ccs: Connecticut child surveys
In ALA: Data sets and examples for the book "Applied Longitudinal Analysis"

Description Usage Format Details Note Author(s) Source References Examples

The data are from two surveys of children's mental health. A standardized measure of childhood psychopathology was used both by parents (Child Behavior Checklist, CBC) and teachers (Teacher Report Forms, TRF) to assess children in the study.

data(ccs)

A data frame with 2501 observations on the following 5 variables.

id: a factor with 2501 levels
problems: a factor with levels good and bad
parent.status: a factor with levels otherwise and single
respondent: a factor with levels parent and teacher
behaviour: a factor with levels no and yes

The response variable is derived from the externalizing scale, which assesses delinquent and aggressive behavior. The scale has been dichotomized at the cut point for borderline/clinical psychopathology.

Because of the multiple levels of permissions and reporting, a substantial number of children were missing the TRF. The data set consists of 1428 children who had both parent and teacher responses, and an additional 1073 children with only a parent response; a total of 2501 children participated in the study. The two sources or respondents are the children's parents and teachers; in the psychiatric literature, these sources are often referred to as “informants”. Explanatory variables include single parent status and child's physical health problems.

Original variable names have been adapted to R conventions. Data were reshaped from wide to long format and the response named behaviour. Factor problems level bad corresponds to “fair to bad” in Fitzmaurice's et al. source.

Sebastian P. Luque spluque@gmail.com

http://biosun1.harvard.edu/~fitzmaur/ala

Zahner GEP, Jacobs JH, Freeman DH and Trainor KF (1993) Rural-urban child psychopathology in a northeastern U.S. state: 1986–1989. Journal of the American Academy of Child and Adolescent Psychiatry 32:378-387.