Innovative Approaches and Methods to the Study of Children=s Conduct Problems


                                                                   Daniel S. Shaw

                                                             University of Pittsburgh





Daniel Shaw, Ph.D.

Department of Psychology

University of Pittsburgh

210 S. Bouquet Street, 4101 Sennott Square

Pittsburgh, PA 15260-0001

Phone: 412 624-1836

F ax: 412 624-8827



            Innovative Approaches and Methods to the Study of Children=s Conduct Problems In this special issue, the focus is on research that introduces novel approaches and methods for studying the development of children=s conduct problems.  During the past two decades, a consensus has been reached that the development of conduct problems (CP) has multiple pathways and correlates. During this period there has been a marked increase in the number of studies devoted to establishing relationships between risk factors and child CP.  Domains of study have included child (e.g., Bates, Moffitt, Raine), family (Egeland, Patterson, Pettit), peer (Coie, Dishion, Dodge) and broader community (Earls, Sampson, Tolan) factors.  The result has been a rapid increase in our knowledge base of the effects of individual risk factors on children=s CP.  More recently, research in the area has focused on integrating the influence of multiple risk factors within or across domains, investigating both mediating and moderating relationships among risk factors in relation to CP. At the microsocial level, much research has been generated on patterns of reciprocal influences children have with caregivers, and more recently, with peers.  At the macro level, researchers have also spent considerable effort to apply Sameroff=s (1990) transactional perspective to the study of CP, demonstrating bidirectional influences between child and proximal factors over time (Coie & Dodge, 1998; Patterson, Reid, & Dishion, 1992).  Advances in quantitative methods have also permitted researchers to more fully capture microsocial processes within observational assessments (Stoolmiller) and macrosocial processes over time (Brown, Nagin, Methun, Raudenbusch).

The present group of papers provides a sampling of these and other innovative directions within the field.  These papers apply novel methodologies, test the validity of understudied constructs, and apply previously studied risk factors to novel populations or contexts. The development periods studied range from infancy to adolescence, and the domains of influence include child attributes, family relations, and peer friendships. A focus on mechanisms that underlie social processes in the development or maintenance of conduct problems is a common thread.  Individually the papers represent the state-of-the art in developmentally-guided social research.  Cumulatively, they should prove to be of considerable heuristic value, encouraging researchers in the field to expand their own conceptual frameworks and methods for pursuing related research.

New Methods for Studying Social Processes

The authors of the first two papers have generated creative methods for capturing ongoing social relationships youth have with peers and parents, respectively.  Granic and Dishion use a dynamic systems framework to examine the influence of deviant talk in adolescent friendships on youth=s later antisocial activities.  Extending the earlier work of Dishion and Owen (2002), they find that an increasing slope of deviant talk during a peer interaction predicts serious antisocial behavior and drug abuse three years later, even after accounting for the duration of deviant talk during the peer observation and the youth=s current problem behavior.  The methodology introduced provides a significant advance for those who use microsocial coding procedures.  Instead of limiting analyses to computing mean rates of target behaviors between individual dyads, the authors compute the slope of deviant talk within the observation. This provides a more proximal estimate of the dyad=s mutual interest in antisocial activities.   

The paper by Snyder and colleagues also raises the >methodological bar,= this time in the area of emotion regulation.  Despite the plethora of recent interest on this topic (e.g., Thompson, 1994) and its relation to different types of child psychopathology, methods used to date have not fully captured the richness of ongoing interchanges between children and caregivers.  Applying hazard rate analyses, the authors describe a method for using observational data to assess how children=s ability to down-regulate anger is moderated by parental expression of negativity.  Children=s ability to regulate anger is also related to chronic covert antisocial behavior.  Although the substantive findings are again of interest, the method used to assess emotion regulation and its moderation by parental behavior, should be of interest to those with microsocial data looking for ways to better optimize its value.  The methodologies used in the Granic and Snyder papers could easily be applied to other dyads and across developmental periods.  For example, Granic and Dishion=s method could be used in the study of parent-child relationships in early or middle childhood.  Similarly, Synder=s use of hazard analysis could be extended to the study of peer relationships on the playground or teacher-child relationships in the classroom.

New Facets of Parenting

Three of the papers delve into relatively untapped domains of parenting in relation to CP.  Following in the footsteps of Patterson=s (1982) work on coercion, most research in this area has focused on negative aspects of parenting, including hostility, inconsistency, control, and neglect.  Gardner=s work, in contrast, has been dedicated to studying the infrequency of positive interaction among mothers and preschool children with CP, most notably the limited use of proactive strategies (Gardner, Sonuga-Barke, & Sayal, 1999).  In the current paper, she and colleagues draw attention to a another context for observing quality of parent-child interaction: play.  Using unstructured observations measured in the home environment, the authors demonstrate that the amount of time spent in joint play at age 3 predicts improvement in conduct problems at age 4, accounting for the influence of initial child behavior, negative parent-child interactions, and other indices of family stress.   As Gardner has noted elsewhere, it is ironic that most researchers interested in parenting devote much of their time to studying bouts of negative interaction that occur relatively infrequently even among families with disruptive children.

Following a similar path, Criss and Shaw examine parent-child synchrony in relation to conduct problems among a sample of low-income, school-age boys.  Until recently, synchrony was a construct that had been applied predominantly by researchers studying parent-child relationships in early childhood, and for the most part in relation to children=s developing competencies.  Using a revealed differences discussion task, the authors find that low levels of parent-child synchrony are associated with youth antisocial behavior and that of their best friends, after controlling for prior youth adjustment and other child and parenting factors.  The use of such an approach provides an  observationally-based alternative for assessing congruence in communication style and affective ties shared by parents and youth at a molar level.  Similar to the work of Gardner, the findings stress the significance of the lack of positivity shared between parent-child dyads in relation to youth=s antisocial behavior.

Laird and colleagues provide a new perspective on a previously researched construct: monitoring.  Aspects of involvement in and supervision of children=s activities have been established as a correlate of youth antisocial behavior, most notably during adolescence. Research in this area has recently been concerned with reciprocal relations in monitoring, particularly the role played by youth (Stattin & Kerr, 2001).  The focus in Laird and colleagues= paper is on youth=s perception of parents= monitoring knowledge.  They find that greater concurrent levels of monitoring knowledge to be associated with lower rates of antisocial behavior and greater enjoyment in parent-youth relationships.  Further, the relation between youth antisocial behavior and monitoring knowledge is found to be partially mediated by parent-youth interaction quality.  Similar to the findings of  Gardner and Criss papers, it suggests that the absence of closeness and positivity in parent-child relationships plays a significant role in the development and maintenance of children=s CP.  Note that this convergence was evident despite the range in the age of children studied, which included preschool, school-age, and adolescent youth.

Novel Contexts

The final two papers investigate previously researched constructs, hostile parenting

and children=s social information processing, but do so in novel ways.  Scaramella and Conger provide a new perspective on hostile parenting, assessing its continuity across two generations of parents.  Moderate continuity in hostile parenting was found between generation one (G1), assessed during adolescence, and generation two (G2), assessed in early childhood.  G2 hostile parenting was in turn, related to early child CP in generation three (G3).  As intergenerational studies with both prospective and observational data on parenting are rare, the finding on parenting is notable in and of itself.  However, by applying a bidirectional framework, the authors further demonstrate that relations between G2 parental hostility and G3 conduct problems are moderated by G3 negative emotionality in early childhood.  The study provides novel data on family processes involved in maintaining continuity in antisocial behavior across generations, and the value of examining developmentally-appropriate risk factors across domains over time.

Finally, Schultz and Shaw trace the origins of biases in social information processing (SIP).  Long established as a consistent predictor of CP, few studies have explored the antecedents of SIP beginning in early childhood.  Findings indicate that children=s maladaptive response generation at school age is predicted by a history of maternal depressive symptomatology and socioeconomic disadvantage.  Children more likely to generate aggressive, retaliatory responses were also more likely to report higher rates of antisocial behavior at age 10, consistent with previous research on SIP.  Whereas much data have been gathered on direct linkages between risk factors and child CP, fewer studies have been directed at uncovering the developmental antecedents of proximal risk factors such as SIP and other common targets of study C hostile parenting, deviant peer affiliation, and child emotion regulation.  The Schultz paper provides an example of a structure from which other researchers with longitudinal data could examine such precursors.

Implications for Future Work

As in most research in the area, these papers offer few definitive answers.  They do, however, provide signposts for future work.  Beginning with the more quantitatively- oriented initial papers, Granic and Snyder acknowledge that their work represents a first attempt at applying an innovative method.  They plan to continue to pursue refinement of these approaches and hopefully their efforts will inspire others to do the same.  The work on joint play, synchrony, and youth=s knowledge of parental monitoring also represent first efforts at using these constructs to better understand CP across childhood.  These studies elicit more questions than they resolve, but provide substantive stepping stones for further inquiry.  Intergenerational data sets are rare, particularly those that employ prospective and observationally-based methods.  The Scaramella paper thus provides a unique perspective on parenting and child effects across generations.  This study too represents only a first glimpse, a snapshot, of an unfolding story.  It will be interesting to see if the pattern of findings remains similar as more G3 children arrive and move into the preschool and school-age periods.  Many of the results of the Schultz paper are consistent with theories on the development of SIP; however, these findings and some of the null results raise issues .  For instance, why are maternal depression and socioeconomic disadvantage related only to child maladaptive response generation and not hostile biases in SIP?  Would the same associations be found for girls? Are there other factors that underly the relation between poor maternal well being and low socioeconomic status and children=s aggressive response style?  Future research can hopefully provide us with a more proximal perspective on mechanisms underlying these associations.

In closing, let me extend my appreciation to all of the contributing authors and in particular, Frances Gardner and Jim Snyder.  Drs. Gardner and Snyder provided their own contributions to the Special Issue, but also each served as the Action Editor on one of the manuscripts. I greatly appreciate their contributions.




Coie, J.D., & Dodge, K.A. (1998).  Aggression and antisocial behavior.  In W. Damon (Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp. 779-862). New York: Wiley.

Dishion, T. J., & Owen, L. D. (2002). A longitudinal analysis of friendships and substance use: Bi-directional influence form adolescence to adulthood. Developmental Psychology, 38(4), 480B491.

Gardner, F., Sonuga-Barke, E., and Sayal, K. (1999). Parents anticipating misbehavior: An observational study of strategies parents use to prevent conflict with behavior problem children. Journal of Child Psychology and Psychiatry, 40, 1185-1196.

Patterson, G. R (1982).  Coercive family processes (Vol. 3).  Eugene, OR:  Castalia.

Patterson, G.R., Reid, J.B., & Dishion, T.J. (1992).  Antisocial boys.  Eugene, OR: Castalia. Publishing Company.           

Sameroff, A. J. (1990).  Prevention of developmental psychopathology using the transactional model: Perspectives on host, risk agent, and environment interactions.  Paper presented at the Conference on the Present Status and Future Needs of Research on Prevention of Mental Disorders, Washington, DC.

Stattin, H., & Kerr, M. (2001). Parental monitoring: A reinterpretation.  Child Development, 71, 1072-1085.

Thompson, R. A. (1994).  Emotion regulation: A theme in search of definition.  In The Development of Emotion Regulation, Monographs of the Society for Research in Child Development, Fox N, ed: 59, No. 2-3, 25-52.