0
$\begingroup$

How can I perform a two sample t-test if one sample is small (n<30) and one sample is large (n>30). The usual method is giving wrong results. What should I do?

$\endgroup$
  • $\begingroup$ welcome to the site. Indicate results/output. $\endgroup$ – Subhash C. Davar Dec 31 '16 at 11:41
  • $\begingroup$ This should work perfectly well so there may be a slip in your calculations. Please edit in how you know it is giving the wrong results. $\endgroup$ – mdewey Dec 31 '16 at 13:31
  • 1
    $\begingroup$ I don't see a problem either but what do you mean by "giving the wrong results"? $\endgroup$ – Michael Chernick Dec 31 '16 at 13:53
0
$\begingroup$

When you perform t test for unequal sample size, your test loses in strength. Maybe your p value doesn't express the "right" answer because of it. Please check other people's questions about it, as here.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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