Using omega squared after regression

[Evgeniya Sherstneva]

Dear Forum,
Can I use omega squared after regression to determine effect size if my independent variables of interest are quantitative?
Thank you!

[Evgeniya Sherstneva]

To be precise, I want to measure effect sizes because the regression coefficients are not meaningful. In addition, I would like to compare the effects of different factors related to compliance with COVID-19 preventive measures, but my independent variables of interest have different units. However, I am not sure if I can use omega squared after regressions to measure effect sizes if my dependent variable and explanatory variables of interest are quantitative.

eststo sob3: reg compliance sob_factor $SD $PT $ES
estat esize, omega

[Maarten Buis]

As long as you use the factor variable notation to include your categorical variable in your model, you should fine.

[Evgeniya Sherstneva]

Thank you for your answer Maarten Buis!
As I understand if the sob_factor is a continuous variable then using omega squared is a wrong decision?

[Maarten Buis]

Oh, I misread your question. No, you can use omega for both quantitative and qualitative variables, but if it is for qualitative you need to tell Stata it is qualitative (which means use factor variable notation). In the manual (the pdf to which the help file links) you see an example where they use it to interpret the effect of a quantitative variable. So, you are fine either way.

[Evgeniya Sherstneva]

Thank you very much for your help!