Learning About Nonrespondents' Characteristics Using Standard Exploratory Data Analysis (EDA) Tools

• Activity Number: 617
• Type: Contributed
• Date/Time: Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Korean International Statistical Society
• Author(s): MoonJung Cho* and Larry Lang
• Companies: Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: Classification and regression tree ; Classical multiple dimensional scaling (MDS) ; Survey nonresponse ; Spectral clustering ; U.S. International Price Program Survey

Abstract

We examined the characteristics of non-respondents using standard EDA tools such as classical multiple dimensional scaling (MDS), spectral clustering, and topological network. In addition, we applied classification and regression tree methods to identify the important variables which could better assess and interpret non-response rates. We applied these tools to analyze non-respondents' characteristics at the initiation in the U.S. International Price Program Survey.