Implementation and Results of an Experiment Using Mahalanobis Distance in Responsive Design to Reduce Nonresponse Bias

• Activity Number: 288
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Jennifer Cooney*+ and Peter Siegel and Melissa Cominole and Bryan Shepherd
• Companies: RTI International and RTI International and RTI International and RTI International
• Keywords: Mahalanobis Distance ; Nonresponse Bias ; Responsive Design

Abstract

Objective: Test whether implementation of a Mahalanobis distance measure as part of a responsive design reduces nonresponse bias in the 2008/2012 Baccalaureate and Beyond Longitudinal Study.

Methods: Prior to data collection, the entire sample was randomized to the treatment group or the control group. At three different time periods, the Mahalanobis distance value was calculated, a cut point was determined, and the sample was divided into high distance and low distance. The high distance cases who were nonrespondents and also randomized to the treatment group were given a data collection treatment. The purpose of the treatment was to increase the likelihood that a person would respond to the survey and thus help to reduce bias.

Results: Data collection treatments did significantly increase response rates for the treatment group versus the control group. Two survey outcomes were significantly different between treatment and control. Nonresponse bias estimates were not significantly different but did move in the expected direction.