|
Organizers |
Dealing with missing data
by
Judi Scheffer
Massey University
Missing Data will occur in Censuses and Surveys even with the most careful data collection and the most rigorous follow up. How the missing data is dealt with can cause the difference between a biased study and an unbiased one. Weighting can be used to deal with unit nonresponse, but I will focus on item nonresponse.
There are a variety of useful methods for dealing with item nonresponse, the most common still being case deletion. Imputation techniques are many and varied, and the choice of method depends firstly on the type of data, and then the missingness mechanism. Different methods of imputation are needed for different circumstances, and I will discuss some in this talk.
Date received: September 3, 2001
Copyright © 2001 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cahg-88.