I am currently testing two interaction terms individually in OLS regressions. Both interaction terms are significant (p < .5) when tested individually (i.e., two regression equations, one for each interaction term) but both interaction terms become insignificant (p > .1) when tested at the same time (i.e., two interaction terms in a single regression equation). The correlation between my two moderators is moderate ($r = 0.44$). What might be some possible reasons for the terms becoming insignificant when tested simultaneously?

  • $\begingroup$ Can you clarify what you mean by simultaneous testing? What are the p-values in each case? $\endgroup$ – Bryan Krause May 6 at 16:08
  • $\begingroup$ Thanks. I tried to make it more clear. $\endgroup$ – user18075 May 6 at 16:26
  • $\begingroup$ Could you be more specific about how you are "testing at the same time"? Are you consulting the individual p-values or are you perhaps examining the F statistic for the two interactions together? $\endgroup$ – whuber May 6 at 16:43
  • $\begingroup$ Re the edit: this is an example of a phenomenon discussed in many good posts. See the hits at stats.stackexchange.com/…. $\endgroup$ – whuber May 6 at 17:06
  • $\begingroup$ Sorry if it is still confusing. When I test both interaction terms in a single regression equation both interaction terms become insignificant (p > .1). $\endgroup$ – user18075 May 6 at 17:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.