Logistic regression with an interaction term of two predictor variables. It's correct to interpret non-significant coefficients?

[Andres Salman]

Hello,

I am a bit curious about the interpretation of results given from: https://stats.idre.ucla.edu/other/mu...ic-regression/

It is correct to interpret the interaction even if the coefficients are not statistically significant?

Here is the excerpt from the tutorial.

Thank you,
Andres

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - Stata code in code delimiters, readable Stata output, and sample data using dataex.

There are two sides to this. On the one hand, we can't be sure that the effect of math varies between women and men. On the other hand, using the interaction is probably the best prediction of the dv given the results.