1
$\begingroup$

Hi & thank you all in advance for your help!

I sent out a survey [to examine teacher satisfaction] to 3 school districts in New York state.

District A is large, district B is medium, & district C is small. Let’s say the response rate from District A is 25%, from District B is 3%, and from District C is 90%.

I don’t want to report that the NY state response rate is 25 + 3 + 90/3. I’d like to account for the size differential among the districts and the disparate response rate among the districts. How do I do this? Is there a weighing methodology/formula for this?

Thank you, NJ

$\endgroup$
2
  • 1
    $\begingroup$ Because the differences among response rates may be informative, why not report them by district? Reporting some kind of statewide average could deceive your readers. $\endgroup$
    – whuber
    Commented May 2, 2016 at 18:25
  • $\begingroup$ Hi..thanks for your response. I am reporting to the feds. They collect and then aggregate data from various states. $\endgroup$
    – nora jones
    Commented May 2, 2016 at 19:24

1 Answer 1

0
$\begingroup$

The canonical ways to calculate response rates are given by the Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys by the American Association for Public Opinion Research. See http://www.aapor.org/Publications-Media/AAPOR-Journals/Standard-Definitions.aspx. The edition as of August 2016 when I am typing this is the 9th edition. The Standard Definitions are revised every two-three years to accommodate the changes in the survey protocols, frames, technologies, sampling practices, etc.

You need to think of this as the industry standard. If you report anything else that you come up with, this would look amateurish.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.