Psych 458 - Questions to ponder
4/19/99
Greenwald, Carnot, Beach, & Young, 1987:
1.What do Greenwald et al. mean when they refer to self-erasing errors of prediction?
2. Besides the effect of prediction, in what other ways might Greenwald et al.=s phone calls have produced a greater number of registered voters (for both predicters and non-predicters)?
3. What was the effect of prediction on voter registration rates? Was the effect significant? What about the effect on voting rates?
4. What are some possible limitations on the generalizability of the results? Besides voting behavior, in what other contexts could you apply Greenwald et al.=s (and Sherman=s) findings? Do you think prediction might also increase the performance of UNdesirable behaviors? Why or why not?
5. AThe effect is small, but it could be of great importance@ -- why is this so?
6. Why do the authors suggest that registration may be less desirable to the subjects in Study 1 in the first study? Why else do they think this group might be less likely to perform the behavior?
Houston, Doan, & Roskos-Ewoldsen, in press
7. In the first study, how did Houston et al. manipulate whether there was a positive campaign or a negative campaign? In addition to the kind of campaign, what else about the candidate varied?
8. What affected the ratings of the shared ideology candidate? Was there an interaction effect and if so, what was it?
9. Did the valence (i.e., positive or negative) of the opposing ideology candidate=s campaign affect the ratings of the opposing ideology candidate? How so? Did valence of the shared ideology candidate=s campaign affect the ratings of the opposing ideology candidate? How so? Was there an interaction?
10. What do Houston et al. mean when they talk about Asynergy@ in campaigns? According to their results, can we see the effects of negative campaigning by looking at just one campaign? (This question relates to the key, novel finding of this article.)
11. Under what particular circumstances does negative campaigning appear to affect people=s likelihood of voting? Do the authors find an interaction effect in the likelihood results? Would you expect one?
12. How does the strength of messages from Aoutgroups@ (groups we aren=t a part of) compare to the strength of messages from Aingroups@ (groups we are a part of)? Under what circumstances is this general pattern changed?
13. What change do Houston et al. make from Study 1 to Study 2?
14. Which results from Study 1 were replicated in Study 2?
15. Which candidate appears to be most affected by the absence of explicit evidence for campaign claims? How does the absence of explicit evidence affect likelihood to vote? Why do these effects occur?