Are effects of travel feedback programs correctly assessed?

It has been claimed that travel feedback programs (TFPs) are effective in reducing levels of car-use related congestion, noise, and air pollution. This article examines this claim, noting that many evaluation studies have employed less than optimal research designs. Demonstrating the use of meta-analysis, the results of 15 Japanese TFPs show that the effect sizes estimated for the frequently used research design lacking adequate control groups differ from the effect sizes estimated for research designs including adequate control groups. In addition, estimates of the homogeneity of treatment effects appear to differ, thus suggesting that inferences of causes of the effectiveness of a TFP vary with research design.