I used SPSS HAYES macro to test for indirect effect, and got the results as below.
I used bootstrap method to derive confidence intervals, and it seems the indirect effect is significant at p<.05
However, SPSS macro does not produce the exact p-value for the indirect effect it has calculated via bootstrap method. Can I just manually calculate the p-value by looking up the t-distribution table using the standardized effect and bootstrapped standard error?
If possible, what df value should I use for the t statistic? should I use the df value for the entire model with the mediator included, or the total effect model which does not include the mediator?