Rja. Little et al., ASSESSMENT OF WEIGHTING METHODOLOGY FOR THE NATIONAL COMORBIDITY SURVEY, American journal of epidemiology, 146(5), 1997, pp. 439-449
The authors studied weighting adjustments for the National Comorbidity
Survey (1990-1992), a large-scale national epidemiologic investigatio
n of the prevalence, risk factors, and consequences of psychiatric mor
bidity and comorbidity in the United States. Weighting adjustments for
differential selection within households, new construction, unit nonr
esponse, and poststratification were examined separately and in combin
ation. Specific issues addressed included the magnitude of the bias in
curred from ignoring the weights, the added variance from weighting an
d how well this was predicted by simple formulae, and the performance
of methods for trimming the weights. Weights had quite modest effects
on point estimates of prevalences but resulted in major increases in v
ariance unless trimmed, The weights after trimming and poststratificat
ion appeared to work well. It is suggested that the added variance fro
m weighting be carefully monitored in similar surveys. Alternatives to
the use of trimming for controlling variance are worth exploring.