This Discussion Assignment involves writing closed-ended survey questions that could be included in a self-administered questionnaire. You should assume that the interview in which your questions would appear will be completed just once by a probability sample of 1500 US adults. You should also assume that the questions you write will be a small part of the overall questionnaire. There are several steps to the prompt, so please read this through carefully.
1. Identify your hypothesis. For this step, you need to state a testable relationship between two concepts of interest. On pages 53 54 of Carr et al. (2021) there are several examples. You can use one of those or come up with one of your own. Is your hypothesis a hypothesis of difference, a hypothesis of association, or a causal hypothesis?
2. Write 4 or 5 closed-ended questions that would measure the two concepts in your hypothesis and that would be suitable for inclusion in the questionnaire. Make sure that you follow the rules discussed in chapter 7 (on survey research) of the textbook (Carr et al., 2021). Also make sure to include any skip patterns that might be needed to collect the appropriate information from respondents (see page 199 of the textbook). Only write questions that are relevant for the two concepts in your hypothesis. You may use questions from existing sources (with proper citation), but do not assume that previously used questions are necessarily the best.
3. How would you pretest these questions? Both the textbook and the Groves et al. chapter (2009) Download Groves et al. chapter (2009) suggest methods for evaluating survey questions. What technique(s) do you think would work best for your questions? Why do you say that?
4. Remember that the goal for writing survey questions is to test your hypothesis. Do you see any obstacles to doing so? Obviously, if these measures are not valid and reliable there could be a problem. There might also be issues presented by the design of the survey (self-administered, one time only, etc.) or the nature of the topic (causing embarrassment for the respondent, for instance). You can assume that potential confounds will be measured as part of the questionnaire if you choose to comment about that.
References:
Carr, D., Boyle, E. H., Cornwell, B., Correll, S., Crosnoe, R., Freese, J., & Waters, M. C. (2021). The art and science of social research (2nd ed). New York, NY: W. W. Norton & Co.
Groves, R. M., Fowler, F. J., Couper, M. P., Lepkowski, J. M., Singer, E., & Tourangueau, R. (2009). Survey methodology (2nd ed.). Hoboken, NJ: Wiley, pp. 259 273.
