|
Abstract:
|
Randomized control trials (RCTs) have been the gold standard to evaluate a new treatment or policy. However, many RCTs assume that study units are willing to share their (potentially sensitive) data, specifically their response to treatment. This assumption, while trivial at first, is becoming difficult to satisfy in the modern era, especially in online settings. The paper presents a new, simple experimental design that is differentially private, one of the strongest notions of data privacy. Also, using works on noncompliance in experimental psychology, we show that our design is robust against "adversarial" participants who may distrust investigators with their personal data and provide contaminated responses to bias the experiment. Under our new design, we propose unbiased and asymptotically Normal estimators for the average treatment effect. We also present a doubly robust, covariate-adjusted estimator to improve efficiency. We conclude by using the proposed design to evaluate online statistics courses at the University of Wisconsin-Madison during the Spring 2021 semester, where many classes went online due to COVID-19. This is joint work with Manjusha Kancharla.
|
