Trends in girls' delinquency and the gender gap: Statistical assessment of diverse sources

Darrell Steffensmeier, Jennifer Schwartz

Research output: Chapter in Book/Report/Conference proceedingChapter

24 Scopus citations


One of the most consistent and robust findings in criminology is that, for nearly every category of crime, females commit much less crime and delinquency than males. The gender gap in offending is particularly notable for more serious and violent offenses. In recent years, however, the extent and character of this gender difference in crime is increasingly being called into question by statistics and media reports suggesting a greater involvement of women-and particularly girls-in the criminal justice system. During the past couple of decades, girls' delinquency as reported in offi cial sources of data has undergone substantial changes relative to boys' delinquency. Uniform Crime Report (UCR) statistics showing marked increases in girls' arrests for aggravated and, especially, simple assault have encouraged the growing perception in the media and among some criminologists that girls' violence is on the rise and that the gender gap in violent offending is closing. For example, between 1980 and 2003 in the United States, girls' arrests nationwide increased 42.7 percent, while arrests of boys actually decreased by 10.2 percent (Federal Bureau of Investigation [FBI] 2004). Within this same time period, arrests of girls for index violent offenses-homicide, rape, robbery, and aggravated assault-increased by 75.2 percent and arrests of girls for "simple assaults" increased by 318.5 percent, whereas boys' arrests declined 11.3 percent for index violent offenses but increased 130.5 percent for simple assault. These arrest trends, along with high-profi le cases of female delin quency, are the main source for media headlines such as "Girls getting increasingly violent," "Girls catching up with boys in delinquency and crime," and "Girls not all sugar and spice." Although little systematic theorizing has been done, commentators have proposed a variety of reasons for what they see as real increases in girls' violence as inferred from the arrest trends. The combination of increased stress and greater female independence is believed to have increased girls' opportunities and motivations for violent crime. However, because arrest counts are a product of both delinquent behavior and responses to it, researchers and policy makers face a dilemma about how to interpret the arrest statistics. Do the arrest gains indicate real changes in underlying behaviors of girls-the Behavior Change Hypothesis- or are the gains artifactual, a product of recent changes in public sentiment and enforcement policies for dealing with youth crime and violence that have elevated the visibility and reporting of girls' " delinquencies" and "violence"-the Policy Change Hypothesis. The statistical information about longitudinal trends in girls' and boys' delinquency and violence comes from three main sources: offi cial, self-report, and victimization surveys (in which the victim identifi es the sex and age of the offender). Each source of data has its strengths and weaknesses, and each offers at least a slightly different picture of crime. Offi cial data on delinquency are collected by local government agencies and disseminated by state and national organizations. Arrest data, collected from local police agencies and disseminated by the FBI, are one of the main sources of offi cial data. In contrast, self-report and victimization data are collected independently of the criminal justice system. Monitoring the Future (MTF), the National Crime Victimization Survey (NCVS), and the newer National Youth Risk Behavior Survey (NYRBS) are highly regarded nationally representative, longitudinal surveys.1 Because these sources of data differ in how they measure delinquency and violence, they are particularly useful for evaluating whether trends in girls' delinquency are a product of changes in the underlying violent and delinquent behavior of girls or changes in juvenile justice policies that have enhanced the visibility and reporting of girls' delinquency. Our confi dence that girls' violence and delinquency have changed is enhanced if all these sources agree on the nature of the trends, despite measurement differences and dissimilar sources of limitations, while that confi dence is diminished if the sources disagree. The Behavior Change Hypothesis contends that arrest gains indicate real changes in underlying behaviors of girls. Girls' lives and experiences have changed dramatically, perhaps in ways that increase their propensities or opportunities to commit violent crimes. Girls today face greater struggles in maintaining a sense of self and confronting a complex, often contradictory, set of behavioral scripts that specify what is appropriate, acceptable, or possible for girls to do. For example, greater exposure to media messages portraying or condoning girls as violent, such as those in movies like Charlie's Angels and Kill Bill and video games like Tomb Raider, might facilitate changing gender-role expectations toward greater female freedom, assertiveness, and male-like machismo and competitiveness. As it becomes more socially acceptable, girls may increasingly turn to violence as a coping strategy or means of solving interpersonal confl icts with school offi cials, parents, or other authority fi gures and peers. The latter may involve arguments with boys in dating contexts but also fi ghts with other girls over ownership of males and defense of one's sexual reputation (Anderson 1999; Campbell 1993; Ness 2004). In addition to gender-role strain, economic and familial strains may have intensifi ed for girls. Finally catching up to girls may be breakdowns in family, community, and other institutions that once buffered girls against victimization and involvement in violence. Female role models are fewer as mothers are forced back into the job market or are rendered ineffective because of increased substance abuse, particularly in depressed urban areas (Almgren et al. 1998; Brown and Gilligan 1992; Hall 2004). This combination of heightened role strain, changing normative expectations, and increased economic and familial stresses could contribute to girls' greater involvement in physical aggression and violence. The Policy Change Hypothesis contends that gains are an artifactual product of recent changes in public sentiment and enforcement policies for dealing with youth crime and violence that have elevated the visibility and reporting of girls' "delinquencies" and "violence." At least four interrelated policy shifts appear to have escalated the arrest proneness of girls today relative to girls in prior decades and relative to boys (listed in order of their overall importance). For a fuller treatment of these policy shifts, see Steffensmeier and others (2005). • Netwidening, or the criminalization of less serious forms of "violence," will escalate female arrests since their violent offending is less serious and less chronic. Criminalization includes but goes beyond "zero-tolerance" policies to encompass, quite broadly, the (a) targeting of minor or trivial forms of lawbreaking and (b) charging up minor types of lawbreaking into offense classifi cations representing greater seriousness and harsher statutory penalties. Recent enforcement practices have lowered the tolerance for low-level crime and misdemeanors, a policy shift that will disproportionately produce more arrests of less serious offenders. Analysts of crime trends point out that this netwidening has been particularly robust in ambiguous offense categories like simple or aggravated assault, where it is more the practice today that (a) disorderly conduct, harassment, endangering, resisting arrest, and so forth, will be categorized as simple assaults and (b) former simple assaults will be "charged up" to aggravated assault (Blumstein and Wallman 2000; Garland 2001; Steffensmeier and Harer 1999; Zimring 1998). These more expansive defi nitions of what constitutes "violence" or an "assault" have led to enhanced sanctioning for aggressive conduct among youths overall (Blumstein and Wallman 2000; Fuentes 1998) but even more so among girls who tend to commit the milder, less serious forms of physical attacks or threats (Chesney-Lind 2002; Steffensmeier 1993; Steffensmeier and Schwartz 2004). What now tends to be dealt with formally was formerly often ignored or dealt with informally. • The criminalization of violence occurring between intimates and in private settings such as at home or school will portray levels of female violence that more closely approximate levels of male violence because girls' violence is more likely to take place in this context than in public settings against strangers. Several recent studies of girls' violence document the trend toward treating domestic and school violence as a criminal matter and establish the impact of these policy shifts on girls' assault arrest trends. A review of girls' "person-to-person" cases referred to Maryland's juvenile justice system revealed the majority to be family-centered "assaults" that involved such activities as a girl hitting or throwing an object at her mother, who subsequently pressed charges (Mayer 1994; as cited in Chesney-Lind 2002). Another study of nearly one thousand girls' fi les from four California counties concludes that "most of these [assault] charges were the result of nonserious, mutual combat situations with parents" (Acoca 1999, pp. 7-8; see also Schaffner 1999). Case descriptions of girls arrested ranged from "father lunged at daughter while she was calling the police about a domestic dispute [when] daughter hit him" (self-defense) to "throwing cookies at her mother" (trivial argument).

Original languageEnglish (US)
Title of host publicationThe Delinquent Girl
PublisherTemple University Press
Number of pages34
ISBN (Print)9781592139514
StatePublished - 2009

All Science Journal Classification (ASJC) codes

  • Social Sciences(all)


Dive into the research topics of 'Trends in girls' delinquency and the gender gap: Statistical assessment of diverse sources'. Together they form a unique fingerprint.

Cite this