Home‎ > ‎BRITE Seminar Series‎ > ‎

Jon de Quidt - Dec 9, 2016

Friday, December 9th
(Institute for International Economic Studies

"Measuring and Bounding Experimenter Demand"
Rm 1199 Nancy Nicholas Hall (WI Idea Room)
2:30-3:45 PM

Abstract:

Experimenter demand effects present a threat to external validity that can be difficult to quantify. In this paper we propose an intuitive technique for assessing robustness of behavioral measures and treatment effects to demand. The basic premise is that by deliberately inducing demand in a structured way we can measure its influence and construct reasonable bounds on demand-free behavior. We provide formal restrictions on choice that validate our method, and a simple Bayesian model that microfounds them. Six pre-registered experiments with eleven canonical laboratory games and around 18,000 subjects demonstrate the technique. We additionally illustrate how demand sensitivity varies by task, subject pool, gender, real versus hypothetical incentives, and subject attentiveness, and provide both reduced-form and structural analyses of demand effects.

Comments