Reducing Behavior Problems in the Classroom: Evidence-Based Recommendations

1. Reducing Behavior Problems in the Classroom: What we know and what we need to know Michael H. Epstein, Ed.D. University of Nebraska-Lincoln mepstein1@unl.edu Krista Kutash, Ph.D. University of South Florida kutash@fmhi.usf.edu The Institute of Education Sciences (IES) 5th Annual Research Conference June 2010 - National Harbor, MD

2. Today’s Presenters Mike Epstein Ph.D. Krista Kutash Ph.D. How you can visualize us

3. Released in 2008

4. Panel Members Mike Epstein (Chair) University of Nebraska-Lincoln Marc Atkins University of Illinois-Chicago Doug Cullinan North Carolina State University Krista Kutash University of South Florida Robin Weaver Harmony Hills Elementary School Principal (Silver Spring, MD)

5. SRI Staff Michelle Woodbridge (Study Director, author) Mary Wagner (Senior Consultant, author) Jennifer Yu (Researcher, author) Carl Sumi (Researcher)

6. WWC Practice Guides: Purpose • Use expertise and judgment of a panel to identify the most important research relevant to the recommendations. • Bring the best available evidence to bear on challenges that cannot currently be addressed by single interventions or programs.

7. WWC Practice Guides: Purpose • Supply discrete recommendations that are intended to be actionable. • Provide a coherent approach to a multifaceted problem. • Explicitly connect each recommendation to the level of evidence supporting it (strong, moderate, or low)

8. WWC Evidence Standards Reviewed studies receive 1 of 3 ratings: • Meets Evidence Standards • Strong evidence of the intervention’s effectiveness • Meets Evidence Standards with Reservations • Weaker evidence of the effectiveness • Does Not Meet Evidence Standards • Insufficient evidence of the effectiveness

9. WWC Study Rating Randomization Yes No Attrition High Low Equivalence Yes No Meets Evidence Standards Meets Evidence Standards with Reservations Does Not Meet Evidence Standards *Figure from WWC Procedures and Version 2 Standards Handbook, December 2008

10. WWC Study Rating Rating affected by: • Rigor of design • RCT or QED studies • Regression discontinuity and single subject standards under development • Validity and reliability of outcome measures • Equivalence of groups at baseline • Overall and differential attrition • Absence of confounds due to: • N of 1 • Intervention combined with another intervention • Not implemented as designed

11. Levels of Evidence for Practice Guides • Strong • High internal and external validity, including: • Evidence of effectiveness with no contradictory evidence, OR • One large, well-designed, randomized, multisite effectiveness trial with no contradictory evidence • Moderate • High internal but moderate external validity, OR • Moderate internal but high external validity, including: • Small sample sizes or conditions that limit generalizability, OR • Nonequivalent comparison groups at pretest but consistently enhanced outcomes for participants, OR • Correlational research with strong controls for selection bias • Low • Evidence not meeting standards above

12. Practice Guide: Scope • Primary audiences include: • General education elementary school teachers who will implement practices • Elementary school/district administrators who will promote practices • Final product is “more like a consensus panel report than a meta-analysis” in terms of breadth and complexity of topic addressed.

13. Recommendation 1 • Identify specifics of problem behavior & conditions that prompt and reinforce it. • Level of Evidence: Moderate • Implementation Guidelines • Concretely describe the behavior problem and its effect on learning. • Observe and record frequency and context of the problem behavior. • Identify what prompts and reinforces the problem behavior.

14. Recommendation 1Level of Evidence • WhyModerate? • Multiple single-subject research studies demonstrated effectiveness of interventions tailored to antecedents and consequences of behavior problems • Only emerging evidence on feasibility of general educator applying assessment-based approaches

15. Recommendation 1 - Need • Identify specifics of problem behavior and conditions that prompt and reinforce it. • Develop screening measures that are time efficient and psychometrically sound. • Develop progress monitoring systems that are valid and reliable. • Demonstrate that teachers can independently implement functional assessment systems.

16. Recommendation 2 • Modify the classroom learning environment to decrease problem behavior. • Level of Evidence: Strong • Implementation Guidelines • Revisit, re-practice, and reinforce classroom expectations. • Modify the classroom environment to encourage instructional momentum • Adapt or vary instructional strategies to increase opportunities for academic success and engagement.

17. Recommendation 2Level of Evidence • WhyStrong? • 3 RCTs, 1 QED, and 6 single-subject research studies demonstrated empirical support for: • Preventive classroom management • Direct and differentiated instruction • Peer tutoring

18. Recommendation 2 - Need • Modify the classroom learning environment to decrease problem behavior. • Investigate the influence of student choice of tasks on student behavior. • Develop professional development programs to teach behavior management skills to teachers.

19. Recommendation 3 • Teach and reinforce new skills to increase appropriate behavior and preserve a positive classroom climate. • Level of Evidence: Strong • Implementation Guidelines • Identify where the student needs explicit instruction for appropriate behavior. • Teach skills by providing examples, practice, and feedback. • Manage consequences so that reinforcers are provided for appropriate behavior and withheld for inappropriate behavior.

20. Recommendation 3Level of Evidence • WhyStrong? • 5 RCTs and 3 single-subject research studies demonstrated effectiveness of teaching and reinforcing replacement behaviors to reduce inappropriate behaviors • Attention seeking • Social skills • Problem solving • Self management • Self control

21. Recommendation 3 - Need • Teach and reinforce new skills to increase appropriate behavior and preserve a positive classroom climate. • Develop manualized programs that teach social behavior skills based on sound instructional practices.  • Investigate the contribution of degree of implementation and dosage of treatments to student outcomes. • Conduct research investigating the value added effects of multi-component programs.

22. Recommendation 4 • Draw on relationships with colleagues and families for guidance and support. • Level of Evidence: Moderate • Implementation Guidelines • Collaborate with other teachers for continued guidance and support. • Build collaborative professional partnerships with school, district, and community behavior experts who can consult with teachers when problems are serious enough to warrant help from outside the classroom. • Encourage parents and other family members to participate as active partners in teaching and reinforcing appropriate behavior.

23. Recommendation 4Level of Evidence • Why Moderate? • 1 QED and 1 single-subject study demonstrated peer teacher relationships (i.e., coaching) improved student social skills and engagement • 1 RCT confirmed effectiveness of teachers’ consulting with behavioral experts • Limitation: study conducted specifically with ADHD population • 2 RCTs confirmed positive effect of teacher-parent partnerships • Limitation: study conducted with specific teacher-parent program

24. Recommendation 4 - Need • Draw on relationships with colleagues and families for guidance and support. • Focus consultation research on student outcomes as opposed to teacher-centered outcomes. • Investigate teacher assistance teams decision making process.  • Test efficacy and effectiveness of consulting models. • Investigate the value added effects of school based mental health programs.   • Understand parent engagement process. • Develop alternative delivery models of parenting programs.

25. Recommendation 5 • Implement schoolwide strategies to reduce negative and foster positive interactions. • Level of Evidence: Moderate • Implementation Guidelines • Address schoolwide behavior issues by involving a school improvement team • Collect information on the hot spots throughout the school, such as the frequency of particular schoolwide behavior problems and when and where they occur • Monitor implementation and outcomes using an efficient method of data collection and allow ample time for the program to work. • If warranted, adopt a packaged intervention program that fits well with identified behavior problem(s) and the school context.

26. Recommendation 5Level of Evidence • Why Moderate? • 1 QED study demonstrated positive effects of schoolwide changes (e.g., structure, organization) on student social relationships and acceptance • 4 RCTs and 1 single-subject study documented impact of schoolwide intervention program on reduced problem behaviors • Limitation: RCTs support only specific schoolwide program, not all components of recommendation.

27. Recommendation 5 Need • Implement schoolwide strategies to reduce negative and foster positive interactions. • Test the efficacy and effectiveness of schoolwide programs.  • Understand how school climate/culture influences the implementation and outcomes of schoolwide programs. • Study the economic benefits of schoolwide programs.

28. To download and print http://ies.ed.gov/ncee/wwc/publications/practiceguides/

29. Thank You Thank you for participating today