Skip to main content
SearchLoginLogin or Signup

The role of psychopathic traits and developmental risk factors on offending trajectories from early adolescence to adulthood: A prospective study of incarcerated youth

Published onJan 01, 2015
The role of psychopathic traits and developmental risk factors on offending trajectories from early adolescence to adulthood: A prospective study of incarcerated youth
·

Abstract

Purpose: Criminal career research has recently identified that symptoms of psychopathy are more prevalent among offenders following chronic offending trajectories. In the current study, the ability of psychopathy to predict involvement in chronic offending trajectories above other criminogenic risk factors is examined. Methods: Criminal convictions were measured for Canadian male (n = 262) and female (n = 64) offenders at each year between ages 12 and 28. Semi-parametric group based modeling identified four unique trajectories labeled bell-shaped adolescent limited offenders (27.9% of sample), slow desisters (28.5%), slow rising chronic offenders (19.0%), and high rate chronic offenders (24.5%). Results: The four and three factor model of the PCL: YV were associated with the most chronic and serious offending trajectory even after controlling for a variety of relevant criminogenic risk factors. Self-reported involvement in weekly physical fights was a significant predictor of trajectory group membership, and criminogenic risk factors were more informative of the strength of the relationship between higher symptoms of psychopathy and offending trajectories than of a direct effect of a specific risk factor on the unfolding of offending trajectories. Conclusions: More research is needed to more fully integrate the construct of psychopathy into criminal career research.

Keywords

Criminal trajectories; group-based trajectory modeling, juvenile delinquency, PCL: YV, psychopathy


Despite Farrington’s (2005) call ten years ago for more systematic attempts at integrating psychopathy into criminological theories, the psychopathy construct has only recently been incorporated within the criminal career perspective. Instead, largely relying on various versions of Hare and his colleagues’ “gold standard” Psychopathy Checklist (PCL), the full psychopathy construct was utilized primarily to predict shorter-term violent, non-violent, and general recidivism outcomes (Corrado, Vincent, Hart, & Cohen, 2004; Edens & Campbell, 2007; Edens, Skeem, Cruise, & Cauffman, 2001; Salekin & Lynam, 2010; Vincent, Odgers, McCormick, & Corrado, 2008). Key symptoms of the psychopathy construct, such as a callous and unemotional disposition, usually combined, were more frequently incorporated into developmental psychology and developmental criminology child/adolescent/young adult studies (e.g., Loeber et al., 2001). Yet, these symptoms do not encompass the full psychopathy construct, nor have they been integrated within a criminal career framework. Only two studies have utilized instruments designed to examine the broader construct of psychopathy and its relationship with criminal career trajectories in full adulthood (e.g., through age 28; McCuish, Corrado, Lussier, & Hart, 2014) and middle adulthood (e.g., through age 40; Piquero et al., 2012). Although the age demarcation of these two adult stages are somewhat arbitrary, they nonetheless convey fundamental developmental changes associated with not only lifestyle turning points (e.g. legal drinking ages, occupation choices, longer-term intimate relationships), but also as discussed more recently by Corrado and Mathesius (2015), substantial maturation of the adult brain involving key executive functions.

The current study uses a sample of male (n = 262) and female (n = 64) offenders, initially recruited while incarcerated during a period of their adolescence, to expand on these two trajectory studies. The relationship between symptoms of adolescent psychopathy as measured by the Psychopathy Checklist: Youth Version (PCL: YV) and types of criminal trajectories measured through age 28 is examined while also considering several important criminogenic risk factors. Although there are few studies of this relationship, it is important to review key empirical findings and theoretical implications involving psychopathy. First, concerns regarding the extension of the adult psychopathy construct to childhood and adolescent developmental stages are discussed.

Extending the Psychopathy Construct to Childhood and Adolescent Developmental Stages

Lochman, Powell, Boxmeyer, Young, and Baden (2010) argued that the identification of high-risk subtypes among children and adolescents, historically, was a critical initial step in eventually relating child and adolescent manifestations of psychopathy to long-term criminal trajectories. Initial studies in criminology, though not discussing psychopathy specifically, indicated that this construct may be operating on early-onset antisocial behavior and persistent criminal behavior. Patterson and colleagues (1989, 1998), for example, asserted that early antisocial behavior was a developmental trait that was expressed in different forms at subsequent stages throughout the life course, including chronic offending by age 18. Similarly, Moffitt (1993) and Loeber and Stouthamer-Loeber (1998) labeled this early onset pathway of serious antisocial behavior and subsequent long term offending as life-course persistent (LCP) offenders. LCP offenders were thought to represent a small group of chronic offenders, roughly less than ten percent of the population, that Wolfgang, Figlio, and Sellin (1972) identified as being responsible for the majority of all crime (also see, DeLisi, 2005; Jennings & Reingle, 2012; Vaughn et al., 2011). Moffitt and Caspi (2001) identified parenting, neurocognitive functioning, and very early child temperament and behavioral problems as key correlates of the LCP antisocial behavior subtype. Similarly, although not finding the same early childhood-based temperament risk factors, Aguilar, Stroufe, Egeland, and Carlson (2000) identified high stress single parent families, an early childhood avoidant attachment style, and childhood abuse, including neglect or other forms of inadequate parenting as correlate of this LCP sub-type. Along with other developmental life-course criminologists (e.g., Farrington, 1989; LeBlanc & Fréchette, 1989; Loeber et al., 2008; Ribeiro da Silva, Rijo, & Salekin, 2012; Vaughn, Howard, & DeLisi, 2008), these initial studies asserted the existence of multiple risk pathways to long term offending, recidivism, and violent offending, with at least one pathway that included childhood onset of antisocial behavioral indicators typical of the antisocial domain of psychopathy. Vaughn and DeLisi (2008) asserted that the small number of chronic offenders and the small number of individuals with the strongest symptoms of psychopathic personality disturbance (PPD) was not coincidental but rather suggested that the two groups were actually comprised of the same individuals. There were no studies in this initial phase of research, though, that included a validated youth psychopathy instrument and examined its association with criminal trajectories into adulthood.

Two thematic changes occurred that facilitated the abovementioned line of empirical inquiry. Most importantly were the developments of the PCL: YV (Forth, Kosson, & Hare, 2003) and the PCL:SV (Hart, Hare, & Cox, 1995) instruments along with the self-administered (e.g., the subject or their parents and/or teachers) child and adolescent psychopathy screening instruments such as the Child Psychopathy Scale (CPS; Lyman, 1997), the Antisocial Process Screening Device (Frick & Hare, 2001), the adolescent Youth Psychopathy Traits Inventory (YPI; Andershed, Gustafson, Kerr, & Stattin, 2002), and the Psychopathy Content Scale (PCS; Murrie & Cornell, 2000). These instruments relied on the downward extension of the adult-established psychopathy personality disorder to adolescence and childhood. As expected, given intense and controversial debate about various validity issues concerning the use of the PCL for adults, especially concerning predictive validity (e.g., tautological concerns regarding the predominance of antisocial behavior items with recidivism; Skeem and Cooke, 2010) and the theoretically justifiable number and labeling of the PCL-R’s factor/facet structure (Cooke & Michie, 2001; Hare & Neumann, 2005), the use of the PCL: YV as well as other instruments raised even more validity issues. The controversy also included ethical concerns regarding the labeling of children and adolescents as psychopaths and its premature use as a risk prediction instrument in juvenile/youth justice settings and in sentencing and treatment planning (e.g., Edens et al., 2001; Salekin, Rosenbaum, Lee, & Lester, 2009). In addition, there have been internal validity concerns about the self-reported scoring of instrument items, particularly by children and adolescents about themselves. A related issue involves the appropriateness of using self-administered instruments in general and community samples of children and youth versus structured instruments for clinical and custodial samples (see Kotler and McMahon (2010) for a comparison of all these instruments and Salekin and Lynam (2010) for broader discussion of these validity issues).

The second theme was the use of these psychopathy instruments in on-going prospective longitudinal studies, such as the 40 year Cambridge Study in Delinquent Development in London, and in more recent longitudinal studies, such as the Pittsburgh Youth Study and the Pathways to Desistance Study. In Canada, the Incarcerated Serious and Violent Young Offender Study utilized the PCL: YV as did Gretton, Hare, and Catchpole (2004) in their young offender study. Many of these studies as well as others examined the relationship between psychopathy and types of offending, including reactive and instrumental violence, property and violent offending, sex and non-sex offending, relational and overt aggression, institutional misconduct, and short-term (i.e., 1-4 years) (Salekin, 2008; Edens & Cahill, 2007) and shorter-term recidivism (e.g., Leistico, Salekin, DeCoster, & Rogers, 2008; the average length of follow-up was 8.56 months). However, in these studies offending was rarely measured through initial developmental stages into adulthood. Additionally, several studies used the various psychopathy instruments either retrospectively with file data solely (Gretton et al., 2004), with a limited number of psychopathy traits (e.g., Loeber at al. 2001), or with psychopathy as a dependent variable and criminal/offending trajectories as one of the independent variables (Piquero et al., 2012).

Integrating the Psychopathy Construct within Criminal Career Research: Conceptual Challenges

In addition to some of the conceptual challenges previously discussed, there are three major concerns, related to construct validity and research design, which may have influenced the paucity of research concerning psychopathy and long-term patterns of offending. First, Edens et al. (2001) noted that some symptoms, asserted to be indicative of the adult psychopathy construct, resemble features of normative adolescent personality/behavior. For example, adolescents are typically more impulsive, more sensation-seeking, and more self-centered than adults and therefore the strength of symptoms of normative adolescent development resemble the strength of symptoms of adult psychopathy. For these and other reasons, Edens and colleagues (2001) argued that psychopathy measures should not be used to make longer-term predictions concerning criminal behavior. As the criminal career perspective emphasizes the unfolding of criminal behavior across multiple developmental stages, use of psychopathy measures would therefore seem contrary to Edens et al.’s (2001) recommendation. However, Forth et al. (2003) emphasized in their PCL: YV rating manual that items are to be rated taking into consideration what would be normative for adolescents, rather than what would be normative for adults. Furthermore, as noted by Cooke, Hart, Logan, and Michie (2012), in addition to symptom strength, functional impairment should also be taken into consideration. In effect, current assessment of adolescent symptoms of psychopathy incorporates, rather than ignores, Edens et al.’s (2001) concerns.

A second validity concern pertains to findings indicating that antisocial behavior markers, rather than more traditional affective and interpersonal symptoms of psychopathy, were more strongly associated with recidivism (e.g., Walters, 2003). As such, Walters (2004) suggested that the psychopathy construct was not a necessary part of the explanation of serious criminality and that low self-control essentially incorporated the antisocial factor of psychopathy and therefore the former construct was sufficient for explaining offending. However, these studies were primarily based on incarcerated samples1, and because most individuals within this population recidivate (e.g., Harris, Rice, & Lalumière, 2001), such a common event likely cannot be explained by a factor (i.e., psychopathy) that is much rarer within this population. As noted by McCuish et al. (2014), psychopathy may be better suited to explaining differences in offending trajectories through adulthood that better encompass the full range of offending patterns within incarcerated samples.

A third concern is related to issues of research design; specifically, the inability to capture the full range of symptoms of psychopathy among a sufficient proportion of the sample population. Although the efficacy of antisocial over affective/interpersonal symptoms of psychopathy has also been observed in community-based studies (e.g., Monahan et al., 2001), the full range of symptoms of psychopathy are rarely observed in these normative samples. For example, using data from the MacArthur Study of Mental Disorders and Violence, Skeem and Mulvey (2001) observed that the average score on the PCL: SV Factor 1 score (emotional detachment) for the sample was 3.11 (out of 12), less than half the average observed in some correctional samples (e.g., Douglas, Strand, Belfrage, Fransson, & Levander, 2005). Similar challenges were experienced when Piquero et al. (2012) first examined psychopathy within the context of criminal career trajectories in the Cambridge Study in Delinquent Development (hereinafter referred to as the Cambridge Study).

Piquero et al. (2012) examined an under-researched theme concerning the correlates of adult psychopathy and distinctive offending trajectories at age 40 utilizing a subsample of 304 of the original 411 boys born recruited in South London in 1953 as part of the Cambridge Study. However, psychopathy was measured at age 48 with the PCL: SV (Hart et al., 1995). In their analyses, they employed the two-factor (F1 Interpersonal-Affective and F2 Antisocial-Lifestyle) four-facet model (F1 interpersonal, F2 affective, F3 lifestyle and F4 antisocial) model, the two-factor (F1 and F2) model, and total PCL: SV scores. A key assertion in this study was that there was sufficient research indicating that psychopathy is stable across the life course, and, therefore, retrospective analysis across the previous life course stages regarding its association with criminal trajectories is theoretically appropriate. Also, unlike typical criminal trajectory studies, the Piquero et al. (2012) study utilized the trajectories as one independent variable among a total of 27 rather than as a dependent variable. Five offense trajectory groups were identified: non-offenders (62.3%), low adolescence peak offenders (18.6%), low rate chronics (11.3%), high adolescence peak offenders (5.4%), and high rate chronics (2.5%). They reported significant relationships regarding the comparisons of different factor-structure measures of psychopathy across the 5 offense trajectory groups. Other than f1 (interpersonal), as anticipated, the most serious offense trajectory, the high-rate chronic group, had the highest average PCL: SV factor/facet scores and total scores (12.17), whereas the least serious trajectory had the lowest (Piquero et al., 2012).

Very importantly, Piquero et al. (2012) examined the impact of offending trajectories on psychopathy scores, controlling for two indexes of individual and environmental factors comprised of 27 unique variables. All factors were based on measures to age 10 and before any criminal activity, which avoided criterion contamination. The individual index was comprised of dichotomous scores for 12 risk/independent variables: (1) low junior school attainment, (2) daring disposition, (3) small height, (4) low nonverbal IQ, (5) nervous/withdrawn boy, (6) high extraversion of boy, (7) high neuroticism of boy, (8) psychomotor impulsivity, (9) dishonest, (10) unpopular, (11) troublesome, and (12) lacks concentration/restless. The environmental index was comprised of 15 environmental risk factors: (1) harsh attitude/ discipline of parents, (2) teen mother at birth of first child, (3) behavior problems of siblings, (4) criminal record of a parent, (5) delinquent older sibling, (6) large family size, (7) poor housing, (8) low family income, (9) parental disharmony, (10) neurotic/depressed father, (11) neurotic/depressed mother, (12) low socio- economic status, (13) separated parents, (14) poor supervision, and (15) high delinquency rate at school. Controlling for these two indexes, the offending trajectories remained significant and were the strongest correlates of PCL: SV scores (Piquero et al., 2012).

Although the Piquero et al. (2012) study remains one of the most elaborate and theoretically insightful examinations of criminal offense trajectories and other correlates associated with psychopathy, it relies on a retrospective utilization of the psychopathy construct, and, as mentioned above, criminal trajectories as an independent variable. In addition, these researchers acknowledged the extensive and continuing controversies concerning construct validity issues as well as related issues concerning the measurement of psychopathy within community samples. Piquero et al. (2012) discussed the difficulty of justifying the minimum cut-off point on the PCL: SV in their community sample. While 18 and higher has been considered one PCL: SV cut-off standard (Hart et al., 1995; Cooke, Michie, Hart, & Clark, 2005), they argued that 16 and higher was appropriate for community samples. Arguably, the main justification for essentially arbitrary cut-off points (i.e., the absence of theoretically/conceptually justified minimums) is for their clinical use and the application of diagnostic categories. These researchers appropriately asserted that the dimensional theoretical/conceptual perspective of personality disorders did not require cut-off points but rather simply relied on the use of ordinal categories (e.g., more or less psychopathy). In their sample, only two cases were categorized as “severe” (i.e., 16 or higher PCL: SV total score) while 33 cases had scores of 10 or higher. The latter group had a higher likelihood of being convicted at age 40 and had a higher number of average convictions than those below the 10-point cut score. From the traditional trajectory perspective, this finding suggested more symptoms of psychopathy decreased the likelihood of desistance. Yet, with only eight individuals in the high chronic group, it is not evident that this community sample, as Piquero et al. (2012) acknowledged, allowed for a more full examination of the relationship between psychopathy and criminal offense trajectories (also see, DeLisi & Piquero, 2011). Samples of young offenders that are followed through multiple developmental stages and contain the full range of symptoms of psychopathy are needed (Piquero et al., 2012).

A more recent and continuing Canadian study of incarcerated serious and violent young offenders (n = 307) by McCuish et al. (2014) combined retrospective and prospective data to explore this relationship. PCL: YV scores measured in adolescence and criminal conviction records from age 12 to age 28 were obtained. In effect, this study explored criminal trajectories from early adolescence through middle adolescence, the transition from late adolescence, to early adulthood, to the full adulthood stage. Using semi-parametric group based modeling, four distinctive criminal offense trajectories were identified: adolescent limited offenders (27.5%), explosive-onset fast desisters (30.6%), high rate slow desisters (14.6%), and high frequency chronic offenders (27.3%). Using a four facet and three facet model, increases in PCL: YV scores were significantly associated with increases in the odds of membership in the high rate slow desister and high frequency chronic offending trajectories compared to the adolescent limited trajectory. However, when the four facets were disaggregated, only the antisocial facet was significantly associated with any of the four trajectories; as expected, the high rate chronics and the high rate slow desisters. The McCuish et al. (2014) study expanded on the Piquero et al. (2012) study by using measures of psychopathy to help explain involvement in chronic offending trajectories, rather than vice-versa. Missing from their models was important additional criminogenic factors that (a) may have their own relationship with offending trajectories and (b) may be responsible for the variance in trajectory membership that was initially accounted for by psychopathy.

Study Aims

Using the same data as McCuish et al. (2014) but with a slightly larger sample (n = 326 versus 307)2, the purpose of the current study was to simultaneously examine the relationship between psychopathy as well as other critical criminogenic risk factors and offending trajectories. Although various studies have demonstrated the importance of psychopathy above and beyond other covariates, this line of analysis has yet to be examined from a criminal career perspective. Reliance on an offender-based sample may limit its generalizability; however, as indicated by Piquero et al. (2012), more research is needed on offense trajectory studies that measure psychopathy within high-risk samples.

Methodology

Sample

Data for the current study were derived from the Incarcerated Serious and Violent Young Offender study conducted in British Columbia, Canada. As part of this study, adolescent offenders between the ages of twelve and nineteen were interviewed in open and secure custody facilities within the Greater Vancouver Regional District and surrounding areas. This sample of Canadian incarcerated adolescent offenders is very specific and generalizing the results of the current study to non-incarcerated populations should be done with caution. The study has been ongoing since 1998 and additional details of the sampling strategy have been discussed at length in prior publications (e.g., Corrado, Cohen, Glackman, & Odgers, 2003; Corrado, Vincent, Hart, & Cohen, 2004; McCuish et al., 2014). Focus within the current study was on a subsample (n = 326) that had been assessed using the PCL: YV. With the exception of seven percent of the sample who were twenty-seven at the time of data collection, convictions for all offenders were coded until age twenty-eight. The sample is overwhelmingly composed of male (80.4%) and Caucasian (60.9%) offenders. The average age of offenders at the time of their assessment was approximately 16.

Procedure

The purpose of the Incarcerated Serious and Violent Young Offender study was to conduct interviews with juvenile offenders and collect file-based information on risk factors associated with the onset, persistence, and desistance of adolescent criminal activity and to identify risk factor profiles associated with the development of serious and violent offending. To recruit research participants, informed consent was first sought and provided by the British Columbia Ministry of Child and Family Development (MCFD). MCFD serves as the legal guardian to all youth in custody, and their consent allowed the research team to approach all youth in custody centers throughout the province of British Colombia. Youth were approached by a member of the research team at their respective custody center and asked whether or not they wanted to participate. Subjects were eligible to participate in the current study provided that they met several criteria: (1) were English-speaking, (2) demonstrated an understanding of interview questions (e.g., had no noticeable deficits in IQ), and (3) were willing to provide accurate information. Of those eligible, only approximately 5% of youth declined to participate. All subjects were informed that they information they provided would be kept, with the exception of the subject making a direct threat against themselves or someone else. To improve the reliability of the subject’ self-reported information, research assistants were granted access to case management files, which contained subjects’ presentence reports and information on their behavior while in the institution. Access to file information prior to interviews allowed RAs to be aware of discrepancies between interview responses and official records.

Measures

Ethnicity and gender were measured through self report interviews. Although some offenders in the current study were in their early thirties, criminal trajectories were only measured to age 28 and therefore it was unnecessary to control for age in subsequent analyses. The primary focus within the current study was on whether symptoms of psychopathy, controlling for other criminogenic factors, was associated with a specific course of offending in adulthood. All criminogenic risk factors were measured at the time of the subject’s interview during their incarceration in adolescence. Seven types of risk were examined for the individual offenders: substance use, school behavior issues, abuse experiences, sexual activity, personality development, residential mobility, and aggression. All of these measures are outlined below in greater detail. Characteristics of the sample are summarized in Table 1.

Psychopathy Checklist: Youth Version (PCL: YV; Forth et al., 2003).3 The PCL: YV is a symptom rating scale that ranges from 0-2 and is scored using information from a 60-90 minute semi-structured interview and a review of file-based collateral information. Access to file information, in addition to interviews, were used to score the PCL: YV. Inter-rater reliability was not conducted in this particular study; however, Vincent (unpublished doctoral dissertation) evaluated inter-rater reliability in a subsample of 30 randomly selected cases and the intraclass correlation coefficient was high (ICC1 = 0.92). The 20 items comprising the PCL: YV were believed to represent the fundamental personality and behavioral traits represent the construct of psychopathy in adolescence. These 20 items represent different facets of the underlying psychopathy construct. Forth et al. (2003) recommended using a four factor model that consists of an interpersonal factor, an affective factor, a lifestyle factor, and an antisocial factor. Cooke and Michie (2001) recommended a three-factor model that excludes Forth et al.’s (2003) Antisocial factor to avoid using prior criminal behavior to predict future criminal behavior. Total scores, aggregated factor scores, and individual factors are presented in Table 1. Total scores did not differ between males and females. Approximately one third of the sample scored what could be considered ‘high’ on the PCL: YV (25 or higher).

Criminogenic Risk Factors. Substance use included separate measures of the age of onset of alcohol and drug use as well as eight dichotomized items (alcohol, marijuana, hallucinogens, ecstasy, cocaine, heroin, crack cocaine, and crystal meth) used to create an aggregate scale of self-reported substance use. Scale reliability was high (0.88) based on the tetrachoric ordinal alpha value, which is more reliable than Cronbach’s alpha for dichotomous items (Gadermann et al., 2012). School behavior issues included the age at which they began getting into trouble at school, the age at which they started skipping school, the number of times that they changed schools, and whether they were attending school prior to their incarceration. Abuse experiences included dichotomous self-report measures of whether the youth had experienced physical abuse and sexual abuse. Sexual activity was measured using one item on the age of onset of consensual sexual activity. Personality development was measured using Schneider’s (1990) Good Citizen’s Scale, a self-report inventory of 15 identity traits coded on a 1-7 scale (Cronbach’s alpha = 0.74), with items reverse-coded so that lower scores indicated a negative identity. Aggression was assessed by asking participants about the frequency of their involvement in physical fights, whether the participant felt they got angry easily, and whether the participant reported that someone had told them they had a bad temper. To measure familial delinquency and disruption, participants were asked to report whether any members of their biological parents or biological siblings had trouble with alcohol or drugs, had experienced physical or sexual abuse, had a criminal record, or had mental illness. These six items were aggregated into a global scale (tetrachoric ordinal alpha = 0.78). Residential mobility measured whether the participant had left home willingly for more than a day to live somewhere else, whether the participant had been kicked out of their home for more than a day, whether the participant was raised by their biological parents, and whether the participant lived in foster care or other forms of ministry care.

Measures of Offending. Offending was measured using official data from British Columbia Corrections’ computerized system, Corrections Network (CORNET), which contains information on an offender’s movement in and out of custody as well as the exact criminal offense, date of conviction, and sentence type received. CORNET data includes only offenses committed within the province of British Columbia. Using data from this computerized system, every criminal charge that resulted in a conviction was coded for the entire sample from age 12, the age of criminal responsibility in Canada, to age 28. In line with prior studies measuring offending trajectories, for the seven percent of offenders who had not reached age 27, their offending for age 28 was coded as missing (Eggleston, Laub, & Sampson, 2004; Livingston, Stewart, Allard, & Ogilvie, 2008; van der Geest et al., 2009). Also in accordance with these studies, because 11 offenders died (3.4%) and six (1.8%) moved outside the province, convictions for these offenders after the age of death or move were coded as missing rather than as ‘zero’. For this sample, the average number of charges for which the individual was convicted for was 23.28 (SD = 17.46). The median number of convictions was 19.5, showing that the higher number of convictions was not an artifact of a small subgroup of individuals. The vast majority of the sample (84.0%) had been convicted of a violent offense. Age of onset, based on age at first court appearance was, on average, fourteen years old. In total, 13.5 % and 27.0 % of the sample first appeared in court at 12 and 13 years old, respectively. Total time spent in custody was also calculated and controlled for. On average, offenders spent 1,166 (SD = 1167) days in custody. The median number of days in custody was 771 and twenty-five percent of the sample spent at least 1,875 days in custody from age twelve to twenty-eight.

--Insert Table 1 about Here--

Analytic Strategy

The number and shape of the offending trajectories that best fit the data were identified by using semi-parametric group based modeling (SPGM; Nagin & Land, 1993). Unlike cluster analysis and other grouping methods that identify groups ex ante, the SPGM method allows developmental trajectories to emerge from the data (Nagin, 2005). SPGM method was used to identify offending trajectories for the sample (n = 326) and was measured using all convictions between age 12 and 28. To control for time at risk, exposure time was built into the SPGM model by adapting Piquero et al.’s (2001) original formula. This adaptation adjusted for high standard errors and improbable rates of offending by constraining the minimum exposure time4 to equal approximately 0.2:

Exposureji = 1 - (Number of Days Incarcerated/455)

where j is the respondent and i is the year of observation.

The association between trajectories, psychopathy, and criminogenic factors were examined in a series of bivariate analyses. All significant criminogenic risk factors and measures of psychopathy were then included in a multinomial logistic regression analysis to examine whether these factors helped predict a particular course of offending (i.e., the offending trajectories).

Results

Model Identification and Interpretation

The SPGM analyses in the current study proceeded in two stages, the first involved model identification, which focused on identifying the number and shape of the offending trajectories that best fit the data. Trajectory analyses were conducted in SAS 9.4 using the Proc TRAJ add-on developed by Jones and colleagues (2001; see also Jones & Nagin, 2007). The zero-inflated Poisson (ZIP) model with quadratic functional form was used to estimate the distribution of the offending trajectories. Bayesian Information Criteria (BIC) values were used to identify the number of offending trajectories that best represented the data. A four group quadratic model resulted in a BIC value of -8530, which was closer to zero than both a three group model (BIC = -8803) and a five group model (BIC = -8543). BIC values for a four group solution with quadratic functional form were also closer to zero than the same model with cubic functional form. In addition, Jeffrey’s scale of evidence based on the Bayes factor approximation was used to determine whether there were substantive differences in BIC values between models specifying a different number of trajectory groups (e.g., Nagin, 2005). Jeffrey’s scale of the evidence of the Bayes factor is calculated as e BICi – BICj where values of Bij greater than ten indicate strong evidence for model ‘i’ (see Nagin, 2005). Based on the BIC values of a ZIP model with quadratic functional form, there was strong evidence for a four group model over a five group model (Bij >10) but not for a three group model over a four group model (Bij <10). The parameters of the four group model are outlined in Table 2 and help support the retention of a four group model. Classification accuracy based on the average posterior probability of accurately assigning individuals to a particular trajectory was high for all four trajectories (range of 0.92-0.94). Finally, odds of correct classification (OCC) was used to help provide confidence that individuals were assigned to the appropriate trajectory group. OCC values for each trajectory group were calculated as:

OCCg = (AvePPg/ (1-AvePPg)) / (∏g/ (1-∏g))

where ∏g is the estimated size of group g (see Skardhamar, 2010).

As indicated in Table 2, OCC values for the four trajectories ranged from 11-15, higher than both Nagin (2005) and Skardhamar’s (2010) recommendation that OCC values of at least five be used to indicate high classification accuracy.

--Insert Table 2 about Here--

Figure 1 presents the four trajectories. It should be noted that these trajectories are nearly identical to those identified in McCuish et al. (2014), as should be expected given that the same study with a slightly larger sample size was used. However, the labels of the trajectories presented here are not the labels given to the trajectories in McCuish et al. (2014). These new labels were thought to better describe each trajectory group. The bell-shaped trajectory (28.5% of the sample) represented a group of low rate offenders. Offending typically began at age 13-14, peaked at age 15, reached a near-zero rate by 19 and an absolute zero rate by 23. The second trajectory group, slow desisters, comprised just over a quarter of the sample (28.0%). This group resembled the bell-shaped trajectory but differed in that individuals in this group continued to offend, albeit at a low rate, throughout their twenties. The third trajectory group, referred to as slow-rising chronics (SRC; 19.0%) offended during mid-adolescence at a rate that was similar to offenders in the bell-shaped and slow desister trajectories. However, by age sixteen, offending frequency began to escalate and offenders maintained a steady rate of offending throughout their twenties. Finally, the high-rate chronic (HRC) offending trajectory (24.5%)5 began offending much earlier than the other three trajectories and showed a steady increase in offending throughout adolescence. However, during adulthood, the rate of offending declined for individuals in this group and was surpassed by the offending rate in the SRC trajectory. As indicated in Table 2, age of onset, offending frequency, and time in custody were higher for the two chronic trajectories (SRC and HRC) compared to the two lower-rate trajectories (bell-shaped and slow desisters). The association between trajectories, demographic characteristics, criminogenic risk factors, and psychopathy is outlined in Table 3.

For the bivariate analyses that involved ANOVA, Bonferroni (equal variances assumed) or Tamhane (equal variances violated) post-hoc comparisons indicated significant differences between trajectories. Eta squared was used to provide an indication of effect size. In terms of demographic characteristics, males comprised the vast majority of the SRC (88.7%) and HRC trajectories (95.0%). It appeared that chronic offending, as defined in this study, was almost exclusively a male phenomenon. If a female chronic offending trajectory does exist, it likely is characterized by a lower rate of offending relative to male chronic offending trajectories. Identifying this female chronic offender trajectory may require conducting SPGM separately for females. In terms of ethnicity, Asian, East Indian, Middle Eastern, and African-Canadian offenders (i.e., those who comprised the ‘Other Ethnicity’ category) were more likely to be in the two low-rate trajectories (bell-shaped and slow desister) compared to the two chronic trajectories. In terms of criminogenic risk factors, five of the seven risk factor domains had at least one risk factor that differed between trajectories. For the SRC trajectory, higher scores on the substance use scale, getting into fights weekly, and a negative self-identity stood out compared to the bell-shaped trajectory. An earlier age of getting into trouble at school, an earlier age of onset of sexual activity, and a negative self-identity differentiated the HRC trajectory from the bell-shaped trajectory. Significant differences in criminogenic risk factors were not observed between the slow desister trajectory and the other three trajectories.

Comparisons between psychopathy and offending trajectories revealed a number of important differences. When the PCL: YV four factor model was examined, the HRC and SRC trajectories were observed to have significantly higher scores compared to the bell-shaped trajectory and the slow desister trajectory. However, when a three factor model was examined, the HRC trajectory but not the SRC trajectory had significantly higher scores compared to the other two trajectories. When the four PCL: YV factors were examined independently, the SRC trajectory had significantly higher scores than the bell-shaped and slow desister trajectory on measures of the antisocial factor but not the other three factors. In contrast, the HRC trajectory scored significantly higher on the affective, lifestyle, and antisocial factors compared to the bell-shaped and slow desister trajectories. This indicated that symptoms of psychopathy may be more salient for the HRC trajectory, whereas antisocial markers best characterized the SRC trajectory.

--Insert Table 3 about Here--

The HRC Trajectory: Are They Really Desisting?

It was expected that symptoms of psychopathy would predict membership in the most chronic offending trajectory through age 28. Seemingly in contrast to this expectation, individuals in the SRC trajectory offended at a higher rate in adulthood compared to individuals in the HRC trajectory, yet symptoms of psychopathy seemed to better characterize the latter trajectory. However, it was possible that the SRC trajectory offended at a higher rate whereas the HRC trajectory committed more serious and violent offenses that would result in lengthier sentences and reduce additional offending opportunities. Clearly, based on Figure 1, the HRC trajectory offended at a higher rate in adolescence. Therefore, the specific focus of comparisons between these two trajectories was on their patterns of offending through adulthood. By age 20, the SRC trajectory’s rate of offending appeared to surpass the HRC trajectory’s rate of offending. The analyses were thus focused on examining differences in the offending patterns of these two trajectories between age 20 and 28.

Although the SRC trajectory averaged a significantly greater number of convictions over this period, the SRC and HRC trajectories spent an equal length of time in custody over this same period. This suggested that although the HRC trajectory was committing fewer offenses, the offenses they did commit were resulting in lengthier sentences (i.e., longer periods of incarceration). Instead of looking at total convictions, total violent convictions from age 20 to 28 were measured. Despite the SRC group committing twice as many general offenses in adulthood compared to the HRC group, the average number of violent convictions between age 20 and 28 for the HRC group (1.51) did not differ from the SRC group (1.78). This suggested that when HRC offenders were convicted of an offense, it was more likely to be violent compared to the SRC group. Indeed, a violence specialization coefficient was calculated by dividing total violent convictions between ages 20-28 by total general convictions for the same period. The HRC offenders’ proportion of violent offenses (0.15) was significantly higher than the average proportion of violent offenses for SRC offenders (0.08). Since violent offenses typically receive more punitive sanctions, it was possible that an offense seriousness metric would also help explain why the HRC trajectory appeared to be desisting from offending despite having the strongest association with symptoms of psychopathy. The seriousness metric was determined by dividing the number of days spent in custody by the number of convictions incurred during the specified time period. In effect, the seriousness metric assessed the average length of an offender’s sentence. The HRC trajectory averaged significantly higher scores on the seriousness metric compared to the SRC trajectory.

--Insert Table 4 about Here--

Taken together, although initial findings shown in Figure 1 indicated that the SRC trajectory was a more serious group of offenders between the age of 20 and 28, a closer look at the patterns of offending of the HRC group indicated that they were offending less often, but committing more serious offenses relative to the SRC trajectory. It should also be kept in mind that the formula for exposure time had to be constrained to avoid improbable rates of offending. Although this constraint would impact all trajectories, it would have the largest impact on the group that had the lowest ratio of convictions to time incarcerated. The HRC group appeared to best resemble this type of group. In fact, on 93 occasions between age 20 and 28, an offender in the HRC trajectory was incarcerated for at least 365 consecutive days. As such, despite appearing to be a group of offenders in a phase of desistance, the HRC group is perhaps better characterized as a group of individuals with high symptoms of psychopathy that engaged in a variety of offenses in adolescence but transitioned towards more serious and violent offending in adulthood. The effect of symptoms of psychopathy on trajectory group membership was examined in greater detail by controlling for demographic characteristics and important criminogenic risk factors.

Psychopathy and other Covariates of Offending Trajectories

Multinomial logistic regression analyses were performed to examine covariates associated with the offending trajectories. In addition to psychopathy, all significant variables examined in Table 3 were examined in subsequent analyses. Three different models were produced to examine the predictive utility of the four factor structure, three factor structure, and the four individual factors: interpersonal, affective, lifestyle, and antisocial. Correlations between all variables were examined to check for multicollinearity, especially because of the concern that the four individual factors would be highly correlated with one another. All correlations were low to moderate (none over 0.5). The bell-shaped trajectory had the lowest rate of offending and was least associated with all risk factors and therefore was used as the reference category. In the first regression model in Table 5, as scores on the four factor model increased, the odds of being in the HRC trajectory increased (OR = 1.17) compared to the bell-shaped trajectory. Four factor model scores were not predictive of membership in the other trajectories. However, offenders who fought on a weekly basis in adolescence were over three times more likely to be in the SRC trajectory compared to the bell-shaped trajectory. Identical results were observed for the second regression model that included the three factor PCL: YV model. This analysis indicated that the relationship between psychopathy and offending was not due to inclusion of prior offending behavior in the measure of psychopathy. However, as illustrated in the third regression model, the antisocial and lifestyle factors, but not the interpersonal and affective factors, significantly increased the odds of membership in the HRC trajectory compared to the bell-shaped trajectory. Interestingly, higher scores on the affective factor of the PCL: YV predicted membership in the slow desister trajectory compared to the bell-shaped trajectory. Consistent with the previous two models, individuals involved in weekly physical fights were approximately four times more likely to be in the SRC trajectory compared to the bell-shaped trajectory. Other criminogenic risk factors were not significant in any of the three models examined.

--Insert Table 5 about Here--

Discussion

The focus of research on the intersection of psychopathy and criminal behavior has been primarily limited to studies that utilized the ‘next offense’ as the dependent variable. While important in theorizing about the utility of psychopathy as a potentially important construct in explaining serious criminal offending, Farrington (2005) advocated that it now was necessary examine the relationship between the psychopathy construct and criminal careers. From a developmental criminological perspective (e.g., Loeber & LeBlanc, 1990), criminal career parameters are particularly useful for understanding serious young offenders, especially incarcerated offenders, who typically have high base-rates of recidivism. However, numerous studies that have utilized various instruments to measure symptoms of psychopathy typically focused on the ‘next offense’ instead of a fuller account of the offenses that comprise the broader criminal career. Given that base rates of recidivism were quite high among incarcerated offenders, these recidivism studies have inherent limitations in identifying long-term (e.g., chronic) offenders. The purpose of the current study was to expand on the earlier work of Piquero et al. (2012) and McCuish et al. (2014) by using symptoms of psychopathy to predict offending trajectories, rather than vice-versa, while controlling for theoretically relevant criminogenic risk factors. The four trajectories identified in the current study, unsurprisingly, were nearly identical to the McCuish et al. (2014) study given that the current study used the same data with a slightly larger sample. The trajectories in the current study were labeled differently, not because of changes in the shape of trajectories between studies, but to more accurately depict their shape. The high-rate chronic (HRC) trajectory appeared to begin a process of desistance in adulthood. However, the HRC trajectory comparisons with the slow-rising chronic (SRC) trajectory, which had the highest rate of offending in adulthood, indicated that the crime mix of the HRC trajectory included the highest proportion of violent offenses and the more punitive sentences. As anticipated, given the extensive research that indicated a strong association between psychopathy and serious criminal offending, the HRC trajectory group had the highest symptoms of psychopathy, assessed using both a four and three factor model of the PCL: YV.

Equally important, the current study indicated that both a four and three factor model of the PCL: YV were associated with the most chronic and serious offending trajectory even after controlling for a variety of relevant criminogenic risk factors. Specifically, of the six significant risk factors that emerged from initial bivariate analyses, only self-reported involvement in weekly physical fights was a significant predictor of trajectory group membership. In effect, the criminogenic risk factors were more informative of the strength of the relationship between higher symptoms of psychopathy and offending trajectories than of a direct effect of a specific risk factor on the unfolding of offending trajectories. However, an important construct validity issue regarding psychopathy and its relationship to criminal offending phenomena, generally, was evident. The socially deviant PCL: YV factors, antisocial and lifestyle, but not the affective and interpersonal PCL: YV factors, had the strongest relationship with the HRC group. Accordingly, this finding supports Walters’ (2003) and Salekin and Lynam’s (2010) perspective that the absence of evidence that these core symptoms of psychopathy related to offending restricted its theoretical utility. It is important to examine this persistent criticism further, including whether the non-association between core affective and interpersonal symptoms of psychopathy and chronic offending reflects a conceptual or operational measurement weakness of psychopathy instruments such as the PCL-R and its several derivatives including the PCL: YV, used in this study (Kotler & McMahon, 2010).

Core Personality Features of Psychopathy and Offending Trajectories

Analyses examining both short-term recidivism (e.g., Corrado et al., 2004; Walters, 2003) and a longer-term recidivism study (Gretton et al., 2004) have indicated that interpersonal and affective measures are less indicative of recidivism or more prolonged offending compared to lifestyle and antisocial measures. Now that these findings were also evident in this trajectory study suggests the need to focus theoretical exploration on explicating with greater specificity the relationship between psychopathy, its different symptoms, and chronic offending trajectories. The central question is whether the prominence of socially deviant PCL: YV factors reflects problems with, as Cooke et al. (2012) described, the map (i.e., the measure), or the terrain (i.e., the psychopathy construct). In effect, Cooke et al. (2012) asserted the need to explore a more comprehensive or detailed mapping of the complex and possible additional domains and related symptoms of the terrain of psychopathy. Lynam (2010) too argued earlier for the need for a more complex domain and symptom mapping of psychopathy utilizing the traditional five-factor/30 facet model of personality, which he asserted was a more theoretically valid basis for the psychopathy construct. This fundamental construct concern, therefore, raised the following two questions regarding this study’s key findings. First, are the core affective and interpersonal symptoms long asserted to comprise the essential personality dimension of the construct of psychopathy (i.e., the terrain) simply unrelated to longer-term chronic offending trajectories? Or, is the PCL: YV’s overreliance on behavioral markers and limited range of affective and interpersonal symptoms (e.g., Dawson, McCuish, Hart, & Corrado, 2012; Skeem & Cooke, 2010) an explanation for why the affective and interpersonal factors (i.e., the map) failed to distinguish between chronic and non-chronic offending trajectories? Second, why are interpersonal and affective deficits theoretically central to explaining the impact of psychopathy on long-term chronic general offending trajectories?

With respect to the psychopathy construct generally and the last question more specifically, it is important to review the early research on the prototypicality of different symptoms of psychopathic personality disturbance (see Cooke et al., 2012; Hoff, Rypdal, Mykletun, & Cooke, 2012; Kreis & Cooke, 2011; Kreis, Cooke, Michie, Hoff, & Logan, 2012) and possible profiles of these symptoms related to certain types of criminal offending patterns. These prototypicality studies identified a broad range of affective and interpersonal symptoms, including a lack of attachment, empathy, caring, and commitment towards others, a lack of remorse, emotional depth, and emotional stability across multiple situations, a domineering, antagonistic, insincere, and manipulative interpersonal style that features a focus on self, including entitlement, aggrandizing, and justification, and glibness, garrulousness, and verbosity in interactions with others. Regarding symptom profiles and types of crime, from a criminological theoretical perspective, strong symptoms of callousness towards others combined with a domineering interpersonal style more likely facilitate involvement in violent offenses especially the former symptoms for engaging instrumental violence (Porter & Woodworth, 2006). In contrast, interpersonal and affective deficits, arguably, may be less central to explaining property offenses, especially where interpersonal contact is avoided, and drug trafficking where larger criminal organizations working together benefit over independent entrepreneurs (e.g., Tremblay, Bouchard, & Petit, 2009). Similarly, violations of court orders, one of the more common offenses for adolescent offenders, have appeared to be related to a lack of stake in conformity, impulsivity, drug addiction, homelessness, and a difficulty following direction (Corrado et al., 2003; Corrado, Odgers, & Cohen, 2000). In effect, although interpersonal and affective traits may influence involvement in more serious types of violent offending, the influence that these symptoms have on more common, less serious offenses that comprise a large proportion of the crime mix of chronic offenders may be quite limited. Instead of the expectation that interpersonal and affective traits are necessary components of theories of chronic general offending, it may be more appropriate to incorporate these psychopathy symptoms as primary in the explanation of continued violent offending across the criminal career.

Regarding the measurement of psychopathy theme (i.e., the map), it is necessary to more fully explicate the symptoms of psychopathy in order to describe the hypothesized theoretical relationship between psychopathy symptoms and long term patterns of violent offending. This entails developing a psychopathy instrument that encompasses the full range of interpersonal and affective symptoms associated with psychopathic personality disturbance. Arguably, the mapping of the PCL: YV’s affective and interpersonal items (glibness, grandiosity, pathological lying, manipulative, lacks remorse, shallow affect, callous/lack of empathy, and a failure to take responsibility) onto the prototypical symptoms of psychopathy (Cooke et al., 2012; Hoff et al., 2012; Kreis et al., 2011; Kreis et al., 2012), suggests that the PCL: YV does not fully cover the complex multi-domain and related wide array symptom terrain. In contrast, the Comprehensive Assessment of Psychopathic Personality Disorder (CAPP), which was constructed based on these prototypicality ratings, appears to include a fuller range of affective and interpersonal symptoms than the PCL: YV. For the current study then, a non-association between interpersonal affective symptoms and chronic offending may also be related to the inability of the PCL: YV to fully capture these symptoms. In effect, there may be chronic offenders with substantial interpersonal and affective deficits that fall outside the scope of the PCL: YV. Sandvik et al. (2012) have shown that the CAPP has high inter-correlations with the PCL-R, suggesting both instruments tap into the same underlying construct. However, the CAPP also diverged from the PCL-R in several critical ways. Most importantly, both qualitative and quantitative examinations of the CAPP indicated that this instrument emphasized affective symptoms and de-emphasized antisocial and criminal behavior acts (Dawson et al., 2012; Sandvik et al., 2012). Although the PCL: YV is still considered the gold standard in measuring the psychopathy construct, it is important to not equate the measure with the full construct. Nonetheless, the antisocial and lifestyle factors appeared important in explaining the unfolding of the criminal career. .

Lifestyle and Antisocial Factors in the Unfolding of the Criminal Career

Higher PCL: YV scores on the lifestyle and antisocial factors were both helpful in identifying individuals who are associated with the HRC trajectory. The importance of the antisocial factor seems obvious since past behavior has long been identified as a strong predictor of future behavior (Robins, 1978). In addition, the possibility of criterion contamination also may have contributed to this finding because, although the majority of this study was prospective, most adolescents’ PCL: YV ratings were assessed at age 15-16 but criminal records were examined from age 12 onward. However, the lifestyle factor did not have the same validity issues, and emerged as an important predictor of long-term chronic offending. The items comprising the lifestyle factor (stimulation seeking, impulsivity, irresponsibility, parasitic orientation, lacks realistic goals) certainly appear, according to most individually focused criminological theories, related, both directly and indirectly, with continued offending across the life course (e.g., Gottfredson & Hirschi, 1990; Moffitt, 1993). Regarding the direct relationship, the item measuring impulsivity could serve as a proxy for the essential part of the key criminological construct of low self-control (DeLisi, 2009). Consistently committing financially-motivated crimes seems necessary for a parasitic orientation life-style. Highly irresponsible individuals typically have difficulty following rules, implying proneness to violate court-orders and frequently being ‘breached’ by probation officers. Regarding the indirect relationship, Forth et al. (2003) argued that stimulation seeking individuals were prone to extensive and versatile substance use. Property crimes are not an uncommon means to sustain this drug-using lifestyle, particularly for individuals with high levels of irresponsibility and unrealistic goals. Whether specific symptoms of psychopathic personality disturbance are more strongly associated with specific types of offending is a key theme that should be addressed in future research.

Limitations

This study focused on individuals that have been incarcerated during a period of adolescence. The specificity of this sample may limit generalizability, but, nonetheless, it is the sample-type needed to explore the relationship between psychopathy and chronic offending. Although a population-based community sample typically is more generalizable, neither psychopathy nor chronic offending is sufficiently prevalent in such samples to describe the relationship between these two concepts. In addition, the current study was conducted in British Columbia, Canada and therefore the percentage of Aboriginal offenders was dissimilar from incarcerated adolescent offenders in the United States (e.g., Teplin et al., 2013). However, the over-representation of Aboriginal offenders in Canada is not dissimilar to the over-representation of African American and Hispanic offenders in the United States (Cross, 2008).

Also, official data usually underestimates offending rates; therefore, individuals with high symptoms of psychopathy may be particularly adept at avoiding police detection for each criminal offence. Yet, in self-report surveys these same individuals also may be more likely to purposefully under-report the level of their involvement in criminal behavior. Further, serious offenses are the most likely to be purposefully under-reported in self-report surveys (see Stouthamer-Loeber, Loeber, Stallings, & Lacourse, 2008), and individuals with psychopathic personality disturbance are most likely to be consistently involved in these types of offenses. Taken together, self-report surveys may have difficulty accurately capturing the rate and severity of offending patterns of individuals with psychopathic personality disturbance. . Nevertheless, a research design that includes both official and self-reported offending will facilitate a more complete exploration of this theoretically complex relationship.

Conclusions and Future Research

The current study is a necessary initial step in explicating the relationship between psychopathy and criminal careers (see also McCuish et al., 2014; & Piquero et al., 2012). Very importantly, higher scores on the PCL: YV’s three and four factor model were associated with chronic general offending even after controlling for several key criminogenic risk factors. Given the paucity of research on this theme, this study’s results have several theoretical implications and related research questions. Specifically, interpersonal and affective symptoms possibly are more appropriate in explaining chronic violent offending than chronic general offending (see McCuish, Corrado, Hart, & DeLisi, this issue). There is the need to explore more fully the extensive theorizing and research concerning comorbidity, especially given the more recent related research on genetics, epigenetics, and the more complex developmental models of personality across the life-course e.g. (DeLisi & Vaughn, 2014; Sevecke & Kosson, 2010; Vaughn & DeLisi, this issue; Viding & Larsson, 2010).

Only the PCL: YV and PCL: SV (Piquero et al., 2012) to date have been used to examine criminal careers. Similar research questions should be examined with different research instruments, especially those that emphasize personality dimensions rather than behavioral markers. Finally, although SPGM is perhaps the most popular analytic strategy for modeling offending trajectories (Piquero et al., 2008), this type of analysis does not capture qualitative differences in criminal career severity. For example, individuals with strong symptoms of psychopathy who commit primarily very serious offenses are unlikely to be associated with high-rate chronic offending trajectories. Instead, this type of offender appears more likely to be found in a low-rate trajectory that continues throughout the life course rather than a high rate trajectory that consists of more mundane offenses This pattern may explain why the PCL: YV’s affective factor was significantly associated with the slow desister trajectory in the current study. To explicate this theme, future studies should examine whether affective and interpersonal symptoms are associated with violence specialization. Another approach involves examining the ‘crime mix’ within different trajectories. There is no one trajectory that encompasses all individuals with high symptoms of psychopathy. Instead, within different offending trajectories, individuals with the highest symptoms of psychopathy may be responsible for the largest proportion of more serious (e.g., violent, lengthier sentences) offenses.

References

Aguilar, B., Sroufe, L., Egeland, B., & Carlson, E. (2000). Distinguishing the early-onset/persistent and adolescence-onset antisocial behavior types: From birth to 16 years. Development and Psychopathology12(02), 109-132.

Andershed, H., Gustafson, S. B., Kerr, M., & Stattin, H. (2002). The usefulness of self‐reported psychopathy‐like traits in the study of antisocial behaviour among non‐referred adolescents. European Journal of Personality16(5), 383-402.

Tremblay, P., Bouchard, M., & Petit, S. (2009). The size and influence of a criminal organization: A criminal achievement perspective. Global Crime10(1-2), 24-40.

Cooke, D. J., & Michie, C. (2001). Refining the construct of psychopathy: towards a hierarchical model. Psychological Assessment13(2), 171.

Cooke, D. J., Hart, S. D., Logan, C., & Michie, C. (2012). Explicating the construct of psychopathy: Development and validation of a conceptual model, the Comprehensive Assessment of Psychopathic Personality (CAPP). International Journal of Forensic Mental Health11(4), 242-252.

Cooke, D. J., Michie, C., Hart, S. D., & Clark, D. (2005). Assessing psychopathy in the UK: concerns about cross-cultural generalisability. The British Journal of Psychiatry186(4), 335-341.

Cooke, D. J., Michie, C., Hart, S. D., & Clark, D. A. (2004). Reconstructing psychopathy: Clarifying the significance of antisocial and socially deviant behavior in the diagnosis of psychopathic personality disorder. Journal of Personality Disorders, 18, 337–357.

Corrado, R. R., Cohen, I. M., Glackman, W., & Odgers, C. (2003). Serious and violent young offenders’ decisions to recidivate: An assessment of five sentencing models. Crime & Delinquency49(2), 179-200.

Corrado, R. R., Odgers, C., & Cohen, I. M. (2000). Incarceration of female young offenders: Protection for whom? Canadian Journal of Criminology42(2), 189-208.

Corrado, R. R., Vincent, G. M., Hart, S. D., & Cohen, I. M. (2004). Predictive validity of the Psychopathy Checklist: Youth Version for general and violent recidivism. Behavioral Sciences & the Law22(1), 5-22.

Corrado, R., & Mathesius, J (in press). Developmental psycho-neurological research trends and their importance for reassessing key decision-making assumptions for children, adolescents, and young adults in juvenile/youth and adult criminal justice systems. Bergen Journal of Criminal Law and Criminal Justice.

Corrado, R.R. & McCuish, E.C. (2014). The contribution of mental health disorders to antisocial behavior pathways. In J. Morizot & L. Kazemian (Eds.), The development of criminal and antisocial behavior: Theoretical foundations and practical applications. New York, NY: Springer.

Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98.

Cross, T. (2008). Native Americans and juvenile justice: A hidden tragedy. Poverty and Race Research Action Council. Washington, DC. Retrieved from http://www.prrac.org.

Dawson, S., McCuish, E., Hart, S. D., & Corrado, R. R. (2012). Critical issues in the assessment of adolescent psychopathy: An illustration using two case studies. International Journal of Forensic Mental Health11(2), 63-79.

DeLisi, M. (2005). Career criminals in society. Thousand Oaks, CA: SAGE.

DeLisi, M. (2009). Psychopathy is the unified theory of crime. Youth Violence and Juvenile Justice, 7(3), 256-273.

DeLisi, M., & Piquero, A. R. (2011). New frontiers in criminal careers research, 2000–2011: A

state-of-the-art review. Journal of Criminal Justice, 39(4), 289-301.

DeLisi, M., & Vaughn, M. G. (2014). Foundation for a temperament-based theory of antisocial

behavior and criminal justice system involvement. Journal of Criminal Justice, 42(1), 10-25.

Douglas, K. S., Strand, S., Belfrage, H., Fransson, G., & Levander, S. (2005). Reliability and validity evaluation of the Psychopathy Checklist: Screening Version (PCL: SV) in Swedish correctional and forensic psychiatric samples. Assessment12(2), 145-161.

Edens, J. F., & Cahill, M. A. (2007). Psychopathy in adolescence and criminal recidivism in young adulthood longitudinal results from a multiethnic sample of youthful offenders. Assessment14(1), 57-64.

Edens, J. F., & Campbell, J. S. (2007). Identifying youths at risk for institutional misconduct: A meta-analytic investigation of the psychopathy checklist measures. Psychological Services, 4(1), 13.

Edens, J. F., Skeem, J. L., Cruise, K. R., & Cauffman, E. (2001). Assessment of “juvenile psychopathy” and its association with violence: a critical review. Behavioral Sciences & the Law19(1), 53-80.

Eggleston, E. P., Laub, J. H. & Sampson, R. J. (2004). Methodological sensitivities to latent class analysis of long-term criminal trajectories. Journal of Quantitative Criminology, 20(1), 1-25.

Farrington, D. P. (1989). Early predictors of adolescent aggression and adult violence. Violence and victims4(2), 79-100.

Farrington, D. P. (2005). The importance of child and adolescent psychopathy. Journal of Abnormal Child Psychology33(4), 489-497.

Forth, A.E., Kosson, D.S., & Hare, R. (2003). The Hare Psychopathy Checklist: Youth Version. New York, NY: Multi-Health Systems.

Forth, A.E., Kosson, D.S., & Hare, R. (2003). The Hare Psychopathy Checklist: Youth Version. New York, NY: Multi-Health Systems.

Frick, P. J., & Hare, R. D. (2001). Antisocial Process Screening Device (APSD). Toronto, ON: Multi-Health Systems.

Gadermann, A. M., Guhn, M., & Zumbo, B. D. (2012). Estimating ordinal reliability for Likert- type and ordinal item response data: A conceptual, empirical, and practical guide. Practical Assessment, Research & Evaluation, 17, 1-13.

Gottfredson, M. & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press.

Gretton, H. M., Hare, R. D., & Catchpole, R. E. (2004). Psychopathy and offending from adolescence to adulthood: a 10-year follow-up. Journal of consulting and clinical psychology72(4), 636.

Gretton, H. M., Hare, R. D., & Catchpole, R. E. (2004). Psychopathy and offending from adolescence to adulthood: a 10-year follow-up. Journal of Consulting and Clinical Psychology72(4), 636.

Hare, R. D., & Neumann, C. S. (2005). Structural models of psychopathy. Current Psychiatry Reports7(1), 57-64.

Harris, G. T., Rice, M. E., & Lalumière, M. (2001). Criminal violence: The roles of psychopathy, neurodevelopmental insults, and antisocial parenting. Criminal Justice and Behavior28(4), 402-426.

Hart, S. D., Hare, R. D., & Cox, D. N. (1995). The Hare psychopathy checklist: Screening version (PCL: SV). Multi-Health Systems, Incorporated.

Hoff, H. A., Rypdal, K., Mykletun, A., & Cooke, D. J. (2012). A prototypicality validation of the Comprehensive Assessment of Psychopathic Personality model (CAPP). Journal of Personality Disorders26(3), 414-427.

Jennings, W. G., & Reingle, J. M. (2012). On the number and shape of developmental/life-course

violence, aggression, and delinquency trajectories: A state-of-the-art review. Journal of Criminal Justice, 40(6), 472-489.

Jones, B. L., & Nagin, D. S. (2007). Advances in group-based trajectory modeling and an SAS procedure for estimating them. Sociological Methods & Research35(4), 542-571.

Jones, B. L., Nagin, D. S., & Roeder, K. (2001). A SAS procedure based on mixture models for estimating developmental trajectories. Sociological Methods & Research, 29(3), 374-393.

Kotler, J.S. & McMahon, R.J. (2010). Assessment of child and adolescent psychopathy. In R.T. Salekin & D.R. Lynam (Eds.), Handbook of child and adolescent psychopathy (pp. 79-112). New York, NY: Guilford Press.

Kreis, M. K., & Cooke, D. J. (2011). Capturing the psychopathic female: A prototypicality analysis of the Comprehensive Assessment of Psychopathic Personality (CAPP) across gender. Behavioral Sciences & the Law29(5), 634-648.

Kreis, M. K., Cooke, D. J., Michie, C., Hoff, H. A., & Logan, C. (2012). The Comprehensive Assessment of Psychopathic Personality (CAPP): content validation using prototypical analysis. Journal of Personality Disorders26(3), 402-413.

Le Blanc, M., & Frechette, M. (1989). Male criminal activity from childhood through youth: Multilevel and developmental perspective. New York, NY: Springer-Verlag.

Leistico, A. M. R., Salekin, R. T., DeCoster, J., & Rogers, R. (2008). A large-scale meta-analysis relating the Hare measures of psychopathy to antisocial conduct. Law and Human Behavior32(1), 28-45.

Livingston, M. Stewart, A., Allard, T., Ogilvie, J. (2008). Understanding juvenile offending trajectories. Australian & New Zealand Journal Of Criminology, 41(3), 345-362.

Lochman, J.E., Powell, N.P., Boxmeyer, C., Young, L., & Baden, R. (2010). Historical conception of risk subtyping among children and adolescents. In. R.T. Salekin & D.R. Lynam (Eds.), Handbook of child and adolescent psychopathy (pp. 49-78). New York, NY: Guilford Press.

Loeber, R., & Stouthamer-Loeber, M. (1998). Development of juvenile aggression and violence: Some common misconceptions and controversies. American Psychologist, 52(2), 242-259.

Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., Moffitt, T. E., Caspi, A., & Lynam, D. (2001). Male mental health problems, psychopathy, and personality traits: Key findings from the first 14 years of the Pittsburgh Youth Study. Clinical Child and Family Psychology Review4(4), 273-297.

Loeber, R., Farrington, D.P., Stouthamer-Loeber, M., & White, H.R. (2008). Introduction and key questions. In R. Loeber, D.P. Farrington, M. Stouthamer-Loeber, and H.R. White (Eds.), Violence and serious theft: Development and prediction from childhood to adulthood (pp. 3-24). New York, NY: Routledge.

Lynam, D. R. (1997). Pursuing the psychopath: capturing the fledgling psychopath in a nomological net. Journal of Abnormal Psychology106(3), 425.

McCuish, E. C., Corrado, R., Lussier, P., & Hart, S. D. (2014). Psychopathic traits and offending trajectories from early adolescence to adulthood. Journal of Criminal Justice42(1), 66-76.

Moffitt, T.E. (1993). "Life-course-persistent" and "adolescent-limited" antisocial behavior: A developmental taxonomy. Psychological Review 100(4), 674-701.

Moffitt, T. E., & Caspi, A. (2001). Childhood predictors differentiate life-course persistent and adolescence-limited antisocial pathways among males and females. Development and Psychopathology13(02), 355-375.

Monahan, J., Steadman, H. J., Silver, E., Appelbaum, P. S., Mulvey, E. P., Roth, L. H., Grisso, T., & Banks, S. (2001). Rethinking risk assessment: The MacArthur study on mental disorder and violence. New York, NY: Oxford University Press.

Murrie, D. C., & Cornell, D. G. (2000). The Millon Adolescent Clinical Inventory and psychopathy. Journal of Personality Assessment, 75(1), 110–125.

Nagin, D.S. (2005). Group-based modeling of development. Cambridge, MA: Harvard University Press.

Patterson, G. R., Forgatch, M. S., Yoerger, K. L., & Stoolmiller, M. (1998). Variables that initiate and maintain an early-onset trajectory for juvenile offending. Development and Psychopathology10(03), 531-547.

Patterson, G.R., DeBaryshe, B.D., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American Psychologist, 44, 329–335

Piquero, A. R., Blumstein, A., Brame, R., Haapanen, R., Mulvey, E. P., & Nagin, D. S. (2001). Assessing the impact of exposure time and incapacitation on longitudinal trajectories of criminal offending. Journal of Adolescent Research16(1), 54-74.

Piquero, A. R., Farrington, D. P., Fontaine, N. M., Vincent, G., Coid, J., & Ullrich, S. (2012). Childhood risk, offending trajectories, and psychopathy at age 48 years in the Cambridge Study in Delinquent Development. Psychology, Public Policy, and Law18(4), 577.

Porter, S. & Woodworth, M. (2006). Psychopathy and aggression. In C.J. Patrick (Ed.), Handbook of psychopathy (pp. 481-494). New York, NY: Guilford Press.

Ribeiro da Silva, D., Rijo, D., & Salekin, R. T. (2012). Child and adolescent psychopathy: A

state-of-the-art reflection on the construct and etiological theories. Journal of Criminal Justice, 40(4), 269-277.

Robins, L. N. (1978). Sturdy childhood predictors of adult antisocial behaviour: Replications from longitudinal studies. Psychological Medicine, 8(4), 611-622.

Salekin, R. T. (2008). Psychopathy and recidivism from mid-adolescence to young adulthood: Cumulating legal problems and limiting life opportunities. Journal of Abnormal Psychology, 117(2), 386-395.

Salekin, R. T., Rosenbaum, J., Lee, Z., & Lester, W. S. (2009). Child and adolescent psychopathy: Like a painting by Monet. Youth Violence and Juvenile Justice, 7, 239-255.

Salekin, R.T., & Lynam, D.R. (2010). Child and adolescent psychopathy: An introduction. In. R.T. Salekin & D.R. Lynam (Eds.), Handbook of child and adolescent psychopathy (pp. 1-14). New York, NY: Guilford Press.

Sandvik, A. M., Hansen, A. L., Kristensen, M. V., Johnsen, B. H., Logan, C., & Thornton, D. (2012). Assessment of psychopathy: Inter-correlations between Psychopathy Checklist Revised, Comprehensive Assessment of Psychopathic Personality–Institutional Rating Scale, and Self-Report of Psychopathy Scale–III. International Journal of Forensic Mental Health11(4), 280-288.

Schneider, A. L. (1990). Deterrence and juvenile crime. New York, NY: Springer.

Sevecke, K. & Kosson, D. S. (2010). Relationship of child and adolescent psychopathy to other forms of psychopathology. In. R.T. Salekin & D.R. Lynam (Eds.), Handbook of child and adolescent psychopathy (pp. 284-314). New York, NY: Guilford Press.

Skardhamar, T. (2010). Distinguishing facts and artifacts in group-based modeling. Criminology48(1), 295-320.

Skeem, J. L., & Cooke, D. J. (2010). Is criminal behavior a central component of psychopathy? Conceptual directions for resolving the debate. Psychological Assessment22(2), 433.

Skeem, J. L., & Mulvey, E. P. (2001). Psychopathy and community violence among civil psychiatric patients: results from the MacArthur Violence Risk Assessment Study. Journal of Consulting and Clinical Psychology69(3), 358-374.

Statistics Canada. (2013). British Columbia (Code 59) (table). National Household Survey (NHS) Aboriginal Population Profile. 2011 National Household Survey. Statistics Canada Catalogue no. 99-011-X2011007. Ottawa. Released November 13, 2013. http://www12.statcan.gc.ca/nhs-enm/2011/dp-pd/aprof/index.cfm?Lang=E (accessed December 1, 2016).

Stouthamer-Loeber, M., Loeber, R., Stallings, R., & Lacourse, E. (2008). Desistance from and persistence in offending. In Violence and serious theft: Development and prediction from childhood to adulthood (pp. 269-308). New York, NY: Routledge.

Teplin, L.A., Abram, K.M., Washburn, J.J., Welty, L.J., Hershfield, J.A., & Dulcan, M.K. (2013). The Northwestern Juvenile Project: Overview. OJJDP Bulletin. Washington, DC: US Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

van der Geest, V., Blokland, A., & Bijleveld, C. (2009). Delinquent development in a sample of high-risk youth: Shape, content, and predictors of delinquent trajectories from age 12 to 32. Journal of Research in Crime and Delinquency, 46(2), 111-143.

Vaughn, M. G., & DeLisi, M. (2008). Were Wolfgang's chronic offenders psychopaths? On the

convergent validity between psychopathy and career criminality. Journal of Criminal Justice, 36(1), 33-42.

Vaughn, M. G., DeLisi, M., Gunter, T., Fu, Q., Beaver, K. M., Perron, B. E., & Howard, M. O.

(2011). The severe 5%: A latent class analysis of the externalizing behavior spectrum in the United States. Journal of Criminal Justice, 39(1), 75-80.

Vaughn, M. G., Howard, M. O., & DeLisi, M. (2008). Psychopathic personality traits and

delinquent careers: An empirical examination. International Journal of Law and Psychiatry, 31(5), 407-416.

Viding, E., & Larsson, H. (2010). Genetics of child and adolescent psychopathy. In. R.T. Salekin & D.R. Lynam (Eds.), Handbook of child and adolescent psychopathy (pp. 284-314). New York, NY: Guilford Press.

Vincent, G. M., Odgers, C. L., McCormick, A. V., & Corrado, R. R. (2008). The PCL: YV and recidivism in male and female juveniles: A follow-up into young adulthood. International Journal of Law and Psychiatry31(3), 287-296.

Walters, G. D. (2003). Predicting institutional adjustment and recidivism with the psychopathy checklist factor scores: a meta-analysis. Law and Human Behavior, 27(5), 541-558.

Walters, G. D. (2004). The trouble with psychopathy as a general theory of crime. International Journal of Offender Therapy and Comparative Criminology, 48(2), 133-148.

Wolfgang, M.E., Figlio, R.M., & Sellin, T (1972). Delinquency in a birth cohort. Chicago, IL: University of Chicago Press.


Tables and Figures

Table 1. Descriptive information of the sample

Individual characteristics

n (%)

Mean (SD)

Demographic Characteristics

Gender

Male

262 (80.4)

Female

64 (19.6)

Ethnic origin

Caucasian

195 (60.6)

Aboriginal

81 (25.2)

Other

46 (14.3)

Measures of psychopathy

Total PCL: YV Score

Four factor model

Three factor model

Interpersonal factor

Affective factor

Lifestyle factor

Antisocial factor

Criminogenic risk factors- offender

Age of onset – alcohol use

Age of onset – drug use

Substance use versatility scale

Enrolled in school

Age of onset – skipping school

Age of onset – trouble at school

Number of different schools

Physical abuse

Sexual abuse

Age of onset – sexual activity

Positive self identity

Prosociality

Obedience

Hyper-masculinity

Fighting – weekly basis

Angers easily

Bad temper

Criminogenic risk factors- family

Family disruption scale

Left home for 24hr

Kicked out of home for 24hr

Raised by biological parents

161 (50.0)

148 (46.5)

72 (22.9)

82 (28.0)

176 (56.6)

234 (74.8)

240 (76.4)

141 (45.8)

203 (65.3)

21.19 (6.37)

19.50 (5.82)

12.41 (4.56)

3.00 (2.04)

4.36 (2.01)

5.04 (2.03)

7.09 (2.26)

11.97 (2.14)

11.75 (2.15)

4.31 (2.11)

12.29 (1.98)

9.73 (3.14)

6.31 (6.17)

13.05 (1.67)

71.16 (10.41)

19.06 (4.09)

24.53 (5.14)

19.58 (3.81)

2.76 (1.48)

Criminal career measures

Days in custody

1,166 (1,167)

Age of onset

14.09 (1.55)

Offending frequency

23.60 (18.03)

Table 2. Fit statistics for zero inflated poisson model (n = 326)

 

Offending Trajectories

Bell Shape

Slow Desisters

SRC

HRC

n (%)

93 (28.5)

91 (28.0)

62 (19.0)

80 (24.5)

Estimated model parameters

Intercept

-26.51

5.76

-3.22

-19.26

Linear

3.66

-0.39

0.46

3.50

Quadratic

-0.12

0.01

-0.01

-0.18

Model fit characteristics

Peak age

15

15

17

16

Median group probabilities

0.99

0.97

0.99

0.99

Range

0.29-1.00

0.55-1.00

0.43-1.00

0.54-1.00

Mean probability-Bell Shape

0.92 (0.16)

0.06 (0.08)

0.00 (0.00)

0.00 (0.02)

Mean probability-Slow Desisters

0.02 (0.07)

0.94 (0.08)

0.02 (0.07)

0.04 (0.11)

Mean probability-HRC

0.00 (0.01)

0.02 (0.06)

0.04 (0.09)

0.94 (0.10)

Mean probability-SRC

0.00 (0.02)

0.02 (0.08)

0.93 (0.13)

0.05 (0.09)

OCC

11.39

15.51

13.03

15.51

Criminal career parameters

Age of onset

14.46 (1.56)d

14.37 (1.37)d

14.11 (1.86)d

13.31 (1.15)abc

Total convictions

7.95 (5.65)

14.10 (8.84)acd

39.01 (16.31)ab

40.49 (11.85)ab

Total custody length (days)

380 (705)cd

682 (872)cd

1859 (987)ab

2088 (1084)ab

Note. HRC = high rate chronic, SRC = Slow rising chronic. a Significantly different from Bell Shape, b Significantly different from Slow Desisters, c significantly different from SRC, d Significantly different from HRC


Table 3. Trajectory groups and different individual-level characteristics (n = 326)

Trajectories 

Bell-Shape

Slow Desister

SRC

HRC

χ2/F, p, Φ/η2

n (%)

93 (28.5)

91 (28.0)

62 (19.0)

80 (24.5)

 

Demographic characteristics

Male

57 (62.6)

74 (79.6)

55 (88.7)

76 (95.0)

χ2 (3)=31.76, p < .001, Φ=.31

Ethnicity

Caucasian

56 (62.9)

47 (51.6)

36 (58.1)

56 (70.0)

n.s.

Aboriginal

15 (16.9)

24 (26.4)

22 (35.5)

20 (25.0)

n.s.

Other

18 (20.2)

20 (22.0)

4 (6.5)

4 (5.0)

χ2 (3)=15.7, p < .01, Φ=.22

Measures of psychopathy

Four factor model

17.71 (6.07)cd

17.97 (5.31)cd

20.36 (5.91)ab

22.58 (4.61)ab

F (3) = 14.3, p < .001, η2 = .12

Three factor model

11.55 (4.87)d

11.46 (4.12)d

12.73 (4.82)

14.24 (3.93)ab

F (3) = 7.2, p < .001, η2 = .06

Interpersonal factor

2.88 (1.98)

2.73 (1.98)

3.18 (2.11)

3.33 (2.11)

n.s.

Affective factor

4.11 (2.17)d

4.20 (2.09)d

4.20 (2.09)

5.03 (1.65)ab

F (3) = 4.2, p < .01, η2 = .04

Lifestyle factor

4.57 (2.20)d

4.60 (1.85)d

5.35 (2.00)

5.87 (1.73)ab

F (3) = 8.61, p < .001, η2 = .07

Antisocial factor

6.22 (2.39)cd

6.51 (2.30)cd

7.63 (1.76)ab

8.35 (1.68)ab

F (3) = 18.72, p < .001, η2 = .15

Criminogenic factors

Age of onset-alcohol use

12.09 (2.27)

12.24 (1.92)

12.16 (2.03)

11.38 (2.24)

n.s.

Age of onset- drug use

11.80 (2.19)

11.88 (2.27)

11.78 (2.19)

11.55 (1.92)

n.s.

Substance use versatility scale

3.86 (2.12)c

4.39 (2.15)

4.92 (1.81)a

4.29 (2.19)

F (3) = 3.0, p < .001, η2 = .03

Enrolled in school

51 (56.0)

47 (51.1)

22 (36.1)

41 (52.6)

n.s.

Age of onset- skipping school

12.58 (1.98)

12.72 (1.78)

11.80 (2.06)

11.91 (2.01)

F (3) = 3.7, p < .05, η2 = .04

Age of onset – school trouble

10.67 (3.16)d

9.79 (2.99)

9.49 (3.17)

8.90 (3.05)a

F (3) = 4.0, p < .01, η2 = .04

Number of different schools

5.63 (5.44)

6.85 (7.88)

6.50 (5.62)

6.32 (5.08)

n.s.

Physical abuse

39 (43.3)

47 (51.6)

28 (46.7)

34 (44.2)

n.s.

Sexual abuse

26 (29.2)

25 (27.5)

10 (17.2)

11 (14.5)

n.s.

Age sexually active

13.46 (1.74)d

13.10 (1.58)

13.07 (1.64)

12.54 (1.61)a

F (3) = 4.0, p < .01, η2 = .04

Positive self identity

74.15 (9.95)cd

70.64 (10.18)

69.88 (9.87)a

69.29 (11.04)a

F (3) = 3.7, p < .05, η2 = .04

Fighting – weekly basis

11 (13.4)

21 (25.6)

28 (48.3)

22 (31.0)

χ2 (3)=21.0, p < .001, Φ=.27

Angers easily

48 (53.3)

47 (54.0)

38 (62.3)

43 (58.9)

n.s.

Bad temper

60 (66.7)

65 (74.7)

52 (85.2)

57 (76.0)

n.s.

Family disruption scale

2.38 (1.76)

2.25 (1.61)

2.79 (1.50)

2.84 (1.55)

n.s.

Left home for 24hr

71 (78.9)

65 (73.9)

42 (70.0)

62 (81.6)

n.s.

Kicked out of home for 24hr

39 (44.8)

44 (50.6)

26 (43.3)

32 (43.2)

n.s.

Raised by biological parents

60 (68.2)

60 (68.2)

34 (57.6)

49 (64.5)

n.s.

Note. SRC = slow rising chronic, HFC = high rate chronic. a Significantly different from bell-shaped, b Significantly different from slow desister, c significantly different from SRC, d Significantly different from HRC. Asymptotically F distributed

Table 4. Comparison of measures of offending between age 20-28 for SRC and HRC trajectories

 

SRC (n = 62)

HRC (n = 80)

χ2/t, p, Φ/d

 

m (sd)/% (n)

m (sd)/% (n)

Severity of Offending

Days incarcerated†

1277 (641)

1101 (797)

t(133.8)= 1.38, n.s., d= .24

Total convictions

23.34 (10.58)

11.87 (7.24)

t(135)= 7.53, p < .001, d= 1.26

Violent convictions

1.78 (1.62)

1.51 (1.69)

t(135)= 0.92, n.s., d= .16

Violence specialization†

0.08 (0.08)

0.15 (0.16)

t(113.7)= -2.81, p < .01, d= .55

Seriousness metric†

63.16 (43.23)

155.44 (374.97)

t(80.81)= -2.17, p < .05., d= .35

Levene’s test of equal variance violated

Table 5. Coefficients of Risk Factors by Trajectory Group (n = 326)

 

Model 1

Model 2

Model 3

SD

SRC

HRC

SD

SRC

HRC

SRC

HRC

Covariates

OR

OR

OR

OR

OR

OR

OR

OR

OR

Demographics

‘Other’ ethnicity

2.41

0.52

0.63

2.40

0.52

0.66

2.67

0.44

0.63

Female

0.38

0.20*

0.03***

0.38

0.20*

0.03**

0.46

0.17**

0.03**

Measures of psychopathy

Four factor model

1.00

1.07

1.17**

.

.

.

.

.

.

Three factor model

.

.

.

1.02

1.08

1.17**

.

.

.

Interpersonal

.

.

.

.

.

.

0.87

1.01

0.87

Affective

.

.

.

.

.

.

1.32*

1.00

1.24

Lifestyle

.

.

.

.

.

.

0.99

1.20

1.47*

Antisocial

.

.

.

.

.

.

0.87

1.07

1.39*

Criminogenic factors

Substance use versatility

1.19

0.90

1.14

1.18

0.90

1.18

1.16

0.88

Age of onset- skip school

1.15

0.96

0.98

1.16

0.96

0.96

1.18

0.98

1.03

Age of onset- trouble

0.91

0.92

0.91

0.91

0.91

0.90

0.88

0.92

0.91

Age sexually active

1.06

1.34

1.11

1.08

1.32

1.05

1.04

1.38

1.21

Positive self identity

1.01

1.00

1.00

1.01

1.00

1.00

1.01

1.00

1.02

Fighting – weekly basis

1.60

3.53*

0.68

1.54

3.73*

0.87

1.65

3.95*

0.81

Model Fit

-2LL = 448.66, χ2= 79.5, df = 27, p<.001

-2LL = 456.39, χ2= 71.8, df= 27, p<.001

-2LL = 428.66, χ2= 99.0,df = 36, p<.001

 Note: Bell-shape trajectory group is reference category. SD = slow desister. SRC = slow rising chronic. HRC = high rate chronic

* p < .05, ** p < .01, *** p < .001. All significant OR do not contain ‘1’ based on 95% CIs.

[CHART]

Figure 1. Offending trajectories from age 12 to 28

Comments
0
comment
No comments here
Why not start the discussion?