Today when I read the esttab command of Stata, I saw the star options description:

star[(symbol level [...])] causes stars denoting the significance of the
    coefficients to be printed next to the point estimates. This is the
    default. Type nostar to suppress the stars. The default symbols and
    thresholds are:  * for p<.05, ** for p<.01, and *** for p<.001.
    Alternatively, for example, type star(+ 0.10 * 0.05) to set the
    following thresholds: + for p<.10 and * for p<.05. Note that the
    thresholds must lie in the (0,1] interval and must be specified in
    descending order.

I know how to convert it to * for p<0.1, ** for p<0.05, *** for p<0.01 (the commonly-used p-value to set the levels of significance). But I am wondering why Stata set the default like that, is there anything important that I missed?


1 Answer 1

  1. Stata is not designed specifically for economics. Different disciplines have different conventions.

For example, while in economics ** signifies p-value of $0.05$ in physics ** is commonly used for $5\sigma$ significance which corresponds to p values of $0.0000006$ (for two tailed test). Although some papers there avoid the use of stars altogether.

You should note you are actually allowed to set these stars to any value you want. This is why in most research papers in table captions you will always see written stars denote this or that significance. Usually fields/subfields settle on some convention but its just that a convention and thus you should always say somewhere in the paper what is your definition of those stars.

  1. The esttab command is not official stata command, it is user written command, written by Ben Jann from Institute of Sociology, University of Bern. You can contact the author if you want to know reason for such selection. I conjecture that the field/subfield where the author works uses * for $0.05$, ** for $0.01$ convention, as it made some sense to him if he coded it this way, only way to know for sure is to contact the author and ask.
  • 2
    $\begingroup$ Just adding the paper talking about this asterisk here $\endgroup$ Commented Jun 30, 2021 at 23:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.