Question 5 of my final exam for Design and Analysis of Sample Surveys

5. Which of the following better describes changes in public opinion on most issues? (Choose only one.)

(a) Dynamic stability: On any given issue, average opinion remains stable but liberals and conservatives move back and forth in opposite directions (the “accordion model”)

(b) Uniform swing: Average opinion on an issue can move but the liberals and conservatives don’t move much relative to each other (the disribution of opinions is a “solid block of wood”)

(c) Compensating tradeoffs: When considering multiple survey questions on the same general topic, average opinion can move sharply to the left or right on individual questions while the average over all the questions remains stable (the “rubber band model”)

Solution to question 4

From yesterday:

4. Researchers have found that survey respondents overreport church attendance. Thus, naive estimates from surveys overstate the percentage of Americans who attend church regularly. Does this have a large impact on estimates of time trends in religious attendance?

Solution: Yes. See this article by Hadaway, Marler, and Chaves, who write, “We suspect that the actual attendance rate has declined since World War II, despite the fact that the survey rate remained basically stable.”

13 thoughts on “Question 5 of my final exam for Design and Analysis of Sample Surveys

  1. On question #4, you don’t provide any idea of whether there’s variation in the over-reporting. If the amount of over-reporting is constant, then there’s no bias.

  2. In this context “better” (i.e. vs best) makes me cringe, though the more I think about it the more it seems the better word to use however strange it sounds.

  3. My pick would be #3. Although I imagine all of this is very disputable and very probably not generalizable over all issues. So far, the degree of subjectivity on your exam questions is pretty amazing to me.

    • DK:

      I aim to amaze. Also, it’s a political science class not a statistics class. Also, the exam covers topics we’d already discussed in class.

    • There’s lots of polling data about lots of issues. If one of these models really does apply most of the time then I don’t see what’s subjective about the question.

  4. I am gonna guess b? That’s the only one that makes intuitive sense to me based on my knowlege of how opinion on say, gay marriage, has changed over time.

    On the other hand I know the country has become more polarized over the years so not sure if B makes sense in light of that…

  5. Pingback: Question 6 of my final exam for Design and Analysis of Sample Surveys « Statistical Modeling, Causal Inference, and Social Science

  6. I’m completely naive about the subject, but I’m curious about the following: does what counts as the best model interact at all with the amount of time you look at?

  7. 4. Interesting result from Hadaway, et al. I was treating p(say you attend church | don’t attend church) as a constant, which results in near stability of the trend. If anything, I would have thought the need for the socially desirable response would decline once church attendance declines as a social practice, and p(say you attend church | don’t attend church) would therefore decline. But instead it increases.

  8. I just read an article based on the Pew report on response rates (which notes the response rate on telephone surveys is now 9%, down from 36% in 1997) also notes “survey participants tended to be significantly more engaged in civic activity than those who do not participate, suggesting that telephone surveys could lead to overestimates of behaviour such as church attendance, contact with elected officials and attendance at campaign events.”

    http://www.research-live.com/4007408.article

  9. This is a point people who do polling on climate change don’t get. They get all excited if the % saying they believe climate change is happening etc. goes up a few ticks & ignore that the degree of partisan division is not changing in any meaningful way. It’s the change in correlation with partisan self-identification in the responses, not the intercept, that tells you whether any headway is being made in dissolving political deadlock.

Comments are closed.