Social-ethnic school composition and school misconduct. Does sense of futility clarify the picture? more

by Jannick Demanet

Published in 'Sociological Spectrum'

Sociology of Education, Agnews General Strain Theory, and Sociology Of Deviance

This article was downloaded by: [Demanet, Jannick] On: 9 February 2011 Access details: Access Details: [subscription number 933060683] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 3741 Mortimer Street, London W1T 3JH, UK Sociological Spectrum Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713666965 SOCIAL-ETHNIC SCHOOL COMPOSITION AND SCHOOL MISCONDUCT: DOES SENSE OF FUTILITY CLARIFY THE PICTURE? Jannick Demaneta; Mieke Van Houttea a Research Group CuDOS, Department of Sociology, Ghent University, Gent, Belgium Online publication date: 02 February 2011 To cite this Article Demanet, Jannick and Van Houtte, Mieke(2011) 'SOCIAL-ETHNIC SCHOOL COMPOSITION AND SCHOOL MISCONDUCT: DOES SENSE OF FUTILITY CLARIFY THE PICTURE?', Sociological Spectrum, 31: 2, 224 — 256 To link to this Article: DOI: 10.1080/02732173.2011.541343 URL: http://dx.doi.org/10.1080/02732173.2011.541343 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Sociological Spectrum, 31: 224–256, 2011 Copyright # Taylor & Francis Group, LLC ISSN: 0273-2173 print=1521-0707 online DOI: 10.1080/02732173.2011.541343 SOCIAL-ETHNIC SCHOOL COMPOSITION AND SCHOOL MISCONDUCT: DOES SENSE OF FUTILITY CLARIFY THE PICTURE? Jannick Demanet Mieke Van Houtte Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Research Group CuDOS, Department of Sociology, Ghent University, Gent, Belgium This article assesses whether social-ethnic composition influences students’ school misconduct. Based on general strain theory, we expect to find that individual sense of futility acts as an intermediate mechanism. Starting from cultural deprivation and oppositional culture theory, however, we hypothesize that a shared school culture of futility acts as mediator. Multilevel analyses of data from the Flemish Educational Assessment, consisting of 11,872 students in 85 schools, showed that ethnic composition has an impact on school misconduct: there is more deviance in ethnically mixed schools than in ethnic concentration schools due to a greater student sense of futility. The presence of a culture of futility did not affect the probability of being deviant. These results were true only for natives: school factors did not affect migrants’ deviancy. We conclude that, although a dispersal of ethnically diverse students across all schools is favorable, this can yield negative side effects. Schools are major socializing agents. In addition to transferring knowledge, they have to teach appropriate social behavior. Therefore, it is important for a school to counter both school-specific deviant behavior and more general deviancy. We define school-specific deviant behavior here as any kind of behavior that counters the school rules. In the past, students performing this kind of behavior have reported more depression (Slee 1995), lower academic achievement (Bryant et al. 2000), and a higher risk of engaging in risky Address correspondence to Jannick Demanet, Research Group CuDOS, Department of Sociology, Ghent University, Korte Meer 3-5, 9000 Gent, Belgium. E-mail: Jannick.Demanet@ ugent.be Social-Ethnic School Composition and School Misconduct 225 behaviors, such as cigarette use (Bryant et al. 2000) and dropping out of school (Jenkins 1995). More generally, having order within a school is important for the development of a positive school climate (S. Davies 1999; Teddlie and Reynolds 2000, p. 148). The opposite is also true: a conflict-ridden school climate diverts attention from an academic orientation and facilitates deviant behavior in other students attending the school (Kasen, Johnson, and Cohen 1990; Kuperminc et al. 1997). To counteract these negative consequences, we need a thorough understanding of the determinants of school deviance. School deviance has been widely studied and researchers have provided numerous explanations for its occurrence (e.g., Cohen 1955; L. Davies 1984; Hirschi 1969). Research based on social control theory (Hirschi) has indicated that students’ involvement at school acts as an impediment to deviance (e.g., Stewart 2003; Welsh 2000), while research based on general strain theory (Agnew 1985, 1992) holds that students experiencing blockage are more prone to misbehaving. However, other researchers have stated that institutional causes exist as well (Reynolds and Jones 1978), making it likely that deviancy rates will vary between schools. Studies focusing on this have shown that schools indeed influence the likelihood that their students are deviant (e.g., Gottfredson et al. 2005; Welsh 2000; Welsh et al. 2000). However, in this kind of research, one dimension of the school climate (C. Anderson 1982; Welsh et al. 2000) remains underemphasized—the social milieu, consisting mainly of compositional school features, which are the aggregated characteristics of the student body, such as mean socioeconomic status (SES) and ethnic and gender composition. Although some recent studies have investigated the impact of ethnic concentration on school violence (Eitle and Eitle 2003; Stretesky and Hogan 2005), few studies have concentrated on these compositional features. The aim of this article is to fill this significant gap in scientific literature. We will specifically focus on the link between school compositional features and school deviance. While studies on ethnic student composition’s relationship to school deviance looked at harsh displays of deviancy, such as homicide, rape, drug abuse, or weapon possession (Eitle and Eitle 2003; Stretesky and Hogan 2005), we will extend this line of research by pinpointing school misconduct, a minor form of deviancy, consisting of rule-breaking behavior such as smoking at school, skipping school, and arriving late at school (e.g., Van Houtte and Stevens 2008). In terms of theory, we start from general strain theory (Agnew 1985, 1992) and (sub)cultural theories (Cohen 1955; Ogbu 1978) to hypothesize that either the individual feelings of futility or the shared feelings—that is, a culture—of Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 226 J. Demanet and M. Van Houtte futility mediate the relationship between school social-ethnic composition and school misconduct. This leads us to address two questions: whether the social-ethnic composition of schools influences the individual level of self-reported school misconduct in the context of Flemish education and whether futility, either individual or shared, is at least in part responsible for the observed influence(s). THEORETICAL PERSPECTIVES Since the 1970s, research has considered whether schools may influence the deviancy of their students (Power et al. 1967; Power et al. 1972). School features that have been shown to relate to school misconduct include school size (Gottfredson and Gottfredson 1985; Stewart 2003), student involvement in decision-making (Gottfredson and Gottfredson 1985), the disciplinary climate (Gottfredson et al. 2005; Reynolds and Jones 1978), and the psychosocial climate (Gottfredson et al. 2005), which includes factors such as respect for students, fairness of rules (Welsh 2000), and feelings of unsafety (Palardy 2008). However, research has largely neglected one important aspect of a school’s climate: the social milieu of schools. This refers to the average background characteristics of the actors at school, such as teachers and students (C. Anderson 1982; Welsh et al. 2000). This is normally assessed using compositional variables (see e.g., Opdenakker and Van Damme 2001; Thrupp et al. 2002). The research strand investigating school influences on students’ deviancy mainly regards these as exogenous structural variables (Gottfredson et al. 2005), adding them to models only as control variables (e.g., Gottfredson et al. 2005; Stewart 2003; Welsh 2000). There are a number of problems inherent to the use of compositional variables, most importantly, that effects sometimes arise as statistical artifacts (Harker and Tymms 2004; Nash 2003). To be sure of a genuine effect, all relevant variables need to be entered at the individual level, and, for social scientists, this task is nearly impossible. We argue that, when statistics closely correspond to theoretical expectations, the evidence for the existence of genuine compositional effects is strengthened. However, studies that do incorporate these effects offer no theoretical explanation for them (e.g., Gottfredson and Gottfredson 1985), mainly because they are entered as control variables. This therefore raises the possibility that they are merely ‘‘phantom effects.’’ To try to counter this problem, we offer a theoretical framework that links school social-ethnic composition to school misconduct. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Social-Ethnic School Composition and School Misconduct 227 Strain and Deviance Strain theory offers one of the most prominent explanations for school misconduct (for an extended discussion, see Froggio 2007), stating that feelings of strain cause school misbehavior (Agnew 1985, 1992; Merton 1938). Initially, researchers argued that goal blockage evokes these feelings: because certain students perceive the school as blocking them from realizing their full academic potential (Merton 1938), they become frustrated, leading them to react against that school. Agnew (1985, 1992) revised this in his general strain theory (GST), adding the blockage of pain-avoidance behavior as an important source of strain: being unable to escape aversive situations can engender frustration as well. Agnew stated that this source is particularly applicable to adolescents, who, generally, lack the power to control their own fate. Certain environments can provide a context for both sources of strain. Schools may supply such a context (Agnew 1985). Whether schools cause strain or not depends on several school characteristics. Van Houtte and Stevens (2008), for example, showed that being in a vocational track can cause feelings of strain, resulting in school misconduct. The authors measured feelings of strain by the students’ sense of futility. This measure assesses the students’ perception of mastery concerning educational goals: if students feel futile, they perceive that they cannot manipulate their own scholastic success (Brookover et al. 1978; Brookover and Schneider 1975). This is clearly a valid indicator of goal blockage in a school context. School Composition and Strain Brookover and colleagues (1978) stated that feelings of futility are more likely to arise in schools with a certain student composition. First, schools with a majority of migrants—we call these henceforth schools with a high ethnic concentration (cf. Leman 2002, see Variables section)—were expected to engender more of these feelings. This makes sense because they are often regarded as low-achieving, disadvantaged contexts where students perform worse (Bankston and Caldas 1996, 1998). Moreover, in these schools, teachers expect less from their students, resulting in lower academic standards (Brookover et al. 1978). Researchers argued that, as a result, also the students expect less from their academic career (Bankston and Caldas 1998; Brookover et al.; Miller 1980): they would perceive the disadvantaged context as a blockage. These schools may also evoke a higher perception of blockage of pain-avoidance behavior, Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 228 J. Demanet and M. Van Houtte Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 because we can expect that students see them as aversive environments. Since in Belgium, as in other countries (e.g., Khan 1995), schooling is compulsory, this aversive context is not easy to escape. This can engender feelings of strain and frustration as well. Because students perceive that the school is blocking them, they may ‘‘act out’’ against that institution. Schools characterized by a working-class population can evoke feelings of strain as well. First, these schools are often negatively valued (Willms 1992), and provide a context where lower academic norms are set and teachers’ expectations are low (Brookover et al. 1978; Van Houtte 2003). Brookover and colleagues (1978) argued that, as a result, high levels of academic futility characterize their students. They see high poverty schools as blocking institutions that impede academic achievement, because, although they share the same goals as those set by the mainstream culture, their aspirations are offset by reality: the majority of the students leaving the school experience no upward mobility (Stretesky and Hogan 2005). Because of this, the students see these institutions as mainly aversive environments and therefore do not see the point of attending school. Moreover, high poverty students have an even slighter chance of quitting school legally. Not only is education compulsory, they simply do not have the means to search for a different, more highly valued school, which is often located at a greater distance from their home (Mahieu 2002). School Misconduct as a Shared Response General strain theory is an individualistic approach (Froggio 2007): it explains why individuals demonstrate deviant behavior. (Sub)cultural theories (Cohen 1955; Ogbu 1978; Willis 1977), on the other hand, stress the shared character of delinquency, arguing that it results from a shared deviant or oppositional subculture among adolescents. Cultural deprivation theory relates this to socioeconomic status (Cohen 1955; Willis 1977), oppositional culture theory to ethnicity (Ogbu 1978). Cultural deprivation theory states that working-class children perceive that they compete for status on unequal grounds (Cohen 1955; Willis 1977): schools are middle-class dominated institutions, and, because middle-class children learn the appropriate values at home while working-class children do not, the former have an advantage in the quest for status. According to Cohen (1955), the working-class child searches for other means to gain status. Status is nothing else than recognition from other people: the necessary condition for this Social-Ethnic School Composition and School Misconduct 229 strategy to be successful is that these new sources of status are accepted by others—in other words, that they form a culture. The resultant subculture explicitly counters the values of the mainstream middle-class society: deviancy is highly valued. As such, the culture becomes a deviant subculture. Cohen (1955, p. 155) argued that a student is evaluated against the entire student body of a school. It follows that working-class children have the lowest chance of competing for status successfully when confronted with a middle-class majority in their schools: in this context, it is easy to make the choice to turn to a deviant subculture. However, another possibility arises. If a group with specific attitudes is large enough, that group can impose their attitudes on the entire school ethos (Wilson 1959): in this manner, individual attitudes can grow into a subculture. Since even schools with a predominantly working-class population are seen as middle-class dominated institutions, blue-collar students may still have the impression that they compete for status with unequal chances of success. Thus, the seed is sown for the emergence of a deviant subculture, and if the majority of students in a school adhere to the subcultural values, this may influence other students to follow the subculture as well. If this expectation is correct, an overrepresentation of working-class students in schools will engender more school misconduct among their students. Oppositional culture theory links these ideas to ethnic minority status (Ogbu 1978). The theory distinguishes three groups of minorities: autonomous minorities, immigrant minorities, and caste-like minorities. The first two categories migrated more or less voluntarily to their host countries. As a result, they use the people back in their home country as a yardstick to compare their own situation against; this comparison turns out mostly positive. Because caste-like minorities did not migrate voluntarily, they lack such a valid yardstick, and, as a result, they compare their own situation to the natives of their host country. Relative to them, they are often disadvantaged: they have more difficulty in achieving upward social mobility. This leads to a loss of motivation, causing them to turn against the schools, which are seen as ethnic majority dominated institutions. An oppositional culture takes root, much in the same way as described above. Although this theory was developed in the American context, researchers argue that it can be applied to the European context (Luciak 2004), where the majority of immigrants could be seen as caste-like minorities (Roosens 1989). Researchers assert that this oppositional culture is more prevalent in high ethnic concentration schools—schools where ethnic minority students are overrepresented Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 230 J. Demanet and M. Van Houtte Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 (Fordham and Ogbu 1986). In fact, research has shown that, when high concentration exists, people may oppose themselves to the context that is segregated (E. Anderson 1994; Massey 2004), in this case, the schools. However, the presence of an optimistic culture in high ethnic concentration schools contradicts this line of reasoning (Frost 2007; Goldsmith 2004): in such schools, members of ethnic minority groups, especially, would be more likely to plan to finish high school and pursue higher education, and to develop more proschool attitudes and therefore more proschool behavior. Goldsmith states that the optimistic feelings are not purely individual, but that a true culture of optimism exists in these schools. The proschool attitudes may stem from individual students, but eventually they become shared attitudes, affecting other students in the school as well, much like the mechanism described by Wilson (1959). Following Goldsmith’s thinking, we can expect to find that an optimistic culture characterizes ethnically concentrated contexts, which may prevent students from exhibiting deviant behavior. Futility as a Key Concept The key concept found in the above-mentioned theories, then, is sense of futility, but while strain theory places it at the individual level, the (sub)cultural theories hold that this futility is shared, forming a culture of futility. This theoretical exploration leads to several hypotheses: first, that ethnic and SES composition are related to school misconduct, although we cannot yet be sure about the direction of the effects; second, that futility, either individual or shared, is at least in part responsible for the influence(s) observed. METHODOLOGICAL FRAMEWORK Research Design The research question at hand is situated on two levels: characteristics at the school level are hypothesized to affect an individual outcome. Multilevel analysis has been found to be the best technique for investigating this (Snijders and Bosker 1999; Thrupp et al. 2002). As is common for delinquency measures (Crosnoe 2002; Stewart 2003), the dependent variable was significantly skewed (1.58, SE ¼ 0.023) toward its lower end. Using other techniques, we tested whether this Social-Ethnic School Composition and School Misconduct 231 affected our results.1 The same picture emerged whether we used linear or more complex, nonlinear models. For ease of interpretation, we present the linear multilevel results in this paper. First, we estimated an unconditional ‘‘null’’ model to determine the school-level variance. The social-ethnic composition variables were entered in a subsequent model. Compositional effects are established when the composite effect is found to be significant over and above the individual effect (Wilson 1959). We should point out that a high correlation exists between the two compositional features (r ¼ À0.784; p < .01). Because other studies have established an effect of ethnic composition in addition to the effect of SES composition (e.g., Eitle and Eitle 2003), we chose to use the variables simultaneously, although the results should be regarded with caution (possible multicollinearity). In the next model, we controlled for other features. At the school level, we entered school sector. Migrants in Flanders mostly attend public schools, because private schools in Flanders are mainly Catholic schools, while migrants are generally Muslim. Public schools are also overrepresented in urban areas, which is where the majority of migrants and working-class children in Flanders live. We also controlled for school size, because most studies have shown that students in larger schools have a higher chance of showing school misconduct (Stewart 2003). At the student level, we controlled for SES, gender, grade, ethnicity, and parental support. Research has namely shown that males, students in their early adolescence, and those receiving less parental support, are at higher risk for school misconduct (Ando et al. 2005; Demanet 2008; Ma 2002). Moreover, student’s SES and minority status have to be controlled for in order to establish true compositional effects (Wilson 1959), and because studies have provided indications that lower-SES students and migrants are at greater risk to feel academically futile (Brookover 1978; Miller 1980). We also took into account whether students attended a vocational track. The Flemish school system can be categorized as ‘‘explicit school-level tracking to different school types catering to specific student groups,’’ using achievement as a selection criterion (Trautwein 2006, p. 789; Van Houtte et al. 2010). These explicit tracking systems mostly have a profound impact on the students, determining their educational and occupational success. Hence, attending a vocational 1 We used HLM6 to perform an overdispersed Poisson model with constant exposure, which yielded the same basic image as the linear multilevel model. Following that, we also took the square root of our dependent variable, reducing its skewness to normality. This mechanism produced the same basic results as the ones shown in Tables 3–5. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 232 J. Demanet and M. Van Houtte Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 track in Flanders is rarely a positive choice, and vocational students are all too aware of their low status in society, yielding more antischool attitudes, more feelings of futility, and more misconduct (for an extended discussion of the Flemish tracking system, see Van Houtte et al. 2010; Van Houtte and Stevens 2008). Because of all this, and because migrant and lower SES students are more likely to attend that track in Flanders, it is important to account for this in our analyses. Individual futility was added in the next step, followed by adding the culture of futility into the next model. Finally, we also controlled for prior achievement. This was done only in the last step, because our measure of prior achievement was a rather crude one (see Variables section) and had a large number of missing values. However, as prior achievement has been linked to both school deviance and our compositional characteristics (Bankston and Caldas 1996, 1998; Roeser and Eccles 1998), it was necessary to control for this. As is common in multilevel analysis, all but the dichotomous variables were grand mean centered to ensure model stability. Data The data we used was part of the FlEA (Flemish Educational Assessment), and was gathered in the school year 2004–2005 in 85 Flemish secondary schools. Flanders is the northern, Dutch-speaking part of Belgium. For the data acquisition, we used multistage sampling. At first, we selected proportional-to-size postal codes, with the size defined by the number of schools within each postal code, so that municipalities with more schools would have a greater chance of being selected. The Flemish Educational Department provided this information. From the 240 Flemish postal codes, we selected 48. This resulted in the desired overrepresentation of the larger municipalities. Our second step was to select all regular secondary schools in the chosen postal codes that provided a third and fifth grade (which corresponds to grades 9 and 11 in the American system), ultimately yielding a response rate of 31 percent. This low response rate is due to schools in Flanders being swamped with research requests. Schools choose the research they want to take part in on a first-come, first-served basis, and, as such, no systematic biases occurred (Van Houtte et al. 2005). In the participating schools, we asked all third- and fifth-grade students present at the time to fill out the questionnaire. Students filled out questionnaires in class, under the supervision of one or two members of the research team and a teacher. A total of 11,945 students completed the questionnaire, of Social-Ethnic School Composition and School Misconduct 233 which 11,872 (87%) proved valid: 6,081 in the third grade, and 5,791 in the fifth grade. The questionnaires were not anonymous because we needed to couple other data provided by the school to the students’ responses. Ultimately, however, we removed all names. Variables Outcome We measured our dependent variable, school misconduct, using a self-reported scale, inspired by Stewart (2003, pp. 602–604). It consists of 17 items, some referring to ‘‘school misbehavior,’’ others to ‘‘school beliefs.’’ Students were asked how often they performed deviant acts, such as ‘‘smoked at school,’’ ‘‘been late for school,’’ ‘‘done drugs at school,’’ and so forth. Students could answer using a 5-point scale, ranging from never (1) to very often (5). These scores were summed up, resulting in a scale ranging from 17 to 85 (M ¼ 30.04, SD ¼ 8.47, Cronbach’s alpha ¼ 0.87, N ¼ 11,566; see Table 1). It has been shown that using self-reported measures is not ideal for measuring deviant acts (Crosnoe 2002), but it nonetheless remains the most common method of gathering such information (e.g., Gottfredson et al. 2005; Stewart 2003). We interpolated missing values by item correlation substitution (Huisman 1999), a technique that assigns the score of the closest correlated item of the scale to the missing item. The dependent variable was significantly skewed (1.58, SE ¼ 0.023) toward its lower end. For this reason, we performed nonlinear techniques on the data, but the same basic picture emerged. We present the results of the easier to interpret linear technique here. School-Level Variables Ethnic composition. We measured this compositional characteristic by the proportion of immigrants in school. We asked the administrators to estimate this; however, 12 (14.12%) of the 85 administrators chose not to answer this question. Additionally, we computed the proportion of immigrants at school using individual-level data (see below). The correlation of 0.88 (p < .001) between the two measures validates this aggregated measure. The 85 schools in our sample cover the entire range of ethnic composition, from 0 percent (six schools) to 88.20% (one school). In Flanders, researchers usually distinguish between schools with a more than 50% migrant student population—from now on, ‘‘high concentration schools’’—and schools with a less than 50% migrant population (Leman 2002; Mahieu 2002). Within the 50% or less schools, Leman (2002) distinguished Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Table 1. Descriptive statistics for variables: Frequencies (%), means (M), standard deviations (SD), Cronbach’s alpha, and n % M SD Cronbach’s alpha n Variables Dependent 30.04 16.45 40.0% 48.2% 4.80 10.09 Public 50.6% 10.00 Female Third Native 51.2% 88.8% 5.20 31.36 Vocational 21.2% 69.42 9.22 10,713 2.10 5.40 51.4% 3.20 0.75 11,615 11,843 11,872 11,870 11,137 11,529 1.23 0.77 Low concentration school Medium concentration school 21.70 8.47 0.87 11,561 85 School misconduct School level Ethnic composition 234 SES composition Culture of futility School sector 85 85 85 Student level Sense of futility Gender Grade Ethnicity SES Parental involvement Vocational track Prior achievement Social-Ethnic School Composition and School Misconduct 235 between ‘‘monocultural schools’’ (0%–5% migrants)—from now on, ‘‘low concentration schools’’—schools with 5%–20% migrants, and ‘‘multicultural schools’’ (20%–50%). We combined the latter two categories, because they are both a mixed situation—from now on, we call these ‘‘medium concentration schools.’’ This created a categorical variable with three categories. Our sample contained 34 low concentration schools, 41 medium concentration schools, and 10 high concentration schools. In order to compare high concentration schools to both other types of school we created two dummy variables for our analyses—low concentration (coded 1), and medium concentration schools (coded 1)—with high concentration schools as reference category. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 SES composition. As is common (see e.g., Opdenakker and Van Damme 2001), this compositional school characteristic was measured by calculating the mean of the SES (1 ¼ unskilled manual labor; 8 ¼ professionals and large proprietors; see the Appendix) of the students per school. The schools in our sample ranged from 2 to 6.72, with a mean of 4.80 (SD ¼ 1.23; see Table 1). There was a significant difference between the ethnic school types (low concentration: 5.71, SD ¼ 0.715; medium concentration: 4.52, SD ¼ 0.995; high concentration: 2.89, SD ¼ 0.534; p < .001). Culture of futility. We constructed the culture of futility scale by aggregating the individual sense of futility to the school level. This individual scale (based on Brookover et al. 1979; see below) consists of items such as ‘‘People like me will never do well in school even though we try hard’’ (for all items, see the Appendix). We aggregated this scale by calculating the mean sense of futility in each school, a common way to aggregate individual beliefs (see e.g., Hofstede et al. 1990). However, we could not simply assume that this culture of futility actually exists: we needed to investigate whether sense of futility is really shared at the school level. To assess this, we used the index of ‘‘mean rater reliability’’ (Glick 1985; Shrout and Fleiss 1979; Van Houtte and Stevens 2008), which is based on the intraclass correlation in a one-way analysis of variance. The intraclass correlation (ICC), which should be greater than 0.60 for a legitimate aggregation, was calculated by (Between Mean Square-Within Mean Square)=Between Mean Square. The ICC was 0.84: the schools differed significantly on their mean level of sense of futility (F ¼ 6.249; p < .001). The mean value for this measure was 10.09 (SD ¼ 0.77; see Table 1). This differed significantly between the ethnic types (low concentration: 9.80, SD ¼ 0.62; medium concentration: 10.27, 236 J. Demanet and M. Van Houtte SD ¼ 0.84; high concentration: 10.37, SD ¼ 0.68; p < .05), although the difference between medium and high concentration schools was not significant. School-level control variables. School sector is a dichotomous variable (0 ¼ private school, 1 ¼ public school). It is important to note that in the Flemish educational system, no distinction is made between public and private schools with respect to state support. In the data, 50.6% of the schools are public, which is a slight overrepresentation of the Flemish situation. This is because we oversampled larger municipalities. Because private schools in Flanders are mainly Catholic, high concentration schools are overrepresented among the public schools (16.3% in the public sector, 7.1% in the private sector); the opposite is true for low concentration schools (25.6% in the public versus 54.8% in the private sector). School size was measured by asking the administrators for the number of students in school (M ¼ 461.55; SD ¼ 285.27). However, we obtained information from only 83 of the 85 schools. As multilevel analysis does not permit missing values at the second level, and our analyses showed that school size exerts no influence on school misconduct,2 we omitted this variable. Individual-Level Independent Variables Sense of futility. This is the main independent variable on the individual level. We used a scale consisting of five items (inspired by Brookover et al. 1979) to measure it. We should point out that Brookover and his colleagues (1975; 1978; 1979) saw sense of futility as an aspect of school climate. However, they measured sense of futility at the individual level. Moreover, mastery (Pearlin et al. 1981), situated at the individual level, is closely linked to the sense of futility (r ¼ À0.359, p < .001). Therefore, we hypothesize that sense of futility can be seen as an individual characteristic. This proved to be a successful assumption in the past (see Van Houtte and Stevens 2008). Examples of items are: ‘‘People like me will not have much of a chance to do what we want to in life’’ and ‘‘At school, students like me don’t have any luck.’’ (for all items, see the Appendix). Students had five possible answers, ranging from absolutely do not agree to totally agree (1–5). We computed missing items by item correlation substitution (Huisman 1999). The answers were summed up, yielding 2 Ã Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 c ¼ 0.000 (SE ¼ 0.001); p > .05. Social-Ethnic School Composition and School Misconduct 237 a range of 5 through 25 (Cronbach’s alpha ¼ 0.75, N ¼ 11,620; see Table 1). Individual control variables. With respect to gender (0 ¼ boy, 1 ¼ girl), our sample was evenly divided: 51.4% were girls. Grade was also evenly distributed: 51.2% attended the third grade. We assessed ethnicity using multiple questions. The principal criterion was the birthplace of the maternal grandmothers. If this was missing (1%), we considered the nationality of students’ mothers and fathers, as most immigrants are second- and third-generation citizens and have Belgian nationality. As is common practice in European research, only West European birthplaces and nationalities were considered as native descent (e.g., Timmerman et al. 2002). Additional criteria, in case of missing data regarding nationality (father: 4%, mother: 3.3%), were the language spoken at home (other than Dutch), religion (Islam), and the student’s name (e.g., Felouzis 2003). This resulted in a dichotomous variable (0 ¼ native, 1 ¼ immigrant); 11.2% (see Table 1) were immigrants. The SES was measured by the occupation of the father or the mother (Erikson et al. 1979), or, if they were unemployed, their last profession. If both worked, we used the highest ranked profession as the SES of the family (for coding, see the Appendix). The respondents in our data covered the entire range of SES (1 ¼ unskilled manual labor; 8 ¼ professionals and large proprietors). The mean was 5.20 (SD ¼ 2.10; see Table 1). To measure parental involvement, we used an index consisting of 10 items (inspired by Muller 1995; 1998; Rumberger 1995). The items ranged from general questions, which could be answered using a 5-point scale with answers ranging from never to always (e.g., ‘‘Are your parents interested in what happens at school?), to questions indicating parental involvement in school decisions (‘‘Are [is] [one of] your parents a member of the parental board?’’) and questions about whether parents were acquainted with the parents of fellow students. The resulting index ranged from 10–45 (M ¼ 31.36; SD ¼ 5.40; see Table 1). We also distinguished students who attend a vocational track (1 ¼ vocational track). Among respondents, 21.2% attended that track (see Table 1; 6.8% in low; 34.3% in medium; and 59.3% in high concentration schools). We measured prior achievement by GPA (Grade Point Average) from the preceding school year. To grade their students, Flemish high schools use a percentage, hence, grades range from 0% to 100%, with 50% being the passing grade. In our data, students ranged from 41% to 100%, with a mean of 69.42% (SD ¼ 9.22; see Table 1), corresponding grossly to a ‘C’ in the U.S. high school system. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 238 J. Demanet and M. Van Houtte This measure should be considered carefully. As no standardized tests (for example, state-administered tests) exist in Flemish education, it is hard to compare measures of academic achievement across schools. Furthermore, as this is a self-reported measure, it could contain biases due to memory problems and cover-up strategies. As a result, it has a large number of missing values (9.8%). Consequently, we entered the measure only in the very last step. RESULTS It is clear from the unconditional model (see Table 2) that levels of school misconduct vary significantly across schools: the unconditional model shows that 7.1 percent (r2 ¼ 67.107; s0 ¼ 5.156; p < .001) of its variance is between schools. School composition influences school misconduct (see Table 3, Model 1). We found that school misconduct is more likely in schools with a lower mean SES (cÃ ¼ À0.111; p < .05). Although there was no significant difference between low and high concentration schools (cÃ ¼ 0.114; p > .05), there was a significant difference between medium and high concentration schools (cÃ ¼ 0.131; p < .01): students attending the former were more likely to exhibit school misconduct. Again, there are no indications that we are dealing with multicollinearity, but we should regard this with caution, given the high correlation between SES-context and ethnic composition. After adding the control variables (see Table 3, Model 2), this association diminishes slightly, but remains significant (cÃ ¼ 0.081; p < .05). The effect of SES composition, however, disappears (cÃ ¼ À0.043; p > .05), due to vocational education: students in lower SES schools are more likely Table 2. HLM unconditional model characteristics: Variation between schools in student school misconduct Characteristic Intercept Parameter variance Within school Between schools HLM reliability estimate Proportion of variance between schools Value 30.180ÃÃÃ 67.107 5.156 0.862 0.071ÃÃÃ Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Note. HLM ¼ hierarchical linear modelling. v2 (84, N ¼ 11,872) ¼ 887.970. ÃÃÃ p .001. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Table 3. The association between school composition, individual and shared sense of academic futility, and individual school misconduct. Results of stepwise multilevel analysis Model 1 28.345ÃÃÃ (0.765) c cÃ c cÃ c cÃ c cÃ c cÃ 29.696ÃÃÃ (0.811) 29.516ÃÃÃ (0.865) 29.805ÃÃÃ (0.878) Model 2 Model 3 Model 4 Model 5 30.050ÃÃÃ (0.790) Variables Intercept School level SES composition Low concentration school Medium concentration school À0.766Ã À0.111Ã (0.305) 1.975 0.114 (1.119) 2.211ÃÃ 0.131ÃÃ (0.755) À0.299 À0.043 (0.210) 1.302 0.075 (0.783) 1.373Ã 0.081Ã (0.630) 1.332ÃÃÃ 0.079ÃÃÃ (0.320) À0.223 À0.032 (0.208) 1.010 0.058 (0.816) 1.174 0.069 (0.674) 1.306ÃÃÃ 0.077ÃÃÃ (0.308) 239 c cÃ c cÃ c À3.153ÃÃÃ À0.186ÃÃÃ (0.220) 0.797ÃÃÃ 0.094ÃÃÃ (0.109) À0.496 À3.077ÃÃÃ À0.182ÃÃÃ (0.218) 0.880ÃÃÃ 0.104ÃÃÃ (0.108) À0.479 School sector Culture of futility À0.039 À0.006 (0.254) 0.680 0.039 (0.890) 0.930 0.055 (0.752) 1.258ÃÃÃ 0.074ÃÃÃ (0.294) 0.455 0.041 (0.292) 0.002 0.000 (0.223) 0.299 0.017 (0.723) 0.914 0.054 (0.613) 1.232ÃÃÃ 0.073ÃÃÃ (0.238) 0.380 0.035 (0.172) Student level Gender Grade Ethnicity À3.068ÃÃÃ À0.181ÃÃÃ (0.223) 0.874ÃÃÃ 0.103ÃÃÃ (0.108) À0.468 À2.795ÃÃÃ À0.165ÃÃÃ (0.219) 0.717ÃÃÃ 0.085ÃÃÃ (0.118) À0.527 (Continued ) Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Table 3. Continued Model 1 cÃ c cÃ c cÃ c cÃ c cÃ c cÃ Model 2 Model 3 Model 4 Model 5 Variables SES Parental involvement Vocational track À0.018 (0.377) 0.243ÃÃÃ 0.060ÃÃÃ (0.042) À0.387ÃÃÃ À0.247ÃÃÃ (0.019) 0.986ÃÃ 0.048ÃÃ (0.332) Sense of futility 240 4.674ÃÃÃ 10.052ÃÃÃ 1.526ÃÃÃ 0.382ÃÃ 3.333 0.023 0.011ÃÃ 2.198ÃÃ À0.018 (0.369) 0.268ÃÃÃ 0.067ÃÃÃ (0.043) À0.351ÃÃÃ À0.224ÃÃÃ (0.019) 0.747Ã 0.036Ã (0.311) 0.356ÃÃÃ 0.135ÃÃÃ (0.033) À0.017 (0.373) 0.268ÃÃÃ 0.066ÃÃÃ (0.043) À0.351ÃÃÃ À0.224ÃÃÃ (0.019) 0.710Ã 0.034Ã (0.313) 0.354ÃÃÃ 0.134ÃÃÃ (0.034) Prior achievement À0.019 (0.387) 0.292ÃÃÃ 0.072ÃÃÃ (0.046) À0.346ÃÃÃ À0.221ÃÃÃ (0.018) 0.972ÃÃ 0.047ÃÃ (0.299) 0.255ÃÃÃ 0.096ÃÃÃ (0.039) À0.150ÃÃÃ À0.163ÃÃÃ (0.014) 10.762ÃÃÃ 1.582ÃÃÃ 0.376ÃÃ 3.420 0.024 0.011ÃÃ 1.719ÃÃ 0.038ÃÃÃ 10.675ÃÃÃ 1.718ÃÃÃ 0.371ÃÃ 3.691 0.023 0.011ÃÃ 1.816ÃÃ 0.038ÃÃÃ 8.329ÃÃÃ 1.566ÃÃÃ 0.549ÃÃ 3.777 0.039 0.007ÃÃÃ 1.188Ã 0.058ÃÃÃ 0.008ÃÃÃ Variance components Intercept Gender Grade Ethnicity SES Parental involvement Vocational education Sense of futility Prior achievement U0 U1 U2 U3 U4 U5 U6 U7 U8 Note. The unstandardized (c) and standardized (cÃ ) gamma coefficients are presented, with the standard errors appearing in parentheses, and with the variance components U. Ã p .05; ÃÃ p < .01; ÃÃÃ p .001. Social-Ethnic School Composition and School Misconduct 241 to attend vocational tracks, and therefore, they seem to have a heightened chance for school misconduct (see also Van Houtte and Stevens 2008). The social-ethnic compositional variables explain 9.35 percent of the variance at the school level. We can now answer our first research question: school compositional features do influence school misconduct, although this is true only for ethnic composition. Students attending medium concentration schools are more likely to report school misconduct than students in high concentration schools. Adding the sense of futility alters the picture considerably (see Table 3, Model 3). While individual futility, as expected, has a significant effect (cÃ ¼ 0.135; p < .001) on school misconduct, the difference between medium and high concentration schools disappears (cÃ ¼ 0.069; p > .05): it is clear that the difference is due to lower feelings of futility in high concentration schools. Furthermore, sense of futility partly mediates the effect of attending a vocational track (cÃ ¼ 0.747; p < .05). Adding culture of futility does not affect the other relations, and has no significant relation to school misconduct (cÃ ¼ 0.041; p > .05; see Table 3, Model 4). This answers our second research question: the individual sense of futility, rather than the culture of futility, is responsible for the higher levels of school misconduct in medium concentration schools. While prior achievement does have a significant relation to school misconduct (cÃ ¼ À0.163; p < .001; see Table 3, Model 5), it does not alter the picture. These results raise an important question: Do they reflect the situation of migrants, natives or both? To ascertain this, we conducted the same stepwise multilevel model separately for migrants and natives. Tables 4 and 5 show the results of these analyses. It appears that the relationships we found concern only native students. Students from this group who attend medium concentration schools have a higher chance of school misconduct than their counterparts in high concentration schools (cÃ ¼ 0.095; p < .05, see Table 4, Model 2). No difference occurs between low and high concentration schools (cÃ ¼ 0.095; p > .05; see Table 4, Model 2). The significant difference between medium and high concentration schools disappears (cÃ ¼ 0.083; p > .05; see Table 4, Model 3) when sense of futility is added to the model, which by itself has a rather strong positive influence on school misconduct (cÃ ¼ 0.133; p < .001; see Table 4, Model 3). For migrants (see Table 5) a different picture emerges. Sense of futility has an impact on school misconduct (cÃ ¼ 0.144; p < .01; see Table 5, Model 3), but none of the school variables exerts any influence whatsoever on the level of school misconduct. It is also important to note that the unconditional model for migrants shows that only 3.1% of the variance in school misconduct was between Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Table 4. Association between school composition, individual and shared sense of academic futility, and school misconduct, for natives. Results of stepwise multilevel analysis Model 1 28.887ÃÃÃ (1.123) c cÃ c cÃ c cÃ c cÃ c cÃ 29.161ÃÃÃ (0.895) 29.007ÃÃÃ (0.913) 29.309ÃÃÃ (0.957) Model 2 Model 3 Model 4 Model 5 29.318ÃÃÃ (0.935) Variables Intercept School level SES composition Low concentration school 242 c cÃ c cÃ À3.164ÃÃÃ À0.189ÃÃ (0.199) 0.850ÃÃÃ 0.101ÃÃÃ (0.113) Medium concentration school À1.098ÃÃ À0.161ÃÃ (0.318) 1.724 0.101 (1.275) 1.930 0.115 (0.991) À0.194 À0.028 (0.198) 1.624 0.095 (0.891) 1.645Ã 0.095Ã (0.783) 1.473ÃÃÃ 0.088ÃÃÃ (0.308) À0.100 À0.015 (0.194) 1.254 0.073 (0.917) 1.401 0.083 (0.817) 1.442ÃÃÃ 0.086ÃÃÃ (0.294) School sector Culture of futility 0.098 0.014 (0.247) 0.890 0.052 (0.978) 1.122 0.067 (0.878) 1.397ÃÃÃ 0.083ÃÃÃ (0.282) 0.442 0.041 (0.301) 0.123 0.018 (0.212) 0.730 0.043 (0.893) 1.224 0.073 (0.813) 1.428ÃÃÃ 0.085ÃÃÃ (0.244) 0.281 0.026 (0.276) Student level Gender Grade À3.065ÃÃÃ À0.183ÃÃÃ (0.195) 0.928ÃÃÃ 0.111ÃÃÃ (0.112) À3.053ÃÃÃ À0.182ÃÃÃ (0.198) 0.924ÃÃÃ 0.110ÃÃÃ (0.112) À2.767ÃÃÃ À0.165ÃÃÃ (0.197) 0.773ÃÃÃ 0.092ÃÃÃ (0.128) Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 SES c cÃ c cÃ c cÃ c cÃ c cÃ Parental involvement Vocational track 0.222ÃÃÃ 0.051ÃÃÃ (0.044) À0.405ÃÃÃ À0.255ÃÃÃ (0.021) 1.046ÃÃ 0.049ÃÃ (0.350) Sense of futility 0.248ÃÃÃ 0.057ÃÃÃ (0.046) À0.370ÃÃÃ À0.233ÃÃÃ (0.021) 0.851Ã 0.040Ã (0.339) 0.353ÃÃÃ 0.133ÃÃÃ (0.034) 0.248ÃÃÃ 0.057ÃÃÃ (0.046) À0.370ÃÃÃ À0.233ÃÃÃ (0.021) 0.825Ã 0.038Ã (0.338) 0.350ÃÃÃ 0.132ÃÃÃ (0.035) Prior achievement 0.272ÃÃÃ 0.062ÃÃÃ (0.049) À0.361ÃÃÃ À0.227ÃÃÃ (0.020) 1.061ÃÃ 0.049ÃÃ (0.328) 0.249ÃÃÃ 0.094ÃÃÃ (0.039) À0.160ÃÃÃ À0.173ÃÃÃ (0.013) 10.951ÃÃÃ 0.916ÃÃÃ 0.391ÃÃ 0.035 0.013Ã 1.838ÃÃ 0.035ÃÃÃ 9.624ÃÃÃ 0.853ÃÃÃ 0.657ÃÃ 0.054Ã 0.011ÃÃ 1.364Ã 0.051ÃÃÃ 0.006ÃÃ 243 5.040ÃÃÃ 11.120ÃÃÃ 0.901ÃÃÃ 0.393ÃÃ 0.026 0.015 2.038ÃÃ 11.072ÃÃÃ 0.879ÃÃÃ 0.387ÃÃ 0.035 0.013Ã 1.784ÃÃ 0.035ÃÃÃ Variance components Intercept Gender Grade SES Parental involvement Vocational education Sense of futility Prior achievement U0 U1 U2 U4 U5 U6 U7 U8 Note. The unstandardized (c) and standardized (cÃ ) gamma coefficients are presented, with the standard errors appearing in parentheses, and with the variance components U. p ¼ .054; Ã p .05; ÃÃ p < .01; ÃÃÃ p .001. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Table 5. Association between school composition, individual and shared sense of academic futility, and school misconduct, for immigrants. Results of stepwise multilevel analysis Model 1 29.341ÃÃÃ (0.856) c cÃ c cÃ c cÃ c cÃ c cÃ 30.798ÃÃÃ (1.690) 30.329ÃÃÃ (1.692) 29.960ÃÃÃ (1.649) Model 2 Model 3 Model 4 Model 5 29.548ÃÃÃ (1.613) Variables Intercept School level SES composition Low concentration school 244 c cÃ c cÃ À2.830ÃÃÃ À0.154ÃÃÃ (0.671) 0.822Ã 0.089Ã (0.326) Medium concentration school 0.152 0.021 (0.391) 1.864 0.098 (1.710) 0.425 0.023 (1.052) School sector 0.162 0.022 (0.316) À0.304 À0.016 (1.172) À0.238 À0.013 (0.571) 0.474 0.026 (0.528) 0.109 0.015 (0.322) À0.456 À0.024 (1.178) À0.562 À0.031 (0.624) 0.548 0.030 (0.504) Culture of futility À0.149 À0.020 (0.319) 0.201 0.011 (1.203) À0.109 À0.006 (0.659) 0.535 0.029 (0.496) À0.642 À0.054 (0.389) 0.150 0.020 (0.399) À0.131 À0.007 (1.433) À0.135 À0.007 (0.889) 0.553 0.030 (0.597) À0.176 À0.015 (0.437) Student level Gender Grade À2.536ÃÃ À0.138ÃÃ (0.709) 0.901ÃÃ 0.097ÃÃ (0.324) À2.657ÃÃ À0.145ÃÃ (0.732) 0.904ÃÃ 0.098ÃÃ (0.325) À1.976Ã À0.108Ã (0.765) 0.791Ã 0.085Ã (0.328) Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 SES c cÃ c cÃ c cÃ c cÃ c cÃ Parental involvement Vocational track 0.369Ã 0.090Ã (0.141) À0.291ÃÃÃ À0.199ÃÃÃ (0.053) À0.123 À0.007 (0.694) Sense of futility 0.345ÃÃ 0.084ÃÃ (0.122) À0.224ÃÃÃ À0.152ÃÃÃ (0.055) À0.450 À0.025 (0.671) 0.374ÃÃ 0.144ÃÃ (0.104) 0.351ÃÃ 0.085ÃÃ (0.122) À0.225ÃÃÃ À0.154ÃÃÃ (0.054) À0.265 À0.014 (0.658) 0.392ÃÃ 0.151ÃÃ (0.105) Prior achievement 0.347Ã 0.084Ã (0.134) À0.228ÃÃÃ À0.156ÃÃÃ (0.059) À0.150 À0.008 (0.731) 0.291Ã 0.112Ã (0.112) À0.091ÃÃ À0.106Ã (0.031) 30.049ÃÃ 11.278ÃÃ 1.996 0.080 0.044ÃÃ 5.363Ã 0.206 22.862Ã 8.918Ã 1.513 0.164 0.039ÃÃ 2.994 0.202 0.011 245 2.196ÃÃ 20.486Ã 8.471ÃÃ 1.832 0.240 0.045ÃÃ 6.318 30.517Ã 11.150ÃÃ 1.903 0.071 0.043ÃÃ 5.915Ã 0.210 Variance components Intercept Gender Grade SES Parental involvement Vocational education Sense of futility Prior achievement U0 U1 U2 U4 U5 U6 U7 U8 Note. The unstandardized (c) and standardized (cÃ ) gamma coefficients are presented, with the standard errors appearing in parentheses, and with the variance components U. Ã p .05; ÃÃ p < .01; ÃÃÃ p .001. 246 J. Demanet and M. Van Houtte schools (r2 ¼ 81.641; s0 ¼ 2.618; p < .001), compared to 8.4% for the natives (r2 ¼ 65.173; s0 ¼ 5.999; p < .001). CONCLUSIONS AND DISCUSSION School deviance is a widely studied topic. While researchers such as Agnew (1985) and Hirschi (1969) have shown that the source of deviant behavior can be found in students, others have indicated school features as a cause (e.g., Gottfredson et al. 2005; Reynolds and Jones 1978; Rutter 1973; Welsh 2000). Although this research has produced interesting insights, one dimension of the school climate (see Welsh et al. 2000) remains underexposed (Dumay and Dupriez 2007; Thrupp 2001): the social milieu, which consists mainly of compositional school features. The aim of this article has therefore been twofold: to assess whether the SES and ethnic composition of schools affect the probability that students exhibit deviant behavior and to determine whether individual feelings, rather than a shared culture, of futility are responsible for eventual influences, as hypothesized in our theoretical model. General strain theory (Agnew 1985; 1992) is an individualistic account of deviancy (see Froggio 2007), stating that deviancy results from feelings of frustration. These originate in perceptions of goal blockage and blockage of pain-avoidance behavior. We expected that both are more prevalent in high ethnic concentration and lower SES schools: these contexts give rise to feelings of futility. Cultural deprivation theory (Cohen 1955) and oppositional culture theory (Ogbu 1978) state that deviancy results from a shared subculture that manifests as a counterculture. The theories hold, respectively, that lower SES youth and migrants are at greater risk of engaging in such an oppositional culture, especially in schools with a particular student composition. As the shared character of deviancy is a trademark of these theories, we expected to find that a school culture of futility was responsible for the associations. Our multilevel analyses showed that both the SES and the ethnic composition of schools have an influence on school misconduct. Students attending schools with a lower SES are more prone to deviancy, but further analyses suggested that this might be due to their track: students in lower SES schools have a higher probability of following the vocational track, which seems to be responsible for their increased school misconduct. Hence, it seems that our theoretical assumptions based on general strain theory apply to attending a vocational track, but not to the SES context of schools. Apparently, low SES schools neither engender feelings of frustration nor give rise to a delinquent Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Social-Ethnic School Composition and School Misconduct 247 subculture. The influence of attending a vocational track has been recorded by numerous previous studies (Blomme 1988; Hargreaves 1967; Van Houtte and Stevens 2008). In accord with results by Van Houtte and Stevens (2008), this relation is partly due to greater feelings of futility associated with attending a vocational track. The most significant finding of this article is the influence of the ethnic composition. Counterintuitively, we found students in Flanders to be less deviant in high ethnic concentration schools compared to medium concentrated schools; this is attributable to their lower sense of futility. These results however apply only to native students. Apparently, compared to their counterparts in medium concentrated schools, natives attending high ethnic concentration schools are less likely to perceive their efforts to advance in academic achievement as pointless. Therefore, they have less of a need to react against the school. Contrary to our theoretical expectations, immigrants seem to be unaffected by the ethnic makeup, at least in reference to school misconduct. It is clear that high ethnic concentration, at least in the Flemish school context, does not trigger the emergence of oppositional feelings in migrant students, as was stated by previous researchers (E. Anderson 1994; Fordham and Ogbu 1986; Massey 2004). The relation between sense of futility and deviancy accords with our expectations based on general strain theory, but the link between ethnic composition and futility does not. We hypothesized that high ethnic concentration schools would trigger feelings of goal blockage because of their disadvantaged situation (Bankston and Caldas 1998), but it is clear that their students do not perceive this disadvantage, or do not rate it as blockage. A possible explanation for these counterintuitive results in Flanders lies in reference group theory (Kelley 1952; Merton 1949; Shibutani 1955). Reference group taking is the process in which people compare themselves to individuals or groups. For reference group taking to occur, the group has to be salient for the comparing actor (Richer 1976). This concept contains two interrelated conditions: visibility and meaningfulness. Kao and Tienda (1998) suggest that migrants compare themselves mainly with ingroup members, especially in segregated contexts, because they are more prevalent there and thus more visible. As a result, comparison is not likely to turn out negatively for them, and, hence, does not seem to result in feelings of futility and deviancy. For natives, however, who are by definition the minority group in highly concentrated schools, the reference group mainly consists of migrant students. These have fewer positive prospects in Flanders: they are more likely to be unemployed, to be involved in temporary work, or to be Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 248 J. Demanet and M. Van Houtte employed in poorly esteemed sectors (Vertommen and Martens 2005). Natives in Flemish high ethnic concentration schools may thus be seen as a privileged minority. If so, comparison should turn out mostly positive for them. We suggest that for their counterparts in medium concentrated schools—where relatively fewer migrants are enrolled—comparison results more in relative deprivation, because their reference group consists more of same-status students. The outcome of comparison there seems to be feelings of futility and, ultimately, deviancy. This supports the view of some researchers that relative deprivation theory can be seen as a strain theory (Froggio 2007). This picture of the positive influence of concentration schools for natives in Flanders contrasts sharply with the negative image that exists among its policy makers (e.g., Mahieu 2002). It is, however, in accord with research, both American and Flemish (Frost 2007; Goldsmith 2004; Kao and Tienda 1998; Van Houtte and Stevens 2010), that reports about the optimistic culture of students in concentration schools, which results in higher occupational and educational aspirations. This applies both to immigrants and to natives (Van Houtte and Stevens). From our study, it is clear that there is also less antischool behavior, if only for natives. By showing that natives in ethnically mixed school contexts are at greater risk of school misconduct because of their heightened sense of futility, this study in a specific Flemish context adds to the debate in the United States concerning (de)segregation in schools. Our results fit in with findings of recent U.S. studies that show that desegregation can bring unwanted side effects (e.g., Eitle and Eitle 2003; Goldsmith 2004; Moody 2001). Goldsmith (2004), for example, shows that levels of school conflict rise with increasing ethnic heterogeneity at school. Especially students in biracial schools with equally-sized ethnic groups are at higher risk of engaging in interethnic conflict (Goldsmith 2004; Moody 2001), as this kind of behavior is used to gain control over the school ‘‘turf’’—a viewpoint also known as the group threat theory (Blalock 1967; Longshore 1982). Moreover, teachers working in ethnically diverse school contexts report less job satisfaction and more difficulty in establishing meaningful connections to their students (Freeman et al. 1999). Lastly, in line with our Flemish results concerning deviancy, ethnically desegregated school contexts in the United States yield more school violence, especially under conditions of large ethnic inequality in the communities surrounding the school (Eitle and Eitle 2003). Hence, this study adds to the growing evidence in the United States that desegregated school contexts can bring forth negative consequences for their Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Social-Ethnic School Composition and School Misconduct 249 students, especially when the ratio of students of different ethnic groups tends to fifty-fifty. In the light of the current debate in the U.S. about (de)segregation in schools, it is important that subsequent research focuses on school misconduct and feelings of futility in students attending desegregated schools as well. In answer to our second research question, our results seem to contradict the expectations that we based on oppositional culture theory (Ogbu 1978): individual feelings of futility, not the culture of futility in schools, affect deviancy. Clearly, school misconduct is an individualistic expression of feelings of futility, and does not require a shared subculture. As a result, we cannot conclude that school deviance is an alternative way of gaining status, as was predicted by the (sub)cultural theories. For status acquisition, a shared frame of reference is needed, and we found no proof of this in our data. Moreover, this would suggest that the school context would affect the disadvantaged more, and, as our results apply only to natives, this is clearly not the case. This study draws attention to sense of futility as an important source of deviancy. Subsequent research should focus on the origins of these feelings of blockage, with special attention to school compositional features and aspects of the school organizational process. If it is true that schools, places where children spend most of their childhood, can produce these feelings, we should try to understand these mechanisms thoroughly, so that prevention programs can focus more effectively on the relevant school-level factors. Subsequent research should also take into account more family and peer characteristics, as there are strong indications that these are important determinants of deviancy (e.g., Christie-Mizell 2003; Crosnoe 2002; Crosnoe et al. 2002; Mouttapa et al. 2004). Research should explicitly focus on the situation in high ethnic concentration schools, especially on the interor intra-ethnic peer influences that occur there (see Kao and Tienda 1998). Moreover, further studies should focus on other intermediate mechanisms besides feelings of futility, such as school involvement (Hirschi 1969; Jenkins 1997; Stewart 2003) or self-esteem (e.g., McCarthy and Hoge 1984). In general, research needs to pay more attention to compositional school features, especially the role of ethnic composition in reference to school deviancy, as only a handful of studies exist on this topic (e.g., Eitle and Eitle 2003; Stretesky and Hogan 2005). This study adds to the growing evidence in the US that desegregation, although desirable, can impose some negative influences on their students. These results should, however, not lead to the conclusion that desegregation efforts should be abandoned, as a myriad Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 250 J. Demanet and M. Van Houtte Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 of research, both sociological and educational, has pointed to the various positive consequences of placing students of different ethnicities together (eg. Bankston and Caldas 1996; Rumberger and Palardy 2005; Stearns 2010; Van Houtte and Stevens 2009). No wonder that policy makers in Flanders and the U.S. alike endeavour a dispersal of ethnically diverse students across all schools (Mahieu 2002; Goldsmith 2004). Our results do show, however, that desegregation efforts should be well thought through before implemented, as they can sometimes bring with them unwanted side effects. It is important that research investigates the various outcomes of ethnic composition thoroughly. As few studies have focused on the impact and effects of ethnic composition of schools on various outcomes in Flanders (Sierens et al. 2006), and research on this topic has declined steadily in the last 20 years in the United States (Schofield and Hausmann 2004), further research on this topic is badly needed. REFERENCES Agnew, Robert. 1985. ‘‘A Revised Strain Theory of Delinquency.’’ Social Forces 64:151–167. ———. 1992. ‘‘Foundation for a General Strain Theory of Crime and Delinquency.’’ Criminology 30:47–87. Anderson, Carolyn S. 1982. ‘‘The Search for School Climate: A Review of the Research.’’ Review of Educational Research 52:368–420. Anderson, Elijah. 1994. ‘‘The Code of the Streets.’’ The Atlantic Monthly 273:80–94. Ando, Mikayo, Takashi Asakura, and Bruce Simons-Morton. 2005. ‘‘Psychosocial Influences on Physical, Verbal and Indirect Bullying among Japanese Early Adolescents.’’ Journal of Early Adolescence 25:268–297. Bankston, Carl and Stephen J. Caldas. 1996. ‘‘Majority African American Schools and Social Injustice: The Influence of De Facto Segregation on Academic Achievement.’’ Social Forces 75:535–555. Bankston, Carl and Stephen J. Caldas. 1998. ‘‘Race, Poverty, Family Structure, and the Inequality of Schools.’’ Sociological Spectrum 18:55–76. Blalock, Hubert M. 1967. Toward a Theory of Minority-Group Relations. New York: Wiley. Blomme, Johan. 1988. ‘‘Sociale Klasse, Schoolstatus en Deviant Gedrag’’ [Social Class, School Status and Deviant Behavior]. Tijdschrift voor Sociale Wetenschappen 33:293–300. Brans, Marleen, Dirk Jacobs, Marco Martineillo, Andrea Rea, Marc Swyngedouw, Ilke Adam, Pascal Balancier, Eric Florence, and Tinne Van Der Straeten. 2004. Recherche et Politiques Publiques, le Cas de l’Immigration en Belgique [Research and Policy: A Case Study of Immigration in Belgium]. Gent, Belgium: Academia Press. Social-Ethnic School Composition and School Misconduct 251 Brookover, Wilbur B. and J. M. Schneider. 1975. ‘‘Academic Environments and Elementary School Achievement.’’ Journal of Research and Development in Education 9:82–91. Brookover, Wilbur B., John H. Schweitzer, Jeffrey M. Schneider, Charles H. Beady, Partricia K. Flood, and Joseph M. Wisenbaker. 1978. ‘‘Elementary School Social Climate and School Achievement.’’ American Educational Research Journal 15:301–318. Brookover, Wilbur B., Charles Beady, P. K. Flood, John H. Schweitzer, and J. M. Wisenbaker. 1979. School Social Systems and Student Achievement. Schools Can Make a Difference. New York: Praeger. Bryant, Alison L., John Schulenberg, Jerald G. Bachman, Patrick M. O’Malley, and Lloyd D. Johnston. 2000. ‘‘Understanding the Links among School Misbehavior, Academic Achievement, and Cigarette Use: A National Panel Study of Adolescents.’’ Prevention Science 1:71–87. ´ Christie-Mizell, C. Andre. 2003. ‘‘Bullying: The Consequences of Interparental Discord and Child’s Self-concept.’’ Family Process 42:237–251. Cohen, Albert K. 1955. Delinquent Boys. The Culture of the Gang. New York: The Free Press. Crosnoe, Robert. 2002. ‘‘High School Curriculum Track, and Adolescent Association with Delinquent Friends.’’ Journal of Adolescent Research 17:143–167. Crosnoe, Robert, Kristan G. Erickson, and Sanford M. Dornbusch. 2002. ‘‘Protective Functions of Family Relationships and School Factors on the Deviant Behavior of Adolescent Boys and Girls—Reducing the Impact of Risky Friendships.’’ Youth & Society 33:515–544. Davies, Lynn. 1984. Pupil Power: Deviance and Gender in School. London: Falmer Press. Davies, Scott. 1999. ‘‘Subcultural Explanations and Interpretations of School Deviance.’’ Aggression and Violent Behavior 4:191–202. Demanet, Jannick. 2008. ‘‘Populair of Verstoten? Een Netwerkanalytische Studie naar de Sociale Kenmerken van Pesters in het Vlaamse Secundaire Onderwijs’’ [Popular or Isolated? A Network Analytical Study for the Social Characteristics of Bullies in Flemish Secondary Schools]. Tijdschrift voor Sociologie 29:397–423. Dumay, Xavier and Vincent Dupriez. 2007. ‘‘Accounting for Class Effect Using the TIMSS 2003 Eighth-Grade Database: Net Effect of Group Composition, Net Effect of Class Process, and Joint Effect.’’ School Effectiveness and School Improvement 18:383–408. Eitle, David and Tamela M. Eitle. 2003. ‘‘Segregation and School Violence.’’ Social Forces 82:589–616. Erikson, R., J. H. Goldthorpe, and L. Portocarero. 1979. ‘‘Intergenerational Class Mobility in 3 Western European Societies—England, France and Sweden.’’ British Journal of Sociology 30:415–441. ´ ´ ´ ´ Felouzis, G. 2003. ‘‘La Segregation Ethnique au College et Ses Consequences’’ [Ethnic Segregation at Secondary School and Its Consequences]. Revue Franc aise ¸ de Sociologie 44:413–447. Fordham, Signithia and John U. Ogbu. 1986. ‘‘Black Students’ School Success: Coping with the Burden of ‘Acting White’.’’ Urban Review 18:176–206. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 252 J. Demanet and M. Van Houtte Freeman, Donald J., Susan M. Brookhart, and William E. Loadman. 1999. ‘‘Realities of Teaching in Racially=Ethnically Diverse Schools—Feedback from Entry-Level Teachers.’’ Urban Education 34:89–114. Froggio, Giacinto. 2007. ‘‘Strain and Juvenile Delinquency: A Critical Review of Agnew’s General Strain Theory.’’ Journal of Loss & Trauma 12:383–418. Frost, Michelle B. 2007. ‘‘Texas Students’ College Expectations: Does High School Racial Composition Matter?’’ Sociology of Education 80:43–65. Glick, W. H. 1985. ‘‘Conceptualizing and Measuring Organizational and Psychological Climate: Pitfalls in Multilevel Research.’’ Academy of Management Review 10:601–616. Goldsmith, Pat A. 2004. ‘‘Schools’ Racial Mix, Students’ Optimism, and the Black-White and Latino-White Achievement Gaps.’’ Sociology of Education 77:121–147. Gottfredson, Gary D. and Denise C. Gottfredson. 1985. Victimization in Schools. New York: Plenum. Gottfredson, Gary D., Denise C. Gottfredson, Allison A. Payne, and Nisha C. Gottfredson. 2005. ‘‘School Climate Predictors of School Disorder: Results from a National Study of Delinquency Prevention in Schools.’’ Journal of Research in Crime and Delinquency 42:412–444. Hargreaves, D. H. 1967. Social Relations in a Secondary School. London: Routledge & Kegan Paul. Harker, Richard and Peter Tymms. 2004. ‘‘The Effects of Student Composition on School Outcomes.’’ School Effectiveness and School Improvement 15:177–199. Hirschi, Travis. 1969. Causes of Delinquency. Berkeley: University of California Press. Hofstede, Geert, B. Neuijen, D. D. Ohayv, and G. Sanders. 1990. ‘‘Measuring Organizational Cultures—A Qualitative and Quantitative Study across 20 Cases.’’ Administrative Science Quarterly 35:286–316. Huisman, M. 1999. ‘‘Imputation of Missing Item Responses: Some Simple Techniques.’’ Pp. 91–119 in Item Nonresponse: Occurrence, Causes, and Imputation of Missing Answers to Test Items, edited by M. Huisman. Leiden, The Netherlands: DSWO Press, Leiden University. Jenkins, Patricia H. 1995. ‘‘School Delinquency and School Commitment.’’ Sociology of Education 68:221–239. ———. 1997. ‘‘School Delinquency and the Social Bond.’’ Journal of Research in Crime and Delinquency 34:337–367. Kao, Grace and Marta Tienda. 1998. ‘‘Educational Aspirations of Minority Youth.’’ American Journal of Education 106:349–384. Kasen, Stephanie, Jim Johnson, and Patricia Cohen. 1990. ‘‘The Impact of School Emotional Climate on Student Psychopathology.’’ Journal of Abnormal Child Psychology 18:165–177. Kelley, H. 1952. ‘‘Two Functions of Reference Groups.’’ Pp. 410–414 in Readings in Social Psychology, edited by G. Swanson, T. Newcomb, and E. Hartley. New York: Henry Holt. Kelly, Elinor and Tessa Cohn. 1988. Racism in Schools—New Research Evidence. Stoke-on-Trent, UK: Trentham Books. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Social-Ethnic School Composition and School Misconduct 253 Khan, Anwar N. 1995. ‘‘Compulsory School Attendance in Britain.’’ Journal of Law & Education 24:91–128. Kuperminc, Gabriel P., Bonnie J. Leadbeater, Christine Emmons, and Sidney J. Blatt. 1997. ‘‘Perceived School Climate and Difficulties in the Social Adjustment of Middle School Students.’’ Applied Developmental Science 1:76–88. Leman, Johan. 2002. ‘‘Historiek, Kritiek en Mogelijke Toekomst van de Nondiscriminatieverklaring (NDV) in het Onderwijs van de Vlaamse Gemeenschap’’ [History, Criticism and Possible Future of the Non-discrimination Declaration in Education of the Flemish Community]. Pp. 167–82 in Allochtone Jongeren in het Onderwijs: Een Multidisciplinair Perspectief, edited by Christiane Timmerman, Philip Hermans, and Johan Hoornaert. Leuven–Apeldoorn, The Netherlands: Garant. Longshore, Douglas. 1982. ‘‘Race Composition and White Hostility: A Research Note on the Problem of Control in Desegregated Schools.’’ Social Forces 61:73–78. Luciak, Mikael. 2004. ‘‘Minority Status and Schooling: John U. Ogbu’s Theory and the Schooling of Ethnic Minorities in Europe.’’ Intercultural Education 15: 359–368. Ma, Xin. 2002. ‘‘Bullying in Middle School: Individual and School Characteristics of Victims and Offenders.’’ School Effectiveness and School Improvement 13:63–89. Mahieu, Paul. 2002. ‘‘Desegregatie in Functie van Integratie’’ [Desegregation in Function of Integration]. Pp. 205–232 in Allochtone Jongeren in het Onderwijs: Een Multidisciplinair Perspectief, edited by Christiane Timmerman, Philip Hermans, and Johan Hoornaert. Leuven-Apeldoorn, The Netherlands: Garant. Massey, Douglas S. 2004. ‘‘Segregation and Violent Crime in Urban America.’’ Pp. 217–346 in Problem of the Century: Racial Stratification in the United States, edited by Douglas S. Massey and Elijah Anderson. New York: Russell Sage Foundation. McCarthy, John D. and Dean R. Hoge. 1984. ‘‘The Dynamics of Self-esteem and Delinquency.’’ American Journal of Sociology 90:396–410. Merton, Robert K. 1938. ‘‘Social Structure and Anomie.’’ American Sociological Review 3:672–682. Merton, Robert K. 1949. Social Theory and Social Structure. Glencoe, IL: Free Press. Miller, Stephen K. 1980. ‘‘Raising Low Socioeconomic=Minority School Achievement: Overcoming the Effects of Student Sense of Academic Futility.’’ Michigan Academician 12:437–453. Mouttapa, M., T. Valente, P. Gallaher, L. A. Rohrbach, and J. B. Unger. 2004. ‘‘Social Network Predictors of Bullying and Victimization.’’ Adolescence 39:315–335. Muller, C. 1995. ‘‘Maternal Employment, Parent Involvement, and Mathematics Achievement among Adolescents.’’ Journal of Marriage and the Family 57:85–100. ———. 1998. ‘‘Gender Differences in Parental Involvement and Adolescents’ Mathematics Achievement.’’ Sociology of Education 71:336–356. Nash, Roy. 2003. ‘‘Is the School Composition Effect Real? A Discussion with Evidence from the UK PISA Data.’’ School Effectiveness and School Improvement 14:441–457. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 254 J. Demanet and M. Van Houtte Ogbu, John U. 1978. Minority Education and Caste. The American System in Cross-cultural Perspective. New York: Academic Press. Opdenakker, Marie C. and Jan Van Damme. 2001. ‘‘Relationship between School Composition and Characteristics of School Process and Their Effect on Mathematics Achievement.’’ British Educational Research Journal 27:407–432. Palardy, Gregory J. 2008. ‘‘Differential School Effects among Low, Middle, and High Social Class Composition Schools: A Multiple Group, Multilevel Latent Growth Curve Analysis.’’ School Effectiveness and School Improvement 19:21–49. Pearlin, L. I., E. G. Menaghan, M. A. Lieberman, and J. T. Mullan. 1981. ‘‘The Stress Process.’’ Journal of Health and Social Behavior 22:337–356. Power, M. J., M. R. Alderson, C. M. Philipson, E. Schoenberg, and J. N. Morris. 1967. ‘‘Delinquent Schools?’’ New Society 10:542–543. Power, M. J., R. T. Benn, and J. N. Morris. 1972. ‘‘Neighbourhood, School and Juveniles before the Courts.’’ British Journal of Criminology 12:111–132. Reynolds, David and Dee Jones. 1978. ‘‘Education and the Prevention of Juvenile Delinquency.’’ Pp. 21–44 in Alternative Strategies for Coping with Crime, edited by N. Tutt. Oxford, UK: Basil Blackwell. Richer, Stephen. 1976. ‘‘Reference-group Theory and Ability Grouping—Convergence of Sociological Theory and Educational Research.’’ Sociology of Education 49:65–71. Roeser, Robert W. and Jacquelynne S. Eccles. 1998. ‘‘Adolescents’ Perceptions of Middle School: Relation to Longitudinal Changes in Academic and Psychological Adjustment.’’ Journal of Research on Adolescence 8:123–158. Roosens, Eugene. 1989. ‘‘Cultural Ecology and Achievement Motivation. Ethnic Minority Youngsters in the Belgian System.’’ Pp. 85–106 in Different Cultures, Same School. Ethnic Minority Children in Europe, edited by L. Eldering, J. Kloprogge, and L. Kloprogge. Lisse, The Netherlands: Swets & Zeitlinger. Rumberger, Russell W. 1995. ‘‘Dropping Out of Middle School—A Multilevel Analysis of Students and Schools.’’ American Educational Research Journal 32:583–625. Rumberger, Russell W. and Gregory J. Palardy. 2005. ‘‘Does Segregation Still Matter? The Impact of Student Composition on Academic Achievement in High School.’’ Teachers College Record 107:1999–2045. Rutter, Michael. 1973. ‘‘Why Are London Children So Disturbed?’’ Proceedings of the Royal Society of Medicine 66:1221–1225. Schofield, Janet W. and Leslie R. M. Hausmann. 2004. ‘‘School Desegregation and Social Science Research.’’ American Psychologist 59:538–546. Shibutani, T. 1955. ‘‘Reference Groups as Perspectives.’’ American Journal of Sociology 60:562–69. Shrout, P. E. and J. L. Fleiss. 1979. ‘‘Intraclass Correlations: Uses in Assessing Rater Reliability.’’ Psychological Bulletin 86:420–428. Sierens, Sven, Mieke Van Houtte, Patrick Loobuyck, Katelijn Delrue, and Koen Pelleriaux. 2006. Onderwijs Onderweg in de Immigratiesamenleving [Education on Its Way in the Immigration Society]. Gent, The Netherlands: Academia Press. Slee, Phillip T. 1995. ‘‘Peer Victimization and Its Relationship to Depression among Australian Primary School Students.’’ Personality and Individual Differences 18:57–62. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Social-Ethnic School Composition and School Misconduct 255 Snijders, T. A. B. and R. J. Bosker. 1999. Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. London: Sage. Stearns, Elizabeth. 2010. ‘‘Long-Term Correlates of High School Racial Composition: Perpetuation Theory Reexamined.’’ Teachers College Record 112: 1654–1678. Stewart, E. A. 2003. ‘‘School Social Bonds, School Climate, and School Misbehavior: A Multilevel Analysis.’’ Justice Quarterly 20:575–604. Stretesky, Paul B. and Michael J. Hogan. 2005. ‘‘Segregation and School Disorder.’’ Social Science Journal 42:405–420. Strohmeier, D., C. Spiel, and P. Gradinger. 2008. ‘‘Social Relationships in Multicultural Schools: Bullying and Victimization.’’ European Journal of Developmental Psychology 5:262–285. Teddlie, Charles and David Reynolds. 2000. The International Handbook of School Effectiveness Research. London: Falmer Press. Thrupp, Martin. 2001. ‘‘Recent School Effectiveness Counter-critiques: Problems and Possibilities.’’ British Educational Research Journal 27:443–457. Thrupp, Martin, Hugh Lauder, and Tony Robinson. 2002. ‘‘School Composition and Peer Effects.’’ International Journal of Educational Research 37:483–504. Timmerman, Christiane, Philip Hermans, and Johan Hoornaert. 2002. Allochtone Jongeren in het Onderwijs: Een Multidisciplinair Perspectief [Migrant Youngsters in Education: A Multidisciplinary Perspective]. Leuven, The Netherlands: Garant. ¨ ¨ ¨ Trautwein, Ulrich, Oliver Ludtke, Herbert W. Marsh, Olaf Koller, and Jurgen Baumert. 2006. ‘‘Tracking, Grading, and Student Motivation: Using Group Composition and Status to Predict Self-concept and Interest in Ninth-grade Mathematics.’’ Journal of Educational Psychology 98:788–806. ¨ Van Houtte, Mieke. 2003. ‘‘Reproductietheorieen Getoetst. De Link Tussen SES-compositie van de School en Onderwijscultuur van Leerkrachten en Directie [Reproduction Theories Tested. The Link Between School SES-Composition and the Educational Culture of Teachers and Administration].’’ Mens en Maatschappij 78:119–143. Van Houtte, Mieke and Peter A. J. Stevens. 2008. ‘‘Sense of Futility: The Missing Link Between Track Position and Self-Reported School Misconduct.’’ Youth and Society 40:245–264. ———. 2009. ‘‘School Ethnic Composition and Students’ Integration Outside and Inside Schools in Belgium.’’ Sociology of Education 82:217–239. ———. 2010. ‘‘School Ethnic Composition and Aspirations of Allochthonous Students in Flanders.’’ British Educational Research Journal 36:209–237. Van Houtte, Mieke, Peter A. J. Stevens, An Sels, Karen Soens, and Ronan Van Rossem. 2005. De Invloed van Structurele en Compositorische Schoolkenmerken op Prestaties en Welbevinden van Leerlingen in het Secundair Onderwijs: een Verklaring Via Cultuur. [The Influence of Structural and Compositional School Features on Achievement and Well-Being of Students in Secondary Education: An Explanation Through Culture]. First research report (not published). Ghent: Universiteit Gent, vakgroep sociologie, onderzoeksgroep jeugd, educatie en geslacht. Van Houtte, Mieke, Jannick Demanet, and Peter A. J. Stevens. 2010. ‘‘Self-esteem of Academic and Vocational Students: Does Inter-school Tracking Sharpen the Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 256 J. Demanet and M. Van Houtte Difference?’’ Paper presented at the 9th Conference of the European Sociological Association, Lisbon, Portugal. Vertommen, S. and A. Martens. 2005. ‘‘Allochtone Werknemers op Lokale Arbeidsmarkten’’ [Migrant Employees on Local Labor Markets]. Over-Werk 2–3:73–78. Welsh, Wayne N. 2000. ‘‘The Effects of School Climate on School Disorder.’’ Annals of the American Academy of Political and Social Science 567:88–107. Welsh, Wayne N., Robert Stokes, and Jack R. Greene. 2000. ‘‘A Macro-level Model of School Disorder.’’ Journal of Research in Crime and Delinquency 37:243–283. Willis, Paul E. 1977. Learning to Labour: How Working Class Kids Get Working Class Jobs. Farnborough, UK: Saxon House. Willms, Jon D. 1992. Monitoring School Performance: A Guide for Educators. London: Falmer Press. Wilson, Alan B. 1959. ‘‘Residential Segregation of Social Classes and Aspirations of High School Boys.’’ American Sociological Review 24:836–845. Downloaded By: [Demanet, Jannick] At: 08:59 9 February 2011 Appendix. Coding of SES measure (based on Erikson et al. 1979) 1. 2. 3. 4. 5. 6. 7. 8. Unskilled manual labor Specialized manual labor Skilled manual labor Routine non-manual employees Farmers and smallholders Lower-grade employees and administrators Higher-grade administrators and executives Professionals and large proprietors Items of sense of futility scale (inspired by Brookover et al. 1979) 1. People like me will not have much of a chance to do what we want to in life 2. People like me will never do well in school even though we try hard 3. I can do well in school if I work hard (reversed) 4. At school, students like me don’t have any luck 5. There is no use in working hard at school; a good job is not reserved for people like me

Social-ethnic school composition and school misconduct. Does sense of futility clarify the picture? more

Log In

Reset Password