One significant limitation in research extending the construct of psychopathy to youth has been the absence of longitudinal studies testing the stability of psychopathic traits prior to adulthood. To begin to address this limitation, the current study estimated the stability of psychopathic traits over a 4 year period in a sample of non-referred children in the third, fourth, sixth, and seventh grades at the first assessment. For parent ratings of psychopathic traits, stability estimates using intra-class correlation coefficients ranged from 0.80 to 0.88 across 2–4 years, with a stability estimate of 0.93 across all four assessments. There were also distinct trends in the patterns of stability found in the sample. Specifically, children rated as being initially high on these traits were more likely to be rated lower at later assessments than was the case for children rated initially low on these traits. Finally, the child's level of conduct problems, the socioeconomic status of the child's family, and the quality of parenting the child received were the most consistent predictors of stability of psychopathic traits. Copyright © 2003 John Wiley & Sons, Ltd.