“What We Can Learn from Five Naturalistic Field Experiments That Failed to Shift Commuter Behavior”, 2019-12-23 (; backlinks; similar):
Across 5 field experiments with employees of a large organization (n = 68,915), we examined whether standard behavioral interventions (‘nudges’) successfully reduced single-occupancy vehicle commutes. In Studies 1 and 2, we sent letters and emails with nudges designed to increase carpooling. These interventions failed to increase carpool sign-up or usage.
In Studies 3a and 4, we examined the efficacy of other well-established behavioral interventions: non-cash incentives and personalized travel plans. Again, we found no positive effect of these interventions.
Across studies, effect sizes ranged from Cohen’s d = −0.01 to d = 0.05. Equivalence testing, using study-specific smallest effect sizes of interest, revealed that the treatment effects observed in 4⁄5 of our experiments were statistically equivalent to zero (p < 0.04).
The failure of these well-powered experiments designed to nudge commuting behavior highlights both the difficulty of changing commuter behavior and the importance of publishing null results to build cumulative knowledge about how to encourage sustainable travel.