Perceived Pubertal Timing and Deviant Peer Processes Predicting Substance Use Initiation: The Moderating Role of Impulsiveness

Abstract

The present study aimed to test if perceived pubertal timing was related to marijuana and alcohol use through deviant peer affiliation. Additionally, we examined if the deviant peer pathway was moderated by impulsiveness, gender, or both. Data were collected from 342 youth, most of whom had a family history of substance use disorder. Youth completed assessments every six months from age 13 to 16. For girls only, longitudinal analyses revealed that perceived pubertal timing was indirectly related to substance use through higher levels of deviant peer affiliation. This pathway was moderated by impulsiveness such that the association between perceived pubertal timing and deviant peer affiliation was only present for girls with average to high levels of impulsiveness. These findings elucidate a developmental pathway from perceived pubertal timing to substance use through deviant peer affiliation for high-risk girls, although low levels of impulsiveness were protective.

Keywords

puberty/pubertal development peer relationships personality substance use/alcohol and drug use

Understanding the complex social and biological transitions during adolescence is crucial to reduce the likelihood of maladjustment, particularly the early onset of substance use. Puberty is perhaps the most obvious indicator of the transition into adolescence and is central to the biological changes defining this life stage. Some youth may experience difficulties (Cance et al., 2013; Castellanos-Ryan et al., 2013; Dick et al., 2000; Ullsperger & Nikolas, 2017) during this transitional state, in particular those who perceive themselves to be more physically developed compared to their same-age, same-sex peers (i.e., perceptions of early pubertal timing).

An important mediator of the association between perceptions of early pubertal timing and maladjustment is affiliating with peers who engage in delinquent or deviant behaviors (e.g., substance use, theft). While deviant peer affiliation may be partly responsible for the association between perceptions of pubertal timing and delinquent behavior, including substance use, other factors have been shown to modify this pathway. Previous research has focused mainly on contextual factors (e.g., parenting behaviors) that influence the relationship between perceptions of pubertal timing and deviant peer affiliation (Westling et al., 2008). We aimed to extend prior research by testing the possibility that personality factors such as impulsiveness may also modify the association between perceptions of pubertal timing and affiliation with deviant peers. The overall goal of the present study was to examine if impulsiveness moderated the indirect association between perceptions of pubertal timing and substance use through deviant peer affiliation. Additionally, differences between girls and boys in this hypothesized pathway were examined because of sex differences in pubertal maturation. To accomplish this, we analyzed data collected longitudinally from a sample of youth, most of whom were at risk for substance use problems.

Pubertal Timing and Deviant Peer Affiliation: Links to Maladjustment

One consequence of perceived early maturation is that youth who view themselves to be more physically developed relative to same-age, same-sex peers may begin to affiliate with older, more deviant peers who are engaging in a high rate of delinquent and deviant acts (Ge & Natsuaki, 2009; Klopack et al., 2020). Due to normative age-graded changes (Moffitt, 1993), older adolescents may be engaging in delinquent behavior at a rate that is seen as “typical” for their age but “deviant” for younger adolescents. Early-maturing youth may select into deviant peer groups because their perceptions of advanced physical development are more closely aligned with older, more deviant peers than similar-aged peers who appear less physically mature (Magnusson et al., 1985). In the deviant peer context, early-maturing youth may be rewarded and encouraged to engage in delinquent behavior (Brechwald & Prinstein, 2011; Dishion et al., 1996). Thus, the deviant peer context is one risk factor through which perceptions of early pubertal maturation are linked to maladaptive outcomes, such as the early initiation of substance use.

Importantly, early-maturing youth may be especially susceptible to the deviant peer context considering their chronological age (Stattin et al., 2011). According to the stage termination hypothesis (Caspi & Moffitt, 1991), youth who perceive themselves to be developing early compared to same-age, same-sex peers may be emotionally ill-equipped to navigate the multifaceted changes that occur during adolescence. For example, early-maturing youth may be susceptible to the deviant peer context because they have yet to develop the capacity to resist peer influence (Steinberg & Monahan, 2007). Indeed, previous research has shown that early-maturing youth are more vulnerable to the influence of deviant peers and more likely to engage in delinquent behavior as a consequence compared to on-time and late-maturing youth (Costello et al., 2007; Fergusson et al., 2007). Thus, youth who perceive themselves as more physically developed may be especially vulnerable to the deviant peer context during early adolescence because of immaturities in cognitive and emotional development.

According to the stage termination hypothesis, perceptions of pubertal timing should be related to peer relationships during early adolescence because individual differences are accentuated during this transitional state. However, it does not make explicit predictions about the development of peer relationships. Instead, two other frameworks offer guidance. The attenuation hypothesis (Copeland et al., 2010; Skoog & Kapetanovic, 2022) proposes that while perceptions of early maturation may be a risk factor for maladjustment during early adolescence, these effects diminish over time. As youth develop psychologically and differences in perceived physical maturation become less evident, this hypothesis suggests that any influence of early pubertal timing does not persist past early adolescence. Thus, the attenuation hypothesis would posit that the influence of perceived pubertal timing on deviant peer affiliation is isolated during early adolescence. Perceptions of pubertal maturation would not be expected to be related to developmental change in deviant peer affiliation because individual differences in pubertal maturation are less apparent as they age.

Alternatively, the selective persistence hypothesis (Senia et al., 2018) proposes that perceptions of early pubertal timing continue to influence youth development beyond early adolescence. This is expected because early-maturing youth may be caught in “snares” (e.g., school truancy or dropout) that alter developmental trajectories (Mendle, Ryan, & McKone, 2019; Moffitt, 1993). Thus, according to this hypothesis, perceptions of pubertal timing would be related to deviant peer affiliation during early adolescence and developmental change in deviant peer affiliation. This is expected because these developmental “snares” might increase the rate at which youth affiliate with deviant peers. There is empirical support for both the attenuation (Jormanainen et al., 2014; Senia et al., 2018) and selective persistence hypotheses (Bucci & Staff, 2020; Carter et al., 2018; Hay et al., 2018; Haynie & Piquero, 2006; Mendle, Ryan, & McKone, 2019), although there is stronger support for the former.

Further, other contextual factors may modify the relationship between perceptions of pubertal timing and deviant peer affiliation. For instance, the association between perceptions of pubertal timing and deviant peer affiliation is diminished among youth whose parents monitor their behavior (Marceau & Jackson, 2017; Westling et al., 2008). The present study examined whether impulsive personality traits might moderate the relationship between perceptions of pubertal timing and deviant peer affiliation. Youth with difficulty managing their impulsive tendencies may be less accepted by their peers because they are disobedient and uncooperative (Cillessen et al., 1992). For example, Gomes and Livesey (2008) found that greater impulsiveness was related to poor relationships with peers, as rated by teachers. Subsequently, impulsive youth may have unstable relationships with conventional peers and seek out deviant peer groups (Chapple, 2005). Other researchers, as well, have shown that impulsive tendencies can be harmful to forming healthy social relationships, which in turn can increase the likelihood of delinquent behavior (Bagwell et al., 2001; Snyder et al., 2004). While research has illustrated a direct role of impulsiveness in peer-group processes, we aim to extend this rationale by testing for a moderating role. Perceptions of early maturation alone may not be sufficient for youth to affiliate with deviant peers. Instead, early-maturing youth who are impulsive may disconnect from conventional peers and turn to deviant peers, whereas early-maturing youth who are less impulsive may maintain their ties with conventional peers. Thus, while perceptions of pubertal timing and impulsiveness may play a direct role, the two may also act synergistically as a pathway to deviant peer affiliation.

Gender Moderation

The influence of perceptions of pubertal timing on adjustment may operate differently for girls and boys (see Mendle et al., 2007; Mendle & Ferrero, 2012 for review). For example, prior research has shown that early-maturing girls are more likely to affiliate with deviant peers and in turn engage in externalizing behaviors, whereas this pathway does not operate for boys (Marceau & Jackson, 2017; Westling et al., 2008). This difference could be due to variations in the typical sequence of puberal development between girls and boys. The initial stages of puberty for girls include the maturation of secondary sexual characteristics that are more observable (e.g., growth spurt, breast development). In contrast, the initial stages for boys include the maturation of primary sexual characteristics that are less observable, e.g., gonadarche (Wolf & Long, 2016).

Additionally, on average, girls begin puberty earlier than boys. Taken together, girls may appear physically mature earlier than boys of a similar chronological age; thus, perceptions of pubertal timing may be more consequential for girls than boys. This difference could partially explain why previous research has found that the deviant peer pathway to maladjustment through perceptions of pubertal timing is specific to girls.

Measures of Pubertal Development

As reviewed in Mendle, Beltz, et al. (2019), pubertal status can be assessed with reports of the physical changes associated with puberty, such as breast development and voice changes. Objective measures of pubertal status can be obtained during a physical exam from a trained medical professional, or subjective measures can be obtained by asking youth about their physical development. To measure pubertal timing, researchers can standardize scores for pubertal status by age and sex. Thus, this method of measuring pubertal timing assesses the youth’s relative physical development compared to similar peers. Another method of assessing pubertal timing is to directly ask youth about their perceived development relative to other peers—henceforth referred to as perceived pubertal timing—instead of creating standardized scores of physical development.

Notably, perceived pubertal timing may be more strongly linked to youth adjustment than assessments of physical development (Mendle, 2014). For example, Carter et al. (2013) found that perceived pubertal timing was positively related to depression and delinquency, above and beyond the effects of physical assessments of puberty. Youth who perceive themselves as more mature than similar peers may begin to spend time with deviant peers regardless of their physical development. As such, perceived pubertal timing was used in the present study rather than traditional measures in which assessments of physical development (e.g., body hair growth) are standardized within age and sex. Though, standardized measures of subjective physical development were included as a covariate to examine the unique role of perceived pubertal timing.

The Present Study

The overall goal of the present study was to identify pathways from perceived pubertal timing (PPT) to marijuana and alcohol use initiation through deviant peer affiliation among a cohort of youth at high risk for substance use involvement. The present study is novel in three important ways: (1) extending the research on the role of pubertal timing on deviant peer affiliation to a longitudinal cohort that primarily consists of high-risk youth with a family history of substance use disorder, (2) examining if PPT is indirectly related to substance use through deviant peer affiliation above and beyond the influence of standardized scores based on physical development, and (3) examining if the indirect association of PPT to substance use through deviant peer affiliation is moderated by impulsiveness.

Consistent with the stage termination hypothesis, we hypothesized that PPT would be related to initial levels of deviant peer affiliation. Though there are competing hypotheses regarding developmental change in deviant peer affiliation, there has been more empirical support for selective persistence than the attenuation hypothesis. As such, we hypothesized that early PPT would be related to increases in deviant peer affiliation from early-to mid-adolescence.

We also examined pathways through which PPT is related to substance use. We hypothesized that early PPT would be related to substance use initiation through higher initial levels of and change in deviant peer affiliation. Moreover, we also hypothesized that impulsiveness would moderate the association between PTT and deviant peer affiliation. We also expected that this pathway to substance use would operate for girls but not for boys, given prior research (Marceau & Jackson, 2017; Westling et al., 2008).

Methods

Participants

The present study was a secondary analysis of a longitudinal cohort of 386 parent–adolescent dyads¹ (51.6% of the adolescents were girls) who were recruited into two groups: youth who have a biological father with a substance use disorder (79.02%) and youth whose parents and grandparents did not have any substance use disorder (20.98%). A higher proportion of high-risk youth were recruited. Specifically, one youth without a family history of substance use disorder was recruited to participate in the study for every three youth with a family history of substance use disorder. This sampling strategy was used to increase the probability of substance use initiation during the observation period. Most youth identified as Hispanic ethnicity (78.2%); 86.8% identified as White, 10.9% as Black, and the remainder (2.3%) identified as another race. Youth were excluded from the cohort of 386 (i.e., were not invited to participate in the study) if they had initiated substance use, had a psychiatric disorder (except for oppositional defiant disorder, attention-deficit/hyperactivity disorder, or conduct disorder because they tend to be comorbid with having a family history of substance use disorder), a significant developmental disability or cognitive impairment, or positive breath alcohol test, drug panel urinalysis, or pregnancy test at study entry.

All recruitment and study procedures were approved by the Institutional Review Board. To ensure the integrity and confidentiality of the data, informed consent and assent were obtained from parents and youth, respectively. A Certificate of Confidentiality was obtained from the Department of Health and Human Services. The cohort was recruited starting in 2008 in the south Texas area through the community via media announcements and flyers. Parent–adolescent dyads completed a baseline assessment when the youth was between 10 and 12 years of age (M = 11.54, SD = .91) and follow-up assessments every six months thereafter. Trained research assistants collected data from the youth and a parent (typically the mother) during structured interviews in a lab setting and administered self-reported questionnaires. To ensure privacy and confidentiality, the youth and parent completed appointments in separate rooms. See Ryan et al. (2016) for further details regarding recruitment and cohort description.

The data for the present study were restructured to treat time as discrete age from 13 to 16 years old at six-month intervals rather than the wave of assessment. Of the 386 participants in the initial cohort, 44 were excluded from the present study because they did not have available data for all exogenous (i.e., pubertal development, impulsiveness) and endogenous (i.e., deviant peer affiliation, substance use) variables for at least one time-point between the ages of 13 and 16 years old. This resulted in an analytic sample size of 342. Analyses revealed no difference between those included in the analysis (n = 342) and those excluded (n = 44) in demographics (i.e., gender, race/ethnicity, family history of substance use disorder) or baseline measures of variables used in the present study (i.e., pubertal status, impulsiveness, deviant peer affiliation). Attrition rates between study time points ranged from .9 to 12.2% (average = 7.9%), and the median number of time points completed by participants was six out of a possible seven visits.

Measures

Family History Status

Parents completed the Family History Assessment Module (Janca et al., 1992; Rice et al., 1995) that assessed if the youth had a family history of substance use disorder or other psychiatric disorders. If the youth had a biological father with a substance use disorder, they were classified as family-history positive (FH+, coded as 1). Conversely, if the youth had no parents or grandparents with a substance use disorder, they were classified as family-history negative (FH–, coded as 0).

Perceived Pubertal Timing (Age 13)

Perceived pubertal timing (PPT) was assessed with a single item from the Pubertal Development Scale (PDS; Petersen et al., 1988) asking youth “Do you think your development is any earlier or later than most other boys/girls your age?” with possible scores ranging from 1 = much later to 5 = much earlier. Thus, higher scores reflect earlier PPT relative to same-age, same-sex peers. Previous research has supported the reliability of the single PDS item as an index of PPT (Dubas et al., 1991; Graber et al., 2004).

Pubertal Status (Age 13)

Pubertal status was included as a covariate and measured with youth reports of the PDS, which reflects the degree of pubertal maturation with possible responses ranging from 1 = has not begun to 4 = seems completed. For boys and girls, pubertal status was assessed with items relating to growth in height, skin changes associated with puberty, and body hair growth (i.e., pubic, underarm). Additionally, girls responded to items that assessed breast growth and the onset of menarche (coded as 1 = no menarche or 4 = begun menarche), and boys also responded to items that assessed voice changes and facial hair growth. Thus, both girls and boys responded to five items total, two of which were sex-specific. Pubertal status was operationalized as the average of the five items with higher scores reflecting more advanced maturation (α = .62 for girls and α = .72 for boys). The correlation between PPT and pubertal status in the study sample was r = .23, p < .05.

Impulsiveness (Age 13)

Impulsiveness was assessed with youth reports of the 30-item Barratt Impulsiveness Scale (BIS), a well-established measure that captures the frequency of impulsive tendencies such as inattentiveness, difficulty planning, and poor self-control (Patton et al., 1995). An example item included “I do things without thinking” with responses ranging from 1 = rarely/never to 4 = almost always/always. The BIS has been previously demonstrated to be a valid and reliable measure of impulsiveness (Mathias et al., 2018; Stanford et al., 2009). Total scores were used in the present study, with higher scores indicating a greater propensity towards impulsiveness (α = .81).

Deviant Peer Affiliation (Ages 13–16)

Deviant peer affiliation (Loeber et al., 1998) was assessed with 15 items that asked adolescents how many of their close friends engaged in delinquent behavior (e.g., skipped school, used marijuana, hit someone) in the past six months with possible responses ranging from zero = none of them to 4 = all of them. First, scores were summed and then log-transformed due to the positive skewness of the distribution. Prior research has used a similar approach (Marceau & Jackson, 2017). For all analyses, log-transformed scores of deviant peer affiliation were used (Cronbach’s α = .83–.89).

Marijuana and Alcohol Use (Ages 13–16)

Youth were interviewed about their substance use by trained research assistants using the drug history questionnaire (DHQ; Sobell et al., 1995) and timeline follow-back measure (TLFB; Sobell & Sobell, 1992; Miech et al., 2020). For the DHQ, youth were asked about their lifetime use of various substances and their current pattern of use, including whether they used in the past six months. For the TLFB, youth were interviewed about the frequency and amount of substance use in the past six months. The interview was aided by a calendar that included holidays to improve the accuracy of recollection. At the beginning of appointments, youth also provided a urine sample for a drug panel analysis as an objective assessment of recent substance use.

Seven binary indicators were used to measure the onset of marijuana and alcohol use between the ages of 13 and 16 years old at a six-month interval. Substance use initiation was coded as negative (0 = Has not yet initiated substance use) if the youth reported no use during the DHQ or TLFB interviews and had a negative urinalysis (marijuana only). Conversely, substance use initiation was coded as positive (1 = Initiated substance use) if the youth reported use in the past six months during the DHQ or TLFB interviews or had a positive urinalysis for marijuana and did not report initiation at an earlier age. After substance use was initiated, the following age time points were coded as missing because the youth is no longer at risk of initiation. The seven binary indicators were used to model a discrete-time survival analysis (described in more detail in the analysis plan and Supplemental Material), which measured the propensity to initiate substance use between the ages of 13 and 16 years old.

The frequency of marijuana use at age 16 was also modeled as an outcome with possible responses including zero = No use (in the past six months), 1 = <1 use a month, 2 = ≤1 use in a week, and 3 = several times per week. Because alcohol use was infrequent (only five participants reported using alcohol once a week or more frequently), alcohol use at age 16 was dichotomized whether youth used it in the past six months (coded 1) or not (coded 0).

Analysis Plan

Analyses were conducted in Mplus 8.1 software (Muthén & Muthén, 2017). Full-information maximum likelihood estimation with robust standard errors (MLR) was used to account for missing data. The pattern of missing completely at random (MCAR) was met based on Little’s MCAR test (Little, 1988), $χ^{2}$ (260) = 261.40, p = .464. Substance use initiation was treated as a binary outcome; for that reason, a logit link function was used when the variable was included in analyses. The primary analysis that tested the main study hypotheses was performed via structural equation modeling to integrate the various analytic approaches, including growth curve modeling, survival analysis, and moderated mediation. Preliminary analyses were first conducted to assess the model fit of each component of the primary analysis separately. When applicable, model fit was assessed using the comparative fit index (CFI) and the root mean square error of approximation (RMSEA). A CFI of >.90 was considered to meet the threshold for acceptable model fit, whereas an RMSEA of <.05 and <.05 to .10 was considered to meet the threshold for good and moderate fit, respectively (Hu & Bentler, 1999). Bivariate correlations and descriptive statistics are provided in the supplemental material (see Table S1).

For the preliminary and primary analyses, a multi-group model was estimated to test for differences between girls and boys. Initially, all model parameters were freely estimated between girls and boys. Next, model parameters were systemically constrained to be equal between the two groups. A chi-square difference test was computed comparing the model with fewer parameters (i.e., more constraints) to the model with more parameters. If there was no significant decrease in model fit, then the model with fewer parameters was retained, suggesting there was no difference between girls and boys. Alternatively, if there was a significant decrease in model fit, then the model with more parameters was retained, suggesting there was a difference between girls and boys. For simplicity, the most parsimonious model is described, but all model comparisons are provided as Supplemental Material.

Preliminary Analyses

Growth Curve Model

A growth curve model was estimated to account for individual differences in the stability and developmental change in deviant peer affiliation. This was accomplished by estimating an intercept growth factor by fixing each of the seven age time points to one and a latent change growth factor by fixing the loading at age 13 to 0, fixing the loading at age 16 to 1, and freely estimating the intermediate loadings from age 13.5 to 15.5. Therefore, the intercept growth factor represents individual differences in deviant peer affiliation at age 13, and the latent change growth factor represents individual differences in the developmental change (linear and nonlinear) in deviant peer affiliation from age 13 to 16. Variance parameters were also estimated for the intercept and latent change factors, and a covariance between the intercept and latent change factor was freely estimated.

Discrete-Time Survival Analysis

Substance use initiation was measured as a latent variable via a discrete-time survival analysis (Muthén & Masyn, 2005; Raykov et al., 2017). Specifically, each factor loading for the seven binary indicators that measured the discrete age (e.g., 13.0, 13.5…16.0) at which the youth initiated substance use was fixed to one. The factor mean and variance was fixed to zero. The thresholds for each time-point were freely estimated; thus, they represent the time-specific conditional probability of initiating substance use, given that use was not initiated at an earlier age. The outcome of interest was the latent variable, which can be interpreted as the propensity to initiate substance use between the ages of 13 and 16 years old. Separate discrete-time survival analyses were conducted for marijuana and alcohol use. More details regarding the model specifications are provided as Supplemental Material.

Primary Analysis

For the primary analyses, PPT and impulsiveness were modeled to predict initial levels of and developmental change in deviant peer affiliation. An interaction term was also modeled to predict initial levels of and developmental change in deviant peer affiliation to test the hypothesis that impulsiveness would moderate the association between PPT and deviant peer affiliation. In turn, PPT, impulsiveness, the interaction term, and the deviant-peer growth factors were modeled to predict substance use. Mediation analyses were conducted to test if PPT was indirectly related to substance use through the deviant-peer growth factors (described below). Family history status and pubertal status were included as covariates. No covariances were estimated between the exogenous predictors (e.g., impulsiveness, PPT). Similar to the preliminary analyses, a multi-group model was estimated to determine if the relationship between the study variables differed between girls and boys. The syntax for the final model is available on OSF, https://osf.io/9hc74?view_only=ff18eae0679140ac9269c505608b1588. For brevity, the results focus on marijuana and alcohol use initiation. The findings pertaining to marijuana use frequency at age 16 and alcohol use at age 16 are kept brief, but full results are reported as supplemental material.

Moderated Mediation

Mediation was tested to examine if PPT indirectly predicted substance use initiation through deviant peer affiliation. The estimates for indirect effects were obtained with bias-corrected bootstrapped estimates, as described by Preacher and Hayes (2008). Rather than point estimates, 95% confidence intervals are reported because sampling distributions of indirect effects tend to be non-normal. The indirect effect was considered significant if neither the lower nor upper bound of the confidence interval contained zero.

Moderation was also tested to determine if the relationship between PPT and deviant peer affiliation is conditional on impulsiveness. To accomplish this, tests of simple slopes (Aiken & West, 1991) were conducted by centering impulsiveness at different values to determine the levels of impulsiveness at which PPT was related to deviant peer affiliation. Impulsiveness was centered at the mean, one standard deviation (SD) below the mean, and one SD above the mean. PPT was kept centered at the middle category (pubertal development is about the same as other girls/boys of a similar age).

Results

Preliminary Analyses

Deviant Peer Growth Model

The model that compared developmental change in deviant peer affiliation from age 13 to 16 years between girls and boys had an acceptable model fit, $χ^{2}$ = 75.86, df = 58, p = .058; CFI = .98; RMSEA = .04. As shown in Figure 1, girls reported higher levels of deviant peer affiliation at age 13 than boys. Both girls and boys increased in deviant peer affiliation from age 13 to 16 at the same rate. For girls and boys, there was significant variability in the intercept and change factors. Model comparisons are provided in Table S2, and final model estimates are provided in Table S3 in the supplemental material.

Survival Analysis

There was no difference in the likelihood of marijuana, $∆ χ^{2} (7) = 4.04, p = . 776$ , or alcohol use initiation from age 13 to 16, $∆ χ^{2} (7) = 9.95, p = . 191$ , between girls and boys (i.e., all parameters were constrained to be equal). As can be seen in Figure 2, the likelihood of marijuana and alcohol use initiation was similar and generally increased from age 13 to 16 years old. By age 16, 28.7% of girls and 29.3% of boys have initiated marijuana use, χ² (1) = .02, p = .900. By age 16, 27.3% of girls and 28.2% of boys have initiated alcohol use, χ² (1) = .03, p = .855.

Primary Analyses

Direct Effects for Deviant Peer Affiliation

Model comparisons for the structural relationships are provided in Table S4 in the supplemental material. Table 1 shows the results for marijuana use initiation, and Table 2 shows the results for alcohol use initiation. As shown in the tables and Figure 3, the interaction between PPT and impulsiveness predicting the deviant peer intercept factor was significant for girls (see the Girl Model column in the Effects for Deviant Peer Intercept Factor section). Tests of simple slopes revealed that PPT was positively related to the deviant peer intercept factor but only at the mean and one SD above the mean for impulsiveness. For boys (see Figure 4), neither PPT nor the interaction term was related to the deviant peer intercept factor (see the Boy Model column). For both girls and boys, PPT, impulsiveness, nor their interaction were related to the deviant peer change factor (see Effects for Deviant Peer Change Factor section). Overall, girls who perceive themselves to be more physically developed than other same-age girls were more likely to affiliate with deviant peers, and this relationship was moderated by impulsiveness.

Table 1.

Model Results for Marijuana Use Initiation.

Effects for Deviant Peer Intercept Factor	Girl Model			Boy Model
Effects for Deviant Peer Intercept Factor	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	.19a	.05	<.001	.06a	.09	.492
Impulsivity (IMP)	.88a	.14	<.001	.71a	.15	<.001
PPT × IMP	.31a	.15	.035	−.39a	.26	.126
Family history status	.27	.06	<.001	.27	.06	<.001
Pubertal status	.07	.07	.294	.07	.07	.294

Effects for Deviant Peer Change Factor	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	−.01	.08	.882	−.01	.08	.882
Impulsivity (IMP)	−.09	.16	.564	−.09	.16	.564
PPT × IMP	−.23	.19	.233	−.23	.19	.233
Family history status	.03	.11	.809	.03	.11	.809
Pubertal status	−.05	.09	.606	−.05	.09	.606

Effects for Marijuana Use Initiation	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	−.02	.18	.933	−.02	.18	.933
Impulsivity (IMP)	.07	.44	.883	.07	.44	.883
PPT × IMP	.42	.52	.421	.42	.52	.421
Deviant peer intercept factor	1.60	.31	<.001	1.60	.31	<.001
Deviant change factor	.67a	.51	.190	1.54a	.43	<.001
Family history status	1.00	.54	.065	1.00	.54	.065
Pubertal status	.46	.23	.040	.46	.23	.040

EST = unstandardized estimate, SE = standard error. For family history status, 0 = no family history of substance use disorder, 1 = family history of substance use disorder. PPT was centered at the middle category indicating perceived on-time maturation, and impulsivity was mean centered.

^aModel parameter was freely estimated between girls and boys.

Table 2.

Model Results for Alcohol Use Initiation.

Effects for Deviant Peer Intercept Factor	Girl Model			Boy Model
Effects for Deviant Peer Intercept Factor	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	.19a	.05	<.001	.07a	.09	.484
Impulsivity (IMP)	.87a	.14	<.001	.72a	.15	<.001
PPT × IMP	.30a	.15	.040	−.40a	.25	.112
Family history status	.27	.06	<.001	.27	.06	<.001
Pubertal status	.07	.07	.297	.07	.07	.297

Effects for Deviant Peer Change Factor	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	−.02	.08	.835	−.02	.08	.835
Impulsivity (IMP)	−.10	.17	.551	−.10	.17	.551
PPT × IMP	−.26	.19	.174	−.26	.19	.174
Family history status	.02	.11	.845	.02	.11	.845
Pubertal status	−.03	.09	.716	−.03	.09	.716

Effects for Alcohol Use Initiation	EST	SE	p	EST	SE	p
Perceived pubertal timing (PPT)	.14	.19	.473	.14	.19	.473
Impulsivity (IMP)	−.24	.52	.648	−.24	.52	.648
PPT × IMP	−.10	.66	.875	−.10	.66	.875
Deviant peer intercept factor	1.64	.33	<.001	1.64	.33	<.001
Deviant peer change factor	.34a	.46	.465	1.04a	.35	<.001
Family history status	.73	.52	.160	.73	.52	.160
Pubertal status	.34	.22	.125	.34	.22	.125

^aModel parameter was freely estimated between girls and boys.

Direct and Indirect Effects for Substance Use Initiation

For both girls and boys (see the Girl and Boy Model columns in Effects for Marijuana Use Initiation sections in Table 1 and Effects for Alcohol Use Initiation in Table 2), higher initial levels of deviant peer affiliation (i.e., intercept factor) were related to an increased likelihood of marijuana and alcohol use initiation. For boys only, a faster rate of developmental change in deviant peer affiliation (i.e., the latent change factor) was also related to an increased likelihood of marijuana and alcohol use initiation. While PPT was not directly related to marijuana or alcohol use initiation, there was a significant indirect effect that was conditional on impulsiveness. For girls only, the hypothesized indirect effect of PPT on marijuana and alcohol use initiation through higher initial levels of deviant peer affiliation was significant but only at average (i.e., at the mean) to high levels (i.e., one SD above the mean) of impulsiveness (see Table 3 for all indirect effects). This conditional indirect effect was also replicated for marijuana use frequency at age 16 and alcohol use at age 16. Specifically, for girls only, PPT was indirectly related to marijuana use frequency at age 16 and alcohol use at age 16 through higher initial levels of deviant peer affiliation at average to high levels of impulsiveness. See Tables S5 and S6 for the model estimates for marijuana use frequency at age 16 and alcohol use at age 16 and Tables S7 and S8 for the indirect effects. For boys, an indirect effect was not possible because PPT was unrelated to the deviant peer intercept or change factor, regardless of levels of impulsiveness.

Table 3.

95% Confidence Intervals for Indirect Effects Predicting Substance Use Initiation.

Indirect Effects Through Deviant Peer Intercept Factor	Model for Marijuana Use Initiation
	Girl Model			Boy Model
	EST	Lower Bound	Upper Bound	EST	Lower Bound	Upper Bound
Perceived pubertal timing
Low levels of impulsivity	.14	-.07	.34	.32	−.15	.73
Average levels of impulsivity	.31	.10	.50	.10	−.19	.46
High levels of impulsivity	.48	.12	.80	−.12	−.53	.53
Pubertal status	.11	−.13	.37	.11	−.13	.37
Impulsivity	1.40	.83	1.95	1.13	.49	1.77
Family history status	.44	.17	.70	.44	.17	.70
	Model for alcohol use initiation
Perceived pubertal timing
Low levels of impulsivity	.15	−.06	.35	.34	−.12	.75
Average levels of impulsivity	.32	.10	.53	.11	−.19	.49
High levels of impulsivity	.49	.13	.84	−.12	−.55	.46
Pubertal status	.12	−.11	.41	.12	−.11	.41
Impulsivity	1.43	.79	2.14	1.17	.57	1.85
Family history status	.45	.19	.73	.45	.19	.73

Note. EST = unstandardized estimate. For family history status, 0 = no family history of substance use disorder, 1 = family history of substance use disorder. Indirect effects are shown when impulsivity was centered at one standard deviation below the mean (low levels), at the mean (average levels), and at one standard deviation above the mean (high levels). Estimates in bold indicate a significant indirect effect (i.e., the 95% confidence interval does not contain zero).

Discussion

The present study aimed to examine whether the pathway from PPT to substance use through deviant peer affiliation was moderated by impulsiveness in a sample of youth, most of whom had a family history of substance use disorder. Additionally, we examined whether these processes were different for girls and boys. For both girls and boys, initial levels of deviant peer affiliation were directly related to an increased likelihood of initiating marijuana and alcohol use between the ages of 13 and 16. For boys only, a faster rate of developmental change in deviant peer affiliation was also related to an increased likelihood of marijuana and alcohol use initiation. For girls only, in support of the primary hypothesis, PPT was indirectly related to marijuana and alcohol use initiation through initial levels of deviant peer affiliation. Notably, this pathway was conditional on girls’ impulsiveness, such that PPT was only associated with initial levels of deviant peer affiliation at average to high levels of impulsiveness. The findings were also replicated for other substance use outcomes, including frequency of marijuana use at age 16 and alcohol use at age 16. Overall, the novel findings from the present study elucidated a developmental pathway from PPT to substance use through deviant peer affiliation in a sample at risk for developing substance use problems.

Contrary to our expectations, our findings do not support the selective persistence hypothesis, such that the association between PPT and deviant peer affiliation was observed during early adolescence but was unrelated to developmental change in deviant peer affiliation. Consistent with our expectations, our results support the stage termination hypothesis. This theory posits that individual differences in PPT would be accentuated during transitional developmental stages, such as early adolescence. In support of this notion, for girls only, we found that perceptions of early maturation were related to deviant peer affiliation during early adolescence. Critically, considering their chronological age, youth with perceptions of early maturation may be ill-equipped to resist the influence of deviant peers, which may explain the increased risk for the early initiation of substance use.

Our findings extend upon prior research that has examined the role of pubertal development in the deviant peer pathway to adolescent substance use in two unique ways. First, we found that perceptions of maturation relative to peers are also pertinent to the deviant peer pathway, even after controlling for self-reports of physical growth. Consistent with Mendle (2014), youth who perceive themselves as more physically mature may begin to affiliate with deviant peers because the social environment matches their self-perceptions. Second, our results build upon the literature by demonstrating that the relationship between PPT and deviant peer affiliation was moderated by trait impulsiveness. Prior research has primarily examined the possibility that contextual factors modify the association between pubertal timing and deviant peer affiliation. For example, parental monitoring (Marceau & Jackson, 2017; Westling et al., 2008) has been shown to reduce the likelihood of deviant peer affiliation. To our knowledge, this is the first study to demonstrate that impulsiveness also moderates this pathway. Thus, contextual and personality factors may affect the psychosocial consequences of perceived pubertal development.

Pubertal Timing and the Deviant Peer Pathway to Substance Use

Collectively, our findings elucidate a pathway from PPT to substance use for adolescent girls. PPT alone was insufficient for girls to affiliate with deviant peers. Rather, trait impulsiveness acted synergistically with PPT. Theory suggests that youth who are impulsive may experience difficulty maintaining relationships with conventional peer groups because they are uncooperative and prone to act out (Chapple, 2005; Cillessen et al., 1992; Gomes & Livesey, 2008) and may be drawn to deviant peer groups if they perceive themselves to be more physically developed (Caspi & Moffitt, 1991). Notably, the current investigation found that the association between PPT and deviant peer affiliation was present from average to high levels of impulsiveness. Thus, impulsiveness may not be a risk factor per se; rather, low impulsiveness may be a protective factor. Girls with perceptions of early maturation who can manage their impulsive tendencies are deterred from affiliating with deviant peers, perhaps because these youth maintain ties to conventional peer groups. Conversely, girls with perceptions of early pubertal development with average to high levels of impulsiveness may be drawn to deviant peers because they perceive themselves as more physically developed. This is problematic because there is a mismatch between the youth’s physical and cognitive development, such that early-maturing youth may perceive themselves to be more developed, but unable to resist the influence of peers who are engaging in delinquent behavior at a rate that is inappropriate for their chronological age.

The moderating role of impulsiveness in the association between PPT and deviant peer affiliation may be especially relevant to girls with a family history of substance use disorder. One consistent finding in the literature is that youth with a family history of substance use disorder tend to be more impulsive than youth without a family history of substance use disorder (Sanchez-Roige et al., 2016; Saunders et al., 2008; Wasserman et al., 2020). Considering this group difference, early-maturing girls with a family history of substance use disorder may be at the greatest risk of affiliating with deviant peers and subsequent substance use. Consequently, prevention efforts may be most effective for high-risk girls with difficulty managing their impulsive tendencies.

Our findings are largely consistent with previous research demonstrating a link between pubertal timing and delinquent behavior through deviant peer affiliation. Yet, some have found that this pathway is specific to girls (Marceau & Jackson, 2017; Westling et al., 2008), while others find no difference between girls and boys (Negriff & Trickett, 2012). The reason for gender differences may be related to the timing of changes in appearance due to puberty. The onset of puberty among girls typically involves more observable physical changes, such as breast development and a growth spurt, whereas boys typically experience less readily observable changes, like gonadarche. The more observable nature of pubertal development among girls may have particular salience due to social comparison and heightened awareness of their appearance differing from their peers, which may partly explain the differences in the findings between girls and boys.

Limitations and Strengths

While the findings from the present study extend upon previous research, there are limitations worth noting. First, while we controlled for youth reports of pubertal status, assessments of pubertal development conducted by a health professional are likely more objective, given that self-reports are still contingent on subjective perceptions. Regardless, it is intriguing that perceptions of maturation compared to same-age peers are weakly correlated with self-reports of pubertal status. Relatedly, there are components of pubertal development beyond timing (e.g., tempo) that may have implications for youth’s substance use that were not assessed in the present study. Second, all measures in the present study were youth reports; thus, shared-reporter variance potentially confounds any relationship. Youth may also be unwilling to disclose their substance use, though we mitigated this possibility by ensuring confidentiality, building a rapport with youth, and supplementing self-reports with a urinalysis. Youth may also under- or over-estimate the delinquent behavior of their peers (Meldrum et al., 2009; Rebellon & Modecki, 2014; Young & Weerman, 2013).

There are strengths of the present study as well. First, our sample was enriched by including youth with a family history of substance use disorder. This strategy allowed us to study pathways that may be more indicative of problems with substance use than general population sampling. However, a limitation of this strategy is that the results may not generalize to other populations because a convenience sample that consists primarily of high-risk youth was recruited for participation in the study. Second, all youth at age 13 had not initiated substance use; thus, we were able to determine developmental processes that existed before initiation or regular use. Lastly, the statistical rigor of the present investigation allowed for a robust test of hypotheses. Although, a limitation worth noting is that the final model was complex and estimated with a relatively small sample size. This limitation was amended by bootstrapping procedures to enhance model stability (Ievers-Landis et al., 2011). It is also worth noting that most model parameters were constrained to be equal between boys and girls (28 out of 32 model parameters were constrained), reducing model complexity.

Implications

The findings from the present study have implications for the measurement of pubertal timing. Previous research has primarily utilized measures of physical development (e.g., standardized or residual scores of pubertal status), particularly when examining the relationship between pubertal timing and deviant peer affiliation. As Mendle (2014) has theorized, perceptions of development may be more rooted in psychosocial processes; thus, they are more relevant to the changes in peer-group processes resulting from pubertal development. As such, researchers should carefully consider their research question to determine which measure of pubertal timing is most appropriate. If researchers are hypothesizing a biological mechanism, then measures of physical development may be more appropriate, whereas subjective measures may be more appropriate for hypotheses pertaining to psychosocial mechanisms (Dorn et al., 2006; Mendle, Beltz, et al., 2019). We do note that self-reports of physical development are still, in part, subjective and based on self-perception. In the present study, self-reports of pubertal status were correlated with PPT, albeit weakly (r = .23). Overall, puberty is a multifaceted construct that can be measured in a variety of ways. Neither method of assessing puberty (objective or subjective) is innately superior, but depends on the study context. Careful consideration is required about whether more objective markers of pubertal maturation or youth perceptions of their own development are best suited to address the research question (see Mendle, Beltz, et al., 2019 for a review).

The present study also has implications regarding the prevention of adolescent substance use. Puberty is a normal aspect of adolescent development; therefore, prevention efforts should concentrate on the context through which puberty occurs (Graber et al., 2010). The parent–adolescent relationship is one context that can protect against maladjustment during the pubertal transition. Specifically, parental monitoring has been consistently shown to prevent deviant peer affiliation and reduce the negative influence of deviant peers (Marceau & Jackson, 2017; Schofield et al., 2015; Westling et al., 2008). Other research has shown that teacher support can also attenuate the influence of deviant peers on youth adjustment (Wang & Dishion, 2012). These findings imply that parents and teachers can act as protective factors against the influence of deviant peers and reduce the probability of substance use initiation, which may be particularly important for early-maturing girls who are impulsive. However, it is possible that parents and teachers might not be sensitive to respond to the needs of early-maturing girls who have high levels of impulsiveness. This might be especially true for youth who have a parent with a substance use disorder. Parents with a substance use disorder may be unable to adapt their parenting to meet the needs of their child (Keller et al., 2008); for example, increase monitoring efforts to reduce the likelihood that their child affiliates with deviant peers.

Another approach to preventing adolescent substance use based on the findings from the present study is to utilize interventions designed to improve self-regulation. Such interventions have shown to be highly effective at improving components of self-regulation that are necessary for adaptive social skills, such as conflict resolution (Pandey et al., 2018). Early maturing girls, in particular, may benefit from these interventions by learning to regulate their impulsive tendencies through improvements in self-regulation (Nigg, 2017; Pandey et al., 2018). Consequently, early-maturing girls may be less likely to affiliate with deviant peers and initiate substance use because they spend time with peer groups less inclined to engage in delinquent behavior. Overall, the present study suggests two potential targets to prevent adolescent marijuana use: parenting strategies to monitor their child’s behavior and utilizing interventions designed to improve youth’s social skills and ability to regulate their impulsive tendencies.

Conclusion

In conclusion, puberty is a dynamic biopsychosocial event, and the broader context (including both environment and personality) through which it occurs should be considered to understand its impact on adolescent well-being. The present study demonstrates that girls’ impulsiveness moderated the association between PPT and deviant peer affiliation. Specifically, perceptions of early pubertal timing were only associated with deviant peer affiliation at average to high levels of impulsiveness. Thus, the capacity to regulate impulsive behavior has implications for peer-group processes. In turn, deviant peer affiliation was associated with marijuana and alcohol use initiation. To summarize, these findings elucidate a pathway from early pubertal timing to substance use through deviant peer affiliation, although the capacity to regulate impulsive tendencies may buffer against this risk pathway.

Supplemental Material

Supplemental Material - Perceived Pubertal Timing and Deviant Peer Processes Predicting Substance Use Initiation: The Moderating Role of Impulsiveness

Supplemental Material for Perceived Pubertal Timing and Deviant Peer Processes Predicting Substance Use Initiation: The Moderating Role of Impulsiveness by A. M. Wasserman, T. E. Karns-Wright, C. W. Mathias, T. J. Moon, N. Hill-Kapturczak, and D. M. Dougherty in The Journal of Early Adolescence.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by NIDA of the National Institutes of Health under award numbers R01-DA026868 and T32-DA031115. Dr. Dougherty is also supported by the William and Marguerite Wurzbach Distinguished Professorship. The funding entities had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

ORCID iDs

A. M. Wasserman

C. W. Mathias

Supplemental Material

Supplemental material for this article is available online.

Note

Author Biographies

A. M. Wasserman is a Visiting Assistant Professor at the Ohio State University. His research examines the role of personality traits in the etiology of adolescent risk behavior, particularly among at-risk youth. He also teaches quantitative courses and aims to apply cutting-edge quantitative methods to his program of research.

T. E. Karns-Wright’s area of expertise is substance use treatment research, with formal training and experience in psychology and analyses of complex datasets generated in clinical research projects. She has substantial experience in clinical trials, data management, analysis, and interpretation and she is the Director of Data and Analytics for Be Well Texas, a network of substance use treatment providers in the state of Texas funded by Texas Health and Human Services.

C. W. Mathias is a behavioral health researcher and Associate Professor at the University of Texas Health Science Center in San Antonio. His research has been influential in defining how impulsiveness is measured and interpreted. For over 20 years, Dr Mathias has applied his expertise advancing the understanding of the personal characteristics underlying aggressive, suicidal, and substance use behaviors that can result in contact with law enforcement. This scholarship has transitioned from study of individual risk, to focus on how interaction with safety-net systems impact health outcomes, particularly for those with mental illness. His efforts now focus on analyses and communications translating research to policy.

T. J. Moon, Ph. D., is an Assistant Professor in the Department of Psychiatry and Behavioral Sciences at the UT Health San Antonio. He earned his doctoral degree in journalism and mass communication from the University of Wisconsin, Madison. His areas of research interest include the assessment of the roles of social support in the prevention and treatment of substance use and the development of technology-implemented treatment using mobile devices and remote sensors.

N. Hill-Kapturczack is an Associate Professor at the University of Texas Health Science Center in San Antonio in the Department of Psychiatry and Behavioral Sciences. She has a broad background in chemistry and molecular biology and have developed a particular interest and expertise in the detection and quantification of drug and alcohol use (through both biological and self-report methods).

D. M. Dougherty is Professor and Chair of the Department of Psychology at the University of North Texas. His research focuses broadly on the development of impulse control and factors (e.g., substance abuse) that may negatively impact these trajectories.

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Sage Publications, Inc.

Bagwell

C. L.

Molina

B. S. G.

Pelham

W. E.

Hoza

(2001). Attention-deficit hyperactivity disorder and problems in peer relations: Predictions from childhood to adolescence. Journal of the American Academy of Child & Adolescent Psychiatry, 40(11), 1285–1292. https://doi.org/10.1097/00004583-200111000-00008

Brechwald

W. A.

Prinstein

M. J.

(2011). Beyond homophily: A decade of advances in understanding peer influence processes. Journal of Research on Adolescence, 21(1), 166–179. https://doi.org/10.1111/j.1532-7795.2010.00721.x

Bucci

Staff

(2020). Pubertal timing and adolescent delinquency. Criminology, 58(3), 537–567. https://doi.org/10.1111/1745-9125.12245

Cance

J. D.

Ennett

S. T.

Morgan-Lopez

A. A.

Foshee

V. A.

Talley

A. E.

(2013). Perceived pubertal timing and recent substance use among adolescents: A longitudinal perspective. Addiction, 108(10), 1845–1854. https://doi.org/10.1111/add.12214

Carter

Halawah

Trinh

S. L.

(2018). Peer exclusion during the pubertal transition: The role of social competence. Journal of Youth and Adolescence, 47(1), 121–134. https://doi.org/10.1007/s10964-017-0682-8

Carter

Silverman

W. K.

Jaccard

(2013). Race and perceived pubertal transition effects on girls’ depressive symptoms and delinquent behaviors. Journal of Youth and Adolescence, 42(8), 1155–1168. https://doi.org/10.1007/s10964-012-9885-1

Caspi

Moffitt

T. E.

(1991). Individual differences are accentuated during periods of social change: The sample case of girls at puberty. Journal of Personality and Social Psychology, 61(1), 157–168. https://doi.org/10.1037//0022-3514.61.1.157

Castellanos-Ryan

Parent

Vitaro

Tremblay

R. E.

Séguin

J. R.

(2013). Pubertal development, personality, and substance use: A 10-year longitudinal study from childhood to adolescence. Journal of Abnormal Psychology, 122(3), 782–796. https://doi.org/10.1037/a0033133

10.

Chapple

C. L.

(2005). Self-control, peer relations, and delinquency. Justice Quarterly, 22(1), 89–106. https://doi.org/10.1080/0741882042000333654

11.

Cillessen

A. H. N.

van Ijzendoorn

H. W.

van Lieshout

C. F. M.

Hartup

W. W.

(1992). Heterogeneity among peer-rejected boys: Subtypes and stabilities. Child Development, 63(4), 893. https://doi.org/10.2307/1131241

12.

Copeland

Shanahan

Miller

Costello

E. J.

Angold

Maughan

(2010). Outcomes of early pubertal timing in young women: A prospective population-based study. American Journal of Psychiatry, 167(10), 1218–1225. https://doi.org/10.1176/appi.ajp.2010.09081190

13.

Costello

E. J.

Sung

Worthman

Angold

(2007). Pubertal maturation and the development of alcohol use and abuse. Drug and Alcohol Dependence, 88(Suppl 1), 50–59. https://doi.org/10.1016/j.drugalcdep.2006.12.009

14.

Dick

D. M.

Rose

R. J.

Viken

R. J.

Kaprio

(2000). Pubertal timing and substance use: Associations between and within families across late adolescence. Developmental Psychology, 36(2), 180–189. https://doi.org/10.1037/0012-1649.36.2.180

15.

Dishion

T. J.

Spracklen

K. M.

Andrews

D. W.

Patterson

G. R.

(1996). Deviancy training in male adolescent friendships. Behavior Therapy, 27(3), 373–390. https://doi.org/10.1016/S0005-7894(96)80023-2

16.

Dorn

L. D.

Dahl

R. E.

Woodward

H. R.

Biro

(2006). Defining the boundaries of early adolescence: A user’s guide to assessing pubertal status and pubertal timing in research with adolescents. Applied Developmental Science, 10(1), 30–56, https://doi.org/10.1207/s1532480xads1001_3

17.

Dubas

J. S.

Graber

J. A.

Petersen

A. C.

(1991). A longitudinal investigation of adolescents’ changing perceptions of pubertal timing. Developmental Psychology, 27(4), 580–586. https://doi.org/10.1037/0012-1649.27.4.580

18.

Fergusson

D. M.

Vitaro

Wanner

Brendgen

(2007). Protective and compensatory factors mitigating the influence of deviant friends on delinquent behaviours during early adolescence. Journal of Adolescence, 30(1), 33–50. https://doi.org/10.1016/j.adolescence.2005.05.007

19.

Natsuaki

M. N.

(2009). In search of explanations for early pubertal timing effects on developmental psychopathology. Current Directions in Psychological Science, 18(6), 327–331. https://doi.org/10.1111/j.1467-8721.2009.01661.x

20.

Gomes

Livesey

(2008). Exploring the link between impulsivity and peer relations in 5- and 6-year-old children. Child: Care, Health and Development, 34(6), 763–770. https://doi.org/10.1111/j.1365-2214.2008.00878.x

21.

Graber

J. A.

Nichols

T. R.

Brooks-Gunn

(2010). Putting pubertal timing in developmental context: Implications for prevention. Developmental Psychobiology, 52(3), 254–262. https://doi.org/10.1002/dev.20438

22.

Graber

J. a

Seeley

J. R.

Brooks-Gunn

Lewinsohn

P. M.

(2004). Is pubertal timing associated with psychopathology in young adulthood. Journal of the American Academy of Child & Adolescent Psychiatry, 43(6), 718–726. https://doi.org/10.1097/01.chi.0000120022.14101.11

23.

Hay

Widdowson

Young

B. C.

(2018). Self-control stability and change for incarcerated juvenile offenders. Journal of Criminal Justice, 56, 50–59. https://doi.org/10.1016/j.jcrimjus.2017.08.008.

24.

Haynie

D. L.

Piquero

A. R.

(2006). Pubertal development and physical victimization in adolescence. Journal of Research in Crime and Delinquency, 43(1), 3–35. https://doi.org/10.1177/0022427805280069

25.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

26.

Ievers-Landis

C. E.

Burant

C. J.

Hazen

(2011). The concept of bootstrapping of structural equation models with smaller samples: An illustration using mealtime rituals in diabetes management. Journal of Developmental and Behavioral Pediatrics, 32(8), 619–626. https://doi.org/10.1097/DBP.0b013e31822bc74f

27.

Janca

Bucholz

Janca

(1992). Family history assessment module. Washinton Univeristy School of Medicine.

28.

Jormanainen

Fröjd

Marttunen

Kaltiala-Heino

(2014). Is pubertal timing associated with involvement in bullying in middle adolescence? Health Psychology and Behavioral Medicine, 2(1), 144–159. https://doi.org/10.1080/21642850.2014.881259

29.

Keller

P. S.

Cummings

E. M.

Davies

P. T.

Mitchell

P. M.

(2008). Longitudinal relations between parental drinking problems, family functioning, and child adjustment. Development and Psychopathology, 20(1), 195–212. https://doi.org/10.1017/S0954579408000096

30.

Klopack

E. T.

Simons

R. L.

Simons

L. G.

(2020). Puberty and girls’ delinquency: A test of competing models explaining the relationship between pubertal development and delinquent behavior. Justice Quarterly, 37(1), 25–52. https://doi.org/10.1080/07418825.2018.1472291

31.

Little

R. J. A.

(1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Association, 83(404), 1198–1202. https://doi.org/10.1080/01621459.1988.10478722

32.

Loeber

Farrington

D. P.

Stouthamer-Loeber

Van Kammer

(1998). Antisocial behavior and mental health problems: Exploratory factors in childhood and adolescence. Lawrence Erlbaum Associates.

33.

Magnusson

Stattin

Allen

V. L.

(1985). Biological maturation and social development: A longitudinal study of some adjustment processes from mid-adolescence to adulthood. Journal of Youth and Adolescence, 14(4), 267–283. https://doi.org/10.1007/BF02089234

34.

Marceau

Jackson

(2017). Deviant peers as a mediator of pubertal timing–substance use associations: The moderating role of parental knowledge. Journal of Adolescent Health, 61(1), 53–60. https://doi.org/10.1016/j.jadohealth.2016.12.020

35.

Mathias

C. W.

Stanford

M. S.

Liang

Goros

Charles

N. E.

Sheftall

A. H.

Mullen

Hill-Kapturczak

Acheson

Olvera

R. L.

Dougherty

D. M.

(2018). A test of the psychometric characteristics of the BIS-brief among three groups of youth. Psychological Assessment, 30(7), 847–856. https://doi.org/10.1037/pas0000531

36.

Meldrum

R. C.

Young

J. T. N.

Weerman

F. M.

(2009). Reconsidering the effect of self-control and delinquent peers: Implications of measurement for theoretical significance. Journal of Research in Crime and Delinquency, 46(3), 353–376. https://doi.org/10.1177/0022427809335171

37.

Mendle

(2014). Beyond pubertal timing: New directions for studying individual differences in development. Current Directions in Psychological Science, 23(3), 215–219. https://doi.org/10.1177/0963721414530144

38.

Mendle

Beltz

A. M.

Carter

Dorn

L. D.

(2019). Understanding puberty and its measurement: Ideas for research in a new generation. Journal of Research on Adolescence, 29(1), 82–95. https://doi.org/10.1111/jora.12371

39.

Mendle

Ferrero

(2012). Detrimental psychological outcomes associated with pubertal timing in adolescent boys. Developmental Review, 32(1), 49–66. https://doi.org/10.1016/j.dr.2011.11.001

40.

Mendle

Ryan

R. M.

McKone

K. M. P.

(2019). Early menarche and internalizing and externalizing in adulthood: Explaining the persistence of effects. Journal of Adolescent Health, 65(5), 599–606. https://doi.org/10.1016/j.jadohealth.2019.06.004

41.

Mendle

Turkheimer

Emery

R. E.

(2007). Detrimental psychological outcomes associated with early pubertal timing in adolescent girls. Developmental Review, 27(2), 151–171. https://doi.org/10.1016/j.dr.2006.11.001

42.

Miech

R. A.

Johnston

L. D.

O’Malley

P. M.

Bachman

J. G.

Schulenberg

J. E.

Patrick

M. E.

(2020). Monitoring the Future national survey results on drug use, 1975–2019: Volume I, Secondary school students. Institute for Social Research, The University of Michigan.

43.

Moffitt

T. E.

(1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100(4), 674–701. https://doi.org/10.1037/0033-295x.100.4.674, http://www.ncbi.nlm.nih.gov/pubmed/8255953

44.

Muthén

Masyn

(2005). Discrete-time survival mixture analysis. Journal of Educational and Behavioral Statistics, 30(1), 27–58. https://doi.org/10.3102/10769986030001027

45.

Muthén

L. K.

Muthén

B. O.

(2017). Mplus user’s guide (8th ed.). Muthén and Muthén.

46.

Negriff

Trickett

P. K.

(2012). Peer substance use as a mediator between early pubertal timing and adolescent substance use: Longitudinal associations and moderating effect of maltreatment. Drug and Alcohol Dependence, 126(1–2), 95–101. https://doi.org/10.1016/j.drugalcdep.2012.04.018

47.

Nigg

J. T.

(2017). Annual research review: On the relations among self-regulation, self-control, executive functioning, effortful control, cognitive control, impulsivity, risk-taking, and inhibition for developmental psychopathology. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 58(4), 361–383. https://doi.org/10.1111/jcpp.12675

48.

Pandey

Hale

Das

Goddings

A. L.

Blakemore

S. J.

Viner

R. M.

(2018). Effectiveness of universal self-regulation-based interventions in children and adolescents a systematic review and meta-analysis. JAMA Pediatrics, 172(6), 566–575. https://doi.org/10.1001/jamapediatrics.2018.0232

49.

Patton

J. H.

Stanford

M. S.

Barratt

E. S.

(1995). Factor structure of the Barratt impulsiveness scale. Journal of Clinical Psychology, 51(6), 768–774. https://doi.org/10.1002/1097-4679(199511)51:6<768::aid-jclp2270510607>3.0.co;2-1

50.

Petersen

A. C.

Crockett

Richards

Boxer

(1988). A self-report measure of pubertal status: Reliability, validity, and initial norms. Journal of Youth and Adolescence, 17(2), 117–133. https://doi.org/10.1007/BF01537962

51.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. https://doi.org/10.3758/BRM.40.3.879

52.

Raykov

Gorelick

P. B.

Zajacova

Marcoulides

G. A.

(2017). On the potential of discrete time survival analysis using latent variable modeling: An application to the study of the vascular depression hypothesis. Structural Equation Modeling: A Multidisciplinary Journal, 24(6), 926–935. https://doi.org/10.1080/10705511.2017.1315305

53.

Rebellon

C. J.

Modecki

K. L.

(2014). Accounting for projection bias in models of delinquent peer influence: The utility and limits of latent variable approaches. Journal of Quantitative Criminology, 30(2), 163–186. https://doi.org/10.1007/s10940-013-9199-9

54.

Rice

J. P.

Reich

Bucholz

K. K.

Neuman

R. J.

Fishman

Rochberg

Hesselbrock

V. M.

Nurnberger

J. I.

Schuckit

M. A.

Begleiter

(1995). Comparison of direct interview and family history diagnoses of alcohol dependence. Alcoholism: Clinical and Experimental Research, 19(4), 1018–1023. https://doi.org/10.1111/j.1530-0277.1995.tb00983.x

55.

Ryan

S. R.

Acheson

Charles

N. E.

Lake

S. L.

Hernandez

D. L.

Mathias

C. W.

Dougherty

D. M.

(2016). Clinical and social/environmental characteristics in a community sample of children with and without family histories of substance use disorder in the san Antonio area: A descriptive study. Journal of Child & Adolescent Substance Abuse, 25(4), 327–339. https://doi.org/10.1080/1067828X.2014.999202

56.

Sanchez-Roige

Stephens

D. N.

Duka

(2016). Heightened impulsivity: Associated with family history of alcohol misuse, and a consequence of alcohol intake. Alcoholism: Clinical and Experimental Research, 40(10), 2208–2217. https://doi.org/10.1111/acer.13184

57.

Saunders

Farag

Vincent

A. S.

Collins

F. L.

Sorocco

K. H.

Lovallo

W. R.

(2008). Impulsive errors on a Go-NoGo reaction time task: Disinhibitory traits in relation to a family history of alcoholism. Alcoholism: Clinical and Experimental Research, 32(5), 888–894. https://doi.org/10.1111/j.1530-0277.2008.00648.x

58.

Schofield

T. J.

Conger

R. D.

Robins

R. W.

(2015). Early adolescent substance use in Mexican origin families: Peer selection, peer influence, and parental monitoring. Drug and Alcohol Dependence, 157, 129–135. https://doi.org/10.1016/j.drugalcdep.2015.10.020

59.

Senia

J. M.

Donnellan

M. B.

Neppl

T. K.

(2018). Early pubertal timing and adult adjustment outcomes: Persistence, attenuation, or accentuation? Journal of Adolescence, 65, 85–94. https://doi.org/10.1016/j.adolescence.2018.03.003

60.

Skoog

Kapetanovic

(2022). The role of pubertal timing in the development of peer victimization and offending from early- to mid-adolescence. The Journal of Early Adolescence, 42(1), 5–32. https://doi.org/10.1177/02724316211002265

61.

Snyder

Prichard

Schrepferman

Patrick

M. R.

Stoolmiller

(2004). Child impulsiveness-inattention, early peer experiences, and the development of early onset conduct problems. Journal of Abnormal Child Psychology, 32(6), 579–594. https://doi.org/10.1023/B:JACP.0000047208.23845.64

62.

Sobell

L. C.

Kwan

Sobell

M. B.

(1995). Reliability of a drug history questionnaire (DHQ). Addictive Behaviors, 20(2), 233–241. https://doi.org/10.1016/0306-4603(94)00071-9

63.

Sobell

L. C.

Sobell

M. B.

(1992). Timeline follow-back. In Litten

R. Z.

Allen

J. P.

(Eds.), Measuring alcohol consumption (pp. 41–72). Humana Press. https://doi.org/10.1007/978-1-4612-0357-5_3

64.

Stanford

M. S.

Mathias

C. W.

Dougherty

D. M.

Lake

S. L.

Anderson

N. E.

Patton

J. H.

(2009). Fifty years of the Barratt impulsiveness scale: An update and review. Personality and Individual Differences, 47(5), 385–395. https://doi.org/10.1016/j.paid.2009.04.008

65.

Stattin

Kerr

Skoog

(2011). Early pubertal timing and girls’ problem behavior: Integrating two hypotheses. Journal of Youth and Adolescence, 40(10), 1271–1287. https://doi.org/10.1007/s10964-011-9696-9

66.

Steinberg

Monahan

K. C.

(2007). Age differences in resistance to peer influence. Developmental Psychology, 43(6), 1531–1543. https://doi.org/10.1037/0012-1649.43.6.1531

67.

Ullsperger

J. M.

Nikolas

M. A.

(2017). A meta-analytic review of the association between pubertal timing and psychopathology in adolescence: Are there sex differences in risk? Psychological Bulletin, 143, (9), 903–938. https://doi.org/10.1037/bul0000106

68.

Wang

M.-T.

Dishion

T. J.

(2012). The trajectories of adolescents’ perceptions of school climate, deviant peer affiliation, and behavioral problems during the middle school years. Journal of Research on Adolescence, 22(1), 40–53. https://doi.org/10.1111/j.1532-7795.2011.00763.x

69.

Wasserman

A. M.

Mathias

C. W.

Hill-Kapturczak

Karns-Wright

T. E.

Dougherty

D. M.

(2020). The development of impulsivity and sensation seeking: Associations with substance use among at-risk adolescents. Journal of Research on Adolescence, 30(4), 1051–1066. https://doi.org/10.1111/jora.12579

70.

Westling

Andrews

J. A.

Hampson

S. E.

Peterson

(2008). Pubertal timing and substance use: The effects of gender, parental monitoring and deviant peers. Journal of Adolescent Health, 42(6), 555–563. https://doi.org/10.1016/j.jadohealth.2007.11.002

71.

Wolf

R. M.

Long

(2016). Pubertal development. Pediatrics in Review, 37(7), 292–300. https://doi.org/10.1542/pir.2015-0065

72.

Young

J. T. N.

Weerman

F. M.

(2013). Delinquency as a consequence of misperception: Overestimation of friends’ delinquent behavior and mechanisms of social influence. Social Problems, 60(3), 334–356. https://doi.org/10.1525/sp.2013.60.3.334

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.45 MB