An Examination of Peer,Family,and Community Context Risk Factors for Alcohol Use and Alcohol Use Intentions in Early Adolescents

Abstract

This study examines the relationship between peer, family, and community context risk factors and alcohol use; gender is examined as a potential moderator of these relationships. Hierarchical logistic regressions conducted in a sample of 781 seventh grade students found that normative beliefs about peers’ alcohol use emerged as the most consistent predictor of current use, lifetime alcohol initiation, and intentions to initiate drinking. Poorer parent–child relationship quality also increased adolescents’ odds of lifetime initiation and current alcohol use. After accounting for peer and family factors, greater perceived availability of alcohol was significantly associated with increased odds of current and lifetime use. Both community context risk predictors were significantly associated with increased risk of intending to initiate alcohol use among nonusers. Results demonstrated some support for gender moderating these relationships; poor family relationship quality increased girls’ odds of initiating alcohol use while more permissive community norms increased boys’ odds of initiating alcohol use.

Keywords

alcohol use gender differences early adolescence prevention

Introduction

A hallmark of the transition from childhood to adolescence is the increase in risk behaviors, including the onset of alcohol use (Hawkins, Catalano, & Miller, 1992; Schulenberg, Bryant, & O’Malley, 2004; Windle et al., 2008). Alcohol initiation rates increase during the middle school years (Kosterman, Hawkins, Guo, Catalano, & Abbott, 2000) with earlier onset of use associated with a host of negative outcomes, including increased risk of developing later alcohol abuse and dependence (Grant & Dawson, 1997). While girls have historically had an older age of onset and lower rates of alcohol use, recent epidemiological findings show that females are initiating alcohol use at the same or younger ages than males (Johnston, O’Malley, Bachman, & Schulenberg, 2004; Substance Abuse and Mental Health Services Administration, 2009). In addition, rates of current alcohol use prevalence (defined as past 30-day use) are now similar among adolescent boys and girls aged 12 through 17, with 14.3% of girls and 15.1% of boys reporting past 30-day use (Substance Abuse and Mental Health Services Administration, 2010). Despite similar initiation and use rates, particular risk factors associated with initiation and use may differ for males and females during this developmental period.

Risk factors can be organized into an ecological framework that views individual development occurring within a range of socioenvironmental contexts. Peers and family represent proximal influences on development, while neighborhood, community, and larger cultural contexts constitute more distal spheres of influence (Bronfenbrenner, 1989). Both parenting practices, such as limit setting and monitoring, and parent–child relationship characteristics, such as warmth and trust, appear to be important risk and protective factors for adolescent alcohol use (Dishion & McMahon, 1998; Schinke, Fang, & Cole, 2008; Schulte, Ramo, & Brown, 2009; Webb, Bray, Getz, & Adams, 2002). Peer influences, including close friends’ use, friends’ approval of use, and normative beliefs regarding alcohol use among their same aged peers, have consistently been found to be the strongest predictors of alcohol use (Bahr, Hoffmann, & Yang, 2005; Flannery, Vazsonyi, Torquati, & Fridrich, 1994; Hawkins et al., 1992; Kung & Farrell, 2000). Beyond parents and peers, research has increasingly found that community context can influence early alcohol use through factors such as the perceived availability of alcohol, or tolerant community norms regarding underage drinking (Dent, Grube, & Biglan, 2005; Kuntsche, Kuendig, & Gmel, 2008; Song, Smiler, Wagoner, & Wolfson, 2012).

Gender Differences in Key Risk Factors

Some gender differences have emerged with regard to family risk factors for alcohol use. Poor parental monitoring (e.g., lack of knowledge about where children are and with whom they spend time) and inconsistent rules appear to be greater predictors of alcohol use for males than females (Borawski, Ievers-Landis, Lovegreen, & Trapl, 2003; Flannery et al., 1994; Griffin, Scheier, Botvin, & Diaz, 2000; Rai et al., 2003). Conversely, interpersonal family-level risk variables, including poor parental support and parental trust, greater family conflict, and low levels of bonding to family are stronger influences on girls’ drinking than boys (Borawski et al., 2003; CASA, 2003; Kumpfer & Turner, 1990; Kung & Farrell, 2000). In addition, greater parental involvement is associated with decreased likelihood of female but not male alcohol use (Eisenberg, Neumark-Sztainer, Fulkerson, & Story, 2008; Fisher, Miles, Austin, Camargo, & Colditz, 2007).

There is a mixed picture of how gender influences the risk that peers pose in adolescents’ own use of alcohol. Studies have shown that having close friends who use alcohol is a stronger predictor of alcohol use for males than females (Barber, Bolitho & Bertrand, 1998; Steinberg, Fletcher, & Darling, 1994). However, other studies have shown that having close friends who use alcohol or who engage in other deviant behavior is associated with increased alcohol use by girls compared to boys (Averna & Hesselbrock, 2001; Callas, Flynn, & Worden, 2004; Chang et al., 2007; Dick et al., 2007; Pope, Smith, Wayne, & Kelleher, 1994; Simons-Morton, Haynie, Crump, Eitel, & Saylor, 2001; Yeh, Chiang, & Huang, 2006). Peer drinking norms (perceptions of what percentage of same aged peers drink alcohol) may be equally risky for adolescent males and females (Epstein, Botvin, Baker, & Diaz, 1999).

Community contextual or environmental variables also influence adolescent alcohol use. Greater availability of alcohol, measured with a variety of methodologies (i.e., self-reported ease of obtaining alcohol, alcohol outlet density) is associated with greater use of alcohol (CASA, 2003; Dent et al., 2005; Harrison, Fulkerson, & Park, 2000; Hawkins et al., 1992; Johnston et al., 2004; Jones-Webb et al., 1997; Newcomb & Felix-Ortiz, 1992; Roski et al., 1997) but few studies have explicitly examined whether gender moderates this relationship. Some evidence suggests that boys are more influenced by their school and community environments than girls (Jones-Webb et al., 1997; Kumpfer & Turner,1990; Substance Abuse and Mental Health Services Administration, 2002).

Hypotheses

The current article examines a set of demographic, family, peer, and community context variables to determine their importance as predictors of lifetime and 30-day alcohol use as well as intentions to initiate alcohol use within nonusing adolescents. Family and peer risk factors are expected to predict alcohol use and intentions to initiate alcohol use after the relationship with demographic variables are accounted for. Further, community context variables are expected to contribute significant and unique variance over and above that accounted for by all other variables. In addition, gender interaction terms for each risk predictor are included to explore whether gender moderates the relationship between risk predictors and alcohol use and intentions to use. It is hypothesized that poor parental monitoring will be a stronger predictor of use and intentions to use for males than for females and poor parent–child relationship quality will be to a stronger predictor of alcohol use and intentions to use for females than for males. Further, perceived availability and tolerant community norms are hypothesized to be stronger predictors of use and intentions to use for males than females. Due to mixed findings for peer influences on alcohol use, no gender-specific is hypothesized.

Method

Participants

The sample includes 781 seventh grade students, with a total of 389 males and 392 females, from 12 middle schools located in the northeastern United States, who completed a self-report survey (2004-2005). EM imputation was used to address the 75 cases containing missing data. EM imputation is a 2-step iterative maximum-likelihood procedure that generates expected values based on all complete data points (Hedderley & Wakeling, 1995). In the expectation step, the starting values for parameters are obtained with all available data and regression methods are used to impute the values for the missing data based on the initial values. In the maximization step, the newly imputed data with the original observed data are used to calculate new values for the parameters. This two-step process continues until the values converge and the difference in the estimates between iterations is negligible (Bennett, 2001).

Procedure

Written parental consent was obtained from parents for students to participate in this study. Students from 12 different middle schools completed self-report questionnaires, which included items on demographics, lifetime and past 30-day use of alcohol, intentions to drink over the next year, perceptions of peers’ and parental behaviors and attitudes regarding alcohol use, as well as perceptions regarding the availability of alcohol and other substances in their social environments.

Measures

Dependent Variables

Current alcohol use

Current alcohol use was assessed with a single item asking students to identify on “how many occasions during the past 30 days (if any) have you had at least one drink.” Responses were originally coded on a 7-point Likert-type scale. The majority of the sample reported no current use (93%) and only 1% reported greater than 1 day of alcohol use in the past 30 days. Due to this low variability in item responses, it was therefore recoded into a dichotomous variable, with no alcohol use coded as “0” and any days of alcohol use in the past 30 coded as “1.”

Lifetime alcohol use

Lifetime alcohol use, a measure to reflect any alcohol initiation by this sample of early adolescents, was coded as a dichotomous variable. Respondents were asked at what age they first used alcohol and all respondents who stated that they had never had a drink of alcohol were coded as “0” and all other ages of first use coded as “1.”

Alcohol use intentions

Alcohol use intentions was assessed with a single item that captures the likelihood of engaging in alcohol use in the next 12 months. Students were asked, “Do you think you will use any of these within the next year: Beer, wine, wine coolers or hard liquor (excluding use during religious ceremonies)?” Youth rated their intentions to use alcohol on a 5-point Likert-type scale ranging from 1 (definitely not) to 5 (definitely will). This scale was converted to a dichotomous variable, with adolescents endorsing “definitely not” coded as “0” and all other response categories coded as “1.”

Demographics

Race was coded as a dichotomous variable, students identifying as non-White coded as “0” and students identifying as White coded as “1.” Lunch status, a proxy for socioeconomic status, was coded as a dichotomous variable, with students endorsing a free or reduced free lunch status as “0” and full pay lunch status coded as “1.” Gender was also dichotomous, with males coded at “-1” and females coded as “1.”

Family Risk Predictors

Parental monitoring practices

The Poor Family Management Scale (Center for Substance Abuse Prevention, 2003) includes six items; examples of these statements include, “My parents ask if I have gotten my homework done” and “The rules in my family are clear.” Student’s responses to these statements ranged from 4 (NO!), 3 (no), 2 (yes), 1 (YES!). Reliability for this scale is 0.79.

Relationship quality

Poor relationship quality between an adolescent and his or her parents was measured using the 7-item Parent-Child Affective Quality Scale (Spoth, Redmond, & Project Family Research Group, 1997). Examples from this scale include, “During the past month when you and your parent have spent time talking or doing things together, how often did your parent let you know he/she really cares about you” and “When you do something wrong, how often does your parent lose his/her temper and yell at you?” This scale assesses the student’s degree of closeness and positive reinforcement behavior from parents on a 7-point Likert-type scale with responses ranging from 1 (Always) to 7 (Never). The Cronbach α for this measure is .80.

Peer Risk Predictors

Friend approval

Friend approval was assessed through a single item assessing, “How would your close friends feel if they thought you had five or more drinks once or twice a week? (USDHHS, 1997). The scale response options were: 3 (approve), 2 (wouldn’t care), 1 (disapprove).

Normative beliefs of peer use

The peer normative beliefs was measured by one item assessing youths’ perceptions regarding the amount of people their own age drink alcohol (Botvin, Baker, Renick, Filazzola, & Botvin, 1984). The scale response options include: 1 (none), 2 (less than half), 3 (about half), 4 (more than half), and 5 (all or almost all).

Community Context Risk Predictors

Perceived availability

The Perceived Availability of Drugs Scale (Arthur, Hawkins, Pollard, Catalano, & Baglioni, 2002) included 1 item, “How easy would it be for you to obtain liquor, including beer or hard liquor” from 1 (very hard) to 4 (very easy).

Tolerant community norms

The Laws and Norms Favorable to Drug Use Scale, (Arthur et al., 2002), is a 6-item scale that assesses community norms regarding use and perceived enforcement for violating use laws. The first three questions focus on community norms regarding use. These items use the same stem, “How wrong would most adults (over 21) in your neighborhood think it was for kids your age” and ask about neighborhood perceptions regarding marijuana use, drinking alcohol, and smoking cigarettes. Responses to these items range from 1 (very wrong) to 4 (not wrong at all). The last three items assess youth’s perceptions about how likely it would be for kids to get caught in their neighborhood using alcohol or marijuana, or get caught for carrying a handgun. Responses to these three questions include, 4 (NO!), 3 (no), 2 (yes), and 1 (YES!). The Cronbach α for this scale is .78.

Data Analyses

Analyses were conducted using SPSS version 19.0 (SPSS, 2009). Missing data were handled using EM imputation. Since surveys were collected from students in 12 different middle schools, intraclass correlations were conducted on each of the study variables. Analyses revealed small intraclass correlations between the independent risk variables and school (.001-.06), as well as between the dependent variables and school (.000-.009). Due to low intraclass correlations, it was determined that the analyses did not need to account for the clustering of students within different middle schools.

Three multivariate hierarchical logistic regression analyses were conducted to assess the relationship between demographic, family, peer, and environmental context risk variables and three dependent variables: current use of alcohol, lifetime use, and intentions to initiate alcohol use. All the predictor variables were standardized by centering each variable and dividing by the standard deviation in order to allow for comparison across different scales. Interaction terms for each of the predictors were then created by multiplying the value of each predictor by the dummy coded gender term (male = −1, female = 1). The first logistic regression analysis examined the associations between the set of predictor variables and current alcohol use. Gender, race (White vs. non-White), and lunch status (full pay vs. reduced fee lunch status) were entered into Block 1 of the regression model. The four peer and family risk variables, (a) friend approval of use, (b) peer normative beliefs, (c) parental monitoring, and (d) parent–child relationship quality were entered into Block 2 of the model. To examine whether the two community context variables contributed substantially to the prediction of current use, perceived availability and tolerant community norms, were entered as Block 3 in the regression equation. Lastly, gender interaction terms were entered into Block 4 of the regression model to test whether gender is a moderator of these peer, family, and community context risk factors. A second hierarchical regression for lifetime alcohol use was conducted with the predictors entered in the same order as the first regression analysis.

A third hierarchical logistic regression was conducted to examine the associations between the same set of risk predictors and intentions to initiate alcohol over the next year. Youth who had already initiated alcohol at the time of the survey (n = 126) were dropped from this analysis to assess seventh grade nonusers and therefore only 655 adolescents were included in this analysis. The demographic variables (gender, race, lunch status) were entered as predictors into Block 1 in this regression model. The four peer and family variables were again entered into Block 2 of the regression model. The community context risk factors were entered into Block 3 and gender interaction terms for the family, peer, and community context variables were entered as Block 4 in this regression model.

Results

Descriptive statistics of the sample are summarized in Table 1. The average age of the participants was 12.45 years (SD = 0.57); most participants were White (66%). The overall sample reported a lifetime rate of alcohol initiation of 16%, with 7% endorsing current use within the past month. Table 2 summarizes the rates of lifetime and current use, as well as alcohol use intentions over the next 12 months. Among nonusers, 87% reported that they “definitely” would not use alcohol in the next year, however, 13% of these early adolescents endorsed riskier behavioral intentions to initiate alcohol use over the next 12 months. Even though males reported slightly higher rates of lifetime and current use and intentions to use compared to females, χ² analyses revealed that these differences were not statistically significant. In contrast, both peer risk factors demonstrated mean differences by gender. Females endorsed that a greater number of their peers drink alcohol than males (t = −2.30, df = 779, p = .02) while males endorsed greater friend approval of alcohol use than females (t = 4.61, df = 779, p < .001). In addition, males were more likely to report more tolerant community attitudes towards use compared to females (t = 2.55, df = 779, p = .01).

Table 1.

Demographic Characteristics of the Sample.

	Males(n = 389)	Females(n = 392)	Total(N = 781)
Variable	N (%)	N (%)	N (%)
Race
White	257 (66)	262 (67)	519 (67)
Living situation
Two-parent household	294 (76)	298 (76)	592 (76)
Single parent household	95 (24)	94 (24)	189 (24)
SES-proxy
Reduced or free lunch	165 (42)	156 (40)	321 (41)
Full pay	224 (58)	236 (60)	460 (59)

Table 2.

Alcohol Use Outcome.

	Males	Females	Total
Variable	N (%)		N (%)
Past 30-day use
Yes	33 (8)	21 (5)	54 (7)
No	356 (92)	371 (95)	727 (93)
Lifetime use
Yes	73 (19)	53 (14)	126 (16)
No	316 (81)	339 (86)	655 (84)
Intentions to use alcohol
Yes	39 (12)	46 (15)	85 (13)
No	300 (88)	270 (85)	570 (87)

Note. Intentions to use alcohol are reported only for the portion of the sample that has not initiated alcohol use (n = 655).

Predictors of current use

Results from the logistic regression analyses demonstrated four significant predictors of current alcohol use, summarized in Table 3. Students who identified as White had a significantly increased odds of current alcohol use (OR: 3.47, 95% CI [1.49, 6.41]), gender and lunch status did not predict use. In partial support of the proposed hypotheses, one peer and one family risk variable predicted greater likelihood of current alcohol use. Youth who perceived that a greater percentage of their same aged peers drank alcohol had about twice the odds of current use (OR: 2.08, 95% CI [1.52, 2.83]). In addition, endorsing poor relationship quality with parents was also associated with significantly increased odds of current drinking (OR: 1.59, 95% CI [1.13, 2.24]) across the sample. Neither friend approval of use nor poorer parental monitoring of behavior predicted current use. After accounting for the significant effect of demographic, peer and family factors, greater perceived availability of alcohol was associated with an almost two-fold increase in risk of current use (OR: 1.87, 95% CI [1.29, 2.70]). Tolerant community norms and laws were not associated with increased odds of current use in the sample. None of the hypothesized gender interactions emerged as significant predictors of current alcohol use.

Table 3.

Hierarchical Logistic Regression Analyses for Current and Lifetime Alcohol Use.

n = 708	Past 30-day alcohol use			Lifetime alcohol use
Variables	OR	95% CI	p	OR	95% CI	p
InterceptBlock 1: Demographics Gender White Lunch status	0. 01 0.723.47** 1.34	0.46-1.121.49-8.11 0.57-3.15	< .001 .145 .004.505	0.05 0.74* 1.23 3.17**	0.57-0.97 0.69-2.17 1.76-5.71	< .001 .03.485 < .001
Block 2: Peer and family risk factors Peer normative beliefs Friend approval Parental monitoring Parent–child relationship	2.08 1.140.931.59	1.52-2.83 0.88-1.490.67-1.30 1.13-2.24	<.001 .326 .681.008	1.86 1.02 1.111.39	1.49-2.31 0.83-1.260.88-1.41 1.09-1.78	< .001 .825.009.367
Block 3: Community context risk factors Perceived availability of alcohol Tolerant community norms/laws	1.87* 1.16	1.29-2.700.82-1.65	.001.408	1.65* 1.19	1.29-2.120.88-1.41	< .001 .169
Block 4: Interaction effects (× gender) Peer normative beliefs × Friend approval × Parental monitoring × Parent–child relationship quality × Perceived availability of alcohol × Tolerant community norms/laws ×	0.94 1.081.10 1.060.970.95	0.69-1.29 0.83-1.400.79-1.54 0.75-1.51 0.70-1.33 0.66-1.38	.707.585.576 .727.835 .791	1.02 1.031.11 1.30* 1.01 0.72*	0.82-1.27 0.83-1.260.88-1.41 1.01-1.67 0.80-1.27 0.55-0.93	.829 .789.371.037 .939.013

Note. *indicates odds ratios significant at p < .05. **indicates odds ratios significant at p < .01.

Lifetime alcohol use

Table 3 also summarizes the significant predictors associated with having initiated any alcohol use within this sample, even if they had not reported any past 30 day use. Results showed that two demographic factors were associated with lifetime use; both being female and reduced fee lunch status, a proxy for lower socioeconomic status, were associated with decreased odds of lifetime use. Similar to current use, greater normative perceptions of use (OR: 1.86, 95% CI [1.49, 2.31]), poorer relationship quality with parents (OR: 1.39, 95% CI [1.09, 1.78]), and greater perceived availability of alcohol (OR: 1.65, 95% CI [1.29, 2.12]) were associated with greater odds of alcohol initiation after accounting for the significant effect of demographic variables. In addition, a couple hypothesized gender differences emerged. Poorer relationship quality was associated with greater odds of having initiated alcohol use for adolescent females while more tolerant community norms was associated with greater increased odds of lifetime use for males compared to females. None of the other predicted gender interactions were significant.

Alcohol use intentions

Demographic factors were not significantly associated with intentions to initiate alcohol in the next year. Study results demonstrated that the belief that a greater percentage of their same age peers drink alcohol was also significantly associated with intentions to initiate alcohol use among the portion of the sample that had not reported any lifetime alcohol use (OR: 1.56, 95% CI [1.19, 2.04]). None of the family risk predictors were associated with risk for initiating alcohol over the next year, however, both tolerant community norms and laws and perceived availability of alcohol were associated with increased odds of initiating alcohol use over the next year among nonusers after accounting for the significant effect of peer normative beliefs. One significant interaction effect was found for peer normative beliefs by gender; peer normative beliefs were more predictive of girl’s intentions to initiate alcohol use over the next 12 months compared to males (see Table 4). None of the hypothesized gender interactions were significant.

Table 4.

Hierarchical Logistic Regression for Intentions to Initiate Alcohol Over Next 12 Months.

	Alcohol use intentions
Variables	OR	95% CI	p
InterceptBlock 1: Gender White Lunch status	0.07 1.46 1.941.10	0.81-2.640.92-4.11 0.52-2.31	< .001 .210 .083 .809
Block 2: Peer & family risk factors Peer normative beliefs Friend approval Parental monitoring Parent–child relationship	1.56** 1.031.20 1.18	1.19-2.04 0.79-1.34 0.90-1.590.88-1.59	.001.821.221 < .001
Block 3: Community context risk factors Perceived availability of alcohol Tolerant community norms/laws	1.82** 1.46*	1.38-2.40 1.11-1.91	.266.006
Block 4: Interaction effects (× gender) Peer normative beliefs × Friend approval × Parental monitoring × Parent–child relationship quality × Perceived availability of alcohol × Tolerant community norms/laws ×	1.39* 0.931.010.870.89 1.09	1.06-1.82 0.72-1.21 0.76-1.35 0.65-1.17 0.68-1.16 0.82-1.45	.016 .604 .927.355.402.766

Note. *indicates odds ratios significant at p < .05. **indicates odds ratios significant at p < .01.

N = 655 for this analysis since it reflects only adolescents who had not yet initiated alcohol use.

Discussion

Results of the analyses supported the predictive role of demographic factors in risk for alcohol use among early adolescents. As supported by nationally representative prevalence data which has demonstrated higher rates of alcohol use among White adolescents (Substance Abuse and Mental Health Services Administration, 2010; Blum et al., 2000), as well as adolescents with higher income (Blum et al., 2000), we found that identifying as White lead to a more than three-fold increase in risk of current alcohol use and higher SES (as measured by lunch status) was also associated with a similar increase in odds of having initiated alcohol use. Gender significantly predicted a greater risk for having initiated alcohol use by seventh grade for boys with 19% of boys reporting having used alcohol at least once compared to only 14% of girls in the sample. This gender difference did not extend to risk of having used alcohol in the past month or in risk of initiating use within the next year.

The results of this study also provide strong support for peer norms influencing alcohol use outcomes among early adolescents and modest support for the presence of gender differences during this developmental period. These findings are consistent with previous studies that have also show that peers norms are important predictors of early alcohol use among adolescents (Bahr et al., 2005; Dick et al., 2007; Flannery et al., 1994; Hawkins et al., 1992; Kung & Farrell, 2000; Pope et al., 1994; Simons-Morton et al., 2001; Yeh et al., 2006). As males and females enter adolescence, perceptions regarding their peers’ attitudes and behaviors appear to play a central role in influencing adolescent’s own intentions regarding the use of alcohol and prior to initiating alcohol, perceptions regarding how normative alcohol is among their same aged peers may be particularly important for girls. In contrast, perceptions of friends’ attitudes towards alcohol use did not significantly predict any alcohol use or intentions to initiate use. This suggests that measuring general perceptions regarding peer alcohol norms may be most salient for assessing risk among early adolescents.

Family-level risk factors demonstrated a less consistent relationship with alcohol initiation and use within the sample. As expected, adolescents who indicated that they had poorer relationships with their parents were more likely to report that they had initiated alcohol use as well as used alcohol within the past month, indicating that the nature and quality of parental relationships with their children as they enter adolescence can influence the trajectory of alcohol use behavior for both males and females. Moreover, poorer relationship quality may place girls at particular risk of initiating alcohol use. Contrary to studies demonstrating an inverse relationship between a high degree of parental monitoring and likelihood of alcohol and other substance use among adolescents (Borawski et al., 2003; DiClemente et al., 2001; Dishion & McMahon, 1998; Griffin et al., 2000; Li, Stanton, & Feigelman, 2000; Steinberg et al., 1994), the current study failed to find support for parental monitoring having a direct effect on early adolescents’ alcohol use or intentions to initiate use. Shared variance between these two family variables may have contributed to the lack of significant findings. The correlation between the parental monitoring and the relationship quality scales was .40 which was the largest bivariate correlation between any of the risk predictors and this likely resulted in decreased power to find significant effects for both variables in the analyses. The authors choose to examine these risk factors separately rather than creating a composite of family-level risk since previous research demonstrated different result by gender for these different variables (Borawski, Ievers-Landis, Lovegreen, & Trapl, 2003; CASA, 2003; Flannery et al., 1994; Griffin et al., 2000; Kumpfer & Turner, 1990; Kung & Farrell, 2000; Rai et al., 2003).

In addition, the current study demonstrates the significant contribution of youth’s social environment, especially perceived availability or ease of access to obtain alcohol. With the notable exception of peer normative beliefs, greater perceived availability was the most consistent predictor of alcohol use or intentions to initiate alcohol use after accounting for significant demographic, peer or family risk factors. More permissive community attitudes towards alcohol use were also associated with a 46% increase in the odds of intending to initiate use within the next year among nonusers. Study findings also provide limited support for more tolerant community norms having a bigger impact of adolescent males’ risk of having initiated alcohol use compared to females. This gender difference is supported by other studies of community-level variables that also found that boys’ use of substances was especially impacted by their community or neighborhood environment compared to girls (Jones-Webb et al., 1997; Kumpfer & Turner, 1990; Substance Abuse and Mental Health Services Administration, 2002). Measuring youth’s social environment, including adolescents’ perceptions, as well as other community indicators of access to alcohol may provide vital information for identifying appropriate avenues for preventing initiation and escalation of alcohol use for both males and females; this may be particularly influential for males’ alcohol use.

Limitations

Results of the current study need to be interpreted within the context of several limitations of the sample and study design. First, data analyses were cross-sectional and caution should be taken when attributing causal relationships between variables. Second, community context variables were measures of youth’s perceptions of alcohol availability and community norms regarding use rather than empirical data describing adolescents’ community environment such as alcohol outlet density, or rates of illegal sales to minors from alcohol vendors. Similarly, parent and peer risk variables also represent adolescents’ perceptions regarding their peers and parents’ behaviors. Cohen & Rice (1997) found that youth’s perceptions of parents and parenting behaviors were associated with substance use outcomes, while parents’ own self-ratings failed to significantly predict substance use outcomes. Youth perceptions appear to capture adolescents’ views of their social world and community context and therefore can offer important insights into larger social systems within which adolescents exist (Cohen & Rice, 1997). Although survey research often relies on youth-report, the single source of the data can inflate the relationship between variables due to shared method variance. Third, data was not available on antisocial behavior or affiliation with deviant peers, both robust predictors of substance use (Hawkins Catalano, & Miller 1992). Fourth, due to low rates of current alcohol use which are consistent with rates reported among community samples of middle school age youth (Substance Abuse and Mental Health Services Administration, 2010), only dichotomous outcomes were examined (e.g., any past 30-day use, no past 30-day use) limiting the ability to examine factors predicting frequency or quantity of alcohol use within the sample. Finally, although EM imputation is a superior approach than deletion and other single imputation methods for handling missing data, it can potentially lead to biased estimates if data is not missing at random (Bennett, 2001; Schlomer, Bauman, & Card, 2010).

Prevention Implications

Tailored interventions, incorporating knowledge of gender differences, designed to address salient ecological risk factors may be more likely to result in more robust outcomes for youth and families. Consistent with other risk factor research, the study maintains the need to address normative peer beliefs regarding alcohol use. Since the empirical literature is inconsistent, greater attention is also needed to examine multiple types of peer influences (i.e., close friends’ use vs. normative beliefs regarding peer use; offers to use alcohol or drug from peers) across adolescence in order to gain a clearer picture of whether and how gender influences these key sources of risk. This study adds support for the effectiveness of social skill training and other preventive interventions targeting social norms for youth prior to middle school (e.g., addressing misperceptions regarding the percentage of peers that use alcohol), especially for adolescent females.

As youth transition to adolescence, the role of family influences, including parent and child relationship quality, still represent an important point of intervention to prevent the onset of underage drinking, and especially during early adolescence when peer influence begins to gain more prominence (Windle, 2000; Wood, Vinson, & Sher, 2001). In addition to prevention interventions that acknowledge the substantial role of peers in adolescent behavior, interventions that increase family involvement and prosocial interactions among parents and adolescents can provide protection against alcohol use. This appears to be especially important factor to delay onset of alcohol prior to age 12, since parent–child relationship quality emerged as a significant predictor of having already initiated alcohol use and having used in the past month among seventh graders, suggesting the need for family interventions in preteen years.

In addition, perceived availability or ease of obtaining alcohol emerged as another consistent risk factors for alcohol use or intentions to initiate use for both males and females. Over the past several decades, the substance abuse prevention field has placed greater emphasis on impacting environmental-level risk factors such as youths’ access to alcohol, as well as changing tolerant community norms and policies related to underage drinking. This study’s findings underscore the importance of environmental prevention strategies that specifically target community context risk variables, such as availability of alcohol or tolerant community norms regarding underage use. In addition, these findings contribute to the field’s understanding of how gender may influence the relationship between greater availability and alcohol use. Males appear to be at greater risk for early alcohol initiation when they perceive more permissive laws and community attitudes towards alcohol. Further attention should be focused on the role of gender in understanding the effectiveness of these types of intervention strategies, especially given the current attention and resources dedicated to community and environmental prevention efforts. Moreover, future research should continue to examine outcomes separately by gender among a larger set of risk predictors in order to gain a more complete picture of how gender moderates their effects on alcohol use outcomes at different points in development.

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: This work was supported by grant 5 UD1 SP09363-03 through the Substance Abuse and Mental Health Services Administration (SAMHSA) which funded the research study described in this manuscript.

Author Biographies

Jessica E. Nargiso, PhD, is currently a psychologist at Massachusetts General Hospital and instructor at Harvard Medical School in Boston, Massachusetts. Her research interests include adolescent substance use, gender differences, the evaluation of community-based prevention strategies, and health disparities.

Karen Friend, PhD, is currently a research scientist at the Pacific Institute for Research and Evaluation, located in Pawtucket, Rhode Island. She is also adjunct research faculty at the Warren Alpert Medical School of Brown University at the Center for Alcohol and Addiction Studies in Providence, Rhode Island. Her research interests include the evaluation of community-based prevention strategies and substance use treatment programs, and training and technical assistance to community organizations.

Paul Florin, PhD, is a professor in the Department of Psychology at the University of Rhode Island, located in Kingston, Rhode Island. His research interests include adolescent substance use, and the evaluation of community and state-level substance abuse prevention initiatives.

References

Arthur

M. W.

Hawkins

Pollard

Catalano

R. F.

Baglioni

A. J.

(2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The communities that care youth survey. Evaluation Review, 26, 575-601.

Averna

Hesselbrock

(2001). The relationship of perceived social support to substance use in offspring of alcoholics. Addictive Behaviors, 26, 363-374. doi:S0306-4603(00)00112-X[pii]

Bahr

S. J.

Hoffmann

J. P.

Yang

(2005). Parental and peer influences on the risk of adolescent drug use. Journal of Primary Prevention, 26, 529-551.

Barber

J.G.

Bolitho

Bertrand

L.D.

(1998). Age and gender differences in the predictors of adolescent drinking. Social Work Research, 22(3), 164-172.

Bennett

D. A.

(2001). How can I deal with missing data in my study? Australian and New Zealand Journal of Public Health, 25, 464-469.

Blum

R. W.

Beuhring

Shew

M. L.

Bearingwe

L. H.

Sieving

R. E.

Resnick

M. H.

(2000). The effects of race/ethnicity, income, and family structure on adolescent risk behaviors. American Journal of Public Health, 90, 1879-1884.

Borawski

E. A.

Ievers-Landis

C. E.

Lovegreen

L. D.

Trapl

E. S.

(2003). Parental monitoring, negotiated unsupervised time, and parental trust: The role of perceived parenting practices in adolescent health risk behaviors. Journal of Adolescent Health, 33, 60-70.

Botvin

G. J.

Baker

Renick

Filazzola

A. D.

Botvin

E. M.

(1984). A cognitive-behavioral approach to substance abuse prevention. Addictive Behaviors, 9, 137-147.

Bronfenbrenner

(1989). Ecological systems theory. In Vasta

(Ed.), Annals of child development (Vol. 6, pp. 187-249). Greenwich, CT: JAI Press.

10.

Callas

P. W.

Flynn

B. S.

Worden

J. K.

(2004). Potentially modifiable psychosocial factors associated with alcohol use during early adolescence. Addictive Behaviors, 29, 1503-1515. doi:10.1016/j.addbeh.2004.02.028 S0306460304000346 [pii]

11.

CASA (2003). The Formative Years: Pathways to Substance Abuse Among Girls and Young Women Ages 8-22. The National Center of Addiction and Substance Abuse at Columbia University. Retrieved April 22, 2012, from http://www.casacolumbia.org/articlefiles/380-Formative_Years_Pathways_to_Substance_Abuse.pdf

12.

Center for Substance Abuse Prevention (2003). Core measures initiative. Phase 1 recommendations. Retrieved from http://preventionplatform.samhsa.gov/Macro/Csap/dss_portal/

13.

Chang

S. L.

Yeh

H. H.

Lin

Y. S.

Chiang

Y. M.

T. K.

Yang

W. C.

(2007). The effect of centaurein on interferon-gamma expression and Listeria infection in mice. Toxicology and Applied Pharmacology, 219(1), 54-61. doi:S0041-008X(06)00455-8 [pii] 10.1016/j.taap.2006.11.026

14.

Cohen

D. A.

Rice

(1997). Parenting styles, adolescent substance use, and academic achievement. Journal of Drug Education, 27, 199-211.

15.

Hedderley

Wakeling

(1995). A comparison of imputation techniques for internal preference mapping using Monte Carlo simulation. Food Quality and Preference, 6, 281-297

16.

Dent

C. W.

Grube

J. W.

Biglan

(2005). Community level alcohol availability and enforcement of possession laws as predictors of youth drinking. Preventive Medicine, 40, 355-362. doi:S0091743504003305 [pii] 10.1016/j.ypmed.2004.06.014

17.

Dick

D. M.

Pagan

J. L.

Holliday

Viken

Pulkkinen

Kaprio

Rose

R. J.

(2007). Gender differences in friends’ influences on adolescent drinking: A genetic epidemiological study. Alcoholism: Clinical & Experimental Research, 31, 2012-2019. doi:ACER523 [pii] 10.1111/j.1530-0277.2007.00523.x

18.

DiClemente

R. J.

Wingood

G. M.

Crosby

Sionean

Cobb

B. K.

Harrington

M. K.

(2001). Parental monitoring: Association with adolescents’ risk behaviors. Pediatrics, 107, 1363-1368.

19.

Dishion

T. J.

McMahon

R. J.

(1998). Parental monitoring and the prevention of child and adolescent problem behavior: A conceptual and empirical formulation. Clinical Child and Family Psychology Review, 1(1), 61-75.

20.

Department of Health and Human Services (USDHHS). Substance Abuse and Mental Health Services Administration. Office of Applied Studies. National Household Survey on Drug Abuse, 1997. ICPSR02755-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2008-10-23. doi:10.3886/ICPSR02755.v2

21.

Eisenberg

M. E.

Neumark-Sztainer

Fulkerson

J. A.

Story

(2008). Family meals and substance use: Is there a long-term protective association? Journal of Adolescent Health, 43, 151-156.

22.

Epstein

J. A.

Botvin

G. J.

Baker

Diaz

(1999). Impact of social influences and problem behavior on alcohol use among inner-city Hispanic and Black adolescents. Journal of Studies on Alcohol, 60, 595-604.

23.

Fisher

L. B.

Miles

I. W.

Austin

Camargo

C. A.

Colditz

G. A.

(2007). Predictors of initiation of alcohol use among US adolescents. Archives of Pediatrics & Adolescent Medicine, 161, 959-966.

24.

Flannery

D. J.

Vazsonyi

A. T.

Torquati

Fridrich

(1994). Ethnic and gender differences in risk for early adolescent substance use. Journal of Youth and Adolescence, 23, 195-213.

25.

Grant

B. F.

Dawson

D. A.

(1997). Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: Results from the National Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 9, 103-110. doi:S0899-3289(97)90009-2 [pii]

26.

Griffin

Scheier L.M., Botvin

G. J.

Diaz

(2000). Ethnic and gender differences in psychosocial risk, protection, and adolescent alcohol use. Prevention Science, 1(4), 199-212.

27.

Harrison

P. A.

Fulkerson

J. A.

Park

(2000). The relative importance of social versus commercial sources in youth access to tobacco, alcohol, and other drugs. Preventive Medicine, 31, 39-48.

28.

Hawkins

D. J.

Catalano

R. F.

Miller

J. Y.

(1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention.

29.

Iowa Youth and Families Project (1989). Wave A (technical report) Iowa State University Center for Family Research in Rural Mental Health, Ames, IA.

30.

Johnston

L. D.

O’Malley

P. M.

Bachman

J. G.

Schulenberg

J. E.

(2004). Monitoring the future national survey results on drug use, 1975-2003. Volume I: Secondary school students. Bethesda, MD: National Institute on Drug Abuse.

31.

Jones-Webb

Toomey

T. L.

Short

Murray

D. M.

Wagenaar

Wolfson

(1997). Relationships among alcohol availability, drinking location, alcohol consumption, and drinking problems in adolescents. Substance Use & Misuse, 32, 1261-1285.

32.

Kosterman

Hawkins

J. D.

Guo

Catalano

R. F.

Abbott

R. D.

(2000). The dynamics of alcohol and marijuana initiation: Patterns and predictors of first use in adolescence. American Journal of Public Health, 90, 360-366.

33.

Kumpfer

K. L.

Turner

C. W.

(1990). The social ecology model of adolescent substance abuse: Implications for prevention. International Journal of the Addictions, 25(4A), 435-463.

34.

Kung

E. M.

Farrell

A. D.

(2000). The role of parents and peers in early adolescent substance use: An examination of mediating and moderating effects. Journal of Child and Family Studies, 9, 509-528.

35.

Kuntsche

Kuendig

Gmel

(2008). Alcohol outlet density, perceived availability and adolescent alcohol use: A multilevel structural equation model. Journal of Epidemiology and Community Health, 62, 811-816. doi:62/9/811 [pii] 10.1136/jech.2007.065367

36.

Stanton

Feigelman

(2000). Impact of perceived parental monitoring on adolescent risk behavior over 4 years. Journal of Adolescent Health, 27(1), 49-56. doi:S1054-139X(00)00092-6 [pii]

37.

Newcomb

M. D.

Felix-Ortiz

(1992). Multiple protective and risk factors for drug use and abuse: Cross-sectional and prospective findings. Journal of Personality and Social Psychology, 63, 280-296.

38.

Pope

S. K.

Smith

P. D.

Wayne

J. B.

Kelleher

K. J.

(1994). Gender differences in rural adolescent drinking patterns. Journal of Adolescent Health, 15, 359-365.

39.

Rai

A. A.

Stanton

Ying

Galbrith

Cottrell

Burns

(2003). Relative influences of perceived parental monitoring and perceived peer involvement on adolescent risk behaviors: An analysis of six cross-sectional data sets. Journal of Adolescent Health, 33, 108-118.

40.

Roski

Perry

C. L.

McGovern

P. G.

Williams

C. L.

Farbakhsh

Veblen-Mortenson

(1997). School and community influences on adolescent alcohol and drug use. Health Education Research, 12, 255-266.

41.

Schinke

S. P.

Fang

Cole

K. C.

(2008). Substance use among early adolescent girls: risk and protective factors. Journal of Adolescent Health, 43, 191-194. doi:S1054-139X(08)00096-7 [pii] 10.1016/j.jadohealth.2007.12.014

42.

Schlomer

G. L.

Bauman

Card

N. A.

(2010). Best practices for missing data management in counseling psychology, Journal of Counseling Psychology, 57(1), 1-10.

43.

Schulenberg

J. E.

Bryant

A. L.

O’Malley

P. M.

(2004). Taking hold of some kind of life: How developmental tasks relate to trajectories of well-being during the transition to adulthood. Development and Psychopathology, 16, 1119-1140.

44.

Schulte

M. T.

Ramo

Brown

S. A.

(2009). Gender differences in factors influencing alcohol use and drinking progression among adolescents. Clinical Psychology Review, 29, 535-547. doi:S0272-7358(09)00070-1 [pii]10.1016/j.cpr.2009.06.003

45.

Simons-Morton

Haynie

D. L.

Crump

A. D.

Eitel

S. P.

Saylor

K. E.

(2001). Peer and parent influences on smoking and drinking among early adolescents. Health Education & Behavior, 28(1), 95-107.

46.

Song

Smiler

A. P.

Wagoner

K. G.

Wolfson

(2012). Everyone says it’s OK: Adolescents’ perceptions of peer, parent, and community alcohol norms, alcohol consumption, and alcohol-related consequences. Substance Use & Misuse, 47, 86-98. doi:10.3109/10826084.2011.629704

47.

Steinberg

Fletcher

Darling

(1994). Parental monitoring and peer influences on adolescent substance use. Pediatrics, 93, 1060-1064.

48.

Substance Abuse and Mental Health Services Administration. (2002). The national cross-site evaluation of high-risk youth programs. Rockville, MD: Center for Substance Abuse Prevention.

49.

Substance Abuse and Mental Health Services Administration. (2009). Results from the 2008 National Survey on Drug Use and Health: National Findings HHS Publication No. SMA 09-4434 (Vol. NSDUH, Series H-36). Rockville, MD: Office of Applied Studies.

50.

Substance Abuse and Mental Health Services Administration. (2010). Results from the 2009 National Survey on Drug Use and Health: Volume I. Summary of National Findings No. HHS Publication No. SMA 10-4586 (Vol. NSDUH, Series H-38A). Rockville, MD: Office of Applied Studies.

51.

Webb

J. A.

Bray

J. H.

Getz

J. G.

Adams

(2002). Gender, perceived parental monitoring, and behavioral adjustment: Influences on adolescent alcohol use. American Journal of Orthopsychiatry, 72, 392-400.

52.

Windle

(2000). Parental, sibling, and peer influences on adolescent substance use and alcohol problems. Applied Developmental Science, 4(2), 98-110.

53.

Windle

Spear

L. P.

Fuligni

A. J.

Angold

Brown

J. D.

Pine

Dahl

R. E.

(2008). Transitions into underage and problem drinking: Developmental processes and mechanisms between 10 and 15 years of age. Pediatrics, 121, S273-S289.

54.

Wood

M. D.

Vinson

D. C.

Sher

K. J.

(2001). Alcohol use and misuse. In Baum

Revenson

Singer

(Eds.), Handbook of health psychology (pp. 281-318). Hillsdale, NJ: Erlbaum.

55.

Yeh

M. Y.

Chiang

I. C.

Huang

S. Y.

(2006). Gender differences in predictors of drinking behavior in adolescents. Addictive Behaviors, 31, 1929-1938. doi:S0306-4603(05)00312-6 [pii] 10.1016/j.addbeh.2005.12.019