审稿人的意见：Regarding methodological issues, to my knowledge, the interaction effect exists if the coefficient is significant and it must also to fulfill a second condition: it must involve a significant improvement of the power of the explanation of the variables, in comparison to the previous model in which only the control variables and the main effects are included (*Cohen & Cohen, 1983). Therefore, authors should test or explain if the improvement from 0.2658 in R2 to 0.2690 and to 0.2698 (Table 4) are significant. I don’t know if the methods employed require less conditions, but authors must explain it supporting the explanation of methods with methodological references. In their response to my previous comments, authors argue that “according to statistical methods and relevant references, if the interaction between independent variable and moderating variable in the model is significant, the variable plays a moderating effect in the relationship between independent variable and dependent variable”. Please, you can refer to these relevant references in your paper.

请问应该如何做啊~