Racial/ethnic and educational differences in the estimated odds of recent nitrite use among adult household residents in the United States: An illustration of matching and conditional logistic regression

This article compares estimates of the relative odds of nitrite use obtaine d from weighted unconditional logistic regression with estimates obtained f rom conditional logistic regression after poststratification and matching o f cases with controls by neighborhood of residence. We illustrate these met hods by comparing the odds associated with nitrite use among adults of four racial/ethnic groups, with and without a high school education. We used ag gregated data from the 1994-B through 1996 National Household Survey on Dru g Abuse (NHSDA). Difference between the methods and implications for analys is and inference are discussed. [Translations are provided in the Internati onal Abstracts Section of this issue.]