A comparison of two survey questions illustrates response bias:
Version A: 'Do you support or oppose stricter regulations on industrial pollution?'
Version B: 'Do you support job-killing regulations that burden businesses with excessive pollution compliance costs?'
Which of the following best describes the expected effect of Version B compared to Version A?
- ABoth versions will produce the same results because they ask about the same topic
- BVersion B will produce lower support because of its loaded wordingCorrect
- CVersion B will produce more nonresponse because it is a longer question
- DVersion B will produce higher support for regulations because it mentions jobs
Why B is right
Version B uses negatively loaded language ('job-killing,' 'burden,' 'excessive') that frames regulations in an unfavorable light. This biased wording primes respondents to view regulations negatively, reducing reported support compared to the neutral Version A. This is a textbook example of how question wording can create response bias.
Why the other 3 choices miss
- A
- Incorrect. Decades of survey research demonstrate that question wording significantly affects responses, even when the underlying topic is the same. Loaded language consistently shifts opinions.
- C
- Incorrect. While slightly longer questions can sometimes increase nonresponse, the primary effect of Version B is to shift the content of responses through biased wording, not to change response rates.
- D
- Incorrect. The mention of jobs is framed negatively ('job-killing'), which would decrease rather than increase support for regulations.