Abstract:
Ruck and Wortley examine perceptions of differential treatment relating to school disciplinary practices in a racially and ethnically diverse sample of high school students. The implications of these findings are discussed, and suggestions for future research are outlined.

Full Text:

This study examined perceptions of differential treatment relating to school disciplinary practices in a racially and ethnically diverse sample (Black, South Asian, Asian, White, and "other" racial/ethnic background) of high school students (N = 1870). Participants completed detailed individual questionnaires assessing general perceptions of school disciplinary practices and various aspects of the school environment. Results indicated that racial/ethnic minority students are much more likely than White students to perceive discrimination with respect to teacher treatment, school suspension, use of police by school authorities, and police treatment at school. Multivariate analyses revealed that such perceptions are particular strong for Black students. In addition, a number of other variables such as gender, socioeconomic status, age of immigration, and views of school climate also predicted students' perceptions of differential treatment toward members of their racial/ethnic group. The implications of these findings are discussed, and suggestions for future research are outlined.

INTRODUCTION

There is no shortage of empirical evidence highlighting the difficulties encountered within the educational system for minority students in the United States and Canada. Racial and ethnic minority students often face poorer educational outcomes in variety of areas than do their mainstream counterparts. Some of the negative outcomes associated with schooling which minority children face, reported in the American literature, include lower test scores, poor grades, low attendance, grade retention, and early school leaving (Connell et al., 1994; Ensminger et al., 1996; Ford, 1993; Fordham and Ogbu, 1986; Gay, 1989; Humphreys, 1988; Irvine, 1991; Shapiro et al., 1993; Taylor et al., 1994; Whaley and Smyler, 1998). Canadian research also suggests that minority students, especially Black students, are more likely to be enrolled in basic and general level academic programs (Brown, 1993; Cheng, 1995; Cheng et al., 1993, all cited in Dei et al., 1995; James 1990; Samunda et al., 1989; Solomon, 1992) and show disproportionately higher levels of school dropout than do other students (Brathwiate and James, 1996; Brown, 1993; Dei et al., 1995, 1997).

In addition, and of primary concern with regard to the present investigation, racial/ethnic minority students are also more likely to be disciplined at school than White students. For example, there is overwhelming American evidence that students of color, especially African American males, are much more likely to be suspended from school than are their mainstream counterparts (Banks and Banks, 1993; Bennett and Harris, 1982; Calabrese and Poe, 1990; Darling-Hammond, 1994; England et al., 1982; Felice, 1981; Irvine, 1991; Jones, 1989; Kaeser, 1979; Sheets and Gay, 1996). Although a variety of disciplinary actions can be taken by schools, suspensions are one of the most severe in that they result in the removal of students from the school and thus a decrease in student's instructional time (Williams, 1989). Available information also suggests that disciplinary problems, such as suspensions, are a factor contributing to the high incidence of early school leaving among minority students (Children's Defense Fund, 1974; Dei et al., 1997; Jones, 1989; Rumberger, 1983; Williams, 1989; Wu et al., 1982).

While there is some evidence that racial and ethnic minority students often believe that they will receive harsher or more public punishment for engaging in the same behavior than White students (Marcus et al., 1991; Murray and Clarke, 1990), there has been little systematic research directly exploring racial/ethnic minority student's views and beliefs pertaining to school disciplinary practices. Two recent ethnographic studies, however, provide some direct evidence pertaining to how racial minority high school students' may view school disciplinary practices.

In a Canadian study of a group of Black, predominately male, low socioeconomic status (SES), urban high school students, Solomon (1992) found that with regard to school disciplinary practices these students tended to view school discipline as being administered arbitrarily by school authorities. For example, these students perceived that they were more often suspended than White students for engaging in the same types of behaviors. In addition, Solomon noted that the Black students in his study felt that the schools' regular use of police at school-related events was primarily to control and supervise the behavior of members of their racial/ethnic group. In a later investigation, Sheets (1996) examined a small group of racially and ethnically diverse students' (e.g., African American, Chicano, European, and Filipino) perceptions of classroom discipline in an urban American high school. One of the main findings reported was that while there were differences in perceptions of disciplinary practices according to students' race/ethnicity, perceptions tended to be more similar among students of color. In addition, the racial minority students interviewed felt that racism was the main reason for the perceived differential application of school disciplinary actions. Taken together the findings from these studies, although limited because of the size of the samples, are instructive in that they offer some insight into how minority high school students' perceive school discipline and experience the educational system. However, there is clearly a need for more quantitatively oriented research in this area.

The importance of examining student's perceptions or views should not be underestimated. How an individual perceives his or her environment may be more important than "objective reality," in that one's perceptions will influence how one responds to the environment (Bronfrenbrenner, 1979). Thus, if certain groups of students perceive unequal or unfair treatment than that perception is their reality (Holliday, 1985; Marcus et al., 1991), which has important implications if schools are to develop measures to ensure that students from all racial and ethnic groups perceive equal educational opportunity. In addition, examining students' perceptions of various aspects of their educational experiences also provides a unique view of how young people interpret aspects of their social world (Berndt et al., 1989, 1990, all cited in Ford, 1993).

This study was conducted with a large multiethnic sample of high school students from the Metropolitan Toronto area. Canada, like the United States, has historically been a racially and ethnically heterogeneous society (Moodley, 1995), and minority students in Canadian schools are no less immune to the detrimental effects of racism, discrimination, and prejudice than are their American counterparts (Alladin, 1996; Braithwaith, 1989; Cheng and Soudak, 1994; Dei et al., 1995, 1997; James, 1990; Karaucharey, 1992; Solomon, 1992). Moreover, no published studies have systematically examined a large and sociodemographically heterogeneous sample of students' perceptions pertaining to differential treatment with regard to school disciplinary practices.

Furthermore, over the last decade the police have become a major presence in Canadian schools in dealing with unacceptable student behaviors (e.g., fights, thefts, etc.), often through the laying of criminal charges (Ministry of Education and Training, 1994). While there is a general dearth of research examining minority students' perceptions relating to school disciplinary action, there is even less information available examining racial/ethnic minority students' perceptions of using the police to control student behavior. However, the available research examining racial and ethnic minorities' perceptions of discriminatory treatment in the criminal justice system (e.g., police treatment) may provide some insight into how young people from diverse racial/ethnic groups are likely to perceive police treatment at school. For instance, American as well as Canadian studies have found that racial and ethnic minorities believe that the police are more likely to discriminate against members of minority groups than against Whites (California Judicial Council, 1993; Hagan and Albonetti, 1982; Wortley, 1996). Determining whether similar perceptions are held in a racially/ethnically diverse sample of Canadian students with regard to differential treatment relating to school authorities and sanctions, including the use of the police, is an important goal of the current study.

This study seeks to move beyond the primarily qualitative and descriptive work on this topic. By including participants from a wide range of racial/ethnic backgrounds, we were able to investigate whether perceptions of school disciplinary actions vary across different racial and ethnic groups rather than being limited to traditional Black-- White comparisons. We also employed a multivariate approach to examine the various factors (e.g., race/ethnicity, gender, attitudinal variables) that may predict students' perceptions of differential treatment. The lack of empirical research in this area necessitated that this study be regarded as primarily exploratory in nature.

METHOD

Sample

Data were collected by the Commission on Systemic Racism in the Ontario Criminal Justice System (Report of the Commission on Systemic Racism in the Ontario Criminal Justice System, 1995), as part of a series of projects examining the extent of systemic racism in procedures, policies, and practices that make up the Ontario criminal justice system.

The sample consisted of 1870 students from Grade 10 and 12 classes from 11 randomly selected high schools from a racially and ethnically diverse school district in the Metropolitan Toronto area of Ontario, Canada. According to available class lists a total of 2487 students were enrolled in the classes selected for study. The response rate was approximately 65%. Eleven percent of eligible students were either not present at school or unavailable when questionnaires were administered. Students took part in the study only if their parent or guardian had given written informed consent in response to a letter describing the study sent home with each student by the principals of the school. In addition, student verbal assent was obtained from all students taking part in the study. Approximately 9% of students did not return parental consent forms and were thus not permitted to take part in the study. Less than 2% of participants actually refused to participate.

In addition to other demographic questions, participants were asked to indicate their racial/ethnic background. The sample was 49% White or European descent, 18% Asian descent, 14% Black or African descent, and 8% South Asian descent. Students from "other" racial/ethnic backgrounds (e.g., Hispanic, Aboriginal, and mixed racial background) combined accounted for 11% of the sample. The sample was approximately equally divided by gender (56% female, 44% male), grade (54% Grade 10, 46% Grade 12), and age group (54% 15-16 years; 46% 17-18 years). More detailed demographic characteristics of the entire sample are provided in Table I.

Survey Development and Procedures

Data were provided via anonymous individual questionnaires administered in supervised group settings (e.g., home classrooms, school auditorium). Surveys focused on demographic information, perceptions of differential treatment, and general perceptions of the school environment. Most survey items employed a standard Likerttype response format. A number of open-ended questions served to provide additional information.

Becaue of the lack of existing measures, questionnaire items used in the current study were developed on the basis of conceptual considerations, a review of the appropriate research literature on survey and scale development (see Dawes, 1987; Schwarz, 1999; Schwarz et al., 1998) and focus groups. For example, during focus groups certain variables (e.g., school violence and segregation) were identified by participants as factors that may impact on perceptions of differential school treatment. In addition, all questionnaire items were developed in consultation with school board research staff. The consultation process provided the opportunity to address any potential concerns about various items and allowed items to be either modified, omitted, or added to the survey (e.g., several items not deemed appropriate were dropped from the survey). The questionnaire was piloted with a group of 30 racially and ethnically diverse high school students from a school not taking part in the study. On the basis of feedback from pilot testing, we were able to ensure that all items were comprehensible, linguistically appropriate and meaningful.

Table I.

Prior to distributing the questionnaires, students were reminded of the confidential and voluntary nature of the study. Students were also reminded not to write their names or make any other identifying marks on the questionnaires and were assured of complete anonymity. Completion of questionnaires took approximately 45-60 min.

Measures

Demographic Information

Information was obtained on participant's race/ ethnicity, age, grade, gender, socioeconomic status, family structure, country of birth, and age immigrated to Canada (for foreign-born participants). The multivariate analyses controlled for race/ethnicity, age, gender, socioeconomic status, family structure, and age of immigration.

Four dummy variables representing, Black students, Asian students, South Asian students, and students from "other" racial/ethnic backgrounds were created. White students served as the comparison group. Dummy variables for the age and gender of the respondents were also included. It should be noted that age was highly correlated with grade (r = 0.91). Almost all of the respondents (98%) in the youngest category were in Grade 10, while 93% of the respondents in the oldest age category were in Grade 12. Therefore, to prevent problems with multicollinearity, grade is left out of the multivariate analyses presented below.

Socioeconomic status is often a very difficult variable to measure when studying student populations (Brownfield, 1986). Our pilot study, for example, revealed that most students could not provide an accurate estimate of their family income or even report their parents' educational level. However, most respondents were able to report whether their parents were working or not and the type of occupation their parents held. Thus, in the current study parental occupation serves as a measure of SES. Parental occupation was coded employing categories adapted from Pineo et al. (1977). In cases were both parents were employed the highest ranking occupation was chosen. There was a significant difference between racial/ethnic groups in terms of parental occupation (chi^sup 2^ = 93.51, p < 0.001). Thirty-eight percent of the Black students were from homes where parents were unemployed while among White, Asian, South Asian, and students from "other" racial/ethnic backgrounds the percentages were 17, 23, 26, and 28%, respectively.

Family structure was determined by asking respondents if they currently live with both parents, mother only, father only, or have some other type of living arrangement (e.g., they live with another relative). The analyses included a dummy variable for lives with both parents (intact) versus all other forms (nonintact). Among racial groups there was also a significant difference in terms of family structure (chi^sup 2^ = 125.51, p < 0.001). Fifty-one percent of Black students in this study were from nonintact families compared to 35% of those from "other" racial/ethnic backgrounds, 24% of White students, 15% of Asian students, and 11% of South Asian students.

In terms of age of immigration to Canada, preliminary analyses indicated that while many students were recent immigrants, many others had migrated to Canada at a very young age and had essentially grown up "Canadian." Thus to distinguish between recent immigrants and those who had lived in Canada for many years, the current analyses focuses on age of immigration rather than birthplace. There were significant racial group differences in terms of the age at which foreign-born students immigrated to Canada, F(4, 1869) = 103.76, p < 0.001. Specifically, Tukey post hoc tests (p < 0.01 or better for all significant contrasts) for pairwise comparisons between ethnic/racial groups (means provided in parentheses) indicated that South Asian students (M = 6.04 years) immigrated at an older age than those from all other racial/ethnic groups. In addition, Black (M = 4.63 years) and Asian (M = 4.75 years) students immigrated at a significantly older age than those from "other" racial/ethnic backgrounds (M = 2.59) and White students (M = 0.58 years). Finally, students from "other" racial/ethnic backgrounds immigrated at an older age than White students.

Perceptions of Differential Treatment

Teacher Treatment. Students' perception of how they felt students from their racial group were treated by teachers at school was measured using a 6-point scale ("much better," "better," "the same," "worse," "much worse," and "don't know") for the following item: "Do you think teachers at your school treat students from your racial4 group better, worse or the same as students from other racial groups."

School Suspension. A single item was used to assess students' perceptions of differential treatment with regard to the likelihood of members of their racial/ethnic group being suspended from school. Using a 4-point scale ("more likely," "less likely," "no difference," "don't know"), participants responded to the question: "Do you think students from your racial group are more likely to be suspended than students from other racial groups for engaging in the same types of behavior."

Use of Police at School. A single item was used to assess students' perception of differential treatment with regard to having the police called on members of their racial/ethnic group. Using a 4-point scale ("more likely," "less likely," "no difference," "don't know"), participants responded to the question: "Do you think students from your racial group are more likely to have the police called on them than students from other racial groups?"

Police Treatment at School. A single item was used to examine students' perception of how members of their racial/ethnic group would be treated by the police at school. Using a 6-point scale ("much better," "better," "the same," "worse," "much worse," and "don't know"), students responded to the question: "If the police were called to your school, do you think they would treat students from your racial group better, worse or the same as students from other racial groups?"

For the current analyses each of the response categories described above for students' perceptions of differential treatment were recoded into dummy variables (e.g., response options for teacher treatment were recoded into 1 = treated worse or much worse; 0 = treated the same, better, much better, or don't know). This recoding procedure allows us to clearly identify those respondents who feel that students from their racial/ethnic group receive negative treatment at school.

General Perceptions of the School Environment

Intergroup Association. The extent to which respondents perceive that students associate or interact with peers from different racial groups at school was measured by having students respond to a single item using a 5-point scale, from 0 (not at all true) to 4 (very true). Students responded to the question: "In general, students of different races usually hang out with students of the same race." Higher mean scores indicated greater perceptions of students segregating themselves by race.

School Violence. Perceptions of personal safety and violence at school were obtained by having students respond to three items on a 5-point scale, ranging from 0 (strongly disagree) to 4 (strongly agree). The items included "I am concerned about my personal safety when I'm at school," "Fighting among students is common at my school," and "many students carry weapons school." These 3 items were combined to from a single school violence measure, with higher mean scores indicating greater perceptions of school violence. The scale possessed moderate internal consistency (coefficient alpha = 0.63).

Severity of School Punishment. The extent to which students believed that various types of behaviors should lead to severe forms of punishment was assessed for 10 different behaviors. Using 3 response categories (0 = no; 1 = don't know; 2 = yes), participant were first asked to indicate whether each of 10 behaviors should lead to being suspended from school and then asked if they felt that the school should have the police to deal with students who engage in each of the 10 behaviors. Typical types of behaviors were fighting at school-no weapons involved; fights involving weapons at school or during school-related activities; student threatening a another student; student threatening a teacher; sexually harassing someone at school, sexually assaulting someone at school; bringing weapons to school; bringing/using drugs or alcohol at school; creating a disturbance in class. The 20 responses to the items were combined into a single punishment scale. Higher means scores on the scale indicate the more punitive the respondents' attitude toward dealing with issues of student misbehavior. The scale possessed good internal consistency (coefficient at = 0.78).

RESULTS

Three types of analyses were conducted. First, analyses employing chi-square statistics for students' responses to individual items for perceptions of differential treatment. Second, one-way analyses of variance and post hoc comparisons (using Bonferroni contrasts) for student's responses to the various subscale items. Third, multivariate analyses employing logistic regression procedures were utilized to examine the independent variables that predict students' perceptions of differential treatment.

Perceptions of Differential Treatment

Table II provides the percentage of students by each racial/ethnic group indicating that members of their group would be subject to discriminatory treatment for each of the 4 types of differential school treatment items. As shown in Table II, chi-square tests for each type of differential treatment revealed overall findings of significant racial/ethnic group differences. To determine the exact nature of these differences for each type of differential treatment, pairwise comparisons (using chi-square tests corrected for continuity) between each racial/ethnic groups were conducted. The large number of multiple comparisons performed necessitated an alpha level of 0.01 to control for Type I errors.

For the question examining students' perceptions of differential teacher treatment, Black students were significantly more likely than South Asian, Asian, White, and students from "other" racial/ethnic backgrounds to perceive that teachers at their school treat students from their racial group worse or much worse than students from other racial groups (all ps < 0.001). In addition, South Asian students were more likely than White students to report that they would be treated worse or much worse by teachers at their school compared to students from other racial groups (p < 0.001).

In terms school suspension practices, Black students were more likely than South Asian, Asian, White, and students from "other" racial/ethnic backgrounds to believe that students from their racial group would face discriminatory treatment compared to students from other racial groups (all ps < 0.001). In addition, South Asian students were significantly more likely than Asian and White students to perceive discriminatory suspension practices directed toward members of their own racial group (all ps < 0.001). Students from "other" racial/ethnic backgrounds were significantly more likely than White students to believe that members of their own racial group were more likely to be suspended compared to students from other racial groups (p < 0.001).

For the question assessing students' perceptions of differential treatment with regard to the schools' use of the police, Black students were significantly more likely than South Asian, Asian, White, and students from "other" racial ethnic backgrounds to believe that students from their racial group were more likely to have the police called on them than did those from other racial groups (all ps < 0.001). Furthermore, South Asian students were more likely than Asian and White students to perceive discriminatory treatment in terms of the schools use of the police (p-values of <0.002 and <0.001, respectively). In addition, Asian students and students from "other" racial/ethnic backgrounds were significantly more likely than White students to believe that they would be subject to differential treatment with regard to the schools use of the police (all ps < 0.001).

Finally, in terms of students' perceptions of how the police would treat members of their racial group at school, Black students were significantly more likely than South Asian, Asian, White, and students from "other" racial/ethnic backgrounds to believe that they would be treated worse or much worse by the police at school (all ps < 0.001). Finally, South Asian, Asian, and students from "other" racial/ethnic backgrounds were all significantly more likely than White students to perceive discriminatory treatment in terms of how they would be treated by the police at school (all ps < 0.001).

General Perceptions of the School Environment

In terms of students' perceptions of the degree to which students from different racial/ethnic groups associated with each other at school there was a significant effect of race/ethnicity, F(4, 1869) = 8.99, p < 0.001. Subsequent post hoc tests (see Table III) indicated that Black and South Asian students viewed their schools as more racially segregated in terms of who students "hang out or associate with" than did White students and students from "other" racial/ethnic backgrounds. Significant differences according to race/ethnicity were also found in terms of students' perceptions of personal safety and violence at school, F(4, 1869) = 17.04, p < 0.001. Post hoc analyses (see Table III) indicated that South Asian and Black students were significantly more likely than Asian and White students to see their schools as violent and unsafe. In addition, South Asian students reported more violence at school than did those from "other" racial/ethnic groups, and students from "other" racial/backgrounds perceived more school violence than did Asian students. Finally, there were also significant racial/ethnic groups differences in students' belief regarding the severity of punishment with regard to various school-based offences, F(4, 1869) = 2.77, p < 0.05. According to post hoc tests (see Table III) South Asian students were more punitive in their views regarding school punishment than Black students.

Predictors of Perceptions of Differential Treatments

A series of logistic regression models were produced to determine the independent variables that are statistically significant predictors of students' perceptions of differential treatment after controlling for relevant demographic and attitudinal variables. Logistic regression was utilized as it makes no assumptions about the distribution of the predictor variables and can be used with any combination of discrete, continuous, and dichotomous variables, as compared to most alternatives (Tabachnick and Fidell, 1996). In the present series of analyses each logistic regression coefficient, which represents the natural log of the odds ratio, is estimated with all other predictors held constant (i.e., each predictor is evaluated while controlling for all other predictors simultaneously). The odds ratio (exponentiated coefficient) represents the increase or decrease of being in a particular outcome category of the dependent variable.5 As a further point of clarification, for the predictor variable of race/ethnicity White students were the comparison group and thus not entered into the regression equations; however, all remaining predictor variables include students from all 5 racial/ethnic groups. Logistic regression coefficients and odds ratios predicting students perceptions of differential treatment for teacher treatment, suspension, police contact, and police treatment at school are summarized in Table IV.

The results of the logistic regression analyses indicated that (1) minority students were all more likely than White students to perceive discriminatory treatment in terms of teacher treatment, school suspension practices, the use of police, and police treatment. In general, Black students were the most likely to perceive discriminatory treatment, followed by South Asian students, next students from "other" racial/ethnic backgrounds and followed by Asian students. Teacher treatment is the only case where this general pattern was not observed with Asian students being slightly more likely than students from "other" racial/ethnic backgrounds to perceive differential treatment; (2) males were more likely than females to perceive that members of their racial/ethnic group would be suspended, have the police called on them, and be treated worse by the police at school than other students. Gender, however, is not related to perceptions of teacher treatment; (3) SES was negatively related to the perception of racial/ethnic bias in police treatment. In general, lower SES students were more likely to perceive that students from their racial/ethnic groups will be treated worse by the police at school than higher SES respondents. SES, however, was not related to perceptions of bias in teacher treatment, school suspension practices or the use of the police in school; (4) students who were born in Canada or immigrated to Canada at an a young age were more likely to perceive differential treatment in terms of suspension, police involvement, and police contact toward members of their racial/ethnic group. However, age of immigration was not related to perceptions of teacher treatment; (5) students who view their school as being racially segregated in terms of who student associate or hang out with (i.e., low levels of intergroup association) were more likely to perceive that students from their racial/ethnic group will be treated worse by teachers, face suspension, have the police called on them, and be treated worse by the police at school; (6) students who feel their school is a violent and unsafe place were more likely to perceive biased treatment toward students from their racial/ethnic group in terms of all 4 dependent variables; and (7) finally, students who held conservative views regarding the severity of punishment for various school-based offences (i.e., believe in harsh or severe punishment for student misbehavior) were less likely to perceive that students from their racial/ethnic group would be discriminated against in terms of teacher treatment, school suspension, police contact, and police treatment.

DISCUSSION

The purpose of the present investigation was to examine Canadian high school students' perceptions of differential treatment pertaining to school disciplinary practices. A major strength of the current study was the inclusion of a large racially and ethnically heterogeneous sample. The results provide an interesting picture of students' views of differential treatment in the high school context, and raise some questions that require further research. The main findings are reviewed in detail below along with the study's general conclusions and directions for future research.

One of the most striking findings was that, in general, students from all four racial/ethnic minority groups were significantly more likely than White students to perceive discrimination in terms of various aspects of their treatment at school. In addition, the results also revealed that in comparison to students from all other minority groups, Black students viewed themselves as being at an obvious disadvantage, especially with regard to their dealings with the police at school. For example, Black students were approximately 32 times more likely than White students to perceive discrimination with respect to the use of police at school and 27 times more likely to perceive that they would be treated worse by the police at school. These findings are consistent, in part, with the research described earlier (e.g., California Judicial Council, 1993; Hagan and Albonetti, 1982; Wortley, 1996) which has also found that Blacks tend to be much more likely than members of other racial and ethnic groups to perceive discrimination in the criminal justice system. On the basis of the present findings, minority status appears to be an extremely important predictor of student's perceptions of inequality regarding how they are treated by both school authorities and police at school.

Although race and ethnicity emerged as the strongest predictors of perceptions of differential treatment, the results of the logistic regression analyses also demonstrated that a number of additional factors, including gender, SES, age of immigration, and students' general perceptions of the school environment (e.g., attitudes toward school punishment) were also significant predictors of students' perceptions of differential treatment. The fact that males in this study were more likely than females to believe that there would be bias in terms of their treatment relating to school punishment concurs with the available research on school discipline, which indicates that males are far more likely than females to be disciplined at school (Bennett and Harris, 1982; Shaw and Braden, 1990). In addition, the finding that SES predicted students' perceptions of police treatment is in line with earlier research which has found that students from low-SES communities tend to have more negative attitudes toward police than their higher socioeconomic counterparts (Derbyshire, 1968; Griffths, 1982).

Age of immigration was associated with students' perceptions of differential treatment, such that the longer respondents had lived in Canada the more likely they were to perceive bias toward members of their racial/ethnic group. We speculate that this finding might reflect the fact that these students having had a longer exposure to the social inequalities or social mechanisms such as racism, discrimination, and prejudice, that exist in Canadian society, and thus they may be more likely than students who are recent immigrants, to have developed negative expectations about how members of their racial or ethnic group will be treated. Another possibility is that students who are relatively recent immigrants may have experienced greater "oppression" or more severe patterns of school discipline in their countries of origin and therefore how they are presently treated is by comparison not seen as overly problematic.

Furthermore, we found that students who were conservative in their views regarding school punishment (i.e., held punitive views regarding school disciplinary practices) were less likely to perceive differential treatment toward members of their racial/ethnic group. One possible interpretation is that these students may accept harsher sanctions and punishments as a legitimate way for their school to maintain order and control. Therefore, these students may see the individual student as being responsible for the punishment and treatment he or she receives, rather than there being any sort of racial/ethnic bias operating on part of the school. However, additional research is clearly required to determine the exact nature of this finding.

How might we explain the fact that, independent of race/ethnicity and other factors, students who viewed their schools as more racially and ethnically segregated in terms of student social interaction were more likely to believe that students from their racial/ethnic group would receive differential treatment by both school authorities and the police? There is available evidence that students who segregate themselves by race or ethnicity may hold less positive attitudes toward students from other racial or ethnic groups (Phinney et al., 1997). Therefore, in this study, it may be case that students who view their schools as being places where students do not generally associate with students from different racial/ethnic groups may themselves engage in fewer cross-race contacts and as a result hold less positive attitudes toward students from other racial/ethnic groups. Further, we can speculate that, within the context of the school, negative attitudes toward students from other racial/ethnic groups could manifest themselves in a variety of ways, with one such way being that students might come to see their group as receiving less favorable treatment compared to students from all other racial/groups. This view is illustrated in the comments of a White female student who described what she felt was differential treatment toward students from her racial/ethnic group, "If anything I believe that White students are being discriminated against. If I were to wear a shirt which said "proud to be White" I'd be suspended ... But other races can wear those types of shirts [at school] and it's expected." However, until additional research provides support for this finding, caution is warranted, with regard to the present interpretation.

The findings also indicate that students who perceived their schools as violent and unsafe places were more likely to believe that students from their racial/ethnic group would receive differential treatment by the school and police. One possible explanation for this finding is that these students are in fact attending schools that have high levels of violence and when school violence increases, so does the surveillance of students by teachers and other school authorities. Increased surveillance, in turn, may lead to an increase in the number of students who are punished or disciplined. Thus students who attend schools marked by violence and personal safety issues may feel unfairly targeted by school authorities and feel that the extra monitoring of student behavior is unwarranted. This may be particularly true for racial and ethnic minority students who may feel that they are frequently blamed or scapegoated for the violence that takes place in the school context (Murray and Clark, 1990; Sheets, 1996; Solomon, 1992).

Finally, the present results also suggest an apparent hierarchy with respect to students' perceptions of differential treatment with Black students generally being the most likely to perceive bias followed by South Asian students, students' from "other" racial/ethnic backgrounds, Asian students, and finally White students. This is consistent with the view that the darker the skin color the greater the social penalties that exist (West, 1993). How skin color and other racial features influence minority students' perceptions of opportunity within the school context is an area that needs to be further addressed.

There are some limitations that must be considered with regard to this study. First, the problems inherent with self-report data, such as impression management or deception, are well documented (Hartmann, 1992), especially when the data is potentially sensitive in nature. While we believe that the conditions of anonymity used in this study may have limited certain social desirability tendencies; caution is still appropriate with regard to the generalizability of these findings. Second, although the results clearly suggest that racial/ethnic minority students believe that they are more likely to face suspension or be subjected to police contact at school than White students, there is no quantitative evidence to either refute or support whether this belief reflects the true state of affairs, as Canadian schools do not compile disciplinary statistics by race or ethnicity (see Paquette, 1992, 1990). Nevertheless, although one could question the relative "accuracy" of these perceptions, we suggest the point to be emphasized here is that if these students' perceive differential treatment, even if no differential treatment exists, these beliefs have a psychological reality for students which undoubtedly impacts on their schooling experience.

These limitations notwithstanding, this study provides valuable insight into how students from ethnically and racially diverse populations perceive an important aspect of their educational experience. Future research is needed to determine how these perceptions influence educational opportunities as well as other life outcomes for Canadian children of color. For instance, how do these types of perceptions influence minority students' attitudes and behaviors toward schooling? What is the influence of these perceptions on minority students' interactions with teachers? In addition, does the perception of school injustice reflect a wider minority perspective that acknowledges discrimination in all aspects of day-to-day life? Furthermore, future research should also examine intragroup variability to assess the range of views found within students from various racial and ethnic groups. While we deliberately limited this study to between-group comparisons, we are acutely aware of the dangers of portraying particular racial and ethnic groups as being monolithic in their thinking and attitudes.

In addition, we hope that future studies will be conducted in other settings (e.g., the United States) to help determine the generalizible features of our present findings. Finally, it is also our hope that continued research of this nature will serve to further legitimize the views of racial and ethnic minority students.

ACKNOWLEDGMENTS

We express our appreciation to the students and schools whose participation made the study possible. We also thank William Cross, Michelle Fine, and Daniel Keating for their helpful comments and suggestions on an earlier version of the manuscript. The technical assistance of Denese Coulbeck and Sue Elgie is also appreciated.