When people are asked to make comparative judgments, the direction of comparison has been found to influence the result of the comparisons. Comparing A to B often does not yield comparable results from comparing B to A. There are a number of ways comparative questions can be worded, but is there a more robust one? Our results suggest that there may be.
In this study, questions about whether industry or traffic contribute more to air pollution, whether TV or newspaper does a better job reporting news, and whether other people or oneself has more luck in life were asked in three formats. In the unidirectional comparison format, one of the compared objects served as the subject of the comparison while the other served as the referent (i.e. Does industry contribute more or less to air pollution than traffic?). In the balanced directional comparison format, two directional comparisons were presented such that both objects had a chance to be the subject or referent of the comparisons (i.e. Does industry contribute more to air pollution than traffic, traffic contribute more to air pollution than industry, or do they contribute about the same to air pollution?). In the third, non-directional comparison format, a comparison is framed such that neither objects clearly served as the subject or referent of the comparison (i.e. Which of the following two sources contribute more to air pollution, traffic or industry?). Within each format, two versions were constructed to control for order effects.
250 college students participated in an online experiment. Participants were randomly assigned to each of the six conditions when they logged on to the web site.
Preliminary analysis of the data partially supported previous findings of direction of comparison effect. The data also suggest that non-directional comparison yields less biased results than the other two formats.