I am investigating a common discrepancy between male and female self-reports on a survey about sexual experiences. Generally, women report ~2/3 higher rates then men do. I have modified the survey and now need to find out if/how the discrepancy between males' and females' reporting rates may have changed. Males and Females each fill out both verisons (modified and original) of there gender-specific survey. So, what i want to compare is the difference between male-and-female difference scores on the original survey, and male-and-female differences on the newer/modified survey. What specific statistical test could I use here?

  • 1
    $\begingroup$ It depends on your design, of which there aren't enough details. ... e.g. was the order of the two surveys randomized? Or were the particpants assigned by design to either a survey 1 first or a survey 2 first treatment sequence? ... Or did everyone do one first and then the other? $\endgroup$ – Glen_b Mar 21 '13 at 5:18
  • 1
    $\begingroup$ the order of the two surveys was not randomized. each participant took the original version of the survey first, based on gender. $\endgroup$ – Walter Jul 18 '13 at 22:29
  • 1
    $\begingroup$ furthermore, my samples are not matched, so i have an uneven number of males and females. $\endgroup$ – Walter Jul 18 '13 at 22:32

Your Answer

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

Browse other questions tagged or ask your own question.