JSM 2011 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Abstract Details

Activity Number: 305
Type: Contributed
Date/Time: Tuesday, August 2, 2011 : 8:30 AM to 10:20 AM
Sponsor: Section on Survey Research Methods
Abstract - #301460
Title: Nonresponse Adjustment for a Vector of Outcomes and the MAR Assumption
Author(s): Joel Wakesberg*+ and Karen Christine Masken and J. Michael Brick
Companies: Westat Inc. and IRS Research, Analysis, and Statistics and Westat Inc.
Address: 1600 Research Blvd , Rockville, MD, 20850,
Keywords: nonresponse ; missing data mechanism ; vector of outcomes ; raking
Abstract:

The IRS National Research Program conducts annual studies of individual taxpayer compliance based on a stratified random sample. As with most studies, not all of the selected taxpayers participate in an audit, resulting in nonresponse that could be due to various reasons including missing returns, unlocatable taxpayers, or taxpayers who never respond to any IRS correspondence. Historically, the IRS has adjusted the respondents weights using assumptions about the missing data mechanism based on the reason for nonresponse. This approach deals with one important dimension and addresses total noncompliance, the key estimate from the survey. However, some single line item entries such as self-employed business income or charitable contributions are of great interest to analysts as well and the adjustment may be less effective for these items. In our paper, we review the rationale for treating all nonresponse as missing at random. We then explore ways to adjust for the nonresponse bias when analysts are interested in a vector of estimates, not just one point estimate. Finally, we demonstrate that raking to multiple variables is very effective as compared to the traditional approach.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2011 program




2011 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.