This research examined how people’s ability to detect behavior change in simulated child targets is affected by their clinical experience and the assessment method they use. When using summary assessment methods that are widely employed in research and clinical practice, both inexperienced and experienced clinical staff detected changes in the overall frequency of targets’ aggressive behavior, but were not uniquely influenced by changes in targets’ reactions to social events. When using contextualized assessment methods that focused on conditional reactions, experienced staff showed greater sensitivity than novices to context-specific changes in targets’ aggressive and prosocial reactions to aversive events. Experienced staff also showed greater sensitivity to context-specific changes in their overall impressions of change, but only for aggression. The findings show how clinically experienced judges become more attuned to if…then… contingencies in children’s social behavior, and how summary assessment methods may hamper the detection of change processes.

This research examined the independent and interactional contributions of peer experiences and group aggression to youth behavioral adjustment in short-term residential treatment. Participants were 219 youth (M age = 12.70, SD = 2.76; 71 % male) nested in 28 same-age, same-sex treatment groups. Sociometric interviews assessed social preference and victimization. Daily behavioral observations by staff assessed overall levels of treatment group aggression, as well as aggressive, withdrawn, and prosocial responses to specific social events. End-of-summer behavioral responses (to all events; to peers; to adults) were predicted, controlling for initial levels of these responses. Social preference predicted higher end-of-summer prosocial responses, and victimization predicted lower prosocial and higher withdrawn responses. Each interacted with group aggression in some analyses, with more positive peer experiences only predicting more favorable responses in groups that were low or average in aggression. Interactant-specific analyses revealed that some of these associations were broad, whereas others applied only to adults. For example, group aggression moderated the association between social preference and aggressive responses to adults but not peers. Gender differences were also interactant-specific. Results highlight the importance of peer experiences in group treatment and underscore the value of both aggregation and disaggregation over interactants in analyses of behavioral adjustment.

A series of experiments examined how summary assessment measures influence people's ability to detect change in behavior over time and across situations.Two measures that are often used to assess child behavior (Teacher Report Form) and adult personality (Five Factor Inventory) were examined.Each instrument led perceivers to focus on the overall frequency of targets' behavior, even when targets differed both in how they reacted to social events and in how often they experienced those events in their interactions with others.Although people adopted an overall frequency perspective when using summary measures, they detected changes in events and targets' if … then … reactions to events when using alternative context-specific measures.The findings demonstrate how summary trait methods can shift perceivers' attention away from situational factors and thereby yield trait scores that are insensitive to context-specific but potentially important changes in targets' social behavior.

This research examined how a contextual approach to personality assessment can reveal change processes that are obscured by measures of overall behavior frequencies. Using field observations of 336 children from three summers at a program for at-risk youth, we illustrate how children's social experiences change over time, how their reactions to these experiences change, and how both processes contribute to changes in the overall frequencies of their prosocial, aggressive, and withdrawn behavior. Children showing opposite patterns of change in their environments and their reactions to them were nevertheless similar in their overall amount of change. The results clarify how changes in reactions and social experiences can be disentangled and reintegrated in order to deepen our understanding of personality change processes. Implications for change assessments that rely on overall behavior summaries are highlighted for program, individual, and intra-individual levels of analysis.

This research examined deviant talk during summer residential treatment using peer nominations and extensive field observations. Participants were 239 youth (M age = 12.62, SD = 2.60; 67% male), nested in 26 treatment groups. Deviant talk was present in this setting, showed individual differences, and increased over time, especially for younger boys. As expected, its relationship to treatment response was moderated by peer behavior. Initial levels of individual deviant talk were related to clinical improvement, but primarily when peer deviant talk was low. Initial levels of peer deviant talk were related to higher than expected end of treatment aggression, especially for youth who were high in deviant talk. Deviant talk effects were observed for staff impressions of change and observations of aggression and adjustment. Initial antisocial behavior affected whether individual or peer levels of deviant talk more heavily influenced treatment response. Implications for clinical assessment and treatment monitoring are discussed.

This study examined how a contextual approach to child assessment can clarify the meaning of informant discrepancies by focusing on children's social experiences and their if...then reactions to them. In a sample of 123 children (Mage=13.30) referred to a summer program for children with behavior problems, parent-teacher agreement for syndromal measures of aggression and withdrawal was modest. Agreement remained low when informants assessed children's reactions to specific peer and adult events. The similarity of these events increased consistency within informants but had no effect on agreement between parents and teachers. In contrast, similarity in the pattern of social events children encountered at home and school predicted informant agreement for syndromal aggression and for aggression to aversive events. Our results underscore the robustness of informant discrepancies and illustrate how they can be studied as part of the larger mosaic of person-environment interactions.

Two studies examined the interactive effect of receptive verbal intelligence measured by the Peabody Picture Vocabulary Test and self-regulatory competencies measured in the delay of gratification paradigm on boys’ aggression. Study 1 participants (N=98) were middle school, low-income boys primarily ethnic minority. Participants for Study 2 (N=59) were drawn from a treatment camp for boys from low-income neighborhoods with behavioral adjustment problems. In both studies, the interaction between verbal intelligence and self-regulation was significant such that verbal intelligence was associated with lower aggression to a greater extent among boys who had effective self-regulatory skills than among those who had ineffective self-regulatory skills. The implications of these findings for interventions and for a theory of risk factors in aggression are discussed.