The stability of child behavior disorders: a one year test-retest study of Adelaide versions of the Conners Teacher and Parent Rating Scales.

The stability of children's disorders was studied, using Adelaide versions of the Conners teachers and parent rating scales, with 5- to 12-year-olds. Of the 20 scales, 14 had high or moderate test-retest stability over the 1-year interval. Most scales also showed discriminant validity over this time. All but 2 of the 14 stable scales gave moderate stability coefficients at each of three age levels. Three teacher scales (Conduct Problem, Socially Rejected, and Antisocial) gave very low stability for the youngest group. Stability was not generally affected by teacher "practice." Little effect of subject selection or repeated-rating bias on scale means was evident. There was a general reduction of some scale means on the second assessment, and the contributions of practice, age, and study-entry effects were examined.