2
$\begingroup$

I'm writing my thesis on people's attitude. The problem is, I have way more female (75%) than male (25%). I wanted to run an analysis on the different attitude between men and women, but I'm not sure which test to use, since T-test would require the 2 groups to be more or less similar sizes and variance. Thank you for your help in advance!

$\endgroup$

2 Answers 2

2
$\begingroup$

Assuming that your response variable fulfills the assumptions made for a t-test (normality of sample mean etc.) and your only concern is that your two groups are different in size and/or variance, then you can use Welch's t-test, which takes care of exactly those concerns while addressing the same issue, namely whether the means are statistically different or not.

$\endgroup$
1
$\begingroup$

Welcome to Cross Validated!

Actually, there is no reason why you shouldn't perform a mean comparison test. Equal sample size and varaince are not a requirement.

You may want to check this to see how it can be done (unknown means and unknown standard deviations)

$\endgroup$

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.