Using data on participants born around 100 years ago, Laub and Sampson asserted that early developmental risk factors are not informative of social outcomes in adulthood. More specifically, they claimed that early risk factors were not informative of adulthood ...
Purpose: Using data on participants born around 100 years ago, Laub and Sampson asserted that early developmental risk factors are not informative of social outcomes in adulthood. More specifically, they claimed that early risk factors were not informative of adulthood informal social control, persistent offending, and the degree to which informal social control protects against later offending. We offer a contemporary comparison point to these claims that carry theoretical implications for developmental and life-course criminology. Methods: The current study used prospective longitudinal data on 518 male and female participants from the Incarcerated Serious and Violent Young Offender Study. Early developmental risk for persistent offending was measured via the Psychopathy Checklist: Youth Version (PCL:YV). Informal social control was measured in emerging adulthood via the Community Risk Needs Assessment (CRNA). Offending trajectories were measured for an average of 15 years following participants' CRNA ratings.
Results: PCL:YV scores were negatively associated with informal social control (selection effects), positively associated with persistent offending, and negatively related to lower rates of offending, even when informal social control was high (treatment effect heterogeneity). Conclusions: Findings align with Elder's conceptualization of life-course theory that acknowledges the potential for individual, social, and macro-level factors to shape human development.
Keywords: Developmental criminology, informal social control, life-course criminology, offending persistence, selection effects, treatment effect heterogeneity
Elder’s (1975, 1994) life-course theory proposed that the interrelationship between individual-level characteristics and social environment early in the life-course was informative of self-selection into adult social roles. Sampson and Laub (1993) borrowed this principle and others from Elder’s theory to develop life-course criminology as a framework for developing theories regarding offending and desistance. While we appreciate Elder’s seminal contribution to life-course research, Laub and Sampson’s interpretation of Elder’s life-course theory principles is the focus of the current study. Sampson and Laub’s (1993) age-graded theory of informal social control is the quintessential theory to emerge from the life-course criminology paradigm (Laub et al., 2018). Initially, this theory aligned with developmental principles embedded in Elder’s life-course theory. For example, Sampson and Laub (1993) acknowledged the importance of early development for various outcomes in adulthood. Their key assertion was that informal social controls in adulthood (e.g., employment, parenthood, marriage) served as turning points by providing stake-in-conformity and altering routine activities in ways that influenced desistance. In line with the age-graded theory, studies have reported a positive relationship between informal social control and desistance (Blokland & Nieuwbeerta, 2005; Laub et al., 1998; Sampson et al., 2006; Savolainen, 2009). However, motivated by new analyses from the Unraveling Juvenile Delinquency (UJD) Study, Laub and Sampson (2003) revised their theory in ways that refuted developmental principles that linked early development to adult outcomes. Notably, contrary to an emphasis on the relationship between early development and persistent offending (e.g., Moffitt, 1993), Sampson and Laub (2003) argued that “differences in adult criminal trajectories cannot be predicted from childhood” (p. 5881).
Historical context has implications for the degree of association between early developmental risk factors and offending (e.g., Nilsson et al., 2017; Van Hofer, 2014). Laub and Sampson’s (2003) revisions to their theory were based on analyses of data from a unique period (i.e., the first half of the 20th century) and may not align with the circumstances of youth adjudicated by contemporary justice systems (see Paternoster et al., 2015). Given the prominence of the age-graded theory of informal social control in contemporary criminology (Cohn et al., 2017), a contemporary comparison point to findings from the UJD Study and associated revisions to the age-graded theory are needed. The current study uses longitudinal data from British Columbia Canada on 518 boys and girls followed through age 40 as part of the Incarcerated Serious and Violent Young Offender Study (ISVYOS). We address two conceptual critiques raised by others regarding Laub and Sampson’s (2003) revised age-graded theory of informal social control. One of these critiques is that the theory fails to explain its assumption that informal social controls are a normative occurrence in the lives of persons from difficult circumstances who would seem to have limited resources and opportunities for positive life events (see Thomas et al., 2021, 2022). A second critique concerns Laub and Sampson’s (2003) assertion that desistance occurs by default and therefore informal social control influences desistance even for persons from difficult circumstances and who repeatedly engage in crime (see Paternoster et al., 2015). Overall, Laub and Sampson (2003) argued that informal social control was a normative part of the life-course, that turning points were structurally-induced, and that desistance occurred by default. Each of these assertions is inconsistent with descriptions of cumulative disadvantage included in earlier versions of their theory (Sampson & Laub, 1997). If informal social control is normative and structurally-induced, then there is no longer room in Laub and Sampson’s theory for notions of structurally-constrained life chances.
We are not responding to claims made in the initial version of the age-graded theory of informal social control. Rather, we focus on three of Laub and Sampson’s (2003) revisions to this theory that refute principles of developmental perspectives. The first developmental principle they rejected concerns selection effects, which describe how individuals tend to self-select into social environments that align with their personality, values, and beliefs (Caspi & Herbener, 1990; Caspi et al., 2005; Wright et al., 1999). Contrasting this viewpoint, Laub and Sampson (2003) asserted that informal social controls like marriage and employment were a normative part of the life course for all persons, regardless of personality, values, and beliefs, because America’s social structure funneled persons into adult roles even without their conscious awareness. Others have raised concerns that this type of assertion fails to explain how positive adult roles occur for high-risk persons from extremely marginalized circumstances (e.g., Hirschi & Gottfredson, 1995; Paternoster et al., 2015; Thomas et al., 2021; Wright et al., 2001). Their theoretical assertion is also contrasted by empirical research on assortative mating that demonstrates that an individual’s personality traits influence the nature of adult roles like marriage (e.g., Krueger et al., 1998). There is a lack of research on whether selection effects apply to contemporary youth involved in serious offenses who are followed through adulthood.
The second developmental principle Laub and Sampson (2003) rejected relates to the importance of early development to persistent offending (e.g., Moffitt, 1993). Based on their analyses of the UJD Study, Sampson and Laub (2003) concluded that they had “failed to find convincing evidence that a life-course-persistent group can be prospectively or even retrospectively identified based on theoretical risk factors at the individual level… adult trajectories of offending among former delinquents cannot be reduced to the past” (p. 588)2. An important caveat to this conclusion is that Sampson and Laub’s analyses were underpowered. Their “high rate chronic” trajectory included only 11 participants when followed through age 32 (Laub et al., 1998) and included as few as five participants when followed through age 70 (Sampson & Laub, 2003). Further, the Gluecks’ data were collected during an era where individual-level risk factors were measured subjectively and unreliably (Cronbach & Meehl, 1955). Sampson and Laub (2003) examined the relationship between early risk factors and adult offending using unreliable measures of psychopathology (e.g., Rorschach tests measuring self-control; see Hirschi & Gottfredson, 1995; Glueck, 1960) and measures irrelevant to the prediction of persistent offending (see Robins, 2005). Improvements to theories of measurement, construct validation, and risk and needs assessment of individual-level risk factors (e.g., Hart et al., 2016; Hoge, 2002; Olver et al., 2009) provides an opportunity to examine Sampson and Laub’s assertions through a new lens.
The third developmental principle Laub and Sampson (20030 rejected relates to treatment effect heterogeneity, sometimes referred to as life-course interdependence, which acknowledges that subgroups who vary in terms of their early developmental history or other pre-existing characteristics will also vary in the degree to which informal social control influences desistance (Nguyen & Loughran, 2018). Such subgroups are often distinguished on the basis of their criminal propensity, which is represented by stable traits that emerge early in the life-course and are a common cause of both a negative social environment (i.e., self-selection) and persistent offending (Nagin & Paternoster, 2000). Opposite this principle, Laub and Sampson (2003, pp. 278-279) suggested that desistance occurred by default because informal social control influenced behavioral change even for subgroups with a particularly negative early developmental history (also see Laub et al., 2018; Sampson & Laub 2016). Thus, from Laub and Sampson’s (2003) perspective, subgroups of individuals with similar developmental histories but who differ in level of informal social control should also vary in their probability of desistance. Others (see Hirschi & Gottfredson, 1995; Paternoster et al., 2015; Polaschek, 2019) have raised concerns that Laub and Sampson (2003) fail to explain why persons who are unmotivated to change and engage in serious offenses would respond to adult roles in the same way as persons who are motivated to change and engage in minor offenses. Support for (Blokland & Nieuwbeerta, 2005; Kubrin & Stewart, 2006; Reisig et al., 2007; Wang et al., 2014) versus against (Bersani et al., 2009; Doherty, 2006; King et al., 2007; Wright et al., 2001) treatment effect heterogeneity varies by sampling strategy. There appears to be greater evidence of treatment effect heterogeneity in justice system samples (e.g., Wang et al., 2014). Individuals who are at a higher risk of recidivism (e.g., higher criminal propensity) tend to show limited investment in positive long-term intimate relationships and are characterized by poor performance and experience limited satisfaction when it comes to employment (e.g., Andrews et al., 2006; Wormith et al., 2007). However, there remains a lack of research using reliable measures of criminal propensity to evaluate treatment effect heterogeneity over the long-term (Nguyen & Loughran, 2018).
Population heterogeneity theoretical perspectives use the concept of criminal propensity to describe early time-stable between-person differences in proneness to offending, selection effects, and treatment effect heterogeneity (Nagin & Paternoster, 2000). However, population heterogeneity perspectives are unclear regarding how propensity should be measured (Caspi et al., 1994; Nagin & Paternoster, 2000). For example, Moffitt (1993) suggested that early neuropsychological deficits were informative of cumulative disadvantage resulting in a lack of exposure to opportunities for positive adult roles, persistent offending, and a lack of responsivity to social roles. However, Moffitt did not specify which neuropsychological deficits were most informative of adult outcomes. Although low self-control, low constraint, negative emotionality, and cognitive skills have been proposed and examined as measures of neuropsychological deficits (e.g., Beaver et al., 2010; Loeber et al., 2008), psychopathy may be a more appropriate indicator of criminal propensity because it meets Caspi et al.’s (1994) recommendation to investigate stable, multidimensional personality traits.
Psychopathy is an important predictor of recidivism (Braga et al., 2023), with effect sizes resembling the strength of the relationship between heart surgery and angina pain (Hart, 1998). Some suggested that the same people comprise the small group of persons responsible for the majority of all offending and the small group of persons presenting with psychopathy traits (Vaughn & DeLisi, 2008). Psychopathy is believed to relate to long-term patterns of offending through persistent and pervasive negative interactions with others (i.e., interpersonal traits), a lack of attachment or caring needed to help deter offending and respond positively to treatment (i.e., affective traits; see Olver et al., 2013), and a level of sensation-seeking, impulsivity, and irresponsibility (i.e., behavioral traits) that increasing the likelihood of acting upon opportunities to engage in crime (Corrado et al., 2015). Longitudinal studies have identified a positive relationship between psychopathy and persistent offending (McCuish et al., 2021). Compared to low self-control, psychopathy is especially helpful in discriminating membership in desistance-based offending trajectories (Altikriti et al., 2020). However, studies that examine the intersection between psychopathy, informal social control, and continued offending are missing from this literature. It thus remains unclear as to whether Laub and Sampson’s (2003) assertions that informal social control is a normative part of the life course and leads to desistance by default holds true for persons with psychopathy traits.
Regarding selection effects, studies of adults show that psychopathy is negatively associated with positive sources of informal social control, including employment, education, friendships, and intimate partner relationships (Eisenbarth et al., 2022; Hemphälä & Hodgins, 2014; Herpers et al., 2016; Lee & Kim, 2022; Ploe et al., 2022; Seto & Davis, 2021). Research on assortative mating indicates that persons with psychopathy traits tend to become involved in relationships with persons with similar personality traits (Kardum et al., 2017) and that this can increase the likelihood of intimate partner violence (e.g., Forth et al., 2022). However, most of this research is cross-sectional and from the perspective of persons who have been victimized. Looking specifically within adolescence among participants from the Pathways to Desistance Study, Lee and Kim (2022) found a positive relationship between psychopathy traits and various negative social outcomes, including associations with criminogenic peers. A prospective relationship between the early development of psychopathy traits and adult social outcomes would align with Moffitt’s (1993) hypotheses regarding self-selection and cumulative disadvantage.
Regarding treatment effect heterogeneity, there is indirect evidence in the literature to suggest that psychopathy traits represent the type of pre-existing difference in early developmental history that decreases the likelihood of positive responses to informal social control (Nguyen & Loughran, 2018). For example, whereas Laub and Sampson (2003) suggested that informal social control would lead to desistance because individuals would not want to jeopardize their social roles, persons with psychopathy traits disregard, or even mock, social norms and the opinions of others (Logan & Johnstone, 2010) and therefore may be less concerned with whether their behavior jeopardizes their social roles and relationships. Whereas Laub and Sampson (2003) suggested that social roles would structure a person’s time in ways that reduce offending, persons with psychopathy traits tend to have histories of using relationships and positions of power to victimize others through aggressive, domineering, and manipulative behavior (Boddy, 2014; Forth et al., 2033; Mathieu & Babiak, 2016). Further, whereas Laub and Sampson (2003) suggested that adult roles influence identity changes, identity change in adult roles may be unlikely for persons with psychopathy traits given that such traits include an uncaring and unemphatic attachment style, cognitive inflexibility, and sense of entitlement, even in stable relationships with prosocial persons (Forth et al., 2022). Whereas Laub and Sampson (2003) suggested that new social roles increase time spent with capable guardians who deter criminal behavior, because psychopathy is negatively associated with perceptions of risk (Altikriti et al., 2021), persons with psychopathy traits may be less responsive to capable guardians. In sum, although empirically untested, a lack of responsivity to informal social control among persons with psychopathy traits would challenge Laub and Sampson’s (2003) assertion that desistance occurs by default.
The current study is a prospective longitudinal examination of whether psychopathy traits in adolescence negatively impact both the development of informal social control in emerging adulthood and the likelihood that these informal social controls influence lower levels of offending. We address sources of disagreement between two theoretical perspectives. On the one hand, Laub and Sampson’s (2003) theoretical perspective portrays informal social control and desistance as a normative part of the life course, even for persons involved in serious offenses, and therefore early development is not informative of adult outcomes. As noted by Paternoster et al. (2015), Laub and Sampson’s (2003) theory endorses Becker’s (1964) structuralist perspective that social systems can be used to “coerce people into behaving as we want them to”, seemingly with no interest in developing “deep and lasting interests” of the person (pp. 52-53). Polaschek (2019) suggested that this version of life-course criminology seems to discount the importance of early intervention and correctional treatment programming. In contrast, developmental theories view early development as important to adult outcomes and therefore early treatment and prevention matters (e.g., Kazemian et al., 2019; Morizot & Kazemian, 2014). From this perspective, informal social controls are not normative and desistance does not happen by default because factors like psychopathy have implications for the degree to which persons will be open and motivated to change, exposed to positive social environments, and engaged in treatment/intervention (e.g., Olver et al., 2013; Polaschek & Ross, 2010; Polaschek & Daly, 2013) Clarifying the accuracy of these different theoretical perspectives is important given gaps in nuanced, theory-informed, and evidence-based strategies for promoting desistance among correctional populations (Nguyen & Loughran, 2018; Thomas et al., 2022).
We used prospective longitudinal data from the ISVYOS (n = 518) to examine three claims made by Laub and Sampson (2003): (1) informal social controls are a normative part of the life course (i.e., rejection of selection effects), (2) early development is not informative of persistent offending, and (3) positive responses to informal social control occur regardless of developmental history and thus desistance happens by default (i.e., rejection of treatment effect heterogeneity). To investigate selection effects, we examined whether psychopathy traits in childhood/adolescence were associated with self-selection into informal social controls in emerging adulthood. To investigate questions about the role of early development in predicting persistent offending, controlling for informal social control, we examined whether psychopathy was associated with offending trajectories measured from emerging adulthood through age 40. To investigate treatment effect heterogeneity, we examined whether psychopathy traits were resistant to the influence of informal social control on lower rates of offending.
The Ministry of Children and Family Development is responsible for the care of incarcerated youth in the province of British Columbia, Canada, and gave the ISVYOS consent to enter its custody centers to recruit participants (see McCuish et al. (2022) for a cohort profile that describes sampling and procedures in greater detail). Recruitment took place between 1998-20113. Among the 1,719 youth recruited, 1,640 participated in an intake interview conducted by ISVYOS research assistants (RAs) and 535 of these participants subsequently received a semi-structured interview conducted by RAs as part of procedures for scoring the Psychopathy Checklist: Youth Version (PCL:YV). A completed intake interview was the only criteria for determining eligibility to receive an interview to rate the PCL:YV. The discrepancy between the number of participants with an intake interview and the number of participants with a PCL:YV rating is because (1) the PCL:YV was not formalized until 2003 (Forth et al., 2003), (2) the PCL:YV required time, specialized training, and funding that was not available over the full 20 years of the study, and (3) the focus of the study changed over time and thus completing PCL:YV ratings was not always prioritized (see McCuish et al., 2022). Youth who spent more time in custody would have more opportunity to participate in the study and therefore may have been more likely to receive a PCL:YV rating.
Interviews occurred in a private room away from other youth and custody staff. To obtain assent4, participants were read and given a copy of an information sheet explaining the purpose of the study, how data would be collected, that they could withdraw at any time, and that information provided during the interview would be kept confidential unless they made a direct threat against themselves or someone else. Research Assistants (RAs) explained that participating in the study would not affect a participant’s stay at the custody center. The structured intake interview was approximately 90 minutes and the PCL:YV interview lasted up to three hours. Interviews were typically conducted across multiple sessions. Data on participant outcomes in adulthood, including justice system involvement and informal social controls, were collected through the Corrections Network (CORNET), a client management software program that allows justice system professionals to record and store information about their clients. CORNET allowed for the measurement of offending and other data (e.g., informal social controls) beginning at age 12 for all participants (i.e., the age of criminal responsibility in Canada) and extending through adulthood. Offending outcomes were measured to an average age of 35.69 (SD = 4.31) for this sample. Seventeen of the 535 participants with a PCL:YV rating were excluded due to CORNET file information being sealed or missing.
Descriptive statistics for the measures described below (see Table 1) pertain to the current study’s sample of 518 participants who were born between 1979-1998 and who were an average age of 16.09 (SD = 1.34) at the time of the intake interview. For all participants, psychopathy was measured in adolescence (Mean age = 16.47; SD = 1.39), informal social control was measured in emerging adulthood (Mean age = 19.33; SD = 1.71), and adult offending outcomes were measured at each year of age beginning after the measurement of informal social control (Mean age = 20.33; SD = 1.71) and extending for an average of 15.36 years (SD = 4.53), meaning that participants were about 35 years old at the end of the follow-up period. The mi chained command in Stata (StataCorp, 2019) was used to impute values on individual items before constructing the PCL:YV, informal social control, and family adversity scales described below. Unlike full information maximum likelihood, mi chained does not assume that data are missing completely at random (e.g., Shin et al., 2017). About 75% of the sample had complete data across all measures. Participants with missing items on the CRNA were involved in lower rates of offending in adulthood. It is possible that their level of justice system involvement meant that probation officers had less information available to rate the CRNA.
--Insert Table 1 about Here--
Control variables. The sample’s demographic characteristics resemble official statistics regarding the gender and ethnic composition of youth in custody in British Columbia (Malakieh, 2017). Most participants self-reported as male (82.6%) and as White (59.7%). Indigenous youth were overrepresented in the sample (25.4%) relative to the general population. The remainder of participants (14.9%) self-identified as a member of a non-Indigenous minority group (e.g., Black, East Asian, South-East Asian, Hispanic). Analyses controlled for family adversity given its relevance in theories of persistent offending (Moffitt, 1993) and to the development of psychopathy (Farrington & Bergstrom, 2018). Family adversity was measured during the intake interview by asking participants if any family member had a: (1) problem with alcohol, (2) problem with drugs, (3) history of physical abuse, (4) history of sexual abuse, (5) criminal record, or (6) mental illness. These dichotomous items were summed to create a scale (M = 2.88; SD = 1.61; tetrachoric ordinal alpha = 0.72). Given the age-graded nature of informal social control, we controlled for participant age at the time informal social control was measured (M = 19.33; SD = 1.71). Finally, we controlled for number of youth convictions5 (M = 12.74, SD = 9.00) to address Sampson and Laub’s (2005b) argument that “conditioned on delinquency or crime… we cannot predict long-term trajectories of offending” (p. 75).
Psychopathy. Psychopathy was measured using the PCL:YV (Forth et al. 2003), which is a 20-item symptom rating scale. PCL:YV items reflect interpersonal (e.g., impression management, manipulation), affective (e.g., shallow affect, lack of remorse), lifestyle (e.g., stimulation seeking, parasitic orientation), and antisocial behavior (e.g., early behavior problems, poor anger control) domains. Items are scored on a three-point scale and ratings are determined by assessing the extent to which an item is stable across developmental history and social context (Forth et al., 2003). For example, to score high on the PCL:YV item reflecting a lack of remorse requires that the person lack remorse in their negative interactions across a range of relationships and social domains (e.g., family, friends, employers, co-workers) examined over time. Raters therefore considered file and self-report information from the full life course. Twenty-eight graduate and undergraduate RAs underwent training from one of the developers of the PCL. Interrater reliability was evaluated by assigning six RAs to three pairs. Each pair rated 10 different participants. Interrater reliability was excellent (intraclass correlation coefficient = 0.92). Our main analyses focused on PCL:YV total scores (M = 21.19; SD = 6.50; polychoric ordinal alpha = 0.84). We also report findings using the three factor model that excludes the antisocial facet to avoid tautological concerns with past offending predicting future offending.
Adult informal social control. Dr. Bill Glackman developed the Community Risk-Need Assessment (CRNA) as a structured assessment tool to be completed every six months by British Columbia justice system professionals as part of mandated case management practices (Gress, 2010). The CRNA reflects the risk-need-responsivity model and the central eight risk factors (Bonta & Andrews, 2007). Probation officers in British Columbia are well-educated and show consistency and accuracy in applying justice system principles and policies to decision-making across a diverse set of clients (Corrado et al., 2010). Although the ISVYOS did not have direct self-report measures of informal social control, justice system professionals rate the CRNA using information reported to them by the participant. They also consult with persons familiar with the participant (e.g., other professionals, family members, intimate partners) and reference past reports (e.g., psychiatric assessments, pre-sentence reports). Prior studies identified strong concordance between self- and professional-reports of informal social controls as well as predictive validity between professional reports and client outcomes (Polaschek et al., 2022). The justice system professionals who rated the CRNA were blind to participants’ PCL:YV scores.
The CRNA informal social control scale includes five items scored on a four-point scale: family relationships, intimate relationships, living arrangements, employment, and academic/vocational skills (Lussier et al., 2016). These items are not scored based on answers to a single, specific question. Rather, items are scored holistically by using information reported by the participant, collateral contacts (e.g., family members), and past reports (e.g., pre-sentence reports). Items are scored based on their influence on reducing offending, which aligns with recent conceptualizations of informal social control turning points (Laub & Sampson, 2003; Nguyen & Loughran, 2018). Family and intimate partner relationships reflect the extent to which parents, siblings, and marital partners are prosocial versus antisocial sources of support6. The measure of living arrangements reflects stability and security of housing needs, which can contribute to investments in community that strengthen ties to social institutions (Sampson & Laub, 1990) and is related to desistance (Mustaine & Tewksbury, 2011). The employment item more directly measures the type of adult role that Laub and Sampson (2003) emphasized was important to desistance. Academic and vocational skills reflect barriers to obtaining employment. Sampson and Laub (1990) emphasized that vocational training was critical for developing social networks and stakes-in-conformity during young adulthood, which is the age-period in which informal social control was measured in the current study. Our analyses focused on the CRNA informal social control scale rather than individual items given that acquiring informal social control is a gradual and cumulative process (Savolainen, 2009). We focused on participants’ first CRNA in emerging adulthood (ages 18-25), which occurred at an average age of 19.33 (SD = 1.71). The items were reverse coded so that higher scores reflected positive levels of informal social control (M = 7.43; SD = 5.41; Cronbach’s alpha = 0.85). Approximately 75% of the sample had complete data for all five items. Missingness was imputed at the item-level rather than imputing scale scores.
Adult justice system involvement. CORNET provides information regarding official justice system involvement within the province of British Columbia7, including each offense for which a person receives a conviction and the date of that conviction. While cross-sectional self-report data are helpful for measuring undetected offenses, self-report data can be unreliable in studies using repeated measures of offending (Kirk, 2006; Lauritsen, 1998). Official data are helpful for measuring more serious offenses that tend to not be self-reported, even in confidential settings (McCuish et al, 2021). Official data are also useful for examining long-term patterns of offending as individuals tend to underestimate their age at last offense (Farrington et al., 2014). CORNET included data on whether a participant was deceased (n = 48; 9.3%) and whether they emigrated outside of British Columbia (n = 42; 8.1%). The date of emigration or date the participant became deceased was used to represent the end of the follow-up period. Practitioners’ ratings on the CRNA may be influenced by their client’s past justice system involvement. To establish temporal order between the CRNA and offending, convictions were measured at each year of age, beginning at the next year of age following the CRNA rating (M = 20.33; SD = 1.71) and continuing up to, on average, age 35.69 (SD = 4.31). Participants averaged 13.45 convictions (SD = 15.62) over this approximately 15-year period that represents data collected through the year 2022. To account for exposure time, data from CORNET pertaining to movements in and out of custody were used to identify the number of days spent incarcerated at each year of age. Participants averaged 81.3% of their follow-up period in the community. Compared to the United States, incarceration is used more sparingly in Canada (Tonry, 2013), which makes it possible to incur a greater number of convictions.
There are three phases to the analytic strategy that correspond with Laub and Sampson’s (2003) rejection of three developmental principles. First, we revisited Laub and Sampson’s assertion that informal social control is a normative part of the life-course (i.e., rejection of selection effects) by using ordinary least squares regression analyses to examine whether PCL:YV scores measured from data in childhood/adolescence were informative of scores on the CRNA informal social control scale measured in emerging adulthood. Second, we revisited Sampson and Laub’s (2005b) conclusion that “we cannot predict long-term trajectories of offending… the childhood (developmental?) paradigm cannot provide the answer” (p. 75) by examining in multinomial logistic regression analyses whether PCL:YV scores were associated with conviction trajectories measured for an average of 15.36 years (SD = 4.53) following participants’ CRNA rating. Third, to revisit debate for (e.g., Blokland & Nieuwbeerta, 2005) versus against (e.g., Laub & Sampson, 2003) treatment effect heterogeneity, we investigated the interrelationship between psychopathy and informal social control in terms of its implications for future offending. To evaluate Laub and Sampson’s perspective that a person’s early developmental history does not impact the influence of informal social control on desistance, we examined whether higher scores on the CRNA informal social control scale moderated the relationship between PCL:YV total scores and conviction trajectories. To evaluate the treatment heterogeneity perspective that subgroups who differ in their developmental history will also differ in the extent to which informal social control influences desistance (Nguyen & Loughran, 2018), we examined whether participants (n = 67) who scored high on the PCL:YV (i.e., 30+; see Hare, 2003) did not differ in their probability of trajectory assignment despite differences in scores on the CRNA informal social control scale.
Semi-parametric group-based modeling (SPGM; Nagin, 2005) was used to model conviction trajectories. Trajectory group assignment tends to not vary depending on whether self-report or official data are used (Fontaine et al., 2014). SPGM is useful under the assumption that (a) representing an entire population by an average trajectory poorly accounts for the heterogeneity of individual offending patterns but (b) measuring individual trajectories for an entire sample requires an inordinate number of parameters. As a compromise between fit and parsimony, SPGM uses finite mixture modeling to identify within an unknown continuous distribution the number of aggregate trajectories that best fit the data (Nagin, 2005). Criticisms regarding SPGM revolve around its appropriateness for testing taxonomic theories and for making conclusions about the number of ‘groups’ underlying the aggregate age-crime curve (e.g., Sampson & Laub, 2005c). These criticisms are not relevant to the current study as SPGM is used only as a data reduction device to capture heterogeneity in offending trajectories. SPGM is not the only way to model longitudinal patterns of offending, but regression analyses do not capture the time component of offending persistence and analyses like growth curve modeling assume linear growth, which contrasts with what is known about the age-crime curve (Farrington, 1986). To assess the robustness of findings, models predicting conviction trajectories were replicated using generalized estimating equations (GEE) and are presented in the Supplemental Materials.
We modeled trajectories of the number of convictions incurred at each year of age following participants’ CRNA rating and continuing through age 40. Each wave represents a single year of age, but because the CRNA was completed when participants were different ages, participants at wave 1 of the follow-up period ranged in age from 18-25. This was accounted for by controlling for age at CRNA rating. Time incarcerated was also directly accounted for through its inclusion as an exposure variable in the trajectory models. All models specified a zero-inflated Poisson distribution with quadratic functional form. We retained a four-group solution (see Figure 1) based on consideration of the following: (1) Bayesian Information Criteria values, (2) accuracy of trajectory assignment as indicated by posterior probabilities of correct classification, and odds of correct classification values, (3) parsimony (i.e., absence of similarly-shaped trajectories), and (4) absence of trajectories comprising less than five percent of the sample. Average posterior probabilities were greater than 0.90 and odds of correct classification values were greater than five for each trajectory, which represents excellent classification accuracy (Nagin, 2005). The length of the follow-up period did not vary across trajectories. Sharp wave-over-wave changes for the High Rate trajectory (e.g., Waves 4, 7, 19) resulted from large wave-to-wave fluctuations in exposure time due to periods of incarceration. The High Rate trajectory peaked at six expected convictions, which is over twice the peak in arrests for the “high rate chronic” trajectory from Sampson and Laub’s (2003) analyses. Over the follow-up period, the High Rate trajectory (15.3% of the sample) averaged 38.46 (SD = 16.37) convictions, the Slow Increasing trajectory (16.4%) averaged 11.86 (SD = 9.47) convictions, the Slow Decreasing trajectory (25.1%) averaged 19.41 (SD = 9.16) convictions, and the Low Rate trajectory (43.2%) averaged 1.61 (SD = 2.12) convictions. Despite a decline in expected convictions, through waves 15-20, the High Rate and Slow Increasing trajectories averaged a conviction per year of age.
--Insert Figure 1 about Here--
Model 1 of Table 2 shows that, controlling for family adversity, demographic characteristics, convictions between ages 12-17, and age at CRNA rating, PCL:YV total scores based on traits in childhood/adolescence were associated with significantly lower scores on the CRNA informal social control scale measured in emerging adulthood. The significant and negative relationship between youth convictions and adult informal social control also reiterates that experiences in adolescence influences selection into adult roles. This model accounted for 24.2% of the variance in the CRNA informal social control scale. Total scores on the PCL:YV three factor model, which is calculated as the sum of the interpersonal, affective, and lifestyle facets, were also significantly and negatively related to the CRNA informal social control scale. Findings were also consistent when using the last informal social control scale from emerging adulthood, which was approximately three years after the first assessment (M = 22.92, SD = 2.29). Thus, the relationship between PCL:YV scores and informal social control was not contingent on the timing of the measurement of informal social control. Model 2 examined the interaction between PCL:YV scores and family adversity to assess Moffitt’s (1993) cumulative disadvantage hypothesis that the influence of early neuropsychological deficits on negative outcomes in adulthood is exacerbated by a person’s social environment. The interaction term in Model 2 was not significant, which implies that psychopathy traits in adolescence prospectively predicted lower levels of informal social control in adulthood irrespective of adolescence exposure to an adverse family environment.
--Insert Table 2 about Here--
All variables from Table 2 plus the CRNA informal social control scale were included in a series of multinomial logistic regression analyses with conviction trajectories as the outcome of interest. Model 1 of Table 3 shows that an increase in PCL:YV totals scores significantly increased the relative risk of assignment to each of the Slow Decreasing, Slow Increasing, and High Rate conviction trajectories compared to the Low Rate conviction trajectory. Although the relative risk ratios presented in Table 3 may seem small, it is important to consider the wide range of possible PCL:YV scores (0-40). When converting PCL:YV scores to z-scores, a one standard deviation increase from the mean PCL:YV score increased the relative risk of being assigned to the Slow Decreasing trajectory versus the Low Rate trajectory by 165%, by 149% when comparing the High Rate trajectory to the Low Rate trajectory, and by 133% when comparing the Slow Increasing trajectory to the Low Rate trajectory. The average marginal effect of PCL:YV scores on assignment to the Low Rate conviction trajectory is -.011 (z = -3.65; p < .001), meaning that a one-unit increase in PCL:YV test score decreased the relative risk of assignment to the Low Rate conviction trajectory by 1.1%. The average marginal effect of PCL:YV scores on assignment to the Slow Decreasing conviction trajectory is .007 (z = -3.65; p < .001). Marginal effects were not significant for the other two conviction trajectories. Figures S3 and S4 of Supplemental Materials show that the relationship between PCL:YV scores and trajectory assignment was not moderated by the family adversity scale nor youth convictions. Thus, the relationship between PCL:YV scores and persistent offending through age 40 was not contingent on a person’s social environment/justice system experiences early in the life-course, which contrasts with Sampson and Laub’s (2005b) assertion that “conditioned on delinquency or crime… we cannot predict long-term trajectories of offending” (p. 75). Findings also did not change when using scores on the PCL:YV three factor model that excludes items measuring prior antisocial behavior.
Although PCL:YV scores were our main focus, it is also noteworthy that the CRNA informal social control scale was significantly related to conviction trajectories. A one-unit increase on the CRNA informal social control scale significantly decreased the relative risk of assignment to the Slow Decreasing and High Rate trajectories compared to the Low Rate conviction trajectory. The average marginal effect of the informal social control scale on the Low Rate conviction trajectory is .018 (z = 4.98; p < .001), meaning that a one-unit increase in informal social control increased the relative risk of assignment to the Low Rate conviction trajectory by 1.8%. The average marginal effect of the informal social control scale on assignment to the Slow Decreasing trajectory is -.014 (z = -3.46; p < .01). Marginal effects for the Slow Increasing and High Rate conviction trajectories were not significant. Findings were consistent when using the CRNA informal social control scale that was measured during a later period of emerging adulthood
--Insert Table 3 about Here--
Contrary to Laub and Sampson’s description of desistance by default, higher scores on the CRNA informal social control scale did not moderate the negative relationship between PCL:YV scores and assignment to the Low Rate conviction trajectory compared to the other three trajectories (see Model 2 of Table 3). Figure 2 helps visualize this relationship. Panel A shows that regardless of their score on the CRNA informal social control scale, participants with a score of at least 30 on the PCL:YV (n = 67), which is considered to represent a high score on the PCL (Hare, 2003), did not differ in their probability of assignment to the Low Rate conviction trajectory. For example, the marginal probability of assignment to the Low Rate conviction trajectory for participants with a high PCL:YV score and the lowest possible score on the CRNA (0.16; 95% CI = 0.02 to 0.31) did not significantly differ when compared to participants with a high PCL:YV score and the highest possible score on the CRNA (0.40; 95% CI = 0.17 to 0.64). Panel B further illustrates that informal social control influences lower rates of offending, but not at higher levels of psychopathy. For example, for participants with PCL:YV scores lower than 30, informal social control helps differentiate the probability of assignment to the Low Rate conviction trajectory (i.e., confidence intervals do not overlap). However, for participants with a high PCL:YV score, informal social control does not help differentiate the probability of assignment to the Low Rate conviction trajectory (i.e., confidence intervals overlap). GEE models presented in the Supplemental Materials further support the conclusion that informal social control fails to influence lower rates of convictions among persons with higher PCL:YV scores.
--Insert Figure 2 about Here--
After reanalyzing data from the Gluecks’ UJD Study and revising their age-graded theory of informal social control, Sampson and Laub (2003) concluded that “differences in adult criminal trajectories cannot be predicted from childhood” (p. 588). Laub and Sampson (2003) further argued that informal social control was a normative part of the life course, even among persons from the most marginalized circumstances (i.e., rejection of selection effects). They also suggested that desistance occurred by default, meaning that informal social control influenced desistance irrespective of a person’s past developmental history or criminal propensity (i.e., rejection of treatment effect heterogeneity). These revisions distanced life-course criminology from Elder’s original life-course theory and pitted developmental criminology and life-course criminology paradigms against each other. Some criticized Laub and Sampson’s (2003) revisions to their theory on the basis that their revisions were premised on the analysis of data that are not suited for examining risk for persistent offending (see McCuish et al., 2021; Robins, 2005). Others argued that Laub and Sampson (2003) failed to explain why persons in the 21st century who experience difficult life circumstances are nevertheless destined for adult roles and will desist from offending despite a lack of motivation to change (see Paternoster et al., 2015; Thomas et al., 2021). Despite these criticisms, tests of Laub and Sampson’s assertions as they relate to contemporary correctional populations is a gap in developmental and life-course criminology research (Nguyen & Loughran, 2018; Thomas et al., 2022) and thus questions remain about selection effects, the relationship between criminal propensity and persistent offending, and treatment effect heterogeneity.
Every generation has unique problems and unique social responses to these problems, which is why historical context is central to life-course theory (Elder, 1975). We used Canadian data from the ISVYOS (n = 518) to provide a contemporary comparison point to Laub and Sampson’s (2003) conclusions about offending over the life course that were based on UJD Study data from the mid-20th century. We used PCL:YV ratings of childhood/adolescence psychopathy traits to capture the type of neuropsychological deficit Moffitt (1993) described as important to persistent offending (also see Lynam et al., 2007). Informal social control was measured several years after participants’ PCL:YV rating and was based on a five-item scale from the CRNA, a risk assessment tool completed by probation officers as part of case management protocols. Offending trajectories were measured for an average of 15 years across a follow-up period that began after the completion of the CRNA. Thus, temporal order was established between our key constructs of interest. Approximately two-thirds of the sample, distributed across two trajectories, were associated with low and declining rates of convictions early into the follow-up period. This reiterates that patterns that resemble desistance are common even for youth involved in particularly serious and violent offenses (McCuish et al., 2021). Although persistent offending is not inevitable even in this high-risk sample, neither is desistance. Unlike Laub and Sampson’s (2003) claims about the lack of evidence for persistent offending, nearly a third of the sample, represented across two trajectories, averaged approximately one observed conviction per year (i.e., even without statistically adjusting for exposure time) during the final five years of the follow-up period.
Psychopathy is considered a prototypical indicator of criminal propensity (DeLisi, 2016) and aligns with Moffitt’s (1993) description of how neuropsychological deficits influence risk for persistent offending. PCL:YV scores were (1) negatively associated with prosocial sources of informal social control in emerging adulthood, (2) positively associated with persistent offending trajectories through age 40, and (3) negatively associated with the Low Rate conviction trajectory, even in the presence of higher levels of informal social control. Overall, findings align with Elder’s (1994) emphasis on the interrelationship between early development and social environment in adulthood and the importance of personality in shaping investment in adult roles over the life course (also see Caspi & Bem, 1990; Clausen & Gilens, 1990; Lodi-Smith & Roberts, 2007). Given the need for theoretical development regarding the life course of correctional samples (Nguyen & Loughran, 2018; Thomas et al., 2022), we consider the implications of our findings for selection effects, persistent offending, and treatment effect heterogeneity and Laub and Sampson’s (2003) revised theory of informal social control.
Laub and Sampson (2003) felt that there was too much random variation over the life course for early development to matter for adult outcomes. Controlling for family adversity, youth convictions, and demographic characteristics, PCL:YV scores based on personality traits in childhood and adolescence were significantly and negatively associated with the CRNA informal social control scale measured in emerging adulthood. At least in emerging adulthood, positive sources of informal social control were not normative for persons with a developmental history of early psychopathy traits. This contrasts with Laub and Sampson’s (2003) endorsement of Becker’s (1964) structuralist perspective that individual-level intervention strategies are unnecessary to address the influence of various forms of psychopathology (e.g., personality disorder, serious mental health problems, intellectual disabilities), criminogenic needs (e.g., antisocial attitudes, substance abuse, self-control, anger management), and barriers to social integration (e.g., homelessness, immigration status, sex offender status, gang involvement) on social outcomes in adulthood. Laub and Sampson’s (2003) perspective that ‘nothing is needed’ when it comes to correctional intervention resembles Martinson’s (1974) since-debunked (Gendreau & Ross, 1983) report that developmental treatment programs in corrections are ineffective.
Laub and Sampson’s (2003) viewpoint that informal social controls are structurally-induced and a normative occurrence in the lives of persons from extremely marginalized circumstances implies that “nothing is needed” when it comes to correctional intervention. A background assumption of Laub and Sampson’s (2003) perspective is faith in the American status quo, or at least the evergreen ability to shift social systems to meet changing needs. However, the expectation that all will be well if society is adequately resourced with jobs and the hope of raising a family is incongruent with the range of difficult life circumstances experienced by incarcerated populations (Harris et al., 2010; Turney & Wakefield, 2019). These inequities, combined with factors like psychopathy that influence poorer quality adult roles, reiterate the importance of correctional psychosocial treatment programs that help address needs that are linked to both informal social control and reoffending (Bonta & Andrews, 2007; Gendreau & Ross, 1983; Polaschek & Skeem, 2018).
Controlling for adult informal social control, family adversity, youth convictions, and demographic characteristics, increases in PCL:YV scores were associated with significant increases in the relative risk of assignment to each of the three trajectories defined by higher rates of convictions throughout adulthood compared to the Low Rate conviction trajectory. Consistent with Altikriti et al. (2020), based on marginal effects, PCL:YV scores were more informative of a low probability of assignment to a low rate conviction trajectory as opposed to a high probability of assignment to a high rate conviction trajectory. These findings were replicated when using the PCL:YV three factor model that excludes items that reflect prior criminal/antisocial behavior. Thus, the relationship between PCL:YV test scores and convictions was not merely a matter of prior criminal behavior predicting future criminal behavior. A series of moderation analyses showed that PCL:YV test scores predicted reconviction outcomes irrespective of a person’s level of family adversity and frequency of convictions in adolescence (see Supplemental Materials). In other words, it was not necessary for participants to have experienced a negative social environment for PCL:YV test scores to predict reconvictions. Given the potential for results to be sensitive to SPGM, analyses were replicated in the Supplemental Materials using GEE models with conviction rate and conviction probability over each wave of the follow-up period as the outcomes of interest. PCL:YV test scores remained a significant predictor of conviction outcomes.
Overall, these analyses contradicted Sampson and Laub’s (2003) conclusion that “differences in adult criminal trajectories cannot be predicted from childhood, contra the National Summits of the policy world and apparently much yearning among criminologists” (p. 588). The difference in findings may relate to the lack of reliable measures of criminal propensity in the UJD Study (Robins, 2005). Additionally, despite presenting the UJD Study participants were primarily involved in status offenses and wayward delinquency (see Figure 7 from Sampson & Laub, 2003), which differs from the serious offender samples that scholars had in mind when developing theories about persistent offending (Paternoster et al., 2015). White boys housed in reformatory schools in the mid-20th century are unlikely to resemble youth from the 21st century whose incarceration resulted from involvement in serious offenses (Blokland & Nieuwbeerta, 2005; McCuish et al., 2021; van der Geest et al., 2009). Moreover, as opposed to being exposed to, for example, helpful social programs like the GI Bill, youth experiencing incarceration in the 21st century are those who came of age in an era associated with major forms of disadvantage (e.g., crack cocaine epidemic, declines in social welfare, mass incarceration), who are caught in the revolving door of the justice system (Padfield & Maruna, 2006; Snedker et al., 2017), and who suffer from mental health problems (Butler et al., 2022; Sugie & Turney, 2017). Our findings reemphasize the importance of contemporary samples when addressing gaps in theory-informed strategies for correctional populations (Thomas et al., 2022).
Understanding whether the influence of informal social control on lower rates of offending is limited to persons who are relatively low risk (i.e., absence of psychopathy traits) is a critical gap to address in studies of offending over the life-course (Nguyen & Loughran, 2018). Our examination of the impact of the interrelationship between psychopathy and informal social on future offending revealed two important conclusions. First, although higher scores on the CRNA informal social control scale in emerging adulthood significantly increased the odds of membership in the Low Rate conviction trajectory compared to each of the other three trajectories, informal social control failed to significantly improve the probability of assignment to the Low Rate conviction trajectory for participants with higher PCL:YV scores. Informal social control influences lower rates of offending, just not at higher levels of psychopathy. Second, participants with a high score (i.e., 30+) on the PCL:YV (n = 67) did not differ in their probability of assignment to the Low Rate conviction trajectory, regardless of their score on the CRNA informal social control scale. In contrast, participants with a score of less than 30 on the PCL:YV were significantly more likely to be assigned to the Low Rate conviction trajectory if they scored higher versus lower on the CRNA informal social control scale. In other words, in line with the treatment effect heterogeneity perspective that a person’s developmental history matters for how they respond to informal social control (Nguyen & Loughran, 2018), social control mattered for a lower rate of convictions, just not when it came to persons with psychopathy traits. GEE models supported this conclusion (see Supplemental Materials).
These findings are consistent with the risk assessment literature’s emphasis that protective factors may not operate as anticipated for persons with psychopathy traits (de Voegel et al., 2012). Findings also align with the perspective that personality disorders like psychopathy are associated with maladaptive social and identity functioning (Cooke et al., 2004; Sharp & Wall, 2021), even in positive social environments (Forth et al., 2022). Psychopathy is associated with a tendency to act in socially inappropriate ways when in adult roles like marriage and employment (Brazil & Volk, 2022). Qualitative interviews with persons who had former relationships with partners with features of psychopathy revealed experiences of financial abuse, social isolation, and emotional, professional, and physical harm (Forth et al., 2022). Employees of persons with features of psychopathy experienced decreased job satisfaction (Boddy, 2014). Accordingly, the mechanisms linking informal social controls to desistance, such as altering identities, forming close bonds, and knifing off from negative peers, may not elicit the presumed uniform response (Laub & Sampson, 2003) when psychopathy traits are present (also see Polaschek & Skeem, 2018). Findings also have implications for other desistance theories. For example, through manipulation, grandiosity, and pathological lying, persons with psychopathy traits may use their adult roles and relationships to avoid punishment by giving off the false perception of a “respectability package” (Giordano et al., 2002, p. 311). Indeed, rather than acting as a hook for change, at least in the short-term, people with psychopathy traits may establish adult roles and leverage perceptions of a respectability package to create opportunities to offend (Forth et al., 2022). This is consistent with descriptions of the relationship between psychopathy and a fast life history strategy that prioritizes mating effort over long-term relationships (Kardum et al., 2017).
Informal social control was measured using the CRNA, which is a case management tool completed by trained, well-educated probation officers as part of mandated practices and required consideration of information self-reported by their clients (Gress, 2010). It is possible that participants would have been more forthcoming in a confidential interview conducted for research purposes. That said, persons with histories of serious and violent offending can have poor insight into their relationships (e.g., Polaschek et al., 2022) and thus there are benefits to using the CRNA. We used an aggregate measure of informal social control to avoid reliability issues that come with single item indicators of a construct (Cronbach & Meehl, 1955). Prior research also showed that the accumulation of informal social control is more important to desistance than individual sources (Savolainen, 2009). The fact that the CRNA was informative of conviction trajectories helps substantiate the validity of this approach to measuring informal social control.
Structuralists like Becker (1964) might suggest that individual-level differences in risk are driven by macro-level factors and therefore the relationship between psychopathy and offending is spurious. However, exposure to structural inequality is not particularly informative of psychopathy traits (Cooke et al., 2001; Horan et al., 2015; McCuish et al., 2018). Although it seems unlikely that the relationship between psychopathy and negative adult outcomes is driven by social environment, we also did not account for all early developmental factors and thus cannot rule out the possibility of a spurious association. The current study highlights challenges simultaneously investigating selection effects and treatment effect heterogeneity. Specifically, because of selection effects, only 12 of the 67 participants with a high PCL:YV score also scored high (i.e., one standard deviation above the mean) on the CRNA informal social control scale. Thus, few participants were available to examine treatment effect heterogeneity. That said, our findings remained consistent when defining a high PCL:YV score more liberally (i.e., on standard deviation above the mean), which increased the number of participants with both a high PCL:YV score and high score on the CRNA informal social control scale (n = 17).
Some may argue that we measured informal social control too early in adulthood and that with more time, positive sources of informal social control would become a normative experience even for individuals with psychopathy traits. However, our findings did not change when we used a measure of informal social control that came later in emerging adulthood (Mean age = 22.92, SD = 2.29). Furthermore, if Laub and Sampson’s (2003) argument is that informal social control causes desistance, it makes little sense to measure informal social control well-after the peak in the age-crime curve. If informal social control does not happen until middle adulthood, it cannot be a cause of desistance, at least not for the majority of persons, even in high risk samples, whose offending declines in early adulthood (McCuish et al., 2021; Skardhamar & Savolainen, 2014).
Laub and Sampson (2003) argued that informal social controls were a normative part of the life-course, that early development was irrelevant to continued offending in adulthood, and that desistance occurred by default. We found that psychopathy traits in childhood/adolescence were (1) negatively related to informal social control in emerging adulthood (i.e., selection effects), (2) positively associated with persistent offending through age 40, and (3) negatively associated with lower rates of offending even when informal social control was high (i.e., treatment effect heterogeneity). Our findings support Elder’s original conceptualization of life-course theory that acknowledges the potential for individual, social, and macro-level factors to shape human development. Laub et al. (2008) argued that their theory was a response to their perception that the developmental paradigm was too deterministic and “losing sight of the humanity of individuals behind the statistics and their capacity for change” (p. 328). This assertion overlooks the developmental paradigm’s emphasis on person-centered intervention strategies and trauma-informed practices that recognize the capacity for change (Kazemian et al., 2019; Morizot & Kazemian, 2014). Laub and Sampson’s (2003) endorsement of Becker’s (1964) emphasis on shaping macro-level characteristics to “coerce people into behaving as we want them to”, seemingly with or without their consent, and with no interest in developing “deep and lasting interests” of the person (pp. 52-53) seems far more deterministic and far less humanistic than the emphasis in non-American jurisdictions that social service interventions help reduce inequalities (e.g., Brickman & Fristad, 2022; Lappi-Seppälä & Tonry, 2011; Tonry, 2022). The revised age-graded theory assumes that desistance is unintentional and occurs without any change to the individual (Paternoster et al., 2015; Thomas et al., 2022). We find that there is variability in sources of informal social control and that psychopathy helps make sense of this variability. Laub and Sampson’s yearning for structural factors to induce change among people affected by long-lasting personal and social issues is a leap of faith that ignores 40 years of correctional psychology research (e.g., Gendreau & Ross, 1983).
Altikriti, S., Nedelec, J. L., & Silver, I. (2021). The role of arrest risk perception formation in the association between psychopathy and aggressive offending. Youth violence and juvenile justice, 19(4), 402-422.
Altikriti, S., Theocharidou, K., & Sullivan, C. J. (2020). Specific theories of crime? A longitudinal assessment of the competing effects of psychopathy and self-control. Journal of Crime and Justice, 43(5), 547-567.
Andrews, D. A., Bonta, J., & Wormith, J. S. (2006). The recent past and near future of risk and/or need assessment. Crime & delinquency, 52(1), 7-27.
Beaver, K. M., DeLisi, M., Vaughn, M. G., & Wright, J. P. (2010). The intersection of genes and neuropsychological deficits in the prediction of adolescent delinquency and low self-control. International Journal of Offender Therapy and Comparative Criminology, 54(1), 22-42.
Becker, H. (1964). Personal change in adult life. Sociometry, 40-53.
Bersani, B., Laub, J., & Nieuwbeerta, P. (2009). Marriage and desistance from crime in the Netherlands: Do gender and socio-historical context matter? Journal of Quantitative Criminology, 25(1), 3-24.
Blokland, A., & Nieuwbeerta, P. (2005). The effects of life circumstances on longitudinal trajectories of offending. Criminology, 43, 1203-1240.
Boddy, C. (2014). Corporate psychopaths, conflict, employee affective well-being and counterproductive work behaviour. Journal of Business Ethics, 121(1), 107-121.
Bonta, J., & Andrews, D. (2007). Risk-need-responsivity model for offender assessment and rehabilitation. Rehabilitation, 6(1), 1-22.
Braga, T., de Castro Rodrigues, A., Cruz, A. R., Pechorro, P., & Cunha, O. (2023). The four facets of the Psychopathy Checklist, Youth Version and recidivism: A meta-analysis. Aggression and Violent Behavior, 101824.
Brazil, K., & Volk, A. (2022). Cads in Dads’ Clothing? Psychopathic Traits and Men’s Preferences for Mating, Parental, and Somatic Investment. Evolutionary Psychological Science, 1-17.
Brickman, H., & Fristad, M. (2022). Psychosocial treatments for bipolar disorder in children and adolescents. Annual Review of Clinical Psychology, 18, 291-327.
Butler, A., Nicholls, T., Samji, H., Fabian, S., & Lavergne, M. (2022). Prevalence of mental health needs, substance use, and co-occurring disorders among people admitted to prison. Psychiatric Services, 73(7), 737-744.
Caspi, A. & Bem, D. (1990). Personality continuity and change across the life course. In L.A. Pervin (Ed) Handbook of personality (pp. 549-75). Guilford.
Caspi, A., & Herbener, E. S. (1990). Continuity and change: Assortative marriage and the consistency of personality in adulthood. Journal of Personality and Social Psychology, 58(2), 250–258.
Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality development: Stability and change. Annual Review of Psychology, 56, 453-484.
Clausen, J., & Gilens, M. (1990). Personality and labor force participation across the life course: A longitudinal study of women's careers. Sociological Forum, 5, 595-618).
Cohn, E., Farrington, D., & Iratzoqui, A. (2017). Changes in the most-cited scholars and works over 25 years: The evolution of the field of criminology and criminal justice. Journal of Criminal Justice Education, 28(1), 25-51.
Cooke, D., Kosson, D., & Michie, C. (2001). Psychopathy and ethnicity: Structural, item, and test generalizability of the Psychopathy Checklist—Revised (PCL-R) in Caucasian and African American participants. Psychological Assessment, 13(4), 531–542
Cooke, D., Michie, C., Hart, S., & Clark, D. (2004). Reconstructing psychopathy: Clarifying the significance of antisocial and socially deviant behavior in the diagnosis of psychopathic personality disorder. Journal of Personality Disorders, 18(4), 337.
Corrado, R., Gronsdahl, K., MacAlister, D., & Cohen, I. (2010). Youth justice in Canada: Theoretical perspectives of youth probation officers. Canadian Journal of Criminology and Criminal Justice, 52, 397-426.
Cronbach, L., & Meehl, P. (1955). Construct validity in psychological tests. Psychological Bulletin, 52, 281–302.
DeLisi, M. (2016). Psychopathy as unified theory of crime. Palgrave Macmillan.
Doherty, E. (2006). Self‐control, social bonds, and desistance: A test of life‐course interdependence. Criminology, 44(4), 807-833.
Eisenbarth, H., Hart, C., Zubielevitch, E., Keilor, T., Wilson, M., Bulbulia, J., ... & Sedikides, C. (2022). Aspects of psychopathic personality relate to lower subjective and objective professional success. Personality and Individual Differences, 186, 111340.
Elder, G. (1975). Age differentiation and the life course. Annual Review of Sociology, 1, 165-190.
Elder, G. (1994). Time, human agency, and social change: Perspectives on the life course. Social Psychology Quarterly, 4-15.
Farrington, D. (1986). Age and crime. Crime and Justice, 7, 189-250.
Farrington, D., & Bergstrøm, H. (2018). Family background and psychopathy. In C. J. Patrick (Ed.), Handbook of psychopathy (pp. 354–379). The Guilford Press.
Farrington, D., Ttofi, M., Crago, R., & Coid, J. (2014). Prevalence, frequency, onset, desistance and criminal career duration in self-reports compared with official re-cords. Criminal Behaviour and Mental Health, 24, 241–253.
Fontaine, N., Lacourse, E., Vitaro, F., & Tremblay, R. (2014). Links between trajectories of self‐reported violent and nonviolent offending and official offending during adolescence and adulthood. Criminal Behaviour and Mental Health, 24, 277–290.
Forth, A., Kosson, D., & Hare, R. (2003). The Hare Psychopathy Checklist: Youth Version. Multi-Health Systems.
Forth, A., Sezlik, S., Lee, S., Ritchie, M., Logan, J., & Ellingwood, H. (2022). Toxic relationships: The experiences and effects of psychopathy in romantic relationships. International Journal of Offender Therapy and Comparative Criminology, 66(15), 1627-1658.
Fox, B., Jennings, W., & Farrington, D. (2015). Bringing psychopathy into developmental and life-course criminology theories and research. Journal of Criminal Justice, 43, 274-289.
Gendreau, P., & Ross, R. (1983). Correctional treatment: Some recommendations for effective intervention. Juv. & Fam. Ct. J., 34, 31.
Giordano, P., Cernkovich, S., & Rudolph, J. (2002). Gender, crime, and desistance: Toward a theory of cognitive transformation. American Journal of Sociology, 107(4), 990-1064.
Glueck, S. (1960). Ten years of "Unraveling Juvenile Delinquency": An examination of criticisms. The Journal of Criminal Law, Criminology, and Police Science, 51, 283-308.
Glueck, S., & Glueck, E. (1950). Unraveling juvenile delinquency. Commonwealth Fund.
Gress, C. (2010). Revealing research & evaluation. https://www2.gov.bc.ca/assets/gov/law-crime-and-justice/criminal-justice/corrections/research-evaluation/issue-3.pdf.
Hare, R. (2003). The Psychopathy Checklist–Revised. Toronto, ON,
Harris, A., Evans, H., & Beckett, K. (2010). Drawing blood from stones: Legal debt and social inequality in the contemporary United States. American Journal of Sociology, 115(6), 1753-1799.
Hart, S. (1998). Psychopathy and risk for violence. Legal & Criminological Society, 3, 121-137.
Hart, S., Douglas, K., & Guy, L. (2016). The Structured Professional Judgement Approach to violence risk assessment: Origins, nature, and advances. The Wiley handbook on the theories, assessment and treatment of sexual offending, 643-666.
Hay, C., & Forrest, W. (2006). The development of self‐control: Examining self‐control theory's stability thesis. Criminology, 44(4), 739-774.
Hemphälä, M., & Hodgins, S. (2014). Do psychopathic traits assessed in mid-adolescence predict mental health, psychosocial, and antisocial, including criminal outcomes, over the subsequent 5 years? The Canadian Journal of Psychiatry, 59, 40-49.
Herpers, P., Klip, H., Rommelse, N., Greven, C. & Buitelaar, J. (2016). Associations between high callous–unemotional traits and quality of life across youths with non-conduct disorder diagnoses. European Child & Adolescent Psychiatry, 25(5), 547-555.
Hirschi, T., & Gottfredson, M. (1995). Control theory and the life-course perspective. Studies on Crime and Crime Prevention 4, 131-142.
Hoge, R. D. (2002). Standardized instruments for assessing risk and need in youthful offenders. Criminal Justice and Behavior, 29(4), 380-396.
Horan, J., Brown, J., Jones, S., & Aber, J. (2015). Assessing invariance across sex and race/ethnicity in measures of youth psychopathic characteristics. Psychological Assessment, 27(2), 657.
Kardum, I., Hudek-Knezevic, J., Gračanin, A., & Mehic, N. (2017). Assortative mating for psychopathy components and its effects on the relationship quality in intimate partners. Psihologijske teme, 26(1), 211-239.
Kazemian, L., Farrington, D., & Piquero, A. (2019). Developmental and life-course criminology. In D. Farrington, L. Kazemian, & A. Piquero (Eds.), The Oxford handbook of developmental and life-course criminology. Oxford University Press.
King, R., Massoglia, M., & Macmillan, R. (2007). The context of marriage and crime: Gender, the propensity to marry, and offending in early adulthood. Criminology, 45, 33–66.
Kirk, D. (2006). Examining the divergence across self-report and official data sources on inferences about the adolescent life-course of crime. Journal of Quantitative Criminology, 22, 107-129.
Krueger, R., Moffitt, T., Caspi, A., Bleske, A., & Silva, P. (1998). Assortative mating for antisocial behavior: Developmental and methodological implications. Behavior genetics, 28, 173-186.
Kubrin, C., & Stewart, E. A. (2006). Predicting who reoffends: The neglected role of neighborhood context in recidivism studies. Criminology, 44(1), 165-197.
Lappi-Seppälä, T., & Tonry, M. (2011). Crime, criminal justice, and criminology in the Nordic countries. Crime and Justice, 40(1), 1-32.
Laub, J. & Sampson, R. (2003). Shared beginnings, divergent lives: Delinquent boys to age 70. Harvard University Press.
Laub, J., & Sampson, R. (1993). Turning points in the life course: Why change matters to the study of crime. Criminology, 31, 301-325.
Laub, J., Nagin, D., & Sampson, R. (1998). Trajectories of change in criminal offending: Good marriages and the desistance process. American Sociological Review, 63, 225-238.
Laub, J., Rowan, Z., & Sampson, R. (2018). The age-graded theory of informal social control. In The Oxford handbook of developmental and life-course criminology (pp. 295-322). Oxford Press.
Laub, J., Sampson, R., & Sweeten, G. (2008). Assessing Sampson and Laub’s life-course theory of crime. In F. Cullen, J.P. Wright, & K. Blevins (Eds.), Taking stock: The status of criminological theory (pp. 313-333). Routledge.
Lauritsen, J. (1999). Limitations in the use of longitudinal self-report data. Criminology, 37, 687.
Lee, Y., & Kim, J. (2021). Psychopathic traits among serious juvenile offenders: Developmental pathways, multidimensionality, and stability. Crime & Delinquency, 67(1), 82-110.
Lodi-Smith, J., & Roberts, B. (2007). Social investment and personality: A meta-analysis of the relationship of personality traits to investment in work, family, religion, and volunteerism. Personality and Social Psychology Review, 11(1), 68-86.
Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., & White, H. R. (Eds.). (2008). Violence and serious theft: Development and prediction from childhood to adulthood. Taylor & Francis.
Logan, C., & Johnstone, L. (2010). Personality disorder and violence: Making the link through risk formulation. Journal of Personality Disorders, 24(5), 610.
Lussier, P., & McCuish, E. (2016). Desistance from crime without reintegration: A longitudinal study of the social context and life course path to desistance in a sample of adults convicted of a sex crime. International Journal of Offender Therapy and Comparative Criminology, 60(15), 1791-1812.
Lynam, D. R., Caspi, A., Moffitt, T. E., Loeber, R., & Stouthamer-Loeber, M. (2007). Longitudinal evidence that psychopathy scores in early adolescence predict adult psychopathy. Journal of Abnormal Psychology, 116(1), 155-165.
Malakieh, J. (2017). Youth correctional statistics in Canada, 2015/2016. Ottawa, ON.
Martinson, R. (1974). What works? Questions and answers about prison reform. The Public Interest, 35, 22-54.
Mathieu, C., & Babiak, P. (2016). Corporate psychopathy and abusive supervision: Their influence on employees' job satisfaction and turnover intentions. Personality and Individual Differences, 91, 102-106.
McCuish, E. C. (2020). Seductions of exposure time: The mismeasurement of desistance among persons involved in frequent and serious offending. Canadian Journal of Criminology and Criminal Justice, 62(3), 29-50.
McCuish, E., Lussier, P., & Corrado, R. (2021). The life-course of serious and violent youth grown up: A twenty-year longitudinal study. Routledge.
McCuish, E., Lussier, P., & Corrado, R. (2022). Cohort Profile: The Incarcerated Serious and Violent Young Offender Study. Journal of Developmental and Life-Course Criminology, 8(2), 315-335.
McCuish, E., Mathesius, J., Lussier, P., & Corrado, R. (2018). The cross-cultural generalizability of the Psychopathy Checklist: Youth Version for adjudicated Indigenous youth. Psychological Assessment, 30, 192.
Moffitt, T. (1993). "Life-course-persistent" and "adolescent-limited" antisocial behavior: A developmental taxonomy. Psychological Review 100, 674-701.
Mustaine, E., & Tewksbury, R. (2011). Residential relegation of registered sex offenders. American Journal of Criminal Justice, 36(1), 44-57.
Nagin, D. (2005). Group-based modeling of development. Harvard University Press.
Nguyen, H. & Loughran, T. (2018). On the measurement and identification of turning points in criminology. Annual Review of Criminology, 1, 335-358.
Nilsson, A. Estrada, F., & Bäckman, O. (2017). The unequal crime drop: Changes over time in the distribution of crime among individuals from different socioeconomic backgrounds. European Journal of Criminology, 14, 586-605.
Olver, M. E., Lewis, K., & Wong, S. C. (2013). Risk reduction treatment of high-risk psychopathic offenders: The relationship of psychopathy and treatment change to violent recidivism. Personality Disorders: Theory, Research, and Treatment, 4, 160-167.
Olver, M. E., Stockdale, K. C., & Wormith, J. S. (2009). Risk assessment with young offenders: A meta-analysis of three assessment measures. Criminal Justice and Behavior, 36(4), 329–353
Padfield, N., & Maruna, S. (2006). The revolving door at the prison gate: Exploring the dramatic increase in recalls to prison. Criminology & Criminal Justice, 6(3), 329-352.
Paternoster, R., Bachman, R., Bushway, S., Kerrison, E., & O’Connell, D. (2015). Human agency and explanations of criminal desistance: Arguments for a rational choice theory. Journal of Developmental and Life-Course Criminology, 1(3), 209-235.
Ploe, M. L., Berluti, K., Ibonie, S. G., Villanueva, C. M., Marsh, A., & Gruber, J. (2023). Psychopathy and associations with reward responsiveness and social networks in emerging adults. Journal of Research in Personality, 103, 104357.
Polaschek, D. & Skeem, J. (2018). Treatment of adults and juveniles with psychopathy. In C. J. Patrick (Ed.), Handbook of psychopathy (2nd Ed.) (pp. 710–731). Guilford Press.
Polaschek, D. (2019). The psychology of desistance. The Wiley international handbook of correctional psychology, 315-336.
Polaschek, D. Bell, R. Casey, A. Dickson, S. & Yesberg, J. (2022). Do triarchic psychopathy components of New Zealand high-risk parolees predict probation officer relationship quality, quality of life on parole, and recidivism? International Journal of Offender Therapy and Comparative Criminology, 66(15), 1682-1702.
Polaschek, D. L., & Daly, T. E. (2013). Treatment and psychopathy in forensic settings. Aggression and Violent Behavior, 18(5), 592-603.
Polaschek, D. L., & Ross, E. C. (2010). Do early therapeutic alliance, motivation, and stages of change predict therapy change for high‐risk, psychopathic violent prisoners? Criminal Behaviour and Mental Health, 20(2), 100-111.
Reisig, M. Bales, W., Hay, C., & Wang, X. (2007). The effect of racial inequality on black male recidivism. Justice Quarterly, 24(3), 408-434.
Robins, L. (2005). Explaining when arrests end for serious juvenile offenders: Comments on the Sampson and Laub study. Annals of the American Academy of Political and Social Science, 602, 57-72.
Salekin, R. (2016). Psychopathy in childhood: Toward better informing the DSM–5 and ICD-11 conduct disorder specifiers. Personality disorders: Theory, research, and treatment, 7(2), 180.
Salekin, R. T., Andershed, H., Batky, B. D., & Bontemps, A. P. (2018). Are callous unemotional (CU) traits enough? Journal of Psychopathology and Behavioral Assessment, 40, 1-5.
Sampson, R., & Laub, J. (1990). Crime and deviance over the life course: The salience of adult social bonds. American Sociological Review, 609-627.
Sampson, R., & Laub, J. (1993). Crime in the making: Pathways and turning points through life. Harvard University Press.
Sampson, R., & Laub, J. (2003). Life‐course desisters? Trajectories of crime among delinquent boys followed to age 70. Criminology, 41, 555-592.
Sampson, R., & Laub, J. (2005a). A life-course view of the development of crime. The Annals of the American Academy of Political and Social Science, 602, 12-45.
Sampson, R., & Laub, J. (2005b). When prediction fails: From crime-prone boys to heterogeneity in adulthood. The Annals of the American Academy of Political and Social Science, 602, 73-79.
Sampson, R., & Laub, J. (2005c). Seductions of method: Rejoinder to Nagin and Tremblay's “Developmental trajectory groups: Fact or fiction?”. Criminology, 43(4), 905-913.
Sampson, R., & Laub, J. (2016). Turning points and the future of life-course criminology: Reflections on the 1986 criminal careers report. Journal of Research in Crime and Delinquency, 53, 321-335.
Sampson, R., Laub, J., & Wimer, C. (2006). Does marriage reduce crime? A counterfactual approach to within‐individual causal effects. Criminology, 44, 465-508.
Sanchez, J. & Tostlebe, J. (2022, October 10). Career reflection: Computer cards, theory, and research with John Laub (No. 56). In The Criminology Academy. Spotify.
Savolainen, J. (2009). Work, family and criminal desistance: Adult social bonds in a Nordic welfare state. The British Journal of Criminology, 49(3), 285-304.
Seto, E., & Davis, W. (2021). Authenticity predicts positive interpersonal relationship quality at low, but not high, levels of psychopathy. Personality and Individual Differences, 182, 1-6.
Sharp, C., & Wall, K. (2021). DSM-5 level of personality functioning: Refocusing personality disorder on what it means to be human. Annual Review of Clinical Psychology, 17, 313-337.
Shin, T., Davison, M. L., & Long, J. D. (2017). Maximum likelihood versus multiple imputation for missing data in small longitudinal samples with nonnormality. Psychological Methods, 22(3), 426–449.
Skardhamar, T., & Savolainen, J. (2014). Changes in criminal offending around the time of job entry: A study of employment and desistance. Criminology, 52(2), 263-291.
Snedker, K., Beach, L., & Corcoran, K. (2017). Beyond the “revolving door?”: Incentives and criminal recidivism in a mental health court. Criminal Justice and Behavior, 44, 1141-1162.
StataCorp. (2019). Stata statistical software: release 16. StataCorp LLC, College Station, TX.
Sugie, N., & Turney, K. (2017). Beyond incarceration: Criminal justice contact and mental health. American Sociological Review, 82(4), 719-743.
Thomas, K., Nguyen, H., & Jackson, E. (2022). Value orientations, life transitions, and desistance: Assessing competing perspectives. Criminology, online first, 1-29.
Thomas, K., Pogarsky, G., & Loughran, T. (2021). Paternoster on human agency and crime: A rejoinder to critics on his behalf. Journal of Developmental and Life-Course Criminology, 7, 524-542.
Tonry, M. (2013). “Nothing” works: Sentencing “Reform” in Canada and the United States. Canadian Journal of Criminology and Criminal Justice, 55(4), 465-480.
Tonry, M. (2022). Punishments, Politics, and Prisons in Western Countries. Crime and Justice, 51, 1-52.
Turney, K., & Wakefield, S. (2019). Criminal justice contact and inequality. RSF: The Russell Sage Foundation Journal of the Social Sciences, 5(1), 1-23.
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, 111-143.
Van Hofer, H. (2014). Crime and reactions to crime in 34 Swedish birth cohorts: From historical descriptions to forecasting the future. Journal of Scandinavian Studies in Criminology and Crime Prevention, 15, 167-181.
Vaughn, M., & 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.
Wang, X., Hay, C., Todak, N., & Bales, W. (2014). Criminal propensity, social context, and recidivism: A multilevel analysis of interactive relationships. Criminal Justice and Behavior, 41(3), 300-317.
Wormith, J. S., Olver, M. E., Stevenson, H. E., & Girard, L. (2007). The long-term prediction of offender recidivism using diagnostic, personality, and risk/need approaches to offender assessment. Psychological Services, 4(4), 287.
Wright, B, Caspi, A., Moffitt, T., & Silva, P. (2001). The effects of social ties on crime vary by criminal propensity: A life‐course model of interdependence. Criminology, 39, 321-348.
Wright, B. R. E., Caspi, A., Moffitt, T. E., & Silva, P. A. (1999). Low self‐control, social bonds, and crime: Social causation, social selection, or both? Criminology, 37(3), 479-514.
Table 1. Descriptive statistics for all variables (n = 518) | ||||
|---|---|---|---|---|
Variable | Mean/Percent | SD | Min | Max |
Conviction Trajectories | ||||
Low Rate | 43.2% | - | 0 | 1 |
Slow Decreasing | 25.1% | - | 0 | 1 |
Slow Increase | 16.4% | - | 0 | 1 |
High Rate | 15.3% | - | 0 | 1 |
Informal Social Controls | ||||
CRNA Informal Social Control Scale | 7.43 | 5.41 | 0 | 15 |
Psychopathy | ||||
PCL:YV Total Scores | 21.19 | 6.50 | 4 | 37 |
Control variables | ||||
Female | 17.4% | - | 0 | 1 |
Ethnicity | ||||
White | 59.7% | - | 0 | 1 |
Indigenous | 25.4% | - | 0 | 1 |
Non-Indigenous Minority | 14.9% | - | 0 | 1 |
Age of CRNA Assessment | 19.33 | 1.71 | 18 | 25 |
Family Adversity Scale | 2.88 | 1.61 | 0 | 6 |
Youth Convictions | 12.74 | 9.00 | 0 | 50 |
Table 2. Ordinary least squares regression predicting scores on the CRNA informal social control scale (n = 505) | ||
|---|---|---|
Model 1 | Model 2 | |
Variable | b coef (SE) | b coef (SE) |
Psychopathy | ||
PCL:YV Total Scores | -0.09 (0.03)* | -0.08 (0.03)* |
Control Variables | ||
Female | 1.16 (0.58) | 1.15 (0.58) |
Whitea | . | . |
Indigenous | -0.67 (0.51) | -0.69 (0.50) |
Non-Indigenous Minority | 0.23 (0.63) | 0.16 (0.63) |
Age at Follow-Up | 1.26 (0.13)*** | 1.25 (0.13)*** |
Family Adversity Scale | -0.19 (0.14) | -0.18 (0.14) |
Youth Convictionsb | -0.34 (0.13)* | -0.34 (0.13)* |
Interaction Term | ||
PCL:YV Score * Family Adversity Scale | . | 0.04 (0.02) |
Model Fit | F (7, 497) = 22.64, p < .001; R2 = 0.242 | F (8, 496) = 20.14, p < .001; R2 = 0.245 |
a Reference category b Used logarithmic transformation * p < .05; ** p < .01; *** p < .001 Notes. PCL:YV and family adversity scales were centered when creating the interaction term. Sample size changed from 518 to 505 due to missing data for ethnicity and sealed youth conviction records | ||
Table 3. The relationship between PCL:YV test scores, CRNA informal social control scale, and offending trajectories (n = 505)a | ||||||
|---|---|---|---|---|---|---|
Model 1 (Risk for Persistent Offending) | Model 2 (Treatment Effect Heterogeneity) | |||||
Slow Decreasing | Slow Increasing | High Rate | Slow Decreasing | Slow Increasing | High Rate | |
Variable | RRR (95% CI) | RRR (95% CI) | RRR (95% CI) | RRR (95% CI) | RRR (95% CI) | RRR (95% CI) |
Informal Social Control | ||||||
CRNA Informal Social Control Scale | 0.87*** (0.83-0.92) | 0.95 (0.90-1.01) | 0.88*** (0.82-0.95) | 0.87*** (0.83-0.92) | 0.95 (0.90-1.01) | 0.88*** (0.82-0.95) |
Psychopathy | ||||||
PCL:YV Total Scores | 1.08** (1.04-1.13) | 1.04* (1.00-1.09) | 1.06* (1.01-1.12) | 1.08** (1.04-1.13) | 1.04 (0.998-1.08) | 1.06* (1.00-1.11) |
Control Variables | ||||||
Female | 0.27** (0.13-0.56) | 0.58 (0.29-1.14) | 0.03** (0.004-0.23) | 0.27** (0.13-0.56) | 0.58 (0.30-1.15) | 0.03** (0.004-0.24) |
Whiteb | - | - | - | - | - | - |
Indigenous | 1.39 (0.74-2.58) | 2.03* (1.08-3.83) | 2.40* (1.21-4.74) | 1.38 (0.74-2.58) | 2.03* (1.08-3.82) | 2.40* (1.21-4.74) |
Non-Indigenous Minority | 0.94 (0.46-1.93) | 0.74 (0.34-1.64) | 0.23* (0.06-0.82) | 0.94 (0.46-1.95) | 0.75 (0.34-1.67) | 0.23* (0.06-0.83) |
Age at Follow-Up | 0.76** (0.64-0.90) | 0.82* (0.69-0.98) | 0.69** (0.54-0.88) | 0.76** (0.63-0.90) | 0.81* (0.68-0.97) | 0.69** (0.54-0.88) |
Family Adversity Scale | 1.00 (0.84-1.18) | 0.85 (0.72-1.02) | 0.96 (0.78-1.18) | 0.99 (0.84-1.18) | 0.85 (0.71-1.01) | 0.97 (0.79-1.19) |
Youth Convictionsc | 1.21* (1.01-1.44) | 1.11 (0.94-1.30) | 1.43* (1.08-1.90) | 1.20* (1.01-1.44) | 1.10 (0.93-1.30) | 1.43* (1.08-1.90) |
Interaction Term | ||||||
PCL:YV Total Score * CRNA Informal Social Control Scale | - | - | - | 1.00 (0.99-1.01) | 1.00 (0.99-1.01) | 1.00 (0.99-1.01) |
Model Fit | LL = -556.23; χ 2 = 200.97; p < .001; McFadden’s Pseudo R2 = 0.153 | LL = -555.88; χ2 = 201.68; p < .001; McFadden’s Pseudo R2 = 0.154 | ||||
Notes. RRR = relative risk ratio. Sample size changed from 518 to 505 due to missing data for ethnicity and sealed youth conviction records a Low Rate conviction trajectory used as reference category b Used as reference category c Used logarithmic transformation * p < .05; ** p < .01; *** p < .001 | ||||||
Chart Description automatically generated with low confidence
Figure 1. Plot of trajectories based on expected convictions
Notes. A wave represents an entire year of age. Participants’ ages at Wave 1 ranged from 18 to 25. Dotted lines represent 95% confidence intervals. Sharp declines/increases in expected convictions reflect fluctuations in exposure time across waves. Even when including exposure time in the model, a high number of convictions in a given wave implies spending lengthy periods of time incarceration at the next wave (see McCuish, 2020).
A screenshot of a computer screen Description automatically generated
Figure 2. Probability of trajectory assignment as a function of CRNA informal social control scale scores across PCL:YV total scores
Notes. Consistent with the PCL literature, a high PCL:YV score (n = 67) is represented as a score of 30 or higher (Hare, 2003). A score of 2.00 or lower represents less than one standard deviation below the mean on the CRNA informal social control scale (n = 122). A score of 12.85 or higher represents greater than one standard deviation above the mean on the CRNA informal social control scale (n = 147). Shaded areas represent 95% confidence intervals.
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC-BY-NC-ND 4.0)
