This paper experimentally examines whether the decoy, which is dominated by the target but not by an outside option, affects choices differently when using different types of products.
Intervention
We experimentally study the dominance effect using vertically and horizontally differentiated products. To do this, we run a laboratory experiment and let subjects to choose when the target, decoy, and money options are available.
Experimental design
The experiment consists of two tasks: (i) product evaluation task and (ii) target evaluation task. Between the tasks, we have an intermediate step called the product pair selection step. In the product evaluation task, we elicit subjective quality values of consumer products in dollars. In the product pair selection step, based on the elicited quality values, we select one pair of vertically differentiated products and another pair of horizontally differentiated products. In the target evaluation task, we measure the values of the target in dollars when the decoy is absent and when it is present. Using the measures of the targets, we can study the dominance effect. See the attached file in the Analysis Plan section for the details of the experimental design.
Design details
Randomization method
Randomization is done by the computer server that hosts the experiment.
Randomization unit
In the product evaluation task, a subject faces 10 rounds, and each round consists of 50 sub-rounds. The order of the rounds is randomized at the subject level, but the order of the sub-rounds is not randomized. In the target evaluation task, a subject faces 6 rounds, and each round consists of 80 sub-rounds. Likewise, the order of the rounds is randomized at the subject level, but the order of the sub-rounds is not randomized. In the payment stage, one round of the tasks is randomly selected with equal probability. Given the selected round, one sub-round is randomly selected with equal probability. The final payment is determined based on the choice of the selected sub-round.
Clusters
Based on the simulation results, including power analysis, the target number of subjects is 150 students. The details of the simulation are presented in the attached file called "Analysis Plan."
Observations
The main variables are "switch points of the target option." Each subject reports 6 switch points of the target option in total. Thus, for the main analysis, we expect to observe 900 switch points.
Arms
Based on the simulation results, including power analysis, the target number of subjects is 150 students. See the attached file in the Analysis Plan section for the details of the simulation.
Primary outcomes
Switch point of the target option without a decoy option
Switch point of the target option with a decoy option
Cited by
No citation data available for this registration yet.