Testing the interaction effects of task complexity in computer training using the social cognitive model

Using a Modified Social Cognitive Theory framework, this study examines the behavior modeling and lecture-based training approaches to computer training. It extends the existing Social Cognitive Model for computer training by adding the task complexity construct to training method, prior performance, computer self-efficacy, outcome expectations, and performance. A sample of 249 students from a large state university served as participants in a laboratory experiment that was conducted to determine the task complexity*training method and task complexity*self-efficacy interaction effects on performance. Structural equation modeling with interaction effects was used to analyze the data. The results show that behavior modeling outperforms lecture-based training in a measure of final performance when task complexity is high. Further, it is found that computer self-efficacy has a greater positive effect on performance when task complexity is high than when task complexity is low. Prior performance is also found to be an important variable in the model.