0
$\begingroup$

I need some help.

I am making a comparison between the percentage of responses to a questionnaire in my target population (N=210) and the responses at a regional level to the same questionnaire (the only data I have is, for example, that 45% responded A, 55% B - but I don't know the size of the sample) given from the Ministry of Health.

I guess I cannot do a chi squared , right?

Do you know can I do?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

Without any assumption on the size of the population this is impossible. The best thing you can probably do is make some assumptions about the size of the population.

With 45% of responses A the minimum number of samples must be at the very least 20. If you assume 20 samples and find a significant effect you'll know that it also applies if there were more samples.

You could also see how many samples you need to assume to get what p-value. You can then make an educated guess about whether there really would be at least that many samples.

Improve this answer
$\endgroup$

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.