P.Mean: Survey results from nine out of thirty six employees (created 2008-07-21).

Hi, hope you can help a struggling grad student in health promotion and education. If I administer a questionnaire to 9 out of 36 staff members, are the results statistically significant or is the survey respondents number too small? This is a needs assessment questionnaire-what the staff feels they need from an educational standpoint. Or am I floating off course and hopeless? Thank you for your time and help!

I'm glad to help as long as you don't ask me to pay off your student loans.

I would suggest that you already know the answer to this. A sample of nine subjects will have very little credibility in the research community. It is indeed possible to draw strong inferences from a sample of this size, but only in carefully controlled designs, such as a crossover trial, or in settings where tight environmental controls are possible, such as in a laboratory.

Now a small sample may still be useful, especially if you treat the results as exploratory.

You can't afford to have a low return rate on your questionnaire. Even a 30% refusal rate would be fatal here.

It would help tremendously if your sample of nine could be a truly random sample rather than a convenience sample.

When you summarize your results, do so in a qualitative rather than quantitative fashion. Avoid any p-values like the plague.

If you do report quantitative results, be sure to examine the confidence limits for those quantities. If you're trying to estimate a percentage and your confidence limits go from 2% to 94%, then that tells you that your estimate is worthless.

Also look at the homogeneity of responses. If there is a strong degree of similarity among the nine responses, that gives you a reasonable basis to believe that the remaining 27 subjects would have similar responses. If, instead, each of the nine responses looks quite different, then there is probably some important but uncollected information among the remaining 27 subjects..

Creative Commons License This work is licensed under a Creative Commons Attribution 3.0 United States License. This page was written by Steve Simon and was last modified on 2010-04-01. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Small sample size issues.