Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
response rates
Message posted by Cindy on October 19, 2000 at 12:00 AM (ET)
My institution recently
conducted a census of its
first-year class. The response
rate was 60%. Many of our
students dismiss the results
by noting that we did not hear
from 40% of the class so our
results are not representative
of the entire class. How should
I respond to this?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: response rates
Message posted by SOF on October 20, 2000 at 12:00 AM (ET)
Even if your sample had been random or probabilistic there would be a known probability of it being non-representative of the population. Your problem now, is to show that the non-respondents are similar to the respondents or differ in known way that can be compansated for. You can also claim that you have a random sample of that part of the population that does tend to respond to surveys - call them the 'conformists' .
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.