Please take our brief survey

Blueprints Programs = Positive Youth Development

Return to Search Results

Promising Program Seal

Group Teen Triple P - Level 4

Blueprints Program Rating: Promising

An 8-week group-based parent-training program designed to improve parenting skills, manage family problems, and enhance positive family relationships, ultimately to prevent problem behavior among youth.

  • Matthew Sanders, Ph.D.
  • Director
  • The University of Queensland
  • Parenting and Family Support Center
  • School of Psychology
  • Brisbane, QLD 4072
  • Australia
  • 61 (7) 3365 7290
  • Close Relationships with Parents
  • Conduct Problems

    Program Type

    • Parent Training

    Program Setting

    • Community (e.g., religious, recreation)

    Continuum of Intervention

    • Universal Prevention (Entire Population)
    • Selective Prevention (Elevated Risk)

    An 8-week group-based parent-training program designed to improve parenting skills, manage family problems, and enhance positive family relationships, ultimately to prevent problem behavior among youth.

      Population Demographics

      The program targets parents of teenage youth aged 12-15 and, through the parents, teens themselves.


      • Early Adolescence (12-14) - Middle School


      • Male and Female


      • All Race/Ethnicity

      Subgroup Details

      The study did not examine racial/ethnic or gender subgroup impacts of the intervention.

      • Individual
      • Family
      Risk Factors
      • Individual: Favorable attitudes towards antisocial behavior
      • Family: Family conflict/violence*, Parent stress, Poor family management*
      Protective Factors
      • Individual: Prosocial behavior
      • Family: Parent social support*

      *Risk/Protective Factor was significantly impacted by the program.

      See also: Group Teen Triple P - Level 4 Logic Model (PDF)

      Group Teen Triple P - Level 4 consists of four 2-hour group sessions, delivered over 8 weeks, with up to 12 parents of teenage adolescents. The sessions provide opportunities for parents to gain knowledge and skills for reducing parent-child and marital conflict, depression, and high levels of parenting stress that can lead to negative outcomes in young adults. Parents learn new skills through a process of observation, discussion, practice, and feedback, and between sessions complete homework tasks designed to reinforce the content of the group sessions.

      Group Teen Triple P - Level 4 consists of four 2-hour sessions, delivered over 8 weeks, in groups of up to 12 parents. Parents actively participate in a range of exercises to learn about the causes of adolescent behavior problems, setting specific goals, and using strategies to promote a teenager's skills development, manage inappropriate behavior and teach emotional self-regulation.

      Specifically, during group sessions parents are coached in a variety of behavior management skills including: monitoring problem behavior; providing brief contingent attention following desirable behavior; arranging engaging activities; using directed discussion for minor problem behavior; making clear, calm requests; and backing up instructions with logical consequences. For more difficult or well-established adolescent behavioral problems, parents learn to use family meetings, and behavior contracts to make necessary changes to family routines.

      Sessions Cover:

      Session 1: Positive parenting. This session provides parents with an introduction to positive parenting, factors that influence teenagers’ behavior, and how to set goals for change. Parents submit a completed assessment booklet at the beginning of this session.

      Session 2: Encouraging appropriate behavior. During this session, the practitioner discusses how to develop positive relationships with teenagers, increase desirable behavior, teach new skills and behaviors, use behavior contracts, and hold family meetings.

      Session 3: Managing problem behavior. In this session, parents learn how to develop family rules, deal with non- cooperation, acknowledge emotions, and use behavior contracts. They have an opportunity to rehearse these routines in the session, to promote emotional self-regulation.

      Session 4: Dealing with risky behavior. This session covers identifying risky situations, routines to deal with risky behavior, and family survival tips. Parents also prepare for their individual consultation sessions.

      Session 5-7: Implementing parenting routines 1-3. The practitioner provides feedback from initial assessments that the family completed and then uses a self-regulatory feedback model to assist parents to review their implementation of parenting strategies and risky behavior plans. From this, parents set goals for the further refinement of their routines, if needed.

      Session 8: Program close. Parents return for a final group session to review progress and family survival tips, look at ways to maintain changes and problem-solve for the future, and to close the program. If necessary, referral options are discussed.

      Teen Triple P is based on social learning principles, with a strong emphasis placed on parents acknowledging and encouraging the growing autonomy and independence of their teenage children. The program is also skill-oriented, in that parents learn about and practice skills that are used to promote a healthier family environment.

      • Skill Oriented
      • Social Learning

      Families with a child between the ages of 12 and 15 were recruited using a community outreach approach in Auckland, New Zealand. Though both parents in intact families were encouraged to participate, only mothers’ assessment data was used. In total, 72 families completed baseline measures before being randomly assigned to treatment or control conditions. Most families (96%) were assessed at posttest, 10 weeks later, and 81% were retained at 6-month follow-up.

      The treatment group improved 13 of 22 outcomes across parent and youth reports at posttest and 6-month follow-up, compared to controls; however, some impacts were not maintained. Generally, the intervention group improved risk and protective factors related to fostering positive relationships between parents and children as well as parent and self-reported adolescent problem behavior.

      Compared to controls, the treatment group improved:

      • Adolescent and parent-reported problem behavior
      • Both adolescent and parent-reported measures of positive child-parent relationships

      Improvements to risk and protective factors:

      • Family Conflict
      • Family Cohesion
      • Parent Self-Efficacy
      • Family Management

      Not assessed.

      Effect sizes reported in Chu et al. (2014) were small (d =.14-.22), small-medium (d =.32-.41), and large (d = .71-1.10). These effect sizes, however, were overstated since the authors used the F-test to calculate effect sizes (a method largely used in meta-analyses when means and standard deviations are typically not reported). The authors therefore recalculated effect sizes using the standard formula (i.e. the adjusted condition differences from the ANCOVA divided by the pooled pretest standard deviation). The recalculated effect sizes (not reported in the article) were small (d =.00-.27), small-medium (d =.32-.38), medium (d =.42-.52), medium-large (d =.61-.67), and large (d =.75-1.26).

      All participating families were recruited from Auckland, New Zealand, and appeared representative of families in that country.

      • Many measures came from parents, who helped deliver the program, but youth reports are also available.
      • Groups differed at baseline on only 1 of 22 measures, parent over-reactivity, but this is controlled in analysis.
      • No significance test for differential attrition by group or socioeconomic background. However, attrition was less than 5% at posttest, and at six-month follow-up there were no significant differences between completers and non-completers.

      • Blueprints: Promising

      Ralph, A., & Sanders, M. R. (2003). Preliminary evaluation of the Group Teen Triple P program for parents of teenagers making the transition to high school. Australian e-Journal for the Advancement of Mental Health, 2(3), 169-178.

      Chu, J. T. W., Bullen, P., Farruggia, S. P., Dittman, C. K., & Sanders, M. R. (2015). Parent and adolescent effects of a universal group program for the parenting of adolescents. Prevention Science, 16, 609-620.

      Bradley Thomas
      Chief Executive Officer
      Triple P America, Inc.
      Head Office Address: 1201 Lincoln Street, Suite 201, Columbia, SC 29201
      Postal Address: PO Box 12755, Columbia, SC 29211
      Phone: (803) 451-2278 x204

      Chu, J. T. W., Bullen, P., Farruggia, S. P., Dittman, C. K., & Sanders, M. R. (2015). Parent and adolescent effects of a universal group program for the parenting of adolescents. Prevention Science, 16, 609-620.


      Recruitment/ Assignment: Families with teenage children were recruited from Auckland, New Zealand using a community outreach approach in schools, media outlets, and public events. Eligible families had a child between 12 and 15 years old who did not have developmental or intellectual disabilities and whose parent was not receiving professional help for his/her own or the child’s behavioral or emotional problems. While 107 parents were screened, 72 were deemed eligible, completed baseline assessments, and randomized to treatment (n= 35) or a care-as-usual control (n=37) condition.

      Assessment: Parents completed measures at baseline, posttest (10 weeks post-baseline), and at 6-month follow-up. While both parents of intact families were encouraged to participate in the evaluations, only the mother’s data was used. Most completed the posttest assessment (n= 69; 96% of baseline sample) and 81% (n= 58) were retained at 6-month follow-up, though the analysis used data from the full baseline sample.


      Most mothers were married (67%), many to the father of their child (65%), with an average age of 45. More than half (52%) had a university degree and were currently employed (81%), with 35% earning more than the average New Zealand household income ($81,067) and 70% reporting no major difficulties paying for household expenses over the last year. The adolescents targeted by the intervention were mostly male (60%) and averaged 13 years of age. Relatively few were indigenous (10%) or belonged to other racial/ethnic minorities (17%).


      Mothers completed 13 measures pertaining to parenting, parental relationships, parental adjustment, and adolescent problem behavior. Adolescents completed 9 measures on family relationships, adjustment, problem behavior, and parental monitoring. For testing, the parent-reported outcomes and the adolescent-reported outcomes were each treated as part of six conceptually related groups: 1) family relationship, 2) parental relationship, 3) parenting, 4) adolescent problem behavior, 5) adolescent adjustment, and 6) parental adjustment.

      Family Cohesion and Conflict was assessed by both parents and children using 2 subscales, each containing 9 items, from the Family Environment Scale. Both measures demonstrated good reliability (a = .84, .83).

      Parent-Adolescent Conflict was measured using 8 items from the Parent Conflict Questionnaire, and had high internal consistency for both mothers and adolescents (a = .82, .84).

      Inter-Parental Conflict Over Child Rearing was measured with 16 items from the Parent Problem Checklist which provides an index of the number of disagreements and the severity these disagreements. Single mothers did not complete the measure. Both index frequency and severity had acceptable reliability (a = .82, .84).

      Parent Relationship Satisfaction was assessed with 6 items from the Relationship Quality Index, which had very high internal consistency (a = .93).

      Parent Laxness and Over-Reactivity came from an adaptation of the parent-reported Parenting Scale with 13 items. Reliability of the laxness measure was good (a = .89), but for over-reactivity was low (a = .61).

      Parental Monitoring was assessed by parents and children using the Parental Monitoring Scale, with 8 items measuring how well parents keep track of their child’s whereabouts. Parent and adolescent reports both had high consistency (a= .91).

      Parent Self-Efficacy was measured from parent responses to 13 items from the 35-item Parental Self-Efficacy Scale. The reliability of the measure was good (a = .92).

      Parent Depression and Stress were assessed using subscales from the Depression Anxiety Stress Scales-21. While anxiety was also measured it was not used, as reliability was considerably lower than for the other measures (a = .87 for depression, .83 for stress).

      Adolescent Problems—Emotional, Conduct, Hyperactivity, & Peer were measured by parents and children using relevant 5-item subscales from the Strengths and Difficulties Questionnaire. The items were summed to create an overall score, which had acceptable reliability for both parents and adolescents (a = .79, .81).

      Adolescent Problem Behavior was measured across multiple dimensions (e.g. school-related deviance, risk taking, substance use) with the 22-item Problem Behavior Checklist. The adolescent-reported measure had moderate internal consistency (a = .72).

      Adolescent Autonomy in Decision-Making was assessed across 12 topical areas with the Autonomy Scale. The youth- eports had moderate reliability (a = .73).

      Adolescent Self-Esteem was assessed using the 10-item Rosenberg Self-Esteem Scale, which demonstrated good consistency with the youth reporting (a = .89).

      Adolescent Empathy was measured using the caring subscale from Positive Youth Development. The 9-items had moderate reliability (a = .75).


      Intervention effects were estimated using a series of multivariate and univariate analyses of covariance (MANOVA and ANCOVA). Separate models were used to assess posttest and longer-term impacts (at 6-month follow-up), with baseline outcomes as covariates, rather than adjusting for repeated measures. Many univariate tests (44) were performed without adjustment for multiple tests, though they were done in combination with the MANOVA tests for conceptually related groups of outcomes.

      In accordance with intent-to-treat, the analyses appeared to use data from the full baseline sample regardless of treatment adherence, though the method for including missing data was mentioned only briefly. The authors stated that, “Expectation maximization was used to estimate values for the intent-to-treat sample.”


      Implementation Fidelity: While no figures were presented, the study states “facilitators completed session checklists to ensure treatment integrity and reduce protocol drift during the trial.”

      Baseline Equivalence: One pre-intervention difference was observed for parent over-reactivity, with the treatment group reporting higher levels than controls. The measure was controlled for in subsequent analyses.

      Differential Attrition: Attrition was minimal (<5%) at posttest, but retention rates appeared to differ at 6-month follow-up (77%, treatment; 84%, control) and the difference did not appear to be tested for statistical significance. Also, there were no tests for attrition by sociodemographic background. However, “[a] series of one-way ANOVAs revealed no significant differences… between completers and non-completers at the 6-month follow-up on any of the dependent variables.” In addition, missing values for the outcomes were found to be missing completely at random.

      Posttest: For parent behavior at posttest, 7 of 12 tests for parent-reported measures and 4 of 4 tests for adolescent-reported measures reached statistical significance. Medium-large improvements were noted for parent and youth-reported parent-youth conflict and parental monitoring, and youth-rated family conflict and empathy and large impacts were noted for mother-reported family conflict, cohesion, laxness, over-reactivity, parenting self-efficacy, and youth-reported family cohesion. For parent behavior at the 6-month follow-up, the results were similar, with 5 of 12 tests for parent-reported measures and 4 of 4 tests for adolescent-reported measures reaching statistical significance.

      For adolescent problem behavior and adjustment at posttest, 1 of 1 parent-reported measure and 0 of 5 adolescent-reported measures reached statistical significance. A large improvement (d = .90) was found for the parent-reported measure of total problems. At the 6-month follow-up, 1 of 1 parent-reported measure and 3 of 5 adolescent-reported measures reached statistical significance. The newly significant outcomes included two measures of adolescent-reported total problems and one measure of caring.