بزهکاری طبقاتی و خودگزارش شده نوجوانان: مدرک از کشور ترکیه

Abstract This study examined the relationship between social class and self-reported various juvenile delinquent acts in Ankara, the capital of Turkey. Data included 1,710 high school students using a two-stage stratified cluster sample. Such uncommon measures of social class as students' perceptions of their family economic status, the type of place where middle school was finished, home ownership, and car ownership were employed as well as often used measures of social class. The findings indicated that most of these new measures were not related to delinquency and that the class variables had low explanatory power. Various indicators of social class also presented contrary results. While some indicators of class showed that the relationship between class and delinquency was positive, other similar measures indicated the opposite.

Introduction The relationship between class and delinquency had been one of the most important topics among criminologists. While this subject gained much attention in the West, there had been little knowledge on the topic in developing countries. Although some research on the relationship between class and delinquency found that these variables were associated with one another, other research revealed that the association was weak or absent. This finding, however, was inconclusive and culture-specific. It reflected the findings of research in the West, especially the United States. Unlike previous traditional social class measures, such uncommon measures of social class as students' perceptions of their family status, the type of place where middle school was finished, home ownership, and car ownership were also employed. Four measures of class variables had been traditionally used in the literature: gradational (or continuous) measures (see Elliott & Huizinga, 1983), neo-Marxist class measures (see Brownfield, 1986, Farnworth et al., 1994, Hagan & McCarthy, 1992, Jensen & Thompson, 1990 and Messner & Krohn, 1990), underclass measures (see Brown, 1984 and Farnworth et al., 1994), and ecological or aggregate measures (see Tittle & Meier, 1991, and for further information on these measures see Farnworth et al., 1994 and Tittle & Meier, 1990). Studies in the West heavily used such gradational measures as father's occupation (most prevalent), father's education, mother's education, father's income, and mother's income. In the current study, except for the Marxist measures, all of the above measures were utilized. In the West (e.g., the U.S.), the most widely hypothesized relationship between class and delinquency was negative. That is, when the level of socioeconomic status increased, there would be a decrease in delinquency (hereafter, it is called “negative thesis,” see Tittle, 1983). The aim of this study was to explore a group of relatively more comprehensive class variables that might be associated with some delinquent acts (e.g., assault, school delinquency, public disturbance, and total delinquency) and to test the generally accepted hypothesis that lower class individuals are more likely to commit more delinquency than middle and upper classes (e.g., negative thesis) using a self-reported survey of high school students in a developing country, Turkey.

Results Total delinquency Income (e.g., 0–249 monthly family income in million Turkish liras or $0–213 in U.S. dollars) and perceived middle class were the only significant indicators of social class in terms of total delinquency, independent of age, gender, family supervision, broken family, household size, classroom size, school commitment, religiosity, alcohol use, and delinquent friends (see Table 3). In other words, juveniles whose parental monthly family income ranged from 0–249 (e.g., $0–213 in U.S. dollars) were less likely to commit delinquency compared to the juveniles whose parental monthly family income was over 750 (e.g., $642 in U.S. dollars). Those juveniles who perceived their families as middle class likewise were less likely to engage in delinquency compared to those juveniles who perceived their family as rich/very rich. These significant individual measures of social class had negative impacts on total delinquency. Such control variables as being female, family supervision, school commitment, religiosity, using alcohol, and having delinquent friends were statistically significant. While being female, higher level of family supervision, and greater commitment in school had a negative influence on total delinquency, religiosity, alcohol use, and delinquent friends had a positive impact. Among the predictors, school commitment had the greatest influence on total delinquency (beta = − 29.5). Compared to the explained variance by the control variables (adjusted R2 = 41 percent), social class variables accounted for 4 percent, which was very small. Table 3. Regression analysis of class and self-reported delinquency (standardized coefficients) and Independent variables Total delinquency (n = 993) Assault (n = 1,001) School delinquency (n = 1,015) Public disturbance (n = 1,014) Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Constant 26.790 35.375 11.009 14.385 10.241 10.679 5.495 10.126 0–249 millionTurkish liras − .142⁎ − .080⁎ − .099⁎ − .052 − .150⁎ − .089⁎ − .107⁎ − .058 250–499 million Turkish liras − .154⁎ − .062 − .116⁎ − .049 − .158⁎ − .059 − .108⁎ − .042 500–749 millionTurkish liras − .057 .021 − .033 .035 − .103⁎ − .026 − .002 .050 Primary school & below—mother's − .054 − .015 − .051 − .014 − .067 − .033 .019 .040 Middle school—mother's .000 − .017 .007 − .003 − .012 − .028 .002 − .015 High school—mother's − .027 − .013 − .031 − .022 .016 .024 − .073⁎ − .064⁎ Primary school & below—father's − .024 − .013 − .023 − .007 .029 .037 − .100⁎ − .099⁎ Middle school—father's − .044 − .024 − .047 − .024 − .018 − .001 − .046 − .045 High school—father's − .006 − .028 .013 − .015 .014 − .003 − .054 − .067⁎ Perceived middle class − .064 − .074⁎ − .065 − .072⁎ − .038 − .045 − .033 − .044 Perceived below middle & poor class − .057 − .069 − .029 − .057 − .052 − .043 − .052 − .063 Father is not working − .040 − .025 − .045 − .027 − .017 − .005 − .037 − .034 Mother is not working − .039 − .035 − .016 − .007 − .052 − .065⁎ − .033 − .028 Type of place middle school was finished .004 − .025 .002 − .029 .005 − .003 − .016 − .040 Non-ownership of home .022 .023 .037 .045 .016 .017 .004 − .010 Not having car − .030 − .021 − .030 − .025 − .033 − .021 − .026 − .012 Middle social standing district .018 .033 .098⁎ .087⁎ − .058 − .025 .005 .015 Low social standing district .031 .044 .083⁎ .078⁎ − .004 − .003 − .015 .026 State high school − .068 − .007 − .021 .028 − .074 − .024 − .092⁎ − .042 Private high school − .049 − .022 − .007 − .011 − .079⁎ .001 − .037 − .061 State occupational high school − .020 .020 .033 .047 − .024 .053 − .097⁎ − .109⁎ Socioeconomic status − .015 − .031 − .015 − .026 − .003 − .014 − .039 − .052 Age .044 .026 .099⁎ − .045 Gender (= female) − .080⁎ − .191⁎ .005 .061⁎ Family supervision − .236⁎ − .208⁎ − .184⁎ − .184⁎ Broken family − .023 .000 − .025 − .040 Size of household − .020 − .013 − .012 − .020 Classroom size .062 .031 .140⁎ − .063 School commitment − .295⁎ − .177⁎ − .341⁎ − .221⁎ Religiosity .075⁎ .089⁎ .040 .036 Alcohol use .190⁎ .143⁎ .200⁎ .121⁎ Delinquent friends .158⁎ .191⁎ .092⁎ .078⁎ Adjusted R-square .041⁎ .411⁎ .018⁎ .319⁎ .047⁎ .392⁎ .043⁎ .180⁎ a In the above table, 750 and over (in million Turkish liras) monthly family income, university and above education, perceived rich and very rich class, working father and working mother, cities as location of place where middle school was finished, ownership of car, ownership of home, high social standing district, state Anatolian high school, being male, intact family, no use of alcohol, and not having delinquent friends were used as comparison category. b Taking the logarithm of the dependent variables, due to any possibility of skewness, gave similar results as here. ⁎ Corresponds to significance at the .05 level. Table options Assault Among the indicators of social class, middle and low social standing districts were significant and associated positively with assault, independent of the control variables. That is, juveniles residing in low and middle social standing districts tended to engage in more assault than juveniles residing in high social standing districts. Being female, family supervision, school commitment, religiosity, alcohol use, and delinquent friends were significant among the control variables. Whereas being female, greater level of family supervision, and higher school commitment were related negatively to assault, greater religiosity, using alcohol, and having delinquent friends were associated positively with the dependent variable. Being female (beta = − .19), family supervision (− .21), and delinquent friends (.19) were the strongest predictors among the independent variables. While the control variables accounted for 30 percent (e.g., adjusted R2), social class variables explained the small portion of the variance in assault (about 2 percent). School delinquency Income (e.g., 0–249 monthly family income in million Turkish liras or $0–213 in U.S. dollars) and a nonworking mother were the only significant class variables, net of the influence of the control variables. These social class variables had negative impacts on school delinquency (again, the opposite of the negative thesis). Stated differently, adolescents whose family monthly income ranged from 0–249 (e.g., $0–213 in U.S. dollars) were less probable to commit delinquency compared to the juveniles whose family monthly income was over 750 (e.g., $642 in U.S. dollars). Juveniles whose mothers were unemployed had a tendency not to engage in delinquency at schools compared to juveniles whose mothers were working at the time of the administration of the survey. With regard to control variables, age, family supervision, size of classroom, school commitment, alcohol use, and delinquent friends were significant: although age, classroom size, using alcohol, and having delinquent friends had positive impacts on school delinquency, greater family supervision and school commitment had negative impacts. School commitment had the greatest effect on school delinquency (beta = − .34). Whereas class variables accounted for about 4 percent of the variance in school delinquency, the control variables accounted for 34 percent. Public disturbance Fathers with primary school and below education, fathers with high school (e.g., individual social class measures), and state occupational high school (e.g., aggregate social class measure) were significant, irrespective of the effects of the control variables. Both of these individual and aggregate social class measures were associated inversely with public disturbance (e.g., the opposite of negative thesis). In other words, adolescents whose fathers' education was less than high school were less likely to be involved in public disturbances as compared to adolescents whose fathers' education were university and above. Juveniles attending state occupational high schools (for poor students) were less likely to commit public disturbances compared to students attending state Anatolian high schools (for successful students). Family supervision, school commitment, use of alcohol, and delinquent friends were significant. Whereas higher family supervision and school commitment had negative impacts on public disturbance, using alcohol and having delinquent friends had a positive influence on public disturbance. School commitment had the highest influence on public disturbance among the predictors (beta = − .22). Although social class variables explained a small amount of the variance in public disturbance (about 4 percent), the control variables explained 14 percent.