Catherine P. Bradshaw, Mary M. Mitchell, Lindsey M. O’Brennan , and Philip J. Leaf

1. Multilevel Exploration of Factors Contributing to the Overrepresentation of Black Students in Office Disciplinary Referrals Catherine P. Bradshaw, Mary M. Mitchell, Lindsey M. O’Brennan, and Philip J. Leaf Johns Hopkins University • There is an increasing awareness of the over-representation of ethnic minority students- especially African-American males in disciplinary actions. • Dwannal McGahee and Kathy Nutt • EDRS 810

2. Is the problem clearly identified and rationalized? ?Question? Are students of color, specifically African-Americans over-represented in suspensions and office referrals, as well as referrals to Special Education? (Eitle& Eitle, 2004, Raffaele Mendez & Knoff, 2003; Skiba et al., 2008; Skiba, Michael, Nardo & Peterson, 2002.)

3. Does the review of literature elucidate the problem and inform the specific purpose of the research? • According to Irvin, Tobin, Sprague, Sugai & Vincent, “ODRs are widely used by schools to monitor student discipline problems, evaluate the impact of school based interventions and policies and make programmatic decisions regarding support services, both school-wide and for individual students” (p. 508)

4. Does the review of literature elucidate the problem and inform the specific purpose of the research? In addition: • “The referral process is based solely on teachers’ subjective appraisal of the situation” (Irvin et al., 2004.) Also: • “ODR’s may be susceptible to contextual factors or potential bias. This is particularly disconcerting, given that ODRs can have negative effects for students, such as reduced opportunities for learning” (Scott & Barrett, 2004.)

5. Purpose of the Research The purpose of this study: • Determine if there is an over-representation of African-American students in ODRs. • Determine if there is possibly a cultural disparity between students and their teachers. • “This possible mismatch in values can increase the likelihood of a discrepancy between what minority students perceive as being “appropriate” behavior and what teachers and administrators hold as acceptable standards for student behavior” (Skiba et al., 2008.)

6. Purpose of the Research continued • To determine if different disciplinary practices could differ as a result of teacher/student ethnicity. SO… • The authors of this study, “considered the overall level of behavior problems, characteristics of the classroom, (ie. Overall level of disruption) and the teacher’s ethnicity as potential factors that may contribute to the overrepresentation of Black students in ODR data.”

7. Research Questions • Are African-American teachers as likely as white teachers to give ODRs to African-American students? • What are some characteristics of the classroom environment that may increase the chances of students receiving ODRs (including average level of behavior problems within the classroom and the percentage of students in the class receiving ODRs?)

8. Is the specific purpose researchable? Does it follow logically from the problem and relevant literature? • Authors of this study cite prior research: “Prior research on teacher ratings of student behavior has shown small but significantly higher levels of disruptive behavior and other externalizing behavior problems, among Black students compared to their White peers (Epstein et al., 2005; Koth, Bradshaw & Leaf, 2009; Sbarra & Pianta, 2001). • “However, several researchers have theorized that the overrepresentation of minority students in ODR and other disciplinary data reflects a potential cultural bias embedded in school discipline practices (Monroe, 2005; Townsend, 2000).

9. Are data sources and collection methods appropriate to the purpose? Would other or additional data sources and collection methods be more relevant to the purpose of the research? Yield richer and more useful information? • Quantitative Research • Large sample • Multi-level analysis using interpretive tools • More than one type of data collection instruments • ODRs • SWIS

10. Experimental conditions and control • ODR data was obtained from classroom teachers and the School-Wide Information System (SWIS.) • Teacher- reported ODR data indicated on a student demographic form whether a child had been referred to the principal’s office during the school year. • Teacher ratings of each student’s behaviors were obtained using the Teacher Observation of Classroom Adaptation – Revised (TOCA-R)

11. Experimental conditions and control continued • Student-level demographics covariates (variables that the researcher controls for using statistics and that relate to the dependent variable but do not relate to the independent variable): • Gender (1=male, 0= female) • Ethnicity (1= Black, 0= White) • Grade level (0=5/ kindergarten to fifth grade) • Classroom level variables: • TOCA-R- contextual risk factors • Percentage of students with ODRs- covariates

12. Experimental conditions and control continued • All data collected within a single school year. • Minimal bivariate(data for which there are two variables for each observation- teachers and SWIS) correlations between independent variables. • The student-level covariates included ethnicity, gender, grade and teachers’ ratings of each student’s disruptive behaviors.

13. ParticipantsStudents and Teachers • 6,988 elementary students (all either African-American or White) • 381 classroom teachers (all either African-American or White) Within • 21 intervention schools • 42.9% of all student body on free or reduced lunch

14. Data Sources • ODR data were obtained from two sources: • The classroom teachers • The School-Wide Information System (SWIS- an internet-based data system used to manage student discipline referral data.

15. Are appropriate interpretive tools used to analyze the data? • The main effects of the student and classroom level covariates on the receipt of each ODR. • Multilevel analysis: • Child gender X Ethnicity Interaction • Three Way Interaction • Main Effects Model Strength of Study: • Availability of multiple sources of data (ODRs, SWIS and TOCA-C ratings

16. In the Discussion are all issues raised adequately resolved either in the findings or researcher’s explanation? • “The current study aimed to identify factors at the student and classroom levels that may contribute to the overrepresentation of Black students in ODRs” (p. 513.) • The current study is consistent with prior research that indicates that African-American students are more likely to receive ODRs than their White peers.

17. Do implications and directions follow logically from the findings? • Implications of this study; • African-American students do receive more ODRs than their White classmates. BUT… • Student- teacher ethnicity continuity does not exist. • “ODRs are likely influenced by multiple factors including student behavior, the classroom and/or school rules, and teacher factors (efficacy, skills, perceptions of the behavior and/or student.”

18. Limitations and Future Research • The limitations of this study: • focused exclusively on African American and White students as there were too few Hispanic/Latino, Asian-American or Native American participants to examine these groups separately. • Future research: • Why teachers make ODR’s and why African American students are at greater risk to receive them. • Look at discipline in secondary schools. • Additional longitudinal research.