I ran an independent sample t-test to compare two groups, student sample and Mturk respondents. The homogeniety of variance was not met (variances are not equal and sample sizes of the two groups are drastically different). Sample size of the groups are 113 and 14 respectively). I have a significant finding when running the Welch's t-test. My question is:

  1. What effect size do I report for a Welches' t-test? Is Cohen's d appropriate?
  2. Is it approprate to run this test for such drastically different sample sizes?

Mturk respondents (M = 4.23, SD = .69) were more likely to report that they would submit their profiles than undergraduates (M = 3.21, SD = 1.42), t(13.78) = 2.63, p < .05, (Report effect size here).

Thank you for your time and help!


1 Answer 1


May I refer you to this thread: Do Cohen's d and Hedges' g apply to the Welch t-test?

And I’ll add that, if you’re reporting a standardized mean difference (such as Cohen’s d), you may report standardized effect sizes as Hedges’ g which is more appropriate for smaller samples and similar to Cohen’s d for larger samples.

  • $\begingroup$ @Kay if you feel that this response sufficiently addresses your question, you’re encouraged to click on the check mark to denote that you’re satisfied with this response. Else, please let me know where clarity is needed. $\endgroup$
    – giopico
    Jun 17, 2022 at 16:00

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.