0

Trying to figure out a way to balance small samples. Comparing ratios received from a survey, but sometimes I will have 25 survey responses and sometimes I will have 1. The ratios I look at might be the same but I think I need to take into consideration that one ratio is based on 25 surveys and the other is based on 1.

Any way to do that with small samples? Thanks.

Eudora
  • 281
  • How did you derive a ratio when there is only one case? Wouldn't they be either 0 or undefined? I think in order to get a useful answer, you may need to explain the question a bit more. E.g. what is the design, why do you need to balance them, and what kind of ratio you were referring to. – Penguin_Knight Sep 27 '13 at 11:54
  • We receive two numbers on the survey, a count of usage and a count of capacity, which we use to calculate a % of capacity number. The problem comes with the response rate on the survey. If 25 sites respond with a 50% of capacity rate one day, and only 1 site reports back the next week, but also with a 50% rate, it seems like the ratio has stayed flat week over week. I'm trying to determine if/how to balance the survey responses. – Eudora Sep 27 '13 at 15:49

0 Answers0