Take a look at the Recent articles

Abstract

Previous studies on parental support have consistently shown it predicts lower adolescent alcohol use, yet findings regarding the influence of parental monitoring have been mixed. The current study aims to resolve this concern while examining peer selection as a mediator of both parenting factors. The current study used structural equation modeling and bootstrapping mediation analysis on data from 3,027 youth across three waves of the Adolescent Alcohol Prevention Trial to examine these factors. We tested a latent path model where the effect of parental support and monitoring in 7^th grade on adolescent alcohol use in 9^th grade was hypothesized to be mediated by best friends’ alcohol use in 8^th grade. Results: Higher parental support in seventh grade predicted lower adolescent alcohol use in 9^th grade, mediated by lower best friends’ use in eighth grade (ab = -0.025, CI = [-0.152, -0.003]). Yet parental monitoring in seventh grade did not predict alcohol use in 9^th grade when parental support was included as a co-predictor in the model (ab = 0.018, CI = -0.135 – 0.025). There was also no significant mediation effect for the monitoring to youth drinking path. Adolescent’s closeness with their parents may direct them to choose non-drinking friends, which leads to lower alcohol use in high school. Previously suggested effects of parental monitoring may be accounted for by support from parents.

Keywords

Parenting, adolescent alcohol use, parental support, monitoring, peers, peer selection

How to best raise a child may be one of life’s most difficult and debated questions. How can one guarantee that the most feared scenarios are widely evaded-situations like criminality, pregnancy, or underage drinking? The current study aims to address this question by attending to the latter problem: How might parents best curb youth alcohol use? In the literature, this age-old parenting question has recently resurfaced after a decade of being woefully understudied. In recent years debate has arisen about the importance of properly defining parental monitoring. Many argue for differentiating the effects of parent efforts to monitor child activities, parents’ actual knowledge of the child’s alcohol behaviors, and the child’s honest disclosure of those behaviors [1]. Yet across the sizable literature on parenting effects, there are generally two broad constructs theorized to be vital for socialization. In addition to parental control (monitoring, supervision, rules, and discipline), there is parental support [2].

The importance of a combined parental support and monitoring model

Both parental support and control have been theorized to exert notable influence on child outcomes. Parental support has been defined as behaviors toward the child that indicate s/he is loved and valued. It can include praise, approval, encouragement, assistance, companionship, physical affection, and the conveyance of warmth, acceptance, and personal worth to children [2]. Theoretically, warmth and assistance forge powerful bonds between children and parents and promotes adolescents’ social competence, identification with their parents, and development of autonomy [3]. Complementing support, parental control has been defined as “parental behaviors toward children that are intended to direct children’s behavior in a manner acceptable to parents” [2]. Monitoring—a subtype of control—is parents’ active awareness and regulation of children’s activity [4]. The current study focuses on monitoring due to its special relation to support, as we will later discuss. Parental support and control have been shown to have major effects on many forms of youth deviancy [5-7]. Underage drinking is itself a delinquent act and has been repeatedly linked to broader delinquency and other drug use [8]. Such a link suggests the many findings regarding parenting effects on general delinquency may be relevant for alcohol use.

Among forms of delinquency, there is much literature linking parenting to underage drinking. Alcohol use among adolescents is one of the most detrimental and prevalent public health concerns for youth, both in the United States and internationally [9-11]. By the final year of high school, nearly 75% of teenagers have consumed alcohol with 25% engaging in heavy drinking [10]. Understanding the causes and correlates of this phenomenon is important for engaging effectively in its prevention. Among these factors, parenting’s influence has been prominent in past research, but could be notably extended and improved.

Most studies have addressed support and monitoring separately without comparing them within the same population, method, and statistical model. Those focusing solely on support have repeatedly shown it to be a promotive factor predicting lower alcohol use in adolescents [12-17]. Yet findings are mixed for monitoring. Although some studies showed that greater monitoring predicted lower drinking, a significant number did not find such a link [18-32]. For delinquent behaviors in general, several more recent studies have failed to find longitudinal linkages between monitoring and later delinquency [33-35]. In fact, some found monitoring predicted greater later delinquency [36,37]. Thus, whereas support’s impact appears clear, monitoring’s power is yet uncertain.

There are possible reasons why findings may be inconsistent for monitoring when support and monitoring are examined separately. Any apparent effects of parental monitoring may actually be more attributable to the influence of parental support: Children may seek to please, follow, and respect the monitored limits of parents with whom they feel close, as well as be more forthcoming about their true activities and whereabouts. Theorists have argued that higher support generally allows for greater control; the two cannot be separated [38]. Yet without controlling for support, the effect of monitoring is uncertain. Thus, it is most important to review studies that include both variables in the same model. However, most combined support and control studies have not examined monitoring specifically. For example, several grouped together various forms of control as restriction or rule-enforcement and thus did not capture monitoring [12,39,40]. Another study grouped monitoring with rule-enforcement under a singular score [41]. Furthermore, used only a single monitoring item that asked solely about mothers’ knowledge of the peers with whom their child spent time outside the home [42].

When support and monitoring have been measured purely and separately in the same model, studies improperly operationalized monitoring as honesty of disclosure by adolescents [2,43]. In fact, it has been suggested that the unintended assessment of child disclosure may account for monitoring’s effect when it does appear significant [31,44]. Researchers have proposed that parents’ intentional tracking of adolescents through solicitating child report is not the factor that reduces drinking likelihood. Rather, it is the honest disclosure of information by the child—willingness to be truthful with parents [45]. In fact, Kerr, et al. [36] found that adolescent disclosure was a significant longitudinal predictor of both parent’s knowledge of problem behaviors and actual delinquency in both single- and cross-rater models. Yet neither of their models’ monitoring measures—parental control or solicitation—were significant predictors of either knowledge or delinquency. This highlights the important distinction between solicitation of child information by the parent and the child’s willingness to disclose that information without deception. Parent knowledge of the child’s actual whereabouts, friends, and behavior and its influence on alcohol use outcomes are dependent on the child’s disclosure. When a child perceives a parent to be supportive and close, s/he may be more likely to give truthful disclosure. If parental monitoring is improperly assessed as disclosure instead of solicitation and regulation, significant promotive effects of “monitoring” may arise—effects that are better attributable to support. The optimal test of whether higher monitoring predicts lower drinking should 1) assess monitoring as parental solicitation and regulation rather than child disclosure and 2) include parental support to account for its influence alongside monitoring. Thus, no previous study has sufficiently examined the prediction of youth alcohol use by parental support and monitoring together with proper construct assessment.

Peer selection as the mechanism of parenting effects on drinking

A much larger problem is the lack of research on the mechanisms by which supportive, aware parenting protects against alcohol use [46]. Socialization theory again offers possible means of influence. In this theory, socialization starts with parents but grows increasingly peer-emphasized throughout adolescence. Many explicitly family-based models of substance use etiology propose that affiliation with deviant peers is an immediate antecedent of substance use (e.g., Family Interactional Theory, Social Coercion Theory, Social Control Theory, etc. [47]. In fact, a large body of empirical work has uncovered a positive relationship between the delinquency of an adolescent’s peers and the adolescent’s own alcohol use [48-52]. Some studies even find that peers have a greater influence on adolescent drinking than any parenting factor [28]. In fact, Nash, et al. [53] found that a promotive effect of positive family environment predicted weaker strength of peer influence, which then later predicted lower alcohol behaviors (although no mediation analyses were presented). Thus, the impact of parenting factors on youth alcohol use may be partly explained by the effect of parenting on selection of peers and their shared activities-activities that may include drinking.

Examination of both parents and peers suggest parenting factors lead to adolescent bonding with delinquent peers, although no mediation study has addressed support and monitoring together [54,55]. Both higher parental closeness and monitoring have predicted less involvement with deviant peers. Nonetheless, parental support has received the most attention [56-58]. For example, in an early study, parent-child attachment in the warmth-support sense was the strongest promotive factor against delinquency among a diversity of family and demographic characteristics [59]. Parental closeness also increased the likelihood of choosing drug-free friends and decreased the likelihood of drug use in another study [57]. In other research, close parent-child attachment in early adolescence led to greater intolerance of deviance and less rebelliousness, which were related to associating with nondeviant peers later on [56]. Associating with nondeviant peers was then related to less young adult drug use. A negative effect of low parental monitoring on peer deviance has research support as well. In one study, for 10-year-old boys poor parental monitoring was associated with involvement with antisocial peers at age 12 [58]. Another study found that lower monitoring predicted higher peer drug use, which then predicted higher child substance use later on [60].

The current study: parental support and monitoring effects through peer selection

Thus, it is possible that parenting factors influence adolescents to select peers with higher alcohol use, resulting in those adolescents using greater amounts of alcohol themselves. Yet no previous study has analyzed the predictive power of both parental support and monitoring on youth alcohol use while assessing monitoring as solicitation and regulation instead of disclosure and allowing support and monitoring to account for one another. Furthermore, no study has examined the mechanism by which they work in a combined model. Better understanding which parenting factors are most crucial and why can improve recommendations for optimal parenting and thus lower youth drinking. The need for clarity in the effects of these key, long-studied parenting factors call for a study that includes support, monitoring, and peer selection together.

Therefore, the current study aimed to test whether effects of parental closeness and monitoring on youth alcohol use were mediated by the alcohol use of adolescents’ best friends in a combined model. Our methods improve on previous studies of support and control in alcohol use by including both parental support and monitoring as predictors in the same model, along with peer alcohol use as a mediator. Furthermore, we attempted to properly operationalize monitoring by including multiple indicators that assessed solicitation and regulation, not child disclosure. Lastly, we employed a preferred analytical approach superior to those conducted on this topic in the past: Bootstrapping regression-based mediation analysis in the context of structural equation modeling [61]. According to our review, no prior parental support or control study utilized these better methods [62]. Given the uncontested promotive effects of support in previous literature, we hypothesized that when 7^th grade adolescents who experienced greater parental support chose peers who used alcohol less in 8^th grade, they would use alcohol less themselves in early high school (9^th grade). Given mixed findings about monitoring and a lack of studies assessing solicitation, analyses of monitoring’s effect were considered exploratory. Yet as a working hypothesis, we proposed that when 7^th grade adolescents who experienced greater parental monitoring chose peers who used alcohol less in 8^th grade, they would use alcohol less themselves in 9^th grade.

Method

Parent study and participants

All measures and data were derived from one of four cohorts of the Adolescent Alcohol Prevention Trial (AAPT), an 8-year project intended to explore and test interventions for preventing youth substance use [9]. Participants were recruited from 12 junior high schools in Los Angeles and Orange Counties, California over the course of the 1987-1988 school year. The current study utilized the first three waves of AAPT data. At each assessment point participants completed the AAPT questionnaire, a 206-item survey on substance use, deviant attitudes, peers, and family experiences developed specifically for the AAPT. First wave responses were obtained during participants’ 7^th grade year. Subsequent responses were then gathered in 8^th and 9^th grade at approximately one-year intervals. These ages were chosen because initiation of drinking occurs in early to mid-adolescence for most individuals—63.4% have consumed alcohol by grade 9—and the transition to high school is a notable risk factor for increased alcohol consumption [23,63]. We analyzed data from 3,027 participants (52% female). The sample was 44.1% Caucasian, 26.5% Hispanic, 17.4% Asian, 1.9% African American, and 10.1% other. In terms of missing data, 80.2% of students (N = 2,416) were tested at the last measurement.

Measures

All items were taken from the Adolescent Alcohol Prevention Trial Questionnaire (AAPT questionnaire), a 206-item survey developed for the AAPT that inquired about topics such as participant alcohol and drug use, attitudes toward substance use, proximal adult substance use, parenting/parent-child factors, self-perceptions, peer substance use, and participant-peer relationships, among other topics. The four variables of interest in the current study (Parental Support, Monitoring, Best Friend’s Alcohol Use, and Participant Alcohol Use) and covariates were comprised of a subset of specific items chosen from the questionnaire in accordance with theoretical considerations (see Appendix A) and shown to have acceptable fit. For each factor of survey items, we conducted a confirmatory factor analysis (CFA). We also calculated scale reliabilities using a CFA-based method as recommended by [64]. Raykov joins several other authors in arguing that Cronbach’s alpha is insufficient under the conditions of latent variable modeling, suggesting an alternative be used [65].

The Parental Support latent factor had 5 items asking participants how close they felt to their parents, how much free time they spent with their parents, how often parents told him/her they were proud of him/her, how well the participant believed his/her parents understood what was important to him/her, and how much his/her parents cared about him/her (see Appendix A). Both the factor’s fit (CFI = 0.97, TLI = 0.94) and internal reliability (ρ = .914) were acceptable. Parental Monitoring included 3 items asking how often their parents asked them where they were going when they left the house, how often parents refused to let them do things with peers, and how often parents told them what time to be home when they went out with friends. Again, both the factor’s fit (CFI = 1.00, TLI = 1.00) and internal reliability (ρ = .817) were acceptable. Best Friends’ Alcohol Use was indicated by items asking about the alcohol use frequency and degree (i.e. drunkenness) of three people in their grade and school who they considered to be their best friends, as well as how often they were around drunk peers. For this factor, both its fit (CFI = 1.00, TLI = 1.00) and internal reliability (ρ = .905) were acceptable as well. Lastly, Alcohol Use was indicated by 5 items asking how many drinks they had their entire life, in the past 30 days, and in the past 7 days, how many days in the past month they had at least one drink, and the most drinks they had on the day in the past month when they drank the most. This factor also had acceptable fit (CFI = 0.97, TLI = 0.94) and internal reliability (ρ = .766).

Statistical analyses

Analyses employed Mplus 7 software. Structural equation modeling was used to perform a latent variable mediation analysis on the factors previously presented. To account for missing data, we used Full-Information Maximum Likelihood (FIML) estimation for mediation analyses with the MPlus ML estimator. We employed bootstrapping regression-based path analysis [61] with 10,000 bootstraps. Analyses tested causal paths in which parental support and parent monitoring of time with friends in 7th grade predicted participant alcohol use in 9th grade, both mediated by best friends’ substance use in 8th grade (Figure 1). Best friend’s alcohol use in 7^th grade, participant alcohol use in 7^th grade, participant gender, and whether the participant was of Caucasian ethnicity were included in this model as covariates. Seventh grade best friend’s and participant’s alcohol use were latent variables using the same indicators as their second and third wave counterparts. Autocorrelations were specified between the first, second, and third wave iterations of each of the individual item responses within these two covariate-outcome pairs. Ninth grade participant alcohol use was regressed on 8^th grade best friend’s alcohol use, 7^th grade closeness to parents, 7^th grade parental monitoring, and the covariates (7^th grade gender, ethnicity, participant alcohol use, and best friends’ alcohol use); 8^th grade best friends’ alcohol use was regressed on closeness to parents and monitoring and the covariates. Concurrent correlations were specified between all 7^th grade variables. Thus, we estimated all associations between latent variables. The relation between covariates best friends’ alcohol use and participant alcohol use in 7th grade exhibited high collinearity in the path model (r = 0.648). Thus, two separate models were run: One with 7^th grade participant alcohol use removed from the model and one with 7^th grade best friends’ use removed. Both models resulted in identical statistical conclusions regarding prediction and mediation. We present the statistical indices from the model with 7^th grade participant alcohol use removed. Although both models’ conclusions were identical, we present this model because our focus is on mediation by best friends’ use and including the covariate closer in time to its later counterpart is arguably more conservative.

Figure 1. Conceptual Diagram of a Mediation Model Where the Relationship Between Parental Support in 7th Grade (X) and Adolescent Alcohol Use in 9th Grade (Y) and Between Parental Monitoring in 7th Grade (X) and Adolescent Alcohol Use in 9th Grade (Y) is Mediated by Best Friends’ Alcohol Use in 8th Grade (M)

Regression Coefficients: a = Effect of X on M, b = Effect of M on Y, ab = Indirect Effect (Mediation), c’ = Direct effect of X on Y (with M Included in the Model), c = Total Effect of X on Y (with M Not Included in the Model).  *Result was Significant by 95% Bootstrap Confidence Interval.

Results

Descriptive statistics for the observed variables are presented in Table 1. Zero-order correlations between the factors prior to specifying the path model are presented in Table 2. The current model’s fit was good. Although the χ2 for this model was significant (χ2 (167) = 971.486, p < .001), χ2 is notably sensitive to sample size [66]. Given the sample size of the current study (N = 3027), even trivially slight deviations from an ideal model would be statistically significant. For that reason, judgments of goodness of fit were made on the basis of four indices of practical model fit: CFI, TLI (also known as NNFI), RMSEA, and SRMR. RMSEA is incorrect in the case of missing data due to the incorrect use of the full sample size N, which ignores missing data [67]. Accordingly, each RMSEA has been corrected in this study by making use of N’ = (% of values observed) x N as suggested by [68]. Based on the net pattern of RMSEA, TLI, and CFI (CFI = 0.95, TLI = 0.94, RMSEA = 0.048, SRMR = 0.045), the fit of the model was considered good.

Table 1. Descriptive statistics: Means and standard deviations of observed variables.

Table 2. Correlations between latent variables prior to specifying the path model.

Note. *p < .05; **p < .01; ***p < .001

Main results of the pre-to-post mediation analyses are summarized in Figure 1. The coefficients (a, b, and c’) represent standardized regression coefficients for each respective model path. First, the paths emanating from closeness to parents were as expected. Analyses revealed a significant total effect where higher closeness to parents in 7^th grade led to lower youth alcohol use in 9^th (c = -0.134, 95% bootstrap CI = [-0.627, -0.130]). Higher closeness to parents in 7^th grade predicted lower best friends’ alcohol use in 8^th grade (a = -0.074, 95% bootstrap CI = [-0.190, -0.002]) and lower best friends’ use significantly predicted lower youth alcohol use in 9^th grade (b = 0.333, 95% bootstrap CI = [0.507, 0.987]). Most importantly, results supported the hypothesized mediation model: According to a 95% bias-corrected bootstrap confidence interval, the indirect effect of closeness to parents in 7^th grade on youth alcohol use in 9^th grade through lower best friends’ use was statistically different from zero (ab = -0.025, 95% bootstrap CI = [-0.152, -0.003]). Thus, the association between higher closeness to parents in 7^th grade and lower youth alcohol use in 9^th grade was mediated by lower best friends’ alcohol use in 8^th grade. A significant direct effect remained (c’ = -0.110, 95% bootstrap CI = [-0.549, -0.063]).

Parental monitoring paths did not align with hypotheses. There was no total effect for the path from parental monitoring in 7^th grade to alcohol use in 9^th grade (c = -0.004, 95% bootstrap CI = [-0.320, 0.202]). Higher parental monitoring in 7^th grade did not significantly predict lower best friends’ alcohol use in 8^th grade (a = -0.054, bootstrap CI = [-0.169, 0.037]). Again, lower best friends’ use significantly predicted lower youth alcohol use in 9^th grade (b = 0.333, 95% bootstrap CI = [0.507, 0.987]). Yet results did not support the hypothesized mediation model: According to a 95% bias-corrected bootstrap confidence interval, the indirect effect of parental monitoring in 7^th grade on youth alcohol use in 9^th grade through lower best friends’ use was not significant (ab = 0.018, 95% bootstrap CI = [-0.135, 0.025]). Thus, the association between higher parental monitoring in 7^th grade and lower youth alcohol use in 9^th grade was not mediated by lower best friends’ alcohol use in 8^th grade. Note that the direct effect was not significant either (c’ = 0.110, 95% bootstrap CI = [-0.262, 0.240]).

Discussion

Previous studies of the effect of parental support and monitoring on youth alcohol use have had mixed findings regarding parental monitoring’s influence. Since past studies have not combined support and monitoring in the same model, have operationalized parental monitoring as child disclosure, and have not used contemporary mediation methods, studies with improved methods could clarify the uncertain impact of these parenting factors. In addition, research has not sufficiently addressed mechanisms of the promotive effect of parenting factors on adolescent drinking, such as selection of peers-particularly not in a combined model. The current study used structural equation modeling and bootstrapping mediation analysis on data from the Adolescent Alcohol Prevention Trial [9]. We tested a model where the effect of parental support and monitoring in 7^th grade on adolescent alcohol use in 9^th grade was hypothesized to be mediated by best friends’ alcohol use in 8^th grade. Our predictions regarding parental support’s effect were supported, but our exploratory hypotheses regarding parental monitoring’s effect were not. Higher parental support in seventh grade predicted lower adolescent alcohol use in 9^th grade, mediated by lower best friends’ use in eighth grade. When adolescents who experienced more parental love and attention chose close friends who drank less alcohol a year later, they drank less themselves in their first year of high school. Yet in contrast to expectations, when parental support was a co-predictor, parental monitoring in 7^th grade did not predict alcohol use in 9^th grade. There was also no significant mediation effect for best friends’ alcohol in 8^th grade for the monitoring to youth drinking path.

Why Did Support Outperform Monitoring for Alcohol Use?

The finding that higher parental support lowered risk for youth alcohol use is consistent with the vast majority of the literature. Both theory and research have offered explanations for how parental support buffers against alcohol use. Parental warmth and closeness may lead children to respect, agree, and identify with their parents’ wishes, such as wishes that they avoid drinking. Increased satisfaction from relationships with parents, identification with parents, and the promotion of social competence that comes from parental support may keep youth from having the desire to drink for pleasure or for any perceived social effects it may have [3]. Wills, et al. [17] have proposed that support from parents builds self-control and acquaints children to mainstream institutions as well. Similarly, close parent-child attachment in early adolescence leads to greater responsibility, less rebelliousness, and intolerance of deviance (a.k.a. “conventional behavior”)-all factors that would suggest lower likelihood of alcohol use [56]. The novel contribution made by the current study is the revelation that parental support lowers risk for adolescent alcohol use by leading youth to choose peers that use less alcohol, if any. When adolescents feel close to and spend more time with parents, and believe their parents care for and understand them, they not only feel a stronger desire to abide by their parents rules out of admiration and respect, but they have a significant foundation of social support from family as well. Thus, they have stronger intimacy bonds and may feel less of a need to engage in risky behaviors to acquire desired social connection. Consequently, they are less likely to forge close friendships with peers who do not follow their parents’ wishes and rules. They may also feel less need to seek out costly relationships that could lead to punishment, because their needs for closeness are satisfied by major attachment figures (the costs outweigh the benefits). Once adolescents have formed best friendships with non-drinking peers, a wealth of literature shows it is much less likely for them to drink later in life [52].

When accounting for support in our combined model, parental monitoring neither predicted participant alcohol use nor best friends’ alcohol use. Although there is literature suggesting parental monitoring protects against choosing deviant peers and adolescent alcohol use, our study suggests that parental support accounts for the majority of this effect. Similarly, the only other combined study to operationalize monitoring as parent solicitation and regulation in adolescents also found that support significantly predicted youth alcohol use, but monitoring did not (though it was assessed by only one item about mothers [42]. Note that a recent Bolivian study on young adults included both parental monitoring and father-child relationship in one model, finding that only parental monitoring significantly predicted alcohol consumption [69]. Yet the mean age of their Bolivian sample was 19-years-old (the legal drinking age is 18 in Bolivia), they did not report the content of any of their self-constructed items, and they omitted the mother-child relationship. Theorists have suggested that when children feel close with their parents, they are more likely to follow their rules out of respect and a desire to please [38]. In fact, Barnes, et al. [2] found that higher parental support prospectively predicted higher monitoring, which later resulted in lower drinking. As another study demonstrated, parenting styles marked by high warmth and low strictness prevent alcohol use to the same degree as parenting with both high warmth and high strictness [70]. Such results suggest control forms like monitoring may be less relevant in the context of support.

Importantly, children who feel close and supported by their parents may also be more forthcoming. A cohort of researchers following the work of Kerr, et al. [44] have proposed that it is not parental tracking of adolescent children, or solicitation of children’s report, that predicts lower alcohol use [1,45]. Rather, it is the child’s truthful disclosure about activities—the adolescent’s tendency to be open with his or her parents about activities. Parental support variables have longitudinally predicted honest child disclosure in several studies, which may be why support overcomes parental monitoring in our combined model [71,72]. Keijsers, et al. [33] used a bidirectional multi-informant design to study both parental monitoring, questioning about the adolescent’s activities, and adolescent disclosure. They found that parental monitoring did not predict delinquency. Rather, the child’s own disclosure was negatively associated with delinquency. The same results were found by Kerr, et al. [36]. In addition, Keijsers, et al. [33] also revealed that high delinquency predicted low disclosure. Also relevant, Laird, et al. [73] found that adolescents with parents who had relatively less true knowledge about their child had higher levels of delinquent behavior and more deviant peers, both of which increased over time. Similarly, Abar, et al. [74] found that greater discrepancies between parent reports of knowledge about their children and children’s report of parents’ knowledge significantly predicted greater youth alcohol use. Lastly, a sophisticated random-intercept cross-lagged model by Keijsers [1] found that linkages between parental monitoring and delinquency were not explained by causal processes of monitoring operating at the family level, but adolescent disclosure and delinquency were associated [75]. It may be that parental monitoring is only as good as adolescents’ willingness to reveal their true behaviors-an act more likely in the context of parental warmth.

Note that in the most extensive and cited longitudinal studies on parental support and control, both parental support and “monitoring” did predict lower later drinking [43]. Yet this may have occurred only because the included monitoring items truly assessed child disclosure (e.g., “How often do you tell your parents where you’re really going?”) rather than parental solicitation and regulation of child activity.

It is important to make the distinction between solicitation and disclosure for practical reasons. Parents have direct control over their solicitation of children’s information and regulation of child activities, but they do not have control over the truthfulness of children’s replies. Taken together, this study’s results suggest that parents wishing to prevent drinking may benefit most from emphasizing the conveyance of love, warmth, and acceptance and building closeness with their children over tracking their activities. Yet it is important to recognize that control should not be entirely abandoned. Parental support can aid the effectiveness of efforts at discipline and monitoring and may be impactful partly because of its influence on monitoring’s power. Control is an arguably necessary component of effective parenting that should not be done away with simply because support bears the most effect [38].

Limitations

The current study was not without limitations. It is important to recognize that data for this analysis was taken from a trial for a particular intervention. The AAPT study included four different treatment conditions, including 1) lessons about the negative consequences of substances, 2) lessons about consequences and education on resisting peer pressure to use substances, 3) lessons on consequences in addition to lessons correcting erroneous perceptions of prevalence and acceptance of substance use among peers (Normative Education), and 4) lessons on consequences, resistance to peer pressure, and conservative norms of peer use. Note that some of these conditions attempted to teach youth how to resist peer pressure, which could possibly have skewed findings on best friends’ alcohol use predicting participant use. There are reasons to believe the AAPT data are appropriate for the current study though. First, Hansen, et al. [9] found that the programs designed to teach techniques to resist and refuse peer pressure substance use offers had “no discernable positive impact on substance use behavior” (p. 425). Thus, it caused no change in substance use and was unlikely to influence results through reducing peer pressure response. Second, engaging in substance prevention programs in an educational setting is actually a very normative experience for junior high students. Based on a large nationally representative survey of schools, on average, U.S. students receive 1.62 substance use prevention programs during their school years from elementary through high school [76]. Our sample’s experience of such programs during the study arguably increases the external validity of our findings.

Several other possible study limitations exist. Due to the self-report nature of this assessment, young participants may have inaccurately over- or under-reported their behavior. Despite repeated assurance that their reports were confidential, completing the questionnaire in a school environment around teachers may have led students to under-report their deviant behaviors, feeling threatened by potential negative consequences of revealing their substance use. It is also possible that junior high participants gave faulty reports regarding drunkenness. Many participants may not have acquired enough experience with alcohol to understand at what level of intoxication one is reasonably “drunk,” leading to erroneous reporting. In addition, a much larger proportion of participants in this study were White, with a quarter being of Hispanic ethnicity. Such restricted ethnic breadth reduced the external validity of results to other groups.

Future directions

This study is one of several that are beginning to explore the intricacies of the means by which parenting factors influence adolescent drinking. There are many avenues for future research to follow beyond our current foundation. It may be a worthy cause for researchers to include the effect of child variables within the context of parental support. The reciprocal effects model of socialization suggests that child behavior affects parenting just as parenting behavior affects the child, with both children and parents eliciting behavioral patterns from one another [77]. Future research should test transactional models of how children’s influences on parenting eventually affect later alcohol use, as some are already doing [35]. Child factors in general may be important to examine, such as level of child empathy. Some studies suggest children’s trait empathy may affect support’s influence. In several studies children with low levels of dispositional affective empathy were less responsive to parental practices and had greater antisocial behavior [78]. Such child factors may be possible moderators of the influence of parental support on later adolescent alcohol use. Study of other child and parent factors may reveal interaction effects in these paths, such as child temperament, child gender differences, and father-child vs. mother-child relationships [79-81]. It is also important to study other mediators of parental support’s preventative effect on adolescent drinking beyond best friends’ alcohol use, such as sibling substance use, self-regulation and risk-taking tendency [16,82,83].

Future research should continue to devote attention to mediators of the influence of parental factors on alcohol use to improve youth health behavior. Peer bonds are a highly impactful motivator of adolescent behavior. Friend variables are an explanatory factor within the parenting-alcohol use relationship. As suggested by this study, research on such variables is likely to reveal fruitful findings. As the current findings suggest, parenting support and the control it fosters can direct adolescents’ selection of friends, which in turn influences their later alcohol use. Expressing love and understanding and cultivating closeness may be a parent’s best bet for curbing relationships marked by deviance and, eventually, underage drinking.

Funding information

This research was supported in part by the National Institute on Alcohol Abuse and Alcoholism, R01-AA06201. This grant is primarily associated with author John W. Graham, Ph.D.

Competing interests

There are no competing interests or conflicts of interest to disclose for this study.

View Appendix Data

References

1. Keijsers L (2016) Parental monitoring and adolescent problem behaviors: How much do we really know? International Journal of Behavioral Development 40: 271-281.
2. Barnes GM, Reifman AS, Farrell MP, Dintcheff BA (2000) The effects of parenting on the development of adolescent alcohol misuse: A six-wave latent growth model. Journal of Marriage and the Family 62: 175-186.
3. Peterson GW, Leigh GK (1990) The family and social competence in adolescence. In Gullota T, Adams GR, Montemayor R (Eds), Advances in adolescent development: Social competence, Newbury Park, CA: Sage pp: 97-138.
4. Abell E Gecas V (1997) Guilt, shame, and family socialization: A retrospective study. Journal of Family Issues 18: 99-123.
5. de Kemp RAT, Overbeek G, de Wied M, Engels RC, Scholte RHJ (2007) Early adolescent empathy, parental support, and antisocial behavior. The Journal of Genetic Psychology: Research and Theory on Human Development 168: 5-18.
6. Harris-McKoy D, Cui M (2013) Parental control, adolescent delinquency, and young adult criminal behavior. Journal of Child and Family Studies 22: 836-843.
7. Hoeve M, Dubas JS, Eichelsheim VI, van der Laan PH, Smeenk W, et al. (2009) The relationship between parenting and delinquency: A meta-analysis. J Abnorm Child Psychol 37: 749-775. [Crossref]
8. Barnes GM, John WW, Hoffman JH (2002) Relationship of alcohol use to delinquency and illicit drug use in adolescents: Gender, age, and racial/ethnic differences. The Journal of Drug Issues 32: 153 - 178.
9. Hansen WB, Graham JW (1991) Preventing alcohol, marijuana, and cigarette use among adolescents: Peer pressure resistance training versus establishing conservative norms. Prev Med 20: 414-430. [Crossref]
10. Johnston LD, O'Malley PM, Bachman JG (2010) Monitoring the future: National Survey Results on Drug Use, 1975-2000, Bethseda: National Instutite on Drug Abuse I: 10-7584.
11. Atilola O, Stevanovic D, Balhara YP, Avicenna M, Kandemir H, et al. (2014) Role of personal and family factors in alcohol and substance use among adolescents: An international study with focus on developing countries. J Psychiatr Ment Health Nurs 21: 609-617. [Crossref]
12. Anderson AR, Henry CS (1994) Family system characteristics and parental behaviors as predictors of adolescent substance use. Adolescence 29: 405-420.
13. Caldwell CH, Sellers RM, Bernat DH, Zimmerman MA (2004) Racial identity, parental support, and alcohol use in a sample of academically at-risk African American high school students. Am J Community Psychol 34: 71-82. [Crossref]
14. Maton KI, Zimmerman MA (1992) Psychological predictors of substance abuse among urban black male adolescents. Drugs & Society 6: 79-114.
15. Piko B (2000) Perceived social support from parents and peers: Which is the stronger predictor of adolescent substance use? Subst Use Misuse 35: 617-630. [Crossref]
16. Wills TA, Resko JA, Ainette MG, Mendoza D (2004) Role of parent support and peer support in adolescent substance use: A test of mediated effects. Psychol Addict Behav 18: 122-134. [Crossref]
17. Wills TA, Mariani J, Filer M (1996) The role of family and peer relaitonships in adolescent substance use. In Pierce GR, Sarason BR, Sarason IG (Eds), Handbook of social support and the family pp. 521-549.
18. Beck KH, Shattuck T, Haynie D, Crump AD, Simons-Morton B (1999) Associations between parent awareness, monitoring, enforcement and adolescent involvement with alcohol. Health Education Research 14: 765-775.
19. Beck KH, Boyle JR, Boekeloo BO (2003) Parental monitoring and adolescent alcohol risk in a clinic population. Am J Health Behav 27: 108-115. [Crossref]
20. Caldwell LL, Darling N (1999) Leisure context, parental control, and resistance to pressure as predictors of adolescent partying and substance use: An ecological perspective. Journal of Leisure Research 31: 57-77.
21. Cottrell L, Li X, Harris C, D'Alessandri D, Atkins M, et al. (2003). Parent and adolescent perceptions of parental monitoring and adolescent risk involvement. Parenting: Science and Practice 3: 179-195.
22. Duncan SC, Duncan TE, Biglan A, Ary D (1998) Contributions of the social context to the development of adolescent substance use: A multivariate latent growth modeling approach. Drug Alcohol Depend 50: 57-71. [Crossref]
23. Jackson KM, Schulenberg JE (2013) Alcohol use during the transition from middle school to high school: National panel data on prevalence and moderators. Dev Psychol 49: 2147-2158. [Crossref]
24. Ledoux S, Miller P, Choquet M, Plant M (2002) Family structure, parent-child relationships, and alcohol and other drug use among teenagers in France and the United Kingdom. Alcohol Alcohol 37: 52-60. [Crossref]
25. Peterson PL, Hawkins JD, Abbott RD, Catalano RF (1994) Disentangling the effect of parental drinking, family management, and parental alcohol norms on current drinking by black and white adolescents. Journal of Research on Adolescence 4: 203-227.
26. Pettit GS, Laird RD, Dodge KA, Bates JE, Criss MM (2001) Antecedents and behavior-problem outcomes of parental monitoring and psychological control in early adolescence. Child Development 72 583-598.
27. Reifman A, Barnes GM, Dintcheff BA, Farrell MP, Uhteg L (1998) Parental and peer influences on the onset of heavier drinking among adolescents. J Stud Alcohol 59: 311-317. [Crossref]
28. Crawford LA, Novak KB (2002) Parental and peer influences on adolescent drinking: The relative impact of attachment and opportunity. Journal of Child & Adolescent Substance Abuse 12: 1-26.
29. Simons-Morton B (2007) Social influences on adolescent substance use. Am J Health Behav 31: 672-684. [Crossref]
30. Soenens B, Vansteenkiste M, Luyckx K, Goossens L (2006) Parenting and adolescent problem behavior: An integrated model with adolescent self-disclosure and perceived parental knowledge as intervening variables. Dev Psychol 42: 305-318. [Crossref]
31. Stattin H, Kerr M (2000) Parental monitoring: A reinterpretation. Child Dev 71: 1072-1085. [Crossref]
32. Waizenhofer RN, Buchanan CM, Jackson-Newsom J (2004) Mothers' and fathers' knowledge of adolescents' daily activities: Its sources and its links with adolescent adjustment. J Fam Psychol 18: 348-360. [Crossref]
33. Keijsers L, Branje SJT, Van der Valk IE, Meeus W (2010) Reciprocal effects between parental solicitation, parental control, adolescent disclosure, and adolescent delinquency. Journal of Research on Adolescence 20: 88-113.
34. Stavrinides P, Georgiou S, Demetriou A (2010) Longitudinal associations between adolescent alcohol use and parents’ sources of knowledge. Br J Dev Psychol 28: 643-655.
35. Wertz J, Nottingham K, Agnew-Blais J, Matthews T, Pariante CM, et al. (2016) Parental monitoring and knowledge: Testing bidirectional associations with youths' antisocial behavior. Dev Psychopathol 28: 623-638. [Crossref]
36. Kerr M, Stattin H, Burk WJ (2010) A reinterpretation of parental monitoring in a longitudinal perspective. Journal of Research on Adolescence 20: 39-64.
37. Kiesner J, Dishion TJ, Poulin F, Pastore M (2009) Social Development. Temporal dynamics linking aspects of parent monitoring with early adolescent antisocial behavior. Social Development 18: 765-784.
38. Wright JP, Cullen FT (2001) Parental efficacy and delinquent behavior: Do control and support matter? Criminology 39: 677-706.
39. Mogro-Wilson C (2008) The influence of parental warmth and control on Latino adolescent alcohol use. Hispanic Journal of Behavioral Sciences 30: 89-105.
40. Stice E, Barrera M Jr, Chassin L (1993) Relation of parental support and control to adolescents' externalizing symptomatology and substance use: A longitudinal examination of curvilinear effects. J Abnorm Child Psychol 21: 609-629. [Crossref]
41. Tharp AT, Noonan RK (2012) Associations between three characteristics of parent-youth relationships, youth substance use, and dating attitudes. Health Promot Pract 13: 515-523. [Crossref]
42. Tyler K, Torres Stone R, Bersani B (2006) Examining the changing influence of predictors on adolescent alcohol misuse. Journal of Child and Adolescent Substance Abuse p.16.
43. Barnes GM, Hoffman JH, Welte JW, Farrell MP, Dintcheff BA (2006) Effects of parental monitoring and peer deviance on substance use and delinquency. Journal of Marriage and Family 68: 1084-1104.
44. Kerr M, Stattin H (2000) What parents know, how they know it, and several forms of adolescent adjustment: Further support for a reinterpretation of monitoring. Dev Psychol 36: 366-380. [Crossref]
45. Racz SJ, McMahon RJ (2011) The relationship between parental knowledge and monitoring and child and adolescent conduct problems: A 10-year update. Clin Child Fam Psychol Rev 14: 377-398. [Crossref]
46. Thomas HJ, Kelly AB (2013) Parent-child relationship quality and adolescent alcohol use. Australian and New Zealand. Journal of Psychiatry 47: 1093.
47. Kliewer SA, Zaharakis N (2014) Family-Based Models of Drug Etiology. In Scheier LM, Hansen WB (Eds), Parenting and teen drug use: The most recent findings from research, prevention, and treatment. New York: Oxford University Press.
48. Barnow S, Schultz G, Lucht M, Ulrich I, Preuss UW, et al. (2004) Do alcohol expectancies and peer delinquency/substance use mediate the relationship between impulsivity and drinking behaviour in adolescence? Alcohol Alcohol 39: 213-219. [Crossref]
49. Dishion TJ, Loeber R (1985) Adolescent marijuana and alcohol use: The role of parents and peers revisited. Am J Drug Alcohol Abuse 11: 11-25. [Crossref]
50. Jones DJ, Hussong AM, Manning J, Sterrett E (2008) Adolescent alcohol use in context: The role of parents and peers among African American and European American youth. Cultur Divers Ethnic Minor Psychol 14: 266-273. [Crossref]
51. Leung RK, Toumbourou JW, Hemphill SA (2010) The effect of peer influence and selection processes on adolescent alcohol use: A systematic review of longitudinal studies. Health Psychol Rev 8: 426-457. [Crossref]
52. Windle M (2000) Parental, sibling, and peer influences on adolescent substance use and alcohol problems. Applied Developmental Science 4: 98-110.
53. Nash SG, McQueen A, Bray JH (2005) Pathways to adolescent alcohol use: Family environment, peer influence, and parental expectations. J Adolesc Health 37: 19-28. [Crossref]
54. Chassin L, Curran PJ, Hussong AM, Colder CR (1996) The relation of parent alcoholism to adolescent substance use: A longitudinal follow-up study. J Abnorm Psychol 105: 70-80. [Crossref]
55. Goldstein SE, Davis-Kean PE, Eccles JS (2005) Parents, peers, and problem behavior: A longitudinal investigation of the impact of relationship perceptions and characteristics on the development of adolescent problem behavior. Dev Psychol 41: 401-413. [Crossref]
56. Brook JS, Whiteman M, Finch S, Cohen P (2000) Longitudinally foretelling drug use in the late twenties: Adolescent personality and social-environmental antecedents. Journal of J Genet Psychol 161: 37-51. [Crossref]
57. Kandel DB, Andrews K (1987) Processes of adolescent socialization by parents and peers. Int J Addict 22: 319-342. [Crossref]
58. Dishion TJ, Patterson GR, Stoolmiller M, Skinner ML (1991) Family, school, and behavioral antecedents to early adolescent involvement with antisocial peers. Dev Psychol 27: 172-180.
59. Gove WR, Crutchfield RD (1982) The family and juvenile delinquency. The Sociological Quarterly 23: 301-319.
60. Chassin L, Pillow DR, Curran PJ, Molina BS, Barrera M Jr (1993) Relation of parental alcoholism to early adolescent substance use: A test of three mediating mechanisms. J Abnorm Psychol 102: 3-19. [Crossref]
61. Preacher KJ, Hayes AF (2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 40: 879-891. [Crossref]
62. Hayes AF (2009) Beyond Baron and Kenny: Statistical mediation analysis in the new millenium. Communication Monographs 76: 408-420.
63. Eaton DK, Kann L, Kinchen S, Shanklin S, Ross J, et al. (2010) Centers for Disease C & Prevention Youth risk behavior surveillance - United States, 2009. MMWR Surveill Summ 59: 1-142. [Crossref]
64. Raykov T (2004) Behavioral scale reliability and measurement invariance evaluation using latent variable modeling. Behavior Therapy 35: 299-331.
65. Kraemer HC (2013) Toward sound objective evaluation of clinical measures. Am J Geriatr Psychiatry 21: 589-595. [Crossref]
66. Bergh D (2015) Chi-squared test of fit and sample size: A comparison between a random sample approach and a chi-square value adjustment method. J Appl Meas 16: 204-217. [Crossref]
67. Graham JW, Coffman DL (2012) Structural equation modeling with missing data. In Hoyle R (Ed), Handbook of structural equation modelling, New York: Guilford Press pp: 227-295.
68. Graham JW, Hofer SM (2000) Multiple imputation in multivariate research. In T. D. Little, Schnabel KU, Baumert J (Eds.), Modeling longitudinal and multiple-group data: Practical issues, applied approaches, and specific examples. Hillsdale, NJ: Erlbaum pp. 201-218.
69. Guilén N, Roth E, Alfaro A, Fernández E (2015) Youth alcohol drinking behavior: Associated risk andprotective factors. Revista Iberoamericana de Psicología y Salud 6: 53-63.
70. Calafat A, Garcia F, Juan M, Becona E, Fernandez-Hermida JR (2014) Which parenting style is more protective against adolescent substance use? Evidence within the European context. Drug Alcohol Depend 138: 185-192. [Crossref]
71. Blodgett Salafia EH, Gondoli DM, Grundy AM (2009) The longitudinal interplay of maternal warmth and adolescent self-disclosure in predicting maternal knowledge. Journal of Research on Adolescence 19: 654-658.
72. Darling N, Cumsille P, Caldwell LL, Dowdy B (2006) Predictors of adolescents' disclosure to parents and perceived parental knowledge: Between- and within-person differences. Journal of Youth and Adolescence 35: 667-678.
73. Laird RD, Criss MM, Pettit GS, Dodge KA, Bates JE (2008) Parents' monitoring knowledge attenuates the link between antisocial friends and adolescent delinquent behavior. J Abnorm Child Psychol 36: 299-310. [Crossref]
74. Abar CC, Jackson KM, Colby SM, Barnett NP (2015) Parent-child discrepancies in reports of parental monitoring and their relationship to adolescent alcohol-related behaviors. J Youth Adolesc 44: 1688-1701. [Crossref]
75. Adolescent disclosure and adolescent delinquency. Journal of Research on Adolescence 20: 88-113.
76. Kumar R, O'Malley PM, Johnston LD, Laetz VB (2013) Alcohol, tobacco, and other drug use prevention programs in U.S. schools: A descriptive summary. Prev Sci 14: 581-592. [Crossref]
77. Hartup WW (1978) Perspectives on child and family interaction: Past, present, and future. In RM, Lerner GB Spanier (Eds.), Child influences on marital and family interaction: A life-span perspective (pp. 23-46). San Francisco: Academic Press.
78. Oxford M, Cavell TA, Hughes JN (2003) Callous/unemotional traits moderate the relation between ineffective parenting and child externalizing problems: a partial replication and extension. J Clin Child Adolesc Psychol 32: 577-585. [Crossref]
79. Stice E, Gonzales N (1998) Adolescent temperament moderates the relation of parenting to antisocial behavior and substance use. Journal of Adolescent Research 13: 5-31.
80. Borawski EA, Ievers-Landis CE, Lovegreen LD, Trapl ES (2003) Parental monitoring, negotiated unsupervised time, and parental trust: The role of perceived parenting practices in adolescent health risk behaviors. J Adolesc Health 33: 60-70. [Crossref]
81. Choo H, Shek D (2013) Quality of parent–child relationship, family conflict, peer pressure, and drinking behaviors of adolescents in an Asian context: The case of Singapore. Social Indicators Research 110: 1141-1157.
82. Windle M (1992) Temperament and social support in adolescence: Interrelations with depressive symptoms and delinquent behaviors. J Youth Adolesc 21: 1-21. [Crossref]
83. Patock-Peckham JA, Cheong J, Balhorn ME, Nagoshi CT (2001) A social learning perspective: A model of parenting styles, self-regulation, perceived drinking control, and alcohol use and problems. Alcohol Clin Exp Res 25: 1284-1292. [Crossref]