Two studies addressed three research questions, (a) Are behavioral diaries or checklists susceptible to between-ratee contrast effects? The studies found that both diaries and checklists are susceptible to between-ratee contrast effects, (b) Can diaries or checklists help reduce between-ratee contrast effects on corresponding performance evaluations? Study 1 illustrated that diaries not only may be ineffective at reducing contrast effects on corresponding ratings but also may actually strengthen those effects. In Study 2, although checklists compared favorably with diaries in their impact on contrast effects on corresponding evaluations, checklists did not eliminate the contrast effect on evaluations, (c) What are the directional influences of context behavior when contrast effects are observed? Significant contrast effects were consistent with the traditional conception of this effect: upward in a poor context and downward in a good context.

In the 1988 American presidential election, leadership perceptions and perceived platforms were used to predict votes for one of the two final candidates (Governor Michael Dukakis and Vice President George Bush). Both leadership perceptions and perceived platforms were found to be related to voting. However, political involvement moderated the relationship between the perceived platforms and the votes, while leadership perceptions were found to be a consistent predictor of voting across levels of political involvement. Leadership perceptions, perceived platforms, and the interaction between involvement and perceived platforms predicted voting above and beyond the voters’ political affiliations. Consistent with a categorization model of leadership perceptions, the prototypicality of perceived leader traits was found to be strongly related to a tendency to vote for a candidate who was perceived to possess those traits. In comparing general favorability of characteristics versus prototypicality with respect to an effective political leadership category, general favorability played a larger role in uninvolved voters’ decisions than in involved voters’ decisions. Differences in mean leadership prototype ratings were also explored as a function of political affiliation and political involvement. The implications of these findings for campaign strategies and for leadership in organizational contexts are discussed.

No abstract