## LSU Historical Dissertations and Theses

1997

Dissertation

#### Degree Name

Doctor of Philosophy (PhD)

#### Department

Human Resource Education and Workforce Development

Elwood F. Holton, III

#### Abstract

The goal of the present research was to contribute to an understanding of the transfer of training process. This study used a conceptual model based on a comprehensive training evaluation model to examine the extent to which secondary influences on training effectiveness, motivational elements, environmental elements, ability/enabling elements, and learning from computer-based training were associated with and explained the variance in supervisory ratings of performance. Subjects were production operators in a Fortune 500 size petrochemical manufacturing firm. This study examined a revised conceptualization of transfer climate, one in which transfer climate variables were determined based on their referent or source in the work environment. As a set, transfer climate variables were shown to account for the largest proportion $\rm (R\sp2 = .36, p \le .001)$ of the variance in performance ratings. Of seven climate dimensions measured, the most powerful predictors of performance to emerge were interpersonal climate dimensions. Peer support, group resistance to change, and supervisor sanctions all emerged as significant predictors of performance. These findings strongly suggest that interpersonal transfer climate dimensions such as work group members belief about themselves as a group, normative expectations about group members work behavior, and supportive interpersonal relationships are highly influential factors dictating the use of training on the job. Results also showed that content validity was (a) positively and significantly correlated (r =.53, p $\le$.001) with performance utility; (b) correlated to a lesser degree with performance (r =.18, p $\le$.08); and (c) a significant predictor of performance $\rm (\beta = .44,\ p \le .001)$ in a regression model. In terms of the conceptual model used for this study, the results suggest that content validity may be appropriate as a secondary variable influencing performance through its motivational value as well as an ability/enabling variable influencing performance directly. The results of this study provide partial support for a conceptual model based on Holton's (1996) comprehensive training evaluation and measurement model. The findings suggest that only through the use of comprehensive, integrated models of the training process can the cause and effect of training success or failure be reliably identified.

9780591458817

274

COinS