How to conduct a survey

[deleted-324908]

I am trying to conduct a survey on if gender affects color preference. First, I don't know what would be a good sample size. I was thinking I need to have equal parts boy and girls. I thought about asking people but my teacher said there is a difference in people volunteering participate and that they aren't exactly random people. Do you have any suggestions for who to survey? Could I post the survey somewhere here to get responses? Thanks in advanced for all your help.

[deleted-140482]

Hi Mzking64,

In terms of a good sample size, in general I would say the more the better, but I would shoot for at least 30 people if possible, because certain statistical tests work better once you have at least 30 samples.

It's true that there are differences between people who volunteer to participate and a true random sample, but it may not be possible for you to get a completely random sample. I don't think there is anywhere on this site that you can post your survey. I recommend seeing if you can get teachers in your school to give your survey to their classes. It might not be completely random, and it will focus on students about your age, not adults, but this should give you a decent sample size, with a relatively even mix of males and females, without it only being made up of you particular friends.

I hope this helps and good luck with your survey! Let us know if you have any more questions we can help with.
JMP