Exclusionary discipline practices across students' racial/ethnic backgrounds and disability status: findings from the Pacific Northwest.
Article Type:
Report
Subject:
Alternative education (Demographic aspects)
Elementary school students (Demographic aspects)
Ethnicity (Educational aspects)
Race (Educational aspects)
Authors:
Vincent, Claudia G.
Sprague, Jeffrey R.
Tobin, Tary J.
Pub Date:
11/01/2012
Publication:
Name: Education & Treatment of Children Publisher: West Virginia University Press, University of West Virginia Audience: Professional Format: Magazine/Journal Subject: Education; Family and marriage; Social sciences Copyright: COPYRIGHT 2012 West Virginia University Press, University of West Virginia ISSN: 0748-8491
Issue:
Date: Nov, 2012 Source Volume: 35 Source Issue: 4
Product:
Product Code: E197200 Students, Elementary
Geographic:
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
310741890
Full Text:
Abstract

We examined 2009-2010 data on exclusionary discipline practices from one state in the Pacific Northwest of the United States across students' racial/ethnic backgrounds and disability status. Our focus was on proportionate representation in exclusionary discipline actions and in the duration of those disciplinary actions. Descriptive outcomes indicated that among students with disability American Indian/Alaska Native (AI/AN) students were over-represented in removal to alternative education. Among students without a disability, Hispanic students were most notably over-represented in all exclusionary discipline practices. African-American students with and without disability lost approximately twice as many days as White students to exclusionary discipline. Follow-up CM-Square tests showed that non-White students were statistically significantly over-represented in most exclusionary practices. ANOVA results indicated that both disability status and race significantly impacted the duration of exclusions. Recommendations for future research are provided.

Recently, disciplinary exclusions of students have attracted national media attention (Carr, 2010; Schwartz, 2011). According to the most recent data available from the National Center for Education Statistics, 21.6% of all public school students in grades 6 through 12 had been suspended at least once in 2007, and a total of 3.4% had been expelled (Aud, Fox, & KewalRamani, 2010). Disaggregated by racial/ethnic background, African-American students are most affected by exclusionary discipline practices: 42.8% were suspended and 12.8% were expelled in 2007, compared to 15.6% and 1.0% respectively of their White peers (Aud et al., 2010). Students with a disability tend to be excluded from the classroom disproportionately more often and for longer durations than students without a disability (Vincent & Tobin, 2011). Students with emotional and behavioral disorders, depression, or mental illness face an especially high risk of being excluded (Achilles, McLaughlin, & Croninger, 2007; Krezmien, Leone, & Achilles, 2006). Unfortunately, the use of exclusionary discipline practices appears to have increased over the last decade. Cregor and Hewitt (2011) report that between 2001 and 2007 the number of out-of-school suspensions in the Chicago Public Schools quadrupled. Losen and Skiba (2010) report that suspension rates for K-12 students have "at least doubled since the early 1970's for all non-Whites" (p. 2), with the racial gap between Black and White suspension rates increasing from 3 percentage points in 1973 to 10 percentage points in 2000.

One aspect of exclusionary discipline practices is their inconsistent use. Great variability of implementation across geographic regions, locales, and student groups suggests that being excluded from school has less to do with the behavioral violation a student engaged in and more with which school the student attends and the student's racial/ethnic background (Skiba et al., 2011; Wu, Pink, Crain, & Moles, 1982). Skiba et al. (2011) found that African-American elementary students were 3.75 times more likely than White students to be suspended out of school for minor misbehaviors including inappropriate language, defiance, non-compliance, and disruption. At the middle school level, African-American students were more likely than White students to be suspended or expelled for abusive language, bullying, lying and cheating, and tardiness or truancy, while Hispanic students were more likely than White students to be suspended for minor misbehaviors, particularly non-compliance.

The disproportionately high representation of minority students among students who are suspended or expelled has been well documented in the research literature (Kaufman et al., 2010; Krezmien et al., 2006; Theriot, Craun, & Dupper, 2010). African-American students are more likely to be suspended than students of other racial/ethnic backgrounds, and tend to be suspended for subjectively interpretable offenses, such as non-compliance and disrespect (Christle, Jolivette, & Nelson, 2005; costenbader & markson, 1998; Gregory & Ripski, 2008; Skiba & Sprague, 2008). Skiba et al. (2011) examined the distribution of student exclusions in the form of suspension or expulsion across student race/ethnicity. African-American and Hispanic middle school students had statistically significantly (p < .01) higher odds of being suspended or expelled (1.12 and 1.58 respectively) than White students. Townsend (2000) found that African-American students were suspended at a rate three times higher than their enrollment. Based on discipline data from 29,613 middle school students, Raffaele-Mendez and Knoff (2003) found that at the middle school level, 20.69% of African-American students were suspended at least once, compared to 12.8% of Hispanic students and 8.84% of White students.

Suspension and expulsion from school are often meant to punish students, alert parents, and protect other students and school staff. However, unintended consequences may occur. Citing information from the U.S. Department of Education, Sughrue (2003) reports that "44% of expelled students did not have access to alternative educational opportunities" (p. 10). Referrals, suspension, and expulsion may exacerbate academic deterioration, and when students are provided with no immediate educational alternative, student alienation, distrust of teachers, delinquency, crime, and substance abuse may result (Christie et al., 2005; Skiba & Rausch, 2006). For example, Costenbader and Markson (1998) surveyed 620 students (45% middle school students) and found that discipline incidents resulting in suspension tended to generate feelings of anger toward the suspending adult, and that suspension was perceived as unhelpful in solving the problem that led to it. Moreover, students who are excluded from school have a higher likelihood of dropping out or entering the juvenile justice system (Christie et al., 2005; Fabelo et al., 2011).

Much of the literature on disproportionate discipline outcomes focuses on African-American and Hispanic students. However, students from other racial/ethnic backgrounds, especially American Indian/Alaska Native (AI/AN) students are similarly affected by disciplinary inequity and its adverse consequences, including losing valuable instructional time in the classroom (Algozzine, Wang, & Violette, 2011). In general, AI/AN students experience high drop-out rates, low graduation rates, and low academic achievement. From 1997 to 2007, the AI/AN dropout rate showed an increasing trend with 20% of 16 to 24 year olds having dropped out of school in 2007 (Aud et al., 2010; Political Research Associates, 2005).

Our goal in the present study was to examine the extent to which racial disproportionality existed in disciplinary exclusions in one state of the Pacific Northwestern United States. Specifically, our inquiry was driven by the following research questions:

1. Do students from different racial/ethnic backgrounds and disability status experience the same rates of exclusion from the classroom?

2. Do students from different racial/ethnic backgrounds and disability status experience exclusion from the classroom for the same duration?

Method

We used extant data only to answer research questions 1 and 2. Thus, our methodological approach was rooted in exploratory data analysis (EDA), an approach designed to identify patterns and relationships that can shape future research efforts (Leinhardt & Leinhardt, 1980). Our extant data consisted of 2009-2010 disciplinary exclusions collected by the state department of education and provided to the authors for research purposes, as well as publicly available enrollment data posted on the websites of the state department of education.

Sample Sizes and Characteristics

The sample consisted of 147,850 disciplinary exclusions that occurred in 2009-2010 and involved a total of 64,088 unique students in 1,195 schools. In the same year, a total of 559,251 students were enrolled in the public schools of the state. Each discipline record in our dataset contained information about the student's race/ethnicity (American Indian/Alaska Native (AI/AN), Asian/Pacific Islander, Hispanic, African-American, White, Multiracial, and Unknown), gender, grade level, and disability status (yes/no). Of the students involved in disciplinary exclusions, 67.1% were male, 17.5 % were in grades Pre-K to 5, 31.3% in grades 6 to 8, and 51.1% in grades 9 to 12. Because our primary interest was in disproportionate representation of students with and without disabilities in disciplinary exclusions, we focused on students' race/ethnicity and disability status only. Table 1 provides an overview of disciplinary exclusions and state-wide enrollment by disability status and racial/ethnic categories. Because state-wide special education enrollment was reported only for 5 racial/ethnic categories, we focused all our analyses exclusively on those 5 racial/ethnic groups and omitted calculations for multiracial students and students with unknown racial/ethnic backgrounds. We coded student ethnicity following the 2-step procedure described by the U.S. Department of Education (National Forum on Education Statistics, 2008). According to this procedure, students report their ethnicity (Hispanic/Not Hispanic) first, and their race second. All students who report Hispanic ethnicity are counted as Hispanic, regardless of their subsequent race identification.

Measures

The dataset from the state contained information on the type of exclusionary discipline associated with each record and its duration. A total of 5 types of exclusionary discipline existed: (a) in-school suspension (ISS), (b) out-of-school suspension (OSS), (c) removal to alternative education, (d) truancy, and (e) expulsion (EXP). ISS was defined as temporary removal from the regular classroom while remaining under school personnel supervision; OSS was defined as removal from the regular school to another setting; removal to alternative education was defined as the removal of a student with a disability to an alternative setting for not more than 45 days; truancy was defined as an event consisting of eight unexcused absences of a minimum of one-half day in one month; and EXP was defined as removal from the regular school for the remainder of the school year or longer. ISS, OSS, EXP and removal to alternative education were associated with a duration. The duration of these events was measured in half day increments. Truancy, an exclusion from the classroom self-imposed by the student, was not associated with a duration.

Analytical Procedures

For research question 1, we completed descriptive analyses first to assess the extent to which students from specific racial/ethnic groups with and without disability were disproportionately over -or underrepresented in each exclusionary category. For each of the 5 racial/ethnic categories of interest, we subtracted the percentage a given group represented in the total student enrollment from the percent of disciplinary exclusions involving students from that group, so that a negative number indicated disproportionate under-representation, 0 represented proportionate representation, and a positive number represented disproportionate over-representation. We completed separate analyses for students with and without disability. We then followed up on the descriptive outcomes with a series of Chi-Square goodness of fit tests (Howell, 2002). We compared observed and expected percentages for AI/AN, Asian/Pacific Islanders, Hispanic, and African-American students in each disciplinary category with observed and expected percentages for White students, the customary comparison group, in each disciplinary category. Because multiple comparisons resulted in a total of 16 tests for students without disability, and 20 tests for students with disability, we adjusted the alpha values to .003 and .0025 respectively. Results indicated whether differences in exclusion rates between non-White and White students were statistically significant. We calculated coefficient phi as an effect size of the difference in rates between non-White and White students for each disciplinary category.

For research question 2, we also completed descriptive analyses first. We calculated total student days for each racial/ethnic group by multiplying the group's total enrollment by 170 days, the average length of the 2009-2010 school year in the state. We then calculated the number of days each racial/ethnic group lost to ISS, OSS, removal to alternative education, and EXP and computed the percentages of student days lost. To determine whether the percentages of total days lost differed significantly among racial/ethnic groups with and without disability, we calculated total days lost for each unique student and followed-up with a two-way ANOVA test with disability status (yes/ no) as one factor and student race as the second factor. Our data did not meet the assumption of normality, and robustness to violations of normality was limited due to unequal cell sizes (Howell, 2002). Therefore, our ANOVA results need to be interpreted with caution within the exploratory framework of our methodological approach.

Results

Research Question 1 (Rates of exclusion from the classroom by student race/ethnicity and disability status)

Figures 1 and 2 provide an overview of the outcomes of our descriptive analyses examining the extent to which students from different racial/ethnic backgrounds and disability status were proportionately represented in disciplinary exclusions. Figure 1 shows that, among students without a disability, White students were considerably under-represented in all disciplinary actions, with the most substantial under-representations occurring in truancy and ISS. Students from Asian or Pacific Islander backgrounds were also generally under-represented in exclusionary discipline, but to a much lesser degree than White students. Overall, Hispanic students were overrepresented in most exclusionary discipline actions, followed by African-American and AI/AN students.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Figure 2 presents the same disaggregations for students with a disability. For students with a disability, removals to alternative education were recorded in addition to truancy, ISS, OSS, and EXP. Among students with a disability, White students were again underrepresented in all types of exclusions except removal to alternative education. Similarly, Asian/Pacific Islander students were slightly under-represented in all types of exclusions except removal to alternative education. Hispanic students were over-represented in truancy, ISS, OSS, and EXP, but under-represented in removal to alternative education. African-American students were over-represented in OSS and truancy. Of all groups, AI/AN students were most severely overrepresented in removal to alternative education.

The results of the follow-up Chi-Square goodness of fit tests for students with and without disability are summarized in Table 2. White students served as the comparison group. Results indicated that for students without a disability, rates of all types of exclusions differed significantly between non-White and White students. For students with a disability, rates of truancy, ISS, OSS, and expulsion differed significantly between non-White and White students. In removal to alternative education, only AI/AN students' rates differed significantly from those for White students. The average size of this difference in proportions is expressed by the phi-values. It is important to note that phi-values--commonly ranging from 0 to 1 for 2 x 2 contingency tables--exceeded values of 1.0 for students with disabilities. These values indicate that expulsion of students with disabilities were rare events and were highly unequally distributed across racial/ ethnic categories, resulting of widely varying cell sizes.

Research Question 2 (Duration of exclusion by student racial/ethnic background and disability status)

Figures 3 and 4 provide an overview of the percent of student days lost to various exclusionary discipline actions across student race/ethnicity. Figure 3 shows that, among students without a disability, African-American students had the highest percentage of total days lost to disciplinary exclusion, followed by AI/AN students and Hispanic students. White students ranked fourth, and Asian/Pacific Islanders had the lowest percentage of student days. For days lost to ISS, Hispanic students ranked first, followed by AI/AN students with almost identical outcomes; for days lost to OSS, African-American students ranked first followed by AI/AN students, and for EXP, African-American students ranked first, followed by Hispanic students and then AI/AN students.

[FIGURE 3 OMITTED]

Figure 4 shows the same disaggregation for students with disability. African-American students with disabilities had the highest percentage of student days lost overall, followed by Hispanic students, AI/AN students, White students, and Asians/Pacific Islanders. In the category of removal to alternative education, African-American students lost the highest percentage of student days, followed by AI/ AN, Hispanic, White, and Asian/Pacific Islander students.

[FIGURE 4 OMITTED]

Table 3 presents the outcomes of our follow-up ANOVA. Because the interaction between disability status and student race/ethnicity was statistically non-significant, F (4, 60317) = 2.199, p = .066, we interpreted the main effects of race/ethnicity and disability status. The effect of race/ethnicity in "days lost" was significant, F (4, 60317) = 17.927, p <.0005, as was the effect of disability status, F (1, 60317) = 5.505, p = .019. The amount of explained variance associated with race/ethnicity was .10%, while the amount of explained variance associated with disability status was .01%. Both factors explained extremely small amounts of the total variance in our model, suggesting that, while race/ethnicity does have a slightly stronger impact on the number of days students are excluded from the classroom than disability status, other factors not included in our model are likely to play important roles. Follow-up post-hoc Tukey tests indicated that with regards to number of days lost, AI/AN students (M = 4.66, SD = 13.86) differed significantly from Hispanic (M = 5.80, SD = 20.01) students (p = .046), Asian students differed significantly from African-American (M = 4.82, SD = 15.38), Hispanic, and White (M = 4.50, SD = 16.60) students (p = .007, p < .0005, and p = .012 respectively), and Hispanic student differed significantly from African-American (p = .017) and White students (p <0005).

Discussion

Taken together, our analyses showed that students from traditional minority backgrounds, especially students from African-American, Hispanic and Al/AN backgrounds, were disproportionately over-represented in exclusionary discipline actions and lost the greatest number of days to those discipline actions. These patterns were constant across students with and without disability and appear to reflect the persistent national trends documenting poorer school outcomes for non-White students compared to White students (Aud et al., 2010). It is especially noteworthy that AI/AN students disciplinary outcomes were very similar to those for African-American and Hispanic students. AI/AN students lost twice as many days to disciplinary exclusion as their White peers, and AI/AN students with disability experienced the highest rates of removal to alternative education. Given the reciprocal relationship between academic achievement and social behavior (Algozzine et al., 2011), there seems to be little doubt that the persistent disproportionate over-representation of African-American, AI/AN and Hispanic students in exclusionary disciplinary actions is inconsistent with the mission of the United States public education system to educate every child.

Based on what we know about national trends and based on the outcomes of our current analyses, one needs to question what can be done to produce greater equity among students from diverse racial/ ethnic backgrounds. Clearly, much more research is needed to fully understand the differential patterns of discipline outcomes across students from different racial/ethnic groups. Nuanced analyses of data are necessary to provide valuable insights into how discipline outcomes differ precisely across student groups. Often, these research efforts are hindered by lack of available data or unknown reliability of extant data. Datasets that contain large numbers of records with student race/ethnicity recorded as "unknown" make any study of disproportionate discipline outcomes difficult. Similarly, datasets reflecting disciplinary actions whose definitions vary across teachers and schools make the generalizability of patterns across data sources difficult.

Fortunately, availability of information on student race is increasing in the wake of the new U.S. Department of Education policy on recording student race/ethnicity. As indicated in Table 1, only .62% of students represented in our dataset had an ethnicity coded as "unknown." The newly introduced racial/ethnic category of "multiracial" makes comparisons with previous years when this category did not yet exist difficult. However, "multiracial" students are a growing subgroup whose discipline outcomes need to be carefully examined. A focus on disciplinary actions that are clearly defined by a state educational agency (SEA) also facilitates interpretation of patterns within a given state. For example, our dataset reflected exclusionary practices as defined by the SEA. Given that disciplinary incidents leading up to exclusionary discipline actions involve multiple people (student, teacher, administrator) and are therefore observer dependent, a certain amount of variability across observers is to be expected. Although this variability can introduce error into analytical outcomes, consistently recorded data can still yield useful insights into overall patterns.

Based on the outcomes of our analyses, it appears also important to focus on student groups who are relatively small in number. AI/AN students, for example, represent only 1% of the U.S. student population (Office for Civil Rights, 2012). These small numbers of AI/AN students make comparison with other groups analytically difficult. As a result, AI/AN students are often excluded from disproportionality studies. Our dataset reflected discipline outcomes for a state where AI/AN students represented 2.3% of the total student population, the 10th largest AI/AN student population in the U.S. (Faircloth & Tippe-connic, 2010). Inclusion of AI/AN students in our analyses revealed important information about how schools tend to discipline AI/AN students in comparison to students from other racial/ethnic groups.

Because quantitative studies of discipline outcomes for AI/AN students are complicated by small sample sizes making statistical outcomes difficult to interpret, a greater emphasis on qualitative studies might be needed to provide critical insights into how and why AI/ AN students are disciplined. Qualitative studies conducted in collaboration with the Native American community could provide needed insights as well as promote mutual trust between AI/AN students and their families and largely White school personnel. The National Congress of American Indians (NCAI) and the National Indian Education Association (NIEA) have formulated a number of recommendations to improve educational outcomes of AI/AN students. These recommendations include (a) an emphasis on Native language and culture to promote positive identity development of AI/AN students, (b) strengthening awareness and knowledge of Native language and culture among school personnel, and (c) strong parent participation in school policy development (CHiXapkaid et al., 2008; Faircloth & Tippeconnic, 2000). However, the Native American community recognizes that recommendations derived from qualitative data might be difficult to accept for school personnel that is often trained in quantitative data collection and analyses. The NCAI states: "we have limited statistical data showing that Native language instruction directly improves academic success [.....]. Therefore it is critically important to have sustainable funding for research that will demonstrate this statistical correlation." (NCIA/NIEA, 2010, p. 4).

Efforts exist to establish quantitative evidence of AI/AN students' school experiences. The angoing National Indian Education Study (NIES; http://nces.ed.gov/nationsreportcard/nies/study_over-view.asp) biannually collects survey data from AI/AN students and their teachers to examine relationships between the Native American community's recommendations and students' school experiences. Greater engagement with this data source, and greater collaborations across methodological boundaries might ultimately benefit AI/AN students.

Limitations

A number of limitations need to be considered when interpreting the results of our study. First, our dataset contained information only on whether a student had a disability or not; it did not provide information on the type of disability with which the student was classified. Therefore, we were unable to provide a more fine-grained analysis of racial/ethnic disproportionate representations within exclusionary discipline actions by type of disability. Second, state-wide enrollment by disability status was provided only for 5 racial/ethnic categories, omitting multi-racial students and those whose ethnicity was unknown. Outcomes for multi-racial students could therefore not be calculated. Third, small overall numbers of expulsions, and their highly unequal distributions across student groups resulted in difficult to interpret phi-values to assess the effect sizes of Chi-Square analyses. Fourth, unequal cell sizes within our ANOVA model made outcomes difficult to interpret. Because the state had an overall small enrollment of AI/AN, Asian, and African-American students, robust statistical models are difficult to construct and outcomes are therefore somewhat unstable and cannot be generalized to other settings.

Also, the extremely small percentage of variance explained by race/ethnicity and disability status in our ANOVA clearly suggests that there are other factors that need to be included in future analyses to achieve a better understanding of what contributes to racial/ethnic disproportionality in exclusionary discipline. The relationships between student race, disability, and disciplinary outcomes are clearly complex and involve many more variables than were present in our dataset. For example teacher demographics (e.g. race, years of experience, professional development access) and school environment characteristics (urbanicity, enrollment size, majority enrollment) are highly likely to impact discipline outcomes for students with and without disability from various racial/ethnic backgrounds. Further study including these variables and more sophisticated analytical procedures taking nesting of students in classrooms and schools into consideration is clearly needed

Finally, it is important to note that, although our data were based on disciplinary actions defined by the SEA, differential interpretations of SEA guidelines by individual school districts can still occur and introduce error into the analytical outcomes.

References

Achilles, G. M., McLaughlin, M. J., & Croninger, R. G. (2007). Sociocultural correlates of disciplinary exclusion among students with emotional, behavioral, and learning disabilities in the SEELS national dataset. Journal of Emotional and Behavioral Disorders, 15, 33-45.

Algozzine, B., Wang, C., & Violette, A.S. (2011). Reexamining the relationship between academic achievement and social behavior. Journal of Positive Behavior Interventions 13, 3-16.

Aud, S., Fox, M. A., & KewalRamani, A. (July 2010). Status and trends in the education of racial and ethnic groups. (NCES 2010-015). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office.

Carr, S. (April 20, 2010). Louisiana school suspension, expulsion rates criticized as excessive. The Times Picayune. Available at http://www.nola.com/education/index.ssf/2010/04/1ouisiana_school_suspension_ex.html

CHiXapkaid, Banks-Joseph, S. R., Inglebret, E., McCubbin, L. Sievers, J., Bruna, L., & Sanyal, N. (2008). From where the sun rises: Addressing the educational achievement of Native Americans in Washington State. Pullman, WA: Washington State University, Clearinghouse on Native Teaching and Learning.

Christie, A. C., Jolivette, K., & Nelson, C.M. (2005). Breaking the school to prison pipeline: Identifying school risk and protective factors for youth delinquency. Exceptionality, 13, 69-88.

Costenbader, V., & Markson, S. (1998). School suspension: A study with secondary school students. Journal of School Psychology, 36, 59-82.

Cregor, M., & Hewitt, D. (January/February 2011). Dismantling the school-to-prison pipeline: A survey from the field. Poverty and Race, 20(1). Available at www.thedefendersonline.com/2011/01/29/disrnantling-the-school-to-prison-pipeline-a-survey-from-the-field/

Fabelo, T., Thompson, M. D., Plotkin, M., Carmichael, D., Marchbanks, M. P., & Booth, E. A. (2011). Breaking schools' rules: A statewide study of how school discipline relates to students' success and juvenile justice involvement. New York: Council of State Governments Justice Center. Retrieved from http://justicecenter.csg.org/resource/juveniles

Faircloth, S. C., & Tippeconnic, III, J. W. (2000). Issues in the education of American Indian and Alaska Native students with disabilities. ERIC Digest ED448009 2000-12-00.

Faircloth, S. C., & Tippeconnic, III, J. W. (2010). The dropout/graduation rate crisis among American Indian and Alaska Native Students: Failure to respond places the future of native peoples at risk. Los Angeles, CA: The Civil Rights Project/Proyecto Derechos Ci-viles at UCLA. Retrieved from www.civilrightsproject.ucla.edu.

Gregory, A., & Ripski, M. B. (2008). Adolescent trust in teachers: Implications for behavior in high school classrooms. School Psychology Review, 37, 337-353.

Howell, D. (2002). Statistical methods for psychology. (5th ed.) Pacific Grove, CA: Duxbury.

Kaufman, J. S., Jaser, S. S., Vaughan, E. L., Reynolds, J. S., Di Donato, J., Bernard, S. N., & Hernandez-Brereton, M. (2010). Patterns in office discipline referral data by grade, race/ethnicity, and gender. Journal of Positive Behavior Interventions, 12, 44-54.

Krezmien, M. P., Leone, P. E., & Achilles, G. M. (2006). Suspension, race, and disability: Analysis of statewide practices and reporting. Journal of Emotional and Behavioral Disorders, 14, 217226.

Leinhardt, G., & Leinhardt, S. (1980). Exploratory data analysis: New tools for the analysis of empirical data. Review of Research in Education, 8, 85-157.

Losen, D. J., & Skiba, R. J. (2010). Suspended education: Urban middle schools in crisis. Indiana University: Center for Evaluation and Education Policy.

National Congress of American Indians/National Indian Education Association (2010). National tribal priorities for Indian education. Washington, DC: Author. Available at: http://www.niea.org/Policy.aspx

National Forum on Education Statistics, Race/Ethnicity Data Implementation Task Force. (2008). Managing an identity crisis: Forum guide to implementing new federal race and ethnicity categories (NFES 2008-802). National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education. Washington, DC.

Office for Civil Rights (2012). The transformed civil rights data collection. United States Department of Education, Washington, DC.

Political Research Associates (2005). Defending justice: An activist resource kit. Available at: http://www.defendingjustice.org/pdfs/factsheets/10-Face/o20SheetT020-[degrees]/020System%20asT020Racist.pdf

Raffaele Mendez, L. M., & Knoff, H. M. (2003). Who gets suspended from school and why: A demographic analysis of schools and disciplinary infractions in a large school district. Education and Treatment of Children, 26(1), 30-51.

Schwarz, A. (July 19, 2011). School discipline study raises fresh questions. The New York Times.

Skiba, R. J., Homer, R. H., Chung, C., Rausch, M. K., May, S., & Tobin, T. (2011). Race is not neutral: A national investigation of African American and Latino disproportionality in school discipline. School Psychology Review, 40, 85-107.

Skiba, R., & Rausch, M. K. (2006). School disciplinary systems: Alternatives to suspension and expulsion. in G. G. Bear & K. M. Minke (Eds.), Children's Needs III: Development, Prevention, and Intervention (pp. 87-102). Bethesda, MD: National Association of School Psychologists.

Skiba, R., & Sprague, J. (2008). Safety without suspensions. Educational Leadership, 66(1), 8-43.

Sughrue, J. A. (2003). Zero tolerance for children: Two wrongs do not make a right. Educational Administration Quarterly, 39,238-258.

Theriot, M.T Craun, S. W., & Dupper, D. R. (2010). Multilevel evaluation of factors predicting school exclusion among middle and high school students. Children and Youth Services Review, 32, 13-19.

Townsend, B.J. (2000). The disproportionate discipline of African-American learners: Reducing school suspension and expulsions. Exceptional Children, 66, 381-391.

Vincent, C. G., & Tobin, T. J. (2011). An examination of the relationship between implementation of school-wide positive behavior support (SWPBS) and exclusion of students from various ethnic backgrounds with and without disabilities. Journal of Emotional and Behavioral Disorders, 19, 217-232.

Wu, S. C., Pink, W. T., Crain, R. L., & Moles, 0. (1982). Student suspension: A critical reappraisal. The Urban Review, 14, 245-303.

Correspondence to Claudia Vincent, Institute on Violence and Destructive Behavior, 1265 University of Oregon, Eugene, OR 97403-1265; e-mail: clavin@uoregon.edu.

Claudia G. Vincent Jeffrey R. Sprague Tary J. Tobin University of Oregon
Table 1
Number and percent of students with disciplinary
exclusions and well as state-wide enrollments by
disability status and racial/ethnic categories.

                      Students with Disciplinary
                     Exclusions
                     Without        With      Total     Without
                   Disability  Disability             Disability

                       Number      Number     Number      Number
 Race               (Percent)   (Percent)  (Percent)   (Percent)
/Ethnicity
AI/AN                    1507         469       1976        8524
                       (3.06)      (3.17)     (3.08)      (1.82)

Asian/Paclslander        1395         208       1603       23861
                       (2.83)      (1.41)     (2.50)      (5.10)

Hispanic                13313        3045      16358       96529
                      (27.00)     (20.61)    (25.52)     (20.64)

AfrAm                    2618        1054       3672       12169
                       (5.31)      (7.13)     (5.72)      (2.60)

White                   28932        9419      38351      326585
                      (58.67)     (63.74)    (59.84)     (69.64)

Multiracial              1219         506       1725
                       (2.47)      (3.42)     (2.69)

Unknown                   326          77        403
                        (.66)       (.52)      (.62)

Total                   49310       14778     64,088      467668
                        (100)       (100)      (100)       (100)

                   State-wide Enrollment

                         With      Total
                   Disability

                       Number     Number
                    (Percent)  (Percent)

AI/AN                    2242     10,766
                       (3.15)     (1.93)

Asian/Paclslander        2018     25,879
                       (2.84)     (4.63)

Hispanic                12636    109,165
                      (17.77)    (19.52)

AfrAm                    3231     15,400
                       (4.54)     (2.75)

White                   50989    377,574
                      (71.70)    (67.51)

Multiracial                       15,140
                                  (2.71)

Unknown                            5,327
                                   (.95)

Total                   71116    559,251
                        (100)      (100)


Table 2
Comparison of rates of exclusionary discipline
between non-White and White students. (All tests had 1 df)

                  Without Disability          With Disability

                 [X.sup.2]   p-value  Phi  [X.sup.2]  p-value
                   value                         value

Truancy  AI/AN      1531.31   <.0001  .22     163.14   <.0001

         Asian      1160.03    <0001  .19     161.16   <.0001

         Hisp       2666.04   <.0001  .29     366.65   <.0001

         AfrAm      4002.80   <.0001  .35     611.58   <.0001

ISS      AI/AN      1913.55   <.0001  .22     183.51   <.0001

         Asian.     2133.94   <.0001  .23     291.85   <.0001

         Hisp       7470.54   <.0001  .43     905.94   <.0001

         AfrAm      1894.53   <.0001  .22     258.31   <.0001

OSS      AI/AN       908.26   <.0O01  .17   11796.64   <.0001

         Asian       984.21   <.0001  .17   12158.49   <.0001

         Hisp        988.25   <.0001  .17   22271.85   <.0001

         AfrAm      3702.13   <.000l  .34   11327.25   <.0001

EXP      AI/AN        43.69   <.0001  .18    2325.74   <.0001

         Asian        53.86   <.0001  .20     2298.3   <.0001

         Hisp         71.17   <.0001  .23    3102.53   <.0001

         AfrAm        71.27   <.0001  .23     2466.3   <.0001

AltEd    AI/AN                                 46.11   <.0001

         Asian                                  2.05    .1522

         Hisp                                  41.96    .2207

         AfrAm                                  1.86    .1726

          Phi

Truancy   .14

          .14

          .21

          .27

ISS       .12

          .15

          .27

          .14

OSS       .60

          .61

          .82

          .59

EXP      1.31

         1.30

         1.51

         1.35

AltEd     .43

          .09

          .41

          .09


Table 3
ANOVA Summary Table

     Source                SS     df        MS       F       P

Between subjects

Race/Ethnicity       21335.047      4  5333.762  17.927  <.0005

Disability status     1637.900      1  1637.900   5.505    .019

Disability x Race     2616.678      4   654.170   2.199    .066

Error              17945539.46  60317    297.52

      Source       [[eta].sup.2]

Between subjects

Race/Ethnicity             .0012

Disability status          .0001

Disability x Race          .0001

Error
Gale Copyright:
Copyright 2012 Gale, Cengage Learning. All rights reserved.