|
Activity Number:
|
519
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #303939 |
|
Title:
|
Evaluation of Alternative Imputation Methods for the Public Libraries Survey
|
|
Author(s):
|
Irene Brown and Terri L. Craig*+
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
4600 Silver Hill Road, Washington, DC, 20233,
|
|
Keywords:
|
Imputation ; Cell boundaries
|
|
Abstract:
|
The Public Libraries Survey (PLS) imputes for missing data. Although the unit response rate is very high, there is still item nonresponse that requires imputation. PLS uses a variety of imputation methods depending on the item that is missing data. The purpose of this study is to evaluate current imputation methods along with some new alternatives. The imputation methods being evaluated are ratio, cell mean, adjusted cell mean, hot-deck, and combinations thereof. Another objective is to minimize the number of imputation methods being used. The study also compares current methods of determining the imputation cell boundaries to new methods using the population variable.
|
