Get documents about "Slide 1
Shared by: HC12100121565
-
Stats
- views:
- 2
- posted:
- 10/1/2012
- language:
- English
- pages:
- 1
/ 1
Document Sample
Investigating The Effect Of Extra Sessions Of A Structured
Peer Interaction Program On Social Competence
Herle McGowan, University of Michigan; Robert Nix, Pennsylvania State University; Susan Murphy, University of Michigan;
Karen Bierman, Pennsylvania State University; and the Conduct Problems Prevention Research Group*
Background Sample Characteristics Assessment Procedure Intervention and Control Group Differences Regression of Peer Pairing on Social Competence
• Intervention group: n = 410 • Measures to determine need for peer pairing collected at end of 1st grade • Purpose: Estimate the overall effect of participating in Fast Track (1 =
2.8
To make progress in the prevention and treatment of children’s mental 2.6
Teacher Rating of Social
All Intervention
health problems, we must determine not only which intervention programs • Control group: n = 403 • Peer pairing sessions were delivered throughout the fall and spring semesters of 2nd intervention; 0 = control) on children’s social competence in 2nd grade 2.4 Children
Competence
2.2
are effective but also how they work (Kazdin, 2000). Examining dose- grade • n = 762 Subset for whom
• All children in 2nd grade; children who repeated first grade were excluded 2
peer pairing was
response relations between the hypothesized agents of therapeutic change • Social competence outcome measure collected at end of 2nd grade • Standardized β = .07 (p = .05)
1.8 recommended
and children’s functioning in clinically-relevant domains is one way of • All scored above 85th percentile among classmates for oppositional and 1.6 and received
aggressive behavior • Teacher ratings of 9 items on 6-point Likert scale (α > .80) • Conclusion: Compared to control group children, intervention children
1.4
testing underlying theories of disorder and treatment. This can be difficult, 1.2
however, because clinical staff members often end up providing more • 55% minority; 28% female • Typical items: Friendly; controls temper when there is a disagreement displayed more social competence at the end of 2nd grade 1
0 11 22 30
services to those children who demonstrate greater need over time. Thus, Number of Peer Pairing Sessions in 2nd Grade
even when those extra intervention services are successful in preventing
additional problems, a positive dose-response relation can be obscured by Intervention Procedure
unmeasured selection effects. Some natural experiments circumvent this
• In 1st grade, all families were supposed to receive the same amount of
Dose-Response Relation For All Intervention
problem by providing more intervention services based on factors other than Timeline of Assessment and Intervention
children’s need, and they have demonstrated positive dose-response relations intervention services Regarding Peer Pairing Impact of Fast Track on Children's Social
Children
(e.g., Seitz, Apfel, & Rosenbaum, 1991). Other studies have attempted to • Intervention services consisted of social skills training, reading tutoring, Competence • Purpose: Estimate the dose-response relation for all intervention children,
control statistically for measured and unmeasured selection effects peer pairing, friendship groups, parent training groups, and home visits including those who received peer pairing as a result of project guidelines or
associated with children’s need. Depending on the presence or absence of • In 2nd grade, children and families received different amounts of reading
2.54
2.52
staff members’ clinical judgment and those did not receive peer pairing at all
strong instrumental variables, such efforts have led to contradictory results tutoring, peer pairing, and home visits based on project guidelines or clinical 2.5
• n = 328
2.48
regarding dose-response relations even in the same intervention program judgment of assessed need.
(e.g., Foster, in press; Salzer, Bickman, & Lambert, 1999).
Spring Summer Fall Winter Spring 2.46
• Standardized β = -.15 (p = .003)
1st grade 1st grade nd
2nd grade 2nd grade
2.44
2 grade
In this study within Fast Track (Conduct Problems Prevention Research
2.42
2.4
• Conclusion: When use of clinical judgment was believed to result in unmeasured
selection effects, there was a significant negative relation between peer pairing
Group [CPPRG], 1992), a preventive intervention program for children with Description of Peer Pairing Teacher Staff Staff Teacher
2.38
2.36
and social competence
conduct problems, we had the opportunity to examine dose-response ratings and members members ratings of
• Weekly 30-minute structured dyadic play session Intervention Control group • It is possible that staff members used more proximal information, which
relations between one component of the intervention, peer pairing, and one sociomentric could added peer social
outcome measure, social competence. (Comprehensive evaluations of Fast • Designed to further develop intervention children’s social skills nominations change pairing for competence was not recorded, about current functioning to provide more peer pairing
Track are available elsewhere [i.e., CPPRG, 1999; 2002]). We examined • Also designed to foster positive interactions between intervention children and collected to decision some collected.
assess need regarding children to those children with more serious behavior problems and less social
dose-response relations among two groups of children: The group of their classmates and thereby decrease the likelihood of peer rejection
children who met formal project guidelines at the end of 1st grade and for peer peer who were competence, resulting in the observed negative relation
• Intervention child paired with same-sex classmate without behavior problems pairing in pairing not doing
received extra peer pairing in 2nd grade and the entire group of children who • It is possible, although believed unlikely, that peer pairing impeded the
participated in the intervention in 2nd grade. • Classmate rotated each week 2nd grade based on well; they Dose-Response Relation For Children Who Should
children’s also development of social competence for those children who did not need
• Session coached by Fast Track staff member
Study Hypotheses functioning adjusted the Have Got Extra Peer Pairing and Did peer pairing but received it anyway
in new number of
• Fast Track will have a positive impact on children’s social competence • Purpose: Estimate the dose-response relation for children who met project • We cannot distinguish between these alternative explanations without
classroom sessions
• There will be a positive relation between peer pairing and social competence Project Guidelines Regarding Need for Peer Pairing context. children
guidelines and actually received at least one session of peer pairing in both the measures of child functioning, which were believed to influence clinical
received fall and spring
• That relation will be weaker for the children who received peer pairing as a • Clinical range on Teacher Report Form aggressive behavior syndrome judgment, collected throughout 2nd grade
(Achenbach, 1991), or based on • n = 157
result of staff members’ clinical judgment because unmeasured selection
current • Standardized β = .18 (p = .02)
effects will be more problematic • Clinical range on attention problems syndrome (Achenbach, 1991), or functioning.
• Rejected status based on peer sociometric nominations • Conclusion: When intervention services were believed to be less determined by Discussion
Method unmeasured selection effects, there was a positive relation between peer pairing
and social competence
At the end of 2nd grade, the entire Fast Track preventive intervention
program resulted in a small but statistically significant treatment effect on
Study Design Use Clinical Judgment to Recommend Peer Pairing • It is possible that there is an even stronger positive relation between children’s social competence. A positive dose-response relation emerged
• Four sites: Durham, NC; Nashville, TN; Seattle, WA; rural central • Borderline range on aggressive behavior syndrome, or peer pairing and social competence that is reduced by our inability to among the children who met project guidelines at the end of 1st grade and who
Pennsylvania control for the use of clinical judgment inc adjusting the number of peer received at least one session of peer pairing in the fall and spring semesters of
• Borderline range on attention problems syndrome, or 2nd grade. Staff members were not supposed to exercise clinical judgment in
• Intervention and control group children randomly assigned by school Results pairing sessions children actually received
• Controversial status based on peer sociometric nominations, or adjusting services for these children, so it is believed that selection effects were
• All schools served poor, high-crime neighborhoods • It is also possible that the relation between the dose of peer pairing and less problematic. A negative dose-response relation emerged, however, when
• Neglected status based on peer sociometric nominations, or
• Two-stage teacher and parent screening process in three successive cohorts Plans for Analyses social competence among these children is actually due to other factors: considering all intervention children. It is believed that staff members used
• Unique information staff member had about child functioning
• Multiple linear regression equations accurate but unmeasured information about children’s functioning in 2nd grade
For example, staff members may have provided both more sessions of
to determine whether and how much peer pairing some children should receive,
________________________________________________________________ • All analyses controlled for study site, cohort, child sex, and child race peer pairing sessions and higher levels of other components of Fast Track thus confounding their need for peer pairing with their response to it. In sum, it
*Members of the Conduct Problems Prevention Research Group are, in Peer Pairing Implementation • Additional analyses controlling for 15 relevant pretreatment or 1st-grade covariates to children who needed help in improving social competence. is critical to have current and valid measures of functioning at each point in
alphabetical order, Karen Bierman, Pennsylvania State University; John • 65% of intervention children received some peer pairing in 2nd grade produced similar results which the amount of intervention services can be adjusted in preventive
• We cannot distinguish between these alternative explanations without
Coie, Duke University; Kenneth Dodge, Duke University; Michael Foster,
• If children were determined by project guidelines or judged by Fast Track staff • Additional analyses with imputed missing data produced similar results intervention programs. This is of particular importance in multiple component
Pennsylvania State University; Mark Greenberg, Pennsylvania State measures of child functioning, which were believed to influence clinical intervention programs such as Fast Track. Even if the intervention as a whole
members to need peer pairing, they were supposed to receive 22 sessions • Additional analyses with propensity scores produced similar results
University; John Lochman, University of Alabama; Robert McMahon, judgment, collected throughout 2nd grade is known to be beneficial, without such measures recorded each time a decision
University of Washington; and Ellen Pinderhughes, Tufts University. • Children actually received between 1 and 30 sessions • Additional analyses with other outcome measures produced similar results regarding treatment is made for a particular component of the intervention, it
will be impossible to determine the effect of that component.
