Blackwell Publishing Ltd.Oxford, UK and Malden, USAIJTDInternational Journal of Training and Development1360-3736Blackwell Publishing Ltd. 20062006 4252268Original ArticlesTransfer of learning from 10 management development programmesInternational Journal of Training and Development

International Journal of Training and Development 10:4 ISSN 1360-3736

Transfer of learning from management development programmes: testing the Holton model
Cyril Kirwan and David Birchall
Transfer of learning from management development programmes has been described as the effective and continuing application back at work of the knowledge and skills gained on those programmes. It is a very important issue for organizations today, given the large amounts of investment in these programmes and the small amounts of that investment that actually translates into an improved individual and organizational performance. This paper describes the testing of the Holton model of learning transfer and suggests amendments to the model where appropriate. A sample of participants on a management development programme within the Irish health service completed the Learning Transfer System Inventory based on the Holton model. The data obtained were analysed using multiple regression and partial correlation techniques. Results indicate that the model does adequately represent the effects of its factors, although a slightly revised model is presented. Important relationships between factors are identified, and the central role of the learners’ personal capacity for transfer (hitherto unreported in the literature) and their motivation to transfer in facilitating transfer is highlighted. In practical terms, the tested model can be used as a diagnostic tool by identifying individual, training design and work environment transfer issues in need of attention and by developing strate-

r Dr. Cyril Kirwan, consultant in management and organization development, Cyril Kirwan & Associates, Management and Organisation Development, 6 Maywood Park, Raheny, Dublin 5, Ireland. Email: ckirwan@indigo.ie. Prof. David Birchall, Director of the School of Management Knowledge and Learning and of Henley Learning Advisory Services, Henley Management College, Greenlands, Henley-on-Thames, Oxfordshire RG9 3AU, UK. Email: david.birchall@henleymc.ac.uk
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St., Malden, MA 02148, USA.

252 International Journal of Training and Development

gies to deal with them. It can also be used as a framework for the evaluation of training and development interventions, examining factors outside the traditional range of most training evaluation efforts and providing a more complete picture of the success or otherwise of that intervention.

Introduction
In today’s rapidly changing business environment, capable managers are crucial to organizations’ success in gaining and sustaining a competitive advantage. This must be achieved against a background of intensified competition, incessant change, a power shift to the customer, collaborations across organizational and geographical boundaries, and a need to maintain high levels of talent (Tannenbaum, 2002). Each year, billions of dollars are spent in developing management capability through management development programmes. For example, Sugrue (2003) reports that on average, US employers’ organizations spend more than 2.2 per cent of payroll on education and training interventions. However, evidence concerning the value that organizations derive from what they spend on training and development interventions is not encouraging. Some recent estimates range between 10 and 20 per cent (Curry et al., 1994; Kaufman, 2002) of skills and knowledge being fully applied by employees on the job following training. The development of management capability through management development programmes presents a special difficulty in these circumstances. Learning outcomes are often concerned with more than just imparting straightforward knowledge or simple skills as defined by behaviourists (Barrow, 1987; Bridges, 1993). They often deal with more complex skills (see for example Hinchcliffe, 2002, who argues for a view of skill that recognizes the importance of situational understanding and of personal agency). They may be used as a way of changing attitudes as well. These outcomes are not easy to translate into an improved on-the-job performance. In addition, a difficulty inhibiting research in this area is the lack of suitable and valid models with which to investigate the relationships among factors known to affect learning transfer, as well as the relationships between these factors and transfer itself. A number of models have been developed, although they possess shortcomings both in terms of their completeness and in their usability in management training contexts. These will be described. A further model, that of Holton (1996), which addresses many of these shortcomings, will be put forward. The purpose of this paper is to: 1. 2. report on the testing of the Holton model of learning transfer suggest a development of the model if, and where, appropriate.

The model, which has been derived from the literature, will be tested using data from participants on a management development programme conducted within the health service in Ireland. Testing will comprise investigating whether or not the factors included in the model adequately reflect those purported in the literature to affect learning transfer, and also whether or not the predicted relationships between those factors exist in practice. Evidence will be provided from a quantitative survey of 72 participants on the programme. First, however, a brief description of the nature of learning transfer will be presented.

The nature of learning transfer
Transfer, in the context of training and development activity, is discussed in the literature in terms of the generalization of material learned in training (e.g. skills acquired, knowledge gained) to the job context as well as the maintenance of the learned material over a period of time on the job (Baldwin & Ford, 1988). Specifically, Broad and Newstrom (1992, p. 5) define it as: Transfer of learning from management development programmes 253
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Trainee characteristics

Training design

Learning and retention

Generalization and maintenance

Work environment

Figure 1: Factors affecting transfer of training (Baldwin & Ford, 1988).

The effective and continuing application by trainees to their jobs of the knowledge and skills gained in training – both on and off the job.

Research undertaken in a variety of settings has identified a large number of factors that affect the transfer of learning back to the workplace. Baldwin and Ford’s (1988) review is still cited extensively and is a good starting point for an investigation of the relevant studies. Baldwin and Ford (1988) propose a linking of the characteristics as shown in Figure 1. A detailed discussion of these factors is beyond the scope of this paper. Instead, those considered to be of greatest importance are discussed briefly in the succeeding paragraphs. Factors that affect transfer of learning fall into three broad categories – trainee factors, training design factors and work environment factors. The literature concerning characteristics of the trainee that can affect learning transfer focuses a lot of attention on their effect on a trainee’s motivation to learn and to transfer learning. It is clear that motivation plays a major part. In this regard, the trainees’ self-efficacy (Stevens & Gist, 1997; Warr et al., 1999) and their expectancies (Hicks & Klimowski, 1987; Magjuka et al., 1994) have been seen to have an important role to play. Different demographic characteristics (Warr & Bunce, 1995) as well as organizational commitment (Tesluk et al., 1995) and job involvement (Noe & Schmitt, 1986) are reported to exert varying effects. Where they have been studied, personality factors such as locus of control (Baumgartel et al., 1984) and goal orientation (Chiaburu & Marinova, 2005; Fisher & Ford, 1998) have also demonstrated effects. Finally, while cognitive ability appears to have some limited effect on transfer, some relevant cognitive style differences have been identified (Armstrong, 2000). Similarly, research has demonstrated that a wide variety of training design factors affect learning transfer. According to the evidence, training design strategies such as the use by trainers of general principles (Clark, 1990; Schunk, 1996) and of variable training stimuli (Baldwin, 1992; Holladay & Quinones, 2003) enhance transfer. For some types of skills, the presence of identical elements in the training and transfer settings also enhances transfer (Butterfield & Nelson, 1989). Tried and tested conditions of practice such as over-learning (May & Kahnweiler, 2000) and the use of distributed rather than massed learning (Enos et al., 2003) still affect transfer in a positive way. Some other training design approaches have been found to have implications for transfer. The use of an integrative learning strategy, although it has been shown to positively affect satisfaction with training (Bretz & Thompsett, 1992), does not necessarily improve transfer. However, the use of analogical models, based on the limited evidence available, is potentially a very useful strategy in facilitating transfer of learning (Thompson et al., 2000). Finally, the development of adaptive expertise has been shown to be of particular importance for management development programmes (Enos et al., 2003; Mainemelis et al., 2002). Building more complex knowledge structures and encouraging the use of metacognition (Smith et al., 1997) may have more lasting effects on learning, retention and generalization of the complex knowledge and skills inherent in management behaviour. Following training and development programmes, the evidence certainly suggests that goal setting, and interventions that include goal setting such as coaching (Brown 254 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

& Latham, 2000; Smither et al., 2003; Thach, 2002), are effective strategies in facilitating transfer. In these situations, the key to transfer appears to be around making conscious decisions about how the training will be used, and anticipating difficulties and generating strategies to cope with them (Burke & Baldwin, 1999; Richman-Hirsch, 2001) in order to transfer one’s learning. Finally, according to the research evidence, there is little doubt concerning the significance of work environment factors in either inhibiting or facilitating transfer of learning from training programmes. The most important factors concern the creation of an organization climate conducive to transfer (Holton et al., 1997a; Lim & Johnson, 2002), in which supervisory support (Cromwell & Kolb, 2004; Gumuseli & Ergin, 2002) and peer support (Chiaburu & Marinova, 2005) play a major part. The climate thus created affects the trainee’s ability or motivation to transfer.

Models of learning transfer
As seen previously, a wide variety of factors affect learning transfer, and a number of models have been put forward in the literature to explain their many and varied influences. A detailed discussion of these models is beyond the scope of this paper. However, the main contributions are summarized in Table 1, along with their key strengths and weaknesses. One other model that shares the characteristics of those already outlined and which has also been derived by examining relationships and constructs from previous empirical research is that of Holton (1996), shown in Figure 2. The model hypothesizes that HRD outcomes are primarily a function of ability, motivation and environmental influences at three outcomes levels – learning, individual performance and organizational performance. It also considers a role for some secondary influences. First proposed as a model for training evaluation but based on work in the area of learning transfer, it has become the conceptual basis for the Learning Transfer System Inventory (LTSI) (Holton et al., 2000), which measures factors relating to the trainee, training design and the work environment in terms of their effect on transfer. The full model

Secondary influences

Intervention readiness

Intervention fulfilment

Personality characteristics

Job attitudes

Motivation elements

Motivation to learn

Motivation to transfer

Expected utility/ROI

Environmental elements

Reaction

Transfer climate

External events

Outcomes

Learning

Individual performance

Organizational results

Ability/enabling elements

Ability

Transfer design

Linkage to organizational goals

Figure 2: HRD evaluation research and measurement model (Holton, 1996). Transfer of learning from management development programmes 255
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Table 1: Models of learning transfer Core characteristics Simplicity Practical focus Strengths Weaknesses

Model

Baldwin and Ford (1988)

Overview of transfer process

Broad and Newstrom (1992)

Kozlowski and Salas (1997)

Identifies contribution to transfer of trainee, trainer and manager Indicates transfer effects at individual, team and organizational level

Generality No indication of interactions Explains little or nothing of the transfer process Lacks specificity regarding transfer factors

256 International Journal of Training and Development Recognizes complexity of transfer process and relationships between levels Integrates multi-level approach and relationships between inputs and outcomes Identifies factors at organizational level that influence outcomes at individual level Identifies interactions among transfer factors; comprehensive coverage of trainee factors

Machin (2000)

© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Thayer and Teachout (1995)

Indicates transfer effects at different levels and links transfer outcomes to training outcomes Includes a range of factors discussed in the literature

Colquitt et al. (2000)

A meta-analytic study of (mainly) trainee factors

Developed in/for very specific (aviation team) settings No indication of relative strength of factors or interactions between them Does not include training design factors

looks at wider issues than just transfer of learning and is included in its entirety here. Thick arrows represent proposed primary relationships between variables whereas lighter arrows represent secondary relationships. Central to this model is the effect of different motivational elements on outcomes. Motivation to learn, for instance, is proposed as a direct precursor of learning. Similarly, motivation to transfer is expected to have a direct effect on individual performance. In addition, the expectancies concerning the utility of the training should affect organizational results. Holton (1996) also suggests that other factors will have a direct impact on learning, performance and results. Trainee reactions to the training directly affect learning, transfer climate directly affects individual performance and external events directly affect organizational results. Finally, the three primary outcomes are also affected by enabling elements such as ability, which is seen to affect learning, transfer design, which affects individual performance, and linkage of training to organizational goals, which affects organizational results. Although the model is put forward as a model for training evaluation, it is nevertheless important in a discussion on learning transfer as all of the primary outcomes are central components of any transfer model. Furthermore, with the exception of trainee reactions and external events, all the other factors in this model have been discussed in the transfer literature. The construction of the model is based on Noe (1986), who proposes that an individual’s behaviour in training is a function of ability, motivation and work environment factors. These are the primary intervening variables in this model. Another important element that this model adds to the discussion is that both primary and secondary influences on outcomes are included. Holton (1996) admits that his model only describes a sequence of influences on outcomes occurring in a single learning experience and does not demonstrate any feedback loops. For example, although he suggests that success in learning would enhance future motivation, he does not include this possibility in the model. Another aspect missing from the model (and indeed all the previous models discussed) is any indication of interaction between factors of the same type, e.g. organizational factors. For example, there is no indication of whether supervisor support, peer support and resistance to change are separate or related factors. Other factors discussed in the literature such as characteristics of the trainee’s job, personality characteristics or training design characteristics are only mentioned in a general way, with no mention of their components, although this is presumably in the interests of not overcomplicating the model. Finally, the model also lacks any mention of how the factors affect transfer. The relationships indicated are more of a guide to the influences that exist within the model rather than a real examination of the effects of these factors. However, of the models discussed, the Holton model reflects more fully than the others the discussion in the literature concerning different factors that affect transfer. It is the only one, for example, that includes a specific mention of motivation to transfer, a factor cited in the literature as being very important. Furthermore, it is the only model that suggests a role for training design factors, which, according to the literature, also exert significant effects. Finally, an important additional advantage in using this model is that its authors have developed the LTSI based on the model – and including some more detail on individual, transfer climate and motivation factors in the model – with which to measure in a quantitative way the impact of these various factors. Both its validity and reliability are reasonable. The constructs discussed within the model and examined through the instrument have all been developed from the literature. Using the model, the LTSI can be used to generate data either alone or in conjunction with qualitative methods. The model can be tested to see if the data support it, and if not, what alternatives might be proposed.

Research process
In the study on which this paper is based, a total of 112 nurse managers from within the health service in Ireland took part in one of four separate but similar management development programmes, all delivered by the same provider. The programmes were Transfer of learning from management development programmes 257
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

of 12 days’ duration and took place over a period of 6 months. The research process involved the collection of quantitative data concerning participants’ perceptions of a variety of factors (the independent variables) likely to affect their transfer of learning (the dependent variable) from the programme they attended. Overall, of the 112 questionnaires distributed, 72 responses were received. This represents a return rate of 64 per cent. The instrument used for this purpose was the LTSI version II, developed by Holton and colleagues (1997a, 1997b, 2000), and based on the Holton (1996) model being tested. All of the scales were derived from the literature and were developed to assess factors affecting trainees’ ability to transfer learning, their motivation to transfer and transfer environment factors. Definitions and sample items for some of the scales are given in Table 2. Respondents rated each of the questions on a Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree). In terms of construct validity, a significant factor analysis had already been carried out on the instrument, resulting in a 16-factor structure. Cronbach’s alpha reliabilities range from 0.63 to 0.91, with only three of the scales below 0.70. Before undertaking an analysis of the responses, reliability of the scales for this study was tested using the data provided by the sample. Cronbach’s alpha was obtained for each of the 16 factors and is presented in Table 3. With the exception of two scales (negative personal outcomes, α = 0.68; personal capacity for transfer, α = 0.62), the reliability coefficients for the 16 factors are all above α = 0.70. According to Hair et al. (1998), a coefficient of α = 0.70 is widely acceptable. They also suggest that coefficients as low as α = 0.60 are acceptable for exploratory research. For the sample of 72 respondents to the LTSI, correlations between the different constructs were computed. First of all, simple bivariate correlations were sought to determine what factors related most strongly to each other. Further investigation of these relationships was undertaken using partial correlations or linear regression in order to determine their strength and direction. The main strength of using this method to test the transfer model is that the factors in the instrument and the model are similar, enabling a direct comparison. Also, as there is evidence for interrelationships between factors in the model, using partial correlation or regression analysis should help examine this.

Results
LTSI responses to the 16 transfer factors were analysed for the 72 participants who responded. A histogram for each of the variables was examined, skewness and kurtosis values were computed, and finally, a Kolmogorov-Smirnov one-sample test was conducted to determine if the distribution differed significantly from a normal distribution. Values obtained revealed that responses to four factors (perceived content validity, transfer design, effort-performance expectations and positive personal outcomes) were not normally distributed. Based on the LTSI responses, the significant correlations among the 16 transfer factors for that group are shown in Table 4. Only correlations that are significant to at least p = 0.05 are included (*p < 0.05; **p < 0.01). Given also the relatively small size of the sample and its consequences for a normal distribution, Spearman’s rho rather than Pearson’s coefficient was used. Although the relationships between the factors will be discussed fully later, some initial observations on the results in Table 4 can be made. One is the central role of motivation to transfer in the overall scheme. More factors correlated with it more significantly than with any other factor. This contention is reinforced in the literature and in models presented earlier. Another observation is the fact that, in terms of work environment factors, peer support and feedback and coaching correlate with many of the same factors – performance self-efficacy, motivation to transfer and manager support. This may suggest a link between these factors and lend support to the concept of a transfer climate. Third, as well as correlating highly with each other, the factors motivation to transfer and performance self-efficacy also correlate with similar factors – learner readiness, 258 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Table 2: Learning Transfer System Inventory factors (Holton et al., 2000) Definition Sample item No. of items 4 4

Factor

Learner readiness

Motivation to transfer

Before the training, I had a good understanding of how it would fit my job-related development. I get excited when I think about trying to use my new learning in my job. My supervisor sets goals for me that encourage me to apply my training on the job.

Supervisor support

6

Perceived content validity

5 4

Transfer of learning from management development programmes 259

© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Feedback and coaching

The extent to which individuals are prepared to enter and participate in training The direction, intensity and persistence of effort toward utilizing in a work setting skills and knowledge learned The extent to which supervisors/managers support and reinforce use of training on the job The extent to which trainees judge training content to accurately reflect job requirements Formal and informal indicators from an organization about an individual’s job performance

What is taught in training closely matches my job requirements. After training, I get feedback from people about how well I am applying what I learned.

Table 3: Alpha coefficients for the Learning Transfer System Inventory scales used in this study Factor Learner readiness Motivation to transfer Positive personal outcomes Negative personal outcomes Personal capacity for transfer Peer support Manager support Manager sanctions α 0.77 0.82 0.75 0.68 0.62 0.90 0.92 0.85 Factor Perceived content validity Transfer design Opportunity to use Effort-performance expectations Performance-outcome expectations Resistance to change Performance self-efficacy Feedback and coaching α 0.81 0.81 0.74 0.83 0.70 0.84 0.79 0.78

transfer design, perceived content validity and opportunity to use. There is also some indication that these factors may work together. Examination of the model However, the fact that so many factors intercorrelate does not necessarily mean that they directly affect one another. What needs to be done is to determine where the most significant relationships exist before comparing it with the Holton model, as displayed in Figure 2. Using significant correlations between model factors as a starting point, both partial correlation and multiple linear regression techniques will be used. Three factors relating to participant motivation are included in the LTSI framework – motivation to transfer, effort-performance expectations and performance-outcome expectations. First, the three factors correlate highly and significantly with each other. Second, responses for the factor performance-outcome expectations were not normally distributed and were therefore a potential problem in analysis. For these reasons, and in the interests of model parsimony, the means of the three variables were summated (Hair et al., 1998), and the combined new variable called motivation to transfer was defined. Both skewness and kurtosis values for this new variable are well within the limits for normal distribution. The significant correlations present between the factors will now be examined more closely, starting with transfer design and perceived content validity. Application of tests for normal distribution of responses to each of the factors showed that among them, both transfer design and perceived content validity responses were negatively skewed. Thus, the use of partial correlations with these factors could cause problems in terms of predicting relationships with other correlated factors. However, LTSI responses indicated that both transfer design and perceived content validity were regarded very highly overall by participants. In the case of both variables, by far the most frequently occurring score was 4 (on a scale of 1–5). A review of the programme prospectus also pointed to a distinct synergy between the programme content (it was the outcome of an in-depth needs analysis and an evaluation of the previous pilot programme) and training design (on the basis of recommendations in the evaluation, a number of transfer-enhancing elements were incorporated). Both of these factors are highly correlated with performance self-efficacy. In terms of which factor has more effect, the correlation between transfer design and performance self-efficacy is significant (β = 0.27, p = 0.02) when perceived content validity is controlled for. Controlling for transfer design yields a not-so-significant correlation between the other two factors of β = 0.22, p = 0.06. This suggests that transfer design may be a more important influence on performance self-efficacy. Therefore, this model proposes that transfer design and perceived content validity separately and together influence participants’ performance self-efficacy, with transfer design having the greater effect. 260 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Table 4: Significant correlations among Learning Transfer System Inventory factors
Percd content validity Perf selfefficacy Oppty to use Pers cap’y for t’fer Motiv’n to transfer Effortperf expec’s Perfoutcome expec’s Res to chnge M’ger suppt M’ger sncts Peer supp F’back and c’ching Pos pers outc’s Neg pers outc’s

Learner readi’s

T’fer design

0.23* 0.40** 0.41** 0.24* −0.26* 0.34** 0.38** 0.44** 0.26* 0.29* −0.33** 0.58** −0.26* 0.38** 0.34** 0.32** 0.26* 0.37** 0.28* 0.28*

0.28* 0.33** 0.40**

0.34** 0.24*

0.53** 0.29* 0.26* 0.43** 0.27* 0.40**

0.29*

0.27*

0.32** 0.37** 0.36** 0.24* −0.30* 0.30*

0.25* 0.34** 0.43** 0.29*

−0.41**

0.30* 0.36** 0.37**

Learner readiness Transfer design Perceived content validity Performance self-efficacy Opportunity to use Personal capacity for transfer Motivation to transfer Effort-performance expectations Performance-outcome expectations Resistance to change Manager support Manager sanctions Peer support Feedback and coaching Positive personal outcomes Negative personal outcomes

Transfer of learning from management development programmes 261

© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

* p < 0.05. ** p < 0.05.

Table 5: Relationships among significant personal factors Dependent variable Motivation to transfer Personal capacity for transfer Regression model (stepwise) Personal capacity for transfer Learner readiness Performance self-efficacy Motivation to transfer β 0.32 0.25 0.24 0.40 p 0.00 0.02 0.03 0.00

As reported in the literature, many transfer factors affect motivation to transfer. However, among the correlations previously reported, the strongest relationship (r = 0.53, p < 0.01) was that between motivation to transfer and personal capacity for transfer. This raises a question as to the relationship between them, and in particular its direction. Either factor could potentially be a dependent variable in relationships in the model. Therefore, linear regression models to be examined here will include that possibility. Thus, a linear regression model to include personal capacity for transfer, motivation to transfer, and the other individual factors performance self-efficacy and learner readiness was constructed. Effects of the factors on motivation to transfer and on personal capacity for transfer were examined in turn. Taking motivation to transfer as the dependent variable, a stepwise regression results in a primary effect of personal capacity for transfer (β = 0.40, p = 0.00). All the factors exerted significant effects, with partial correlation values in the final model of personal capacity for transfer (β = 0.32, p = 0.00), learner readiness (β = 0.25, p = 0.02) and performance self-efficacy (β = 0.24, p = 0.03). However, when personal capacity for transfer is the dependent variable, the only factor that makes a significant contribution to the equation is motivation to transfer (β = 0.40, p = 0.00). These relationships are summarized in Table 5. What this suggests is that although there may be a reciprocal relationship between motivation to transfer and personal capacity for transfer, the factors learner readiness and performance self-efficacy exert their effects on motivation to transfer only. This concurs with studies reported in the literature concerning the role of motivation to transfer. It also lends credibility to the notion that the learner’s motivation to transfer is an antecedent rather than an outcome of personal capacity for transfer. The relationships in the model examined so far indicate the importance of a number of personal and training programme-related factors that have a significant effect on a participant’s motivation to transfer learning. Transfer design and perceived content validity work together to improve performance self-efficacy and to influence motivation to transfer in this way. Independently of this, learner readiness also appears to affect motivation to transfer directly. It is also proposed at this stage that motivation to transfer influences the learner’s personal capacity for transfer. The next set of relationships to be examined includes those relating to participants’ work environments. First of all, a linear regression model was constructed for all the work environment factors that significantly correlate with motivation to transfer or with personal capacity for transfer – opportunity to use, manager support, manager sanctions, peer support, feedback and coaching and resistance to change. Taking motivation to transfer as the dependent variable first, a stepwise linear regression demonstrated that peer support contributes most significantly to motivation to transfer (β = 0.55, p = 0.00). Independently, opportunity to use also exerts a significant effect (β = 0.39, p = 0.00), whereas the beta value for peer support remains at β = 0.55. The only other variable to have a significant effect (β = 0.27, p = 0.00) was feedback and coaching. The fact that in this third regression model the value for peer support fell to β = 0.44, p = 0.00 whereas opportunity to use remained almost constant at β = 0.38, p = 0.00 suggests a link between feedback and coaching and peer support. If personal capacity for transfer is made the dependent variable, opportunity to use has the strongest effect (β = 0.40, p = 0.00). The only other factor to add significantly to the 262 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Table 6: Relationships with personal capacity for transfer and motivation to transfer Dependent variable Motivation to transfer Personal capacity for transfer Regression model (stepwise) Peer support Opportunity to use Feedback and coaching Opportunity to use Manager support β 0.44 0.38 0.27 0.35 0.22 p 0.00 0.00 0.00 0.00 0.04

Table 7: Partial correlations among transfer-inhibiting factors Factor Manager support Manager support Manager sanctions Correlated with Manager sanctions Resistance to change Resistance to change Controlling for Resistance to change Manager sanctions Manager support β −0.21 −0.23 0.27 p 0.09 0.04 0.02

model is manager support. This effect appears to be largely independent as the value for opportunity to use changes to β = 0.35, p = 0.00 whereas that for manager support is β = 0.22, p = 0.04. These relationships are summarized in Table 6. The differing effects of the factors personal capacity for transfer and motivation to transfer in the models in Table 6 provides further evidence of the direction of the relationship between them. If personal capacity for transfer is added to the first model, it accounts for only 5 per cent of the variance in terms of effects on motivation to transfer. By contrast, in the second model, motivation to transfer is shown to have the strongest effect on personal capacity for transfer, accounting for 17 per cent of the variance. Thus, the proposal to consider motivation to transfer as an antecedent of personal capacity for transfer is further supported. In terms of work environment factors, therefore, the effects of opportunity to use, peer support and feedback and coaching on motivation to transfer, and the effects of opportunity to use and manager support on personal capacity for transfer, are the important relationships. There are some remaining relationships among work environment factors to be examined – the transfer-inhibiting factors resistance to change and manager sanctions. These factors did not directly affect either variable, so it is likely that any effects they exert are indirect. In the original correlation matrix, both manager sanctions and resistance to change showed significant relationships with the factor manager support. These relationships will be explored using partial correlations. Manager support and manager sanctions are not significantly related when the effects of resistance to change are removed (β = –0.21, p = 0.09). However, the relationship of resistance to change with both manager support (β = –0.23, p = 0.04) and manager sanctions (β = 0.27, p = 0.02) is significant. What these data suggest is that the amount of resistance to change in the work environment affects the level of manager support provided. Resistance to change is also related to the level of manager sanctions, although it is not yet clear how exactly this manifests itself. What is likely is that both factors are two aspects of an unsupportive climate, and they work together. These relationships are summarized in Table 7. The two remaining factors included in the LTSI framework are positive personal outcomes and negative personal outcomes. Within the framework, the two factors correlated with each other, and positive personal outcomes correlated positively with manager sanctions. Ratings on both these scales were low. Transfer of learning from management development programmes 263
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Comparison with the Holton model The main similarities and differences between this and the Holton model will now be discussed. First of all, the main relationships predicted among factors in that model exist also in the newly constructed one. Although the labels are different (Holton, 1996), the constructs they describe are the same. Primarily, the central role of motivation to transfer is emphasized. In addition, the links between motivation to transfer and both learner readiness and performance self-efficacy stand up. Also implied in the Holton model is the importance of perceived content validity and of transfer design. In addition, transfer climate factors (mainly manager support and resistance to change) have a stronger relationship with personal capacity for transfer (ability to transfer) than they have with motivation to transfer, (desire to transfer) as predicted. On the other hand, the Holton model does not suggest a direct link between transfer design or perceived content validity and motivation to transfer, a relationship found with this sample. Finally, the construct personal capacity for transfer, although a factor in the LTSI, is not named as such in the Holton model but is rather implied as a personal factor related to aspects of transfer climate. It is suggested that the role of this construct deserves greater recognition. The question of whether or not the Holton model could be used as a basis for a model of learning transfer was posed earlier. The evidence presented on the basis of significant correlations among the factors would appear to broadly support it. Therefore, based on a closer investigation of the relationships among factors in the model, a revised version of the model is presented in Figure 3.

Conclusions and implications
The construction of a model of learning transfer from the significant relationships among its constituent factors has resulted in a model that broadly supports that of Holton (1996). Thus, conditions for learning transfer are optimized in the following way. To begin with, the practical design of the management development programme, demonstrating in a practical way how the training can be best used on the job, combines with a valid, relevant content. Offered to a participant who is ready in the sense that he/she sees the programme as an opportunity to learn and develop his/ her motivation to transfer, i.e. desire to apply learning, is activated. Motivation to transfer is also enabled through a higher level of self-efficacy (confidence that she/he can learn and apply the learning in practice), which an appropriate transfer design will reinforce. In the work environment, a number of other factors exert effects. The amount of support received, particularly from a participant’s peers, as well as the amount of feedback and coaching received from any source independently and positively affects the motivation to transfer. In addition, the support received in various ways from participants’ managers enables them to ‘make the space’ to transfer learning. On top of that, organizational support in the form of opportunities to apply that learning exerts effects on their desire to transfer as well as on their ability to do so. On the negative side, the existence of manager sanctions and a resistance to change (both hindrances to the application of learning) will inhibit the motivation to transfer, although this effect will be moderated by the amount of manager support available. The constructed model also reinforces the combined role of motivation to transfer and personal capacity for transfer in enabling transfer to take place. Many of the factors discussed independently affect the former, which in turn directly affects the latter. These effects were discussed in the literature but were not made explicit in the Holton model. For example, the relationship between manager support and personal capacity for transfer (rather than motivation to transfer) reflects a more specific relationship with trainees’ ability to transfer than their desire to do so. The constructed model also goes further than the Holton model in revealing a greater level of complexity within the transfer system. Relationships between 264 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Learner readiness Training seen as an opportunity To learn and develop

Transfer design Shows in a practical way how training can be best used on the job

Perceived content validity Programme content perceived as relevant to job

Performance self-efficacy Confidence from experience, feedback, etc. builds up in individual regarding ability to learn/transfer

MANAGEMENT DEVELOPMENT PROGRAMME
Feedback and coaching Amount of constructive feedback from any source Motivation to transfer Desire to apply learning Manager support Amount of help from boss in applying learning Resistance to change Reluctance in work environment to try new ways of doing things

Peer support Amount of help in applying new learning from peers

Opportunity to use Organization factors enable or inhibit use of training Personal capacity for transfer Personal factors enable or inhibit use of training Outcomes from non-application of training Negative personal outcomes Manager sanctions Amount of hindrance from boss in applying learning

Outcomes from application of training Positive personal outcomes

TRANSFER

Transfer of learning from management development programmes 265 Figure 3: Model of learning transfer.

© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

factors, particularly work environment factors, which were also suggested in the literature but were not made explicit in the framework, have been demonstrated. Among these is the effect of the workplace climate factor resistance to change on both manager support and manager sanctions, which in turn affect trainees’ ability to transfer (personal capacity for transfer). Likewise, although opportunity to use has some effects on motivation to transfer, it has stronger effects on personal capacity for transfer. Whether the model predicts learning transfer or not is beyond the scope of this paper. Learning transfer from a programme of this nature can be complex and is subject to a range of individual, training design and organizational influences, many of which could not be controlled in this study. Nevertheless, the primary importance of the combined effects of motivation to transfer and personal capacity for transfer has been underlined. One particular limitation with this study was that given the restrictions imposed in terms of numbers, the sample of participants involved (N = 72) was small for quantitative work. A larger sample would have enabled a greater sophistication in the statistical analysis, for example, allowing the use of structural equation modelling to examine causal relationships (Hair et al., 1998). However, no similar programmes for the same target population (i.e. nurse managers) were being planned at that time, and it was considered inappropriate to use data from other development programmes. Its limitations notwithstanding, this study contributes a refinement to the original Holton (1996) model. Although that model included the main factors relevant to learning transfer, it also focused on factors not directly related to transfer. On the other hand, the conceptual framework for the LTSI, which was derived from the model and includes all the factors that were measured in this study, gives no indication of interaction between the factors. What the model constructed as a result of this study provides is a set of relationships that demonstrates the interactions between the most important factors. The demonstration of a central role for personal capacity for transfer is also new. Personal capacity for transfer has not been labelled as such in the literature, although its components such as self-management (Richman-Hirsch, 2001), dealing with situational constraints (Olivero et al., 1997) and developing strategies to overcome them have. This model suggests that its role is critical, it being the factor that enables motivation to transfer (itself a critical factor) to exert its effect. In addition, the combined role of these factors as the constituent components of trainability (Noe, 1986) is emphasized. Previous research, although focusing on the link between motivation to transfer and actual transfer (Colquitt et al., 2000), has not necessarily investigated how one translates into the other. In practical terms, the model described previously can be used as a diagnostic tool by identifying individual, training design and work environment transfer issues in need of attention and by developing strategies to deal with them. It can also be used as a framework for the evaluation of training and development interventions, examining factors outside the traditional range of most training evaluation efforts and providing a more complete picture of the success or otherwise of that intervention. It is clear from the literature and from this research that a range of factors affect the transfer of learning from management development programmes. It is hoped that the previously presented model can help in a practical way by providing a focus and a consistent measuring system for those factors. Through proper diagnosis and focused interventions, they can be manipulated in order to increase the amount of transfer. By increasing the amount of transfer, organizations can improve the rate of return on resources spent on training and development from what is still a very low base (recall that estimates have typically been in the 10–20 per cent range). To achieve this, they need to consider transfer in a structured way, working to increase the effect of the facilitating factors and to reduce the effect of the inhibiting factors. Only in this way will investment in the management capability necessary to deliver competitive advantage be realized. 266 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

References
Armstrong, S. J. (2000), ‘The influence of individual cognitive style on performance in management education’, Educational Psychology, 20, 323–39. Baldwin, T. T. (1992), ‘Effects of alternative modeling strategies on outcomes of interpersonal training’, Journal of Applied Psychology, 77, 147–54. Baldwin, T. T. and Ford, J. K. (1988), ‘Transfer of training: a review and directions for future research’, Personnel Psychology, 41, 63–105. Barrow, R. (1987), ‘Skill talk’, Journal of the Philosophy of Education, 21, 187–201. Baumgartel, H., Reynolds, M. and Pathan, R. (1984), ‘How personality and organization-climate variables moderate the effectiveness of management development programmes: a review and some recent research findings’, Management and Labour Studies, 9, 1–16. Bretz, R. D. and Thompsett, R. E. (1992), ‘Comparing traditional and integrative learning methods in organizational training programs’, Journal of Applied Psychology, 77, 6, 941–51. Bridges, D. (1993), ‘Transferable skills: a philosophical perspective’, Studies in Higher Education, 18, 43–51. Broad, M. L. and Newstrom, J. (1992), Transfer of Training: Action-Packed Strategies to Ensure High Payoff from Training Investments (New York: Addison Wesley). Brown, T. C. and Latham, G. P. (2000), ‘The effects of goal setting and self-instruction training on the performance of unionized employees’, Relations Industrielles, 55, 1, 80–95. Burke, L. A. and Baldwin, T. T. (1999), ‘Workforce training transfer: a study of the effects of relapse prevention training and transfer climate’, Human Resource Management, 38, 3, 227–41. Butterfield, E. C. and Nelson, G. D. (1989), ‘Theory and practice of teaching for transfer’, Educational Technology Research and Development, 37, 5–38. Chiaburu, D. S. and Marinova, S. V. (2005), ‘What predicts skill transfer? An exploratory study of goal orientation, training, self-efficacy and organizational supports’, International Journal of Training and Development, 9, 2, 110–23. Clark, R. E. (1990), ‘A cognitive theory of instructional method’, paper presented at the Annual Meeting of the American Educational Research Association (16–20 April), Boston, MA. Colquitt, J. A., LePine, J. R. and Noe, R. A. (2000), ‘Trainee attributes and attitudes revisited: a meta-analysis of research on training motivation’, Journal of Applied Psychology, 85, 5, 678–707. Cromwell, S. E. and Kolb, J. A. (2004), ‘An examination of work-environment support factors affecting transfer of supervisory skills training to the workplace’, Human Resource Development Quarterly, 15, 4, 449–71. Curry, D. H., Caplan, P. and Knuppel, J. (1994), ‘Transfer of training and adult learning (TOTAL)’, Journal of Continuing Social Work Education, 6, 1, 8–14. In Gregoire, T. K., Propp, J. and Poertner J. (1998), ‘The supervisor’s role in the transfer of training’, Administration in Social Work, 22, 1, 1–18. Enos, M. D., Kehrhahn, M. T. and Bell, A. (2003), ‘Informal learning and the transfer of learning: how managers develop proficiency’, Human Resource Development Quarterly, 14, 4, 369–88. Fisher, S. L. and Ford, J. K. (1998), ‘Differential effects of learner effort and goal orientation on two learning outcomes’, Personnel Psychology, 51, 397–420. Gumuseli, A. I. and Ergin, B. (2002), ‘The manager’s role in enhancing the transfer of training: a Turkish case study’, International Journal of Training and Development, 6, 2, 80–97. Hair, Jr., J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. (1998), Multivariate Data Analysis, 5th edn (Englewood Cliffs, NJ: Prentice-Hall). Hicks, W. D. and Klimowski, R. J. (1987), ‘Entry into training programs and its effects on training outcomes: a field experiment’, Academy of Management Journal, 30, 542–52. Hinchcliffe, G. (2002), ‘Situating skills’, Journal of the Philosophy of Education, 36, 187–205. Holladay, C. L. and Quinones, M. A. (2003), ‘Practice variability and transfer of training: the role of self-efficacy generality’, Journal of Applied Psychology, 88, 6, 1094–103. Holton, E. F. III (1996), ‘The flawed four level evaluation model’, Human Resource Development Quarterly, 7, 5–21. 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, 4, 333–60. Holton, E. F. III, Bates, R. A., Seyler, D. L. and Carvalho, M. B. (1997a), ‘Towards construct validation of a transfer climate instrument’, Human Resource Development Quarterly, 8, 95–113. Holton, E. F. III, Bates, R. A., Seyler, D. L. and Carvalho, M. B. (1997b), ‘Final word: reply to Newstrom and Tang’s reactions’, Human Resource Development Quarterly, 8, 2, 145–9. Kaufman, R. (2002), ‘Resolving the (often deserved) attacks on training’, Performances Impermanency, 41, 6. Kozlowski, S. W. J. and Salas, E. (1997), ‘An organizational systems approach for the implementation and transfer of training’, in K. Kraiger (ed.), Creating, Implementing, and Managing

Transfer of learning from management development programmes 267
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.

Effective Training and Development: State-of-the-Art Lessons for Practice (San Francisco: JosseyBass), pp. 117–59. Lim, D. H. and Johnson, S. D. (2002), ‘Trainees’ perceptions of factors that influence learning transfer’, International Journal of Training and Development, 6, 1, 36–48. Machin, M. A. (2000), ‘Enhancing the transfer of aviation team training’, paper presented at the Fifth Australian Aviation Psychology Symposium (November), Manly, Australia, in K. Kraiger (ed.), Creating, Implementing, and Managing Effective Training and Development: State-of-the-Art Lessons for Practice (San Francisco: Jossey-Bass). Magjuka, R., Baldwin, T. T. and Loher, B. T. (1994), ‘The combined effects of three pretraining strategies on motivation and performance: an empirical exploration’, Journal of Managerial Issues, 6, 3, 282–96. Mainemelis, C., Boyatzis, R. and Kolb, D. A. (2002), ‘Learning styles and adaptive flexibility: testing experiential learning theory’, Management Learning, 33, 1, 5–33. May, G. L. and Kahnweiler, W. M. (2000), ‘The effect of a mastery practice design on learning and transfer in behaviour modelling training’, Personnel Psychology, 53, 353–73. Noe, R. A. (1986), ‘Trainees’ attributes and attitudes: neglected influences on training effectiveness’, Academy of Management Review, 11, 736–49. Noe, R. A. and Schmitt, N. (1986), ‘The influence of trainee attitudes on training effectiveness: test of a model’, Personnel Psychology, 39, 497–523. Olivero, G., Bane, D. K. and Kopelman, R. E. (1997), ‘Executive coaching as a transfer of training tool: effects on productivity in a public agency’, Public Personnel Management, 26, 4, 461–69. Richman-Hirsch, W. L. (2001), ‘Post-training interventions to enhance transfer: the moderating effects of work environments’, Human Resource Development Quarterly, 12, 2, 105–20. Schunk, D. H. (1996). Learning Theories, 2nd edn (Englewood Cliffs, NJ: Prentice-Hall), in Yildirim, Z., Ozden, M. Y. and Aksu, M. (2001), ‘Comparison of hypermedia learning and traditional instruction on knowledge acquisition and retention’, Journal of Educational Research, 94, 4, 207–20. Smith, E. M., Ford, J. K. and Kozlowski, S. W. J. (1997), ‘Building adaptive expertise: implications for training design strategies’, in M. A. Quinones and A. Ehrenstein (eds), Training for a Rapidly Changing Workplace: Applications of Psychological Research (Washington, DC: American Psychological Association), 89–118. Smither, J. W., London, M., Flautt, R., Vargas, Y. and Kucine, I. (2003), ‘Can working with an executive coach improve multisource feedback ratings? A quasi-experimental field study’, Personnel Psychology, 56, 1, 23–44. Stevens, C. K. and Gist, M. E. (1997), ‘Effects of self-efficacy and goal-orientation training on interpersonal skill maintenance: what are the mechanisms?’ Personnel Psychology, 50, 4, 955–78. Sugrue, B. (2003), State of the Industry: ASTD’s Annual Review of US and International Trends in Workplace Learning and Performance (Alexandria, VA: ASTD). Tannenbaum, S. I. (2002), ‘A strategic view of organizational training and learning’, in K. Kraiger (ed.), Creating, Implementing, and Managing Effective Training and Development: State-of-the-Art Lessons for Practice (San Francisco: Jossey-Bass), pp. 10–52. Tesluk, P. E., Farr, J. L., Mathieu, J. E. and Vance, R. J. (1995), ‘Generalization of employee involvement training to the job setting: individual and situational effects’, Personnel Psychology, 48, 607–32. Thach, E. C. (2002), ‘The impact of executive coaching and 360° feedback on leadership effectiveness’, Leadership and Organization Development Journal, 23, 205–21. Thayer, P. W. and Teachout, M. S. (1995), ‘A climate for transfer model (AL/IIR-TP-1995-0035). Brooks Air Force Base, TX. Technical training research division, Armstrong Laboratory’, in K. Kraiger (ed.), Creating, Implementing, and Managing Effective Training and Development: State-ofthe-Art Lessons for Practice (San Francisco: Jossey-Bass), pp. 263–301. Thompson, L., Gentner, D. and Lowenstein, J. (2000), ‘Avoiding missed opportunities in managerial life: analogical training more powerful than individual case training’, Organizational Behavior and Human Decision Processes, 82, 1, 60–75. Warr, P. and Bunce, D. (1995), ‘Trainee characteristics and the outcomes of open learning’, Personnel Psychology, 48, 347–75. Warr, P. B., Allan, C. and Birdi, K. (1999), ‘Predicting three levels of training outcome’, Journal of Occupational and Organizational Psychology, 72, 351–76.

268 International Journal of Training and Development
© 2006 The Authors. Journal Compilation © 2006 Blackwell Publishing Ltd.