The attraction effect predicts the increase of the choice probability of an alternative (the target), caused by the introduction in the choice set of a new alternative (the decoy) that is dominated by the target but not by the competing alternatives. While the attraction effect has received a great interest from practitioners and academics alike and has been highlighted in laboratory and survey settings, there is a debate on the ecological validity of these studies, questioning the relevance and economic significance of the effect.

We test the internal and external validity of the attraction effect in travel booking, with a laboratory experiment reproducing the incentive structure of travel booking (a delayed time-money trade-off), and a large-scale field experiment on a flight aggregator. In those experiments, we test the effect of introducing or highlighting a decoy alternative that is asymmetrically dominated by the cheapest or the fastest travel in the choice set.

We control for individual characteristics in the laboratory, and for session characteristics in the field experiment and obtain mixed results: the internal validity of the attraction effect is weak in the laboratory and we do not find evidence for the attraction effect in the field.

Co-authors : Zakaria Babutsidze, Thierry Delahaye, Rodrigo Acuna-Agost, Nobuyuki Hanaki

Practical information