Probit or logit?

[John Galvin]

Dear community?
Is there any difference in using probit or logit regression when i have binary DV?

[Andrew Musau]

Both give very similar transformations from \(\beta^{\prime}x\) to \(\text{Pr}(\beta^{\prime}x)\) and it is very difficult to distinguish them empirically. Therefore, it matters little which is used. However, some disciplines have a preference for one over the other (e.g., in Economics, there is a preference for probit).

[Richard Williams]

Some people like logit because they can exponentiate the coefficients and get odds ratios. Other times the choice is made based on whatever is customary in your discipline. With more advanced methods, there may be statistical reasons why you have to use logit or probit. But with binary DVs, I've never seen a case where it made any substantive difference whether you used logit or probit.