Specialization in Juvenile Offending Research Paper

This sample Specialization in Juvenile Offending Research Paper is published for educational and informational purposes only. If you need help writing your assignment, please use our research paper writing service and buy a paper on any topic at affordable price. Also check our tips on how to write a research paper, see the lists of criminal justice research paper topics, and browse research paper examples.

Whether and to what extent juvenile offenders specialize in committing certain types of crimes during their delinquent careers has substantial implications for theory and policy. The question of offender specialization arguably first emerged with the advent of the positivist school in criminology in the late nineteenth century when Caesare Lombroso sought to identify criminal types. With the introduction of specialized juvenile justice in laws in the early twentieth century and their fundamental reforms in many countries in the second half of the twentieth century and into the initial decade of the twenty-first century, there have been considerable theorizing, research, and debate about specialization in youthful offending. Political debates in countries such as Australia, Canada, England-Wales, and the United States concerning the reform of youth justice laws have been influenced, in part, by the considerable research on specialization youthful offending. The main theoretical debate has been focused on the attempts to specify potential distinguishing characteristics of certain types of violent juvenile delinquents such as serious sexual offenders, murderers, and gang members, compared to less serious violent offenders, property offenders, and drug users/traffickers, for example. Many diverse labels have been applied to young offenders, such as serial fire-setter/rapist/murderer, career criminal, life-course-persistent/adolescence-limited offender, chronic/ nonchronic offender, and adolescent psychopath/conduct disordered, that typically include attempts to specify types and patterns of crime. Oftentimes these labels or categories have been introduced into the media, political, and legislative circles, and debates emerge about whether to incarcerate, punish, and/or rehabilitate juvenile delinquents depending on what “type” of offender they are.

The essence of the theoretical debate is whether juvenile offenders can be classified into an “offender type” based on the qualitative nature, and/or quantity of their offending, versus whether there is more simply a single type of juvenile offender who engages in diverse and multiple antisocial and criminal behaviors. Not surprisingly, a considerable empirical research theme over the past half a century has focused on the nature and extent to which juvenile offenders demonstrate specialization in offending. The related theoretical debate has centered on considering that if specialization is not evident, then a single or general theory of juvenile delinquency could be sufficient to explain this phenomenon. In turn, criminal justice responses, therefore, could be directed toward a central set of causes or risk/protective factors to prevent juvenile delinquency from beginning as well as reoccurring. In contrast, if research demonstrates specialization or a diversity of types of juvenile delinquent offending patterns, then typologies require far more complex and unique theoretical explanations along with specialized juvenile justice responses, including treatment and intervention strategies, depending on the type of juvenile delinquent or young offender. To better understand how this debate on specialization has evolved over the years, it is necessary to describe the key theoretical controversies historically.

Key Issues And Controversies: Theoretical Debates

Typologies of adult and youth offenders gained popularity in the 1960s with the assertion that offenders could be classified into conceptually valid types based on fundamental differences in the crimes they committed. The most basic and obvious crime-centered typologies generally consisted of nominal categories of specialized criminal behavior consisting of property versus violent offender types. More types emerged as well including white-collar criminals, organized crime, street gangs, and political terrorists, for example. It was then theorized that these criminal types were explained by different sets of causal factors.

By the 1980s, there was a wide range of theories to explain ever-increasing types of offenders. In turn, typologies and related theories generated intense debates concerning basic definitions of deviance, delinquency, and criminality that are essential to assessing youthful offending specialization. For example, the assertion that delinquency was overwhelmingly associated with youth from low-income groups along with those belonging to certain ethnic/racial minorities was challenged by self-report studies of delinquency (e.g., Elliott and Ageton 1980). These studies demonstrated that minor delinquency was common for youth from all income and ethnic groups. A second criticism of initial crime-centered typologies was the reliance on offense records whereby the presence of a particular type of offense (e.g., sexual assault) determined a youth’s “specialization” in offending. Critically, such a snapshot-based examination of offending patterns did not consider variations in offending over time. The longitudinal cohort studies of the 1970s and subsequent major cross-sectional studies provided a more valid basis for establishing typologies and specialization in youthful offending because they took into consideration changes in the frequency of offense types over the entire course of adolescence. Finally, a third key criticism involved arbitrary and income/ethnic group biases of juvenile justice laws regarding charges and convictions up to the 1970s. For example, juvenile justice laws from the early 1900s to the early 1970s virtually criminalized many childhood and adolescent antisocial and deviant behaviors. Typologies ranged from the enormous diversity of trivial antisocial behaviors such as status offenses (e.g., swearing and drinking alcohol) to the most violent crimes (e.g., rape and murder). However, because juvenile justice laws governed the ages between 8 and 21 depending on the country and within-country jurisdiction (e.g., state, province, region), it was abundantly clear that attempts to assess the nature and extent of specialization in juvenile offending had to be based on more sophisticated approaches.

What emerged from both longitudinal cohort and major cross-sectional research was that simple typologies based on the presence of certain offenses in a juvenile’s history were inadequate for assessing specialization for several reasons. For example, status, property and violent crimes committed by juveniles varied substantially in not only degree of seriousness but frequency. To assess specialization, theoretically driven and complex equation-based indices often consisting of ratios of different types of crime such as property compared to violent crimes became central to investigating specialization in youthful offending. Seriousness of types of crimes, for example, based on the extent or amount of damage in dollars involved with property crimes, and the nature and extent of injuries for violent crimes also were incorporated into indices determining offender types. Another dimension was the variation in the extent of switching to different types of crime in certain developmental periods, such as childhood, adolescence, and adulthood. In other words, early crime-centered typologies were unable to address rapidly emerging evidence that delinquent careers were characterized by significant heterogeneity involving different patterns in the type, seriousness, and frequency of offenses committed over childhood and adolescence.

Several of the key concepts from the criminal career paradigm, particularly the concept of the career criminal, evoked intense theoretical and empirical debate. Klein (Klein and Malcolm 1979; Klein 1984) maintained that an abundance of empirical evidence demonstrated that juvenile offending was predominately unpatterned and reflected the notion of “cafeteria-style” offending. Subsequently, Gottfredson and Hirschi (1990) introduced their General Theory of Crime based on Hirschi’s (1969) original research and his Bonding Theory of social control. This theory asserts that specialization does not exist and instead delinquents engage in a wide range of delinquent behaviors, stemming broken bonds to family and other societal institutions such as the church, school, and community. The absence of strong prosocial bonds further explains the time-stable personality trait of low self-control, which in turn is caused by lax and/or inconsistent parenting practices. Low self-control is established in childhood and remains stable throughout life. It is manifested by the pursuit of easy and instant gratifications inherent in deviant behaviors such as smoking, excessive drinking, short-term and multiple sexual conquests, recklessness, and crime, when opportunities are present. In effect, the General Theory of Crime predicts that juvenile offending is best understood as involving substantial offense versatility.

By the 1990s the developmental theoretical perspective of crime introduced more complex predictions about patterns in juvenile offending such as specialization and versatility. This perspective identified risk and protective factors for several forms of antisocial behavior throughout the life course beginning with early childhood and into early adulthood. Loeber (1990) asserted that children who exhibited recklessness, problems with authority, aggression, and coercive/ manipulative behaviors were more likely to follow a more violent pattern in adolescence but also engage in diverse crimes. In contrast, children who were more withdrawn and antisocial were more likely to engage in covert patterns and drug use as adolescents.

The most prominent developmental theory relevant to specialization involved Moffitt’s (1993) developmental taxonomy. She predicted that more sustained aggressive and violent life-course patterns were based on the age of onset of offending. Moffitt proposed two distinct types of offenders in the population: life-course-persistent’ (LCP) offenders and “adolescence-limited” (AL) offenders. LCP offenders are characterized by the interaction, and accumulation, of individual deficits and environmental adversities that begin early in childhood and cascade over subsequent developmental periods. In effect, these individuals begin offending early in childhood/ adolescence, engage in a wide range of deviant and criminal acts including bullying and other violent behavior, and persist into adulthood. In adolescence, therefore, this developmental group was predicted to exhibit frequent, versatile, and serious offending including violence. In contrast, AL offenders are predominately characterized by risk factors that emerge in adolescence. The primary risks involve status frustration driven by the expanding maturity gap between biological and social maturities (i.e., the observation that while adolescents experience physical or biological maturity at earlier ages when compared to the early twentieth century, their social maturity as evidenced by the age they assume adult social roles such as husband, breadwinner, father is delayed), as well as exposure to deviant or delinquent peers and role models. AL offending was predicted to be more specialized than LCP offending, in the context of offenses reflecting the effect of the maturity gap such as vandalism, public order, substance abuse, and status offenses.

As with all classic debates in criminology, measurement of key constructs are often central to resolving competing hypotheses, in this instance, about specialization versus versatility in youthful offending.

The Measurement Of Specialization In Youthful Offending

Historically, methodological and measurement innovations have had to take into consideration the inherent limitations associated with data sources and especially official juvenile justice information recording. Most importantly, there is a consensus that official data significantly underrepresents the total amount of crimes committed, and even more for official juvenile offending data. During the past 30 years, measures based on informal and formal diversion programs have been introduced in many Western countries. Often, informal diversions from formal justice proceedings have not been recorded. In addition, there has been concern over the variations in the offense recording and charging process across jurisdictions even within countries. When only the most serious offense is proceeded with in terms of offense recording, trial, and/or sentencing, both the range and seriousness of actual offenses are under-recorded. This affects the accuracy of officially recorded crime records. Another limitation has been the concern with variations in the number of police charges that are recorded when police practices include deliberately over charging in order to enhance their ability to gain information about other crimes and to an advantage in plea bargaining to gain a confession of guilt. As well, police and prosecutors utilize their ability to increase or decrease the seriousness of the charges since criminal codes include wide discretion concerning the degree within a criminal offense category. For example, a violent assault resulting in death ranges from manslaughter to first-degree murder. All of these concerns raise questions over the validity of measures when using, for example, charges versus convictions in assessing specialization, especially across different juvenile justice jurisdictions included in comparative research designs. Similar validity concerns have been a source of debate with self-report methodologies utilized to assess specialization. Most importantly and especially when youth are involved, memory recall is often problematic in terms of overstating offenses, and the timelines in which they occurred in their past. Nonetheless, despite these limitations, most research on specialization in youthful offending has used a combination of methodologies to identify typologies based on a variety of specific measurement criteria. One advantage with official data is that a more complete delineation of potentially different offending trajectories over time is possible because this data can record patterns across childhood, adolescence, and adulthood. Similarly, complex cohort designs typically include multiple data recording approaches (e.g., including both official and self-report data) to measure the types, frequency, and temporal patterns of offending.

The seminal Philadelphia Birth Cohort Study (Wolfgang et al. 1972) was the most widely recognized initial methodological innovation in juvenile offender specialization research. Wolfgang and colleagues examined crime type patterns from offense to offense as a series of transitions using transition matrices. This method assessed the probability of juvenile offenders having received charges for similar crimes across adjacent offending events. Transition matrices identified whether specialization in certain crime types was evident by evaluating whether the probability of being charged with the same offense consecutively was greater in value than the probability of being charged with a different offense. Importantly, these techniques were also adopted to assess specialization using major cross-sectional research that included the complete court histories of charges and convictions for juvenile offenders (e.g., Farrington et al. 1988).

Related to the transition matrix methodology was the development of statistical tests in the 1980s to measure whether the probability of observed transitions to similar offense types was greater than could be expected on the basis of chance alone. Coefficients representing the level of specialization for specific types of offending in a sample became a standard measure. For example, Bursik (1980) introduced the Adjusted Standardized Residual (ASR) that determined whether the ratio between observed and expected transition values was statistically significant. Farrington et al. (1988) further introduced the Forward Specialization Coefficient (FSC) to control for the influence of infrequent cases that would distort the basic observed to expected ratio. The FSC represents a single value summary of specialization for different offense types based on the sequence of similar offenses in transition matrices. Fisher and Ross (2006) more recently asserted that the FSC is the most widely utilized measure of specialization to date because of its theoretical sophistication and predictive validity.

Despite widespread use, the transition matrix approach generally, and the FSC specifically, has several limitations (e.g., see Britt 1996; Fisher and Ross 2006; LeBlanc and Frechette 1989; Nieuwbeerta et al. 2011; Sullivan et al. 2009). Most importantly, the FSC describes specialization aggregately for a group of delinquents. This aggregate measure is theoretically appropriate for the explanation of the shifting from different types of crime within groups. However, it does not measure shifts in crime for specific individuals. In effect, the aggregate nature of the FSC does not produce information about the individual extent of specialization.

Within-individual change was central to the concept of criminal careers in the 1980s (Blumstein et al. 1986) and the individual-level examination of specialization in juvenile offenders. From this perspective, the measurement focus is on the proportions of similar offense types in individual criminal histories. Here, specialization was defined by the absence of diversity of different types of crime in an individual’s offense history. Regarding a specific measurement, for example, Wikstrom (1987) calculated the total proportion of crimes belonging to a single category of offenses in the histories of offenders, which allowed for the utilization of different percentage thresholds for determining specialization. Wikstrom (1987) asserted that if, regardless of the type of offense, two-thirds or more were of one type, then the offender’s criminal offending history is considered specialized. However, this measurement methodology is not applicable to occasional or limited offending. In a Swedish sample of delinquents, Wikstrom (1987) identified approximately one-third of juvenile offenders as specialists. Theoretically, different thresholds can be asserted as establishing specialization. In addition, this threshold measurement methodology raises validity concerns depending on the number of types of crimes available for analysis. For example, certain studies can and have calculated a threshold based on three categories of crime, such as property, violent, and status offenses. Others have used even more specific crime categories, making cross-study comparisons problematic because the two-thirds threshold used for three crime types does not appear an appropriate threshold for a larger number of categories. In short, determining specialization based on percentage threshold approaches is somewhat arbitrary and ultimately does not easily capture the nature of offense specialization.

More recently, the extent of specialization in an offender’s history has been calculated using the diversity index which also focuses on the proportion of different offenses committed by individuals but also provides a measure of the probability that any two observed offenses drawn randomly from a criminal history belong to dissimilar offense types. In effect, by calculating the proportions of offenses across a specified number of crime categories, this approach provides an individual-level measure of specialization based on the complete offense patterns, rather than the sequence of offenses. This measure calculates the overall amount of similar/dissimilar offenses and allows for the comparison of specialization patterns for both individuals and groups. Again, however, when studies used different numbers of crime categories to calculate the amount of diversity, cross-study comparisons become problematic. In addition, some scholars have asserted that the diversity index does not provide precise information about the type of offense specialization, such as the frequency of different offense types (Sullivan et al. 2009).

Several recent studies have applied new and innovative approaches to measure specialization in juvenile offending (and more generally). These approaches typically involve complex multivariate techniques that have the ability to simultaneously capture the nature and extent of specialization, something that past techniques typically accomplished independently (i.e., the FSC compared to the diversity index). One such approach is Latent Class Analysis (LCA). While not designed specifically for the study of offense specialization, LCA is useful for identifying homogeneous subgroups of offenders that cluster around specific types of offenses within a sample. LCA has been used in the classification of offenders’ criminal career trajectories that are based on the varying frequency of total offending over time (Francis et al. 2004). In effect, this technique has the potential to distinguish groups according to both the frequency and type of offenses committed.

Research Findings

The original empirical studies using transition matrices were characterized by mixed findings because of differences in sample composition, data sources, and several methodologies. Nonetheless, most of these studies identified considerable versatility in juvenile offending behavior, which therefore, provided initial support for the General Theory of Crime and the prediction of “cafeteria-style” offending patterns of juvenile offenders (e.g., Klein and Malcolm 1979; Klein 1984). However, several of the initial studies indicated some support for specialization for certain types of minor offending, such as theft (Wolfgang et al. 1972). Subsequent studies in the 1980s found evidence of specialization, again, for less serious crimes such as property and status offenses. For example, Farrington et al. (1988) demonstrated that the most specialized offenses involved status offenses (e.g., running away, liquor, incorrigibility, curfew, and truancy) and property offenses (e.g., burglary, vehicle theft) as well as drug use. As well, there was some indication of a modest tendency toward crime switching to more serious forms of offending, particularly from status to property and property to violent offenses (e.g., LeBlanc and Frechette 1989; Smith et al. 1984). The same pattern of mixed support for specialization was evident in other studies. For example, females were more likely to specialize in status offenses compared to males, who were comparatively more likely to specialize in serious offending (e.g., Kempf 1986, Farrington et al. 1988). Yet still, other studies did not report such differences. Even regarding ethnicity/racial groups, versatility was more prevalent than specialization despite some studies that indicated black youth were slightly more likely to specialize in more serious property offenses than white youth (e.g., Lattimore et al. 1994). Again, several studies did not report this pattern.

By the late 1990s, studies based on developmental theories’ predictions concerning specialization in juvenile offending revealed even more complex findings. Mazerolle et al. (2000) examined Moffitt’s typology and found AL offenders were more likely to specialize in nonserious offenses compared to LCP offenders. The latter also were more violent (see also Piquero et al. 1999). Subsequent developmental theory-based studies expanded into the adulthood stages revealing that offending trajectories over the life course confirmed the adolescence period had the least amount of specialization with specialization typically emerging later in adulthood. Most importantly, both onset and desistance occurred more often in the adulthood stages than originally predicted. In effect, as offenders aged into adulthood, their offending became more patterned but less frequent.

In sum, three main empirical themes regarding specialization in youthful offending were evident. First, adolescence generally involved the least amount of offense specialization compared to later periods. Second, a small but significant amount of specialization occurred during adolescence. This variation however, typically consisted of the less serious end of the offending spectrum, such as status offenses and theft. However, while this small but significant amount of specialization occurred, versatility was far more prevalent. Third, a small minority of juvenile offenders was responsible for much of the versatile offending committed in adolescence and, most importantly, for more serious violent offending.

Juvenile Justice Policy Implications Related To The Specialization Debate

The original juvenile justice laws reflected early crime-specific theories and crime-centered typologies that presupposed specialization in juvenile delinquency and, in particular, status offenses. Given the apparent extent of nonserious and more specialized offending, welfare/rehabilitation models of juvenile justice were dominant and focused on the family-based causes of delinquency through rehabilitation interventions in juvenile courts. In effect, status offenses were a key target for policy intervention with youth, and in many Western jurisdictions, formal status offender programs proliferated and were based on the hypothesis that youth status offenders would escalate to more serious forms of offending if formal intervention and treatment was not pursued. In other words, one rationale for these programs was that early and formal intervention/treatment was warranted if “status offenders” represented a distinct group of youth who specialized in nonserious misbehavior to prevent them from becoming involved in delinquency and more serious offending. Proponents of this view, therefore, subscribed to the belief that youthful offending was (a) characterized by some degree of specialization and (b) escalated to more serious forms of delinquency in the absence of intervention.

By the mid-1970s, however, the welfare model-based juvenile justice laws and assumptions about specialization in youthful offending were challenged theoretically, empirically, and politically. One reform theme was that diversion of youth outside of the formal juvenile justice system was necessary to avoid the hypothesized delinquent effects of labeling. In other words, the common view was that minor versatile offending such as status offenses and vandalism escalated into more diverse and serious offending types when juvenile justice systems labeled minor offending youth as delinquents. In effect, specialization was conceptualized more in qualitative terms such as status offenders versus serious criminal, property, and violent offenders. However, beyond the theoretical debates, research on specialization in juvenile offending in the 1970s and 1980s did not support the assertion that, in the absence of intervention, the majority of youth who specialized in status offenses escalated to more serious offending types.

The reform debates in several jurisdictions, especially Canada and the United States, focused on specialization themes involving the identification of certain types of offender categories including career criminals, chronic offenders, serious and violent offenders, and sexual offenders. In effect, the political and research debates did not just center on more contemporary themes and measures of specialization but rather on more broad conceptualizations focused on distinguishing types of minor offenders, types of property offenders, types of violent offenders, and combinations of offender types. Youth justice laws were and are still based on this conceptualization of juvenile delinquent or young offender specialization. In many jurisdictions, different sets of procedural processes are utilized depending on the above specialized categories as well as versatile types. For example, as stated above, minor property offenders and minor violent offenders in certain jurisdictions in Australia, New Zealand, Canada, the United Kingdom, and the United States are completely diverted away from juvenile justice court proceedings. In contrast, in Canada, the United States, and most other jurisdictions, the most serious violent juvenile offenders, such as murderers, are either processed in adult criminal courts or subject to adult length custodial sentences. In addition, patterns of offending, especially predominately minor violent offenses, automatically subject such offenders to more punitive sentences. The key assumption informing many of these sentencing options is that the minor violent offending specialization pattern typically indicates the potential for such youthful offenders to escalate into specialized violent and chronic offending.

The mixed results regarding the existence of specialization in youthful offending belie the above assumptions that there is a strong correlation between chronic minor juvenile offending and long-term chronic violent offending. While there is more support for the relationship between serious chronic violent juvenile offending and long-term violent offending trajectories, only an extremely small proportion of all juvenile offenders have been identified within this specialized type. The concern is that juvenile justice laws have become over focused on this rare offender category to the detriment of programming for the overwhelming number of less serious juvenile offenders. In other words, juvenile justice reforms have been influenced by the debate in research concerning specialization, arguably, in a progressive way in some countries (e.g., Australia, New Zealand, Canada, the United Kingdom) and in a regressive manner in several US states, for example, those with life sentences with no parole and formally capital punishment for the most serious juvenile offenders. Even more recently in Canada, a conservative federal government introduced punitive reforms to the national youth justice law based on the assertion that specialized violent and specialized property offenders require more punitive sentences.

Conclusion

There is a consensus that versatility typifies most juvenile delinquents or young offenders. There also is considerable evidence that for a small proportion of juvenile offenders, certain types of offenses distinguish them from other offense patterns. For example, sexual offenses, while not constituting the largest proportion of a young offender’s offending profile, nonetheless, distinguishes such an offender from another young offender who has no sexual offenses in their offending profile. In effect, the conceptualization of offender specialization remains open to different operational definitions. The key to the different conceptualizations is the theory or policy rationales. Major theoretical insights, for example, have emerged from developmental theories of offending that have asserted certain types of specialization as discussed above. The related theoretical debates and empirical research has propelled further theorizing and research that has several benefits. First, multiple measures and indices involving increased methodological sophistication have emerged and enhanced the understanding of juvenile offending trajectories. Second, the specialization research has revealed far more complex and diverse patterns of offending than were originally hypothesized. Third, it is vitally important that sophisticated research designs continue to be employed in assessing specialization because the ongoing juvenile justice policy debates concerning how to react to different types of offenders are best informed by valid and empirically supported categories of offending as opposed to impressions and emotion regarding what constitutes serious youthful offending.

Bibliography:

Blumstein A, Cohen J, Roth JA, Visher CA (1986) Criminal careers and ‘career criminals’, vol 1. National Academy Press, Washington, DC
Britt CL (1996) The measurement of specialization and escalation in the criminal career: an alternative modeling strategy. J Quant Criminol 12:193–222
Bursik RJ (1980) The dynamics of specialization in juvenile offences. Soc Forces 58:851–864
Cloward R, Ohlin L (1960) Delinquency and opportunity: a theory of delinquent gangs. Free Press, New York
Colvin M, Pauly J (1983) A critique of criminology: toward an integrated structural-Marxist theory of delinquency production. Am J Sociol 89:513–551
Elliott DS, Ageton SS (1980) Reconsidering race and class differences in self-reported and official estimates of delinquency. Am Sociol Rev 45:95–110
Farrington DP, Snyder HN, Finnegan TA (1988) Specialization in juvenile court careers. Criminology 26:461–485
Fisher G, Ross S (2006) Beggarman or thief: methodological issues in offender specialization research. Aust N Z J Criminol 39:151–170
Francis B, Soothill K, Fligelstone R (2004) Identifying patterns and pathways of offending behaviour: a new approach to typologies of crime. Eur J Criminol 1:47–87
Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford
Hirschi T (1969) Causes of delinquency. University of California Press, Berkley
Kempf K (1986) Offense specialization: does it exist? In: Cornish D, Clarke R (eds) The reasoning criminal: rational choice perspectives on offending. Springer, New York
Klein MW (1979) Deinstitutionalization and diversion of juvenile offenders: a litany of impediments. In: Crime and justice, vol 1. University of Chicago Press, Chicago, pp 145–201
Klein MW (1984) Offense specialization and versatility among juveniles. Brit J Criminol 24:185–194
Lattimore PK, Visher CA, Linster RL (1994) Specialisation in juvenile careers: Markov results for a California cohort. J Quant Criminol 10:291–316
Le Blanc M, Loeber R (1998) Developmental criminology updated. In: Crime and justice, vol 23. University of Chicago Press, Chicago, pp 115–198
LeBlanc M, Frechette M (1989) Male criminal activity from childhood through youth. Springer, New York
Loeber R (1990) Developmental and risk factors of juvenile antisocial behavior and delinquency. Clin Psychol Rev 10:1–41
Mazerolle P, Brame R, Paternoster R, Piquero A, Dean CW (2000) Onset age, persistence, and offending versatility: comparisons across gender. Criminology 38:1143–1172
Moffitt TE (1993) Adolescence-limited and life-coursepersistent antisocial behavior: a developmental taxonomy. Psychol Rev 100:399–420
Nieuwbeerta P, Blokland AAJ, Piquero AR, Sweeten G (2011) A life-course analysis of offence specialization across age: introducing a new method for studying individual specialization over the life course. Crime Delinq 57:3–28
Piquero AR, Paternoster R, Mazerolle P, Brame R, Dean CW (1999) Onset age and offence specialization. J Res Crime Delinq 36:275–299
Smith DR, Smith WR, Noma E (1984) Delinquent career lines: a conceptual link between theory and juvenile offences. Sociol Quart 25:155–172
Sullivan CJ, McGloin JM, Ray JV, Caudy MS (2009) Detecting specialization in offending: comparing analytic approaches. J Quant Criminol 5:419–441
Wikstrom P-OH (1987) Patterns of crime in a birth cohort: age, sex, and class differences (Project metropolitan: a longitudinal study of a Stockholm cohort, no. 24). University of Stockholm, Department of Sociology, Stockholm
Wolfgang ME, Figlio RM, Sellin T (1972) Delinquency in a birth cohort. University of Chicago Press, Chicago