Does it always help to adjust for misclassification of a binary outcome in logistic regression?
Luan-X; Pan-W; Gerberich-SG; Carlin-BP
Stat Med 2005 Jul; 24(14):2221-2234
It is well known that in logistic regression, where the outcome is measured with error, a biased estimate of the association between the outcome and a risk factor may result if no proper adjustment is made. Hence, it seems tempting to always adjust for possible misclassification of the outcome. Here we show that it is not always beneficial to do so because, though the adjustment reduces the bias, it also inflates the variance, leading to a possibly larger mean squared error of the estimate. In the context of a data set on agricultural injuries, numerical evidence is provided through simulation studies.
Agricultural-industry; Analytical-methods; Analytical-models; Analytical-processes; Injury-prevention; Mathematical-models; Qualitative-analysis; Quality-standards; Risk-analysis; Risk-factors; Statistical-analysis; Statistical-quality-control;
Author Keywords: bias; logistic models; mean squared error; measurement error; sensitivity and specificity; simulation
Bradley P. Carlin, Division of Biostatistics, University of Minnesota, School of Public Health, Box 303 Mayo Memorial Building, 420 Delaware St., Minneapolis, MN 55455-0378, U.S.A
Statistics in Medicine
University of Minnesota Twin Cities