The Gendered Monitoring of Juvenile Delinquents

Abstract

Through the framework of power-control theory (PCT), we provide a model of juvenile offending that places the gendered-raced treatment of juveniles central to the analysis. We test the theory using a unique sample that is predominately African American, poor, and composed entirely of juvenile offenders. Multivariate models compare the predictive power of many variables, including ones central to PCT, on status offenses and other, more serious, offenses. Gender and race variables were found to be significant, but varied in their impact across models. The interaction between these variables suggests that being Black and female increases the likelihood of sanctioning for status offenses, but not other types of offenses. Contrary to the theory, single-mother-headed households do not seem to produce more delinquent girls than other types of households. The overall findings of this study indicate that patriarchy and white privilege are continuing characteristics of the juvenile justice system.

Keywords

delinquency race/ethnicity African American gender power-control theory single-mother

Early scholarship suggested that the gender gap in delinquency, as evidenced by larger proportions of boys than girls committing offenses, is generated by prescribed social roles (Chesney-Lind, 1989; Smart, 1977). Despite a persistent gender gap over time (Steffensmeier, Schwartz, Zhong, & Ackerman, 2005), criminologists do not adequately address this difference via any one, robust theory (Chesney-Lind, 1989; Parker & Reckdenwald, 2008). Moreover, as incarceration rates keep rising, the rate for girls entering juvenile justice is pronounced (Puzzanchera, Adams, & Sickmund, 2010; Snyder & Sickmund, 2006), particularly among girls of color (Hartney & Silva, 2007). This is not due to an increase in serious offenses. Property crime, for instance, is an area where the gender gap has been closing (Grasmick, Hagan, Blackwell, & Arneklev, 1996; Hagan, Boehnke, & Merkens, 2004). Still, girls are sanctioned more frequently for less serious offenses than boys are (Chesney-Lind & Shelden, 2004; Feld, 2009; Stevens, Morash, & Chesney-Lind, 2011), a phenomenon addressed in this article by our comparison of status offenses with more serious offenses. Status offenses might elicit disparate treatment of juveniles as their substance belies deeper, gendered meanings assigned to particular offenses. Through the framework provided by Hagan, Gillis, and Simpson’s (1979, 1985, 1990) power-control theory (PCT; Hagan, Simpson, & Gillis, 1987), we provide a model of juvenile offending that places the gendered treatment of juveniles central to the analysis.

We utilize a sample that is predominately Black and poor, and composed entirely of juvenile offenders, allowing us to concentrate on arguably the most vulnerable among the juvenile population who are also subjected to the most systemic control. Equally important, the patriarchal structure of the criminal justice system itself, a hierarchical institution characterized by such masculine values as dominance and control, and the tendency to label offenders and offenses differentially by gender and race is taken into consideration. In other words, we examine status offenses, and the relationship between gender, race, and the likelihood of juvenile justice intervention.

A gendered evaluation of status offenses is integral to understanding gendered outcomes in delinquency. There is a long history of punishment inflicted upon girls for violating gender norms (Bernstein, Cardascia, & Rose, 1979; Chesney-Lind & Eliason, 2006), specifically for committing status offenses (Chesney-Lind & Shelden, 2004). Juvenile services were originally developed to ensure normative gender conformity, especially those pertaining to female sexuality (Chesney-Lind & Shelden, 2004). To our knowledge, no study to date examines the occurrence of status offenses in a delinquent population. Comparing status offenses with more serious offenses allows for a more detailed examination of how gender and power operate in juvenile justice.

PCT presents a cohesive explanation of how institutional arrangements are mimicked in families and affect levels of delinquency. It is comprehensive, drawing from various dominant theories in criminal justice, but is primarily concerned with understanding how gender functions in explaining delinquency. According to PCT, patriarchal family structures are linked to differential preferences for risk-taking between boys and girls. This theory suggests that a more patriarchal family will produce sons who take more risks and engage in more criminal behavior than daughters. Less patriarchal or egalitarian families should produce sons and daughters who engage in an equal amount of delinquency. Though these assertions appear to easily lend themselves to hypothesis testing, the operationalization of “patriarchy” has proven to be challenging. We intend to address the methodological complications inherent to PCT by providing a more complex understanding of familial and social arrangements, by including variables like race and caregiver criminal history, and by including known criminological variables, like the juvenile’s psychological disposition.

The study’s sample, rich with variables known to affect delinquency, provides a unique opportunity to expand and refine PCT by focusing on juveniles who have been heavily system-involved. PCT researchers have studied relatively minor delinquent infractions in otherwise non-delinquent samples (Blackwell & Reed, 2003; Leiber & Wacker, 1997; McCarthy, Hagan, & Woodward, 1999; Uggen, 2000). And, with the notable exception of Mack and Leiber’s (2005) research, most studies utilizing PCT do not fully incorporate the impact of race. Before offering our methodological approach, we situate our analysis by providing an overview of the voluminous research on PCT and various theoretical and empirical interpretations of the theory.

Conceptualizing Patriarchy in PCT

PCT began as an attempt to explain males’ overrepresentation in criminality (Hagan et al., 1979). In a patriarchal society, like patriarchal families, girls are subject to more social controls than boys are, explaining the greater propensity of boys to commit crimes. As Hagan et al. (1985) write, “The presence of power and the absence of control create conditions of freedom that permit common forms of delinquency” (p. 1174). Hagan and colleagues describe parental controls as being instrumental (direct supervision) and relational (emotional bonding). As girls are more often subjected to both types of control, they are less likely to engage in risk-taking behavior. In single-mother households, viewed as more egalitarian than traditional two-parent households, mothers are presumed less controlling of both sons and daughters (Hagan, Gillis, & Simpson, 1987; Hagan et al., 1990) thus resulting in similar rates of delinquency for both.

The theoretical implication is that risk-taking behavior will increase among girls as women enter the workforce in greater numbers, particularly if they are also single mothers (Grasmick, Blackwell, & Bursik, 1993; Uggen, 2000). This conclusion is reminiscent of Adler’s (1975) liberation hypothesis, suggesting that female delinquency will accompany women’s workforce entry (Chesney-Lind & Shelden, 2004). Society becomes more criminogenic as women are “liberated,” and families exert less control over daughters. This contention has been widely dismissed by feminist scholars (Morash & Chesney-Lind, 1991) and empirical research has found that despite women’s participation in the workplace, delinquency remains a male-dominated activity (Chesney-Lind, 1989; Hagan et al., 2004). Still, findings are mixed and seem to support aspects of PCT, with some research suggesting that single-mother households are linked to female delinquency (Uggen, 2000).

Hagan and colleagues envisioned a measure of parental level of authority afforded in the workplace as important, but the research is undecided. In assessing patriarchal family structures, tests of PCT have relied on survey questions asking respondents to reflect on the level of power parents exert in the workplace. Though parental employment status is a component pertinent to understanding family dynamics, it is a crude measure of patriarchy, as adults operate in a number of other social contexts besides work. However, PCT does not mandate rigid family structure classifications and later conceptualizations suggest that families, whether they be “non-traditional” or not, can be assessed on a continuum of patriarchy rather than, for example, a dichotomy (Mack & Leiber, 2005; McCarthy et al., 1999).

Contemporary studies of PCT that still utilize parental occupational status now include other measures of patriarchy including parental education level and income (Blackwell & Reed, 2003; Leiber & Wacker, 1997; Uggen, 2000). Our measure of familial patriarchy is multifaceted in that we include the parental education level, employment status, household income, and family structure. Parental control of youth, discussed in further detail in the proceeding, will be mitigated by familial structural characteristics.

Family Structure (Instrumental Control)

Children learn appropriate behavior from their parents according to social learning theories (Bandura, 1977; Biblarz & Raftery, 1999), but which familial construction offers the best environment for children is debatable (Murry, Bynum, Brody, Willert, & Stephens, 2001; Waldfogel, Craigie, & Brooks-Gunn, 2010). PCT has been modified to include a variety of intra-familial structures as well as different parental workplace class positions. Uggen’s (2000) study suggests that single working mothers in authoritative positions were more likely to have delinquent girls than others. Women with workplace authority are posited to exert similar levels of control on all children. A parent’s ability to monitor a child’s free time, instrumental control, is a key factor in PCT.

Some evidence suggests that single parenthood and delinquency are related (Matsueda & Heimer 1987; Rebellon, 2002) and researchers who investigated delinquency and single-parent households saw a diminishing gender gap, purportedly because the socialization of sons and daughters were similar (Hagan et al., 1990; McCarthy et al., 1999). Others are doubtful of this and critique PCT for failing to recognize the structural disadvantages single parents face (Mack & Leiber, 2005; Morash & Chesney-Lind, 1991). In Morash and Chesney-Lind’s (1991) test, PCT “did not reveal a positive relationship of delinquency to upper class status, but rather showed the opposite” (p. 369). In short, these authors suggest that other factors, like peer delinquency and poverty, need consideration and examination.

Observing that girls were less delinquent in all types of families, Morash and Chesney-Lind (1991) also find evidence of different levels of control placed on sons and daughters. In particular, single mothers were more likely to exert control over sons than daughters. While single-mother households are generally presumed to be more matriarchal (Hines, 1997; McCarthy et al., 1999), other variables like income, race, and familial criminality may diminish the mother’s relative power in controlling her children’s behavior. With that, we include the impact of the school environment as a potential source of control.

Juveniles are not just rebelling against parental authority; they are reacting against external controls as well. Thus, an understanding of the juvenile’s social positioning is as crucial to explaining their delinquency as their parents’ social location. We consider the school environment as a separate, important source of control on juveniles to be included in PCT analyses. School delinquency has been considered distinct from the overall category of juvenile delinquency because the school setting itself presents a confluence of different social bonds and interactions that act on juveniles (Hirschi, 1969; Welsh, Green, & Jenkins, 1999). A recent study conducted by Smith-Adcock et al. (2013) highlights the complexity of the school environment and the myriad of controls that act upon students. Their study underscores a juvenile’s self-agency, or vested interest in delinquency, by demonstrating how juveniles actively cultivate their identities through delinquent behavior, and that boys are more likely than girls to mold their reputations in this manner. Gender seems to play a role in predicting the specific type of school delinquency (Nichols, Graber, Brooks-Gunn, & Botvin, 2006) as well as the frequency of occurrence, and is partially explainable by gendered bonds to both the school and to parents (Hart & Mueller, 2013).

Familial Criminality (Relational Control)

The repercussions of parental incarceration on children are well documented, including poor academic performance and inappropriate behavior at home and school (Henriques, 1982; Johnston, 1995). It is also well established that positive parental bonding leads to lower delinquency rates (Cernkovich & Giordano, 1987), but it is unclear how previous parental incarceration affects appropriate parent–child bonding. Unsurprisingly, criminal behavior in families is correlated (Van de Rakt, Nieuwbeerta, & Apel, 2009). Sibling involvement in the juvenile justice system is highly correlated to juvenile offending (Haynie & McHugh, 2003; Rowe & Gulley, 1992) especially among same-sex siblings (Rowe & Farrington, 1997). Applications of PCT have largely neglected to include familial criminality.

Race and Family Structures

PCT has yet to properly incorporate race in explaining the delinquency gender gap. A patriarchal system of control, as experienced in the United States, is not explained by capitalism alone. Indeed, critics of Marx’s theory point out that he does not give full credence to how racial classifications either serve the status quo or are affected by it (Gimenez, 2001). The experiences of Black families are different from those of White families and continue to be affected by a myriad of historical trajectories that place children at a disadvantage. Racism, in other words, affects families, family structures, and children’s reactions to internal and external events.

Mack and Leiber (2005) provide a significant contribution to the study of PCT as they incorporate race, as well as single-motherhood, as part of the explanatory framework. Specifically, they found the following: boys were more likely than girls to be delinquent regardless of single-motherhood or race. Clearly, the conceptualization of “delinquency” is equally important as the measurement of other aspects of PCT. Their measure of delinquency was, like many PCT studies, “non-serious” and “minor” and suggested that future research examine the effect of community variables (like joblessness) on Black and White families.

As this discussion illustrates, singular applications of PCT often neglect the multifaceted ways in which patriarchy and privilege operate. Race, for example, is a significant predictor of poverty (Jonson-Reid, Drake, & Zhou, 2013), health, and life chances (Althoff, Karpati, Hero, & Matte, 2009; Barondess, 2008). Race is therefore integral in explaining how patriarchal power structures affect families.

Method

Data Collection Procedures

Previous tests of PCT have relied on non-criminal samples, but this study presents an opportunity to examine the theory in a high-risk, youthful offender population. The data were gleaned from a retrospective study of 248 randomly selected court files of adjudicated juveniles, active between 2007 and 2009. It was possible for cases to be reopened between the sample window and the data collection date, resulting in identification of new data beyond 2009 to the final date of data collection in February 2012. The first charge recorded was in 1998, and the last recorded charge was in 2010. Funded by a grant from the U.S. Office of Juvenile Justice and Delinquency Prevention, the data were collected from the largest juvenile court in a southern state as part of a plan to improve clinical and social services to system-involved youth, and to develop data for research and quality improvement.

Due to the density of information provided, a large number of variables known to affect delinquency could be examined. To develop a file abstraction instrument, a research team extensively reviewed a random sample of 10 cases and carefully recorded what information was available in each section of the court record. Typical files contained within the youths’ records included police reports, probation officer logs, and psychosocial and offense histories. From this review, an abstraction instrument was developed for data collection from the sample.

Sample

The sample of 248 juveniles included 170 males (68.5%) and 78 females (31.5%). The age at first offense ranged from 6 to 17 years, with an average age at first offense being 13 years. About 35% (86 juveniles) of the sample were White and 63% of the sample (162 juveniles) were Black. Two were of Asian descent, one was Latino, and three were multiracial. Race was measured dichotomously as Black = 1 and White = 0 due to the sample’s composition. Gender was also measured dichotomously with female = 1 and male = 0. In Table 1, the descriptive statistics for all research variables are provided. (see Table 1).

Table 1.

Descriptive Statistics for Full Sample and by Gender.

Variable	Full sample (n = 248)		Females (n = 78)		Males (n = 170)
Variable	M or %	SD	M or %	SD	M or %	SD
White	35.24%		26.32%		39.74%
Black	64.76%		73.68%		60.26%
Psychological index	0.21	0.51	0.38	0.69	0.12	0.35
No. of suspensions	2.56	1.52	2.69	1.46	2.49	1.55
No. of disciplinary school problems	1.93	1.09	2.01	1.11	1.89	1.08
Two-parent family	15.88%		14.47%		16.56%
Single mother	56.65%		60.53%		54.78%
Single father	8.58%		5.26%		10.19%
Other family structure	18.88%		19.74%		18.47%
Mother has at a least H.S. diploma	67.7%		75%		64.60%
Father has at least a H.S. diploma	68.29		82.50%		61.54%
Household income	1,702.83	1,618.7	1,579.51	905.9	1,755.81	1,842.81
Mother employed	60.32		57.38		61.72
Father employed	72.90		65.63		76.00
Total siblings	2.36	2.33	2.48	2.32	2.30	2.34
No. of siblings involved in crime	0.40	0.88	0.52	1.25	0.34	0.67
No. of parental guardians involved in crime	0.60	0.79	0.66	0.89	0.57	0.74
Family dysfunction index	2.26	3.33	3.29	3.83	1.78	2.97

Note. H.S. = high school.

Dependent Variables: Offenses

A total of 165 different types of charges were possible given the state criminal code. Of these, 94 different offenses were recorded within the sample at least once. The number of charges per youth ranged from 1 to 31, with an average of 7.52 offenses per youth (SD = 5.80). Girls were charged with a total of 577 offenses, while boys were charged with 1,287, an overall total of 1,864 charges. Of these, 77 youth were charged with 99 offenses classified as a violent or serious offense by the Federal Bureau of Investigation (including murder, rape, aggravated assault, and robbery) or as a Class A felony offense in this court’s state jurisdiction (which includes FBI violent/serious offenses and charges such as arson and kidnapping). Boys were charged with 76 and girls with 23 Violent/Class A felonies.

A total of 291 status offenses were charged to 132 youth. Status offenses are those for which an adult cannot be charged, and are assigned to youth due to their minor status. Such offenses include underage drinking, truancy, running away, and ungovernable behavior. Altogether, 10 such types of charges were identified with the youths’ files. A count variable of status offenses was calculated. In all, 78 boys and 54 girls were charged with at least one status offense. The average number of status offense charges was 1.17 per youth. Girls were charged at a higher rate than boys, with 54 girls charged with 140 status offenses, and 78 boys charged with 151 status offenses, a significant gender effect, t(106) = 3.58, p = .001. The most common status offenses were Children in Need of Supervision (CHINS)–Ungovernable Behavior charges.

The remainder of the charges (1,474) were neither violent nor status offenses. They did not meet the federal definition of a violent offense, nor were they included in the state’s definition of a Class A felony.

Independent Variables: Psychological and School Problems

As emotion dysregulation has been found to be a contributing factor to delinquency across studies (Trupin, Stewart, Beach, & Boesky, 2002), we included an examination of emotional symptoms and their relationship to delinquency. At least one psychiatric diagnosis for Axis I disorders (as defined by the Diagnostic and Statistical Manual of Mental Disorders [4th ed.; DSM-IV]; American Psychiatric Association, 1994) was given to 130 juveniles, according to their records. These diagnoses include juveniles who had attention deficit disorder (ADD), attention-deficit/hyperactivity disorder (ADHD), conduct and/or mood disorders, depression, and substance abuse disorders. To specifically focus upon emotional dysregulation, an emotional symptoms index was computed that included any DSM-IV diagnoses in the youths’ files that are known to affect the ability to regulate emotions, including mood, depression, or anxiety disorders, or to affect executive functioning, including ADD or ADHD. All other diagnoses were excluded from the index. A composite score was needed because many youth had more than one diagnosis, and diagnosticians assigned different diagnoses to youth through multiple assessments over time. Index scores ranged from 0 to 3, with 3 indicating three different diagnoses known to affect emotion regulation/executive functioning. Thirty-nine youth had at least one such diagnosis. Girls’ mean score on this index was .38, compared with the boys’ average of .12, which was significantly higher, t(111.66) = 3.123, p = .002.

Boys and girls were about equally likely to experience school disciplinary problems, and school suspension rates were proportionately similar. Disciplinary problems that were non-violent in nature, but involved disruptive/defiant behavior, were treated as an ordinal variable (1 = none, 2 = 1, 3 = 2-4, 4 = 5-10, 5 = more than 10). The number of suspensions was also coded as a five-level ordinal item in which 1 indicates 0 suspensions and 5 indicates more than 10. Clearly, disciplinary problems and school suspensions were correlated (r = .62, p = .000).

Familial Patriarchy Measures

Family structure served as the primary variable in this category. The last known place of residence was identified for 237 youth, and unknown in 11 cases. Only 22 (8.9%) were living with both biological parents, and one youth was adopted by a two-parent family. More than half of the sample (129) was living with their biological mothers. About 8.5% (21) were living with their biological fathers. About 15% (38) were living with one or both biological grandparents or with extended family. Those who were placed in a residential home (group, foster, etc.) due to child protective services (CPS) involvement comprised about 4% (10) of the sample. Seventeen youth experienced other situations such as emancipation, homelessness, incarceration, and monitoring at a mental health facility. To best adhere to PCT testing conventions, all living situations not involving a single or two-parent household were recoded as “other,” while single-female-headed households, single-father-headed households, and two-parent families were included separately (all measured dichotomously).

Parental custody information was reported in 180 cases. Custody was disrupted for 68 (38% of valid cases) of the juveniles at least once. Girls were more likely to be separated from their parents than boys were (45% of the girls vs. 34% of the boys).

Education Level, Income, and Employment

Like Mack and Leiber (2005), we substitute social class for occupational authority by assessing the education of the parental guardian(s). Due to the high rate of high school dropouts among the parents (67%), we measured education level dichotomously with 1 = high school graduate/GED (General Educational Development) recipient, and 0 = neither a high school graduate nor GED recipient. We also examined income, which was negatively correlated with boys’ overall offending but not girls’, r(142) = −.168, p = .045. Past year’s reported family income was available for 204 of the juveniles’ households and ranged from US$0 to US$17,500 with a mean of US$1,700 a year. Given the low income of these households, it is unsurprising that 176 of the youths’ files indicated that the family had received, or were currently receiving, some type of government assistance. Last, we considered parental employment, measured dichotomously as 1 = employed, 0 = unemployed. Twenty-nine percent of the households reported that no one in the house was working.

Instrumental and Relational Variables

We included number of siblings in the household as a count variable. The number of siblings (including full, half, and step) reported to be delinquent ranged from 0 to 7, with 68 files reporting 108 delinquent siblings. PCT predicts that older sibling delinquency is linked to younger sibling delinquency, but having any sibling involved in delinquent activity, whether younger or older, may influence delinquency. Parental criminogenic influence was considered similarly. Parental guardians were identified as having a criminal history in 105 youths’ files, with 148 total parents/guardians identified as possessing a criminal history.

As may be expected within a juvenile offender sample, a troubling array of family conflict was identified in the youths’ files. The presence of any abuse was identified for 184 youth, 62 girls and 122 boys. For girls, abuse was mentioned in the files in exactly 50% (31) of the cases, and about 24% (29) of the boys’ cases. Nineteen of 45 youth both had abuse identified in their files and their custody was disrupted. Seventeen girls and 8 boys were identified as sexual abuse victims.

To best capture the severity of family dysfunction as recorded in youths’ files, an index variable was computed that included 13 indicators, identifying if CPS referral was recommended, if CPS was mentioned, if abuse was mentioned, if a worker was assigned, or if an investigation was conducted (coded 0,1). Count variables in the index included the number of separate abuse, physical abuse, sexual abuse, and neglect reports, and number of times custody was disrupted by CPS as indicated in the files. It also included out-of-home placements by CPS and if violence between caregivers was noted in the file (0,1). Multiple instances of abuse naturally weighted problem severity by being included as counts. Values on the index ranged from 0 to 18, with a mean of 2.26. Girls’ files indicated higher levels of family dysfunction, t(121) = 3.09, p = .003.

To construct multivariate models predicting offenses, associations between the independent variables and these charges were computed, and provided in Table 2.

Table 2.

Bivariate Associations.

Measure	1	2	3	4	5	6	7	8	9
1. Status offenses	—	.492**	.259**	.122	.322**	.467**	.060	.241**	.190**
2. All offenses		—	−.014	.206**	.294**	.418**	−.075	.331**	.187**
3. Sex^a			—	.124	.234**	.211**	.026	.055	.038
4. Race^a				—	−.014	−.006	.018	.124	.061
5. Emotional symptoms					—	.431**	−.008	.055	.075
6. Family dysfunction						—	.010	.272**	.331**
7. Number of siblings							—	−.098	.031
8. Number of siblings-criminal								—	.174*
9. Number of parents-criminal									—
Measure	1	2	3	4	5	6	7	8	9
M	1.17	7.52	—	—	0.206	2.26	2.36	0.440	0.600
SD	1.63	5.80	—	—	0.519	3.33	2.31	0.963	0.789
Measure	10	11	12	13	14	15	16	17	18
10. Status offenses	—	.492**	.063	.117	−.044	−.200*	−.056	−.116	.032
11. All offenses		—	.110	.068	−.067	−.352**	−.163*	.058	−.024
12. Mom’s education^a			—	.423**	−.035	−.163	−.090	.059	−.075
13. Dad’s education^a				—	.047	−.042	.143	−.045	−.003
14. Mom’s employment^a					—	.364**	.222**	−.023	.034
Measure	10	11	12	13	14	15	16	17	18
15. Dad’s employment^a						—	.212*	.061	−.002
16. Income							—	−.252**	.280**
17. Single mother household^a								—	−.334**
18. Single father household^a									—
Measure	10	11	12	13	14	15	16	17	18
M	1.17	7.52	—	—	—	—	1,702.15	—	—
SD	1.63	5.80	—	—	—	—	1,591.71	—	—
Measure	19	20	21	22	23		24
19. Status offenses	—	.492**	−.048	.170*	.149*		.258**
20. All offenses		—	−.085	.024	.382**		.407**
21. Two-parent family^a			—	−.218**	−.036		−.136*
22. Other living arrangements^a				—	.044		.101
23. School discipline incidents					—		.616**
24. School suspensions							—
Measure	19	20	21	22	23		24
M	1.17	7.52	—	—	1.94		2.53
SD	1.63	5.80	—	—	1.09		1.53

p ≤ .05. **p ≤ .005.

Eta and phi coefficients reported are equal to Pearson correlation coefficients.

Demographic and Psychological Characteristics

First, the relationship between the outcomes of interest and youths’ race and sex were analyzed. Being female was significantly associated with more status offenses (η = .259, p = .000), but not overall offending. Race was significantly associated with overall offending (η = .206, p = .001), but not with status offending. Higher scores on the emotional symptoms index were significantly and positively correlated with status offending, r(246) = .322, p = .000, and overall offending, r(246) = .294, p = .000.

School Disciplinary Problems

Youth with more frequent disruptive behavior problems at school were charged with more status offenses, r(246) = .149, p = .01, and overall offenses, r(246) = .382, p = .000. Similarly, youth who were suspended from school more often were charged with more status offenses, r(246) = .258, p = .000; and overall offenses, r(246) = .407, p = .000.

Family Structure and Social Class

Single-female-headed households, single-male-headed households, and two-parent households were not significantly associated with either of the outcome offense variables. Only “other” living arrangements (e.g., foster home, residential care, detention, homeless shelter, etc.) were significantly associated with status offending (η = .170, p = .007).

Whether or not the youths’ caregivers graduated high school was not significantly associated with either of the offense outcomes. Household income was not significantly correlated with status offending, but was significantly and negatively correlated with overall offending, r(202) = −.163, p = .02. While mothers’ employment was not associated with any offending, fathers’ employment was significantly and negatively associated with overall offending (η = −.352, p = .000) and status offending (η = −.200, p = .032).

Familial Criminality and Dysfunction

Last, we considered criminogenic influence and family dysfunction and their relationships to delinquency. The number of siblings with criminal backgrounds was significantly correlated with status offending, r(246) = .241, p = .000, and overall offending, r(246) = .331, p = .000. The number of parents/guardians with criminal histories was similarly correlated with offending, but not as strongly—r(246) = .190, p = .003, for status offending; r(246) = .187, p = .003, for overall offending. Family dysfunction was also significantly and positively correlated with status offending, r(246) = .467, p = .000, and overall offending, r(246) = .418, p = .000.

Hypotheses

Figure 1 provides a conceptual model of the variables used in the analysis. The primary hypotheses tested are presented below. In keeping with previous studies of PCT, gender and family structure were the key independent variables, and delinquency served as the dependent variable. However, we also tested the effect of race independently and as an interaction term. Gender was assessed for its individual and interactive impact on status offending. The model and hypotheses identify status offenses as the primary dependent variable, but all variables and hypothesized relationships were also tested on overall offenses.

Hypothesis 1a (H1a): Girls will be charged with more status offenses than boys.

Hypothesis 1b (H1b): Girls who are Black will be charged with more status offenses than boys and girls who are White (and boys who are Black).

Figure 1.

Predictors of status offenses: A conceptual model of power-control theory.

Consistent with PCT, we might expect girls living in households headed by women to commit more offenses than girls from households that are more patriarchal and this should be mitigated by race. However, given the gendered treatment of status offenses in the juvenile justice system, we might see a different effect when predicting status offenses as compared with all other crimes. Status offenses reflect gender norms and thus, in comparison with all other types of delinquency, we might expect more status offenses from girls raised in patriarchal households than from girls raised in less patriarchal households.

Hypothesis 2a (H2a): Girls will be charged with more status offenses in single-mother-headed households in comparison with all other types of family structures.

Hypothesis 2a (H2b): Girls who are Black will be charged with more status offenses in single-mother-headed households.

PCT requires “instrumental” variables of parental control that refer to the supervision and monitoring of juveniles. Again, like Mack and Leiber (2005), we examined the number of siblings in the household as satisfying this aspect of PCT. We modified this variable by including the number of siblings engaged in criminality (as well as including a variable that assesses parental/guardian criminality). Due to the special nature of status offenses, we expected that familial criminality would demonstrate a lack of control placed on juveniles and thus dampen the effect of control said to come from the number of family members. Therefore, we did not expect that familial criminality would impact status offenses (but may predict other types of offenses).

Hypothesis 3a (H3a): Sibling criminality will affect the frequency of overall offending, but not status offending.

Hypothesis 3b (H3b): Parental criminality will affect the frequency of overall offending, but not status offending.

PCT also calls for the inclusion of relational variables as aspects of parental controls. Our sample allowed for an indirect measure of this type of control that may offer a stronger predictor of status offenses than previously used self-reflective survey questions. We examine family dysfunction, expecting that girls would be subjected to the negative effects of family dysfunction more so than boys were (as girls are more likely to be “controlled”). Moreover, we posited that parental working status would be positively related to status offenses, in that there is less monitoring of juveniles, particularly of girls.

Hypothesis 4 (H4): Family dysfunction exerts a stronger effect on girls committing status offenses than on boys.

Hypothesis 5 (H5): Parental employment exerts a stronger effect on girls committing status offenses than on boys.

Finally, we included the school environment as an external measure of control on delinquents. Using PCT as a theoretical guide, we might expect that control exercised by school officials would dampen overall levels of delinquency. However, a record of school sanctions may be more an indicator rather than a suppressor of outside delinquency. Thus, no hypotheses are offered but, as will be clear from the analyses, their inclusion increases the model’s explanatory strength.

Results

To assess the impact of gender and race on offenses, interaction variables were created. The dependent variables consist of positively skewed count data. Negative binomial regression was used to analyze status offenses, due to variable overdispersion, whereas Poisson regression was utilized to predict the total number of offenses. Both model types are given in Table 3 for ease of interpretation (specifically, significance and signs are directly comparable). Predicted counts for significant variables are provided in the discussion that follows.

Table 3.

Regression Models by Offense Type.

	Status offenses		Total offenses
	1	2	3	4
	Direct effects	Interaction	Direct effects	Interaction
Gender (1 = female)	.26 (.17)	−.45 (.57)	−.22 (.05)***	−.51 (.19)**
Race (1 = Black)	.50 (.19)**	.29 (.27)	.31 (.06)***	.13 (.09)
Gender × Race	—	.91 (.41)*	—	.08 (.14)
Psychological symptoms index	.38 (.15)*	.39 (.15)*	.22 (.05)***	.22 (.05)***
School suspensions	.05 (.06)	.09 (.07)	.08 (.02)***	.06 (.02)**
Suspensions × Gender	—	−.20 (.11)	—	.09 (.04)*
Disciplinary problems	.11 (.08)	.10 (.08)	.10 (.02)***	.10 (.02)***
Single-mother-headed household	−.32 (.16)*	−.21 (.31)	.08 (.05)	.01 (.10)
Mother × Gender	—	−.01 (.34)	—	−.33 (.11)**
Mother × Race	—	−.17 (.37)	—	.25 (.12)*
Employment status of any parental guardian	−.11 (.16)	−.33 (.21)	−.11 (.05)*	−.19 (06)**
Employment × Gender	—	.66 (.37)	—	.19 (.13)
Sibling criminality	.05 (.07)	.04 (.07)	.11 (.02)***	.10 (.02)***
Guardian criminality	.17 (.11)	.18 (.11)	.01 (.03)	.01 (.03)
Family dysfunction	.09 (.02)***	.08 (.03)**	.04 (.01)***	.04 (.01)***
Dysfunction × Gender	—	.04 (.05)	—	.01 (.01)
Constant	−.93 (.28)***	−.76 (.35)*	1.19 (.08)***	1.38 (.11)***
N	212	212	212	212
Likelihood ratio χ²	68.67***	78.40***	358.94***	379.80***
Pseudo-R²	.11	.12	.22	.23

Note. The standard errors are in parentheses.

p < .05. **p < .01. ***p < .0011.

In predicting status offenses, evidence for H1b was found in Model 2. Black girls were significantly more likely to be sanctioned for status offenses than Black and White boys (and White girls). The predicted mean number of status offenses for Black girls was 2.02, and for boys it was 0.95. The predicted number for White girls was 0.83, whereas for White boys it was 1.02. Even though girls were disproportionately represented among status offenders in this sample, and despite bivariate analyses showing that gender and status offenses were significantly correlated, the inclusion of other variables dampened the gender effect. The interaction of race and gender was more important than either variable alone, thus H1a, that all girls will be charged with more status offenses than boys, was not supported.

The variable “single-mother-headed household” is significant and negatively related to status offenses. Girls living in mother-headed households appear, contrary to PCT, less likely to be charged with status offenses. The predicted mean number of status offenses for mother-headed households (using estimates found in Model 1) was 1.06 versus 1.39 for all other types of households. H2, predicting higher levels of status offending among girls in such households, was not supported in Models 1 and 2.

Higher scores on the psychological symptoms index were related to being charged with more status offenses as was the level of family dysfunction. Sibling or guardian criminality (H3) did not seem to affect status offenses. However, H4, which stated that family dysfunction should affect girls more than boys, did not receive support. H5, stating that parental employment will affect girls more than boys, also received no support in either model.

Turning now to the total offense Models 3 and 4, a different pattern of relationships emerged. Being female was negatively related to being charged with offenses and being Black was positively related, but only in Model 3. Unlike what is seen in the status offense models, the interaction variable suppressed the effect of race such that gender was the only significant predictor of the commission of other offenses. The psychological symptoms index, school suspensions, and school disciplinary problems were consistently and positively related to total offenses in both models. Interestingly, the interaction between being female and school suspensions was significant and positively related to overall offending.

Single-mother-headed households were significant only when interacted with gender and race. That is, girls were charged with significantly fewer offenses (with a mean of 7.02) than boys (7.61) in such households. On average, Black girls were charged with fewer offenses than Black boys (predicted value of 7.54 vs. 8.57) and White girls were charged with fewer offenses than White boys (predicted value of 5.23 vs. 6.17) in mother-headed households.

Parental employment status was negatively related to total offending with a predicted mean value of 6.73 offenses when there was at least one working parent versus 9.33 when there were no working parents/guardians in the household. Sibling criminality was positively related to overall offending in both Models 3 and 4.

Limitations and Discussion

This test of PCT presented findings from an exceptional group of juveniles and found, contrary to expectations, single-mother-headed households did not produce more delinquent girls than other household types. Unlike most studies of delinquency, it was not solely comprised of boys and was not dependent on self-reports of petty delinquency. Moreover, variables assessing PCT concepts like relational control, typically measured by asking youth to recall the identity of the primary rules enforcers, were replaced by using variables like familial criminality (a proxy for emotional bonding). But this is also a shortcoming or our study. As the sample was derived from court records, we were unable to assess subjective perceptions of important PCT variables like parental control. Arguably, a youth’s own perceptions of control and other criminogenic variables should be included in future research along with many of the variables we introduced here.

Difficulty achieving robust and representative sample sizes of girls in delinquency research is a known problem (Zahn et al., 2010), and our research is no different in this sense. The data were time and place specific and therefore not generalizable in the sense that we do not know how much of our observations were due to regional and cultural factors. Still, our research raises some questions that merit replication.

Importantly, we did not find that mother-headed households produce any more delinquency than other types of households, with exception of the race-interaction variable. Black juveniles in mother-headed households were more likely to be processed for criminal offenses. Black girls were more likely to be charged with status offenses than White girls. While obviously speculative, we suspect that our results are indicative of the continued pervasive influences of intersectional patriarchy and White privilege.

Family dysfunction was the more consistent predictor of delinquency in all of the models. Notably, parental criminality did not exert a significant effect on delinquency in any of the models. Sibling criminality was significant in predicting total offenses likely due to the fact that siblings suffer from the same environmental shortcomings as the juveniles included the sample. Taken holistically, these findings suggest that family dysfunction is not necessarily a consequence of structure or of the caregiver’s criminal influence and should be examined more closely along with the psychological problems of juvenile offenders.

Tests of PCT have largely avoided any discussion of the complex interactions between patriarchal family structures, abuse, and delinquency, which are interlocking and causally related variables. A patriarchal family structure may be more likely to exhibit dysfunctional familial relationships in which power and abuse are more intimately connected. Psychological symptoms were significant in every model, but it was impossible to discern which disorders/problems were generated by family dysfunction and which were more individualistic (see Table 2). Studies have found that girls’ greater propensity to have been previously abused, particularly sexually abused, is linked to status offending (Chen, Tyler, Whitbeck, & Hoyt, 2004; Tyler, Hoyt, Whitbeck, & Cauce, 2001). Our study was unable to properly assess this relationship.

In addition to noting the juvenile’s gender, race, and social location, research should specify how and why different types of offenses are charged and processed. Interestingly, in analyzing charge types, we found that the most common charge (n = 340) was a petition to revoke probation, yet no probation revocations were filed during the same time frame. We suspect that the petition to revoke probation is therefore used in a coercive fashion, capitalizing on the court’s power to modify youths’ behavior. Wherein surveillance responsibilities have transferred to other arenas, such as juvenile court and school settings, we see these as surrogate agents of patriarchy in terms of provision of social control.

To combat these disempowering forces, researchers and practitioners must be prepared to address these systemic processes directly. A PCT lens can assist in highlighting these processes, but researchers should be cognizant of the fact that the modern family structure is dynamic, which lessens its predictive value to delinquency, especially if examined in isolation from other, known correlates that also operate as patriarchal controls.

Footnotes

Acknowledgements

The authors wish to thank Dr. James Stefurak, the University of South Alabama Mobile Juvenile Court Collaborative students, and Dr. Jennifer Langhinrichsen-Rohling for their support of this study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The data for this project were funded through a grant from the Office of Juvenile Justice and Delinquency Prevention. Mentoring experiences and domestic violence prevention in adjudicated youth, Grant No. 233051-421910-4200, August 1, 2009 to July 31, 2011.

Author Biographies

Corina Schulze, PhD, is an assistant professor in the Department of Political Science & Criminal Justice at the University of South Alabama.

Valerie Bryan, MSW, PhD, is an associate professor of Social Work in the Department of Sociology, Anthropology, and Social Work at the University of South Alabama.

References

Adler

(1975). Sisters in crime. New York, NY: McGraw-Hill.

Althoff

Karpati

Hero

Matte

(2009). Secular changes in mortality disparities in New York City: A reexamination. Journal of Urban Health: Bulletin of New York Academy Journal of Medicine, 86, 729-744.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Bandura

(1977). Social learning theory. New York, NY: General Learning Press.

Barondess

J. A.

(2008). Health through the urban lens. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 85, 787-801.

Bernstein

I. N.

Cardascia

Rose

C. E.

(1979). Defendants, sex, and criminal court decisions. In Alvarez

Lutterman

K. G.

(Eds.), Discrimination in organizations (pp. 320-354). San Francisco, CA: Jossey-Bass.

Biblarz

Raftery

(1999). Family structure, educational attainment. and socioeconomic success: Rethinking the “pathology of matriarchy.” American Journal of Sociology, 105, 321-365.

Blackwell

S. S.

Reed

(2003). Power-control as a between- and within-family model: Reconsidering the unit of analysis. Journal of Youth and Adolescence, 32, 385-401.

Cernkovich

Giordano

(1987). Family relationships and delinquency. Criminology, 25, 295-319.

10.

Chen

K. A.

Tyler

L. B.

Whitbeck

D. R.

Hoyt

(2004). Early sex abuse, street adversity, and drug use among female homeless and runaway adolescents in the Midwest. Journal of Drug Issues, 34, 1-21.

11.

Chesney-Lind

(1989). Girls’ crime and woman’s place: Toward a feminist model of female delinquency. Crime & Delinquency, 35, 5-29.

12.

Chesney-Lind

Eliason

(2006). From invisible to incorrigible: The demonization of marginalized women and girls. Crime, Media, and Culture, 2, 29-48.

13.

Chesney-Lind

Shelden

R. G.

(2004). Girls, Delinquency, and Juvenile Justice (3rd ed.). Belmont, CA: Wadsworth Publishing.

14.

Feld

B. C.

(2009). Violent girls or relabeled status offenders? An alternative interpretation of the data. Crime & Delinquency, 55, 241-265.

15.

Gimenez

(2001). Marxism and class, gender, and race: Rethinking the trilogy. Race, Gender, & Class, 8(2), 23-33.

16.

Grasmick

H. G.

Blackwell

B. S.

Bursik

R. J.

(1993). Changes in the sex patterning of perceived threats of sanctions. Law & Society Review, 26, 679-705.

17.

Grasmick

H. G.

Hagan

Blackwell

B. S.

Arneklev

B. J.

(1996). Risk preferences and patriarchy: Extending power-control theory. Social Forces, 75, 177-199.

18.

Hagan

(1989). Structural criminology. New Brunswick, NJ: Rutgers University Press.

19.

Hagan

Boehnke

Merkens

(2004). Gender differences in capitalization processes and the delinquency of siblings in Toronto and Berlin. British Journal of Criminology, 44, 659-676.

20.

Hagan

Gillis

A. R.

Simpson

(1979). The sexual stratification of social control: A gender based perspective on crime and delinquency. British Journal of Sociology, 30, 25-38.

21.

Hagan

Gillis

A. R.

Simpson

(1985). The class structure of gender and delinquency: Toward a power-control theory of common delinquent behavior. American Journal of Sociology, 90, 1151-1178.

22.

Hagan

Gillis

A. R.

Simpson

(1990). Clarifying and extending power-control theory. American Journal of Sociology, 95, 1024-1037.

23.

Hagan

Simpson

Gillis

A. R.

(1987). Class in the household: A power-control theory of gender and delinquency. American Journal of Sociology, 92, 788-816.

24.

Hart

C. O.

Mueller

C. E.

(2013). School delinquency and social bond factors: Exploring gendered differences among a national sample of 10th graders. Psychology in the Schools, 50(2), 116-133.

25.

Hartney

Silva

(2007). And justice for some: Differential treatment of youth of color in the justice system. Oakland, CA: National Council on Crime and Delinquency.

26.

Haynie

McHugh

(2003). Sibling deviance: In the shadow of mutual and unique friendship effects? Criminology, 41, 355-391.

27.

Henriques

Z. W.

(1982). Imprisoned mothers and their children: A descriptive and analytical study. Washington, DC: University Press of America.

28.

Hines

A. M.

(1997). Divorce-related transitions, adolescent development, and the role of the parent-child relationship: A review of the literature. Journal of Marriage and Family, 59, 375-388.

29.

Hirschi

(1969). The causes of delinquency. Berkeley: University of California Press.

30.

Joe

K. A.

(1995). The dynamics of running away, deinstitutionalization policies and the police. Juvenile & Family Court Journal, 46(3), 43-55.

31.

Johnston

(1995). Effects of parental incarceration. In Gabel

Johnston

(Eds.), Children of incarcerated parents (pp. 59-88). New York, NY: Lexington Books.

32.

Jonson-Reid

Drake

Zhou

(2013). Neglect subtypes, race, and poverty: Individual, family, and service characteristics. Child Maltreatment, 18, 30-41.

33.

Leiber

M. J.

Wacker

M. E. E.

(1997). A theoretical and empirical assessment of power-control theory and single-mother families. Youth & Society, 28, 317-350.

34.

Mack

K. Y.

Leiber

M. J.

(2005). Race, gender, single-mother households, and delinquency: A further test of power-control theory. Youth & Society, 37, 115-144.

35.

Manning

W. D.

Smock

P. J.

(1997). Children’s living arrangements in unmarried-mother families. Journal of Family Issues, 18, 526-544.

36.

Matsueda

Heimer

(1987). Race, family structure, and delinquency: A test of differential association and social control theories. American Sociological Review, 2, 826-840.

37.

McCarthy

Hagan

Woodward

T. S.

(1999). In the company of women: Structure and agency in a revised power-control theory of gender and delinquency. Criminology, 37, 761-788.

38.

Morash

Chesney-Lind

(1991). A reformulation and partial test of the power-control theory of delinquency. Justice Quarterly, 8, 347-377.

39.

Murry

V. M.

Bynum

M. S.

Brody

G. H.

Willert

Stephens

(2001). African American single-mothers and children in context: A review of risk and resilience. Clinical Child and Family Psychology Review, 4, 133-155.

40.

Nichols

T. R.

Graber

J. A.

Brooks-Gunn

Botvin

G. J.

(2006). Sex differences in overt aggression and delinquency among urban minority middle school students. Journal of Applied Developmental Psychology, 27, 78-91.

41.

Parker

Reckdenwald

(2008). Women and crime in context: Examining the linkages between patriarchy and female offending across space. Feminist Criminology, 3, 5-24.

42.

Puzzanchera

Adams

Sickmund

(2010). Juvenile Court Statistics 2006–2007. Pittsburgh, PA: National Center for Juvenile Justice.

43.

Rebellon

(2002). Reconsidering the broken homes/delinquency relationship and exploring its mediating mechanisms. Criminology, 40, 103-136.

44.

Rowe

D. C.

Farrington

D. P.

(1997). The familial transmission of criminal convictions. Criminology, 35, 177-201.

45.

Rowe

D. C.

Gulley

(1992). Sibling effects on substance use and delinquency. Criminology, 30, 217-233.

46.

Smart

(1977). Women, crime and criminology: A feminist critique. London, England: Routledge.

47.

Smith-Adcock

Lee

S. M.

Kerpelman

Majuta

Young

C. B.

(2013). Reputation enhancement and school delinquency: A prospective study using the National Educational Longitudinal Survey [NELS: 88]. Youth & Society, 45(2), 201-220.

48.

Snyder

H. N.

Sickmund

(2006). Juvenile offenders and victims: 2006 National Report. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

49.

Steffensmeier

Schwartz

Zhong

Ackerman

(2005). An assessment of recent trends in girls’ violence using diverse longitudinal sources: Is the gender gap closing? Criminology, 43, 355-406.

50.

Stevens

Morash

Chesney-Lind

(2011). Are girls getting tougher, or are we tougher on girls? Probability of arrest and juvenile court oversight in 1980 and 2000. Justice Quarterly, 28, 719-744.

51.

Trupin

Stewart

Beach

Boesky

(2002). Effectiveness of a dialectical behavior therapy for incarcerated female juvenile offenders. Child and Adolescent Mental Health, 7, 121-127.

52.

Tyler

K. A.

Hoyt

D. R.

Whitbeck

L. B.

Cauce

A. M.

(2001). The impact of childhood sexual abuse on later sexual victimization among runaway youth. Journal of Research on Adolescence, 11, 151-176.

53.

Uggen

(2000). Class, gender, and arrest: An intergenerational analysis of workplace power and control. Criminology, 38, 835-862.

54.

Van

Rakt

Nieuwbeerta

Apel

(2009). Association of criminal convictions between family members: Effects of siblings, fathers and mothers. Criminal Behaviour and Mental Health, 19, 94-108.

55.

Waldfogel

Craigie

T. A.

Brooks-Gunn

(2010). Fragile families and child well-being. Future Child, 20, 87-112.

56.

Welsh

W. N.

Greene

J. R.

Jenkins

P. H.

(1999). School disorder: The influence of individual, institutional, and community factors. Criminology, 37(1), 73-116.

57.

Zahn

M. A.

Agnew

Fishbein

Miller

Winn

D. M.

Dakoff

. . . Chesney-Lind

(2010). Causes and correlates of girls’ delinquency (OJJDP Girls Study Group Series). Washington, DC: U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.