Behavior Problem Trajectories and First-Grade Cognitive Ability and Achievement Skills: A Latent Growth Curve Analysis Kristen L. Bub, Kathleen McCartney, and John B. Willett Harvard University Using data from the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development, the authors investigated whether there was evidence of intraindividual stability in behavior problems over time as well as whether children with higher levels of behavior problems at 24 months and more rapid increases in behavior problems prior to school entry performed more poorly on 1st-grade tests of cognitive ability and achievement than their peers. Three findings were noteworthy. First, there was evidence of both intraindividual and interindividual variability in behavior problems between 24 months and 1st grade. Second, children with higher initial levels of internalizing and externalizing behaviors at 24 months had lower cognitive ability and achievement scores in 1st grade. Finally, children with more rapid increases in internalizing behaviors over time had lower cognitive ability scores in 1st grade; this association did not exist for externalizing behaviors. Implications for future research are discussed. Keywords: behavior problems, school performance, longitudinal analyses New evidence from a report released by the Kauffman Early Education Exchange (2002) suggests that as many as half of the children entering kindergarten classrooms are not ready to learn, in part because they do not have well-developed cognitive and/or social skills. Children with behavior problems, marked by aggres- sion, disruptiveness, anxiety, and social withdrawal, face particular adversity. In fact, children with higher levels of behavior problems tend to do less well in school (Arnold, 1997; McClelland, Morri- son, & Holmes, 2000), have lower verbal and reading skills (Bow- man, Donovan, & Burns, 2001; Snow, Burns, & Griffin, 1998), and receive more negative feedback from their teachers (Arnold et al., 1999) than their more behaviorally competent peers. The majority of research on behavior problems has focused on children in elementary school or beyond; there is only a small body of literature on the development of behavior problems prior to ele- mentary school. It is important to note that little is known about the links between early changes in behavior problems and subsequent elementary school outcomes, or whether the effects differ across a variety of outcomes. In the present study we addressed these gaps by investigating whether there is within- and between-individual stability in behav- ior problems over time and whether growth trajectories in behavior problems from 2 years of age through first grade are related to early school performance. Of particular interest was whether chil- dren with higher initial levels of behavior problems at 24 months and more rapid increases in behavior problems prior to school entry perform more poorly on tests of cognitive ability and achievement skills in first grade than their peers with average or below average behavior problems at 24 months and minimal or decreasing behavior problems over time. We also tested whether the pattern of findings differed across our two outcomes: chil- dren’s cognitive ability and achievement skills. Behavior Problems in Early Childhood: Prevalence and Stability Research on children’s behavior problems has commonly fo- cused on two specific classes of behaviors: those characterized by overcontrol, referred to as internalizing behaviors, and those char- acterized by undercontrol, referred to as externalizing behaviors (Hinshaw, 1992). Internalizing behaviors are typified by self- focused expressions of distress and are marked by social with- drawal, anxiety, and fearfulness (Achenbach & Edelbrock, 1978; Campbell, 2002). In contrast, externalizing behaviors are ex- pressed against others and are marked by aggression, impulsivity, disruptiveness, and defiance (Achenbach & Edelbrock, 1978). Although the vast majority of behaviors exhibited by young chil- dren are thought to represent normal developmental stages or transgressions, and most children are expected to outgrow them (Campbell, 2002; Mathiesen & Sanson, 2000), some patterns of behavior are cause for concern. In general, within-individual be- haviors are considered serious if they are somewhat stable over time and if they disrupt the child’s ability to negotiate develop- mental events (Campbell, 1995, 2002). It is difficult to determine with absolute certainty the prevalence of behavior problems among young children, in part because Kristen L. Bub, Kathleen McCartney, and John B. Willett, Harvard Graduate School of Education, Harvard University. This project was funded by Grant HD25451 from the National Institute of Child Health and Human Development (NICHD) to Kathleen McCart- ney. We thank Clara Barata, Daniel Berry, Pia Caronongan, Suzanne Graham, and Emily Mann for their helpful comments on earlier drafts of this article. We also thank the investigators in the NICHD Early Child Care Research Network for the data set, the site coordinators and research assistants for their data collection efforts, and the children and their families for their participation in this longitudinal study. Correspondence concerning this article should be addressed to Kristen L. Bub, Harvard Graduate School of Education, 715 Larsen Hall, Appian Way, Cambridge, MA 02138. E-mail: bubkr@gse.harvard.edu Journal of Educational Psychology Copyright 2007 by the American Psychological Association 2007, Vol. 99, No. 3, 653– 670 0022-0663/07/$12.00 DOI: 10.1037/0022-0663.99.3.653 653