| Summary: | 碩士 === 輔仁大學 === 心理學系 === 106 === This study is focused on meta-analysis of the journal literature related to the training effect of Taiwan’s organization. Based on methods of Arthur et al. (2003) and modifies structures of the training model of Baldwin and Ford (1988) to study the training effect of organizations in Taiwan as well as three moderators which influence the training effect: pre-/post-training interventions, work environment, and individual characteristics.
Results of the meta-analysis revealed training effectiveness sample-weighted mean ds of 0.57 (k = 31, N=2839) by learning criteria, and 0.66 (k = 11, N=867) by behavioral criteria. These results, medium to large effect size, are similar to the results of Arthur et al. (2003); namely, the organizational training of Taiwan is quite effective. However, by results criteria, the effect size is only about 0.32, which is very different from the research result of Arthur et al. (2003). One reason may be that the number of papers about results criteria is too small to avoid sampling error. The other reason may be that in Taiwan the training of results criteria does not take other variables into account, and lacks the system concept of the training system because 75% of the data points in the qualitative data do not include individual characteristics and interventions before and after, and because 50% of the data points in the qualitative data do not take work environment into consideration.
Finally, the results also show the effects of interventions before and after training and individual characteristics at the learning and behavioral criteria are considered higher than unconsidered, and work environment have a positive impact on training outcomes when behavioral criteria are included. Showing the system concept is important in the training. It is suggested that the thoughtful analysis of training effects can be used to analyze correlation with variables in the future in order to understand the interactions, and to collect data from behavior and results criteria. Also, it is possible to conduct an in-depth discussion on one of the three variables in order to increase breadth and depth.
|
|---|
