Stability matters: A dynamic process view on self-efficacy in training transfer.

We present a dynamic process view on self-efficacy that introduces the level and stability as two properties of self-efficacy influencing training transfer. The level of self-efficacy refers to the extent to which individuals believe in their capability to perform a task (e.g., 0%–100%). Stability refers to the extent to which the current level of self-efficacy can be predicted from the previous level of self-efficacy, indicating a lack of change in the level over time. According to our view, the level of self-efficacy positively influences the goal intentions to engage in action after training, independent of the stability of self-efficacy. Furthermore, the effect of the level of self-efficacy on transfer performance depends on the stability of self-efficacy. We test our model using a longitudinal data set over 2 years with a total of 1,399 lagged observations from N = 871 participants of an entrepreneurship training intervention. We find that training increases the level of self-efficacy in the short term and that the stability of self-efficacy develops over time during the transfer process. Furthermore, the level of self-efficacy becomes a stronger predictor of transfer performance over time as it becomes more stable. Our results suggest that both properties of self-efficacy develop over time, and the joint consideration of both properties better explains training transfer.