The impact of a written parental consent policy on estimates from a school-based drug use survey.

The authors examine differences between mean, variance, and correlation parameter estimates derived from a full school-based sample and subsamples restricted by the provision of parental consent. A total of 1,607 students at 21 continuation high schools and 1,192 students at 3 traditional high schools completed a survey containing variables related to sociodemographics, drug use, mental health, and violence. The employment of a researcher-initiated home-telephone-call procedure substantially increased the parental response rate over a student-/school-assisted consent method. The subsamples restricted by the written consent criterion showed some small biases in estimates of sociodemographic variables but little or no biases on measures related to mental health, drug use, or violence measures. The augmentation of the written consent samples with verbally consented students reduced observed biases.