Shannon L. Craig Seminar in Industrial Psychology September 29, 1999
Transfer of Training
While corporations hire employees based on the knowledge, skills, and abilities that they possess prior to their appointment, all members of the workforce will participate in training throughout most of their careers. While some people will only participate in job specific training, others will have the opportunity to be trained on various concepts such as time management, stress management, empowerment, and leadership. Muchinsky (1999) cites Goldstein (1991) as referring to training processes as �the systematic acquisition of attitudes, concepts, knowledge, roles, or skills that result in improved performance at work� (p. 171). Tziner, Haccoun, and Kadish (1991) state that �the fundamental purpose of training is to help people develop skills and abilities which, when applied at work, will enhance their average job performance� (p. 167). One last definition of training is that it is �a planned learning experience designed to bring about permanent change in an individual�s knowledge, attitudes, or skills� (Campbell, Dunnette, Lawler, & Weick, 1970, p. 497).
The problem inherent in all of these definitions is that they fail to recognize what is necessary for training to result in improved performance or any sort of permanent change. For these things to happen a transfer of training must take place. �Transfer of training is defined as the extent to which trainees effectively apply the knowledge, skills, and attitudes gained in a training context back to the job� (Muchinsky, 1991, p. 197). Transfer of training may take place in three different ways: positive, which results in enhanced job performance; negative, which results in a deficit from previous job performance; or neutral, which has no effect on job performance. Only by maximizing positive transfer of training can a program attain the goals it was designed to achieve. The occurrence of such transfer is reliant on the generalization of the learned behavior to the job context and its maintenance over a relatively long period of time (Baldwin & Ford, 1988). While it is obvious that transfer of training is important for the training to be worthwhile, there are also financial considerations for corporations. It has been estimated that 100 billion dollars is spent each year on training and development programs (Kelly, 1982), but only about 10 percent of the money spent on such programs results in behavioral changes (positive transfer) once an employee returns to work (Georgenson, 1982).
Kirkpatrick�s Training Criteria The effectiveness of training has traditionally been measured based on Kirkpatrick�s (1967) hierarchical model of training, which is composed of four levels of training outcomes. This taxonomy of training was considered useful in business and academia due to its ability to evaluate training programs in an organized, structured manner that was easily understood (Shelton & Alliger, 1993). The four levels of criteria are reaction, learning, behavior, and results. Reaction. Reaction criteria make up the lowest level of the hierarchy. They measure the participants� feelings and impressions about the criteria, basically the face validity of the training. Alliger, Tannenbaum, Bennett, Traver, and Shotland (1997) further separate this category into affective reactions and utility judgments, or how participants feel about the program as well as how useful they consider the training to have been.
Learning. Learning criteria evaluate how much has been learned and can include measures of outcomes that are cognitive, skills-based, or affective (Cascio, 1998; Muchinsky, 1999). Alliger et al. (1997) further divide this set of criteria into three subcategories: knowledge that is assessed immediately after training, knowledge that is assessed at a later time, and behavior/skill demonstration assessed immediately after training. The vast majority of research done on training has focused on assessments immediately following the training program. Reaction and learning criteria together compose internal criteria because they are assessments internal to the training program itself (Muchinsky, 1999).
Behavior. Behavior criteria assesses actual transfer of training to on-the-job performance. It refers to changes in behavior that are retained and applied to the workplace. Cascio (1998) says that two steps are involved: �the first is to demonstrate behavioral changes in on-the-job behavior, and the second is to demonstrate that such changes are due specifically to training� (p. 284). He also notes that the assessment should be made after a reasonable length of time for the participant to successfully integrate the new material into his/her work environment. He suggests about three months. Alliger et al. (1997) have renamed this level �transfer� to accentuate the distinction between behavior on-the-job and behavior displayed in a test situation immediately following the training program. Results. Results criteria measure the impact to the organization and can include assessments of productivity changes, scrap, accidents, and profit. Together, behavior and results criteria are external criteria - assessments which are external to the training program itself (Muchinsky, 1999). The changes made and categories added to Kirkpatrick�s hierarchy by Alliger et al. (1997) are an indication of growing concern that the traditional method of evaluating training programs may not be as sufficient as was once thought. The four criteria levels are thought to be both vague and overlapping (Alliger et al., 1997). Alliger and Janak (1989) found that the four criteria did not necessarily exhibit a causal relationship or show a positive intercorrelation. Werner, O�Leary, Baldwin, and Wexley (1994) also found results supporting the idea that reaction and learning criteria are not always highly correlated. Noe and Schmitt (1986) also found limited support for the hierarchy. They found a relationship between behavior change and performance improvement, but no indication that trainee satisfaction with the training program affected learning or that learning had an impact on behavior change. Alliger et al. (1997), however, did find that utility reactions were more strongly related to transfer than were affective reactions. While a review of Kirkpatrick�s levels of criteria is important to understand how training programs are evaluated, the focus of this paper is on factors which may affect the transfer of training.
Factors Affecting Transfer
There are a number of factors which have been found to affect the positive transfer of training. These include practice, similarity of training variables to actual performance variables, cognitive ability, attitude or motivation of participants, locus of control, transfer climate, goal-setting, and self-management to maintain changes in behavior. Practice. Practice is believed to be important in the transfer of training because it facilitates the provision of experience with the tasks, concepts, and skills being taught to participants so they are better able to deal with situations that aren�t specific to examples set forth in the training (Cascio, 1998). Cascio (1998) distinguishes between three types of practice: active practice, overlearning, and length of the practice session. Active practice is considered favorable because it is thought to be the only type that provides internal cues that regulate motor performance (Cascio, 1998). The idea behind overlearning is that a task becomes �second nature� after repeated performance. Baldwin and Ford (1988) cite studies in support of the idea that overlearning resulting from continued practice will lead to increased retention of the material presented in the training program. Driskell, Willis, and Copper (1992) suggest that overlearning has one more advantage in addition to those presented above. They conclude that overlearning is effective for cognitive as well as physical tasks, but that the effect is stronger for cognitive tasks. The length of the practice session relates to whether the practice is distributed, with rest intervals between sessions, or massed, where sessions are very close together. Baldwin and Ford (1988) cite research findings supporting the idea that �material learned under distributed practice is retained longer than material learned by massed practice� (p. 67). They cite a study by Holding (1965) which found that for difficult and complex tasks, higher performance resulted when practice sessions were first presented as massed and followed by somewhat more distributed sessions. Baldwin and Ford (1988) also further differentiate practice into whole, in which all the material is presented at once, and part, in which different aspects of the training are practiced at different times. They cite a study which found that the whole method was more effective when the intelligence of the learner was high, practice sessions were distributed, and the training material was highly organized but not complex. It is thought, then, that distributed practice sessions are more effective, but the intelligence of the trainees and complexity of the tasks being practiced can alter the effectiveness of both massed and distributed practice sessions.
Identical Elements. Research has also focused on the effects of providing training programs in which the training variables are similar to variables present in performance. Cascio (1998) suggests that the similarity between the training and the job should be maximized, there should be a strong link between training content and job content, and trainees should understand the principles being taught and how to apply them. The idea of similarity has frequently been referred to as �identical elements.� Osgood (1949) published some of the first material on this concept. He stated that positive transfer results from varied stimuli to a particular response. In other words, training should have set outcome variables and the training variables presented to meet those outcomes should be similar to the outcome, but varied. Baldwin and Ford (1988) cite a variety of studies supporting the concepts postulated by Osgood (1949), including the idea of identical elements, varying stimuli, and the teaching of general rules and underlying theoretical principles. They state that the theory behind varying stimuli is to �avoid the problem of training becoming attached to a narrow range of stimuli and responses� (p. 87). They also point out a problem with identical elements being that research has yet to determine exactly which elements of a training program and job should be identical. Ford, Smith, Weissbein, Gully, and Salas (1998) point out recent research which �has determined that it is the similarity of the information processing between the training and transfer tasks that is most important� (p. 221). Cognitive Ability. Cognitive ability is another factor which has long been thought to affect transfer of training. It is thought that approximately 16 percent of the variance in trainee performance is a result of cognitive ability (cited by Noe, 1986). Noe (1986) says that �the cognitive and psychomotor skills that trainees possess directly influence whether or not they will be able to understand and master the content of the training program� (p. 737). For this statement to be indicative of transfer the relationship between Kirkpatrick�s criteria must be assumed, specifically the idea that learning is highly correlated with behavior. Ford et al. (1998) cite Glaser (1986) as saying that a sufficient knowledge base to recognize what a transfer environment must consist of is a prerequisite for positive transfer. Ford et al. (1998) found a significant link between knowledge and transfer. Baldwin and Ford (1988) also cite studies in support of general intelligence affecting both learning and transfer.
Attitude. Attitude of the participants, including notions of motivation and self-efficacy, has also been shown to be an indicator of positive transfer. Four attitudes of participants which are likely to lead to greater transfer include trainees being confident enough to use the new skills, being aware of appropriate work situations for such use, being perceptive of improvement in their performance resulting from such use, and believing that the new skills will be useful to them in their work environments (Cascio, 1998; Noe & Schmitt, 1986). Cascio (1998) also points out that high self-efficacy is also necessary, as well as motivating factors in the work environment. Self-efficacy is defined as �an individual�s estimate of his or her ability to perform a specific behavior� (Liebert & Spiegler, 1994, p. 430). Ford et al. (1998) cite a study pointing out that the reasoning behind self-efficacy enhancing transfer is that people with greater confidence in their abilities will be more resilient when facing any barriers to implementing their new skills. A significant effect is found for self-efficacy on transfer performance and the researchers conclude that this is evidence of the necessity for trainers to consider the role of affect and motivation on transfer of training, suggesting that the development of self-efficacy and motivation (possibly through the mastering of goals) be provided during the training (Ford et al., 1998). Noe and Schmitt (1986) point out that transfer is higher when participants agree with the assessment of their skill weaknesses, and therefore also accept the training as an acceptable way to overcome those weaknesses. Werner et al. (1994) introduced a pretraining motivational intervention and failed to find any significant effects of this intervention on transfer. They suggest that extending the time over which the intervention is spread may show more of an effect. Baldwin and Ford (1988) identify one problem with research in this area being that there is a lack of theoretical frameworks to guide it, however, at the time this statement was made there was less research available on which to base this judgment.
Locus of Control. A trainee�s locus of control is also thought to affect the extent to which they transfer their newly learned skills (Noe & Schmitt, 1986). Locus of control is a way of describing what individuals view as being the source of their outcomes. People with an internal locus of control believe that their outcomes are determined by their own efforts and decisions. Those with an external locus of control believe that they have no control over their outcomes, which are determined by outside forces (Liebert & Spiegler, 1994). Noe (1986) cites studies which conclude that individuals with an internal locus of control are more likely to transfer new skills because they are more likely to accept feedback and take action to correct their shortcomings. It is also thought that locus of control can affect transfer due to some trainees feeling a greater loss of control in situations where they perceive constraints within their work environment (Tziner et al.,1991). Baldwin and Ford (1988) state, however, that there is �limited support for the effects of locus of control on pretraining motivation and learning� (p. 68), which suggests that it has little effect on transfer if the linkages between motivation, learning, and transfer are assumed.
Transfer Climate. Another factor which is thought to affect transfer is the transfer climate, or the climate within the work environment that the training participant will be returning to following the training program (Baldwin & Ford, 1988). The posttraining environment needs to provide opportunities for trainees to apply their new skills (Ford et al., 1998; Muchinsky, 1999). Muchinsky (1999) discusses the importance of supervisory support. Supervisors much reinforce effective behavior in order for it to be acquired, modified, and sustained (Cascio, 1998). Baldwin and Ford (1988) also point out the importance of feedback and the timing and specificity of the feedback. For positive transfer to occur, supervisors need to provide reinforcement, modeling of behaviors, and goal-setting activities (Muchinsky, 1999). Peer support is also important for transfer (Cascio, 1998; Noe, 1986; Noe & Schmitt, 1986). Another important component of the work environment is the amount of task constraints perceived by the trainees. Constraints such as lack of equipment or financial resources can influence transfer by reducing motivation to learn or application of the newly learned skills to on-the-job behavior (Noe, 1986; Noe & Schmitt, 1986). Alliger et al. (1997) found that utility reactions were more highly correlated with transfer than were measures of learning. They suggest that utility reactions are influenced by knowledge of the work climate and that transfer is therefore more likely to occur if the work environment allows for the application of the newly learned skills. Baldwin and Ford (1988) point out two limitations of research on the effects of the transfer climate. The first weakness they identify is that the studies done are all correlational, which impedes any determination of causality. The second weakness is that all of the studies have relied on self-report measures. They conclude that the concept of �buddy systems� may be a valuable tool within the transfer climate. This system pairs two trainees together �to reinforce each other in order to maintain learning, provide advice, and be alert for signs of relapse in themselves and the buddy� (p. 98). Goal-Setting. Goal-setting activities have also been studied as having an effect on transfer of training. These activities are based on the idea that conscious goals or intentions can regulate overt behavior (Cascio, 1998; Werner et al., 1994). Noe and Schmitt (1986) found that individuals with greater amounts of career planning or those who had a career strategy (career goals) experienced greater changes in on-the-job behavior as a result of the training program. Goal-setting can either be determined by the trainees (participative) or by the trainers (assigned). Wexley and Baldwin (1986) argue in favor of participative goal-setting because it increases ownership of the goals, which leads to increased transfer. In their study, Wexley and Baldwin (1986) found that both assigned and participative goal-setting resulted in greater retention of new skills on-the-job than did a control condition, but assigned goal-setting affected learning and behavior while participative goal-setting affected only behavior. It is suggested that the outward statement of the goals and monitoring by trainers both could have led to the positive transfer of training. Gist, Bavetta, and Stevens (1990) found that trainees in a goal-setting condition were less able (than trainees in a self-management condition) to apply their new skills in novel situations, but they demonstrated their skills more frequently. One study found that trainees in a goal-setting condition reported less satisfaction than those not asked to set goals immediately following the training, but were more satisfied a few months later (Werner et al., 1994). Wexley and Baldwin (1986) suggest that both trainees and supervisors should complete periodic progress reports to assess their goal achievement. Reber and Wallin (1984) found that goal-setting was even more effective for skill transfer when combined with feedback. Werner et al. (1994) conclude that assigned goal-setting is rarely used �apparently because of a belief that trainees will not like assigned goals, and this negative reaction will affect subsequent performance� (p. 171). Cascio (1998) cites a study which concludes that �the effects of goal-setting on performance can be enhanced further by providing information to performers about how to work on a task and by providing a rationale about why the goal and task are important� (p. 276). It is also suggested that in addition to discussing the content and benefits of goals with trainees, supervisors should consider setting goals both before (for improvement) and following (for application) the training program (Wexley & Baldwin, 1986).
Posttraining Strategies. The final factor affecting transfer of training which has received a great deal of research attention is the concept of posttraining strategies, specifically behavioral self-management and relapse prevention. Behavioral self-management is based on the idea that the setting of behavioral targets will lead to higher transfer levels (Wexley & Baldwin, 1986). The concept of behavioral self-management was popularized by Luthans and Davis (Wexley & Baldwin, 1986). They look at behavioral self-management as a way of manipulating stimuli and rewards for ourselves to bring about desired changes (Luthans & Davis, 1979). More specifically, they define it as �deliberate regulation of stimulus cues, covert processes, and response consequences to achieve personally identified behavioral outcomes� (p. 43). Gist et al. (1990) found that trainees in a self-management condition were better able to generalize their new skills to apply them to novel situations. It has also been concluded, however, that behavioral self-management does impact learning retention and behavior modeling, but has little effect on transfer (Werner et al., 1994). The concept of behavioral self-management led Marx (1982) to create what is know as relapse prevention. The concept and effectiveness of relapse prevention has been examined by a variety of researchers (Baldwin & Ford, 1988; Cascio, 1998; Muchinsky, 1999; Noe, 1986; Tziner et al., 1991; Wexley & Baldwin, 1986). Though it was originally designed as a method of preventing relapse in individuals with addictive behaviors (Marx, 1982), relapse in this context refers to a return to behaviors practiced prior to participation in the training program designed to alter those behaviors (Wexley & Baldwin, 1986). Marx (1982) says that the first step in the process is to make trainees aware of the relapse process by asking them to pinpoint high risk situations, or situations in which they may inadvertently revert back to previous behaviors. These situations usually include pressure of some sort - such as time, resources, or managerial. Trainees are then asked to discuss specific situations in which they have suffered from relapse before and ways to avoid such relapse in the future. This is looked upon as an opportunity for learning and change rather than an admission of failure (Marx, 1982). The next step is to teach the trainees coping strategies for when the situations discussed do arise. This may require overcoming skill deficits which can be done through additional training, such as time management or strategic planning (Marx, 1982). The final step involves improving the self-efficacy of trainees by teaching them to feel a sense of accomplishment in simply attempting to maintain newly learned behaviors in difficult situations (Marx, 1982; Noe, 1986). Marx (1982) also suggests practicing the relapse experience so that the trainer can help identify any attitudinal factors that may lead to slips and reinforce any effective coping behaviors. Appendix 1 shows the different outcomes that Marx (1982) contends can be expected both with and without the use of relapse prevention. Tziner et al. (1991) suggest that for relapse prevention to be effective, trainees must view themselves as being capable of and willing to assume control in difficult situations. They therefore conclude that locus of control may impact the effectiveness of any relapse prevention strategy. It was also found that trainees who viewed their work environment as being more supportive and participated in relapse prevention experienced higher levels of transfer (Tziner et al., 1991). Tziner et al. (1991) state, �Several possible reasons exist why RP produces positive results. It may serve to reinforce the perception of applicability of trained skills; it may have strengthened or consolidated the acquired knowledge; it may help trainees develop the behavioural repertoire for dealing with application situations which are not ideal for implementation, and/or it may provide the flexibility in skill usage needed to overcome barriers in unexpected job situations, or it may have helped trainees develop greater levels of self-efficacy in transferring skills� (p. 176). The information presented here on transfer of training should show, if nothing else, that transfer is both affected and clouded by a large number of variables. While training effectiveness has traditionally been measured based on criteria set forth by Kirkpatrick, the sufficiency of this measure is beginning to fall under question. The factors affecting transfer are also numerous: practice, similarity of the training variables to performance variables, cognitive ability, attitude (including motivation and self-efficacy), locus of control, environmental transfer climate, goal-setting activities, and behavioral self-management along with relapse prevention. These factors become clouded as they are found to interrelate with one another. Their influence then becomes even more clouded as the research techniques employed to explore their impact is questioned (Baldwin & Ford, 1988). It becomes obvious, then, that continued research is necessary for a topic as essential to industrial/organization psychology - the topic of transfer of training.
REFERENCES
Alliger, G.M., Tannenbaum, S.I., Bennett, W., Traver, H., & Shotland, A. (1997). A meta-analysis of the relations among training criteria. Personnel Psychology, 50, 341-358.
Baldwin, T.T., & Ford, J.K. (1988). Transfer of training: A review and directions for future research. Personnel Psychology, 41, 63-105.
Campbell, J.P., Dunnette, M.D., Lawler, E.E. III, & Weick, K.R., Jr. (1970). Managerial behavior, performance, and effectiveness. New York: McGraw-Hill.
Cascio, W.F. (Ed.). (1998). Applied Psychology in Human Resource Management, 5th Edition. Upper Saddle River, NJ: Prentice Hall.
Driskell, J.E., Willis, R.P., & Copper, C. (1992). Effect of overlearned on retention. Journal of Applied Psychology, 77, 615-622.
Ford, J.K., Smith, E.M., Weissbein, D.A., Gully, S.M., & Salas, E. (1998). Relationships of goal orientation, metacognitive activity, and practice strategies with learning outcomes and transfer. Journal of Applied Psychology, 83 (2), 218-233.
Georgenson, D.L. (1982). The problem of transfer calls for partnership. Training and Development Journal, 36 (10), 75-78.
Gist, M.E., Bavetta, A.G., & Stevens, C.K. (1990). Transfer training method: Its influence on skill generalization, skill repetition, and performance level. Personnel Psychology, 43, 501-523.
Kelly, J.B. (1982). A primer on transfer of training. Training and Development Journal, 36 (2), 102-106.
Kirkpatrick, D.L. (1967). Evaluation of training. In Craig, R.L., Bittel, L.R. (Eds.), Training and Development Handbook (pp. 87-112). New York: McGraw-Hill.
Liebert, R.M., & Spiegler, M.D. (1994). Personality: Strategies and Issues Seventh Edition. Pacific Grove, CA: Brooks/Cole Publishing Company.
Luthans, F., & Davis, T.R.V. (1979). Behavioral self-management - The missing link in managerial effectiveness. Organizational Dynamics, 8 (1), 42-60.
Marx, R.D. (1982). Relapse prevention for managerial training: A model for maintenance of behavior change. Academy of Management Review, 7 (3), 433-441.
Muchinsky, P.M. (Ed.). (1999). Psychology Applied to Work. Stamford, CT: Wadsworth.
Noe, R.A. (1986). Trainees� attributes and attitudes: Neglected influences on training effectiveness. Academy of Management Review, 11 (4), 736-749.
Noe, R.A., & Schmitt, N. (1986). The influence of trainee attitudes on training effectiveness: Test of a model. Personnel Psychology, 39, 497-523.
Osgood, C.E. (1949). The similarity paradox in human learning: A resolution. Psychology Review, 36, 132-143.
Reber, R.A., & Wallin, J.A. (1984). The effects of training, goal setting, and knowledge of results on safe behavior: A component analysis. Academy of Management Journal, 27, 544-560. Shelton, S., & Alliger, G.M. (1993). Who�s afraid of Level 4 Evaluation? A practical approach. Training & Development Journal, 47, 43-46.
Tziner, A., Haccoun, R.R., & Kadish, A. (1991). Personal and situational characteristics influencing the effectiveness of transfer of training improvement strategies. Journal of Occupational Psychology, 64, 167-177.
Werner, J.M., O�Leary-Kelly, A.M., Baldwin, T.T., & Wexley, K.N. (1994). Augmenting behavior-modeling training: Testing the effects of pre- and post-training interventions. Human Resource Development Quarterly, 5 (2), 169-183.
Wexley, K.N., & Baldwin, T.T. (1986). Management development. Journal of Management, 12 (2), 277-294.
Wexley, K.N,. & Baldwin, T.T. (1986). Posttraining strategies for facilitating positive transfer: An empirical exploration. Academy of Management Journal, 29 (3), 503-520.
Wexley, K.N., & Latham, G.P. (1981). Developing and Training Human Resources in Organizations. Glenview, IL: Scott Foresman.
