Editing data: What difference do consistency checks make?

In 1998, the Florida Department of Health undertook a self-administered sch ool-based survey of tobacco use, attitudes, and behaviors among nearly 23,0 00 public school students in grades 6-12. The survey design did not use ski p patterns; therefore, students had multiple opportunities to contradict th emselves. By using examples from the high school portion (grades 9-12) of t he survey, the authors examined five possible approaches to handling data i nconsistencies and the effect that each has on point estimates. Use of thes e approaches resulted in point estimates of current cigarette use ranging f rom 25.6% to 29.7%. The number of missing respondents varied from 33 (less than 1%) to 1,374 (13%), depending on which approach was used. After strati fication by gender and race, the prevalence estimates changed marginally fo r girls but strikingly for boys. Non-Hispanic White students were substanti ally more likely than non-Hispanic Black students to report current cigaret te use, but the magnitude of this difference varied significantly according to the analytical approach used. The approach used to check data consisten cy may influence point estimates and comparability with other studies. Ther efore, this issue should be addressed when findings are reported.