis ... p 47

Oh dear! Why swear? p 86

Goodness! Tonight, I'll give you a treat, p 89

Good Gracious! You embarrass me, of course I have the money.

Dear me. what a shame! p 247

Oh my sweet. You are clever to talk, aren't you? p 288

How clever you are, my sweet, p 289

Oh dear, you shouldn't have troubled yourself sis ... p 47

Oh! Such a pity! p 198

Praise be to God, p 198

If you become dead wood in the house, goodness, the tramp willtake advantage... p 321

My! Oh my! p 324

Lo and behold after that, the clerk disappears for months.p 325

Very well! You say it is difficult ... p 79


Jariah Mohd. Jan

TABLE 5Gender and distribution of hedges in Salina

No. Types of Hedges Male Female Total

A1.2.3.

Modal Adverbsreallyperhapssurely

1597

231413

Total 19 50

B1.2.3.4.

D1.2.3.4.5.6.

Modal Auxiliarywouldshouldcouldmust

EgocentricI knowI do / I don'tI thinkI hopeI wantI feel

156310

16248462

5222

6151000

208512

c1.2.

Adjectivestooso

Total

Total

34

623

29

11

18

9

45

731

38

22399462

Total 60 22 82

E1.2.3.

Fillerswellof courseyou knowTotal

1381132

65617

19131749

Total (A+B+C+D+E) 186 78 264

Mohd. Jan 1991) that women use these featuresmore than men do.

As can be seen in Table 5, the malecharacters in Salina use more hedges, especiallythe egocentric types such as / do/1 don't (24items) and / know (16 items), compared to thefemale characters. The assertive behaviour ofthe male characters is reflected in their use ofhedges (60 items). On the other hand, thefemale characters appear to be less assertive (22items) and seldom relate their needs and truefeelings about matters that are of importance tothem.

CONCLUSIONThe findings of the study suggest that there aregender differences in the use of the super polite

forms and hedges. The speech styles portrayedin Salina exhibit clearly features of male andfemale language. However, unlike the previousclaims (LakofT 1975; Tannen 1990; Githens 1991;Jariah Mohd. Jan 1999), these linguistic featuresseem not to be exclusively used by one genderonly. Both the male and female characters inthe novel are portrayed as using them. Althoughthe speech characteristics in this study havebeen claimed as associated with female speechstyle, the results of this study show that it is theparticular requirement of the conversations thatmotivate the use by either a male or a femalecharacter. It is suggested that the occurrence ofthe styles portrayed in the male and femalespeech is not determined or influenced by thegender of a speaker, rather, the context itself



has a strong bearing on the choice of styles to beused.

In most studies, claims of gendered speechstyles have been made based on the westernperceptions. In Salina, however, both men andwomen characters have been pictured to haveused similar linguistic features in theirconversations. The characters are also sensitiveto the behaviour of their conversational partnersand thus adapt their speech style accordingly.

The findings suggest that the speech stylesof male and female cannot be generalised forthey differ from one cultural context or societyto another. However, it is reasonable to concludethat writers, regardless of their culture andsocietal background, appear to perceive orconceptualise male and female speech as acollaborative social activity rather than beingrestricted to their personal traits. As such, whenthey write the speech for their characters in anovel, they take into account aspects such as therole and status of the characters within theparticular society and the context of theconversation rather than gender.

Perhaps it would be best to conclude thatthe perception of the different speech stylesbetween men and women lies in our own beliefsand perceptions. To a certain extent, we tendto exaggerate perceived differences in the verbalstyles of men and women because we live in anenvironment that stresses differences rather thansimilarities.

REFERENCESA. SAMAD SAID. 1991. Salina (translated by Hawa

Abdullah). Kuala Lumpur: Dewan Bahasa danPustaka.

ASMAH HAJI OMAR. 1992. The Linguistic Scenery in

Malaysia. Kuala Lumpur: Dewan Bahasa danPustaka.

BROWN, P. and S. LEVINSON. 1987. Universal inLanguage Usage: Politeness Phenomena. InQuestions and Politeness : Strategies in SocialInteraction, ed. E. N. Goody, p. 58-289.Cambridge: Cambridge University Press.

COATES, J. 1986. Women, Men and Language : A

Sociolinguistic Account of Gender Differences inLanguage. 2nd Edition. London: Longman.

COATES, J. 1989. Gossip Revisited : Language in All-Female Groups. In Women in Their SpeechCommunities, ed. J. Coates and D. Cameron.London: Longman.

FISHMAN, P. 1983. Interaction: The Work WomenDo. In Language, Gender and Society, ed. B.Thorne, C. Kramarae and N. Henley, p. 89-101. Cambridge: Newbury House Publishers.

FRANK, F. and F. ASHEN. 1983. Language and the Sexes.Albany: University of New York Press.

GITHENS, S. 1991. Men, Women in Conversation :An Analysis of Gender Styles in Language.(On-line) Available: http//www.george-town.edu.bassr/githens/ottol922.htm

HANNA SHEIKH MOKHTAR. 1993. The Portrayal ofwomen in A Samad Said's Salina. Unpublishedthesis, UPM, Serdang, Selangor.

HOLMES, J. 1995. Women, Men and Politeness. London:Longman.

JAMALIAH MOHD. ALL 1991. The Academic Seminaras an Arena for Cognitive Conflict and ConflictResolution : A Malaysian Context. WorkingPaper No. 9. Project on Culturally ConditionedModels of Conflict Resolution. Gothenburg,Sweden: Centre for East and South East AsianStudies, University of Gothenburg, Sweden.

. 1995a. The Pragmatics of Cross-CulturalCommunication in a Malaysian Context. InRules of Speaking: Verbal Interaction at Play, ed.Zainab Abdul Majid and Loga MahesanBaskaran, p. 112-124. Pelanduk Publications.

. 1995b. Malaysian student seminar : Astudy of pragmatic features in verbalinteraction. Unpublished Ph.D Thesis,University of Malaya, Kuala Lumpur.

JARIAH MOHD. JAN. 1999. Malaysian talk shows : Astudy of power and solidarity in inter-genderverbal interaction. Unpublished PhD Thesis,University of Malaya, Kuala Lumpur, Malaysia.

LAKOFF, R. 1975. Language and Women's Place. NewYork: Harper and Row.

. 1977. Language and Women's Place. NewYork: Harper and Row.

OSTMAN, J. 1981. You Know. A Discourse FunctionalApproach. Amsterdam: John Benjamins.

SHASEL, P. 1997. Restore Malay/Islamic Courtliness.[Online] Available: http://www.ioi.ie/~/afifi/BICNews/shasel/shasel3.htm

SMITH, P. 1985. Language, The Sexes and Society.Cornwall: Basil Blackwell.


Jariah Mohd. Jan

SPENDER, J. and R. HELMREICH. 1975. Masculinity and TAYLOR, M. and J. HALL. 1982. Psychological

Femininity: Their Psychological Dimensions, androgyny: Theories, methods and conclusions.Correlates and Antecedents. Austin, Texas: Psychological Bulletin 92(2): 347-66.

University of Texas Press. WORREL, J. 1978. Sex roles and psychological well-

TANNEN, D. 1990. Gender Differences in Topical being : Perspectives on methodology. JournalCoherence: Creating Involvement in Best of Consulting and Clinical Psychology 46(4):Friend's Talk. In Conversational Organisation 777-91.and Its Development, ed. B. Dorval. Norwood, . , , o n / 1 . .XT i\_* (Received: 26 June 2002)New Jersey: Ablex. J

. 1991. You Just Don't Understand : Womenand Men in Conversation, London: Virago.

1 5 2 PertanikaJ. Soc. Sci. & Hum. Vol. 10 No. 2 2002


Penterjemahan Pragmatik dalam Konsep Masa Arab-Melayu:Satu Analisis Teori Relevan

MUHAMMAD FAUZI BIN JUMINGANJabatan Bahasa Asing

Fakulti Bahasa Moden & KomunikasiUniversiti Putra Malaysia

43400 UPM, Serdang, Selangor, Malaysia

Kata kunci: Teori relevan, pragmatik, penterjemahan Arab-Melayu, konsep Masa

ABSTRAK

Masa itu emas. Masa itu seperti pedang. Begitulah konsep masa diungkapkan dalam budayaMelayu dan Arab bagi melambangkan pentingnya konsep masa itu. Dalam konsep masa ini,bangsa Melayu sering dikonotasikan negatif. Sering kedengaran ungkapan Janji Melayu yangmelambangkan bangsa Melayu tidak begitu menitikberatkan kepentingan masa. Adakah perkaraini benar? Justeru, makalah ini cuba merungkai permasalahan konsep masa melalui pendekatanpragmatik, khususnya Teori Relevan (TR). Dengan menggunakan faktor-faktor bukan linguistik(ekstralinguistik) seperti faktor konteks, latar situasi, dan niat pengujar, TR dijangka mampumenghuraikan permasalahan ini secara tuntas. Dengan menggunakan data kata pinjaman Arabyang sudah diterima pakai dalam bahasa Melayu, khususnya melalui penterjemahan dan adaptasi,konsep masa Arab-Melayu ini akan cuba dibincangkan.

ABSTRACTTime is gold. Time is like a sword. Both these proverbs are used in Malay and Arabic culture forshowing the significance of time. Unfortunately, the concept of time in Malay culture has beenpresented negatively. The Malay race is regarded as a ethnic group that not take care about theimportance of punctuality. Is it true? This article will try to explain that problem by using apragmatic analysis especially through Relevance Theory (TR). Based on the TR concept thatengaged factors such as context, co-text, and intended meaning, TR is capable of solving theproblem of this misconception in Malay culture. In this articles there are some data that are usedto discuss some foreign lexical items that were assimilated in Malay vocabulary. Actually, thechosen lexical items were adapted and translated into the Malay language.

PENGENALANMasa itu emas. Masa itu kehidupan. Begitulahbeberapa ungkapan orang Melayu tentang peripentingnya masa dalam kehidupan mereka.Malah dalam bahasa Arab juga terdapat madahpujangga yang mengungkapkan masa dalamkehidupan mereka. Antaranya seperti Waktu ituseperti pedang; kalau tidak digunakan, pedang ituakan memenggal kepala kita. Malah dalam Islampun, kepentingan masa terakam dalam satu surahkhas yang menceritakan kepentingan masa, iaitusurah al-'Asr. Keprihatinan tentang masa inisebenarnya melambangkan pemikiran masyarakatyang menuturkan bahasa tersebut. Dengan kata

ini, ungkapan konsep masa sebenarnyamelambangkan gaya pemikiran dan bahasasesuatu bangsa (Reha Mustafa 1999).

Sebenarnya, ada banyak kajian dan tulisanyang membincangkan konsep masa khususnyakajian ilmiah yang dibuat oleh pengkaji tempatanseperti Asmah Hj Omar (1994), Reha Mustafa(1999) dan Nor Hashimah Jalaluddin (2001).Kajian mereka dibuat dari perspektif ilmulinguistik dengan cabang utamanya seperti nahu,sosiolinguistik, semantik, dan pragmatik.Bagaimanapun belum ada kajian yang dibuatdalam konteks penterjemahan. Dalam disiplinterjemahan, konsep masa diungkap dengan cara

Muhammad Fauzi bin Jumingan

yang agak berbeza sedikit kerana penterjemahanmelibatkan dua bahasa dan budaya yang berbeza.Oleh itu, makalah ini mencuba untuk menganalisiskonsep masa ini dengan beracukan satu teoriterjemahan moden, iaitu teori relevan (TR).

PERNYATAAN MASALAH

Sesuatu kajian yang benar-benar ilmiah mestilahdilakukan dengan berdasarkan satu kerangkakerja atau teori yang mapan. Tanpa kerangkakerja ini, kajian itu bersifat awang-awangan. Olehkerana aspek masa melibatkan bahasa danpemikiran sesuatu bahasa dan bangsa makabaiknya kajian itu dianalisis dari sudut kerangkateoritis (Nor Hashimah Jalaluddin 2001).Misalnya ungkapan berikut :

Kes 1Ahmad : Bila engkau nak beraya di rumah aku,

Salim?Salim : Lepas Maghrib nanti.

Dalam kes 1 di atas, perkataan Maghrib yangberasal daripada bahasa Arab adalah waktu yangbermula selepas gelincir matahari di kaki langitdan tamat apabila warna merah di ufuk barathilang kesemuanya. Atau dengan kata lain, waktuMaghrib tamat apabila masuk waktu Isyak.Manakala dalam budaya Melayu, ungkapan selepasMaghrib itu mungkin selepas waktu Isyak. Sudanmenjadi kelaziman orang Melayu, sering'terlewat' dalam mengotakan janji waktunyahingga tercipta ungkapan Janji Melayu (NorHashimah Jalaluddin, 48). Memandangkanperbezaan makna masa dalam kedua-dua bahasa,budaya, dan pemikiran itu tadi berbeza makaTR cuba menganalisis interpretasi dan maksudsebenar ungkapan tersebut dan cara terbaikuntuk menterjemahkannya. Lazimnya interpretasiujaran itu akan dianalisis menurut perspektifilmu pragmatik melalui TR. Menurut TR,ungkapan itu mengandungi beberapa andaian.Antaranya seperti :a. waktu itu betul-betul selepas sembahyang

Maghrib, iaitu sebelum waktu Isyak.b. waktu itu mungkin selepas Isyak.c. waktu itu mungkin selepas Maghrib, tetapi

hari yang berikutnya.

Kalau konteks menentukan bahawa Salimini jenis orang yang menepati janji makaungkapan Salim akan menghasilkan kesan

konteks, iaitu Salim akan datang tepat padawaktunya, betul-betul selepas Maghrib dan bukanselepas Isyak. Dan kalau sebaliknya maka andaianb) dan c) akan menjadi makna pada semantiknyasahaja, iaitu makna sebenar Maghrib itu tetapijuga berdasarkan konteks ia diujarkan. BegitulahTR menganalisis sesuatu ujaran secara deduksibagi menghasilkan inferens (kesimpulan).Justeru, satu interpretasi tepat dapat dihasilkan.

Oleh itu kajian ini cuba mengkaji masalahseperti:1. Mencari maksud sebenar sesuatu ujaran yang

mengandungi konsep masa dari perspektifbahasa, budaya, dan pemikiran orang Melayudengan beracukan pendekatan TR.Gabungan ilmu semantik dan pragmatikpenting dalam mencari makna sesuatuujaran. Kebergantungan terhadap ilmusemantik sahaja tidak memadai.

2. Menterjemah konsep masa yang dipinjamdaripada bahasa Arab dengan menggunakanpendekatan terjemahan langsung (directtranslation) seperti yang dicadangkan olehTR. Pendekatan terjemahan langsungdianggap lebih selamat kerana konsep inimengekalkan kesemua aspek teks sumberseperti konteks, kesan konteks dan stilistik.Tambahan pula, kejujuran dan kesamaanterhampir dapat dikekalkan dalampendekatan ini. Oleh itu, ketepatan dalamterjemahan dapat dihasilkan melaluipendekatan terjemahan langsung.

OBJEKTIF KAJIAN

Antara objektif kajian ini ialah:1. Melihat konsep masa dari perspektif bahasa,

budaya, dan pemikiran orang Arab.2. Meninjau penggunaan konsep masa dalam

pemikiran dan budaya orang Melayu,3. Mencadangkan pendekatan terjemahan

langsung dalam menterjemahkan konsepmasa yang dipinjam daripada bahasa danbudaya asing khususnya bahasa Arab.

BATASAN KAJIAN

Kajian ini hanya melibatkan kata masa yangdipinjam daripada bahasa Arab dan konsep masaini telah diterima pakai oleh orang Melayu.Kebanyakan kata pinjaman itu melibatkan istilahmasa dalam agama Islam. Antara yang akandianalisis adalah seperti istilah al-Fajr (Fajar),al-Subh (Subuh) dan al-Sahr (sahur).


Penterjemahan Pragmatik dalam Konsep Masa Arab-Melayu: Satu Analisis Teori Relevan

KERANGKA KERJA

Kerangka kerja atau metodologi yang akandigunakan adalah bersifat kajian kepustakaan(arm chair research). TR akan digunakan dalamkajian ini kerana TR menekankan aspekkerelevanan sesuatu ujaran. Pendengar (dalamkonteks terjemahan dipanggil pembaca kedua)diberi kebebasan untuk berfikir dalammentafsirkan sesuatu ujaran yang didengar atauditerjemahkan. Hal ini bersesuaian dengan carabaru berkomunikasi yang menyatakan bahawamanusia ialah insan istimewa yang mampuberfikir dalam mentafsirkan sesuatu ujaranberdasarkan persekitaran kognitif dan konteksyangada (Sperber dan Wilson 1995). Kemudian,TR akan diterapkan dalam beberapa katapinjaman masa bagi dianalisis maksud sebenarnyadalam sesuatu ujaran orang Melayu selainmenterjemahkan data pinjaman masa itu secaralangsung.

DEFINISI KONSEP

Dalam kajian ini, terdapat beberapa istilah kosailmu yang perlu dijelaskan terutamanya konsepdan pengertian TR. Antaranya adalah sepertiberikut:

Konsep Masa

Menurut Kamus Dewan (1992), masa bermaksudwaktu dan ketika. Waktu itu mungkin sekejap,agak lama, lama atau tertentu sahaja. Perkataanmasa berasal daripada bahasa Sanskrit yangbermaksud bulan. Manakala perkataan waktupula berasal daripada bahasa Arab, iaitu waqt(Reha Mustafa 1999). Sebenarnya sukar bagikita untuk mendefinisikan makna masa keranaia bersifat relatif dan subjektif. Masa ialah suatukontinum yang bukan ruang dan berlakurentetan masa dari masa lepas, masa kini, danmasa depan. Oleh itu, masa tidak dapatdiputarkan balik putarannya. Kontinum masaboleh dilihat daripada rentetan masa sepertisaat, minit, hari, minggu, bulan, tahun, dekad,kurun, dan alaf. Nama-nama rentetan masa tadiadalah ciptaan manusia yang berkaitan dengankehidupan mereka (Reha Mustafa, 16-17). Secararingkasnya, masa bermaksud urutan peristiwaatau tempoh berlakunya peristiwa-peristiwatersebut.

Teori Relevan

Sesungguhnya TR lebih menekankan bentukkomunikasi yang dapat difahami oleh penuturdan pendengar. Dalam hal ini, penutur (dalamterjemahan dipanggil penulis teks sumber)seharusnya menjamin apa yang dibualkan benar-benar relevan dan mudah difahami memahamkanpendengar (dalam terjemahan dipanggilpembaca kedua). Perbualan itu mesti mencapaikerelevanan yang optimal (Nor Hashimah 1992).Apabila sesuatu ujaran itu disampaikan makamenjadi tanggungjawab si penutur memahamkanpendengar (Nor Hashimah Jalaludin, 33). Bagipendengar pula, dia mencuba memperoleh kesankonteks yang mencukupi dan kesan itu mestilahmudah diperoleh. Tujuannya adalah untukmenggalakkan usaha memproses maklumat yangtidak diperlukan. Hal ini seterusnya menghasilkandefinisi TR yang sebenarnya, iaitu:1) Dengan andaian segala perkara seimbang,

semakin banyak kesan konteks maka ujaranitu semakin relevan.

2) Dengan andaian segala perkara seimbang,semakin sedikit usaha memproses maklumatmaka ujaran itu semakin relevan.

Inferens

Menurut TR, komunikasi bukan setakatmengekod, memindah, dan menyahkodmaklumat tetapi memerlukan proses inferensiaitu proses membuat kesimpulan dan rumusanterhadap sesuatu maklumat yang ada (Gutt1998). Misalnya contoh berikut :Kes 2Ani : Lama lagi ke si Karim tu?Tijah : Dia sekarang sedang berurusan dengan

Ali.

Dalam perbualan di atas, Tijah tidakmenjawab secara langsung pertanyaan Ani. Tijahhanya memberitahu bahawa Karim sedang ber-urusan dengan Ali. Katakanlah si Ali itu seorangyang mudah berurusan dengannya makadapatlah disimpulkan bahawa Karim tidak akanlama bersama dengannya. Daripada konteks yangdibekalkan, kita perlu membuat inferensterhadap sesuatu komunikasi seperti kes di atas.Dengan kata lain, menurut TR, makna ujaranmengikut konteks yang ada. Jadi untukmemastikan kejayaan sesuatu komunikasi, konsepmembuat kesimpulan iaitu inferens perlulahdibuat (Gutt, 42).



Konteks

Dalam TR, konteks ialah binaan pemikiran yangbersifat psikologi dan kognitif (Gutt 1998). lamerupakan segugus andaian pendengar ataupembaca kedua/pembaca sasaran (dalam disiplinterjemahan) terhadap alam dunia danpersekitarannya (Sperber dan Wilson 1986).Menurut takrifan ini, konteks meliputi hampirkesemua fenomena dan peristiwa yang dapatditanggap oleh pemikiran manusia. Konteks jugabukan setakat terdapat pada persekitaran fizikalyang boleh dilihat dan dirasa tetapi juga meliputiujaran dan ayat terdahulu (untuk maklumatdalam teks), andaian masa depan, hipotesissaintifik, kepercayaan agama, memori silam,budaya sejagat, dan persepsi manusia terhadapsikap dan mentaliti si penutur sesuatu komunikasi(Gutt, 43). Konteks boleh dibentuk dandikembangkan. Pembentukan konteks bolehmempengaruhi pemahaman sesuatu maklumat.Misalnya manusia tidak mudah mengingati alatdan jenis permainan jika disuruh apatah lagijika yang ditanya itu ialah warna alat permainanyang pertama sekali dibeli semasa hari jadi. Olehitu, usaha mengingat maklumat dan keupayaanminda kita memproses maklumat mempunyaikaitan yang rapat. Kesemua itu terangkum dalamkonsep dan pengertian konteks.

Konteks yang bersesuaian dapatdikembangkan jika maklumat baru ditambah.Setiap maklumat baru yang ditambah maka iamemerlukan konteks yang berbeza (NorHashimahjalaluddin 1992). Pembinaan konteksadalah bahagian terpenting dalam interpretasiujaran. Dalam interpretasi ujaran, pendengarakan menetapkan konteks yang sesuai bagimenghasilkan kesan konteks. Kesan konteks ialahandaian awal pendengar tentang ujaran yangdidengarnya. Dalam perbualan misalnya,pendengar akan menggabungkan ataumemperkuat atau menggugurkan konteks yangdikemukakan oleh penutur bagi menghasilkankesan konteks. Misalnya contoh berikut:

Kes 3 (gabungan maklumat baru dan lama)

Zaiton bangun daripada tidur dengan andaiankonteks seperti berikut:1. Kalan azan subuh kedengaran, dia akan

bangun untuk solat subuh.(Kemudian dia bangun dan kedengaran :(konteks sedia ada / lama)

2. Azan subuh sedang berkumandang (konteksbaru)

Inferens (Rumusan atau kesimpulan):3. Zaiton akan sembahyang subuh (gabungan

konteks lama dan konteks baru)

Kes 4 (menguatkan konteks baru dan lama)

Zaiton bangun daripada tidur kerana terdengarbunyi suara di menara masjid. Lantas Zaitonmembuat andaian iaitu:1. Azan subuh sedang bergema (andaian awal/

konteks lama).

Tetapi apabila Zaiton membuka matanyadan diamat-amati suara itu, rupa-rupanya iamerupakan:2. Bacaan al-Quran yang dibaca oleh pak bilal

di masjid (konteks sebenar— denganmenggugurkan andaian awal).

Begitulah cara sesuatu konteks dibentuk bagimenghasilkan kesan konteks yang sebenarnya.Dengan kata lain, konteks amat memainkanperanan yang cukup penting dalammenginterpretasi makna sebenar sesuatu ujaran.

Kerelevanan Optimal

Menurut Sperber dan Wilson (1986 dan 1995),faktor utama kejayaan sesuatu komunikasi adalahdengan mencapai kerelevanan optimal sama adadi pihak penutur mahupun pendengar .Kerelevanan optimal pendengar dicapai apabila1) pendengar dapat memahami niat dan maksudsebenar penutur tanpa memerlukan usahamemproses yang berlebihan dan tidaksepatutnya, 2) makna yang diniatkan olehpenutur dapat ditangkap oleh pendengarsetimpal dengan usahanya memproses maklumattersebut. Makna atau 'faedah' yang dihasilkanoleh pendengar kerana usaha tadi adalah bersifatpsikologi, iaitu berdasarkan kualiti dan kuantitipengetahuan seseorang. Oleh itu, menurut TR,apabila seseorang ingin berkomunikasi, merekamestilah memastikan bahawa ujarannyamencapai kerelevanan yang optimal (Gutt, 43).Hal ini penting kerana interpretasi yang tepatterhadap sesuatu ujaran dapat dicapai olehpendengar. Justeru, penutur mestilahmembekalkan maklumat dan konteks yangsecukupnya sebelum berkomunikasi. Misalnyaapabila seorang mahasiswa universiti (penutur)yang berbual dengan seorang nelayan(pendengar) maka perbualan mereka mestilahberkisar tentang hal ehwal kehidupan nelayan



seperti cerita tentang laut, ikan, pukat, dansampan. Kalau diubah topik kepada hal ehwaldunia akademik di universiti maka nelayan itutidak akan mendapat maklumat yang cukuptentang perkara yang dibincangkan. Lantaranitu perbualan mahasiswa tadi tidak mencapaikerelevanan yang optimal. Dengan kata lain,penutur mestilah memandu perbualannya agarkomunikasinya dapat difahami oleh pendengardengan mudah serta berjalan lancar. Dengankata lain yang lebih mudah, kerelevanan yangoptimal bermaksud setiap ujaran itu mempunyaikesan konteks yang maksimum yang dapatdirumuskan dalam masa sesingkat mungkin (NorHashimah Jalaluddin, 33).

Penggunaan Bahasa Secara Interpretif danDeskriptif

Dalam TR, terdapat duajenis penggunaan bahasasecara kognitif, iaitu penggunaan interpretif dandeskriptif. Penggunaan bahasa secara deskriptifbermaksud sesuatu ujaran komunikasi itudianggap betul dalam sebarang duniakemungkinan. Manakala penggunaan secarainterpretif pula bermaksud sesuatu ujaran yangdigunakan untuk menyatakan dan menjelaskanapa yang orang kata dan lafazkan. Misalnyaujaran berikut (contoh berikut diadaptasidaripada contoh Gutt: 44):

Kes 6a)Aminah : Ahmad dan isteri pertamanya telah

bercerai.b)Aminah : Talib pernah memberitahu saya,

Ahmad dan isteri pertamanya telahbercerai.

Kedua-dua contoh ujaran di atas mempunyaiujaran yang sama, iaitu Ahmad dan isteripertamanya telah bercerai. Dalam ayat pertama (a),Aminah menganggap bahawa kenyataannyatentang penceraian Ahmad dan isteri pertamanyaadalah benar. Dengan itu, kenyataan dan ujaranAminah itu dianggap deskriptif. Bagaimanapundalam ayat kedua 9b), Aminah tidak menyatakandan mendakwa bahawa Ahmad dan isterinyatelah bercerai. Aminah hanya melaporkandakwaan Talib yang berbunyi Ahmad dan isteripertamanya telah bercerai Dengan kata lain,Aminah hanya menyatakan semula secara penuh(memetik kata-kata) apa yang pernah dinyatakanoleh Talib. Oleh yang demikian, ujaran Aminahitu dianggap sebagai bersifat interpretif. Dalam

ayat kedua (b), ujaran Aminah tidak dianggapsalah sekiranya Ahmad dan isterinya tidakbercerai. Tetapi Aminah akan dianggap bersalahsekiranya Talib tidak pernah berbuat kenyataantentang penceraian Ahmad dan isterinya; seolah-olah Aminah menipu. Dalam ilmu komunikasi,aktiviti terjemahan dianggap sebagai aktivitikebahasaan secara interpretif.

Kesamaan Interpretif dan Kejujuran dalamPenterjemahan

Faktor penting dalam pengunaan bahasa secarainterpretif ialah hubungan antara kesamaaninterpretif (interpretive resemblance) ujaran asaldengan pernyataan tentang ujaran itu (dalamdisiplin terjemahan ialah kesamaan interpretifteks sumber dan teks sasaran). Kedua-dua ujaranitu (yang asal dan yang diujarkan semula olehorang lain) berkongsi makna dan maklumateksplikatur (tersirat) dan implikatur (tersurat).Perkara ini membayangkan bahawa kesamaaninterpretif mempunyai tahap perbezaan (degree).Jika kedua-dua ujaran tadi mempunyai kesamaaninterpretif yang lebih rapat, berkemungkinanmaklumat tersurat dan tersiratnya sama. Petikankata-kata secara penuh (direct quotation) dalamkes 6 (b) di atas menunjukkan kesamaaninterpretif (persamaan makna dan maklumat)yang paling tinggi dan rapat dengan ujaan asal.Petikan kata secara penuh tadi pastinya dapatditafsirkan dan diinterpretasi dalam konteks yangsama dengan konteks ujaran asal (Gutt, 44-45).

Sebaliknya ujaran yang bersifat huraian,ringkasan, tafsiran, agakan dan andaian dianggapmempunyai kesamaan interpretif yang berbeza(kurang sedikit) dengan ujaran asal berbandingujaran yang bersifat petikan kata-kata secarapenuh tadi. Misalnya jika seorang pelajar ditanyaoleh kawannya (yang tidak hadir ke kuliah)tentang kandungan dan isi kuliah yangdisampaikan oleh pensyarahnya, mungkin pelajaritu mempunyai beberapa pilihan untukmenjawab pertanyaan tersebut. Antaranya, diaakan menjawab seperti berikut:1. Dia akan memberitahu secara terperinci isi

kuliah tersebut hingga mungkin mencapaikesamaan interpretif yang tinggi dengan isikuliah asal yang disampaikan olehpensyarahnya.

2. Dia juga mungkin hanya melaporkan danmemberitahu sebahagian kecil isi kuliahsecara terperinci; yang baki mungkindiringkaskan.



3. Dia mungkin hanya memberikan isi-isi utamakuliah tersebut secara ringkas sahaja.

Untuk mencapai kesamaan interpretif,pelajar itu perlu menyampaikan jawapan (1)kerana dia perlu jujur (faithfulness) dengan ujaranasal (dalam konteks ini ialah kuliah pensyarahnya).Hal ini kerana pelajar itu akan mencapaikerelevanan optimal semasa menjawab pertanyaankawannya yang tidak hadir ke kuliah. Dengankata lain, ujaran kita mestilah sejujur dansehampir mungkin dengan ujaran asal. Dalampenterjemahan yang tergolong dalam kelompokpenggunaan bahasa secara interpretif makakonsep inilah yang sepatutnya ditonjolkan. Halini bersesuaian dengan pendapat Wilson danSperber (1988 : 137):

... The speaker guarantees that her utterance is a faithfulenough representation of the original; that is, resembles itclosely enough in relevant respect...

Bagaimanapun, TR yang menekankankonsep kerelevanan ujaran dan konteks makapelajar itu masih boleh memberikan jawapan(2) atau (3) berdasarkan konteks dan kehendakkawannya yang tidak hadir ke kuliah tadi.

Penterjemahan Sebagai Penggunaan Bahasa SecaraInterpretif

Menurut TR, penterjemahan tergolong dalamkonsep penggunaan bahasa secara interpretif(Gutt, 46). Penterjemahan sebenarnya bermaksudmenyatakan semula apa yang orang kata atautulis tetapi dalam bahasa yang lain.Penterjemahan sering dibandingkan dengankonsep memetik dan melaporkan sesuatu ucapandan dialog dalam sesuatu bahasa yang sama(intra linguistic use). Cuma yang membezakankedua-dua jenis penggunaan bahasa ini ialahterjemahan melibatkan bahasa orang lainsedangkan petikan dan laporan hanya melibatkanteks dalam bahasa asal sahaja dan tidakmelibatkan bahasa orang lain. Oleh sebabpenterjemahan tergolong dalam penggunaanbahasa secara interpretif maka ia terpaksamengamalkan konsep kejujuran (faithfulness)dalam menyampaikan warta mesejnya. Dengankata lain, penterjemahan mestilah sedayamungkin untuk setia dan jujur dengan tekssumber (Wilson dan Sperber 1998: 137).

Tipotogi (Jenis) Teks Sebagai Pedoman MencariKerelevanan

Sebagaimana segala fenomena kehidupan yangberlaku di dunia ini maka komunikasi manusiajuga mempunyai bentuk istilah-istilah tertentuberdasarkan jenis komunikasi yang diujarkan.Misalnya terdapat istilah atau laras bahasa yangberbeza dalam cerita rakyat, novel, cerita kartun,buku ilmiah, dan abstrak kajian ilmiah. Menurutteori komunikasi moden, bentuk dan tipologiteks tadi dapat membentuk niat penutur (dalamteks disebut penulis dan dalam terjemahandipanggil penterjemah) menghasilkan karyanya.Khalayak (pembaca atau pendengar) dapatmenjangkakan idea dan mesej yang ingindisampaikan berdasarkan tipologi teks tersebut.Misalnyajika penutur menyatakan ujaran dalambentuk laporan maka dia dapat mengagakrespons dan reaksi khalayaknya berbanding kalaudia menghasilkan ujaran berbentuk ceritakomedi. Dalam TR, jenis dan tipologi tekssebegini akan memandu khalayak mencapaikerelevanan optimal dalam komunikasi mereka.Misalnya kalau komunikasi yang dipersembahkanitu berbentuk novel maka yang terbayang dikepala pembaca ialah konsep plot cerita, watakutama (protagonis), watak jahat (antagonis), dannilai moral yang ingin disampaikan. Oleh itupembaca tidak akan mempersoalkan ketepatandan kerasionalan kejadian peristiwa dalam noveldan tarikh-tarikh penting kerana teks itu bukanberbentuk teks sejarawi (Gutt). Justeru, kosmemproses untuk memahamkan teks juga tidakbanyak. Dan jika teks itu relevan dengankehendak khalayak maka interpretasi sebenarteks yang dihasilkan oleh penulis itu dapatdisingkap. Kesimpulannya tipologi teks akanmembantu khalayak mendapatkan kerelevananoptimal dalam menginterpretasi sesuatu ujarankomunikasi sama ada komunikasi lisan sepertiperbualan ataupun komunikasi bukan lisanseperti novel dan cerpen.

Penterjemahan Langsung

Dalam TR, penterjemahan langsung (directtranslation) dianggap sebagai petikan kata-katasecara langsung (direct quotation). Oleh itu, aktivitipenterjemahan dikategorikan sebagai penggunaanbahasa secara interpretif. Petikan kata-kata secaralangsung bermaksud satu contoh ujarankomunikasi yang dilafazkan atau ditulis bagimenyatakan kesamaan terhampir dengan


Penterjemahan Pragmarik dalam Konsep Masa Arab-Melayu: Satu Analisis Teori Relevan

keaslian maksud teks asal (Sperber dan Wilson,227). Antara keistimewaan petikan langsungadalah kesamaan dan ketepatannya dengan tekssumber (untuk penterjemahan) bukan setakatpada apa yang dimaksudkan (makna) tetapimerangkumi juga cara bagaimana (stilistik) maksuditu dipersembahkan. Dalam penterjemahanlangsung, segala ciri linguistik teks sumber akandikekalkan. Dengan itu, khalayak terjemahanmampu menstruktur semula makna teks sumberberdasarkan makna yang diniatkan oleh penulisasal. Hal ini kerana teks sumber mempunyaiandaian kontekstual tersendiri yang perludikekalkan. Jadi biarlah konteks sumber itu jugadikekalkan dalam penterjemahan (Gutt 1991 :162). Misalnya contoh berikut:

Kes 7John eats like a pig. (John makan seperti babi).

Dalam ayat yang mengandungi bahasafiguratif di atas, konsep penterjemahan langsungharus dibuat. Sebenarnya, penulis asal inginmenyampaikan maksudnya kepada khalayak asal,iaitu orang Inggeris mengikut konteks asal tekssumber itu. Penulis asal mungkin tidakmenjangka karyanya akan diterjemahkan. Dalambudaya Inggeris, ada beberapa andaian tentangmaksud makan seperti babi (Gutt, 167). Antaraandaian yang ingin dimaksudkan adalah seperti:1) John makan terlalu banyak seperti seekor

babi makan.2) John makan gelojoh seperti seekor babi.3) John makan terlalu bising seperti seekor

babi hutan.4) John makan terlalu kotor dan jijik seperti

babi sedang makan.

Andaian-andaian di ataslah yang ditujukankhas oleh penulis teks sumber kepada khalayakasalnya, iaitu orang Inggeris. Terpulanglahkepada mereka untuk mentafsirkan interpretasisebenar ujaran itu berdasarkan konteks masing-masing. Yang pasti perbuatan makan seperti babiitu adalah perbuatan yang tidak senonoh dantidak sopan. Itulah kesan sebenar teks asal. Olehitu, sebagai penterjemah kita terpaksamengekalkan andaian-andaian dan kesan konteksasal tersebut kerana memang itulah kesankonteks yang ingin disampaikan oleh penulisasal. Maka kita tidak boleh menterjemahkanbahasa figuratif itu sebagai makan seperti tikus(dalam budaya Melayu) kerana tikus hanya

melambangkan sifat pengecut dan penakut. Jikaadaptasi ini buat maka kesan konteks asal akanhilang. Jadi untuk mencari kerelevanan optimaldalam komunikasi maka konsep terjemahanlangsung harus digunakan. Hal ini lebih selamatbagi penterjemah sekiranya mereka memandaiuntuk mentafsirkan makna sebenar takut terjebakkalau salah tafsir. Kalau salah tafsir maka salahlahterjemahan itu. Dengan kata lain, terjemahanlangsung adalah langkah paling selamat untukmengelakkan salah tafsir (Gutt, 165-167).

PEMAPARAN DATA

Seperti yang disebut dalam batasan kajian,terdapat beberapa kata masa bahasa Melayu yangdipinjam daripada bahasa Arab. Kata pinjamanini telah sebati dan diterima pakai oleh orangMelayu sejak sekian lama. Antara kata masatersebut ialah :

al-Fajr (Fajar)

Dalam bahasa Arab, al-fajr bermaksud tamatnyawaktu malam kerana munculnya cahaya Subuh(al-Nahw, Khalil 1991 : 370). Manakala dalampengertian syariah Islam (hukum fekah) al-fajrbermaksud satu waktu tatkala terpancar cahayadari celah kegelapan malam kemudian cahayatersebut tersebar di ufuk timur. Cahaya inidinamakan fajar Sadiq (Reha Mustafa, 86). FajarSadiq akan muncul dahulu sebelum masuknyawaktu sembahyang subuh. Waktu fajar dianggapsebagai salah satu waktu yang paling diberkatikerana ibadat pada waktu ini akan disaksikanoleh para Malaikat dan mendapat pahala yangbesar daripada Allah Ta'ala. Menurut tradisiilmu Islam, waktu fajar ialah waktu yang palingsesuai untuk mengulang kaji pelajaran keranaketenangan dan keheningan waktu fajar itu.Inilah makna sebenar Fajar dalam bahasa Arabdan agama Islam.

Orang Melayu sudah menerima pakaiperkataan Fajar ini. Cuma apabila konsep masaini digunakan dalam bahasa dan budaya Melayumaka ia telah diasimilasikan menurut budayatempatan. Misalnya contoh berikut:

Kes 8Jali : Bila engkau nak pergi menoreh?Samad: Sebelum fajar menyingsing.

Dalam kes 8 di atas, apakah yangdimaksudkan oleh ungkapan Sebelum Fajarmenyingsing? Menurut TR, ungkapan itu



Ilmu komunikasi kognitif

Ilmu pragmatik

Teori relevan

(Ciri-ciri)

Gabungan semantik dan pragmatik

Penggunaan bahasa secara interpretif

Penterjemahan langsung

Tipologi teks sebagai pedoman mencari kerelevanan

Kerelevanan optimal

Kesamaan interpretif dan kejujuran dalam penterjemahan

Rajah 1: Cm dan gagasan kemunculan teori relevan

mempunyai beberapa andaian. Antaranyaseperti:a. Betul-betul sebelum waktu Subuh.b. Sebelum terbit Fajar Sadiq.c. Sebelum ayam berkokok.d. Awal pagi, iaitu dinihari (selepas tengah

malam).

Dalam hal ini, konteks akan menentukaninterpretasi sebenar waktu yang dimaksudkanoleh Samad. Kalau Samad lazimnya pergi kekebun untuk menoreh getah sebelum ayamberkokok maka Jali akan memahami waktusebenar Samad keluar dengan mudah tanpa kosniemproses maklumat yang tinggi. Dalam halinijuga, jawapan Samad dianggap telah mencapaikerelevanan optimal dengan syarat Jali telahdibekalkan maklumat yang secukupnya.Maksudnya, Jali tahu benar tentang waktu kerjaSamad (mungkin mereka berdua kawan rapat).Kalau konteks menentukan Jali selalu pergi kekerja pada waktu begitu maka Samad akanmembuat inferens (rumusan) seperti dalamandaian c) tadi. Penentuan konteks pentingkerana ruang lingkup dan tempoh masa dalambudaya orang Melayu terlalu luas, sukar

ditentukan waktunya secara tepat. Justeru sikapbeginilah yang menghasilkan pepatah JanjiMelayu. Bagaimanapun, TR dapat menyingkapmakna sebenar ujaran di atas denganberdasarkan konteks ayat.

Dan mungkin juga maksud Samad waktu ituawal pagi (andaian-d). Hal ini kerana pokokgetah akan mengeluarkan hasilnya denganbanyak pada dinihari. Kemudian si penorehakan berehat untuk solat subuh sebelummenyambung semula kerja untuk mengumpulgetah yang telah ditoreh. Jawapan tepat waktuyang dinyatakan oleh Samad itu adalahberdasarkan konteks, iaitu rutin kerja Samadyang sebenarnya. Apa pun dalam budaya Melayu,makna implisit sebelum fajar menyingsing ialahwaktu terlalu awal dan lazimnya sebelum waktusubuh. Ciri implisit dalam ujaran di atas agakmudah difahami kerana orang Malayu fahambenar akan maksud sebelum fajar menyingsing.Oleh itu, usaha memproses maklumat ini rendahdi samping konsep konteks yang dapat mencariinterpretasi sebenar ujaran tersebut. Jika usahamemproses maklumat rendah dan kesan konteksyang banyak maka ujaran itu akan lebih relevan(Nor Hashimah 2001).


Penteijemahan Pragmatik dalam Konsep Masa Arab-Melayu: Satu Analisis Teori Relevan

Untuk menterjemahkan perkataan al-fajrmaka pendekatan TR, iaitu penterjemahanlangsung harus digunakan. Penterjemahlangsung akan menghasilkan kesamaan dankonteks yang paling hampir dengan teks sumber(Gutt 1991). Oleh itu, kalau teks sumberberbunyi Zahabtu ila al-hadiqah Fajran makaterjemahan langsungnya ialah Say a pergi ke kebunpada waktu Fajar. Perkataan fajar di sini tidakboleh diterjemahkan pagi ataupun Subuh keranakedua-dua perkataan ini mempunyai pengertianyang berbeza dalam bahasa Arab. Begitu jugamakna dari segi bahasa tadi (terbitnya Fajar Sadiq)diberikan maka ini pun tidak dibenarkan.Lagipun istilah Fajar sudah sebati dan diterimapakai oleh orang Melayu sejak sekian lama.

al-Subh (Subuh)

Dalam bahasa Arab al-Subh bermaksud awalpermulaan hari siang (al-Nahw, Khalil 1991 :281). Manakala dalam syariat Islam, Subuhbererti waktu solat Subuh yang bermula apabilaterbit Fajar Sadiq (Reha Mustafa, 87). Terdapatperbezaan antara waktu Subuh dan waktu Fajar.Waktu fajar merupakan waktu merupakan waktupemisah antara malam dengan siang atau antarasuasana gelap dengan terang. Manakala waktuSubuh disandarkan pada solat Subuh itu sendiri,iaitu waktu yang terbit selepas waktu fajar (RehaMustafa, 87). Dengan kata lain, waktu Fajar danwaktu Subuh pasti tidak akan tertembung selama-lamanya.

Dalam bahasa dan budaya Melayu,pengertian dan penggunaan istilah Subuh itumasih sama seperti dalam makna asalnya dalambahasa Arab. Bagaimanapun oleh kerana orangMelayu terkenal kerana 'Janji Melayunya' makapersepsi penggunaan waktu Subuh agak berbezasedikit. TR akan mencuba menganalisis masalahini. Misalnya lihat ujaran berikut:

Kes 9Syed : Bila engkau nak pergi pasar?Jamal : Lepas Subuh nanti.

Dalam contoh untuk kes 9 di atas, TRmembuat beberapa set andaian tentang maksudmasa dalam ujaran itu. Antaranya seperti:a. Betul-betul selepas solat subuh, iaitu selepas

sarapan pagi.b. Selepas mentari pagi naik segalah, iaitu pada

waktu duha.

c. Mungkin juga sebelum masuk waktu Zohor,iaitu sebelum matahari condong sedikit kearah barat dan menghasilkan bayang-bayangsebarang objek di bumi.

Oleh sebab, penggunaan konsep waktudalam budaya Melayu tidak tetap maka agaksukar menentukan masa sebenar waktu yangdinyatakan oleh Jamal itu. Tetapi bagi TR ujaranmasih boleh difahami dengan berdasarkankonteks. Kalau Jamal ini jenis orang yangmenepati janji dan masa (punctuality) makaandaian a) akan menjadi interpretasi sebenarujaran. Lantaran itu, Syed akan membuatkesimpulan (inferens) bahawa Jamal akan pergike pasar betul-betul selepas waktu subuh, lebihkurang pukul 7 - 8 pagi. Sebaliknya juga jikaJamal dikenali di kalangan rakannya sebagaijarang menepati masa maka andaian c) mungkinmenjadi interpretasi tepat bagi ujaran Jamal tadi.

Selain itu, konteks masa dan waktu pasar itudibuka juga menentukan waktu sebenarpemergian Jamal ke pasar. Kalau pasar itudibuka pada pukul 8 pagi maka andaian a)relevan untuk menjadi interpretasi sebenarujaran Jamal. Dan jika pasar itu dibuka pukul10 pagi (kebanyakan pasar raya dibuka padawaktu ini) maka andaian b) merupakan maksudsebenar ujaran Jamal. Dengan kata lain, dalamTR penutur mestilah membekalkan maklumatyang secukupnya kepada pendengar. Penuturmestilah memandu perbualan tersebut agarkerelevanan optimal dapat dicapai. Konteksatau set andaian tadi perlu ada agar kesankonteks dapat dihasilkan. Menurut TR kalausemakin banyak konteks dibekalkan maka kesankonteks mudah difahami tanpa usaha memprosesmaklumat yang tinggi. Dan jika usaha dan kosprosesnya rendah sedangkan kesan konteksnyatinggi maka ujaran itu lebih relevan (NorHashimah Jalaludin, 51).

Selain itu, terdapat satu ungkapan dalamperibahasa Melayu iaitu, Subuh gajah. Dalamilmu semantik, bahasa figuratif itu akanditafsirkan secara eksplisit, iaitu setakat maknaleksikal dan logiknya sahaja ; subuh - waktuselepas terbit Fajar Sadiq dan gajah - spesiesbinatang mamalia darat yang terbesar. Itu sahajayang dapat ditafsirkan oleh ilmu semantik.Tetapi dalam ilmu pragmatik dengan tunjangnyakonsep TR maka peribahasa itu harus dianalisismengikut konteks. Lazimnya orang Melayu akanfaham bahawa peribahasa itu masih melibatkan



waktu dan solat subuh. Makna implisit peribahasaini ialah sembahyang Subuh di luar waktusebenarnya dan lazimnya pada waktu duha.Dengan kata lain, Svbuh gajah itu ialah mengqadasolat Subuh kerana terlewat bangun tidur dengantidak sengaja. Dan qada' itu perlu ditunaikansebaik sahaja kita terbangun daripada tidur.Untuk tujuan analisis, lihat perbualan berikut.

Kes 10Fauzi: Engkau Subuh gajah ke, Jalil?Jalil : Apa boleh buat, dah terlewat bangun.

Dalam kes 10 di atas, Jalil masih memahamimaksud si Fauzi kerana dia tahu tentang maknaimplisit peribahasa itu. Lebih-lebih lagi kontekssi Jalil yang sememangnya terlewat bangun tidur.Maka Jalil faham benar apa yang dimaksudkanoleh Fauzi. Menurut TR ujaran peribahasa Fauzimencapai kerelevanan optimal kerana ia pekakonteks. Penutur sengaja bertanya dalamungkapan sebegitu mungkin kerana memperli(ironi) ataupun cakap berlapik kerana menjagahati kawan. Yang pasti makna implisit peribahasaitu masih boleh difahami. Selain itu, kesankonteks yang dihasilkan juga tinggi keranapendengar memang berada dalam situasi terlewatbangun. Pendengar tidak sukar memahamiujaran itu kerana pertanyaan Fauzi memangdalam konteks yang sepatutnya. Dengan katalain, konteks yang dibekalkan pada pendengarsememangnya cukup. Cukup dalam erti katabahawa peribahasa itu memang difahami olehorang Melayu khususnya golongan muda. Jaditidak ada masalah bagi pendengar menginterpretasimaksud sebenar ujaran itu.

Dalam konteks terjemahan, TR mengamalkankonsep kejujuran dan kesamaan interpretif.Maksudnya bagaimana teks sumber dinyatakandan ditafsirkan maka teks terjemahan pun begitujuga. Oleh itu, untuk mencapai kesamaaninterpretif maka konsep kejujuran mestilahdiamalkan. Kejujuran dalam pengertian TRialah segala bentuk dan ciri teks sumber mestidikekalkan (Gutt, 11). Antara ciri yang perludikekalkan adalah seperti ciri linguistik (semantikdan sintaksis), statistik (gaya pengolahan),konteks, dan kesan konteks. Oleh itu,penterjemahan ungkapan masa dan waktu inipun harus menggunakan pendekatan terjemahanlangsung. Misalnya perkataan sabahun yangbermaksud permulaan siang atau waktu subuh

maka makna itu perlu dinyatakan. Dalamkonteks ini, terjemahan makan pagi kurang sesuaidigunakan kerana ruang lingkup dankontimumnya terlalu luas. Begitu juga ungkapanSabah al-khayr (selamat pagi) yang merupakankata sapaan yang diucapkan sewaktu bertemusesiapa sahaja pada waktu pagi. Bentuk sapaanbegini merupakan satu penghormatan kepadaorang yang disapa. Dalam budaya Arab, katasapaan ini boleh ditujukan kepada sesiapa sahaja,tidak kira orang Islam ataupun bukan Islam.Hal ini berbeza dengan ungkapanassalamualaikum yang dikhususkan untuk orangIslam. Padanan yang tepat untuk masa sabah al-khayr ialah selamat pagi dan Salam Sejahtera.Manakala ungkapan assalamualaikum tidak bolehditerjemahkan menjadi Salam Sejahtera keranakonsep penterjemahan langsung mestilah bersifatinterpretif dan petikan penuh (direct quotation).

Al-Sahr (Sahur)Dalam kamus Arab, al-sahr bermaksud fasaterakhir waktu malam, iaitu waktu malam yangakan tamat sebaik sahaja terbit Fajar Sadiq (al-Nahw, Khalil 1991 : 233). Dalam syariat Islamwaktu sahur ini ialah bahagian terakhir malam,iaitu terletak 1/3 terakhir setiap malam (RehaMustafa, 100). Waktu ini sangat elok untukqiyamullayl (bangun malam) dan solat tahajjud(sembahyang tengah malam). Pada waktu ini,amal ibadat kita mudah diterima oleh AllahTa'ala.

Dalam budaya Melayu, waktu sahur seringdikaitkan dengan makan sahur semasa bulanRamadan. Sebenarnya, sahur itu tergolong dalamkategori masa dan waktu. Makan sahur itubermaksud menjamah makanan atau meneguksegelas dua air pada waktu terakhir setiap malamRamadan (dalam budaya Melayu, selepas jam 12tengah malam dianggap pagi), iaitu sebelumterbit Fajar Sadiq. Orang Islam tidak bolehmakan dan minum apabila Fajar Sadiq telahterbit. Dengan kata lain, apabila masuk waktusubuh maka orang Islam dilarang makan danminum. Begitulah pengertian sebenar Sahurdalam bahasa Arab dan syariat Islam. Cumadalam budaya Melayu ada sedikit asimilasipengertian itu tetapi masih lagi ada kena-mengena dengan makna sebenar sahur itu.Justeru bagaimana TR menganalisis konsep sahurini dalam pengertian budaya Melayu akan terlihatdalam contoh berikut:



Kes 11Ibu : Bila engkau nak habiskan kari ayam ni

Mat?Mamat : Waktu sahur nanti emak?

Waktu sahur dalam pengertian orang Melayuada kalanya bermula selepas jam 12 tengahmalam, sedangkan waktu sebenarnya ialah padadetik-detik terakhir sebelum menjelang fajar.Sering kedapatan orang muda yang malasbangun dan masak pada awal pagi akan bersahurpada tengah malam, iaitu sekitar jaml2 tengahmalam hingga 3 pagi. Jadi waktu sahur orangMelayu cukup lama, lebih kurang 4-5 jam. Olehitu, ujaran waktu sahur itu mengandungibeberapa andaian. Antaranya seperti :a. Waktu 1/3 terakhir setiap malam, iaitu

sekitar 1 jam sebelum masuk waktu Subuh.Waktu ini adalah paling afdal untukbersahur.

b. Waktu imsak, iaitu waktu kritikal untukbersahur. Umat Islam tidak digalakkanbersahur pada waktu ini kerana takut terlajakmakan.

c. Waktu tengah malam, iaitu awal pagi bagipengertian orang Melayu. Bersahur padawaktu begini kurang mendapat fadilat sahur.

Dalam kes 11, kalau lazimnya keluargaMamat akan bersahur pada waktu a) maka si ibuakan memahami ujaran Mamat itu. Dalam halini, jawapan Mamat akan dikira mencapaikerelevanan optimal jika si ibu dibekalkankonteks yang mencukupi, iaitu si ibu tahu waktusahur keluarganya sendiri. Ujaran Mamat dikirapeka konteks kerana Mamat tahu bila ibunyaakan bangun memasak dan bersahur. Si ibupula akan mudah memproses ujaran Mamatkerana maklumat dan konteks yang dibekalkanmencukupi. Dalam TR, jika konteks yangdibekalkan cukup maka kesan konteks tinggi.Justeru, ujaran itu dikira bermanifestasi (jelas)dan relevan (Gutt 1991). Begitujuga jika waktusahur keluarga Mamat seperti pada andaian b)maka ujaran Mamat tetap relevan buat si ibukerana si ibu dibekalkan konteks yangsecukupnya, iaitu dalam erti kata si ibu memangtahu waktu sahurnya keluarganya. Tetapi oranglain seperti jiran atau kawan Mamat mungkinsukar memahami waktu sebenar keluarganyabersahur kerana tidak dibekalkan konteks yangsecukupnya. Maksudnya kawan Mamat tidak tahuwaktu sahur keluarga Mamat. Begitulah TR

menganalisis ujaran waktu orang Melayu yangsedia dimaklumi bahawa konsep waktu orangMelayu terlalu 'luas' dan sukar ditentukandengan tepat.

Dalam konteks terjemahan, TR mengamalkankonsep kejujuran dan kesamaan konteks - tekssumber (seperti yang dijelaskan sebelum ini).Untuk itu, penterjemah harus tahu akan kontekssebenar sesuatu ayat itu diujarkan dalam tekssumber. Kalau teks sumber menyatakan maknawaktu dan masa seperti perkataan sahur makanyatakan maksud masa, iaitu 1/3 terakhir setiapmalam. Jika teks sumber menyatakan perbuatanmakan pada waktu sahur maka nyatakan maknasetiap malam. Dan jika sahur itu bermaksudperbuatan makan pada 1/3 terakhir malamsebelum subuh tetapi bukan dalam konteks bulanRamadan maka nyatakan makna tersebut.Dengan kata lain, konteks asas teks sumber harusdikekalkan dalam penterjemahan langsung.Pengekalan konteks asal amat ditekankan dalamTR kerana penterjemahan diklasifikasikansebagai penggunaan bahasa secara interpretif.Dalam penggunaan bahasa secara interpretif ini,terdapat konsep kesamaaan interpretif (interpretiveresemblance) yang menjadi ciri utamapenterjemahan TR (Gutt 1991).

KESIMPULAN

Sebagai kesimpulan, konsep masa dalampemikiran bahasa dan budaya Melayu terlaluluas. Terlalu luas maksud bahawa ketepatan masaorang Melayu tidak menentu. Mereka jarangberpegang pada ketepatan masa mereka hinggatercipta ungkapan Janji Melayu. Hal ini berbezadengan orang Jepun yang terlalu berpegangpada ketepatan masa mereka hingga kebanyakanmajlis dan aktiviti harian mereka ditentukandengan konsep masa seperti minit dan jam.Lazimnya kita terdengar mesyuarat orang Jepunbermula pada pukul 8.15 pagi dan tamat padajam 10.15 pagi berbanding orang Melayu yangsuka menggunkan konsep jam sahaja sepertipukul 8 pagi, 10 pagi dan 12 pagi.

Bagaimanapun ketidaktentuan masa dalambudaya Melayu ini dapat diselesaikan denganpendekatan TR yang mampu menginterpretasimasa dan waktu sebenar dalam komunikasi danujaran orang Melayu {Rajah 2). Denganberbekalkan konteks, kesan konteks dan usahamemproses maka TR berjaya mencari maknasebenar ujaran orang Melayu.



TR

(melalui PL)

PL (Kebaikan)

rMemudahkan Penterjemahan Mengekalkan Makna Asal

iMakna Tepat

Menghasilkan Set Andaian Menguji Minda

Rajah 2: Kebaikan TR dalam penterjemahan konsep masa Arab-Melayu

Selain itu, konsep TR juga dapat diterapkandalam penterjemahan melalui pendekatanlangsungnya. Penterjemahan langsung sesuaidigunakan dalam menterjemahkan konsep masaArab-Melayu meskipun konsep penggunaan masadalam budaya dan bahasa Melayu agak rumitkerana sifatnya yang luas dan tidak menentu. TRlebih menekankan konteks asal yang terdapatdalam teks sumber kerana itulah sememangnyayang diniatkan oleh penulis asal. Melalui TR,kesamaan interpretif dan kesan konteks yangsama dengan teks dapat dikekalkan dalam teksterjemahan.

RUJUKAN

AL-NAHW, AL-KHALIL. 1991. al-Mu'jam al- Arabiya, al-

Muyassar. Tunis. Larousse.

ASMAH Hj. OMAR. 1994. Bahasa dan Pemikiran. KualaLumpur: Dewan Bahasa & Pustaka.

GUTT, E.A. 1991. Translation and Relevance. Cognitionand Context Oxford: Basil Blackwell.

GUTT, E A. 1998. Pragmatic Aspects of Translation:Some Relevance-Theory Observations. DalamThe Pragmatics of Translation. Toronto:Mitilingual Matters Ltd.

NOR HASHIMAH JALALUDDIN. 1992. Semantik dan

Pragmatik: Satu Pengenalan. Kuala Lumpur:Dewan Bahasa & Pustaka.

NOR HASHIMAH JALALUDDIN. 2001. Masa dan

Masyarakat Melayu dalam Dewan Bahasa. KualaLumpur: Dewan Bahasa & Pustaka.

REHA MUSTAFA. 1999. Konsep masa dalam al-Quran:Satu kajian dalam Juzu' Amma. Tesis MA.Universiti Malaya.

SPERBER dan WILSON. 1986 /1995 . Relevance:Communication and Cognition. Oxford:Blackwell.

(Diterima: 22 Ogps 2002)

164 PertanikaJ. Soc. ScL & Hum. Vol. 10 No. 2 2002


Tingkah Laku Keibubapaan dan Penyesuaian Tingkah Laku Anakdalam Keluarga Berisiko di Luar Bandar

ZARINAH ARSHAT, ROZUMAH BAHARUDIN, RUMAYA JUHARI 8c ROJANAH KAHARJabatan Pembangunan Manusia dan Pengajian Keluarga

Fakulti Ekologi Manusia, Universiti Putra Malaysia43400 UPMy Serdang, Selangvr, Malaysia

Kata kunci: Tahap resiliensi, analisis regresi, estim diri, tingkah laku keibubapaan

ABSTRAK

Kajian ini bertujuan untuk menentukan perkaitan di antara faktor latar belakang keluarga(pendidikan ibu, pendapatan keluarga dan bilangan anak) dan kualiti tingkah laku keibubapaandengan penyesuaian tingkah laku anak (resiliensi, estim diri dan pencapaian akademik) dalamkeluarga berisiko. Seramai 201 pasangan ibu-anak berbangsa Melayu dari kawasan luar bandarnegeri Kelantan, Kedah, Selangor dan Johor yang telah memenuhi kriteria (miskin dan mempunyaisekurang-kurangnya seorang anak berumur 7-12 tahun) kajian dipilih sebagai responden kajian.Responden telah ditemu bual secara bersemuka dengan menggunakan borang soal selidik.Kualiti tingkah laku keibubapaan telah diukur menggunakan skala 'Supportive Parenting (Simonset al. 1992). Tahap resiliensi anak pula diukur menggunakan skala 'Resiliency Attitudes Scales'(Bischoe 8c Harris 1994). Tahap estim diri anak diukur berdasarkan 'Hare Self-Esteem Scale' (Hare1985). Pencapaian akademik anak diukur berdasarkan skor lima mata pelajaran terpilih: BahasaMelayu, Bahasa Inggeris, Sains, Matematik dan Agama Islam. Hasil kajian menunjukkan bahawakualiti tingkah laku keibubapaan dan tahap penyesuaian tingkah laku anak adalah rendah.Pendidikan ibu didapati berkait secara signifikan dengan estim diri anak (r=0.18, p<;0.05) danpencapaian akademik (r=0.19, ps0.05). Keluarga yang berisiko tinggi dan mempunyai bilangananak yang ramai didapati mempunyai anak yang rendah tahap pencapaian akademik danresiliensi. Bilangan anak dalam keluarga didapati mempunyai perkaitan yang negatif dansignifikan (r=-0.16, pssO.05) dengan tahap resiliensi anak. Hasil kajian juga menunjukkan tahaprisiko keluarga mempunyai perkaitan yang signifikan dengan pencapaian akademik anak (r=-0.26,p^O.01). Kualiti tingkah laku keibubapaan juga mempunyai perkaitan secara positif dan signifikandengan tahap estim diri anak (r=0.17, p^O.05). Analisis regresi berganda menunjukkan tahappendidikan ibu (?=0.16, pss.05) dan tingkah laku keibubapaan (?=0.15, ps.05) menyumbangsecara unik terhadap estim diri anak (F=5.55, ps.05). Hasil kajian juga menunjukkan tahappendidikan ibu (?=0.18, p£.O5) menyumbang secara unik ke atas pencapaian akademik anak(F=2.87, ps.05). Kajian ini memberi implikasi bahawa pendidikan ibu dan tingkah laku keibubapaanmempengaruhi penyesuaian tingkah laku anak dalam keluarga berisiko di luar bandar.

ABSTRACT

This study aims to determine the relationships between family background factors (mother'seducational level, family income and number of children) and quality of parental behavior withchildren behavioral adjustment (resiliency, self-esteem dan academic achievement) in at-riskfamilies. A total of 201 Malay mother-<:hild dyads from rural areas in Kelantan, Kedah, Selangorand Johor who fulfill the selection criteria (poor and have at least one child aged 7-12 years) wereselected as respondents for the study. Respondents were interviewed face-to-face using questionaire.Parental behavior was measured using Supportive Parenting Scale (Simons et al 1992). Child'sresiliency is measured using Resiliency Attitudes Scales (Bischoe 8c Harris 1994) while Hare (1985)Self-Esteem Scale was utilized to measure child's self-esteem. The children's academic achievementis indicated by their score on five selected subjects, namely, Bahasa Melayu, Engish, Science,Mathematics and the Islamic Study. Results show that parental behavior and behavioraladjustment of the children are low. Mothers1 education is significantly related to child self-esteem

Zarinah Arshat, Rozumah Baharudin, Rumaya Juhari 8c Rojanah Kahar

(r= 0.18, p < .05) and academic achievement (r= 0.19, p < .05). Families with higher level of riskand with greater number of children are found to have children with low levels of academicachievement and resiliency. Number of children have negative and significant relationship (r= -0.16, p < .05) with child's resiliency. Findings also reveal that level of risk correlate significantlywith child's academic achievement (r= - 0.26, p < .05). In addition the quality of parental behavioris positively related to the level of child's self-esteem (r= 0.17, p < .05). Multiple regressionanalyses indicate that mother's education (? =0.16, p < .05) and quality of parental behavior (?• 0.15, p < .05) are unique predictors of child's self-esteem ( F = 5.55, p < .05). The findings alsoreveal that mother's education (? = 0.18, p < .05) determines child's academic achievement (F =2.87, p < .05). The study imply that mother's education and quality of parental behaviorinfluence behavioral adjustment of children in rural at-risk families.

PENGENALAN

Faktor berisiko merupakan keadaan atau variabelyang memberi impak negatif kepada kesihatan,kesejahteraan dan perlakuan sosial individu(Voydananoff & Donnelly 1998). Selain itu, faktorrisiko yang dihadapi oleh keluarga didapati bolehmempengaruhi kualiti tingkah laku keibubapaandan seterusnya kualiti kesejahteraan anak (Stern,Smith 8c Jang 1999; Gerard & Buehler 1999;Rozumah et al 1999a). Menurut Dumka danRoosa (1995) faktor berisiko merupakan variabelyang boleh mendorong anak mengalamiperkembangan negatif. Ini kerana kehadiranfaktor berisiko tidak menjamin kesan yang negatifsecara langsung, tetapi berpotensi untukmemudahkan kesan terse but berlaku (Small 8cLuster 1994). Kajian oleh Murry dan Brody(1999) mendapati semakin meningkat bilanganfaktor risiko yang wujud dalam keluarga semakinkurang tahap penyesuaian tingkah laku anak.Faktor risiko yang dihadapi oleh keluarga bolehwujud dalam pelbagai bentuk seperti tahappendidikan yang rendah, pendapatan keluargayang sedikit dan tidak menentu serta persekitaranrumah yang tidak selamat.

Walaupun terdedah kepada beberapa faktorrisiko, terdapat keluarga yang masih bolehmenjalankan fungsinya dengan berkesan.Namun, apabila terdapat banyak faktor risikoyang wujud dalam keluarga secara serentak,hanya segelintir keluarga yang mampu untukmenjalankan fungsinya dengan berkesan.Terdapat bukti menunjukkan ibu bapa darikeluarga berisiko berkecenderungan mempunyaimasalah tekanan psikologi yang tinggi (Voydanoff8c Dolley 1998; Mayhew 8c Lempers 1998;VoydanofF 1990). Oleh yang demikian adalahtidak mustahil jika didapati ibu bapa darikeluarga berisiko kurang memberikan sokongan,lebih mengamalkan hukuman fizikal untukmengarwal dan mendisiplinkan anak, gemar

memberikan arahan tanpa penjelasan dankurang memberikan tindak balas terhadapkeperluan emosi anak (Mayhew dan Lempers1998; Hashima dan Amato 1994; Small danLuster 1994). Di samping itu, ibu bapa daristatus sosioekonomi yang rendah didapati kurangmenyediakan persekitaran rumah yang kondusifuntuk pembelajaran anak dan mempunyaijangkaan yang rendah terhadap pencapaian anak(Shumow et al 1998; Miller dan Davis 1997;Solis-Camara dan Fox 1996; Luster dan McAdoo1994; Garret, Ng'andu dan Feron 1994; Dodge,Pettit dan Bates 1994).

Anak-anak daripada keluarga berisikodidapati mengalami kemurungan, gagal dalamakademik, mempunyai estim diri yang rendahdan terlibat dengan tingkah laku antisosial(Rozumah et al 1999b; Shumow et al 1998;Mayhew dan Lempers 1998; Hartos dan Power1997). Bagaimanapun jika anak daripadakeluarga berisiko mempunyai ciri resiliensiseperti mempunyai kemahiran menyelesaikanmasalah, kemahiran interpersonal, mempunyaijangkaan yang tinggi, motivasi yang tinggi danproaktif mereka akan mampu membuat adaptasiterhadap perubahan, tekanan dan kekecewaandalam kehidupan untuk meneruskan kehidupanyang lebih produktif (Franco dan Levitt 1998;Waxrnan dan Huang 1996; Gordon 1996; Fagenand Cowen 1996).

Kajian ini dijangka dapat menjana maklumattentang sejauh mana proses dalam keluargaberisiko terutamanya tingkah laku keibubapaanibu memainkan peranan dalam mempengaruhitahap penyesuaian anak (resiliensi, estim diridan pencapaian akademik). Maklumat sepertiini dijangka dapat mendedahkan proses yangwujud dalam keluarga berisiko di sampingmembantu memperluas kefahaman tentangkesan daripada pengalaman hidup dalamkeluarga berisiko terhadap pembangunan insan.


Tingkah Laku Keibubapaan dan Penyesuaian Tingkah Laku Anak dalam Keluarga Berisiko di Luar Bandar

KERANGKA KERJA KONSEPTUAL KAJIAN

Kajian ini memfokuskan kepada faktor yangdijangka boleh mempengaruhi tahap penyesuaiantingkah laku anak daripada keluarga berisiko diluar bandar. Teori Ekologi Manusia yangdipelopori oleh Bronfenbrenner (1979) telahdipilih sebagai salah satu panduan di dalammembina kerangka kerja konseptual kajian ini.Bronfenbrenner menyatakan bahawapembangunan kanak-kanak dipengaruhi bukanhanya oleh personalitinya tetapi juga olehgabungan pelbagai faktor dan konteks di manakanak-kanak itu berada. Kajian ini juga telahmenggunakan model kecekapan, modelresiliensi, model faktor risiko dan perkembangankanak-kanak dan model faktor pelindung danperkembangan kanak-kanak sebagai panduanuntuk menganalisis impak proses keluargakhususnya keibubapaan ke atas tahappenyesuaian anak (resiliensi, estim diri danpencapaian akademik). Huraian setiap modelini dibentangkan dalam perenggan seterusnya.

Model Kecekapan

Model kecekapan untuk fungsi keluarga (Watersdan Lawrence 1993) dipilih kerana bolehmenerangkan bagaimana proses ibu dan bapamelibatkan diri di dalam mendidik danmembesarkan anak menjadi seorang individuyang cekap dan berpotensi. Andaian model iniialah tingkah laku keibubapaan merupakan salahsatu sumber yang boleh digunakan untukberadaptasi terhadap persekitaran yangmencabar. Ibu bapa secara semula jadinya telahdimotivasikan untuk bertingkah laku bagi'memastikan dunia mereka berjalan denganlancar, membesar dan berubah' (Waters danLawrence 1993). Berdasarkan kepada modelkecekapan, kajian ini menggambarkan tingkahlaku keibubapaan sebagai manifestasi penguasaansepenuhnya terhadap persekitaran. Kajian inimelihat masalah tingkah laku keibubapaansebagai percubaan untuk beradaptasi terhadappersekitaran yang berisiko dengan sedikit kawalandaripada menggambarkan keluarga sebagai tidakberfungsi. Model ini juga mengandaikan ibuyang tinggal di dalam persekitaran mencabarmempunyai kekuatan yang boleh digunakan bagimenerangkan variasi kecekapan di kalangankanak-kanak.

Model Resiliensi

Resiliensi ditakrifkan sebagai kuasa ataukebolehan untuk mengatasi kesakitan, tekananatau kesulitan. Risiko merupakan konsep utamadi dalam model resiliensi yang berasal daribidang perubatan bagi mengkaji berkenaanwabak dan penyakit. Kajian ini mengandaikanbahawa masalah boleh dielakkan denganmengenal pasti proses yang dapat mengurangkankejadian risiko. Pendekatan faktor-risiko adalahamat berguna untuk memahami pembangunanManusia kerana ia menyatakan kewujudan faktorrisiko dalam diri individu, keluarga danpersekitaran boleh mengurangkan prosesperkembangan akan datang (Bogenschneider1996). Di samping itu, para saintis sosialmenunjukkan hasil kajian yang sama denganmodel perubatan apabila mereka mendapatisemakin meningkat timbunan faktor risiko yangwujud kemungkinan ketidakfungsian akanberkurangan (Murry dan Brody 1999; Lusterdan McAdoo 1994). Dengan kata lain, kombinasifaktor risiko (contoh: miskin, tinggal dalamkeluarga induk tunggal dan mempunyai ahlikeluarga yang berpenyakit kronik) akanmengehadkan potensi kanak-kanak untukberjaya.

Bagi menerangkan kenapa penimbunanfaktor risiko mempunyai impak yang berlainanterhadap perkembangan anak, para saintis sosialtelah mula mengkaji kanak-kanak yang tinggaldalam persekitaran yang merugikan denganmenggabungkan proses faktor pelindung danfaktor risiko. Proses perlindungan ditakrifkansebagai individu dan persekitaran yang bolehmeningkatkan keupayaan kanak-kanak untukbertahan terhadap kejadian yang negatif danmempromosikan adaptasi dan kecekapan(Steinberg, Mounts, Lamborn & Dornbusch1991).

Apa yang akan diterangkan seterusnyamerupakan gambaran kasar berkenaan keluargaberisiko yang menekankan kepada faktorpelindung dan risiko berhubung dengan estimdiri dan pencapaian akademik di kalangan kanak-kanak. Estim diri merupakan satu komponenresiliensi kerana dengan mempunyai persepsiyang tinggi terhadap pencapaian dan tanggapanyang positif terhadap diri mampu mewujudkanperhubungan dengan orang lain danmemperoleh kejayaan dalam kehidupan secaraamnya (Burnett 1996; Lawrence 1996).Pencapaian akademik yang baik akan

PertanikaJ. Soc. ScL 8c Hum. Vol. 10 No. 2 2002 167


merangsang aspirasi anak untuk terus menjadiseorang pelajar yang lebih cemerlang dan secaratidak langsung boleh meningkatkan estim diri(Filozof dan Albertin 1998; Fuligni 1997;Schreiber dan Schreiber 1995). Oleh yangdemikian, kanak-kanak yang mempunyaipandangan yang positif terhadap keupayaan dirisendiri mampu mengawal emosi dan keupayaanmental untuk menghadapi persekitaran yangnegatif dan tidak stabil. Ketidakupayaan untukmengawal emosi, perhatian dan mental bolehmenyebabkan kanak-kanak menjadi murung ataumenunjukkan tingkah laku bermasalah(Rozumah et al. 1999b).

Faktor Risiko dan Perkenibangan Kanak-kanakRisiko merupakan konsep yang biasa digunakanoleh bidang perubatan dan menyifatkannyasebagai potensi untuk mendatangkan hasil yangnegatif (Jens dan Gordon 1991). Secaratradisinya, faktor risiko telah digambarkansebagai keadaan yang meningkatkan hasil yangnegatif dan tidak diingini seperti morbidity danmortality. Bagaimanapun pada masa sekarang,definisi risiko telah diperluas lagi denganmelibatkan tingkah laku yang berkaitankesihatan, kesejahteraan dan sosial individu.Kanak-kanak dalam keluarga berisiko mempunyaipeluang yang tinggi untuk terlibat denganmasalah tingkah laku, mudah terpengaruhdengan ajakan negatif rakan sebaya danmempunyai estim diri yang rendah. Kajianterdahulu juga menunjukkan latar belakangkeluarga (contoh: pendidikan ibu, pendapatankeluarga dan bilangan anak) bolehmempengaruhi perkembangan kanak-kanak danfungsi sosial (Fuligni 1997; Pungello et al 1996;Garret et al. 1994). Kanak-kanak keluarga statussosioekonomi yang rendah didapati memper-sembahkan pencapaian yang kurang baik disekolah dan mempunyai lebih banyak masalahsosial dan kognitif jika dibandingkan dengankanak-kanak keluarga status kelas pertengahandan atasan (Shumow et al. 1998; Dodge et al.1994; Small dan Luster 1994). Kanak-kanakkeluarga berpendapatan rendah kebiasaannyatinggal dan bersekolah di kawasan yang miskindan mempunyai kadar jenayah yang tinggi(Connell dan Spencer 1994). Keluarga yangberpendapatan rendah juga cenderung untukmengalami lebih banyak tekanan hidup danketerpencilan sosial. Keadaan ini akanmenambahkan lagi tekanan psikologi yang

dihadapi oleh ibu bapa dan seterusnyamengakibatkan pertengkaran dan konflik diantara ibu bapa-anak meningkat (McLoyd 1990).Pertengkaran dan konflik yang wujud di antaraibu bapa-anak akan mengurangkan penyesuaiantingkah laku anak dalam keluarga berisiko(Vosler dan Proctor 1990).

Faktor Pelindung dan Perkembangan Kanak-kanakKajian tentang faktor pelindung terhasil daripadapengkaji yang berminat dengan prosesperkembangan anak yang terdedah padapsikopatologi dan memfokuskan ke atas variabelyang boleh menjadi penahan {buffer) untukmembantu mereka mengatasi penyakit mental.Faktor pelindung dicirikan sebagai semuatingkah laku dan perihal yang mengurangkankejadian atau hasil yang tidak diingini dari aspekkemahiran, kebolehan dan sumber yang dipunyaioleh ibu bapa dan keluarga (Murry dan Brody1999). Faktor pelindung mungkin termasukkawalan peribadi atau sosial (contoh: pengawasanibu bapa dan kepercayaan agama). Faktorpelindung mungkin mempunyai kesan secaralangsung ke atas perkembangan anak denganmerangsang anak untuk menentang faktor risikoyang wujud dalam keluarga atau mengubahsuaihubungan di antara risiko dan hasil perkem-bangan. Perbezaan dari segi tahap perkembangananak mungkin terjadi antara keluarga yangmempunyai banyak faktor risiko dan faktorpelindung berbanding keluarga yang berisikotinggi diiringi faktor pelindung yang rendah.

Kajian terdahulu menunjukkan tingkah lakukeibubapaan yang positif dalam keluarga berisikomampu menjadi faktor pelindung kepada anak(Franco dan Levitt 1998; Lamborn, Dornbuschdan Steinberg 1996). Ibu bapa statussosioekonomi rendah yang mengamalkanpeneguhan disiplin secara konsisten, kawalandan sokongan yang tinggi serta mempunyaikomunikasi yang positif didapati mempunyaianak yang kurang bermasalah dari segipenyesuaian tingkah laku.

METODOLOGI

Sampel

Sampel kajian terdiri daripada 201 respondenpasangan ibu-anak (berumur 11 tahun),berbangsa Melayu dan tinggal bersama. Sampelkajian ini merupakan subset sampel kajian IRPA"Ekologi Keluarga Berisiko: Proses Interaksi danPerkembangan Resilien Anak" yang berjumlah



1262 responden pasangan ibu dan anak yangberumur 11 tahun, 14 tahun dan 16 tahun.Responden ibu yang mempunyai keluargaberisiko telah dikenal pasti secara rawak melaluisenarai nama ketua isi rumah termiskin yangdiperoleh daripada Daftar dan Profil KeluargaTermiskin 1995/96, Kementerian PembangunanLuar Bandar. Kajian ini telah dijalankan dinegeri Kelantan, Kedah, Selangor dan Johor.

PENGUKURAN ANGKUBAH KAJIANAngkubah Bebas

Latar belakang keluarga yang telah diukur didalam kajian ini ialah taraf pendidikan ibu,pendapatan keluarga dan bilangan anak. Tarafpendidikan adalah jumlah tahun pendidikanyang diterima oleh responden ibu. Jumlahpendapatan keluarga dalam sebulan adalahgabungan pendapatan suami, isteri danpendapatan sampingan yang diperoleh keluargasebulan. Bilangan anak pula diukur berdasarkanjumlah bilangan anak responden.

Tahap risiko telah diukur berdasarkanjumlah faktor risiko yang kemungkinan wujuddalam sesebuah keluarga. Terdapat 19 faktorrisiko yang telah disenaraikan di dalam borangsoal selidik. Antaranya ialah keluarga induktunggal, tiada bilik tidur, keadaan rumahmembahayakan dan mempunyai ahli keluargaberpenyakit kronik. Keluarga yang terlibat dalamkajian mestilah mempunyai sekurang-kurangnyasatu faktor risiko. Skor risiko responden kajianini menjulat dari 1 hingga 11. Skor risikodibahagikan kepada dua tahap iaitu 'Rendahdan 'Tinggi'. Pembahagian dua tahap ini adalahberdasarkan kepada nilai purata yang diperolehkeluarga responden iaitu 5.3. Jumlah risiko yangmelebihi daripada nilai purata menunjukkantahap keluarga berisiko tinggi. Manakala jumlahrisiko yang rendah daripada nilai puratamenunjukkan tahap keluarga berisiko rendah.

Kualiti tingkah laku keibubapaan respondenibu diukur menggunakan skala 'SupportiveParenting' oleh Simons et al (1992) yangmengandungi sembilan (9) item positif. Nilaikoefisien alpha bagi skala tingkah lakukeibubapaan ini dari perspektif ibu ialah 0.79.Skor bagi setiap kenyataan yang dikemukakankepada responden ialah 1 = Tidak pernah, 2 =Hampir tidak pernah, 3 = Kadang-kadang, 4 =Hampir selalu dan 5 = Selalu. Jumlah skor dikiradengan mencampurkan semua skor yang didapatioleh responden bagi setiap item yang

dikemukakan. Jumlah skor minimum yang bolehdiperoleh responden adalah 9 dan skormaksimum adalah 45. Semakin tinggi skor yangdiperoleh ibu, semakin tinggi sokongan yangdiberikan kepada anak. Sokongan yang diberikanmenunjukkan ibu mengambil berat tentangperasaan anak, berminat dalam aktiviti harian,menyatakan perasaan sayang dan penerimaan,menggalakkan tingkah laku yang bersesuaian,membantu menyelesaikan masalah, dan memberipengukuhan terhadap pencapaian.

Angkubah Sandar

Tahap resiliensi anak telah diukur menggunakanskala ' Children i Resiliency Attitudes Scales' (C.R.A.S)(Bischoe dan Harris 1994). Instrumen inimengandungi tujuh subskala iaitu celik akal,berdikari, hubungan, inisiatif, kreativiti dankejenakaan, moraliti dan resiliensi umum.Instrumen ini mengandungi 28 item dengan 14item positif dan 14 item negatif. Setiap subskalamengandungi empat item (2 item positif dan 2item negatif). Responden dikehendaki menjawabsalah satu daripada jawapan yang dikemukakaniaitu l=sangat tidak setuju, 2=tidak bersetuju,3=tidak dapat diputuskan, 4=setuju dan 5=sangatsetuju. Setiap kenyataan negatif akan dikodkansemula iaitu dari 5= sangat tidak setuju hingga1= sangat setuju. Jumlah skor keseluruhan bagiskala resiliensi dikira dengan membahagikanjumlah keseluruhan yang diperoleh respondendengan 140 (jumlah skor maksimum yang bolehdiperoleh untuk skala resiliensi) didarabkandengan 100. Oleh itu skor minimum yang bolehdiperoleh responden ialah 20 dan skormaksimum ialah 100. Jumlah skor keseluruhandianalisis mengikut dua tahap yang ditentukanberdasarkan kepada nilai min yang diperolehresponden. Nilai skor yang tinggi daripada minmenunjukkan tahap resiliensi adalah tinggi.Manakala nilai skor yang rendah daripada minmenunjukkan tahap resiliensi adalah rendah.Nilai koefisien alpha bagi skala resiliensi anakdalam kajian ini adalah 0.64.

Tahap estim diri anak telah diukurmenggunakan skala 'Hare Self-Esteem Scale' (HSS)oleh Hare (1985). Skala ini mengandungi 3subskala yang mengkhususkan 3 elemen yangberkaitan dengan estim diri iaitu rakan sebaya,sekolah dan rumah. Terdapat 30 item (15 itempositif dan 15 item negatif dalam skala ini.Pilihan jawapan adalah berdasarkan kepadal=sangat tidak setuju, 2=tidak setuju, 3=setuju


Zarinah Arshat, Rozumah Baharudin, Rumaya Juhari & Rojanah Kahar

dan 4=sangat setuju. Setiap kenyataan negatifakan dikodkan semula iaitu 1=4, 2=4, 3=2 dan4=1. Skor maksimum yang boleh diperolehresponden ialah 120 dan skor minimum ialah30. Nilai skor yang tinggi daripada minmenunjukkan tahap estim diri yang tinggi. Nilaikoefisien alpha bagi skala ini ialah 0.76.

Pencapaian akademik dinilai berdasarkanjumlah lima (5) mata pelajaran terpilih iaituBahasa Melayu, Bahasa Inggeris, Matematik, Sainsdan Agama Islam. Pencapaian akademik telahdiperoleh daripada rekod keputusan peperiksaanakhir tahun 4. Bagi setiap subjek, markahtertinggi yang boleh diperoleh responden anakialah 100 dan markah terendah adalah 0. Skormarkah maksimum yang boleh diperolehresponden anak ialah 500. Semakin tinggi skormarkah yang diperoleh, semakin tinggipencapaian akademik yang diperoleh respondenanak.

HASIL KAJIAN

Ciri Latar Belakang Keluarga

Jadual 1 menunjukkan maklumat latar belakangkeluarga yang dikaji. Purata umur respondenibu yang terlibat adalah 39.8 tahun dengansisihan piawai 6.9. Majoriti (92.5%) respondenyang terlibat dalam kajian ini adalah keluargalengkap, manakala 7.5% lagi adalah induktunggal. Purata jumlah tahun pendidikan yangditerima oleh responden ialah 5.4 tahun, dengankebanyakan (48.3%) responden mendapatpendidikan di antara darjah 1 hingga darjah 6.Pendapatan bulanan keluarga responden secarapuratanya adalah sebanyak RM505.3 iaitu beradadi bawah tahap garis kemiskinan (RM510) yangdinyatakan dalam Rancangan Malaysia Kelapan(Unit Perancang Ekonomi 2001). Kajian inimendapati kebanyakan (66.7%) respondenkajian tidak bekerja, iaitu merupakan suri rumahsepenuh masa. Purata bilangan anak respondenadalah 5.7 orang.

Tahap Risiko Keluarga

Berdasarkan hasil yang telah diperoleh, kajianini mendapati hampir separuh (44.8%)responden kajian datang daripada keluargaberisiko tinggi. Manakala sebanyak 55.2%responden datang daripada keluarga berisikorendah. Secara umumnya, kebanyakan faktorrisiko yang wujud dalam keluarga adalah sepertiresponden ibu tidak mempunyai pekerjaan tetap(81.1%), keadaan rumah membahayakan(73.1%), pendidikan responden kurang daripada

SRP (72.6%) dan pendapatan bulanan kurangdaripada RM 516.

Kualiti Tingkah Laku KeibubapaanSecara keseluruhannya lebih separuh (55.2%)responden mempunyai kualiti tingkah lakukeibubapaan yang rendah. Jika diteliti mengikutsubsampel keluarga berisiko rendah dansubsampel keluarga berisiko tinggi juga, didapatilebih daripada separuh responden mempunyaikualiti tingkah laku keibubapaan yang rendahiaitu masing-masing 55.9% dan 54.4%. Keadaanini arnat membimbangkan kerana tingkah lakukeibubapaan yang berkualiti merupakan elemenpenting dalam membentuk perkembangansosioemosi dan perkembangan kognitif anakterutama pada peringkat awal kanak-kanak.Analisis data menunjukkan responden ibu yangmempunyai kualiti tingkah laku keibubapaanyang tinggi adalah sebanyak 44.8%. Bagisubsampel keluarga berisiko rendah didapatikualiti tingkah laku keibubapaan yang tinggiadalah sebanyak 44.1%. Manakala subsampelkeluarga berisiko tinggi pula, sebanyak 45.6%responden mempunyai kualiti tingkah lakukeibubapaan yang tinggi.

Penyesuaian Tingkah Laku AnakPenyesuaian tingkah laku anak dalam kajian inimerujuk kepada aspek resiliensi, estim diri danpencapaian akademik. Secara umumnya, tahappenyesuaian tingkah laku anak dalam membuatadaptasi terhadap faktor-faktor risiko yang wujuddalam keluarga berisiko kurang memuaskan(Jadual 2).

Secara keseluruhannya, lebih daripadaseparuh (55.7%) responden anak mempunyaitahap resiliensi yang rendah. Situasi ini amatmembimbangkan kerana anak yang mempunyaitahap resiliensi mudah terpengaruh dengananasir negatif apabila meningkat dewasa kelak.Walau bagaimanapun masih ramai lagi kanak-kanak dalam keluarga berisiko mempunyai tahapresiliensi yang tinggi iaitu sebanyak 44.3%.Keadaan ini menunjukkan bahawa anak darikeluarga berisiko tinggi masih mampu memantulsemula daripada kehidupan dan pengalamanyang sukar (Bland dan Sowa 1994) untuk menjadiseorang insan produktif.

Kajian mendapati peratusan (54.4%)responden anak daripada subsampel keluargaberisiko tinggi yang mempunyai tahap estim dirirendah adalah lebih tinggi jika dibandingkan



JADUAL 1Maklumat latar belakang responden (N=201)

Pemboleh ubah (%) Purata s.p mak.

Umur<35 tahun35-45 tahun46-55 tahun>55 tahun

Taraf PendidikanTidak bersekolahDarjah 1-6Tingkatan 1-2SRPSPMSTPMSijil

Jumlah Tahun Pendidikan<6 tahun6-10>10 tahun

PekerjaanBekerjaTidak bekerja

Pendapatan Bulanan RespondenRMORM100-500RM501-RM1000

Pendapatan Bulanan Keluarga<RM500RM500-RM1000>RM1000

Bilangan Anak<6 orang6-10>10 orang

Bilangan Ahli Isi rumah3-6 orang7-10 orang>10 orang

51 (25.4)106 (52.7)43 (21.4)1 ( 0.5)

47 (23.4)97 (48.3)1 ( 0.5)30 (14.9)22 (10.9)1 ( 0.5)3 ( 1.5)

68 (33.9)109 (54.2)24 (11.9)

67 (33.3)134 (66.7)

133 (66.7)64 (31.8)3 ( 1.5)

95 (47.3)103 (51.2)3 ( 1.5)

109 (54.2)86 (42.8)6 ( 3.0)

77 (38.3)115 (57.2)9 ( 4.5)

39.8 6.9 27 63

5.4 3.7 13

RM84.6

RM505.3

5.7

7.3

RM152.4

RM220.5

2.2

L9

RMO

RM150

RM1000

RM1700

15

15

Nota: s.p=sisihan piawai, min=minimum, mak=maksimum, SRP=Sijil Rendah Pelajaran,SPM-Sijil Pelajaran Malaysia, STPM=Sijil Tinggi Pelajaran Malaysia

dengan peratusan (46.8%) responden anakdaripada subsampel keluarga berisiko rendah.Hasil kajian menunjukkan secara keseluruhannyapencapaian akademik anak adalah memuaskandengan lebih daripada separuh (57.7%)responden anak memperoleh jumlah markahmelebihi nilai purata. Walau bagaimanapun,masih ramai lagi (42.3%) responden anakmempunyai pencapaian akademik yang rendah.

Keadaan ini amat membimbangkan kerana anakyang mempunyai pencapaian akademik yangrendah dikhuatiri berpotensi untuk terlibatdalam pelbagai masalah sosial sama ada disekolah atau di tempat lain.

Perkaitan di Antara Latar Belakang Keluargadengan Penyesuaian Tingkah Laku Anak

Hasil analisis korelasi Pearson menunjukkan



JADUAL 2Kualiti tingkah laku keibubapaan dan penyesuaian tingkah laku anak mengikut tahap risiko keluarga

n (%)

Variabel Keseluruhan(N=201)

SSKR SSKT(n=90)

Kualiti Tingkah laku KeibubapaanTinggiRendah

Purata:s.p:minimum:maksimum:

Tahap Resiliensi AnakTinggiRendah

Min:s.p:minimum:maksimum:

Tahap Estim Diri AnakTinggiRendah -


Tahap Pencapaian Akademik AnakTinggiRendah


90 (44.8)111 (55.2)

32.4

5.21744

89 (44.3)112 (55.7)

53.96.4

34.380.7

101 (50.2)100 (49.8)

83.6

7.961105

116 (57.7)85 (42.3)284.1

76.556439

49 (44.1)62 (55.9)

32.64.5

2142

48 (43.2)63 (56.8)

54.26.4

34.380.7

59 (53.2)52 (46.8)

83.9

8.161105

61 (55.0)

50 (45.0)304.266.6

96439

41 (45.6)49 (54.4)

32.05.91744

41 (45.6)49 (54.4)

53.66.4

34.370.7

41 (45.6)

49 (54.4)

83.27.866

105

50 (55.6)40 (44.4)

259.280.8

56411

Nota: SSKR = Subsampel keluarga berisiko rendahSSKT = Subsampel keluarga berisiko tinggis.p = sisihan piawai

wujud perkaitan yang signifikan dan positif diantara pendidikan yang diperoleh ibu denganesiim diri (r=0.18, p*0.05) dan pencapaianakademik (r=0.19, p^O.01) responden anakkeseluruhan (Jadual 3). Dalam Jadual 3, estimdiri anak mempunyai perkaitan yang signifikandengan tingkah laku keibubapaan (r=0.32,p^O.Ol) dalam keluarga berisiko tinggi. Kajianjuga mendapati pendapatan bulanan keluargatidak mempunyai perkaitan yang signifikandengan penyesuaian tingkah laku anak. Kajianmendapati bilangan anak yang terlalu ramaiboleh mengurangkan tahap resiliensi respondenanak secara keseluruhan (r^-0.16, ps;0.05). Hasil

yang sama juga diperoleh dalam subsampelkeluarga berisiko rendah iaitu bilangan anakmempunyai perkaitan yang signifikan dan negatif(r=0.27, pssO.Ol) dengan resiliensi anak. Analisisdata menunjukkan bahawa tahap estim diridipengaruhi oleh bilangan anak (r=0.21, p^ 0.05)dalam subsampel keluarga berisiko tinggi.

Perkaitan di Antara Tahap Risiko Keluargadengan Penyesuaian Tingkah Laku AnakHasil analisis pekali korelasi Pearsonmenunjukkan terdapat perkaitan yang signifikandan negatif (r=-0.26, p<; 0.01) di antara tahaprisiko dengan pencapaian akademik anak. Ini



JADUAL 3Perkaitan di antara latar belakang keluarga, tahap risiko, tingkah laku keibubapaan dan penyesuaian tingkah

laku anak (resiliensi, estim diri dan pencapaian akademik)

Variabel

Latar belakangKeluargaPendidikan ibuPendapatanKeluargaBilangan anak

Tahap risiko

Tingkah lakuKeibubapaan

Resiliensi

Keseluruhan SSKR

0.050.02

-0.16*

-0.06

0.05

0.100.07

-0.27**

-0.11

0.18

SSKT

-0.03-0.10

-0.04

0.03

-0.08

Estim

Keseluruhan

0.18*0.06

0.06

-0.07

0.17*

Diri

SSKR

0.060.06

0.21*

-0.04

0.11

SSKT

0.32**0.03

-0.07

-0.09

0.23*

Keseluruhan

0.19**0.08

-0.07

-0.26**

0.02

PencapaianAkademik

SSKR

0.19-0.04

-0.05

-0.09

-0.10

SSKT

0.180.05

0.08

-0.00

0.08

Nota: SSKR= Subsampel keluarga berisiko rendahSSKT= Subsampel keluarga berisiko tinggi* psO.05

p*** psO.OOl

menunjukkan bahawa persekitaran keluarga yangberisiko tinggi akan menjejaskan prestasiakademik anak. Walau bagaimanapun, tahaprisiko didapati tidak memberi pengaruh yangsignifikan ke atas tahap estim diri dan resiliensianak.

Perkaitan di Antara Kualiti Tingkah LakuKeibubapaan dengan Penyesuaian Tingkah LakuAnak

Hasil yang diperoleh menunjukkan kualititingkah laku keibubapaan yang tinggi mampumeningkatkan tahap estim diri responden anakkeseluruhan (r=0.l7, ps 0.05). Analisis datajugamenunjukkan tingkah laku keibubapaanmempunyai perkaitan yang signifikan (r=0.23,p^ 0.05) dengan estim diri anak bagi subsampelkeluarga berisiko tinggi. Walau bagaimanapun,tingkah laku keibubapaan didapati tidakmemberi pengaruh yang signifikan ke atas tahapresiliensi dan pencapaian akademik anak.

Prediktor Penyesuaian Tingjkah Laku Anak

Analisis regresi berganda mendapati tahappendidikan ibu (p =.16, p ^05) dan tingkah lakukeibubapaan (P =.15, ps: .05) mempunyaiperkaitan secara kelainan dengan estim diri anakkeseluruhan (F=5.55, p* 0.01) (Jadual4). Kajianmendapati tahap pendidikan ibu (p =.18, p «s.O5)

juga menyumbang secara unik ke atas pencapaianakademik anak (F=2.87, ps: 0.05).

KESIMPULANBerdasarkan hasil yang telah diperoleh, kajianini membuat kesimpulan bahawa hampir separuh(44.8%) responden kajian datang daripadakeluarga berisiko tinggi. Manakala sebanyak55.2% responden datang daripada keluargaberisiko rendah. Kajian mendapati lebih daripadaseparuh (55.2%) responden ibu mempunyaikualiti tingkah laku keibubapaan yang rendah.Secara umumnya, penyesuaian tingkah laku anakdalam keluarga berisiko yang dikaji berada padatahap yang agak membimbangkan. Analisis datamenunjukkan hampir separuh responden anakmempunyai tahap estim diri (49.8%) danpencapaian akademik (42.3%) yang rendah.Manakala lebih daripada separuh (55.7%)responden anak mempunyai tahap resiliensi yangrendah.

Bronfenbrenner (1979) mengandaikanbahawa penyesuaian tingkah laku anak (resiliensi,estim diri dan pencapaian akademik) bolehdipengaruhi oleh faktor dan konteks di manaanak tinggal seperti persekitaran keluarga(pendidikan ibu, pendapatan keluarga, bilangananak, tahap risiko dan tingkah lakukeibubapaan). Sebahagian daripada hasil kajian


JADUAL 4Prediktor penyesuaian tingkah laku anak

pto

Angkubah Resiliensi,p

Latar belakangKeluargaPendidikanPendapatan keluargaBilangan anak

Latar belakangKeluarga danTingkah lakuKeibubapaanPendidikanPendapatan keluargaBilangan anakTingkah lakuKeibubapaan

N(201)

Model 1

.02

.03-.16*

R2=0.03F=L87

Model 2

.02

.03-.16*.02

R2=0.03F-1.41

SSKR(lll)

Model 1

.08

.11-.28**

R2=0.09F=3.67*

Model 2

.06

.11-.26**

.14

R2=0.11F=3.30*

SSKT(90)

Model 1

-.04-.10-.04

R2=0.01F=0.37

Model 2

-.03.-09-.06-.08

R2=0.12F=0.41

Penyesuaian Tingkah

N(201)

Model 1

.19*.04.09

R2=0.04F=2.93*

Model 2

.17*.04.12

.17*

R2=0.07F=3.63**

Estim Diri,

Model 3

.16*

.15*

R2=0.05F=5.55**

Laku Anak

SSKR(llll)

Model 1

.08

.03.21*

R2-0.05F=1.96

Model 2

.06

.04

.23*.13

R2=0.07F=1.97

SSKT(90)

Model 1

.31**.04-.03

RMM0F-3.25*

Model 2

.29**.02.01.19

R2=0.14F-3.40*

Pencapaian Akademik,p

N(201) :

Model 1

.18

.06-.03

R2-0.04F=2.87*

Model 2

.19

.06-.04-.02

R2=0.04F=2.16

SSKR(lll)

Model 1

.02-.03-.05

R2=0.00F-0.15

Model 2

.04-.03-.07-.12

R2-0.02F=0.48

SSKT(90)

Model 1

.19

.04

.10R2=0.04F=1.32

Model 2

.18

.03

.12

.08

R2=0.05F=l.ll

Nota: Model 3 tidak dibina untuk resiliensi, estim din dan pencapaian akademik anak subsampel keluarga berisiko rendah dansubsampel keluarga berisiko tinggi kerana tiada variabel yang didapati signifikan dalam Model 1 dan Model 2.*p=£.O5, **p<;0.001N = Sampel KeseluruhanSSKR • Subsampel Keluarga Berisiko RendahSSKT • Subsampel Keluarga Berisiko Tinggi


ini adalah konsisten dengan andaian yang dibuatoleh Bronfenbrenner. Penyesuaian tingkah lakuanak dalam keluarga berisiko didapatidipengaruhi oleh pendidikan ibu, bilangan anak,tahap risiko dan tingkah laku keibubapaan.Walau bagaimanapun pendapatan bulanankeluarga didapati tidak mempunyai perkaitanyang signifikan dengan mana-mana aspekpenyesuaian tingkah laku anak yang dikaji.

Secara keseluruhan, kajian mendapati wujudperkaitan yang signifikan dan negatif antarabilangan anak dengan resiliensi kanak-kanak.Pola yang sama juga ditunjukkan dalamsubsampel keluarga berisiko rendah. Kajianmendapati bilangan anak yang ramai dalamsubsampel keluarga berisiko rendah memberikesan negatif kepada perkembangan resilienanak. Hasil kajian ini adalah konsisten denganmodel faktor risiko dan perkembangan kanak-kanak (Shumow et al 1998; Dodge et al. 1994;Sumow dan Luster 1994; Jens dan Gordon 1991)yang dibincangkan dalam kerangka kerjakonseptual. Model ini mengandaikan bilangananak yang ramai merupakan faktor risiko yangboleh memberi impak negatif kepadaperkembangan resiliensi anak. Berdasarkankepada model resiliensi (Murry dan Brody 1999;Luster dan McAdoo 1994; Bogenschneider 1996)tahap resiliensi kanak-kanak dalam keluargaberisiko boleh ditingkatkan dengan kehadiranfaktor pelindung. Bagaimanapun dalam kajianini didapati tidak ada satu pun variabel yangdikaji boleh dijadikan faktor pelindung untukmembantu anak memantul semula dari tekananhidup. Keadaan ini menjelaskan sebab tahapresiliensi anak yang terlibat dalam kajian adalahrendah.

Estim diri responden anak keseluruhandidapati dipengaruhi oleh pendidikan yangditerima oleh ibu dan tingkah laku keibubapaan.Bilangan anak pula didapati mempunyaiperkaitan yang signifikan dengan estim diri hanyadalam subsampel keluarga berisiko rendah.Analisis regresi berganda menunjukkan tahappendidikan ibu dan tingkah laku keibubapaanmempunyai perkaitan yang signifikan secarakelainan dengan estim diri anak. Modelkecekapan (Waters dan Lawrence 1993) danmodel faktor pelindung dan perkembangankanak-kanak (Murry dan Brody 1999; Francodan Levitt 1998; Lamborn Dornbusch danSteirnberg 1996) mengandaikan pendidikan

ibu dan tingkah laku keibubapaan bolehdijadikan sumber dan faktor pelindung untukmeningkatkan tahap estim diri anak di kalangankeluarga berisiko. Oleh yang demikian, hasilkajian ini adalah konsisten dengan kedua-duamodel tersebut.

Hasil analisis korelasi Pearson menunjukkanpencapaian akademik anak keseluruhandipengaruhi oleh tahap risiko dan pendidikanyang diterima oleh ibu. Analisis regresi bergandapula menunjukkan tahap pendidikan ibumenyumbang secara unik ke atas pencapaianakademik anak. Oleh yang demikian tahappendidikan ibu merupakan sumber kekuatanyang dipunyai oleh ibu untuk membantunyamendidik dan membesarkan anak dalampersekitaran keluarga yang berisiko. Maka hasilkajian ini menyokong andaian model kecekapanyang telah dibincangkan sebelum ini.

IMPLIKASI

Hasil kajian ini adalah konsisten dengan TeoriEkologi Manusia (Bronfenbrenner 1979) danjuga kajian terdahulu yang mendapati bahawafaktor latar belakang keluarga dan tingkah lakukeibubapaan boleh mempengaruhi perkembanganemosi dan kognitif anak. Kajian ini jugamenyokong andaian model kecekapan (Watersdan Lawrence 1993) yang menyatakan ibu yangtinggal dalam persekitaran yang mencabarmempunyai sumber kekuatan dalaman diri yangboleh digunakan untuk membentuk anak yangcekap dan berpotensi. Oleh yang demikian kajianini memberi implikasi bahawa pendidikan yangditerima oleh ibu merupakan sumber kekuatandiri yang boleh mempengaruhi tahap estim diridan pencapaian akademik anak dalam keluargaberisiko di luar bandar.

Ibu bapa yang berpendidikan tinggikebiasaannya melihat hubungan anak sebagaidua hala, bersifat timbal balik (Dekovic danGems 1992), banyak melibatkan diri dalamaktiviti pembelajaran anak dan mengamalkankomunikasi yang positif dan terbuka. Komunikasiyang terbuka membolehkan ibu bapamenunjukkan kasih sayang melalui sentuhan,kemesraan dan sokongan. Adalah menjadi fitrahmanusia untuk disayangi dan menyayangi. Tanpakasih sayang, seseorang itu mungkin menghadapipelbagai masalah emosi (Kamarudin 1996).Justeru itu, ibu bapa perlu mempelajaribagaimana untuk menjadi seorang pendengar



yang aktif danjuga mengesahkan keperluan danpermasalahan anak-anak mereka (Rak danPatterson 1996).

Andaian model faktor risiko danperkembangan kanak-kanak (Shumow et aL1998; Dodge et aL 1994; Small dan Luster 1994;Jens dan Gordon 1991) yang telah dibincangkansebelum ini adalah konsisten dengan hasil kajianini. Kajian mendapati bahawa bilangan anakyang ramai memberi impak negatif ke atas tahapresiliensi anak. Ini kerana bilangan anak yangramai boleh mengurangkan perhatian dan kasihsayang yang diberikan oleh ibu bapa kepadaanak. Di samping itu, ibu bapa yang mempunyaianak yang ramai lebih cenderung menggunakanpendekatan autoritarian dan permisif dalammendidik dan membesarkan anak yang memberiimpak negatif ke atas perkembangan anak(Luster dan McAdoo 1994; Menaghan dan Parcel1991; McLoyd 1990). Hasil kajian jugamenunjukkan kualiti tingkah laku keibubapaanmempengaruhi tahap estim diri anak. Justeruitu, program intervensi yang inovatif dankomprehensif bagi meningkatkan kemahirantingkah laku keibubapaan dan fungsi keluargaperlu diadakan bagi membantu ibu bapamenjana kemahiran dalam merangsangperkembangan anak.

RUJUKANBISCHOE, B. dan B. HARRIS. 1994. Resiliency Attitudes

Scale. Oklahoma: University of Oklahoma.

BURNETT, P. 1996. Gender and grade differences inelementary school-going children's descriptiveand evaluative self-statements and self-esteem.School Psychology International 17: 159-170.

BRONFENBRENNER, U. 1979. The Ecology of HumanDevelopment. Cambridge, MA: HarvardUniversity.

BOGENSCHNEIDER, K. 1996. An ecological risk/protective theory for building preventionprograms, policies, and community capacity tosupport youth. Family Relations (Online) 45(2):127. Available: http://gw6. epnet.com/fulltext.asp? results... Iding%20prevention%programs&gotoFT-true. [1999, Jan 12].

CONNELL, P. J. dan M. B. SPENCER. 1994. Educationalrisk and resilience in African-American youth:Context, self, action and outcomes in school.Child Development 65: 493-506.

DEKOVIC, M. dan J. R. M. GERRIS. 1992. Parentalreasoning complexity, sosial class, and child

rearing behaviors. Journal of Marriage and theFamily 54: 678-685.

DODGE, K. A., G. S. PETTIT dan J. E. BATES. 1994.Sosialization mediators of the relation betweensocioeconomic status and child conductproblems. Child Development 65: 649-665.

DUMKA, L. E. dan M. W. ROOSA. 1995. Usingresearch and theory to develop preventionprograms for high risk families. Family Relations(Online) 44(1): 78. Available: Wysiwyg://bodyframe.35/http://ehostweb5. epnct.com/fultext.asp. (1999, Sept. 14)

FAGEN, D. B. dan E. L. COWEN. 1996. Relationshipsbetween parent-child relational variables inhighly stressed urban families families. ChildStudy Journal (Online) 26(2): 87. Available: h t t p : / / g w 2 . e p n e t . c o mfulltext.asp? results...child%20relationalMuzzyTerm =&gotoFT=true. [1998, Dec 15]

FRANCO, N. dan M. J. LEVITT. 1998. The sosialecology of middle childhood: Family support,friendship quality and self-esteem (A). FamilyRelations (Online) 47(4): 315. Available: http:// g w 6 . e p n e t . c o m / f u l l t e x t . a s p ? -r e s u l t S . . .-m = & f u z z y T e r m =Family%ecology&gotoFr=true [1999, Jan 11].

FILOZOF, E. M dan H. K. ALBERTIN. 1998. Relationshipof adolescent self-esteem to selected academicvariable. Journal of School Health (Online) 65(2):68. Available: http:/ /gw9.epnet.com/full text. asp?-resu...=adolescent%20se If %2desteem&fuzzyTerm= [1999. (1999, October10)

FULIGNI, A. J. 1997. The academic achievement ofadolescents from immigrant families: The rolesof family background, attitudes and behavior.Child Development 68(2): 351-363.

GARRET, P., N. NG'ANDU dan J. FERRON. 1994. Povertyexperiences of young children and the qualityof their environment. Child Development 65:331-343.

GERARD, J. M. dan C. BUEHLER. 1999. Multiple riskfactors in the family environment and youthproblem behaviors. Journal of Marriage and theFamily 61(2): 343.

GORDON, K. A. 1996. Resilient Hispanic youths' selfconcept and motivational patterns. HispanicJournal of Behavioral Sciences (Online) 18(1):63. Available: http:/ /gw2.epnet.com/fultext.asp?-resultS...tNum=171&boolenTerm=resilient&fuzzyTerm= (1999, July 25).


Tingkah Laku Keibubapaan dan Penyesuaian Tingkah Laku Anak dalam Keiuarga Berisiko di Luar Bandar

HARE, B. R. 1985. The HARE general and areaspecific 9 school, peer and home self-esteemscale. Unpublished manuscript Departmentof Sociology, SUNY Stony Brook, New York.

HARTOS, j . L. dan T. G. POWER. 1997. Mother'sawareness of their early adolescent's stressor:Relation between awareness and adolescentadjustment. Journal of Early Adolescence 17(4):371-390.

HASHIMA, P. Y. dan P. R. AMATO. 1994. Poverty,social support and parental behavior. ChildDevelopment 65: 394-403.

JENS, K. G. dan B. N. GORDON. 1991. Understandingrisk: Implications for tracking high-risk infantsand making early service decisions. InternationalJournal of Disability 38: 211-224.

KAMARUDIN, H. 1996. Sekolah dan Perkembangan Kanak-kanak Sekolah. Kuala Lumpur : Lohprint Sdn.Bhd.

LAMBORN, S. D., S. M. DORNBUSCH dan L. STEINBERG.1996. Ethnicity and community context asmoderators of the relations between familydecision making and adolescent adjustment.Child Development 67: 283-301.

LAWRENCE, D. 1996. Enhancing Self-esteem in theClassroom. London: Paul Chapman Publishing.

LUSTER, T. dan H. P. MCADOO. 1994. Factors relatedto the achievement and adjustment of youngAfrican American children. Child Development65: 1080-1094.

MAYHEW, K P. dan J. D. LEMPERS. 1998. The relationamong financial strain, parenting, parent self-esteem, and adolescent self-esteem. Journal ofMarriage and the Family (Online) 18(2): 145.Available : http://gw9.epnet.com/fulltext.asp?-r e s u . . . = a d o l e s c e n t % 2 0 s e l f % 2 desteem&fuzzyTerm= (1999, Jun 10)

MENAGHAN, E. G. dan T. L. PARCEL. 1991.Determining children's home environment:The impact of maternal characteristics andcurrent occupational and family conditions.Journal of Marrige and the Family 53: 417-431

MCLOYD, V. C. 1990. The impact of economichardship on black families and children:Psychological distress, parenting, andsocioemotional development. Child Development61: 190-198.

MILLER, J. E dan D. DAVIS. 1997. Poverty history,marital history and quality of children's homeenvironments. Journal Marriage & the Family

(Online) 59(4): 996. Available: h t tp : / /g w 6 e p n e t . c o m / f u l l t e x t . a s p /resul t s . . . %20and%20quality% of % 20child&gotoFl>true (1999, Jan 12)

MURRY, V. M dan G. H. BRODY. 1999. Self-regulationand self worth of black children reared ineconomically stressed, rural, single mother-headed families. Journal ofFasmily Issues 20(4):458-484.

PUNGELLO, E . P . , J . B . KUPERSMIDT, M . R BURCHINAL

dan C.J. PATTERSON. 1996. Environmental riskfactors and children's achievement frommiddle childhood to early adolescence.Developmental Psychology 32(4): 755-767.

RAK, C. F. dan L. E. PATTERSON. 1996. Promotingresilience in at-risk families. Journal of Counseling& Development (Online) 74(4): 368. Available :http//gw5.epnet.com/fultext.aspPresultS-... a n d % 2 0 c h i l d r e n 2 0well%2dbeing&gotoFT=true. (1999, Jan 7).

ROZUMAH, B., Y. SITI NOR, M. ABDULLAH AL-HADI, B.MAZNAH dan M. T. AIDAH. 1999a. KajianKeruntuhan Institusi Keiuarga Islam Malaysia.Selangor : Universiti Putra Malaysia.

ROZUMAH, B., Y. SITI NOR, M.K, ABDULLAH AL-HADI,K ROJANAH, M. T. AIDAH dan K.H. KOH. 1999b.Impak Pelbagai Struktur Keiuarga TerhadapKehidupan Anak. Jabatan PembangunanManusia dan Pengajian Keiuarga, FakultiEkologi Manusia, Universiti Putra Malaysia.

SCHREIBER, E. H. dan K N. SCHREIBER. 1995. Usingrelaxation techniques and positive self-esteemto improve academic achievement of collegestudents. Psychological Reports 76(3): 929-930.

SHUMOW, L., D. L. VANDELL dan J. IC POSNER. 1998.Harsh, firm, and permissive parenting in low-income families. Journal of Family Issues 19(5):483-507.

SIMONS, R. L., F. O. LORENZ, R. D. CONGER dan C.Wu. 1992. Support from spouse as mediatorand moderator of the disruptive influence ofeconomic strain on parenting. Child Development63: 1282-1301.

SMALL, S. A. dan T. LUSTER. 1994. Adolescentssexual activity: An ecological, risk factorapproach. Journal of Marriage and the Family56: 18M92.

STEIRNBERG, N. S., L. MOUNTS, S. D. LAMBORN danS.M. DORNBUSCH. 1991. Authoritative parentingand adolescent adjustment. Journal of Researchon Adolescence 1: 19-36.



STERN, S.B., C. A. SMITH dan S.J.JANG. 1999. Urban

families and adolescent mental health. SocialWork Research (Online) 23(1): 15. Available:w y s i w y g : / / b o d y f r a m e . 3 5 / h t t p : / /ehostweb5epnet.com/fulltext.asp. (1999, Sept14).

SOLIS-CAMARA, P. R. dan R. A. Fox. 1996. Parentingpractices and expextations among mexicanmothers with young children. Journal of GeneticPsychology (Online) 157(4): 465. Available:h t t p : g w l O e p n e t . c o m /full text.asp? resul t . . .eanTerm=parental%20behavior%20&fuzzyTerm= [1999, July 26]

VOSLER, N. R. dan E. K. PROCTOR. 1990. Stress and

competence as predictors of child behaviorproblems. Social Work Research and Abstracts(Online) 26(2): 3. Available : h t t p : / /g w 6 e p n e t . c o m / f u l l t e x t . a s p -? re s u l t S . . . o o l e a n T e r m = % 2 0 c h i l d %20behavior&fuzzyTerm=.(1998. Dec 16).

VOYDANOFF, P. dan B. W. DONNELLY. 1998. Parents'

risk and protective factors as predictors ofparental well-being and behavior. Journal ofMarriage and the Family (Online) 60(2): 344.Available: htpp://gwlO.epnet.com/fulltext.-

a s p ? r e s u l t . . . e a n T e r m = p a r e n t a l %20behavior%20&fuzzyTerm= (1999, July 26).

VOYDANOFF, P. 1990. Economic distress and familyrelations: A review of the eighties. Journal ofMarriage and the Family 52: 1099-1115.

WATERS, D. dan E. LAWRENCE. 1993. Competence,

Courage, and Change: An Approach to Therapy.New York: Norton.

WAXMAN, H. C. dan S. Y. L. Huang. 1996. Motivationand learning environment differences in inner-city middle school students. Journal ofEducational Research (Online) 90(2): 93.Available : http://gw2.epnet.com/fulltext-.aspresultS...tNum = 129&booleanTerm=resilient&fuzzvTerm=(1999. July 25).

. 2001. Laporan Rancangan MalaysiaKelapan. Kuala Lumpur: Unit PerancangEkonomi, Jabatan Perdana Menteri.

(Diterima; 29 Mei 2003)



Subject Index for Volume 10 Nos. 1 and 2, 2002

Adjectives 150Agihan pendapatan 117-122, 127Agresif 10, 12, 14-22Akaike's biased corrected information criterion

136Analisis korelasi Pearson 171, 175Analisis regresi 165Analisis reliabiliti 13-14Analisis statistik inferensi 12Analitikal 10, 12, 14-24Angkubah bebas 169Angkubah sandar 169ANOVA 17, 102, 104ARCH see Autogressive Conditional

Heteroskedasticity modelAspirasi keusahawanan 53-55, 57, 59-61Assurance 101Athletic directors 1, 3-7Augmented Dickey Fuller method 77-79, 134, 138Autogressive Conditional Heteroskedasticity model

86-87, 92-93, 138-139Autonomi 10, 12, 14-23Autoregressive 123, 131

Bahasa Malaysia 109-116Banking 109-110, 112-113, 115-116Bayaran pindahan 117Benefits 27Bersandar 10, 12, 14-22, 24Bidang pengkhususan 9, 11-12, 17, 22, 24Bilingualism 109-110

CDLT see Listening Test on Compact DiscCharisma 64Charismatic personality 2Children's Resiliency Attitudes Scales 169Chi-squared critical value 136Ciri-ciri keusahawanan 53-55, 57-58-59Coaches' job satisfaction 1-6Coaches' perception 3Code choice 115Collectivism-individualism 100-101Commercial development 109, 116Clients'communication 115Computers 27, 31-33, 38-39Consumer Price Index 134Contingent reward 64Correlation matrix 102Correspondence 115Cultural values 99Culture studies 97Culture-service quality relationships 102

Currency derivative 86Currency options 86

Deskriptif 157Diagnostics checking 91Discriminant analysis 1, 4Durbin-Watson d statistics 77-81

EGARCH see Exponential GeneralisedAutogressive Conditional Heteroskedasticitymodel

Egocentric 150Ekstrovert 10, 12, 14-23e-Learning 27, 29Empathy 101Employees'communication 115English 109-115ESTAR see Exponential Smooth Transition

AutoregressiveEstim diri 165, 167, 169-170, 173, 175Exchange rates 85-86, 89, 91Executive-level employees 109, 112-113Exponential Generalised Autogressive Conditional

Heteroskedasticity model 87-89, 91Exponential Smooth Transition Autoregressive

133-140

Face-threatening act 146Faktor pelindung 168Faktor risiko 168, 175Familiarity ethnic 46Familiarity non-ethnic 46Female speech styles 145-146Fillers 150Forecasting 85Forecasting accuracy 131Future contracts 74

GARCH see Generalised Autogressive ConditionalHeteroskedasticity model

Gender characteristics 144Gender differences 143-144Gender stereotype 143-144Generalised Autogressive Conditional

Heteroskedasticity model 85, 88-94, 138

Hedging 146, 149-150Hipotesis Kuznets 119-120, 127

Individualized consideration 64Inferensi 15Inspirational motivation 64


Instructional leadership 63, 65, 67, 69-70Instructional leadership behaviour 65, 66-67Intelektual 10, 12, 14-24Intellectual stimulation 64, 66-67Interpretif 157-158, 162-163Intra-cultural variations 102Introvert 10, 12, 14-22, 24Inventori Personaliti Sidek 10, 12-15

Jantina 9, 11-12, 18-19Job performance 2Junior Eysenck Personality Inventory 10

Kaedah kebolehpercayaan alpha Cronbach 14Kaedah persamaan regresi 119Kajian deskriptif 11-12Kawalan peribadi 168KDNK see Keluaran Dalam Negara KasarKejujuran 14Keluaran Dalam Negara Kasar 117, 123-127Keluaran Negara Kasar 121Keluarga berisiko 171-173Kemiskinan 117Kepelbagaian 10, 12, 14-22Kerelevanan optimal pendengar 156Kesamaan interpretif 157-158, 163Ketahanan 10, 12, 14̂ 24Keterbukaan ekonomi 117, 122, 125Keusahawanan 53-57, 59KNK see Keluaran Negara KasarKolmogorov-Smirnov test 45Konsep masa 153Kritik diri 10, 13-21, 24Kualiti tingkah laku keibubapaan 170-176Kuasa dua terkecil see Ordinary least square

Language competency 114Language proficiency 114-116Latar belakang keluarga 170, 173-174Latihan dan pendidikan keusahawanan 53, 58Lead and lag relationships 73-74, 81Leadership behaviours 2Learners' interests 44Lexical traits 145, 148Likelihood ratio test 138Listener characteristics 43-44Listening Test on Compact Disc 45Literacy text 143Ljung-Box portmanteau Q test 137Logistic regression model 1, 4

MacKinnon critical values 79MAFE see Mean Absolute Forecast ErrorMalaysia Derivatives Exchange 75Malaysian music 43-44, 48

Management by exception 64MAPFE see Absolute Percentage Forecast ErrorMasculinity-femininity 100-101, 104Matrikulasi 9-12, 15-19, 21-24MDEX see Malaysia Derivatives ExchangeMean Absolute Forecast Error 132, 140Mean Absolute Percentage Forecast Error 132,

140Mean preference values 45Meeting 115Mengawal 10, 13-22Menolong 10, 13-21, 23Minnesota satisfaction questionnaire 1, 4Modal adverbs 150Modal auxiliary 150Model faktor pelindung dan perkembangan

kanak-kanak 167Model faktor risiko dan perkembangan kanak-

kanak 167, 175Model kecekapan 167, 175Model resiliensi 167Mono-cultural context 102-103Morbidity 168Mortality 168Multicultural 109-110Multilingual 109-110Multiple regression 78Music education 43Music familiarity 43-48Music preference 43-49Music stimuli 45Musical training 45-47

Non-linear time series 131Non-parametric Phillips-Perron unit root test 134

Oratory skills 2Ordinary least square 121Pearson product-moment correlation analyses 46

Pedagogy 43Pekali Gini 117-118, 120-126Pekali kebolehpercayaan alpha Cronbach 14̂ 15Pembolehubah pelaburan langsung asing 120Pencapaian 10, 13-23Pencapaian akademik 170-173Penterjemahan Arab-Melayu 153Penterjemahan langsung 158-159, 161, 164Penyesuaian tingkah laku anak 170-173Petikan langsung 158-159Petikan penuh 162Phonological traits 145, 148Potensi keusahawanan 56-61Power distance 100-101Pragmatik 153-154Prediktor penyesuaian tingkah laku anak 173-174


Preference ethnic 45-46Preference non-ethnic 45-46Profil personaliti 9, 11-12, 15, 17-19, 21-24Purchasing Power Parity 132

Random walk 85Reliability 101Responsiveness 101RMSFE see Root Mean Square Forecast ErrorRoot Mean Square Forecast Error 132, 140

School effectiveness items 68-70School leadership and management 63Self-Excited Threshold Autoregressive 132-133SERQUAL dimensions 99, 101, 103-104Service quality 97-98, 100-101Service quality perception 103-104SETAR see Self-Excited Threshold AutoregressiveSimple random sample 3Sixteen Personality Factor Questionaire 10Smooth Transition Autoregressive 131-135, 140Sokongan 10, 13-22, 24Speech styles 143, 151Sport administrators 2-3, 6Sports settings 2-3, 6-7,Standardized item alpha 13STAR see Smooth Transition AutoregressiveStatistik deskriptif 15, 17Stock index 74, 81Stock index futures 73, 81Stock indices 74Strategic vision 2Stratified random sample 3Struktur 10, 13-21, 23Student's acceptance 31Students 27Students' pattern 27Students' perception 30-31Students' preferences 47, 49

Subordinates' job satisfaction 2, 6Superior diagnostic 2Supportive Parenting Scales 169Syntactic-pragmatic traits 145, 148

Tahap resiliensi 165, 169-170, 173Tahap risiko keluarga 170, 172-173, 175Tangibles 101Teen culture 43-44Tennessee Self-Concept Scale 10Teori Ekologi Manusia 167, 175Teori Relevan 153-158, 160-164Teori Tipologi Holland 22Tingkah laku keibubapaan 165Tipologi teks 158Total familiarity 46Total preference 45-46Transactional leadership 2, 5-7, 63-65Transactional leadership behaviour 65, 68Transformational leaders 2, 6Transformational leadership 63, 65, 69-70Transformational leadership behaviour 1-7, 64, 65-

70Transformational leadership behaviour inventory

1,3Transformational leadership theory 1-3Tret personaliti 9-10, 13, 15-17, 19-20, 22-24

Ujian personaliti 23Ujian-t dua sample tak bersandar 15, 17, 19-20Uncertainty avoidance 100-101United States 1

Value line index 74Value orientation 100-102, 104Value variations 97, 100Variance analysis 97Variasi kecekapan 167Varimax rotation 14Virtual class application 27-40Volatility 85-86


Pertanika Journal of Social Science & HumanitiesAuthor Index for Volume 10 Nos. 1 and 2, 2002

Abd Rahim Bakar 53-61Ahmad Zubaidi Baharumshah 131-141Ain Nadzimah Abdullah 109-116Aminuddin Yusof 1-8Ang, Minni K. 43-51Annuar Mohd Nasir 73-84

Arfah Salleh 27-41

Badriyah Minai 27-41

Choo Wei Chong 85-95

Habibah Elias 53-61Hazman Shah Abdullah 97-107Huson Joher Alliahmed 73-84Jariah Mohd Jan 143

Liew Khim Sen 131-141Loo Sin Chun 85-95

Mahdhir Abdullah 73-84Minni K. Ang see Ang, Minni K.Miranda P. Yeoh see Yeoh, Miranda P.

Mohd Majid Konting 53-61Muhammad Fauzi Jumingan 153-164Muhammad Idrees Ahmad 85-95

Othman Jailani 9-25

Poo Bee Tin 117-129

Rahmah Ismail 117-129Razmi Chik 97-107Rojanah Kahar 165-178RosliTalif 109-116Rozumah Baharudin 165-178Rumayajuhari 165-178

Shamsher Mohamad 73-84Sidek Mohd Noah 9-25

Taufiq Hassan 73-84

Yeoh, Miranda P. 43-51

Zaidatol Akmaliah Lope Pihie 53-61, 63-71Zarinah Arshat 165-178


Acknowledgements

The Editorial Board acknowledges the assistance of the following reviewers in the preparationof Volume 10, Numbers 1 & 2 of this journal

Prof. Dr. Ahmad MahzanProf. Dr. Ahmad Zubaidi BaharumshahAssoc. Prof. Dr. Che Ani MadDr. Christina Tio Ee MingProf. Dr. Daeng NasirProf. Dr. Fatimah ArshadDr. Foo Say FooiProf. Dr. GuptaDr. Hafriza BurhanudeenDr. Imran Ho AbdullahProf. Dr. Jamaliah Mohd AliDr. Karen Yip ChengProf. Dr. Md. Zabid Abd RashidAssoc. Prof. Dr. Mohd. Majid Konting

Assoc. Prof. Dr. Muhammad Bukhari LubisAssoc. Prof. Dr. Muzafar Shah HabibullahAssoc. Prof. Dr. Obiyathulla Ismath BachaMdm Rosnani JusohDr. Saharuddin Abdul AzizDr. Salleh YahyaProf. Dr. Shamsher MohamedProf. Dr. Sharifah NorMr Soaib AsimiranDr. Tan Sooi BengDr. Zaleha Md. NorAssoc. Prof. Dr. Zakaria KasaAssoc. Prof. Dr. Zulkifli Hamid


Preparation of Manuscript

Typing and PaperThe manuscript should be typed double spaced on A4paper with 4cm margins on all sides, it should be limitedto 25 pages including tables, illustrations and references.

Title pageThe title of the paper, name of author and full addressof the institution where the work was carried outshould appear on this page. A short title not exceeding60 characters should be provided for the runningheadlines.

AbstractEnglish and Bahasa Melayu abstracts of 200 words eachshould follow the title page. Papers from outsideMalaysia may be submitted with an English abstractonly.

KeywordsAbout six to ten keywords are required and these shouldbe placed directly above the abstract.

TablesTables should be typed on separate pages and numberedusing Arabic numerals. Each table should be referred toin the text, has a brief title and include explanatorynotes, if necessary, below it. Vertical rules should not beused. Footnotes in tables should be designated bysymbols or superscript small italic letters. Tables shouldconform to page size.

EquationsThese must be clearly typed, triple-spaced. They shouldbe identified by numbers in square brackets placed flushwith the right margin.

Illustrations <jf PhotographsIllustrations including diagrams and graphs are to bereferred to in the text as * figures' and photographs as'plates' and numbered consecutively in Arabic numerals.All photographs (glossy black and white prints) shouldbe supplied with appropriate scales.

Illustrations should be of print quality; output fromdotmatrix printers is not acceptable, Illustrations shouldbe on separate sheets, about twice the size in print. Allletters, numbers and legends must be included on theillustration with the author's name, short title of thepaper, and figure number written on the verso. A list ofcaptions should be provided on a separate sheet.

Spelling & MeasurementsThe Oxford English Dictionary should be consulted forspelling. Metric units must be used for empiricalmeasurements.

Citations and ReferencesCitations to the literature in the text should to be indicatedby the author's name and year of publication inparentheses, e.g. (Barnett and Lewis 1982). If an authoris quoted in the text, the year of publication shouldimmediately follow in parentheses, e.g. Barnett and

Lewis (1982) state that " " Citation of a particularpage follows the date and is preceded by a comma, e.g.Humphrey 1990, p. 26-27.

For works with multiple authors, the first author'ssurname is used followed by et al. The full form ofcitation is used for two authors.

If two or more works by the same author are citedtogether, the author's name is followed by the variousyears of publication arranged in chronological order e.g.(Sulaiman 1979, 1980, 1981). In case of citing anauthor with several works published in the same year,the works should be distinguished by the addition of asmall letter e.g. Choa (1979a); Choa (1979b).

When several authors are cited, they are to be arrangedin chronological order and separated by semicolons, e.g.Zaharah 1960; Yong 1980; Lewis 1990.

Serials are to be abbreviated as in the World List ofScientific Periodicals. The abbreviation for PertanikaJournal of Social Science and Humanities is Pertanikaf. Soc. Sd. Hum.

The following reference style is to be observed:

BookShamsher Mohamed, Shamsuddin Ismail & AnnuarMohd. Nassir. 1989. Asas Belanjawan Modal. 197p.Serdang: Universiti Pertanian Malaysia Press.

Chapter in Edited BookZahid Emby. 1990. The Orang Asli Regrouping Scheme-converting swiddeners to commerical farmers. InMargins and Minorities - The Peripheral Areas and Peoplesof Malaysia, ed. V. T King & MJ.G. Parnwell, p. 94-109. Hull: Hull University Press.

Unpublished Materials (e.g. theses, reports, documents)Shahwahid, H.O. 1989. Price competitiveness anddemand behaviour of Malaysia Meranti lumber andhardwood plywood in the United States' import market.Ph.D. Dissertation, State University of New York.Syracuse.

Ministry of National Unity. 1973. A socio-economicsurvey on the new villages in Perak and Melaka. 67p.Malaysia.

SerialsNoran Fauziah Yaakub. 1990. A multivariate analysis ofattitude towards teaching. Pertanika 13(2): 267-273.

Conference ProceedingsAmir Awang. 1992. Counselling, human resorcesdevelopment and counseling services. In Proceeding ofAsia Pacific Conference on Human Resource Development, ed.Sulaiman M. Yassin, Yahya Mat Hassan Kamariah AbuBakar, Esah Munji anda Sabariah Mohd Rashid, p. 243-246. Serdang: Universiti Pertanian Malaysia.


Volume 10 Number 2 (September) 2002

Contents

Modelling the Volatility of Currency Exchange Rate Using GARCHModel - Choo Wei Chong, Loo Sin Chun and Muhammad [drees Ahmad

The Influence of Value Orientations on Service Quality Perceptions in 97a Mono-Cultural Context: An Empirical Study of Malay UniversityStudents - Hazman Shah Abdullah and Razmi Chik

The Sociolinguistics of Banking: Language Use in Enhancing 109Capacities and Opportunities - Ain Nadzimah Abdullah and RosliTalif

Faktor-faktor Mempengaruhi Agihan Pendapatan di Malaysia 1970- 1172000 - Rahmah Ismail dan Poo Bee Tin

Performances of Non-linear Smooth Transition Autoregfessive and 131Linear Auto regressive Models in Forecasting the Ringgit-Yen Rate- Liew Khim Sen and Ahmad Zubaidi Baharumshah

Interpretation of Gender in a Malaysian Novel: The Case of Salina 143- Jariah Mohd, fan

Penterjemahan Pragmatik dalam Konsep Masa Arab-Melayu: Satu 153Analisis Teori Relevan - Muhammad Fauzi bin Jummgan

Tingkah Laku Keibubapaan dan Penyesuaian Tingkah Laku Anak 165dalam Keluarga Berisiko di Luar Bandar - Zarinah Arshat, RozionahBaharudin, Rumaya Juhari dan Rojanah Kahar

ISSN 0126-7702

09

JSSH Vol. 10 (2) Sep. 2002

Documents

Transcript of JSSH Vol. 10 (2) Sep. 2002