[Home Page] [Previous Page]
[Old Roundtable Archives] [New Roundtable]
Posted by jon on February 03, 2000 at 15:05:16:
In Reply to: Re: Questions, please help posted by John Taylor on February 02, 2000 at 12:35:04:
I agree with John - if your study will receive appreciable scrutiny, your sample is the foundation of your work and needs to be defenceable. I would sample a 100 or so of your non-responders and compare/contrast their responses. If not significantly statistically different, accept, include, and present as no appreciable non-respondent bias. Also, 27% response from a mail sample to a random group is not a bad response rate. I have seen higher rates (45% to 70%) only from targeted groups with a vested interest in the outcome.
Subject: