In order to further verify H6, we tested the significance of the simple slopes when core self-evaluations predicts employee proactive behavior at higher (1 SD above the mean) and lower (1 SD below the mean) levels of follower political skill, and plotted the moderation of follower political skill according to Aiken and West (1991).
According to Table 5 (Model 4), the interaction term between core self-evaluations and follower political skill is significant and positive (β = 0.15, p < 0.05), indicating that follower political skill positively moderates the relation between core self-evaluations and proactive behavior.
The five-factor model includes authentic leadership, psychological empowerment, core self-evaluations, proactive behavior and follower political skill.
We use an 18-item scale adapted from Ferris et al. (2005)'s to measure follower political skill.
H7 predicts that the indirect effect of authentic leadership on employee proactive behavior via core self-evaluations is moderated by follower political skill.
Table 6 presents the indirect effect of authentic leadership on employee proactive behavior at higher (1 SD above the mean) and lower (1 SD below the mean) levels of follower political skill by using the PROCESS program (Model 14).
As shown in Table 6, when follower political skill is high, authentic leadership has an indirect effect on proactive behavior via core self-evaluations (β = 0.0469, 95%CI = [0.0095, 0.1149]).
When follower political skill is low, the indirect effect of authentic leadership on proactive behavior via core self-evaluations is not significant (β =0.0035, 95%CI = [− 0.0161, 0.0349]), where the 95% bias corrected confidence interval contains zero.
Core self-evaluations are positively related to proactive behavior (r = 0.12, p < 0. 10) and follower political skill (r = 0.35, p < 0. 01), which provides initial evidence in support of our hypothesized relationships.
As shown in Fig. 2, the positive relationship between core self-evaluations and employee proactive behavior is stronger when follower political skill is high (β = 0.33, p < 0.01) than when it is low (β = 0.12, n.s).s
