IJTD_2007

The effects of trainingdesign, individual

characteristics and workenvironment on transfer

of training

Raquel Velada, António Caetano,John W. Michel, Brian D. Lyons

and Michael J. Kavanagh

This study aims to gain insight into some of the factors thatdetermine the transfer of training to the work context. Thepresent research examined the relationship between three typesof predictors on transfer of training, including training design,individual characteristics and work environment. Data wascollected at two points in time from 182 employees in a largegrocery organization. The results indicated that transferdesign, performance self-efficacy, training retention and perfor-mance feedback were significantly related to transfer of train-ing. Contrary to expectation, supervisory support was notsignificantly related to transfer of training. These resultssuggest that in order to enhance transfer of training, organiza-tions should design training that gives trainees the ability to

❒ Raquel Velada, PhD student, ISCTE – Institute of Social Sciences and Business Studies, Department ofSocial and Organizational Psychology, Avenida das Forças Armadas, 1649-026 Lisbon, Portugal. Email:[email protected]ónio Caetano,Associate Professor, ISCTE – Institute of Social Sciences andBusiness Studies, Lisbon, Portugal. Email: [email protected]. John W. Michel, Assistant Professor,Department of Management, College of Business and Economics, Towson University, 8000 York Road,Towson, MD 21252, USA. Email: [email protected]. Brian D. Lyons, Assistant Professor, Departmentof Management, Craig School of Business, California State University, Fresno, 5241 N. Maple Ave.,Fresno, CA 93740, USA. Email: [email protected]. Michael J. Kavanagh, Professor Emeritus,Department of Management, School of Business, University at Albany, 1400 Washington Ave., Albany,NY 12222, USA. Email: [email protected] study was supported by a grant to the first author from the Foundation for Science and Technologyof Portugal (SFRH/BD/10558/2002).

International Journal of Training and Development 11:4ISSN 1360-3736

© 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2QD,UK and 350 Main St., Malden, MA 02148, USA.

282 International Journal of Training and Development

mailto:[email protected]�nioCaetano

mailto:[email protected]




transfer learning, reinforces the trainee’s beliefs in their abilityto transfer, ensures the training content is retained over timeand provides appropriate feedback regarding employee job per-formance following training activities.

IntroductionOrganizations spend an immense amount of time and money on training in order tofacilitate employees’ learning of job-related competencies (Cascio, 2000; Noe et al.,2006). According to Training Magazine’s ongoing industry report, US companies spendmore than $50 billion annually on formal training (Dolezalek, 2004). Moreover, invest-ment in training activities has increased all over the world in recent years. As a resultof the financial investments organizations make in training, it is important to provideevidence that training efforts are being fully realized (Cascio, 2000; Dowling & Welch,2005). In other words, it is important for organizations to ensure that training leads todesired work outcomes such as increases in job performance.

Research has demonstrated that training efforts are unlikely to result in positivechanges in job performance unless the newly trained competencies are transferredto the work environment (see Baldwin & Ford, 1988; Montesino, 2002; Rouiller &Goldstein, 1993). As a result, there has been an increased effort to understand theantecedents and consequences of the transfer of training process.

Baldwin and Ford (1988) define transfer of training as ‘the degree to which traineeseffectively apply the knowledge, skills, and attitudes gained in the training context tothe job’ (p. 63). This suggests that transfer of training first requires a trainee to learnnew job-related competencies (Velada & Caetano, 2007). By learning, we are referringto a relatively permanent change in knowledge, skills and behaviors of trainees (Weiss,1990). After learning and retaining the training content, trainees should transfer theknowledge and/or skills accrued to the work context with the intention of improvingjob performance over time (Noe et al., 2006).

However, it has been estimated that only about 10 per cent of all training experi-ences are transferred from the training environment to the job (Baldwin & Ford, 1988).Although this is a lower-bound estimate, Wexley and Latham (2002) suggest thatalthough approximately 40 per cent of content is transferred immediately followingtraining, the amount transferred falls to 25 per cent after 6 months and 15 per centafter 1 year. This suggests that as time passes, trainees may be unable or less moti-vated to retain and use the information gained in the training program. Furthermore,this indicates that much of the time and money invested in training is never fullyrealized, because only a small percentage of the training effectively results in perma-nent transferability to the workplace. As a result, understanding and improving thetransfer of training process has become a primary concern for training researchersand practitioners.

Although several studies have been conducted to understand the transfer of trainingprocess, conceptual models for understanding this process are limited. Kavanagh (1998)developed a multi-level multistage process to help understand the complexities of thetransfer of training process. Specifically, he suggested that training transfer is influ-enced by several variables at different levels of analysis (e.g. individual, supervisor,workgroup and organization) and in different stages in the training process (e.g.pre-training, training and post-training). To date, the extant literature (e.g. Baldwin &Ford, 1988; Holton, 1996, 2005; Tracey et al., 1995) has identified three main determi-nants of training transfer: training design or enabling factors, individual factors or traineecharacteristics, and work environment or transfer climate.

Holton et al. (2000) developed the Learning Transfer System Inventory (LTSI) toevaluate the specific factors on those dimensions affecting the transfer of trainingprocess. Based on the Human Resource Development Research and Evaluation Model(Holton, 1996), the LTSI includes 16 factors that either facilitate or inhibit trainingtransfer. It has been argued that the LTSI is the only research-based instrument for

Transfer of training 283© 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd.

assessing a comprehensive set of factors affecting transfer of learning (Chen et al.,2005; Holton et al., 2000). Whereas studies have been conducted to validate the LTSImeasure (Chen et al., 2005; Holton et al., 2000; Khasawneh et al., 2006), little has beendone to empirically demonstrate the relationship between LTSI measures and transferof training. Additionally, the LTSI appears to exclude important individual differencevariables such as cognitive ability, locus of control and training retention (e.g. Baldwin& Ford, 1988), and environmental factors such as continuous learning culture (e.g.Tracey et al., 1995).

Based on a review of the transfer of training literature, Kavanagh (1998) concludedthat there are significant gaps in the empirical literature for training design, individualvariables and organizational environment factors. This suggests the need for studies toinvestigate the impact of these issues on the transfer of training process. Thus, consid-ering the three main influences on transfer of training previously identified by Holton(1996, 2005), this study aims to contribute to the theory of training transfer by empiri-cally analysing how different sets of variables simultaneously influence the transfer oftraining.

As depicted in Figure 1, we hypothesize that several variables related to trainingdesign, individual characteristics and work environment affect transfer of training. Thefollowing section provides a brief overview of the literature regarding the influence oftraining design, individual characteristics and work environment on transfer of train-ing, giving special attention and theoretical justification to the relationships that will betested in this study.

Influences of training design on transfer of trainingAccording to the training literature, there are several training design factors thatinfluence transfer of training: instructional techniques and learning principles (e.g.Alvarez et al., 2004); self-management and relapse prevention strategies (e.g. Tzineret al., 1991; Wexley & Nemeroff, 1975) and goal setting (e.g. Gist et al., 1990). Thus,organizations should design their training programs to include such factors thatincrease the likelihood of transfer. Accordingly, the LTSI measures such a factor, trans-fer design. Transfer design refers to the degree to which training has been designed anddelivered in such a way that provides trainees the ability to transfer learning back to thejob (Holton et al., 2000). Holton et al. (2000) argue that part of transfer design is thedegree to which training instructions match job requirements. Trainees are more likelyto transfer the training content to the work context when they perceive that the trainingprogram was designed and delivered in such a way that maximizes the trainee’s ability

TRAINING DESIGN

Transfer design

INDIVIDUAL CHARACTERISTICS

Self-efficacy

Training retention

WORK ENVIRONMENT

Feedback

Supervisor support

Transfer of training

Figure 1: Simplified model of transfer of training.

284 International Journal of Training and Development© 2007 The Authors. Journal compilation © 2007 Blackwell Publishing Ltd.

to transfer the training to the job (Holton, 1996; 2005). Consequently, when traineeshave previous knowledge and practice on how to apply the newly learned knowledgeand skills to the job and when training instructions are congruent with job require-ments, an increased likelihood of transfer should exist. Based on these assertions, wehypothesized the following:

H1 Trainees’ perceptions of transfer design will be positively related to transfer oftraining.

Influences of individual characteristics on transfer of trainingIn addition to training design, there are several individual characteristics that affect thetransfer of training process. Some of these characteristics include cognitive ability,locus of control, conscientiousness, achievement motivation, motivation to learn and totransfer, anxiety, self-efficacy, and valence (e.g. Colquitt et al., 2000; Mathieu et al., 1992;Noe, 1986). Others include job involvement, organizational commitment, organiza-tional cynicism and job satisfaction (e.g. Mathieu et al., 1993; Tannenbaum et al., 1991;Tesluk et al., 1995; Velada & Caetano, 2007).

Of these characteristics, performance self-efficacy has been found to strongly relate toboth learning (Gist et al., 1991; Mathieu et al., 1992; Quinones, 1995) and transfer oftraining (e.g. Ford et al., 1998). Additionally, some studies (e.g. Ford et al., 1998) haveindicated that trainees with higher self-efficacy are more likely to transfer the training tothe job. Holton et al. (2000) defined performance self-efficacy as an individual’s generalbelief that they are able to change their performance when desired. Hence, when atrainee feels confident in his or her ability to perform, the more likely he or she willtransfer such knowledge and/or skill to the job. Thus, we hypothesized the following:

H2 Performance self-efficacy will be positively related to training transfer.

As suggested before, trainees must have the ability to retain the knowledge instilledduring the training program to facilitate the transfer process. Similar to cognitiveability, training retention is the degree to which trainees retain the content after train-ing is completed. Baldwin and Ford (1988) argue that learning retention outcomes aredirectly associated with the generalization and maintenance of training effects on thejob. They argue that in order for trained skills to be transferred, they first must belearned and retained. Although we were unable to find any previous research empiri-cally demonstrating this relationship, we believe it is an important aspect in the transferof training process. Therefore, we hypothesized the following:

H3 Training retention will be positively related to training transfer.

Influences of work environment on transfer of trainingWork environment variables have been investigated less often than training design andindividual characteristics (Alvarez et al., 2004; Baldwin & Ford, 1988; Holton et al., 1997;Tannenbaum & Yukl, 1992). However, a number of studies have shown that environ-mental factors are important for understanding the transfer of training process (e.g.Baldwin & Ford, 1988; Lance et al., 2002; Rouiller & Goldstein, 1993; Tracey et al., 1995).

Two dimensions of the work environment that have received attention with regardto transfer of training include organizational culture and climate (e.g. Baldwin & Ford,1988; Rouiller & Goldstein, 1993; Tracey et al., 1995) Tracey et al. (1995) stressed theimportance of both transfer of training climate and continuous learning culture as workenvironment variables that have a significant impact on the post-training behaviors.Some indicators of transfer climate include performance feedback, peer support, super-visor support and supervisor sanctions (e.g. Baldwin & Ford, 1988; Holton et al., 2000;Tracey & Tews, 2005; Tracey et al., 1995).

Research has indicated that when employees perceive that the organizational climateis supportive, they are more likely to apply their new knowledge in the work environ-


ment (see Baldwin & Ford, 1988; Tracey et al., 1995). Performance feedback includes anindication from management about how well one is performing his or her job (Holtonet al., 2000). Specifically, feedback regarding the newly learned knowledge and skillsand how these relate to job performance increases the likelihood of its transfer to thework context (Reber & Wallin, 1984). Therefore, we hypothesized the following:

H4 Feedback regarding the trainee’s performance after training will be positivelyrelated to training transfer.

Supervisor support can be described as the extent to which supervisors support andreinforce the use of newly learned knowledge and skills on the job (Holton et al., 2000).Although there is some contradictory evidence (e.g. Russell et al., 1985), the dominantliterature suggests that when trainees perceive that their supervisors support the appli-cation of newly developed knowledge and skills, they are more likely to transfer thesecompetencies back to the job (e.g. Bates et al., 2000; Brinkerhoff & Montesino, 1995;Colquitt et al., 2000; Noe, 1986; Tracey & Tews, 2005). Thus, we hypothesized thefollowing:

H5 Supervisor support for training transfer will be positively related to trainingtransfer.

MethodProcedure

The initial sample consisted of 336 employees from nine stores of a grocery marketcompany in Portugal. All training occurred in a classroom within each store, althoughthe training content differed to reflect each store’s job composition. In a period of 1month, participants attended one of the following training programs: customer service,environmental issues, security, prevention and hygiene in the workplace. In order toaddress all the training objectives and to facilitate the knowledge acquisition as well asbehavior change, different training methods were used including lecture, discussion,simulations (e.g. role play) and audiovisual techniques (e.g. video tapes).

At the end of the training program, 336 trainees completed a self-report survey toassess perceptions of transfer design, performance self-efficacy and supervisor support(Time 1 data collection). Three months after the training sessions, surveys were sent tothe trainees to complete and return (Time 2 data collection). These surveys measuredthe trainees’ perceptions of training retention, feedback and training transfer. Of the336 surveys distributed, 219 were returned. This loss of 117 respondents over the3-month period was due in part to the high rate of personnel turnover in these jobs. Inaddition, due to incomplete survey completion and incorrect coding, 37 more surveyshad to be discarded, leaving a final sample of 182 matched pairs.

Participants

Of the 182 study participants, 62 per cent were female. The mean age for the partici-pants was 30, with a range of 19–58. The average tenure of the participants in thecompany was 4.5 years with 46.6 per cent having been employed in the company forless than 1 year. In terms of education, 52.3 per cent had a high school degree whileonly 5.2 per cent had a college degree. Furthermore, 79 per cent of the participants werein non-supervisory roles.

Measures

The independent variables of transfer design, performance self-efficacy, supervisorsupport and performance feedback were measured with items from the LTSI (Holtonet al., 2000). The measure of training retention was created specifically for this study.


The measure of the dependent variable, training transfer, was based on Tesluk et al.’s(1995) research. All variables were measured using a 5-point Likert scale.

Transfer designTransfer design is defined as the ‘degree to which (1) training has been designed anddelivered to give trainees the ability to transfer learning to the job, and (2) traininginstructions match job requirements’ (Holton et al., 2000, p. 345). Transfer design wasmeasured with four items with an alpha of 0.78. A sample item is ‘The way thetrainer(s) taught the material made me feel more comfortable I could apply it’.

Performance self-efficacyPerformance self-efficacy was defined as ‘an individual’s general belief that they areable to change their performance when they want to’ (Holton et al., 2000, p. 346).Performance self-efficacy was measured with four items with an alpha of 0.76. Asample item is ‘I am confident in my ability to use new skills at work’.

Supervisor supportSupervisor support was defined as the ‘extent to which supervisors/managers supportand reinforce use of training on the job’ (Holton et al., 2000, p. 345). The measure ofsupervisor support consisted of six items with an alpha of 0.89. A sample item is ‘Mysupervisor meets with me to discuss ways to apply training on the job’.

Performance feedbackPerformance feedback was defined as ‘formal and informal indicators from an organi-zation about an individual’s job performance’ (Holton et al., 2000, p. 346). The measureof performance feedback consisted of three items with an alpha of 0.72. A sample itemis‘After training, I received feedback from people on how well I am applying what Ilearned’.

Training retentionBecause a measure of training retention could not be found in the literature, three itemswere created for this study (a = 0.70). The three items included ‘I still remember themain topics that I have learned in the training course,’ ‘I can easily say several thingsthat I have learned in the training course’ and ‘I had never thought again about thetraining content (reverse coded)’.

Training transferTesluk et al. (1995) developed a three-item scale to measure the extent to which indi-viduals transfer the knowledge and skills presented in training sessions to their corejobs (a = 0.87). A sample item is ‘I have been using the skills presented in the trainingcourse to help improve my performance’.

ResultsFactor analysis

In order to assess the extent to which each of the variables represents a separateconstruct, an exploratory factor analysis (EFA) was performed. An EFA is warrantedbecause some of the items were adapted from previous research to correspond to thepurposes of this study and, as a result, the dimensionality of the survey was not assuredfor this sample. A principal component analysis (PCA) followed by varimax rotationwas performed on the 23 survey items. Two decision rules were imposed to ascertainthe number of components and items comprised in the factor analytic model: (1)eigenvalues over 1.0 were retained and (2) components that discretely loaded fromother factors in the screen plot were retained. As depicted in Table 1, six componentswere deduced from the 23 items and these components accounted for 55.43 per cent ofthe variance in the items.


Table 1: Item means, standard deviations and varimax rotated factor loadings

Item Factor loadings

M SD 1 2 3 4 5 6

Supervisor supportWays to apply training on

the job3.47 0.83 0.85 0.09 0.14 0.13 0.09 -0.01

Problems in usingtraining

3.56 0.84 0.79 0.12 0.04 0.16 0.08 0.07

Interest in the traininglearning

3.71 0.82 0.72 0.21 0.06 0.07 0.05 0.05

Feedback on performanceafter training

3.73 0.82 0.65 0.20 0.15 -0.04 0.31 0.01

Performance goals basedon training

3.63 0.82 0.63 0.14 0.26 -0.03 0.36 0.03

Goals to apply training onthe job

3.77 0.85 0.56 0.22 0.14 0.08 0.21 0.02

Transfer designExamples about ways to

use learning on the job4.13 0.59 0.19 0.77 0.19 0.06 0.07 -0.02

Activities and exercises onhow to apply learning

3.98 0.67 0.10 0.63 0.17 0.19 -0.06 -0.03

Teaching on how to applylearning on the job

3.84 0.70 0.18 0.61 0.10 0.07 0.04 0.07

Understanding of thetraining designers onhow to use learning

3.98 0.69 0.23 0.54 0.18 0.02 0.17 0.01

Performance self-efficacyAbility to use newly

learned skills on the job3.99 0.64 0.08 0.16 0.80 0.08 0.05 0.05

Confidence in the abilityto use new skills atwork

4.00 0.61 0.10 0.22 0.71 0.11 0.01 0.07

Using learning even indifficult situations

3.98 0.61 0.30 0.17 0.51 0.14 0.04 -0.02

Overcoming obstacles touse new skills orknowledge

3.95 0.64 0.17 0.16 0.40 0.21 0.26 -0.01

Training transferUsing new skills to

improve performance3.76 0.71 0.01 0.07 0.14 0.75 0.24 0.25

Training helped toimprove job performance

3.77 0.69 0.13 0.19 0.15 0.73 0.30 0.16

Incorporating learnedskills into daily workactivities

3.77 0.75 0.08 0.17 0.21 0.65 0.31 0.20

Performance feedbackPost-training conversations

with people about howto improve jobperformance

3.38 0.77 0.28 0.08 0.12 0.17 0.72 0.02

Feedback on how well wasthe training application

3.20 0.82 0.14 -0.04 0.01 0.21 0.59 0.05


Overall, the results of the PCA were consistent with the extant literature. Thefirst principal component (six items) was labeled ‘supervisor support’ (a = 0.89); thesecond principal component (four items) was labeled ‘transfer design’ (a = 0.78);the third principal component (four items) was labeled ‘performance self-efficacy’(a = 0.76); the fourth principal component (three items) was labeled ‘training transfer’(a = 0.87); the fifth principal component (three items) was labeled ‘performance feed-back’ (a = 0.72); and finally, the last principal component (three items) was labeled‘training retention’ (a = 0.70). Based on the results from the EFA, we used these 5-factorscores to test the aforementioned hypotheses.

Hypothesis testing

Means, standard deviations, intercorrelations and reliabilities of the variables are pre-sented in Table 2. In general, results indicated that all independent variables of ourstudy were positively and significantly related to training transfer. To determine theextent to which transfer design, performance self-efficacy, training retention, feedbackand supervisor support affect transfer outcome, hierarchical regression analysis wasperformed (Cohen & Cohen, 1983). The independent variables entry was based on thelogic of our hypothesized model. Additionally, demographic variables and the natureof the training were initially included in the analysis but were subsequently discardedbecause they did not significantly predict transfer of training. The results of the hier-archical regression analysis are presented in Table 3.

The results of step 1 showed that transfer design significantly predicted transfer oftraining (b = 0.31, p < 0.01), supporting our first hypothesis. In step 2, we entered theindividual factors. Similarly to transfer design, both performance self-efficacy (b = 0.30,p < 0.01) and training retention (b = 0.30, p < 0.01) significantly predicted transfer oftraining, supporting our second and third hypotheses, respectively. Finally, in step 3 weentered the environmental variables. As shown in Table 3, only feedback significantlypredicted transfer of training (b = 0.42, p < 0.01), providing support for our fourthhypothesis. Although the bivariate correlation between supervisor support and train-ing transfer was positive and significant (r = 0.31, p < 0.001), the hierarchical regressiondid not support this hypothesis (b = -0.05, p = ns) Therefore, the fifth hypothesis wasonly partially supported. Overall, these variables explained 42 per cent of the totalvariance in transfer of training.

Table 1: Continued

Item Factor loadings

M SD 1 2 3 4 5 6

People said things thathelped to improve jobperformance

3.50 0.81 0.17 0.10 0.03 0.26 0.47 -0.08

Training retentionRemember the main topics

learned in the training3.90 0.65 0.00 0.06 -0.07 0.08 0.08 0.99

Easily say several thingslearned in the training

3.82 0.70 0.02 -0.01 0.09 0.10 0.05 0.67

Never thought againabout the trainingcontenta

3.98 0.87 0.05 -0.01 0.04 0.21 -0.12 0.37

Note: Higher item loadings on each factor are underlined. A 5-point Likert response scale wasused for all items.a item was reversed coded.n = 182.


DiscussionThe results of this research demonstrate the importance of training design, individualcharacteristics and the influence of the work environment on transfer of training.

The results for H1 indicated that transfer design, which assessed how applicable thetraining was to the job, positively influenced transfer of training. These results reinforcethe notion that organizations should be aware of how well the content of the trainingprogram, in terms of the use of activities, examples and exercises, is focused on theapplication of on-the-job learning.

The second and third hypotheses involving performance self-efficacy and the reten-tion of the training content were also significantly related to transfer of training. Suchresults suggest that when trainees believe in their capabilities to transfer learning andwhen they retain training content, they are more likely to perceive that they havetransferred the training to the work context.

The impact of the work environment in terms of feedback on transfer of training, aspredicted in hypothesis 4, was also supported in this research, indicating that thefeedback from others in the organization regarding the trainee’s performance aftertraining influences perceptions of training transfer. Contrary to what was predicted,however, hypothesis 5 was not supported. In this instance, supervisor support did notsignificantly predict transfer of training.

Table 2: Scale item means, standard deviations, intercorrelations and reliabilities

Variables M SD 1 2 3 4 5 6

(1) Transfer design 3.96 0.51 (0.78)(2) Performance self-

efficacy3.98 0.47 0.43** (0.76)

(3) Training retention 3.90 0.58 0.05 0.08 (0.70)(4) Performance

feedback3.36 0.64 0.19* 0.25** 0.06 (0.72)

(5) Supervisorsupport

3.65 0.66 0.42** 0.41** 0.08 0.43** (0.89)

(6) Training transfer 3.77 0.63 0.31** 0.40** 0.33** 0.50** 0.31** (0.87)

Note: A 5-point Likert scale was used for all variables.* p < 0.01.** p < 0.001.

Table 3: Hierarchical regression analysis results

Independent variables Dependent variables

Step 1 Step 2 Step 3b b b

Transfer design 0.31** 0.17* 0.14*Performance self-efficacy – 0.30** 0.23**Training retention – 0.30** 0.29**Performance feedback – – 0.42**Supervisor support – – -0.05R2 0.10 0.27 0.42Adj. R2 0.09 0.26 0.41

Note: Dependent variable: training transfer.* p < 0.05.** p < 0.01.


One strength of this research was that it allowed us to compare the results of transferof training research using two different methodologies. One methodology assumesthat in order to evaluate transfer of training, one must have evaluations of changes in jobperformance as measured by persons (sources) other than the trainees (Lance et al.,2002; Tracey et al., 1995). These other persons could include supervisors, co-workersand customers. However, it is usually difficult to gain access to and collect data frompersons other than the trainees. The other methodology used in this study is to collectall data on the research variables from the trainees. Despite the fact that some argueagainst the use of self-report ratings of job performance, it is likely that the trainees arethe most important and valid source of the measurement of job performance as theirperceptions will drive their motivation and performance. Whereas using data collectedfrom other sources (i.e. supervisors, peers and customers) may provide slightly differ-ent ratings, data collected using only a trainee self-report method can provide similarresults as demonstrated in the present research. Furthermore, because it is so importantto measure transfer of training in order to assess the practical and financial aspects oftraining effectiveness (Cascio, 2000), using this latter paradigm may be simpler and lessexpensive for human resource and managerial professionals. The important point isthat it may influence more organizations to evaluate transfer of training with empiricaldata rather than assuming that transfer always occurs.

Theoretical and practical implicationsResults from this study have potentially important implications for future research andpractice. In general, the results of this research argue for examining all aspects of thetraining process when conducting research on transfer of training. Hence, these resultsprovide empirical evidence to the aforementioned theoretical models (e.g. Baldwin &Ford, 1988; Holton, 1996, 2005; Kavanagh, 1998) suggesting that transfer of training isimpacted by the training design, characteristics of the trainee and contextual factorssuch as feedback regarding post-training job performance. Future research shouldexamine pre-training factors in a similar study in order to determine if the combinedfactors provide better prediction of transfer of training. As indicated earlier, withoutthe effective transfer of training from the training context to the organizational envi-ronment, the costs and time spent in training is simply wasted.

More specifically, this study supports the theoretical literature (e.g. Holton, 1996,2005) that suggest the importance of a transfer design that maximizes the trainee’sability to transfer in enhancing training transfer. The present study also extendsBaldwin and Ford’s (1988) work by demonstrating empirically that for training to betransferred, the training content should be retained over time. Finally, the results of thisstudy reinforce the role of performance self-efficacy (Ford et al., 1998) and performancefeedback (Reber & Wallin, 1984) on the explanation of training transfer.

However, results of this study do not support the dominant literature (e.g. Bateset al., 2000; Brinkerhoff & Montesino, 1995) that indicate that supervisory support is acritical variable in transfer of training. The results regarding the hypothesis in which itwas predicted that supervisor support would affect transfer of training could seempuzzling at first because the previous literature has typically supported this relation-ship (Baldwin & Ford, 1988; Lance et al., 2002; Rouiller & Goldstein, 1993; Tracey et al.,1995). However, some gaps persist in the literature regarding the specific supervisoryfactors that influence transfer. Like in the prevalent literature, this study consideredonly the post-training supervisory dimensions like meetings and feedback. Perhapssupervisory interventions in pre-training and during the training could have the stron-ger impact on transfer of training. Clearly, this is an avenue for future research.

An interesting aspect of the findings of this study is a comparison with previousresearch using a different methodology (Baldwin & Ford, 1988; Lance et al., 2002;Rouiller & Goldstein, 1993; Tracey et al., 1995), as well as using participants from acountry other than the United States. Most of the cited transfer of training research hasused pre-post designs using a sample from the United States, whereas this study useda post-test design with a sample from Portugal. Thus, the different results generated


with the supervisor support variable could also be attributed to these methodologicaland cultural differences. More important, however, are the similarities in findingsbetween this study and those done in the United States.

Regarding psychometric implications, the results from this study provided somecriterion validation evidence for the LTSI measure. As noted in the hypotheses testing,the results of this study found strong relationships between some LTSI dimensions andtransfer of training.

Based on the results of this study, we can argue that for organizations to maximizetheir return on investment with regards to training and development, they need tofocus on all three determinants of transfer of training: training design, individualcharacteristics and work environment. First, organizations need to ensure that trainingis designed such that it matches the ability level of trainees. This will help ensure thattrainees have the ability to (1) learn the training material and (2) utilize the knowledgeand skills accrued during training outside of the learning environment. Second, orga-nizations can improve transfer of training by ensuring that trainees believe that theyhave the capabilities to successfully learn the new material and utilize their newknowledge, skills and abilities (KSAs) on the job. This can be improved by (1) showingtrainees that other employees who have received the training have successfullyimproved their job performance, (2) providing trainees the opportunity to experiencemastery of the training material in the training environment and (3) modeling theappropriate behaviors so that trainees can conceptualize how the KSAs can be utilizedoutside of the training context. Organizations should also conduct follow-up assess-ments after the training to ensure that the training content is retained over the time.

Finally, it is important for organizations to create environments that support thetransfer of newly trained KSAs to the work environment. In other words, traineesshould feel that they will receive the support and feedback necessary regarding theirperformance from the organization, supervisor and co-workers in order to effectivelytransfer the training. One way this can be accomplished is by creating a climate inwhich all employees perceive that training is an important aspect of organizational lifethat will help employees become productive members of the organization (Baldwin &Ford, 1988; Tracey et al., 1995).

LimitationsSeveral limitations of this study should be noted. First, common method variance couldbe an issue as data on the predictor and criterion variables were collected from the samesource. However, we controlled for common method bias using both statistical andprocedural methods. Statistically, we conducted Harman’s single-factor test and exam-ined the unrotated factor solution (Podsakoff et al., 2003). Results of the EFA demon-strated that no single factor accounted for the majority of the covariance among themeasures. Procedurally, we controlled for common method bias by collecting ourpredictor variables at time 1 and our criterion variables at time 2 as suggested byPodsakoff et al. (2003).

A second limitation is that transfer of training was measured by self-report ratherthan actual behavior and this may have influenced the pattern of results. Nevertheless,in addition to the above-mentioned argument for using self-report performanceratings, utilizing specific items and anonymous and confidential surveys might haveenhanced the accuracy of the self-report data. Also, previous research has used similarself-report measures of training transfer (e.g. Chiaburu & Tekleab, 2005; Facteau et al.,1995; Tesluk et al., 1995), showing evidence that trainees can accurately self-report theirlevels of training transfer. Finally, as mentioned before, the results of this researchusing self-report data are in fair agreement with previous transfer of training resultsusing both self-report and supervisor evaluations of performance (Lance et al., 2002;Tracey et al., 1995). This convergence of results using different research paradigmsprovides evidence that self-report data, collected anonymously, may be adequate for theevaluation of transfer of training. However, for future research, we do suggest the useof additional measures collected from several sources (e.g. supervisors, peers and


subordinates) in order to reinforce the trainees ratings or, when possible, the use ofdirect and objective measures of the trainee’s on-the-job performance.

A third limitation is the attrition rate of our sample between the two points of datacollection. However, our final sample of 182 participants yielded passable statisticalpower to conduct our hypotheses testing.

Finally, because we collected data from only one organization, these results may notgeneralize to other organizations or industries. This may be especially relevant for thefindings regarding supervisor support. It is likely that in other organizational environ-ments, supervisory support will have a larger impact on transfer of training as evi-denced by previous research (Baldwin & Ford, 1988; Lance et al., 2002; Rouiller &Goldstein, 1993; Tracey et al., 1995). Consequently, future research should examine thegeneralizability of our results in different organizational contexts.

ConclusionsTraining practitioners and researchers have not yet extensively studied the empiricaland simultaneous effects of the three major determinants (training design, individualcharacteristics and work environment) of transfer of training. This study attempted tofill this gap by analysing the influence of these determinants on training transfer. Thefindings indicated that transfer design, performance self-efficacy, training retentionand performance feedback were significantly related to transfer of training over time.This suggests that it is important that training researchers and practitioners examine allaspects of the training process when conducting research on transfer of training.

ReferencesAlvarez, K., Salas, E. and Garofano, C. M. (2004), ‘An integrated model of training evaluation and

effectiveness’, Human Resource Development Review, 3, 385–416.Baldwin, T. T. and Ford, J. K. (1988), ‘Transfer of training: a review and directions for future

research’, Personnel Psychology, 41, 63–105.Bates, R. A., Holton, E. F. III, Seyler, D. L. and Carvalho, M. A. (2000), ‘The role of interpersonal

factors in the application of computer-based training in an industrial setting’, Human ResourceDevelopment International, 3, 19–42.

Brinkerhoff, R. O. and Montesino, M. U. (1995), ‘Partnerships for training transfer: lessons froma corporate study’, Human Resource Development Quarterly, 6, 263–74.

Cascio, W. F. (2000), Costing Human Resources: The Financial Impact of Behavior in Organizations, 4thedn (Cincinnati, OH: South-Western).

Chen, H.-C., Holton, E. F. III and Bates, R. (2005), ‘Development and validation of the learningtransfer system inventory in Taiwan’, Human Resource Development Quarterly, 16, 55–84.

Chiaburu, D. S. and Tekleab, A. G. (2005), ‘Individual and contextual influences on multipledimensions of training effectiveness’, Journal of European Industrial Training, 29, 604–26.

Cohen, J. and Cohen, P. (1983), Applied Multiple Regression/Correlation Analysis for the BehavioralSciences, 2nd edn (Hillsdale, NJ: Lawrence Erlbaum).

Colquitt, J. A., LePine, J. A. and Noe, R. A. (2000), ‘Toward an integrative theory of trainingmotivation: a meta-analytic path analysis of 20 years of research’, Journal of Applied Psychology,85, 678–707.

Dolezalek, H. (2004), ‘TRAINING annual industry report’, TRAINING Magazine, 42, 10. Availableat www.trainingmag.com (accessed 27 July 2005).

Dowling, P. J. and Welch, D. E. (2005), International Human Resource Management: Managing Peoplein a Multinational Context, 4th edn (Mason, OH: Thomson South-Western).

Facteau, J. D., Dobbins, G. H., Russell, J. E. A., Ladd, R. T. and Kudisch, J. D. (1995), ‘The influenceof general perceptions of the training environment on pretraining motivation and perceivedtraining transfer’, Journal of Management, 21, 1–25.

Ford, J. K., Smith, E. M., Weissbein, D. A., Gully, S. M. and Salas, E. (1998), ‘Relationships of goalorientation, metacognitive activity, and practice strategies with learning outcomes and trans-fer’, Journal of Applied Psychology, 83, 218–33.

Gist, M. E., Bavetta, A. G. and Stevens, C. K. (1990), ‘Transfer training method: Its influence onskill generalization, skill repetition, and performance level’, Personnel Psychology, 43, 501–23.

Gist, M. E., Stevens, C. K. and Bavetta, A. G. (1991), ‘Effects of self-efficacy and post-trainingintervention on the acquisition and maintenance of complex interpersonal skills’, PersonnelPsychology, 44, 837–61.


http://www.trainingmag.com

Holton, E. F. III (1996), ‘The flawed four-level evaluation model’, Human Resource Quarterly, 7, 5–21.Holton, E. F. III (2005), ‘Holton’s evaluation model: new evidence and construct elaborations’,

Advances in Developing Human Resources, 7, 37–54.Holton, E. F. III, Bates, R. A, Seyler, D. L. and Carvalho, M. B. (1997), ‘Toward construct validation

of a transfer climate instrument’, Human Resource Development Quarterly, 8, 95–113.Holton, E. F. III, Bates, R. A. and Ruona, W. E. A. (2000), ‘Development of a generalized learning

transfer system inventory’, Human Resource Development Quarterly, 11, 333–60.Kavanagh, M. J. (1998), ‘Transfer of Training: A Multi-Stage Model Designed for Practical Use by

Organizations’, in C. Scholz and J. Zentes (eds), Strategisches Euro-Management (Stuttgart:Schaffer-Poeschel), pp. 301–21.

Khasawneh, S., Bates, R. and Holton, E. F. III (2006), ‘Construct validation of an Arabic version ofthe learning transfer system inventory for use in Jordan’, International Journal of Training andDevelopment, 10, 180–94.

Lance, C. E., Kavanagh, M. J. and Brink, K. E. (2002), ‘Retraining climate as a predictor of retrainingsuccess and as a moderator of the relationship between cross-job retraining time estimates andtime to proficiency in the new job’, Group and Organization Management, 27, 294–317.

Mathieu, J. E., Tannenbaum, S. I. and Salas, E. (1992), ‘Influences of individual and situationalcharacteristics on measures of training effectiveness’, Academy of Management Journal, 35, 828–47.

Mathieu, J. E., Martineau, J. W. and Tannenbaum, S. I. (1993), ‘Individual and situational influ-ences on the development of self-efficacy: implications for training effectiveness’, PersonnelPsychology, 46, 125–47.

Montesino, M. (2002), ‘Strategic alignment of training, transfer-enhancing behaviors, and trainingusage: a post-training study’, Human Resource Development Quarterly, 13, 89–108.

Noe, R. A. (1986), ‘Trainee’s attributes and attitudes: neglected influences on training effective-ness’, Academy of Management Review, 11, 736–49.

Noe, R. A., Hollenbeck, J. R., Gerhart, B. and Wright, P. M. (2006), Human Resource Management:Gaining a Competitive Advantage, 6th edn (Boston, MA: McGraw-Hill Irwin).

Podsakoff, P. M., MacKenzie, S. B., Lee, J. and Podsakoff, N. P. (2003), ‘Common method biases inbehavioral research: a critical review of the literature and recommended remedies’, Journal ofApplied Psychology, 88, 879–903.

Quinones, M. A. (1995), ‘Pretraining context effects: training assignment as feedback’, Journal ofApplied Psychology, 80, 226–38.

Reber, R. A. and Wallin, J. A. (1984), ‘The effects of training, goal setting, and knowledge of resultson safe behaviour: a component analysis’, Academy of Management Journal, 27, 544–60.

Rouiller, J. Z. and Goldstein, I. L. (1993), ‘The relationship between organizational transfer climateand positive transfer of training’, Human Resource Development Quarterly, 4, 377–90.

Russell, J. S., Terborg, J. R. and Powers, M. L. (1985), ‘Organizational performance and organi-zational level training and support’, Personnel Psychology, 38, 849–63.

Tannenbaum, S. I. and Yukl, G. (1992), ‘Training and development in work organizations’, AnnualReview of Psychology, 43, 399–441.

Tannenbaum, S. I., Mathieu, J. E., Salas, E. and Cannon-Bowers, J. A. (1991), ‘Meeting trainees’expectations: the influence of training fulfillment on the development of commitment, self-efficacy, and motivation’, Journal of Applied Psychology, 76, 759–69.

Tesluk, P. E., Farr, J. L., Mathieu, J. E. and Vance, R. J. (1995), ‘Generalization of employeeinvolvement training to the job setting: individual and situational effects’, Personnel Psychology,48, 607–32.

Tracey, J. B. and Tews, M. J. (2005), ‘Construct validity of a general training climate scale’,Organizational Research Methods, 8, 353–74.

Tracey, J. B., Tannenbaum, S. I. and Kavanagh, M. J. (1995), ‘Applying trained skills on the job: theimportance of the work environment’, Journal of Applied Psychology, 80, 239–52.

Tziner, A., Haccoun, R. R. and Kadish, A. (1991), ‘Personal and situational characteristics influ-encing the effectiveness of transfer of training improvement strategies’, Journal of OccupationalPsychology, 64, 167–77.

Velada, R. and Caetano, A. (2007), ‘Training transfer: the mediating role of perception of learning’,Journal of European Industrial Training, 31, 283–96.

Weiss, H. M. (1990), ‘Learning Theory and Industrial and Organizational Psychology’, in M. D.Dunnette and L. M. Hough (eds), Handbook of Industrial and Organizational Psychology, Vol 1(Palo Alto, CA: Consulting Psychologists Press), pp. 171–221.

Wexley, K. N. and Latham, G. P. (2002), Developing and Training Human Resources in Organizations(Englewood Cliffs, NJ: Prentice Hall).

Wexley, K. N. and Nemeroff, W. (1975), ‘Effectiveness of positive reinforcement and goal settingas methods of management development’, Journal of Applied Psychology, 64, 239–46.


IJTD_2007

Documents

Transcript of IJTD_2007