Skip to main content
ARS Home » Research » Publications at this Location » Publication #168925
Title: IMPUTATION STRATEGY FOR A HEALTH AND NUTRITION SURVEY.

Author
item SIMPSON, PIPPA - DELTA NIRI
item GOSSETT, JEFF - DELTA NIRI
item JO, CHAN - DELTA NIRI
item GOEL, RAJIV - DELTA NIRI
item PARKER, JAMES - DELTA NIRI
item Bogle, Margaret

Submitted to: Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 4/1/2004
Publication Date: 8/12/2004
Citation: Simpson, P.M., Gossett, J.M., Jo, C.H., Goel, R., Parker, J.G., Bogle, M. 2004. Imputation strategy for a health and nutrition survey [abstract]. Joint Statistical Meetings (JSM) 2004, August 8, 2004, Abstract Number 301691. Available: http://www.amstat.org/meetings/jsm/2004.

Interpretive Summary:

Technical Abstract: The FOODS 2000 was a cross-sectional telephone survey conducted in the spring of 2000 in the Lower Mississippi Delta regions of Arkansas, Louisiana, and Mississippi to assess health and nutrition. Missing data is a common problem in survey data. The purpose of the survey was to investigate the relationships between health and nutrition factors adjusting for control variables such as income, race, weight, and nutritional assistance program participation. It is particularly troublesome to have missing control variables. If we use all available cases, the sample size varies considerably. We investigate single and multiple imputation strategies. The FOODS 2000 is a complex weighted sample. We discuss strategies for incorporating predetermined weights in the imputation. This work was funded under the Lower Mississippi Delta Nutrition Intervention Research Initiative, USDA ARS grant # 6251-53000-003-00D.