Do Strength-Related Attitude Properties Determine Susceptibility to Response Effects? New Evidence From Response Latency, Attitude Extremity, and Aggregate Indices

A great deal of research has shown that small changes in question wording, format, or ordering can sometimes substantially alter people's reports of their attitudes. Although many scholars have presumed that these so-called response effects are likely to be more pronounced when the attitudes being measured are weak, a number of studies have disconfirmed this notion. This paper presents several new tests of this hypothesis using a variety of measures and analytic techniques. The findings largely replicated previously documented effects and non-effects but also uncovered new effects not previously tested. No single strength-related attitude attribute emerged as a consistent moderator of all response effects. Rather, different individual attributes moderated different effects, and a conglomeration of strength-related dimensions did not emerge as a reliable moderator. Taken together, these results support the conclusions that different response effects occur as the result of different cognitive processes, and that various strength-related attitude attributes reflect distinct latent constructs rather than a single one.