Who gets suspended from school and why: A demographic analysis of schools and disciplinary infractions in a large school district.
The purpose of this investigation was to examine out-of-school suspensions in a large, ethnically diverse school district by race, gender, school level, and infraction type. Such an analysis was undertaken to address gaps in the suspension literature regarding how suspension rates change across school levels for students of different genders and races and what types of infractions result in suspension for students in various demographic groups. Suspension data from the 1996-97 school year was examined for all (N=142) general education schools within one west central Florida school district. Results showed that the over-representation of Black males that has been cited consistently in the literature begins at the elementary school level and continues through high school. Black females also were suspended at a much higher rate than White and Hispanic females at all three school levels. Across school levels, most suspensions were for relatively minor misbehavior. Black males were over-represented in suspensions across almost all infraction types. Findings are discussed in terms of their implications for school discipline reform.
As public concern regarding school safety has grown in recent years, out-of-school suspension has been used with increasing frequency to respond to serious levels of student misbehavior and maintain a positive educational climate in schools (Brooks, Schiraldi, & Zeidenberg, 1999). However, because it involves the exclusion of students from the learning process, suspension frequently is perceived as one of the more extreme responses available to administrators within the continuum of various disciplinary options. Indeed, suspension typically is intended by administrators and perceived by students as a punishment (Costenbader & Markson, 1998; Mellard & Seybert, 1996). That is, in contrast to a consequence, suspension is delivered to punish an already-committed inappropriate act or behavior; it rarely has a logical, functional, or instructive connection to the offense or infraction; and it usually occurs in the absence of additional interventions that focus on teaching or reinforcing students' more prosocial or appropriate responses to difficult situations. Regardless of the rationale underlying it, repeated suspension has been linked to a variety of negative outcomes for students, including academic failure, negative school attitudes, grade retention, and school drop-out (Brooks et al., 1999; Nichols, Ludwin, & Iadicola, 1999).
From a problem-solving, prevention, and intervention perspective, it is important to know how many students are suspended from the different levels (i.e., elementary, middle, and high school) of our schools, whether there are important demographic trends, and what types of incidents are triggering the need for suspensions. These analyses are especially critical (a) in light of historic trends where Black students have been disproportionately suspended from schools across the country, especially when contrasted with White students; (b) given the need to protect special education students' rights when a disability influences the behavior that prompts the suspension; and (c) in the face of recent "zero tolerance" legislation whereby certain specified behaviors incur an immediate suspension (Brooks et al., 1999; Skiba, Peterson, & Williams, 1997; The Civil Rights Project, 2000). Clearly, by understanding the data and their implications, schools may be able to develop and implement more positive and effective scho ol-wide behavioral support systems (Lewis, Sugai, & Colvin, 1998; Sugai & Homer, 1999) such that the need for suspension substantially decreases, and it is used more as a strategic intervention than as a last, reactive resort.
The Incidence and Demographics of School Suspension
In reviewing recent history, out-of-school suspensions have increased during the past decade, and they are one of the most commonly used forms of discipline in the United States (Dupper & Bosch, 1996). In 1993, it was estimated that 1.5 million students missed one or more days of school per year because they had been suspended or expelled (U.S. Department of Education, 1993, cited in Rosen, 1997). By 1997, projections based on data from one-third of the nation's public schools and school districts indicated that over 3.1 million students nationwide (or 6.84% of the student population) had experienced a suspension during one school year (Brooks et al., 1999). More recently, individual state data suggest that suspensions have increased even further. For example, during the 1998-1999 school year, 7.8% of the entire kindergarten to 12th grade student body in Maryland experienced at least one out-of-school suspension (Brooks et al., 1999).
Relative to student age (or grade level) and minority status, the greatest rates of out-of-school suspension tend to be at the middle school and early high school levels, rising steadily from 7th through 8th grades and peaking in 9th grade (Florida State Department of Education, 1995; Massachusetts State Board of Education, 1991). Unfortunately, minority students continue to be grossly over-represented when rates of suspension are compared (Brooks et al., 1999; The Civil Rights Project, 2000; Dupper & Bosch, 1996; Florida State Department of Education, 1995; Massachusetts State Board of Education, 1991; McFadden, Marsh, Price, & Hwang, 1992). More specifically, in 1997, although Black students made up approximately 17% of all students enrolled in public education, they represented approximately 32% of all students who were suspended, and across the United States, Black students were suspended about 2.3 times more often than White students during the 1996-97 school year. By way of contrast, White students made up approximately 63% of public school enrollment while representing 50% of suspensions nationwide (Brooks et al., 1999).
Although it appears that suspensions of Black students have increased substantially over time, it also is true that minority students are more likely to be disciplined (a) for minor offenses and (b) with disproportionately higher levels of punishment or intensive intervention (Morrison & D'Incau, 1997). In urban school districts, suspensions of minority students are especially troubling. For example, a recent report showed that in Phoenix, Arizona, Black students were suspended or expelled 22 times more often than White students; in Austin, Texas, the rate was 4 times that of White students; in San Francisco, it was 3.7 times that of White students; and in Denver, it was 3.2 times that of White students (Brooks et al., 1999).
Behavioral Infractions Resulting in School Suspensions
Given the intensity of a disciplinary action where a student is barred from attending school and the concomitant loss of instruction and academic engaged time, one might think that only the most egregious behaviors would result in school suspension. However, contrary to popular belief, most out-of-school suspensions across the country are for minor infractions of school rules rather than for dangerous or violent acts (Brooks et al., 1999; National School Board Association, 1994; Skiba et al., 1997). Indeed, one recent study of over 100 secondary administrators (Rosen, 1997) found that the most common reasons for out-of-school suspension were defiance of school authority, not reporting to after school detention or Saturday school, and class disruption.
One also might assume, given the frequency with which suspension is used, that it would result in an almost immediate suppression of inappropriate behavior, improved prosocial behavior, and greater conformity and self-control upon a student's return to school. Contrary to this assumption, the impact of school suspensions has been disappointing at best. Indeed, if a critical goal of suspension is to decrease future school suspensions, then this approach to "behavioral intervention" has failed. As an example, in a one-year investigation of a 9-school south Florida district, McFadden et al. (1992) found that 25% of the students receiving an in-school or out-of-school suspension or corporal punishment committed more than five disciplinary offenses, 75% of the students committed between one and five offenses, and less than 1% of the students committed only one offense. Perhaps more troubling, however, are the results of a study by Costenbader & Markson (1998), which showed that 32% of 252 middle and high school st udents (from a total sample of 620 students) who had been suspended from either an inner city or a separate rural school district reported that that suspensions were "not at all" helpful and thought that they would "probably be suspended again."
Based on the data from a number of diverse studies, it appears that school suspension is being used with increasing frequency, in a disproportionate manner relative to minorities, and for infractions that should be handled with less intensive disciplinary strategies. Moreover, these studies indicate that school suspension often is not successful in decreasing students' chronic and inappropriate behavior, and it is related to a variety of negative academic and educational outcomes for students. Despite these clear and consistent findings, there are some important gaps in the literature base. These gaps include (a) limited information on suspension rates by gender, race, and school level combined (e.g., Are Black girls in elementary school over-represented in suspensions compared to their White peers? What percentage of Black middle school boys experience a suspension compared to Black elementary school boys?), and (b) limited information on the types of infractions that result in suspension by gender and race combined (e.g., For what types of infractions are Hispanic boys most often suspended? How does that compare to the types of infractions for which White boys are suspended?)
The present study aimed to address these gaps by looking at the pattern of school suspensions in a large county school district serving just under 146,000 students in west central Florida. In doing this, the study improved upon previous research in four ways: (a) it included elementary, middle, and high schools in the sample so that increases or decreases in school suspensions could be tracked from level to level, (b) it investigated suspension rates by gender, race, and school level combined, (c) it identified and analyzed the specific types of infractions that resulted most frequently in student suspensions, and (d) it identified those students who accounted for the most suspensions in each infraction category by gender and race.
Description of the School District
The district involved in this investigation is one of 67 in Florida, a state that organizes its school districts in county systems. Demographically, the county served by the district had a total population of 834,054 individuals (1990 Census), with approximately 72% of the population identifying itself as White, 13% as Black, and 10% of Hispanic origin (regardless of race). Thirty-three percent (33%) of Black and 10% of White individuals were reported to live below the poverty line, and 76% of adults held at least a high school diploma. In addition, almost 25% of the population (1994 data) was between the ages of 0 and 17, and 19% of all children and youth below the age of 18 lived in poverty. Most children (66%) were reported to live with a married couple, although a sizable minority lived in single parent homes (23%), with other relatives (9%), or outside the family (3%).
The district is the 12th largest public school district in the nation and the 2nd largest in Florida. During the 1996-97 school year, it served just under 146,000 children in 97 elementary schools, 30 middle schools, 15 high schools, and several special centers focused on early childhood education, exceptional student education, alternative education, or adult education. Only the 142 general education schools (which served 138,761 students--or approximately 95% of all students in the district) were included in the analysis.
Schools in the district varied widely from inner-city schools, where over 95% of children received free or reduced price lunch, to schools in suburban areas, where the socioeconomic status of families was considerably higher. The district also included schools in outlying rural areas, where a number of students were from migrant families. At the time data were collected, 56% of students in the 142 general education schools identified themselves as White, 23% as Black, 18% as Hispanic, and 3% as coming from other minority groups (including Asian, Indian, and Multiracial students). (It is noted that White, Black, Hispanic, Asian, Indian, and Multiracial were the specific descriptors used by the district to identify the various racial groups in the student population.) Of the Black students, the great majority were African-American, although students from places such as Africa, Haiti, and the Caribbean islands whose skin tone was black were included in this group as well. Parents chose the racial group they bel ieved best described their child when they registered their child for school.
Forty-nine percent (49%) of students in the district received free or reduced price lunch, with many more receiving a free rather than a reduced price lunch. A considerably smaller percentage of White students (29%) received free or reduced price lunch compared to Black students (78%) and Hispanic students (72%). As would be expected, the percentage of males (51%) and females (49%) in the student population was approximately equal. Eighty-two percent (82%) of the district's teachers identified themselves as White, 12% as Black, and 6% as Hispanic. The average years of teaching experience was 12.48 years at the elementary level, 11.84 years at the middle school level, and 14.47 years at the high school level.
To gather the data for this investigation, the first author worked with individuals from the district's Management Information Systems (MIS) staff to obtain information routinely collected on out-of-school suspensions by the district from its main database. Data for one complete school year (August 1996--May 1997) for all 142 general education schools in the district was included in the analysis. The raw data obtained on out-of-school suspensions listed each school separately and contained the following information: (a) the unduplicated count of out-of-school suspensions by gender and race, (b) the duplicated count of out-of-school suspensions by gender and race, and (c) the total number of each type of incident (e.g., disruptive behavior) that resulted in out-of-school suspension by gender and race. The unduplicated count represents all students who experienced at least one suspension. Students who experienced multiple suspensions are counted only once in the unduplicated count. The duplicated count represen ts all suspensions. Students who experienced multiple suspensions are counted multiple times in the duplicated count.
Five major questions were addressed in the data analysis:
(a) What percentage of students of each race (i.e., White, Black, Hispanic), gender (i.e., male, female), and school level (i.e., elementary, middle, high) experienced at least one suspension during the 1996-97 school year (i.e., the unduplicated count)?
(b) How many suspensions were there per 100 students of each race, gender, and school level during the 1996-97 school year (i.e., the duplicated count adjusted for differences in group size)?
(c) What types of incidents (of the 33 identified by the school district) resulted in the greatest percentage of total suspensions across school levels?
(d) Among the infractions resulting in the greatest number of suspensions, how were suspensions divided across each combined race/gender group?
(e) What types of incidents (organized into 6 broad categories, see Table V) resulted in the greatest number of suspensions at each school level?
Question (a) was addressed by dividing the number of students of each race, gender, and school level who had experienced a suspension during the 1996-97 school year (e.g., the total number of Black females in elementary school who had experienced a suspension) by the total number of students of that same race/gender/school level according to October 1996 school membership figures (e.g., the total number of Black females in elementary school).
Similarly, question (b) was addressed by dividing the total number of suspensions for students in each race/gender/school level group during the 1996-97 school year by the total number of students in that group according to schools' October 1996 memberships. That number was multiplied by 100 to yield the number of suspensions that occurred per 100 students in that group. This number permitted a standardized comparison of suspension rates across each of the race/gender/school level groups since these groups were not equal in size (i.e., there were far fewer Black and Hispanic students than White students in the district). Figures greater than 100 indicate that there were more suspensions within a particular group than members of that group. This was possible only if there were multiple suspensions of one or more students.
To answer question (c), the number of suspensions for each type of infraction used by the district was summed. Each of these sums was then divided by the total number of suspensions during the 1996-97 school year to yield the percentage of suspensions accounted for by each type of infraction.
Question (d) was addressed by first calculating the number of suspensions for each type of infraction for White, Black and Hispanic students combined. (Suspensions of Asian, Indian, and Multiracial students, who made up only about 1% of suspensions combined, were excluded from this analysis.) The number of suspensions for each combined race/gender group was then divided by this total to yield the percentage of suspensions for each type of infraction accounted for by each race/gender group. These were then compared to the percentage of students of each race/gender group in the population of White, Black, and Hispanic students so that the over- or under-representation of each group in the list of reasons for a suspension could be determined.
Finally, for question (e), the 33 types of infractions resulting in suspension listed by the district (ranging from relatively minor offenses such as dress code and parking violation to major infractions such as weapons possession, assault, and arson) were grouped into six broad categories by the first author. (These categories and the behavioral infractions included in each category are shown in Table V.) Subsequently, the total number of infractions resulting in suspension in each category was calculated for each school level (i.e., elementary, middle, high). Each of these figures was divided by the total number of students at each school level and then mul tiplied by 100 to yield the number of infractions per 100 students at each school level resulting in a school suspension. This was necessary to accommodate for and standardize these data across the different numbers of students at the elementary, middle, and high school levels.
Rates of Suspension Across Gender, Race, and School Level
Table I shows the percentage of students experiencing at least one suspension by race, gender, and school level. The numbers in this table are based on an unduplicated count of student suspensions (i.e., students who were suspended multiple times are counted only once). Three important findings are revealed in this table. First, across race and school level, many more males than females experienced at least one suspension. Among White students, for example, 11.95% of males vs. 4.53% of females experienced at least one suspension. Among White and Hispanic students, males were more than twice as likely as females to experience a suspension. Among Black students, males were slightly less than twice as likely as females to experience a suspension.
A second finding that is evident in Table I is that Black students were much more likely to experience a suspension than were White or Hispanic students. This was true across gender and school levels. For example, 26.28% of Black males and 13.64% of Black females experienced at least one suspension compared to 11.95% of White males and 4.53% of White females. Suspension rates among Hispanic students (15.42% of males and 6.48% of females) were more similar to the rates of White students than Black students. It is notable that at the middle school level, almost one-half of all Black males and almost one-third of all Black females experienced at least one suspension. Additionally, it can be seen that the disproportionate percentage of Black students being suspended was evident as early as elementary school, with Black males more than three times as likely as White or Hispanic males to experience a suspension and Black females more than eight times as likely as White or Hispanic females to experience a suspension in elementary school. It is noted as well that a greater percentage of Black females (3.88%) than White (3.08%) or Hispanic males (3.36%) experienced a suspension in elementary school, and a greater percentage of Black females (31.88%) than White males (25%) experienced a suspension in middle school. Similarly, a greater percentage of Black females in high school (20.69%) experienced a suspension than White males (18.90%) in high school. These were the only instances where suspensions of females were higher than those of males within a particular school level.
A third important finding that can be seen in Table I is that rates of suspension increased considerably from elementary to middle school and then dropped off in high school. This pattern was true across race and gender. At the elementary level, 3.36% of all students experienced at least one suspension. That percentage increased to 24.41% at the middle school level and then declined somewhat to 18.46% at the high school level.
Table II shows the number of suspensions per 100 students by race and gender. Information in this table is based on a duplicated count of suspensions (i.e., students who were suspended multiple times are counted multiple times) and focuses on the absolute number of suspension incidents. Again, these data indicate that when differences in the population sizes of each group are controlled, males experienced a higher rate of suspension than females, Black students experienced a higher rate of suspension than White or Hispanic students, and total suspensions increased greatly from elementary school to middle school and then dropped off in high school. Additionally, when contrasting the data in Tables I and II, it is apparent that many suspended students likely experienced multiple suspensions. For example, although there were 52.41 total suspensions per 100 White middle school males, only 25% of White middle school males in the sample experienced a suspension. Similarly, although there were 127.06 total suspensio ns per 100 Black middle school males, only 48.90% of Black middle school males experienced a suspension. The differences in the numbers in the duplicated and unduplicated counts are accounted for by multiple suspensions of the same students. Overall, Black middle school males had by far the highest number of suspensions per 100 students. White and Hispanic elementary school females had the lowest number of suspensions per 100 students.
Types of Infractions Resulting in Suspension
Table III shows the 15 most common infractions that resulted in suspension for all of the students in the district studied. Of the 33 different types of offenses identified by the school district, these 15 infractions made up approximately 90% of suspensions. By far the greatest number of suspensions (20%) was for disobedience/insubordination. Disruptiveness (13% of all suspensions) and fighting (13% of all suspensions) were the next most frequent reasons for suspension. It is notable that possession of a weapon made up less than 1% of the infractions resulting in suspension.
Table IV shows how the suspensions for the infractions in Table III were distributed across combined race/gender groups. The percentage of students in each race/gender group for the district as a whole is noted at the top of the table so that comparisons with the incidence figures can be made. What is clearly evident in Table IV is that Black males experienced a much greater percentage of suspensions across all but three of the infraction categories (i.e., tobacco, narcotics, and alcohol possession) than would be expected given their demographic representation in the population. For example, while Black males made up only 12% of the student population, these students made up one-third or more of the total suspensions for disruptive behavior, fighting, inappropriate behavior, battery, threat/intimidation, left class or campus without permission, and sexual harassment. The only infraction where Black males were underrepresented was possession of tobacco. In addition, their representation in suspensions for alco hol and narcotics was much lower than for other types of infractions. In contrast, White males were over-represented in suspensions having to do with substance possession (i.e., tobacco, narcotics, alcohol) and weapons possession. In general, a much greater percentage of White students than Black or Hispanic students experienced a suspension for tobacco possession. Black females, like Black males, were over-represented in the top 15 infractions resulting in suspension given their percentage in the population, but the over-representation was not nearly as dramatic as it was for Black males.
Across the listed infractions, the percentages of suspensions for White males were consistent with those reflecting their representation in the population (with the exception of suspensions for tobacco possession, weapons possession, narcotics possession, and alcohol possession). White and Hispanic females tended to be under-represented across the different suspension reasons, while Hispanic males were slightly over-represented, except in the area of sexual harassment, where they were suspended twice as often as would be expected given their representation in the population. Relative to weapons possession and sexual harassment, males made up 83% and 95% of suspensions, respectively. Both Black males and White males were greatly over-represented in suspensions for weapons possession. Black males also accounted for four times as many suspensions for sexual harassment as would be expected given their representation in the population (i.e., 12% of the population vs. 48% of suspensions for sexual harassment). Over all, with the exception of the few areas noted, the combined race/gender groups tended to make up similar proportions of each of the suspension reasons.
Table V shows the mean number of incidents of each type (divided into six categories) resulting in suspension per 100 students by school level. The behavioral category resulting in the greatest number of suspensions at every school level was Disobedience. Suspensions for Disobedience increased dramatically from the elementary school level (3.62 per 100 students) to the middle school level (35.75 incidents per 100 students) and dropped off slightly at the high school level (26.74 incidents per 100 students). Violence Against Persons was a distant second reason for suspension at the elementary and middle school levels, while Substance Abuse was a distant second at the high school level. Suspensions for Violence Against Persons increased almost 700% from elementary school (M=1.31 incidents per 100 students) to middle school (M=9.78 incidents per 100 students) and then dropped off significantly in high school (M=3.65 incidents per 100 students). Violence Against Property was the category with the third highest nu mber of suspensions at the elementary and middle school levels, while Violence Against Persons ranked third at the high school level. Violence Against Property, Substance Possession, and Absent Without Permission were the only categories showing an increase from middle school to high school. Weapons Possession, which accounted for a very small proportion of the total number of suspensions, occurred more often at the middle school level than at the high school level (0.28 incidents per 100 students at the middle school level vs. 0.15 incidents per 100 students at the high school level).
The results of this study have a number of implications for educators seeking to improve discipline, behavior management, and safety in their schools while reducing the number of out-of-school suspensions. Four areas in particular stand out as being especially significant: (a) demographics, (b) multiple suspensions, (c) incidence at the middle school level, and (d) reasons for suspension.
Results of this study are clear in showing that males (as compared to females) and Black students (as compared to White and Hispanic students) are at much greater risk of being suspended. Such findings are not new, having been reported numerous times before in the literature (Costenbader & Markson, 1998; Dupper & Bosch, 1996). However, this study, by examining race and gender combined, has added some additional information to what is already known. First, when comparing students of different grades and races, this study found that Black girls are at much greater risk than White or Hispanic girls to experience a suspension. Clearly, in terms of overall numbers, Black males were suspended much more frequently than any other combined gender and race group. However, looking at suspensions across all three levels, Black males were twice as likely to experience a suspension as White males, but Black females were more than three times as likely to experience a suspension as White females. Additionally, this study de monstrated that the over-representation of Black students in suspension rates is a phenomenon that begins in elementary school. This is consistent with previous research (Brooks et al., 1999). Critically, while far fewer students are suspended in elementary school than in middle or high school, Black students are much more likely than their White or Hispanic peers to receive a suspension at the elementary level. Thus, the disproportionate suspension of Black students is not grade-level specific (i.e., different at the elementary vs. middle vs. high school levels); rather, it appears more pervasive. And yet, because of the smaller suspension numbers at the elementary school level, this pervasiveness may be hidden. In the present study, however, the over-representation of Black students in suspensions was very apparent by the middle school level as almost one half of all Black males and one third of all Black females had experienced at least one school suspension. These suspension rates are dramatically higher than those of their White and Hispanic middle school peers.
Importantly, although the over-representation of Black students in suspensions in this study is clearly evident, the reasons for it are not. It could be, for example, that Black students commit greater numbers of infractions that typically result in out-of-school suspension than students of other races.
However, it also is plausible that the over-representation of Black students may be due to cultural and social misunderstanding, lack of teacher and administrator training, classroom and/or school climate, or worse, discrimination. In addition, socioeconomic status must be considered as a moderating variable as a greater percentage of Black (78%) than White (29%) children received free or reduced price lunch (an indicator of poverty), and past studies have shown socioeconomic status to be related to suspension rates as well (Nichols, Ludwin, & ladicola, 1999; Skiba, Peterson, & Williams, 1997). However, the fact that far fewer Hispanic children (72% of whom receive free or reduced price lunch) than Black children experienced a suspension provides some evidence that socioeconomic status is not the primary variable underlying the differences in rates of suspension that were evident in this study. If SES alone explained differences in rates of suspension, it would be expected that the percentage of Hispanic stud ents experiencing a suspension would be similar to the percentage of Black students experiencing a suspension because similar percentages of students from these groups lived in impoverished circumstances. Instead, it appears that something other than SES accounts for the over-representation of Black students in suspensions, although precisely what that factor is cannot be ascertained from the data collected.
The most important implication of this finding is that there is a need for administrators to determine if disproportionate suspensions among various racial groups exist in their schools and, if so, (a) determine why they exist and (b) develop and implement an action plan to remediate the problem without deciding that the over-representation is beyond their control. As Black students are at greater risk for suspension than students of other races, educators need to create secondary prevention strategies targeted toward the circumstances around these suspensions. If Black students are suspended more often than other students because they are committing more serious infractions, then this needs to be understood from an ecological perspective. For example, are frequently used classroom materials and teaching strategies meeting the needs of these students? Do these students feel valued and respected in their classrooms? Are the types of interventions being used with these students working to teach and reinforce ap propriate behavior? If, alternately, the over-representation of Black students being suspended is due to a lack of cultural or socio-behavioral understanding of these students by teachers and administrators and/or to racial discrepancies in how discipline is administered, then solutions lie in increased diversity training for school staff and careful examination--and subsequent reform--of the way various behavior problems are handled across race and gender. Too often, educators assume that the inequity and disproportionate treatment and outcomes in education, such as the overrepresentation of some student groups in suspension, are due solely to factors outside of the school's control (e.g., poverty, parent apathy, negative role models in the community). Adopting an ecological perspective that includes examining students' experiences in the school setting can help schools generate solutions to such problems by changing those things that can be addressed within the context of the school.
It is important for school districts to track the number of multiple suspension students in their schools because these students represent persistent problematic behavior that often is more serious than single suspension cases and is not responding to the "strategic" goals of suspension (i.e., to decrease or eliminate the need for such action in the future). At the same time, each suspension--regardless of the involvement of a student who has or has not been suspended before--still reflects an incident where staff are involved, time is expended, and the academic atmosphere of a class or building is disrupted. In the present study, the difference between the percentage of students experiencing at least one suspension in Table I and the number of suspensions per 100 students in Table II suggests that multiple suspensions of individual students were quite common. As noted, in the absence of staff insensitivity or bias, multiple suspensions of individual students provide evidence that, consistent with previous r esearch (Henderson & Friedland, 1996; Imich, 1994; McFadden et al., 1992), suspension alone often does not curtail inappropriate behavior. In fact, Henderson & Friedland (1996) noted that some students purposely misbehave, using the suspension, for example, to escape from aversive situations at school and/or to experience more desirable conditions at home or out of school. Not surprisingly, students who experience repeated suspensions are at greater risk of dropping out of school (Velez, 1989), a fact that may explain the lower high school versus middle school suspension rates in this study. More explicitly, it is hypothesized that the middle school students with the most serious behavior problems had dropped out of school, such that they were not present to be suspended in high school.
Students with multiple suspensions present a serious dilemma for educators. Although such students sometimes make it difficult to maintain a positive learning environment for other students, the school staff still has a responsibility to analyze and meet the behavioral and educational needs of these students as early and effectively as possible. For students with chronic behavior problems, educators must implement intensive tertiary prevention strategies, including a functional analysis of the behavior and appropriate interventions. Among the functional questions that should be asked are: What is the function or purpose of the acting out behavior? Is it linked to unresolved academic or social problems? Is the teacher receiving adequate support in trying to manage the student's behavior? Unfortunately, although suspension can protect the interests of other students and staff, it often is not functionally linked to the core problem that results in the suspension. Thus, schools need to have procedures in place to functionally analyze the reasons for a student's multiple suspensions, and, when a student is suspended repeatedly, to determine if the behavior is a reflection of a specific disability (IDEA, 1999). This frequently means informally involving professionals such as the school psychologist, counselor, or behavior specialist, or formally involving a student support or special education child study team. Moreover, for students with chronic behavior problems, it is recommended that schools work with families and community agencies to develop strategies for school-linked and wrap-around services and problem-solving interventions. This system might include (a) different school-based intervention strategies to be implemented following a first, second, or third suspension, respectively; (b) specific guidelines to be followed by school personnel when meeting with parents (and/or students) following a suspension; and (c) a referral network to connect parents of students with multiple suspensions to support services i n the community.
It also is very important for schools to have alternatives to suspension available. Although there are times when students must be removed from their regular classrooms, this does not mean that they have to be completely removed from a school's broader learning environment. Some schools have in-school suspension programs available. Others have alternative suspension sites where students can be supervised on the day(s) of their suspension. Still others use Saturday School programs for some suspensions. Regardless, it is recommended that in-school suspension and alternative-to-suspension programs: (a) involve a rehabilitative component (as opposed to being strictly punishment-oriented), (b) actively involve parents (e.g., by requiring parent involvement at the alternative site), and (c) be linked to other support services for students and families (e.g., through collaborative partnerships with local social service agencies). In the end, a focus on the goal of suspension must be maintained: to understand why th e inappropriate behavior is occurring, to develop and implement remedial interventions, and to decrease or eliminate the occurrence of future inappropriate behavior and suspensions.
Incidence at the Middle School Level
The results of this study are consistent with previous research indicating that most suspensions occur at the middle school level (Florida State Department of Education, 1995). In fact, this investigation showed that suspensions rose dramatically from the elementary to the middle school level and then dropped off at the high school level. In this study, the number of suspensions per 100 students increased more than 10-fold from elementary to middle school. In addition, there were more than 7 times as many suspensions, per 100 students, for Violence Against Persons (e.g., fighting) at the middle school level than at the elementary school level.
These findings must be considered within the developmental context of early adolescence, a time of myriad physical, cognitive, social, and emotional changes for students. Some of the typical challenges of this period include coping with peer pressure, making decisions relative to experimenting with various behaviors (e.g., smoking, sexual activity), and beginning to form an identity separate from one's parents. As young adolescents are faced with these multiple changes in their lives, they also make the shift from elementary to middle school, where they encounter school environments that are larger, less personal, and require greater self -control and self-direction. Some researchers (Eccles, Midgley, Wigfield, & Buchanan, 1993) have noted concern about the poor fit between (a) early adolescent students' needs for guidance and support from teachers as they begin to explore who they are as individuals and (b) the ecology of the middle school environment.
Others (Knoff, 2000a) have emphasized that early adolescents need to be taught the prosocial skills to respond to these developmental and ecological changes and that schools need to have meaningful and consistently applied incentives and consequences for student behavior. The tremendous increase in suspensions noted at the middle school level in this study (particularly for Violence Against Persons) suggests a greater need for primary prevention strategies to prevent behavioral problems at this developmental stage. Strategies that give careful consideration to the developmental context of early adolescence are those that are likely to have the highest probability of success. Such strategies include social skills training designed to help students make appropriate decisions in the face of peer pressure, training for teachers in developing effective relationships with students of various cultural backgrounds based on mutual respect and trust, and carefully supervised peer mediation programs that allow students to assume a greater role in helping to negotiate solutions to problems. Many of these strategies could be implemented during the elementary school years, and all of these strategies should be implemented across an entire grade-level, if not school-wide. It is important as well to find alternatives to suspension relative to consequences (as opposed to punishments) for inappropriate behavior. These might include removal of privileges (e.g., not being able to attend a special event at school because of inappropriate behavior in the classroom prior to the special event) or restorative measures (e.g., giving something back to the school community to replace something that was defaced or destroyed).
Reasons for Suspension
Beyond a focus on why individual students are suspended, schools need to look at the collective reasons across students for suspensions in order to identify common patterns of inappropriate behavior and more preventive strategies that can impact large numbers of students. In this context, this study has clearly shown that the majority of suspensions occurring in this district fell under the broad category of Disobedience. Indeed, one-fifth of all suspensions (i.e., almost 6,500) were for infractions labeled by administrators as disobedience/insubordination. While this parallels findings reported in other studies (National School Board Association, 1994), it is notable that a large proportion of these infractions might have been prevented through improved behavior management strategies in the classroom and school-wide behavioral supports. These infractions are far less serious and more amendable to intervention than more serious infractions like weapons and narcotics possession (each of which made up a very sm all number of suspensions in this study). If schools were to focus greater effort on the prevention of classroom misbehavior (e.g., through improved teacher training in effective behavior management techniques), it is likely that suspension rates could be greatly reduced in this school district and others like it. Of course, this would require a tremendous shift in attention and resources away from the punishment-oriented model of intervention toward a prosocial, positive behavioral support model.
It also is true, however, that analyses of suspensions across students in a building do reveal clusters of infractions committed by clearly identified groups of students. Often, these infractions impact or are mediated by peer groups, and they require a more intense, systematic response. Some of these situations arise with some frequency in particular places in a school building (e.g., fighting in the halls). In such circumstances, Knoff (2000b) has suggested that administrators invoke a "special situations analysis" to help link the assessment of peer or setting-oriented problems to solutions that have the highest probability of success. He noted that it often is necessary in these cases to analyze the situation both from an ecological and from a functional perspective. Accordingly, he suggests an assessment of (a) student characteristics, issues, and factors (e.g., Who is fighting? What issues are giving rise to the fights?); (b) teacher/staff characteristics, issues, and factors (e.g., How are teachers res ponding when they see students fighting?); (c) environmental characteristics, issues, and factors, including physical plant and logistical considerations (e.g., Where are most of the fights taking place?); (d) incentives and consequences (e.g., What incentives are in place to promote harmony among students? What consequences are used to deal with fighting on campus?); and (e) resources (e.g., Who or what might be available in the school or in the community to assist in helping students to co-exist more peacefully?). Knoff recommends careful examination of how each of these factors contributes to (and can help to solve) the particular problem being addressed in order to develop a focus on the most relevant issues and generate the most appropriate solution(s).
Limitations and Future Directions
A major limitation of the current study is that it involved only one school district. Thus, the findings may not generalize to districts with different characteristics and types of student populations. The particular district that was chosen, however, did serve a large, racially diverse group of students that approximated the demographic proportion of various racial groups in the United States population. Moreover, the large number of schools included in the analysis (N=142) allowed for inclusion of numerous schools with different proportions of students from different racial and socioeconomic backgrounds.
The current study also has identified several areas where more research is needed. First, the findings of this study indicate that Black students are at considerably greater risk for suspension than their peers of other races. More research is needed to better understand this phenomenon. For example, do schools with a majority of Black teachers and administrators have lower rates of suspension among Black students? How does active school involvement among Black parents impact their children's suspension rates? Are suspension rates lower for Black students attending schools that have adopted Afrocentric approaches? How is busing to achieve racial integration related to rates of suspension? Answers to these questions could help to ascertain the degree to which cultural misunderstandings between teachers and students might be contributing to higher rates of suspension among Black students.
This study also has revealed a tremendous increase in suspensions from the elementary to the middle school level. More research is needed to better understand how to prevent such an increase. Investigations that examine the types of middle school environments that are successful in promoting appropriate behavior in young adolescents would be very helpful in this regard.
This study has shown that race, gender, and school level all have strong relationships to suspension rates. In the district investigated, students at greatest risk for suspension were male, Black, and in middle school. Clearly, students who matched all three of these descriptors were at greatest risk for suspension, with almost half of all Black males at the middle school level experiencing one or more suspensions. However, students matching only one of these descriptors (e.g., Black females in elementary school, White females in middle school) still had much higher rates of suspension than other students who matched none of the descriptors. Results also showed that most suspensions were for relatively minor infractions, with the great majority of them for disobedience. Looking at the top 15 reasons why students were suspended, it is clear that Black males were greatly over-represented in most of the reasons for suspension, with no one type of infraction standing out amongst the others for these students.
Overall, educators who wish to reduce suspension rates at their schools are encouraged to adopt an ecological perspective to understanding why certain gender and racial groups are over-represented in suspension rates. Additionally, for students with multiple suspensions, it is recommended that educators utilize functional assessment strategies to determine the reasons behind students' misbehavior. Finally, given the great number of suspensions that appear to result from relatively minor misbehavior in the classroom (as opposed to some of the more serious types of infractions), it is suggested that preventive school-wide strategies that focus on establishing and maintaining effective classroom discipline are needed. These approaches focus on more positive, prosocial behavior, and they could make a significant difference in decreasing the overall number of suspensions that occur in schools each year.
Table I Percentage of Students Experiencing at Least One Suspension Race White Black Hispanic Gender Male Female Male Female Male Female n=41,564 n=38,353 n=16,390 n=16,045 n=13,057 n=12,154 Level Elementary n=73,841 3.08% 0.40% 12.15% 3.88% 3.36% 0.38% Middle n=29,613 25.00% 9.28% 48.90% 31.88% 33.95% 15.58% High n=34,109 18.90% 8.84% 39.46% 20.69% 27.36% 12.80% M N=137,563 11.95% 4.53% 26.28% 13.64% 15.42% 6.48% Race Gender M N=137,563 Level Elementary n=73,841 3.36% Middle n=29,613 24.41% High n=34,109 18.46% M N=137,563 11.63% Table II Number of Suspensions Per 100 Students Race White Black Hispanic Gender Male Female Male Female Male Female Level n=41,564 n=38,353 n=16,390 n=16,045 n=13,057 n=12,154 Elementary n=73,841 4.84 0.58 22.29 6.31 5.10 0.53 Middle n=29,613 52.41 16.47 127.06 69.13 72.83 27.20 High n=34,109 35.00 14.29 72.60 38.82 56.50 22.52 M N=137,563 22.97 7.61 56.66 27.01 31.44 11.28 Race Gender M Level N=137,563 Elementary n=73,841 4.38 Middle n=29,613 53.34 High n=34,109 33.96 M N=137,563 22.94 Table III 15 Most Common Infractions Resulting in Suspension for All Students Across All District Grade Levels Infraction % of Total Suspensions Disobedience/Insubordination 20% Disruptive 13% Fighting 13% Inappropriate Behavior 11% Noncompliance with Assigned Discipline 7% Profanity 7% Disrespect 6% Tobacco Possession 4% Battery 3% Threat/Intimidation 2% Left Class or Campus without Permission 2% Weapons Possession .7% Narcotics Possession .6% Sexual Harassment .6% Alcohol Possession .3% Table IV Distribution of Suspension Reasons by Race and Gender Gender Male Female Race White Black Hispanic White Percent of Population (a) (30%) (12%) (9%) (28%) Infraction (Total Number of Infractions in Category) (b) Disobedience/Insubordination (n=6489) 27% 28% 13% 11% Disruptive (n=4377) 29% 38% 11% 6% Fighting (n=4187) 28% 34% 13% 7% Inappropriate Behavior (n=3546) 30% 35% 13% 7% Noncompliance with Assigned Discipline (n=2408) 29% 24% 15% 12% Profanity (n=2237) 31% 25% 12% 12% Disrespect (n=1975) 29% 32% 10% 8% Tobacco Possession (n=1168) 64% 5% 7% 22% Battery (n=887) 27% 43% 12% 5% Threat/Intimidation (n=537) 23% 43% 11% 7% Left Class or Campus without Permission (n=509) 28% 33% 12% 11% Weapons Possession (n=225) 48% 24% 11% 5% Narcotics Possession (n=205) 55% 15% 13% 11% Sexual Harassment (n=181) 25% 48% 22% 2% Alcohol Possession (n=91) 48% 14% 6% 13% Gender Female Race Black Hispanic Percent of Population (a) (12%) (9%) Infraction (Total Number of Infractions in Category) (b) Disobedience/Insubordination (n=6489) 16% 5% Disruptive (n=4377) 15% 2% Fighting (n=4187) 15% 5% Inappropriate Behavior (n=3546) 12% 35% Noncompliance with Assigned Discipline (n=2408) 14% 7% Profanity (n=2237) 16% 4% Disrespect (n=1975) 18% 3% Tobacco Possession (n=1168) 1 % 2[degrees]% Battery (n=887) 12% 1% Threat/Intimidation (n=537) 12% 3% Left Class or Campus without Permission (n=509) 9% 7% Weapons Possession (n=225) 10% 2% Narcotics Possession (n=205) 15% 4% Sexual Harassment (n=181) 3% <1% Alcohol Possession (n=91) 11% 8% (a) Population refers to all White, Black, and Hispanic students in the district. (b) Includes only suspensions of White, Black, and Hispanic students. Excludes India, Asian, and Multiracial students. Table V Number of Suspensions Per 100 Students Organized by Incident Categories Elementary Middle High Schools Schools Schools Incident Category (n=97) (n=30) (n=15) Violence Against Persons 1.31 9.78 3.65 Violence Against Property 0.13 1.25 1.62 Substance Possession 0.01 0.85 4.28 Disobedience 3.62 35.75 26.74 Absent Without Permission 0.09 0.61 0.97 Weapons Possession 0.11 0.28 0.15 Note: Infractions included in each category were as follows: Violence Aganist Persons: Battery, sexual assault, sexual harassment, sexual offense, threat/intimidation. Violence Against Property: Arson, breaking and entering, falsification of records, larceny/theft, motor vehicle theft, parking violation, petty theft, robbery, trespassing, vandalism. Substance Possession: Alcohol, narcotics (excluding alcohol), non-controlled substance, tobacco. Disobedience: Disobedience/insubordination, disorderly conduct, disrespectful, disruptive, dress code violation, inappropriate behavior, noncompliance with assigned discipline, profanity. Absent Without Permission: Left class without permission, left campus without permission. Weapons Possession: Possession of any type of weapon on campus (eg., gun, knife etc.)
Brooks, K., Schiraldi, V., & Ziedenberg, J. (1999). School house hype: Two years later. San Francisco, CA: Center on juvenile and Criminal Justice. [On-line]. Available: www.cjcj.org
The Civil Rights Project. (2000). Opportunities suspended: The devastating consequences of zero tolerance and school discipline policies [On-line]. Available: www.law.harvard.edu/civilrights
Costenbader, V. K., & Markson, S. (1998). School suspension: A study with secondary school students. Journal of School Psychology, 36, 59-82.
Dupper, D. R., & Bosch, L. A. (1996). Reasons for school suspensions. Journal for a Just and Caring Education, 2, 140-150.
Eccies, J.S., Midgley, C., Wigfield, A., & Buchanan, C. (1993). Development during adolescence: The impact of stage-environment fit on young adolescents' experiences in schools and families. American Psychologist, 48(2), 90-101.
Florida State Department of Education. (1995). Florida school discipline study: 1994 juvenile justice reform act. A report to the Florida legislature. Tallahassee, FL: Author. (ERIC Document Reproduction Service No. ED 384 981)
Henderson, J., & Friedland, B. (March, 1996). Suspension, a wake-up call: Rural educators' attitudes toward suspension (Report No. RC 020 545). Rural Goals 2000: Building Programs that Work. (ERIC Document Reproduction Service No. ED 394 749) Individuals with Disabilities Education Act Regulations, 34 C.F.R. [section] 300 and 303. (1999).
Imich, A. J. (1994). Exclusions from school: Current trends and issues. Educational Research, 36, 3-11.
Knoff, H. M. (2000a). The stop & think social skills program for children and adolescents: Teaching effective interpersonal and self-management skills. Longmont, CO: Sopris West.
Knoff, H. M. (2000b). Organizational development and strategic planning for the millennium: A blueprint toward effective school discipline, school safety, and crisis prevention. Psychology in the Schools, 37, 17-32.
Lewis, T. J., Sugai, G., & Colvin, G. (1998). Reducing problem behavior through a schoolwide system of effective behavioral support: Investigation of a school-wide social skills training program and contextual interventions. School Psychology Review, 27, 446-459.
Massachusetts Board of Education. (1991). Suspensions in the Massachusetts public schools. Quincy, MA: Office of Planning, Research, and Evaluation. (ERIC Document Reproduction Service No. ED 334 333)
McFadden, A. C., Marsh, G. E., Price, B. J., & Hwang, Y. (1992). A study of race and gender bias in the punishment of school children. Education and Treatment of Children, 15, 140-146.
Mellard, D., & Seybert, L. (1996). Voices about school suspension, expulsion, and safety. Lawrence, KS: Center for Research on Learning. (ERIC Document Reproduction Service No. ED 403 639)
Morrison, G. M., & D'Incau, B. (1997). The web of zero tolerance: Characteristics of students who are recommended for expulsion from school. Education and Treatment of Children, 20, 316-335.
National Schools Board Association. (1994). Towards better and safer schools. American School Board Journal, 152.
Nicholas, J. D., Ludwin, W. G., & Iadicola, P. (1999). A darker shade of gray: A year-end analysis of discipline and suspension data. Equity & Excellence in Education, 32, 43-55.
Rosen, L. (1997). School discipline: Best practices for administrators. Thousand Oaks, CA: Corwin Press.
Skiba, R. J., Peterson, R. L., & Williams, T. (1997). Office referrals and suspension: Disciplinary intervention in middle schools. Education and Treatment of Children, 20, 295-315.
Sugai, G., & Horner, R. H. (1999). Discipline and behavioral support: Practices, pitfalls, & promises. Effective School Practices, 17 (4), 10-22.
Velez, W. (1989). High school attrition among Hispanic and non-Hispanic White youths. Sociology of Education, 62, 119-133.
Address all correspondence concerning this article to: Linda M. Raffaele Mendez, Ph.D., School Psychology Program, University of South Florida, 4202 E. Fowler Avenue, EDU 162, Tampa, FL 33620. E-mail: Raffaele@tempest.coedu.usf.edu
|Printer friendly Cite/link Email Feedback|
|Author:||Mendez, Linda M. Raffaele; Knoff, Howard M.|
|Publication:||Education & Treatment of Children|
|Date:||Feb 1, 2003|
|Previous Article:||Augmenting stimulus intensity with an awareness enhancement device in the treatment of finger sucking.|
|Next Article:||A developmental ecological perspective in systems of care for children with emotional disturbances and their families.|