JSM 2013 Home
Online Program Home
My Program

Abstract Details

Activity Number: 614
Type: Invited
Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
Sponsor: ENAR
Abstract - #307275
Title: Multiple Imputation for Nonignorable Missingness: Evaluating Alternative Nonresponse Bias Adjustment Cells
Author(s): Rebecca Roberts Andridge*+ and Katherine Jenny Thompson
Companies: The Ohio State University College of Public Health and U.S. Census Bureau
Keywords: multiple imputation ; nonignorable nonresponse ; survey data ; nonresponse bias
Abstract:

Distinguishing between ignorable (e.g., missing at random) and nonignorable (missing not at random) missing data is not possible using observed data. Standard implementations of multiple imputation (MI) require assuming that data are missing at random. Andridge and Little (2011) propose MI under a proxy pattern-mixture (PPM) model, providing a simple framework for assessing nonresponse bias with respect to different nonresponse mechanisms, ranging from ignorable to nonignorable. They use the PPM model to quantify the potential bias in survey outcomes, given a predefined set of adjustment cells. We now propose using MI under the PPM model to compare candidate sets of adjustment cells for a survey. In particular we focus on business populations, which are characterized by highly positively skewed populations, and extend the PPM model using a bivariate gamma model more appropriate for such data. This research is motivated by the Service Annual Survey conducted by the U.S. Census Bureau, which uses imputation to account for unit nonresponse. We illustrate our method using empirical data from six years of data collection in ten industries in the health care and transportation sections.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2013 program




2013 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.

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

ASA Meetings Department  •  732 North Washington Street, Alexandria, VA 22314  •  (703) 684-1221  •  meetings@amstat.org
Copyright © American Statistical Association.