تفاوت های جنسیتی در آثار دادگاه های بزهکاری نوجوانان: ارزیابی نظریه هدایت شونده

Abstract Teen Court (TC) is an innovative juvenile diversion program that has spread rapidly across the United States in recent years. Despite its popularity, rigorous research on TC effectiveness is lacking. This study used data from a recent randomized trial of the effectiveness of TCs to examine gender differences and mediators anticipated by labeling theory. The study found gender differences in the effect of TC on delinquency. TC was found to increase delinquency for males and to have no effect for females. Implications related to the findings are discussed.

Introduction Teen court (TC) is an innovative juvenile justice diversion program that involves teens in judicial decision-making about the behavior of other juveniles who have committed misdemeanor offenses. These programs are an attempt to hold minor offenders accountable for their behaviors, but also to reduce formal labels compared to traditional juvenile justice services. TCs incorporate components of restorative justice and positive peer pressure to deter future crime by these juveniles. The program is growing in popularity despite mixed findings on its effectiveness from evaluation studies. This study extended an earlier study of the effectiveness of TCs (Stickle, Connell, Dugas, & Gottfredson, 2008) by applying criminological theory to understand the mechanisms through which TCs might influence subsequent delinquency. It tested the hypothesis, based on labeling theory, that TCs may increase informal labels. Gender socialization research suggests that males and females will subjectively interpret these labels differently, leading to different influences on self-concept and subsequent delinquent behavior.

Results Hypothesis 1 Table 2 shows that the first set of regression analyses supported the first hypothesis; teens with a more positive self-concept were less likely to report all three types of delinquency. Age also predicted all three types of delinquency. It is well known that older teens are more likely to be involved in delinquent behaviors, thus these findings were expected. Finally, as was reported in the original report, being a male or involved in TC was related to greater involvement in delinquent behaviors. Minorities also reported greater involvement in delinquent behaviors. Table 2. Regression analysis examining the impact of self-concept on delinquency Variable (LN) Last month frequency drug use (N = 75) (LN) Variety drug use since teen court (N = 75) (LN) Delinquent behavior since teen court (N = 75) β p β p β p Positive self-concept - 1.999⁎ .002 - 1.962⁎ .001 - 1.709⁎ .000 Male .087 .728 - .148 .497 .442⁎ .007 Teen court .080 .742 .227 .284 .500⁎ .002 Age .282⁎ .000 .329⁎ .000 .093⁎⁎ .043 White - .126 .615 .076 .729 - .537⁎ .001 R2 = .272 R2 = .410 R2 = .396 Note: Unstandardized regression coefficients are reported. ⁎ p < .01, two-tailed test. ⁎⁎ p < .05, two-tailed test. Table options Hypothesis 2 Table 3 presents results testing the hypothesis of gender differences in the effect of TC on the development of a positive self-concept. Individual t-tests compared treatment and control participants, separately by gender. The difference in self-concept between girls in the treatment and control groups was not significant, however, the difference in self-concept between boys in the treatment and control groups was. Boys involved with TC, on average, reported lower levels of self-concept than boys involved with DJS. In fact, TC boys reported the lowest level of self-concept of all four groups, and although not significantly different from DJS females, TC females reported the highest level of self-concept. Not surprisingly, the effect size comparing treatment and control boys was larger than the effect size comparing girls; the signs also differed. An OLS regression was conducted to further explore the relationship among gender, program, and self-concept. Table 3. Means and effect sizes for positive self-concept and delinquency variables, by program and gender Variable Teen court Department of Juvenile Services Females (N = 26) Mean SD N Mean SD N ES Positive self-concept .741 .225 18 .696 .192 8 .208 Last month frequency drug use .378 .480 18 .325 .385 8 .117 Variety drug use since teen court .589 .463 18 .575 .483 8 .030 Delinquent behavior since teen court .198 .281 18 .077 .082 8 .503 Males (N = 49) Mean SD N Mean SD N ES Positive self-concept .582⁎⁎ .156 24 .717 .165 25 - .840 Last month frequency drug use .625 .749 24 .304 .392 25 .540 Variety drug use since teen court .800⁎ .610 24 .440 .548 25 .622 Delinquent behavior since teen court .397⁎ .324 24 .188 .277 25 .695 ⁎ p < .05, two-tailed test. ⁎⁎ p < .01, two-tailed test. Table options The regression results showed that the p-value for the interaction term between gender and TC approached significance (p = .103). The sample size was small, thus this p-value was notable. This finding coupled with the individual t-tests suggested that the effect of TC on self-concept might vary by gender. Hypothesis 3 After finding support for the first hypothesis (teens with a more positive self-concept engaged in less delinquency) and the suggestion of a negative effect of TC on males' self-concept, differences between TC and DJS participants on the measures of delinquency were compared separately by gender. There were no significant differences in reported levels of delinquency when treatment and control girls were compared (Table 3). On the other hand TC boys, on average, reported more drug use since TC and more delinquent behavior since TC. This finding suggested that of all four groups, TC males engaged in the most delinquency. The effect sizes comparing treatment and control males were much larger than female comparisons for two of the three delinquency measures. Thus, TC appears to have negative effects on males compared to the control, whereas TC appears to have little or no effects on females compared to the control. OLS regression was conducted to further explore the relationship between gender, program, and delinquency. Regression analyses showed that the interaction term between TC and gender failed to significantly predict any of the delinquency variables. Results did show that older teens reported higher levels of last month variety drug use and drug use since TC. Minorities also reported higher levels of delinquent behavior since TC. When the regressions were run without the interaction term, all three significant findings were upheld. In addition, as reported in the original analysis, males and those involved with TC reported significantly more delinquent behavior since TC (p < .01). Despite nonsignificance, the p-values suggested that there may be an interaction between gender and TC in predicting variety of drug use ( Table 4) and delinquent behavior since TC. For the model predicting delinquent behavior since TC, the p-value for the interaction term was .123. Considering the small sample size, this value was notable. The p-value was slightly higher for variety drug use since TC (p = .189), but still notable. Table 4. Regression analysis examining the difference between male and female control and treatment cases on levels on self-concept and delinquency Variable Positive self-concept (N = 75) (LN) Last month frequency drug use (N = 75) (LN) Variety drug use since teen court (N = 75) (LN) Delinquent behavior since teen court (N = 75) β p β p β p β p Teen court⁎ male - .155 .103 .466 .388 .628 .189 .561 .123 Teen court .041 .598 - .042 .921 - .379 .311 .231 .414 Male .016 .833 - .106 .809 - .069 .859 .233 .431 Age .013 .333 .258⁎⁎ .001 .305⁎⁎ .000 .073 .146 White - .067 .150 - .006 .983 .179 .445 - .499⁎ .013 R2 = .157 R2 = .177 R2 = .320 R2 = .270 Note: Unstandardized regression coefficients are reported. ⁎ p < .05, two-tailed test. ⁎⁎ p < .01, two-tailed test.