Psychological Distance and the Discrepancy Between Recommendations and Actions

Abstract

Action hypocrisy is the tendency to recommend behavior for others that one would personally be unwilling to undertake. Six studies examine the relationship between action hypocrisy and psychological distance. Studies 1a and 1b and 2 demonstrate action hypocrisy in three populations and in 2 different contexts. Studies 3 to 5 support a psychological distance explanation for action hypocrisy: people are more likely to choose action for themselves in the future versus the present (Study 3), for someone at a distant location versus a nearby location (Study 4), and for someone who appears to be distant versus close in a photograph (Study 5).

Notes

¹Participants reported their races to be 34% Hispanic/Latino, 34% Asian, 11% White/Caucasian, 7% Middle Eastern, 5% Black/African American, 4% Hawaiian/Pacific Islander, and 5% other. As such, this study serves also to suggest that action hypocrisy is not a primarily White American phenomenon.

²We also included the option “I am in favor of comprehensive exams, and therefore am not interested in taking any action” and removed from analysis all participants selecting this option.

³We compared responses to the midpoint of the scale because the midpoint represented a neutral response indicating that the participant did not agree or disagree with the statement. As such, a significant positive one-sample t test indicates that, on the whole, participants agreed with the statement.

⁴Because no participant personally chose Option 1, this effect size is likely an overestimate. However, the effect remained when considering participants' decisions as a continuous 3-point, scale, t(86) = 8.54, p < .001, d = 1.91.

⁵Although one might wonder whether people closer to Louisiana (e.g., in bordering states like Texas or Mississippi) displayed less action hypocrisy than those farther away (e.g., in California or New York), we lack sufficient data to address this question for two reasons. First, only 14 states are represented in both conditions (Mississippi is not one of them), and many have fewer than six participants. Second, because we collected only state-level data, we have no way of knowing how distant our participants were from Louisiana. For instance, a person living in Texas could be as near as 1 mile and as far as more than 800 miles away from Louisiana.

⁶Although we observed a difference between conditions in fairness, fairness did not mediate the relationship between condition and action intention, bias corrected 95% confidence interval [–1.0, .26].

⁷Linear polynomial contrasts are a set of orthogonal contrast codes that test the basic premise that group means follow a linear trend (e.g., A > B > C; see O'Brien, 1979, for more discussion).