Exploration of Data Quality in Work with Administrative Records and Sample Surveys

• Activity Number: 287
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Section on Government Statistics
• Author(s): Daniell Toth*+ and Polly Phipps
• Companies: Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: Recursive partitioning ; non-ignorable nonresponse ; propensity model ; establishment survey ; Classification and Regression Trees (CART)

Abstract

To gain insight into how characteristics of an establishment affect nonresponse, a recursive partitioning algorithm is applied to the Occupational Employment Statistics May 2006 survey data to build a regression tree. The tree models an establishment's propensity to respond to the survey given certain establishment characteristics. It provides mutually exclusive cells based on the characteristics with homogeneous response propensities. This makes it easy to identify interpretable associations between the characteristic variables and an establishment's propensity to respond; something not easily done using a logistic regression propensity model. A linear representation of the tree model is used to test the model obtained using the May data against data from the November 2006 Occupational Employment Statistics survey. This test, on a disjoint set of establishment data, gives compelling evidence that the tree model accurately estimates the response rate of establishments. This representation is then used along with frame-level administrative wage data linked to sample data to investigate the possibility of nonresponse bias. We show that there is a risk that the nonresp