0
$\begingroup$

I have designed a questionnaire asking people about various observations within their workplace. Do I need to do a power analysis for this?

So far I have collected data from 72 people. Is this a good sample size, or should I continue collecting more data?

Improve this question
$\endgroup$
3
  • $\begingroup$ Some people (although certainly nobody who routinely hangs out on this site!) have said things tantamount to "if you have obtained significant results with your questionnaire, then stop; otherwise, if you have the time and budget, administer it to more people until you do obtain significant results." In light of this possibility (and its awful consequences), could you perhaps edit your post to elaborate on what a "good sample size" might mean for you? $\endgroup$ Commented Jul 1, 2018 at 15:14
  • $\begingroup$ Thanks, It was just a small survey and I have significant results. I suppose I was hoping to publish this but was unsure as 72 seems a small sample. $\endgroup$ Commented Jul 4, 2018 at 12:30
  • $\begingroup$ This is called a post hoc power analysis. It doesn't tell you a heck of a lot, because its results are hypothetical and irrelevant to the data you actually obtained. $\endgroup$ Commented Jul 4, 2018 at 14:08

1 Answer 1

Reset to default
0
$\begingroup$

If you want to know if you have a good sample size to attain statistical significance (if there is, in fact, something going on) or to estimate a parameter to a given degree of precision, then yes, you need to do a power analysis.

Improve this answer
$\endgroup$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.