Consequences of the performanceappraisal experience

Michelle BrownDepartment of Management, Faculty of Economics and Commerce,

University of Melbourne, Melbourne, Australia

Douglas HyattRotman School of Management, University of Toronto, Toronto, Canada, and

John BensonSchool of Management, University of South Australia, Adelaide, Australia

AbstractPurpose – The purpose of this paper is to examine the role of low quality performance appraisals(PA) on three human resource management outcomes ( job satisfaction, organisational commitmentand intention to quit).

Design/methodology/approach – Using data from 2,336 public sector employees clusters of PAexperiences (low, mixed and high) were identified. Regression analysis was then employed to examinethe relationship between low quality PA experiences and job satisfaction, organisational commitmentand intention to quit.

Findings – Employees with low quality PA experiences (relative to those with mixed and highquality PA experiences) were more likely to be dissatisfied with their job, be less committed to theorganisation and more likely to be contemplating leaving the organisation.

Research limitations/implications – The data were collected in a large public sector researchorganisation where the results of the appraisal were linked to pay increments. Further research isneeded to determine the applicability of the results to private sector employees.

Practical implications – The quality of the PA experience varies and a low quality experienceresults in lower job satisfaction and organisational commitment and higher quit intentions. Thechallenge for human resource (HR) practitioners is to decide whether the allocation of additionalresources to ensure that all employees have a uniformly high quality PA experience is a worthwhileinvestment.

Originality/value – Research has tended to focus on the relationship between a single feature of aPA process and HR outcomes. Organisations need to acknowledge the importance of the overall PAexperience when evaluating its consequences for HRM outcomes.

Keywords Performance appraisal, Performance management, Job satisfaction, Pay,Organizational behaviour, Employee turnover

Paper type Research paper

There is an increasing use being made of the performance appraisal process (Millwardet al., 2000; Nankervis and Compton, 2006; Wiese and Buckley, 1998) generally

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0048-3486.htm

This research was funded by grants received from the Faculty of Economics and Commerce,University of Melbourne. The authors would like to thank management and staff of public sectorresearch (PSR) and the officials of the Community and Public Sector Union for their support forthe project.

Performanceappraisal

experience

375

Received 7 August 2008Revised September 2008

Accepted 4 July 2009

Personnel ReviewVol. 39 No. 3, 2010

pp. 375-396q Emerald Group Publishing Limited

0048-3486DOI 10.1108/00483481011030557

motivated by an organisational desire to affect employee behaviours and attitudes and,ultimately, organisational performance (Aguinis, 2009; Gardner, 2008; Murphy andCleveland, 1991; Shields, 2007). This occurs as a consequence of the establishment ofgoals at the beginning of the evaluation cycle which provide employees with clearperformance targets, the monitoring of performance during the evaluation cycle (whichcan be used to assist poor performers) and the reinforcement provided for goodperformance through the provision of rewards, usually in the form of higher pay(Milkovich and Wigdor, 1991). This process is seen to encourage employeeperformance in subsequent performance cycles (Heneman and Werner, 2005; Mani,2002).

The capacity to achieve these positive outcomes will be a function of the quality ofthe performance appraisal (PA) experience. Taking a lead from the operationsmanagement field, quality is typically defined as establishing and operating processesthat promote organisational efficiency (for example see: American Society for Quality(available at: www.asq.org). The aim of a quality approach is to reduce variation inevery process in order to obtain greater consistency (Roberts and Sergesketter, 1993).PA is a complex process and there is scope for variation, particularly when thesupervisor is required to make subjective judgments of employee performance (ascompared with an objective performance appraisal where the measurement focuses onthe quantifiable aspects of job performance). Subjective judgements have the potentialto dilute the quality of the PA process as they may be influenced by bias or distortionas a result of emotion (for example, see Longenecker et al., 1987; Murphy andCleveland, 1995). Understanding the impact of PA quality is particularly important asTreadway et al. (2007) suggest that PAs are becoming increasingly subjective.

Assessments of quality typically require the involvement of stakeholders(Ghorpade and Chen, 1995). In the context of performance appraisals, a criticalstakeholder is the employee: the PA process is designed to stimulate employeeperformance (Aguinis, 2009). In this paper we use employees as our assessors of PAquality as it is the behaviours and attitudes of employees that are important to theoverall success of a PA process (Keeping and Levy, 2000). Employees are sensitive toquality variations in PA as its processes are a powerful determinant of employees’prospects (for example, promotion, termination of employment) within theorganisation. Thayer (1987) suggests PA quality variations will generate strongreactions among employees.

Organisational efficiency can be affected by the quality of the performanceappraisal process. In our paper, the focus is on the impact of low quality PA on threeindicators of organisational efficiency: job satisfaction, organisational commitmentand intent to quit. Past research has typically focused on a single outcome of PA (forexample, satisfaction with appraisal feedback: see Jawahar, 2006), though from apractical perspective organisations need to weigh up the combined effects of anappraisal process in order to determine if the advantages outweigh the costs (Tayloret al., 1995).

In this paper, the quality of an employee’s PA experience is determined bycombining aspects of the PA process into a measure of their total experience. Ourdefinition of quality, developed in a subsequent section, focuses on the formal appraisalprocedures, the enactment of these formal appraisal procedures and the interpersonaltreatment of employees during the appraisal process. This approach is a response to

PR39,3

376

the observation that many organisations equate PA with a single practice, for example,goal setting (Fletcher and Williams, 1996), which is inappropriate. Heathfield (2007)advises that organisations need to make a commitment to the whole PA process, andnot just its components, in order to achieve its objectives.

Our particular focus is on those employees who report low quality PA experiencesas this can help HR practitioners identify useful areas for improvement. Past researchhas tended to focus on the positive organisational effects that a single feature of a PAprocess can achieve, but what is less understood is what happens when multipleaspects of the PA process are working less than optimally. We might anticipate that anemployee who has a low quality PA experience does not react at all: the process wasintended to promote a range of positive outcomes and in the absence of a positive PAexperience the employee fails to react at all. Alternatively, the justice (Adam, 1965) andemotion research (for example, see Fugate et al., 2008) suggests that negativeassessments are more powerful drivers of employee attitudes and behaviours than arepositive assessments. Therefore, we might expect that when an employee reports a lowquality PA experience across multiple features of the PA process that they will indeedreact in a negative way. Our study provides an opportunity to test these twointerpretations of employee reactions to low quality appraisals.

Understanding the implications of low quality PA experiences is important:a review of the practitioner and academic journals suggests that low qualityPA is a continuing challenge for organisations (Gardner, 2008; Heathfield,2007; Moats-Kennedy, 1999; Merritt, 2007) so the consequences of low quality PAexperiences are potentially of interest to many organisations. For researchers, focusingon low quality PA experiences potentially challenges our existing approaches to theevaluation of a PA process. It is implicitly assumed that employees enter into a PAprocess prepared to fully engage with its requirements (Milkovich and Wigdor, 1991).However, if an employee’s past experiences with a PA process have been of a lowquality, it may adversely affect the willingness of the employee to fully engage in asubsequent evaluation cycle. Researchers may therefore need to consider building in aPA experience feedback loop when developing models to test the effectiveness of anappraisal process.

The next section outlines why the quality of an employee’s PA experience can vary.This is followed by a discussion of the implications of low quality PA experiences forour three human resource management (HRM) outcomes. Our hypotheses are thentested using data from 2,336 non-managerial employees of a public sector organisationin Australia. The paper concludes by discussing the implications for HR policy,practice and research.

Assessing the performance appraisal experienceThere is a body of empirical research that suggests that performance appraisals doresult in increases in employee performance and productivity (Rodgers and Hunter,1991; Schay, 1988; Taylor and Pierce, 1999). These improvements are seen to derivefrom the greater employee identification with and commitment to, the objectives of theorganisation. Work efforts are directed to activities that will be of benefit to theorganisation. Poor performing employees are identified during the evaluation cycleand given feedback on how to improve. They might also undertake somedevelopmental activities (for example, training) in order to rectify performance

Performanceappraisal

experience

377

deficiencies. Employees are motivated to work at a higher level by the offer of financialincentives that are contingent on their performance, and these financial incentives areimportant in encouraging the retention of high performing employees (Kessler andPurcell, 1992; Milkovich and Wigdor, 1991). When these positive outcomes are notgenerated, it is often concluded that the fault lies with the low quality of the PA process(Blau, 1999; Roberts, 1998).

The attitudes and approach of supervisors to the PA process has been identified asa source of quality variations in PA. One source of quality variation derives from thegeneral attitudes that supervisors have towards the process. Many writers haveidentified the reluctance of supervisors to undertake appraisals. Heathfield (2007, p. 6)notes that “when surveyed about most disliked tasks, managers say they hateconducting appraisals, second only to firing employees”. Wanguri (1995) believes thatmany practitioners see appraisal as a necessary evil in corporate life. As aconsequence, Pettijohn et al. (2001, p. 337) suggest that supervisors conduct appraisalsin an “arbitrary and perfunctory manner” as they believe that “conductingperformance appraisals requires considerable amounts of time and effort, generatesfew rewards and adds considerably to the manager’s level of conflict and stress”.Latham et al. (1993) believe that managers often avoid the appraisal process becausethey either fear the consequences or they feel the potential returns from their appraisalefforts are not worthy of the required investments.

A second source of quality variation derives from the way supervisors exercise theirresponsibilities under an appraisal process. Researchers have demonstrated thatsupervisors make mistakes (for example, halo errors, recency effects, central tendency:see Milkovich and Newman (2002) for a full list of rating errors), are uncomfortableabout providing negative feedback (for example see Chen et al., 2007), often put offdoing appraisals (Bratton and Gold, 1999) or apply their own set of internal valuesabout performance when assessing their employees (Murphy and Cleveland, 1995). Inorder to encourage supervisors to conduct high quality performance appraisals, asupervisor is likely to find the assessment of their performance is partly a function ofthe way they manage the evaluations of those who report to them (Curtis et al., 2005).This discussion demonstrates that supervisors are a major source of quality variationsin PA, but quality is ultimately the assessment made by the recipients of the process,namely employees.

Like supervisors, employees have been found to regard the PA process with “fearand loathing” (Thomas and Bretz, 1994). Taking a lead from the justice literature (seeFortin, 2008 for a review), there are two schools of thought (instrumental andrelational) that identify the criteria that employees apply when assessing PA quality.According to the instrumental school of thought, employees value PA process controlsas it promotes predictability (Thibaut and Walker, 1975). Process controls can beestablished through operation of formal procedures (such as the right to presentevidence to a supervisor) and enhance quality by ensuring that the process applies toall employees equally (Hendrix et al., 1998). The relational school of thought postulatesthat individuals care about PA quality because it signals their status and worth withinthe organisation. The relational school of thought emphasises the needs for belongingand self-esteem and acknowledges the informal actions of someone in a position ofauthority over the employee (Hendrix et al., 1998). Further, an individual’s standing isthought to be conveyed by their interpersonal treatment during social interactions

PR39,3

378

(Lind and Tyler, 1988). Low quality interpersonal treatment conveys the view thatmanagement regards the individual as being of low status and disrespect forindividual rights communicates that their interests are unlikely to be protected (Fortin,2008).

Combining these two schools of thought suggests that PA quality is a function ofboth PA procedures and treatment. Employees evaluate the formal organisationalprocedures (and the way they are enacted) as well as how they are treated by PAdecision-makers when assessing the quality of their PA experience. Having identifiedthe meaning of quality in a PA context, we now turn to the identification of specificindicators that can be used to generate a measure of PA quality. We focus on fourindicators: clarity, communication, trust and PA fairness. The first indicator refers tothe extent to which employees are clear about the role and purpose of the PA. This willinvolve clarity about the PA process and the role it will play in determining anemployee’s fate within the organisation. A second indicator is the extent ofcommunication and information flows between the employee and their supervisor.Opportunities to acquire, supply and evaluate information, in conjunction with clarityabout PA processes, provide employees with the opportunity to exercise a level ofprocess control. Further, giving an employee the opportunity to express his or herviewpoint is valued in itself and validates his or her belongingness in the organisation.A third indicator of PA quality is the level of trust the employee has for theirsupervisor. Employees who believe their supervisor is competent and has a goodknowledge of their employees’ job duties will be more likely to trust their supervisorand rate their PA experience positively (Greenberg, 1986). Moreover, when employeestrust their supervisor, they hold positive expectations about their supervisor’s motives,believing that their manager will act in their best interest. A fourth indicator refers tofair treatment (Bies and Moag, 1986). Employees want to be treated fairly throughoutthe PA process as this is seen to have an impact on the quality of the outcomes of theprocess (Fortin, 2008).

Performance appraisal experience and HRM outcomesThe importance of the quality of the PA experience derives from its impact on HRMoutcomes. The following reviews the literature on three HRM outcomes: jobsatisfaction, organisational commitment and intention to quit. Each of these HRMoutcomes and their relationship with quality of an employee’s PA experience is nowexamined.

Job satisfactionJob satisfaction is the overall degree to which an employee likes his or her job (Priceand Mueller, 1981). Job satisfaction is typically seen as a short-term measure as it canbe affected by transitory events. Job satisfaction has been used as a measure ofemployee wellbeing (Green., 2004) and has also been associated with employeeproductivity and safety (Harter et al., 2002). High quality PA is intended to increase jobsatisfaction (Fletcher and Williams, 1996) for which there is some empirical support(see Masterson et al., 2000).

Job satisfaction is the result of an employee’s perception of how well their jobprovides those things that are viewed as important (Locke, 1976). A high quality PAexperience is likely to enhance employees’ feelings of self-worth, achievement,

Performanceappraisal

experience

379

attitudes about their job and their feelings of a positive standing in the organisation(Lind and Tyler, 1988). It is also likely to generate confidence in the quality of theoutcomes from the PA process (Hendrix et al., 1998). High quality PA experiences aretherefore likely to generate higher levels of job satisfaction (Fried and Ferris, 1987).

A low quality PA experience may result in a lower level of job satisfaction. Theemployee is likely to feel that their contributions are not valued as they have limitedcapacity to exercise any process control and experience some confusion about theperformance expectations of the organisation. The process of trial and error inattempting to live up to vague performance expectations will reduce the employees’feelings of achievement and self-worth and ultimately job satisfaction (Behrman et al.,1982). The results of Campbell et al. (1998) suggest that a low quality PA experiencemay lead to lower levels of job satisfaction. Therefore we hypothesise that:

H1. A low quality PA experience will be associated with lower levels of jobsatisfaction relative to those with high or mixed quality PA experiences.

Organisational commitmentOrganisational commitment reflects attitudes towards the entire organisation and istypically seen as broader than job satisfaction and is more consistent than jobsatisfaction over time (Mowday et al., 1982, p. 28). A PA process is intended to promoteorganisational commitment and ultimately improve organisational effectiveness(Fletcher and Williams, 1996). The psychological contracts and justice literaturesuggest an employee’s experience of the PA process may affect their willingness to becommitted to their organisation.

The psychological contracts literature, especially relational contracts, suggests thatwhen an employer provides a high quality PA experience it will increase theemployees’ perceived obligations to the employer which in turn affect their attitudesand potentially their behaviours (Hendrix et al., 1998). Organisational commitment isdependent on maintaining a relationship of consistency and good faith, which can beaffected by the quality PA experience (Robinson et al., 1994). A high quality PAexperience should lead employees to feel respected by and be proud of the organisation.In turn they are more likely to identify with and internalise the values of theorganisation.

According to justice researchers, a high quality PA experience will cause employeesto have faith in the system, which can result in higher organisational commitment(Sweeney and McFarlin, 1993). A performance appraisal process is intended tocommunicate organisational goals and values and give people a “cause they can rallyaround” (Kuvaas, 2007, p. 381). A high quality PA experience increases the legitimacyof the organisation in the employees’ eyes and thereby their willingness to comply withthe goals of the organisation. High quality PA experiences also enhance the evaluationemployees make of supervisors and their organisation. The implementation of a highquality PA process affirms the employees’ self-worth and conveys that theorganisation and supervisors hold them in high regard and are committed toprotecting their rights in the context of organisational decisions (Taylor et al., 1995).Moreover, Pearce and Porter (1986) suggest that employees who believe theircontributions are highly regarded are likely to be more committed to the organisation.

In the absence of a high quality PA experience, we anticipate that employees willnot feel any sense of reciprocal obligation. If the organisation is not able to provide a

PR39,3

380

high quality PA experience, employees are less likely to know of, internalise, and becommitted to the goals and values of their employing organisation. Further, employeesare less likely to have faith in the PA processes or regard it as a legitimate tool of HRmanagement, both of which are likely to contribute to lower levels of organisationalcommitment. We therefore hypothesise that:

H2. A low quality PA experience will be associated with lower levels oforganisational commitment relative to those with high or mixed quality PAexperiences.

Intention to quitAn objective of PA is typically to retain the highest performing employees, though asGriffeth et al. (2000) observe the focus of much of the turnover research has been on therole of merit rewards as a retention tool. It is rare to see studies that include the qualityof the PA experience as a predictor of turnover or quit intentions, but as Griffeth et al.(2000, p. 480) argue, PA quality has “as much – if not more – to do with encouragingemployees to stay as fair pay amounts”. We assess the relationship between PAquality and quit intentions as the retention of staff is usually seen as a betterinvestment than incurring the costs of recruiting replacements (Mitchell et al., 2001).The withdrawal of employees from the organisation can have a detrimental effectthrough the loss of institutional memory (Moynihan and Pandey, 2007) as well asnegative impacts on the organisations’ reputation and productivity (Hom and Griffeth,1995).

An employee’s decision to leave an organisation is typically a progressive process.While it may be triggered by some type of dissatisfaction (Hom and Griffeth, 1995) or“shock” (Morrell et al., 2008) the decision involves a series of related cognitive andbehavioural considerations before the person actually terminates their employmentwith the organisation (Campbell and Campbell, 2000). The process begins by havingthoughts (withdrawal cognitions) about leaving the current job and then disgruntledemployees progressively enact more extreme manifestations of job withdrawal overtime (Rosse, 1988).

Quit intentions are a form of withdrawal from the organisation and may beinfluenced by the quality of the PA experience. Turnover research has suggested thatemployees tend to withdraw from situations that are counter to their best interests(Fugate et al., 2008). A low quality PA experience could be counter to the best interestsof an employee. Uncertainties about the requirements of the PA process and poorinformation flows means that the employee is likely to feel confused about how to meettheir performance obligations. They are also likely to feel undervalued when theybelieve that their contributions to the organisation are not being effectively assessed orrecognised. This in turn will contribute to negative attitudes about their work andorganisation. Because of the significance of PA processes to an employee’s prospects inthe organisation, we expect that low quality PA experiences will weaken thewillingness of the employee to stay within the organisation so offer the followinghypothesis:

H3. A low quality PA experience will be associated with higher intentions to quitrelative to those with high or mixed quality PA experiences.

Performanceappraisal

experience

381

Methodology and dataParticipants and contextThe study is based on the results of a mail survey of 6,957 employees of a largeAustralian public sector organisation (“PSR”). The confidential survey was sent to allemployees via PSR’s internal mail system. The surveys were returned directly to theresearchers. Completed questionnaires were returned by 3,335 employees, representingan overall response rate of 47.9 per cent. It is important to delineate the reactions ofemployees from the manager/supervisors as research has demonstrated that attitudesto performance management systems are affected by the role an individual playswithin it. Dyer et al. (1976) demonstrate that managers are more supportive ofperformance-based systems. This result reflects the greater control that managersoften have over their work and hence their ability to affect their performance outcomes.The following analysis, therefore, is based on the responses of the non-managerialemployees of PSR as it is this group of employees who work under the PA process andwhose behaviour the system is intended to affect. After excluding managerialemployees (a total of 579 employees) and missing data, the final data set consists of2,336 non-managerial employees.

Employees of PSR (both union and non-union) are covered by an industrialagreement that provides for a nine-level salary system. Each pay grade has a definedpay range (on average plus or minus 12 per cent from the midpoint) and a number ofincrements of predetermined size. Usually employees are appointed at the bottom oftheir pay grade and progression through the increments is based on a positivesupervisory assessment of performance. Each year the supervisor and the employeejointly develop performance objectives, which are monitored during the year. At theconclusion of the 12-month evaluation cycle (March 31, irrespective of the start date),the supervisor makes an assessment of the employee based on the agreed objectives,using a five-point scale. Ratings are used to determine increments (employees mustreceive a three or better in order to move to the next increment), promotions and toprovide career advice to employees.

MeasuresDependent variables. There are three dependent variables in the following analysis: jobsatisfaction, organisational commitment and intention to quit. Job satisfaction wasmeasured by six items from Price and Mueller (1981). A representative item in thisscale is “Most days I am enthusiastic about my job”. Organisational commitment wasmeasured using the nine items from Porter et al. (1974). A representative item in thisscale is “I am willing to put in a great deal of effort beyond that normally required inorder to help PSR be successful”. There is a debate in the literature about the best wayto measure turnover. Griffeth et al. (2000, p. 483) report based on a meta-analysis, thatquit intentions are good precursors of turnover so we use a measure of quit intentionsin our study. Intention to quit was measured by two items from the scale by Porter et al.(1974), with a representative item being “It would take very little for me to changeemployers”. A summary of the definitions and descriptive statistics for all variablesused in the regression analysis can be found in Table I. The correlation matrixpresented in Table II shows no evidence of strong correlations between theindependent variables.

PR39,3

382

Low

qualityPA

experiences

Highqu

alityPA

experiences

Label

Definition

Mean

SDMean

SD

Dependentvariables

Jobsatisfaction

Jobsatisfaction

asmeasuredby

sixitem

sfrom

Price

andMueller(1981),alpha

¼0.89.F

ive-pointscale

with5¼

high

jobsatisfaction

3.24

0.89

3.66

0.73

Organisationalcommitment

Organisationalcommitmentas

measuredby

nine

item

sfrom

Porteret

al.(1974),alph

a¼

0.86.F

ive-

pointscalewith5¼

high

lycommitted

3.17

0.70

3.48

0.54

Intentionto

quit

Intentionto

quitas

measuredby

twoitem

sfrom

Porteret

al.(1974),alph

a¼

0.70.F

ive-pointscale

with5¼

high

willingn

essto

quit

3.02

0.90

2.60

0.78

Independentvariable

Low

qualityPA

experience

Dichotomousvariablewhere

1¼

low

overall

experience,0

otherw

ise

nana

naNa

Controlvariables:demographic

Age

Age

inyears

41.51

9.30

39.61

9.53

Edu

cation

PhD

¼1,0otherw

ise

0.19

0.39

0.19

0.39

Fem

ale

Fem

ale¼

1;male¼

00.35

0.48

0.43

0.50

Income

Midpointapproxim

ated

from

each

levelofthesalary

classification

system

,multipliedby

100

47,990

10,770

47,810

10,510

Negativeaffect

Negativeaffectivity(the

extent

towhich

anindividu

alexperiencesaversive

emotionalstates

over

timeandacross

situations)as

measuredby

threeitem

sfrom

Watsonetal.(1987),alph

a¼

0.86).

Five-pointscalewith5¼

high

negative

affectivity

3.11

0.86

2.67

0.85

Primaryincomeearner

Primaryhouseholdincomeearner

¼1,0otherw

ise

0.72

0.45

0.67

0.47

Union

Union

mem

ber¼

1,0otherw

ise

0.57

0.50

0.49

0.50

(continued)

Table I.Description of regression

variables

Performanceappraisal

experience

383

Low

qualityPA

experiences

Highqu

alityPA

experiences

Label

Definition

Mean

SDMean

SD

Controlvariables:situational

Autonom

yAutonom

yat

workas

measuredby

four

item

sfrom

Tetrick

andLaRocco

(1987),alpha

¼0.71.F

ive-

pointscalewhere

5¼

high

autonomy

3.43

0.73

3.81

0.58

Jobsecurity

Perceptionof

jobsecurity

measuredby

threeitem

sfrom

Oldham

etal.(1986),alph

a¼

0.85.F

ive-point

scalewith5¼

high

jobsecurity

2.50

0.98

2.93

0.92

Resourceadequacy

Three

item

sthat

measure

thelevelof

resource

adequacy

from

Iverson(1992).Five-pointscalewhere

5¼

high

resource

adequacy,1

¼low

resource

adequacy.A

lpha

¼0.80

2.89

0.90

3.46

0.81

Roleconfl

ict

Roleconfl

ictas

measuredby

threeitem

sfrom

Kahn

etal.(1964),alph

a¼

0.74.F

ive-item

scalewhere

5¼

high

roleconfl

ict

3.19

0.74

2.64

0.72

Taskindepend

ence

Levelof

task

independ

ence

injobas

measuredby

amodified

five-item

scalefrom

Kiggu

ndi(1983),

alph

a¼

0.84.F

ive-pointscalewith5¼

high

levelof

task

interdependency

3.60

0.64

3.89

0.52

Table I.

PR39,3

384

12

34

56

78

910

1112

1314

1516

1.Jobsatisfaction

1.00

2.Organisational

commitment

0.56

1.00

3.Intentionto

quit

20.48

20.54

1.00

4.Low

qualityPA

experience

20.20

20.18

0.18

1.00

5.Age

0.17

0.12

20.18

0.06

1.00

6.Edu

cation

0.05

20.05

0.03

0.02

20.12

1.00

7.Fem

ale

20.01

0.01

20.02

20.06

20.13

20.16

1.00

8.Income

0.11

0.01

20.01

0.02

0.32

0.44

20.35

1.00

9.Negativeaffect

20.26

20.17

0.19

0.23

20.02

0.06

20.04

0.07

1.00

10.Primaryincomeearner

0.05

0.04

20.05

0.06

0.16

0.12

20.38

0.27

0.03

1.00

11.Union

0.00

20.04

20.03

0.04

0.23

20.06

20.05

0.15

0.07

0.07

1.00

12.Autonom

y0.36

0.21

20.20

20.24

0.06

0.14

20.05

0.23

20.21

0.03

0.02

1.00

13.Roleconfl

ict

20.16

20.14

0.21

0.31

20.01

20.01

20.09

0.04

0.27

0.06

0.02

20.28

1.00

14.Resourceadequacy

0.16

0.20

20.17

20.26

0.02

20.09

0.15

20.10

20.29

20.10

20.09

0.18

20.28

1.00

15.Jobsecurity

0.17

0.18

20.12

20.15

20.01

20.07

0.04

20.01

20.14

20.07

20.10

0.15

20.13

0.21

1.00

16.Taskindepend

ence

0.49

0.28

20.25

20.20

0.08

0.12

20.09

0.25

20.16

0.08

20.03

0.42

20.13

0.16

0.16

1.00

Table II.Correlation matrix

Performanceappraisal

experience

385

Independent variable. Our measure of the quality of the PA experience combines itemsto assess the overall quality of an employees’ PA experience. The earlier discussion ofemployees’ expectations of a quality PA process suggests that four features should beincluded in our measure: clarity of performance expectations, level of communicationsbetween the employee and their supervisor, trust in the supervisor and the fairness ofthe PA process.

Clarity about the PA process was measured by three items from Tang andSarsfield-Baldwin (1996). A representative item in this scale is: “When you took upyour current position how clear was it that your performance would be periodicallyevaluated”. The extent of communications between the employee and the supervisorwas measured by six items from Tang and Sarsfield-Baldwin (1996). A representativeitem in this scale is “How often is the progress toward your goals set in previousmeetings reviewed by your supervisor with you”. The level of trust in the supervisorwas measured by four items from Tang and Sarsfield-Baldwin (1996). A representativeitem in this scale is “To what extent do you have confidence and trust in yourimmediate supervisor regarding his/her general fairness?”. The fairness of the PAprocess was measured by six items from Tang and Sarsfield-Baldwin (1996). Arepresentative item in this scale is “My last performance appraisal was fair”. Usingcluster analysis, these four scales were combined in order to assess the overall qualityof an employee’s PA experience. The resulting variable is then used as an independentvariable in the regression analysis. A summary of the definitions and descriptivestatistics for all variables used to determine the quality of the PA experience can befound in Table III.

Control variables. There were 11 variables included in the regression analysis tocontrol for individual employee differences and differences in the work context. Sevendemographic characteristics of the employee were included:

(1) age;

(2) education;

(3) gender;

(4) income;

(5) negative affect;

(6) primary income earner; and

(7) union membership.

There were five control variables to account for the situation in which the employeeundertakes their work. In the present study the following factors were designated ascontextual: job autonomy, role conflict, resource adequacy, task independence and jobsecurity perceptions. We include these control variables in recognition of the rolesthese factors have played in furthering our understanding of job satisfaction,organisational commitment and intent to quit (for example see Mobley, 1977; Riketta,2002; VanVoorhis and Levinson, 2006)

Method of analysisThis study uses two forms of analysis. First, cluster analysis is utilised to identify thequality of an employee’s PA experience, using the four features of a PA processoutlined above. Cluster analysis classifies respondents so that each respondent is very

PR39,3

386

Means

Correlations

Labela

Description

Total

High

Mixed

Low

l2test

12

34

1.Clarity

aboutPA

Clarity

abouttheperformance

appraisal

asmeasuredby

threeitem

sfrom

Tang

andSarsfield-Baldw

in(1996),

alph

a¼

0.86.F

ive-pointscalewith

5¼

high

clarity

2.78

4.49

2.73

2.18

285.27

*1.0

2.Com

mun

ications

Extentof

two-way

commun

ication

betw

eentheem

ployee

andtheir

supervisor

inPA

asmeasuredby

six

item

sfrom

TangandSarsfield-Baldw

in(1996).A

lpha

¼0.83.F

ive-pointscale

with5¼

high

two-way

commun

ication

2.51

3.38

2.65

1.91

29.95*

0.35

1.0

3.Trust

insupervisor

Trustinsupervisor

inPAas

measuredby

four

item

sfrom

TangandSarsfield-

Baldw

in(1996).A

lpha

¼0.90.F

ive-point

scalewith5¼

high

levelsof

trust

3.64

4.61

3.98

2.61

246.54

*0.31

0.55

1.0

4.Fairness

Employee

perceptionsof

thefairness

ofperformance

appraisalas

measuredby

sixitem

sfrom

TangandSarsfield-

Baldw

in(1996),alpha

¼0.95.F

ive-point

scalewith5¼

very

fair

3.55

4.27

3.92

2.58

213.51

*0.29

0.41

0.59

1.0

n¼

2,336

n¼

991

n¼

706

n¼

639

Notes:

*p,

0.01;aThe

alph

aon

thesefour

variablesis0.72.Clusteranalysiswas

used

toidentify

experiencesofperformance.T

henu

mberofclustersin

thedata

was

identified

usinghierarchicalclusteranalysis(com

pletelin

kage).Based

ontheoutcom

eof

thehierarchicalanalysis,non-hierarchicalcluster

analysis(k-m

eans)w

asused

toidentify

thefinalclustersolution

(see

Sharma,(1996)).Clusteranalysisgeneratesavariablewiththethreecategories

ofPA

experience.T

hisvariablewas

conv

ertedinto

threedichotom

ousvariables(high,lowandmixed

PAexperiences)andlowqu

alityPAexperiences(relative

totheothertypesof

PA

experience)wereinclud

edin

subsequent

regression

analysis

Table III.Quality of the

performance appraisalexperience

Performanceappraisal

experience

387

similar to others in the cluster on the basis of their experiences of the PA process. Inother words, the objective of cluster analysis is to identify clusters that exhibit highwithin cluster homogeneity and high between cluster heterogeneity (Hair et al., 1999).There are two broad types of cluster analysis: hierarchical and non-hierarchicaltechniques. Hierarchical clustering techniques do not require a priori knowledge of thenumber of clusters and these are ideal for determining the number of clusters in a dataset. Non-hierarchical clustering techniques require knowledge about the number ofclusters but typically yield better cluster solutions than hierarchical methods.Therefore as recommended by Sharma (1996) hierarchical analysis (using the completelinkage method) was first used to determine the appropriate number of clusters. Anon-hierarchical form of cluster analysis (k-means) was then used to identify thecluster solution based on the results from the hierarchical analysis.

The second form of analysis used is multiple regression to evaluate the impact of alow quality of the PA experience on our three HRM outcome variables.

ResultsThe results of the cluster analysis presented in Figure 1 demonstrate that there arethree internally consistent clusters (as shown by the l 2 test presented in Table III).Cluster 1 consisted of 639 employees who reported the lowest levels of trust in theirsupervisor (mean ¼ 2.61), poor communications (mean ¼ 1.91), expressed a lack ofclarity about their performance expectations (mean ¼ 2.18), and perceived the PAprocess as less fair (mean ¼ 2.58) than the other two clusters. This cluster was labelled“low quality PA experiences”. Cluster 3 consisted of 991 employees and was labelled“high quality PA experiences”. This group reported the highest levels of trust in theirsupervisor (mean ¼ 4.61), felt they had good communications (mean ¼ 3.381),expressed clarity about their performance expectations (mean ¼ 4.49) and perceivedthe PA process to be fair (mean ¼ 4.27). Cluster 2 consisted of 706 employees whoreported uniformly higher levels of trust, communications, clarity and fairness than

Figure 1.Profiles of performanceappraisal experience

PR39,3

388

those in cluster 1 but lower relative to those in cluster 3. So this cluster was labelled“mixed quality PA experiences”. High and mixed quality PA experience was theomitted category in the subsequent regression analyses.

The descriptive statistics in Table I show that those classified as having an overalllow quality PA experience are a little older, male and more likely to be a union memberthan those classified as having high quality PA experiences. Employees reporting highquality PA experiences work in jobs that have a higher level of autonomy, less roleconflict, more supportive co-workers and higher perceived job security than employeesreporting low quality PA experiences.

It was hypothesised (number 1) that job satisfaction would be lower amongemployees with low quality PA experiences. The results in Table IV (column 1)demonstrate that low quality experiences (relative to employees with high or mixedquality PA experiences) are associated with lower levels of job satisfaction(b ¼ 20.1050, p , 0.01). H1 is therefore supported.

The second hypothesis suggested a positive relationship between the quality of thePA experience and organisational commitment. Table IV (column 2) reports a significantnegative relationship between low quality PA experiences and organisationalcommitment (b ¼ 20.0979, p , 0.01). These results suggest that organisationalcommitment is adversely affected by low quality PA experiences (relative toemployees with high or mixed quality PA experiences), providing support for H2.

Job satisfactionOrganisationalcommitment Intention to quit

Variable nStandarderror n

Standarderror n

Standarderror

Independent variableLow PA experience 20.1050 * * * 0.0358 20.0979 * * * 0.0300 0.1999 * * * 0.0418

Control variables: demographicAge 0.0137 * * * 0.0017 0.0073 * * * 0.0014 20.0167 * * * 0.0020Education 0.1037 * * 0.0440 20.0383 0.0371 20.0499 0.0515Female 0.0696 * * 0.0336 0.0149 0.0283 20.0387 0.0392Income 20.0057 * * 0.0018 20.0043 * * * 0.0015 0.0086 * * * 0.0021Primary income earner 0.0587 * 0.0341 0.0719 * * 0.0287 20.1027 * 0.0399Negative affect 20.1326 * * * 0.0176 20.0345 * * 0.0149 0.0702 * * * 0.0207Union member 0.071 0.0303 20.0455 * 0.0255 20.0440 0.0354

Control variables: contextualAutonomy 0.1822 * * * 0.0256 0.0626 * * * 0.0215 20.0737 * * 0.0299Role conflict 0.0081 0.0208 20.0126 0.0176 0.1069 * * * 0.0244Resource adequacy 0.0097 0.0187 0.0613 * * * 0.0158 20.0456 * * 0.0219Job security 0.0564 * * * 0.0156 0.0643 * * * 0.0131 20.0424 * * 0.0182Task independence 0.5401 * * * 0.0289 0.2181 * * * 0.0244 20.2483 * * * 0.0338Adj R 2 0.3190 0.1438 0.1455Mean vif 1.26 1.26 1.26

Notes: *p , 0.10 two-tailed test; * *p , 0.05 two-tailed test; * * *p , 0.01 two-tailed test.Unstandardised coefficients. Many models of turnover include job satisfaction as a determinant (forexample, see Hom and Griffeth (1995). When we include job satisfaction as a control variable, the resultsfor our independent variable remain the same. Copies of these results are available from the authors

Table IV.Regression results testingthe relationship between

the quality of the PAexperience and HR

outcomes

Performanceappraisal

experience

389

The third hypothesis (H3) suggested a negative relationship between the quality of thePA experience and intention to quit. From the results contained in Table IV (column 3),we see that low quality PA experiences are significantly and positively related tointentions to quit (b ¼ 0.1999, p , 0.01). The present study therefore suggests thatemployees who report a low quality PA experience are more likely to look foralternative employment[1].

Discussion and conclusionsPerformance appraisals are a foundation element of human resource management: theresults of appraisals are used as the basis for many HR decisions. While organisationsinvest considerable time and resources into PA (for example through training ofsupervisors) our cluster analysis has demonstrated that the quality of the employee PAexperience does vary: some employees had high quality experiences of the PA processwhile at the other end of the spectrum some employees report low quality PAexperiences. Further, our study demonstrates that organisations do pay a price forallowing low quality PA experiences: when employees have low quality PAexperiences the organisation will likely incur a penalty in terms of lower jobsatisfaction and organisational commitment and higher intentions to quit.

Policy and practice implicationsFrom a practice perspective the results present two challenges for HR decision-makers.The first challenge is that improving the quality of an employee’s PA experience willcome at a cost and HR decision-makers need to decide if the costs needed to bring aboutan improvement in perceived PA quality are justified given the circumstances of theorganisation and the labour market in which it operates. An organisation, for example,that relies on highly skilled workers for its competitive success may be more willing toinvest in PA quality than an organisation that relies on low skilled labour drawn froman abundant labour market.

The second challenge is to keep the focus of HR on the whole PA process and not getoverly distracted by individual features of the process. HR decision-makers are oftenreminded (for example, see Schay, 1993) to keep the PA process under constant review,but this may encourage constant tinkering with individual features rather thankeeping the focus on the whole PA process.

Research implications and limitationsPA is typically described as a system that is intended to promote a range of positiveHR outcomes and researchers have demonstrated that individual features of a PAprocess are associated with the intended positive outcomes (Bratton and Gold, 1999;Fletcher and Williams, 1996). What has remained unclear is what happens when thePA process is in place but does not send the intended signals to employees. Our studyshows that when a PA process is in place but provides employees with a low qualityexperience, not only does the process fail to generate any positive HR outcomes, itactually generates some negative outcomes in the form of lower job satisfaction andorganisational commitment as well as higher intent to quit. This is consistent withfindings in the justice and emotions literature that negative assessments are powerfuldrivers of employee attitudes and behaviours (for example, see Fugate et al., 2008).

PR39,3

390

The results of our study raise some issues for future research. First, our study wasconducted in the public sector and the PA process was used to determine payincrements. Employees who achieved a three or better on the five point rating scalewere entitled to an increment and 97.8 per cent of the employees in our sample achievedthis rating. So despite the fact that the outcome of the PA was favourable for theoverwhelming majority of employees (a performance rating sufficient to merit a payincrement) employees’ perceptions of the PA experience influence job satisfaction,organisational and intention to quit, independent of the favourable outcome. It isevident from this study that the PA experience itself, and not just the outcome,influences employee attitudes. It is important, however, to test the generalisability ofour findings in a private sector organisation where there is a greater dispersion inperformance ratings.

A second issue of importance for future research is the effect of “one off” versus anongoing low quality PA experience. Future researchers should examine whetheremployees dismiss a low quality PA experience as a “one off” or whether it colourstheir expectations of the next performance evaluation cycle. Ongoing low quality PAexperiences may generate disillusionment with the PA process and a mistrust ofmanagement’s motives for the PA process among employees (Feldman, 2000). Futuremodels of the PA process should, therefore, take account of past PA experiences whenassessing the effects of a PA process and this relationship may well be moderated by arange of factors including the purpose of the appraisal process. A low qualityexperience in an evaluative PA process is likely to generate stronger reactions amongemployees than in a more developmentally focussed PA process (Youngcourt et al.,2007).

Performance appraisal has attracted a great deal of research attention over anextended period of time but the research has typically been focused on assessing theimpact of individual features of the PA process on a range of HR outcomes. Our paperhas shown that it is now time to move beyond this narrow approach and consider thecombined impact of the PA process and to investigate the unintended as well as theintended consequences of this tool of human resource management.

Note

1. We also ran the three models using only high quality PA experiences as the omittedcategory. The results for poor quality PA experiences in all our models remained significant:job satisfaction model (b ¼ 20.1331, p , 0.01); organisational commitment model(b ¼ 20.1428, p , 0.01) and our quit model (b ¼ 0.2071, p , 0.01). Copies of these resultsare available from the authors.

References

Adams, J.S. (1965), “Inequity in social exchange”, in Berkowitz, L. (Ed.), Advances inExperimental Social Psychology, Vol. 2, Academic Press, New York, NY, pp. 267-99.

Aguinis, H. (2009), Performance Management, 2nd ed., Pearson Prentice-Hall, Upper SaddleRiver, NJ.

Behrman, D.N., Bigoness, W.J. and Perreault, W.D. (1982), “Sources of job-related ambiguity andtheir consequences upon salespersons’ job satisfaction and performance”, ManagementScience, Vol. 27, pp. 1246-60.

Performanceappraisal

experience

391

Bies, R.J. and Moag, L. (1986), “Interactional justice: communication criteria of fairness”,in Lewicki, R., Sheppard, B. and Bazerman, M. (Eds), Research on Negotiation inOrganisations, Vol. 1, JAI Press, Greenwich, CT, pp. 43-55.

Blau, G. (1999), “Testing the longitudinal impact of work variables and performance appraisalsatisfaction on subsequent overall job satisfaction”, Human Relations, Vol. 52 No. 8,pp. 1099-113.

Bratton, J. and Gold, J. (1999), Human Resource Management, 2nd ed., Macmillan Business,London.

Campbell, D.J. and Campbell, K.M. (2000), “Why individuals voluntarily leave: perceptions ofhuman resource managers versus employees”, Asia Pacific Journal of Human Resources,Vol. 39 No. 1, pp. 23-41.

Campbell, D.J., Campbell, K.M. and Chia, H.B. (1998), “Merit pay, performance appraisal, andindividual motivation – an analysis and alternative”, Human Resource Management,Vol. 37 No. 2, pp. 31-146.

Chen, A., Lam, W. and Zhong, J.A. (2007), “Leader-member exchange and member performance: anew look at individual-level negative feedback-seeking behavior and team-levelempowerment climate”, Journal of Applied Psychology, Vol. 92 No. 1, pp. 202-12.

Curtis, A.B., Harvey, R.D. and Ravden, D. (2005), “Sources of political distortions in performanceappraisal”, Group and Organization Management, Vol. 30 No. 91, pp. 42-60.

Dyer, L., Schwab, D.P. and Theriault, R.D. (1976), “Managerial perceptions regarding salaryincrease data”, Personnel Psychology, Vol. 29 No. 1, pp. 233-42.

Feldman, D.C. (2000), “The Dilbert Syndrome: how employee cynicism about ineffectivemanagement is changing the nature of careers in organisations”, The American BehavioralScientist, Vol. 43 No. 8, pp. 1286-300.

Fletcher, C. and Williams, R. (1996), “Performance management, job satisfaction andorganisational commitment”, British Journal of Management, Vol. 7, pp. 69-179.

Fortin, M. (2008), “Perspectives on organizational justice: concept clarification, social contextintegration, time and links with morality”, International Journal of Management Reviews,Vol. 10 No. 2, pp. 93-126.

Fried, Y. and Ferris, G.R. (1987), “The validity of the job characteristics model: a review and ameta analysis”, Personnel Psychology, Vol. 40, pp. 287-322.

Fugate, M., Kinicki, A.J. and Prussia, G.E. (2008), “Employee coping with organisational change:an examination of alternative theoretical perspectives and models”, Personnel Psychology,Vol. 61, pp. 1-36.

Gardner, C.E. (2008), “Employee evaluation: is it worth the effort?”, DVM Magazine, January 1,p. 4F.

Ghorpade, J. and Chen, M.M. (1995), “Creating quality-driven performance appraisal systems”,Academy of Management Executive, Vol. 9 No. 1, pp. 32-9.

Green, F. (2004), “Work intensification, discretion and the decline in wellbeing at work”, EasternEconomic Journal, Vol. 30 No. 4, pp. 615-25.

Greenberg, J. (1986), “Determinants of perceived fairness of performance evaluations”, Journal ofApplied Psychology, Vol. 71 No. 2, pp. 320-4.

Griffeth, R.W., Hom, P.W. and Gaertner, S. (2000), “A meta-analysis of antecedents and correlatesof employee turnover: update, moderator tests and research implications for the nextmillennium”, Journal of Management, Vol. 26 No. 3, pp. 463-88.

Hair, J., Anderson, R., Tatham, R. and Black, W. (1999), Multivariate Data Analysis, 5th ed.,Prentice-Hall International, Englewood Cliffs, NJ.

PR39,3

392

Harter, J.K., Schmidt, F.L. and Hayes, T.L. (2002), “Business-unit-level relationship betweenemployee satisfaction, employee engagement and business outcomes: a meta analysis”,Journal of Applied Psychology, Vol. 87 No. 2, pp. 268-79.

Heathfield, S. (2007), “Performance appraisals don’t work – what does?”, The Journal for Qualityand Participation, Spring, pp. 6-9.

Hendrix, W.H., Robbins, Y., Miller, J. and Summers, T.P. (1998), “Effects of procedural anddistributive justice on the factors predictive of turnover”, Journal of Social Behaviour andPersonality, Vol. 98 No. 1, pp. 611-32.

Heneman, R.L. and Werner, J.M. (2005), Merit Pay: Linking Pay to Performance in a ChangingWorld, 2nd ed., Information Age Publishing, Greenwich, CT.

Hom, P. and Griffeth, R. (1995), Employee Turnover, South-Western College, Cincinnati, OH.

Iverson, R.D. (1992), Employee Intent to Stay: An Empirical Test of a Revision of the Price MuellerModel, University of Iowa, Iowa City, IA.

Jawahar, I.M. (2006), “An investigation of potential consequences of satisfaction with appraisalfeedback”, Journal of Leadership and Organizational Studies, Vol. 13 No. 2, pp. 14-29.

Kahn, R.L., Wolfe, D.M., Quinn, R.P. and Rosenthal, R.A. (1964), Organisational Stress: Studies inRole Conflict and Ambiguity, Wiley, New York, NY.

Keeping, L.M. and Levy, P.E. (2000), “Performance appraisal reactions: measurement, modellingand method bias”, Journal of Applied Psychology, Vol. 85 No. 5, pp. 708-23.

Kessler, I. and Purcell, J. (1992), “Performance-related pay: objectives and application”, HumanResource Management Journal, Vol. 2 No. 3, pp. 16-33.

Kiggundu, M.N. (1983), “Task interdependence and job design: test of a theory”, OrganizationalBehavior and Human Performance, Vol. 31, pp. 145-72.

Kuvaas, B. (2007), “Different relationships between perceptions of developmental performanceappraisal and work performance”, Personnel Review, Vol. 36 No. 3, pp. 378-97.

Latham, G.P., Skarlicki, D., Irvine, D. and Siegel, J.P. (1993), “The increasing importance ofperformance appraisals to employee effectiveness in organizational settings in NorthAmerica”, in Cooper, C. and Robertson, I. (Eds), International Review of Industrial andOrganizational Psychology, Wiley, Chichester, pp. 87-132.

Lind, E.A. and Tyler, Y. (1988), The Social Psychology of Procedural Justice, Plenum, New York,NY.

Locke, E. (1976), “The nature and consequences of job satisfaction”, in Dunnette, M.D. (Ed.),Handbook of Industrial and Organizational Psychology, Rand-McNally, Chicago, IL,pp. 1297-349.

Longenecker, C.O., Sims, H.P. and Gioia, D.A. (1987), “Behind the mask: the politics of employeeappraisal”, Academy of Management Executive, Vol. 1, pp. 183-93.

Mani, B.G. (2002), “Performance appraisal systems, productivity, and motivation: a case study”,Public Personnel Management, Vol. 31 No. 2, pp. 41-159.

Masterson, S.S., Lewis-McClear, K., Goldman, B.B. and Taylor, S.M. (2000), “Integrating justiceand social exchange: the differing effects of fair procedures and treatment on workoutcomes”, Academy of Management Journal, Vol. 43, pp. 738-48.

Merritt, D.M. (2007), “Appraising the performance appraisal”, Supervision, Vol. 68 No. 4, pp. 3-5.

Milkovich, G.T. and Newman, J.M. (2002), Compensation, 7th ed., McGraw-Hill Irwin, Boston,MA.

Milkovich, G.T. and Wigdor, A.K. (1991), Pay for Performance: Evaluating PerformanceAppraisal and Merit Pay, National Academy Press, Washington, DC.

Performanceappraisal

experience

393

Millward, N., Bryson, A. and Forth, J. (2000), All Change at Work?, Routledge, London.

Mitchell, T., Holton, B., Lee, T., Sablynski, C. and Erez, M. (2001), “Why people stay: using jobembeddedness to predict voluntary turnover”, Academy of Management Journal, Vol. 44No. 6, pp. 1102-21.

Moats-Kennedy, M. (1999), “The case against performance appraisal”,Across the Board, January,pp. 51-2.

Mobley, W.H. (1977), “Intermediate linkages in the relationship between job satisfaction andemployee turnover”, Journal of Applied Psychology, Vol. 62, pp. 237-40.

Morrell, K., Loan-Clarke, J., Arnold, J. and Wilkinson, A. (2008), “Mapping the decision to quit:a refinement and test of the unfolding model of voluntary turnover”, Applied Psychology:An International Review, Vol. 57 No. 1, pp. 128-50.

Mowday, R.T., Porter, L.W. and Steers, R.M. (1982), Employee-Organisation Linkages, AcademicPress, New York, NY.

Moynihan, D.P. and Pandey, S.K. (2007), “The ties that bind: social networks,person-organisation value fit and turnover intention”, Journal of Public AdministrationResearch and Theory, Vol. 18, pp. 205-27.

Murphy, K.R. and Cleveland, J.N. (1991), Performance Appraisal: An Organisational Perspective,Allyn & Bacon, Boston, MA.

Murphy, K.R. and Cleveland, J.N. (1995), Understanding Performance Appraisal: Social,Organizational and Goal-based Perspectives, Sage, Thousand Oaks, CA.

Nankervis, A.T. and Compton, A.R. (2006), “Performance management: theory in practice?”, AsiaPacific Journal of Human Resources, Vol. 44 No. 1, pp. 83-101.

Oldham, G.R., Kulik, C.T., Stepina, L.P. and Ambrose, M.L. (1986), “Relations between situationalfactors and the comparative referents used by employees”, Academy of ManagementJournal, Vol. 29 No. 3, pp. 599-608.

Pearce, J.L. and Porter, L.W. (1986), “Employee responses to formal performance appraisalfeedback”, Journal of Applied Psychology, Vol. 49, pp. 24-33.

Pettijohn, C., Pettijohn, L.S., Taylor, A.J. and Keillor, B.D. (2001), “Are performance appraisals abureaucratic exercise or can they be used to enhance sales-force satisfaction andcommitment?”, Psychology and Marketing, Vol. 18 No. 4, pp. 337-52.

Porter, L., Steers, R., Mowday, R. and Boulian, P. (1974), “Organisational commitment, jobsatisfaction and turnover among psychiatric technicians”, Journal of Applied Psychology,Vol. 59 No. 5, pp. 603-9.

Price, J. and Mueller, C. (1981), Professional Turnover: The Case of Nurses, SP Medical andScientific, New York, NY.

Riketta, M. (2002), “Attitudinal organizational commitment and job performance:a meta-analysis”, Journal of Organizational Behavior, Vol. 23 No. 3, pp. 257-66.

Roberts, G.E. (1998), “Perspectives on enduring and emerging issues in performance appraisal”,Public Personnel Management, Vol. 27 No. 3, pp. 301-20.

Roberts, H.V. and Sergesketter, B.F. (1993), Quality Is Personal: A Foundation for Total QualityManagement, The Free Press, New York, NY.

Robinson, S.L., Draatz, M.S. and Rousseau, D.M. (1994), “Changing obligations and thepsychological contract: a longitudinal study”, Academy of Management Journal, Vol. 37,pp. 137-52.

Rodgers, R. and Hunter, J.E. (1991), “Impact of management by objectives on organizationalproductivity”, Journal of Applied Psychology, Vol. 76, pp. 322-36.

PR39,3

394

Rosse, J.G. (1988), “Relations among lateness, absence and turnover: is there a progression ofwithdrawal?”, Human Relations, Vol. 41 No. 7, pp. 517-31.

Schay, B.W. (1988), “Effect of performance-contingent pay on employee attitudes”, PublicPersonnel Management, Vol. 17 No. 2, pp. 237-50.

Schay, B.W. (1993), “In search of the Holy Grail: lessons in performance management”, PublicPersonnel Management, Vol. 22 No. 4, pp. 649-68.

Sharma, S. (1996), Applied Multivariate Techniques, Wiley, New York, NY.

Shields, S. (2007), Managing Employee Performance and Reward, Cambridge University Press,Sydney.

Sweeney, P.D. and McFarlin, D.B. (1993), “Workers’ evaluations of the ends and the means:an examination of four models of distributive justice”, Organisational Behaviour andHuman Decision Processes, Vol. 55, pp. 23-40.

Tang, T. and Sarsfield-Baldwin, L. (1996), “Distributive and procedural justice as related tosatisfaction and commitment”, SAM Advanced Management Journal, Vol. 61 No. 3,pp. 25-30.

Taylor, P.J. and Pierce, J.L. (1999), “Effects of introducing a performance management system onemployees’ subsequent attitudes and effort”, Public Personnel Management, Vol. 28 No. 3,pp. 423-52.

Taylor, S.M., Tracy, K.B., Renard, M.K., Harrison, J.K. and Carroll, S.J. (1995), “Due process inperformance appraisal: a quasi-experiment in procedural justice”, Administrative ScienceQuarterly, Vol. 40 No. 3, pp. 495-523.

Tetrick, L.E. and La Rocco, J.M. (1987), “Understanding prediction and control as moderators ofthe relationship between perceived stress, satisfaction and psychological wellbeing”,Journal of Applied Psychology, Vol. 72 No. 4, pp. 538-44.

Thayer, F.C. (1987), “Performance appraisal and merit pay systems: the disasters multiply”,Review of Public Personnel Administration, Vol. 7 No. 2, pp. 36-53.

Thibaut, J.W. and Walker, L. (1975), Procedural Justice: A Psychological Analysis, Erlbaum,Hillsdale, NJ.

Thomas, S.L. and Bretz, R.D. (1994), “Research and practice in performance appraisal: evaluatingperformance in America’s largest companies”, SAM Advanced Management Journal,Spring, pp. 28-37.

Treadway, D.C., Ferris, G.R., Duke, A.B., Adams, G.L. and Thatcher, J.B. (2007), “The moderatingrole of subordinate political skill on supervisors’ impressions of subordinate ingratiationand ratings of subordinate interpersonal facilitation”, Journal of Applied Psychology, Vol. 92No. 3, pp. 848-55.

VanVoorhis, R.W. and Levinson, E.M. (2006), “Job satisfaction among school psychologists:a meta-analysis”, School Psychology Quarterly, Vol. 21 No. 1, pp. 77-90.

Wanguri, D.M. (1995), “A review, an integration and a critique of cross-disciplinary research onperformance appraisals, evaluations and feedback: 1980-1990”, The Journal of BusinessCommunication, Vol. 32, pp. 267-93.

Watson, D., Pennebaker, J.W. and Folger, R. (1987), “Beyond negative affectivity: measuringstress and satisfaction in the workplace”, Journal of Organizational BehaviourManagement, Vol. 8 No. 2, pp. 141-58.

Wiese, D.S. and Buckley, W.R. (1998), “The evolution of the performance appraisal process”,Journal of Management History, Vol. 4 No. 3, pp. 233-49.

Performanceappraisal

experience

395

Youngcourt, S.S., Leiva, P.I. and Jones, R.G. (2007), “Perceived purposes of performanceappraisal: correlates of individual- and position-focused purposes on attitudinaloutcomes”, Human Resource Development Quarterly, Vol. 18 No. 3, pp. 315-43.

About the authorsMichelle Brown is an Associate Professor in the Department of Management and Marketing,University of Melbourne, Victoria. Her research interests are in the areas of pay and performancemanagement systems, employee involvement and organisational cynicism. Her research seeks tounderstand the unintended consequences of human resource management policies and practices.Michelle Brown is the corresponding author and can be contacted at: brownm@unimelb.edu.au

Douglas Hyatt is a Professor at the Rotman School of Management, University of Toronto.His research interests include compensation, occupational health and safety, workers’compensation and, more recently, earnings and working conditions in the music industry andthe valuation of copyrights.

John Benson is Professor and Head of the School of Management at the University of SouthAustralia. His major research interests are Japanese management and unions, the restructuringof Chinese and Vietnam industry, outsourcing and knowledge work. Hemost recently edited amonograph – Trade Unions in Asia (Routledge, London, 2008) with Associate ProfessorYing Zhu.

PR39,3

396

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints

Reproducedwith permission of the copyright owner. Further reproduction prohibitedwithout permission.