Supervisory Ratings as a Measure of Training Transfer: Testing the Predictive Validity of the Learning Transfer System Inventory

Publication Year:
Usage 114
Downloads 102
Abstract Views 12
Repository URL:
Katsioloudes, Vicky
predictive validity; transfer or learning; training transfer; LTSI
thesis / dissertation description
The primary purpose of this dissertation was to examine the criterion-related validity of all 16 LTSI General and Specific Scales in relation to training transfer. Using an ex-post facto design the researcher analyzed secondary data that included measures of key transfer system variables, as captured by the LTSI scales, and individual training transfer as indicated by supervisory ratings. The dataset comprised of employees from diverse organizations who had participated in different types of organizational instructor-led classroom training programs. Confirmatory factor analyses supported the construct validity of the LTSI General and Specific Scales (N = 619). Hierarchical multiple regression analysis measured the unique impact of four sets of variables (i.e., secondary influences, motivation factors, ability/enabling factors, and work environment factors) on training transfer as measured by the difference score between supervisory pre- and post-training ratings indicating the frequency in which trainees utilized the behaviors on-the-job taught in training (N = 202). Results suggested that as a group the LTSI factors explained 10.3 percent of variance in training transfer. Performance outcome expectations emerged as the only significant predictor. Supplemental analyses controlling for the impact of supervisory pre-training rating scores and using supervisory post-training ratings as the dependent variable identified four significant predictors all stemming from the LTSI General Scales: two factors reflect training-motivation constructs (i.e., performance outcome expectation and transfer effort performance expectations), and the other two reflect work environment factors (i.e., performance coaching and openness to change). LTSI factors explained 10.3% of explained variance in training transfer. Overall, the results supported the notion that the transfer system as a complex of factors is important in explaining training transfer. However, the modest variance in training transfer explained by the LTSI factors in this study should be interpreted in light of the study’s limitations. Avenues for future research are discussed.