Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Aug 1.
Published in final edited form as: Addict Behav. 2023 Mar 21;143:107697. doi: 10.1016/j.addbeh.2023.107697

Daily-level associations between alcohol use cognitions and normative perceptions among adolescents: An intensive longitudinal study

Katja A Waldron 1, Melissa A Lewis 2, Anne M Fairlie 3, Dana M Litt 2, Zhengyang Zhou 4, Dwalyn Bryant 1
PMCID: PMC10122708  NIHMSID: NIHMS1885969  PMID: 36965214

Abstract

Background:

Adolescence is an important developmental period in which to understand the cognitive underpinnings of risky alcohol use. Normative perceptions, such as descriptive and injunctive norms, are one of the strongest and most consistent predictors in adolescent drinking research. Thus, it is essential to examine which drinking cognitions (e.g., attitudes, prototypes, perceived vulnerability) are associated with normative drinking perceptions using repeated daily-level data among adolescents. The present study assessed associations between drinking cognitions and normative perceptions using an intensive daily longitudinal design.

Methods:

Participants were ages 15-17 years (N=306; 61.4% female; Mage (SD)=16.0 (0.8)) who were part of a larger ecological momentary assessment study (EMA) on drinking cognitions and alcohol use. The study design consisted of a 3-week EMA burst design (8 surveys per week, up to 2x/day) that was repeated quarterly over the 12-month study. The present analyses used the afternoon assessment for all measures.

Results:

Our multilevel model results demonstrated that drinking attitudes, prototypes of a typical drinker, and perceived vulnerability were positively associated with both descriptive and injunctive drinking norms between individuals and within individuals across days.

Conclusions:

Current findings have important clinical implications as they demonstrated how specific drinking cognitions were associated with variability in normative perceptions at the daily level. Findings support the delivery of intervention messaging to adolescents on days when drinking attitudes, prototypes of a typical drinker, and perceived vulnerability are elevated.

Keywords: intensive longitudinal design, alcohol use, cognitions, norms, adolescents

1. Introduction

It is well-established in the literature that alcohol use during adolescence typically occurs in the context of peers (Barnes et al., 2006; Ingram et al., 2007; Lewis et al., 2020; Pedersen et al., 2017). Cross-sectional and longitudinal research consistently show that social norms (i.e., cognitions concerning peer alcohol use) are significantly related to one’s own alcohol use (Borsari & Carey, 2001; Dumas et al., 2019; Lewis & Neighbors, 2006; Simons-Morton et al., 2018). Thus, researchers have emphasized the importance of including social norms when investigating adolescent drinking behavior (Litt & Lewis, 2016; Litt et al., 2019; Gerrard et al., 2008; Rivis & Sheeran, 2003). Descriptive normative perceptions refer to the perception of others’ quantity and frequency of a given behavior (i.e., how often an individual believes their peers drink; Cialdini et al., 1990). Injunctive normative perceptions refer to the perception of others’ approval of a given behavior (i.e., how much do others approve of drinking; Cialdini et al., 1990). An individual may implicitly or explicitly use these normative perceptions to determine which behaviors are socially acceptable. A large body of evidence indicates that both perceived descriptive and injunctive norms for drinking behaviors are significantly associated with actual drinking among adolescents (Borsari & Carey, 2001; Lewis & Neighbors, 2006; Neighbors et al., 2015; Simons-Morton et al., 2018).

1.1. Why predict drinking norms?

Much attention has been given to the occurrence of overestimating normative perceptions. When an individual perceives that others drink more than they really do (descriptive normative misperception) and are more approving of drinking than others really are (injunctive normative misperception), they in turn are more likely to drink themselves (Boyle et al., 2020; Neighbors et al., 2007; Lewis et al., 2015, 2020; Pedersen et al., 2017; Ward & Guo, 2020). Reducing overestimated perceptions using personalized normative feedback is a key focus of many drinking interventions (Carey et al., 2007; Cronce & Larimer, 2011; Miller et al., 2013). Although many social norms interventions are successful overall, they do not address day-to-day decisions and situations when personalized normative feedback interventions may be particularly beneficial. A better understanding of daily variation in both descriptive and injunctive normative perceptions for drinking in adolescents’ daily lives is needed. Adolescents’ drinking cognitions are often established prior to an individual engaging in a behavior (Ajzen 1985; Fishbein & Ajzen 1975; Gerrard et al., 2008; Kuntsche & Kuntsche, 2019; Smit et al., 2018); thus, cognitions related to a specific behavior may be indicative of a predisposition to engage in that behavior. Establishing which drinking cognitions are associated with a key target of existing efficacious norms interventions could help address risky drinking cognitions prior to the initiation of drinking on a given day.

1.2. Cognitive predictors of normative perceptions

How cognitions are associated with social norms has been guided by multiple theories, such as the Theory of Reasoned Action (Fishbein & Ajzen, 1975), Theory of Planned Behavior (Ajzen, 1988), Social Learning Theory (Bandura, 1969; 1977), Social Cognitive Theory (Bandura, 1986), and the Prototype Willingness Model (Gibbons et al., 1998). Several drinking cognitions, including attitudes, prototypes, and perceived vulnerability, are associated with descriptive and injunctive normative perceptions (Garcia et al., 2018; Gerrard et al., 2008; Lewis et al., 2020; Litt & Lewis, 2016). Attitudes are comprised of negative and positive evaluations of a behavior (e.g., acceptability of drinking; Armitage & Conner, 2001; Lewis et al., 2020). Prototypes are images of the typical person with certain traits (e.g., sex, age) who engage in a behavior with the two prototype factors being: (1) prototype favorability - the perception that a given prototype is associated with positive attributes; and (2) prototype similarity - the perception of oneself as similar to the prototype (Gerrard et al., 2008; Litt et al., 2019). The perception of being vulnerable to experiencing a consequence if one engaged in a risk behavior such as alcohol use is considered one’s perceived vulnerability to that risk behavior (Garcia et al., 2018; Gerrard et al., 2008).

Attitudes, prototypes, and perceived vulnerability are typically associated with drinking norms and subsequent drinking, but most of this research has been conducted with college students or young adults using cross-sectional designs (Dumas et al., 2019; Garcia et al., 2018; Hamilton et al., 2020; Litt et al., 2019; Lewis et al., 2020; Pedersen et al., 2017). To understand how drinking norms translate into alcohol use, it is important to study predictors of drinking norms among younger adolescents who have less experience with drinking. A daily-level examination of this age group would provide researchers with more precise information of day-to-day variability in drinking cognitions and normative perceptions to help build prevention programs for adolescents before alcohol use initiation.

1.3. The present study

The present study examined daily-level drinking attitudes, prototypes, and perceived vulnerability as predictors of perceived (1) descriptive drinking norms – quantity of alcohol consumed and percentage of friends drinking; and (2) injunctive drinking norms –friends’ general approval of drinking and quantity of alcohol consumed. Within-person effects hypothesized that on days individuals reported more approving attitudes, higher prototypes, and lower perceived vulnerability than average, they would report higher descriptive and injunctive drinking norms. Between-person effects hypothesized that individuals who on average reported more approving attitudes, higher prototypes, and lower perceived vulnerability would report higher descriptive and injunctive drinking norms.

2. Materials and Methods

2.1. Participants

Participants were part of a larger longitudinal measurement burst ecological momentary assessment study reporting drinking cognitions and alcohol use in a 3-week burst design (8 surveys per week) that was repeated quarterly across 12 months (N=1,015). Eligibility for the larger study included individuals (1) between 15-25 years old; (2) if age 18 or older, drink alcohol at least once a month over the last 6 months; (3) if age 15-17, no drinking criteria; (4) residing in Texas, (5) if female, must not be pregnant or trying to get pregnant; (6) be willing to complete a training session via Zoom; and (7) have a schedule that allows for daily surveys.

Data for the current study was restricted to adolescents ages 15-17 at baseline (N = 306; Mage (SD) = 16.0 (0.8); 61.4% female). The sample was limited to ages 15-17 because alcohol use is less likely to occur during this stage of development, but alcohol use cognitions and normative perceptions are starting to form. Racial and ethnic breakdown of the analytic sample was as follows: 27.9% White (non-Hispanic), 14.9% Black, 13.2% Asian, 5.7% Multiracial, 4.3% Another Race, 1.2% American Indian or Alaska Native, 0.5% Native Hawaiian or other Pacific Islander, and 32.3% Hispanic. The survey completion rates were 87.4%, 81.0%, 73.8%, and 70.8% for Bursts 1-4, respectively. The analysis sample included surveys with complete information on the variables included in that model.

2.2. Procedures

Participants were recruited through online and print advertisements, referrals, and flyers. Individuals who expressed interest in participating were emailed invitation letters containing the purpose of the study and a link to the online screening survey to determine eligibility. Those participants ages 15-17 years old consented to the study electronically and were required to provide parental consent from at least one parent or guardian by phone. After participants were deemed eligible and consented to the study, they were invited to complete a 45-minute online baseline survey ($25) followed by a 30-minute Zoom training session ($25) for which participants were compensated ($50 total). Participants were also asked to complete one web-based survey (~45 minutes) at 12 months post baseline for which they were paid $25.

Daily surveys started the first Thursday after completion of the Zoom training session. Over the course of 12 months, participants completed four bursts of daily surveys approximately every three months. Within each week of each burst, there were five weekend assessments (i.e., two on Friday, two on Saturday, and one on Sunday morning), and three assessments during the week (i.e., one during a random weekday, one on the following morning, and one on Thursday afternoon) for a total of eight weekly assessments. Each daily survey took approximately 5-7 minutes to complete. The first daily assessment of each quarter measured past month behavior, to better account for changes in their behavior across the year. Morning (AM) assessments measured drinking behavior and situational factors for the previous day. Afternoon (PM) assessments measured drinking cognitions and affect. The current study only used data from the PM surveys as this was when both drinking cognitions and norms were asked. Participants were allowed to choose a 3-hour window between 6am-12pm to complete their AM assessment, separately for weekdays and weekends. The PM assessment occurred randomly anytime within the designated afternoon block (1pm-6pm). Participants had one hour from the random prompt to complete the PM assessment. At each prompt, participants received a text message and email reminding them that their 3-hour AM assessment or one-hour PM assessment period was starting for completing the survey. Participants were compensated $3 per daily survey completed. Across all phases of the study, participants could earn up to $408.

2.3. Measures

2.3.1. Attitudes

Participants were asked about their personal drinking attitudes, “How many drinks do you think are acceptable for you to drink tonight?” Attitudes were assessed by response options that ranged from 0 (0 drinks) to 15 (15 or more drinks) (intraclass correlation (ICC) = 0.53).

2.3.2. Prototype favorability

Participants were asked, “Think about the typical [male/female] your age who drinks alcohol tonight.” Participants reported how much the following six words described their image of that person: (1) smart, (2) attractive, (3) popular, (4) impulsive, (5) immature, and (6) careless (# 4-6 reverse-coded). Prototype favorability was assessed by response options that ranged from 0 (not at all) to 4 (extremely) (Lewis et al., 2020). Items were internally consistent after adjusting for the multilevel structure (within-person ω = 0.58; between-person ω = 0.84; ICC = 0.55).

2.3.3. Prototype similarity

Participants were asked, “How similar are you to the typical [male/female] your age who drinks alcohol tonight?” Prototype similarity was assessed by response options that ranged from 0 (not at all similar) to 4 (very similar) (Litt et al., 2020) (ICC = 0.59).

2.3.4. Perceived vulnerability

Participants were asked, “How likely is it that something bad will happen to you tonight if you drink [amount of alcohol].” Perceived vulnerability was measured by asking about drinking different amounts of alcohol: (1) 4/5 or more (females/males) alcoholic drinks, (2) 1-3/1-4 (females/males) alcoholic drinks, and (3) any alcoholic drinks. Response options ranged from 0 (not at all likely) to 4 (very likely) (Lewis et al., 2020). Items were internally consistent after adjusting for the multilevel structure (within-person ω = 0.91; between-person ω = 0.97; ICC = 0.59).

2.3.5. Descriptive norms

First, participants were asked, “How many alcoholic drinks, on average, do you think your friends will individually consume tonight?” Responses ranged from 0 (0 drinks) to 15 (15 or more drinks) (ICC = 0.45). Second, participants were asked, “What percentage of your friends do you think will drink alcohol tonight?” (Lewis et al., 2020). Responses were open-ended from 0% to 100%. This resulted in two separate descriptive norms outcome variables (ICC = 0.40).

2.3.6. Injunctive norms

First, participants were asked, “Tonight, your friends think that drinking [amount of alcohol] would be…” The first injunctive norms variable was assessed by asking about each of three different amounts of alcohol (see “perceived vulnerability” measure for items and response options). Items were internally consistent after adjusting for the multilevel structure (within-person ω = 0.87; between-person ω = 0.97; ICC = 0.59). Second, participants were asked, “How many drinks do your friends think are acceptable for them to drink tonight?” Responses ranged from 0 (0 drinks) to 15 (15 or more drinks) (Lewis et al., 2020). This resulted in two separate injunctive norms outcome variables (ICC = 0.54).

2.3.7. Daily covariates

Daily-level covariates were Weekend (coded 1 = Friday or Saturday and 0 = Otherwise; Merrill et al., 2018; Thrul et al., 2018), Month (coded from 0 = January to 11 = December), and Burst (coded 0-3).

2.3.8. Baseline covariates

Participants reported age and biological sex assigned at birth (coded 0 = Female and 1 = Male). Race and ethnicity were recoded into five groups for analyses: White, Non-Hispanic (NH) [reference group]; Asian, NH; Black, NH; Hispanic; and Another Race (e.g., Multiracial, Native American). Participants were asked, “During your lifetime, on how many occasions (if any) have you had any alcoholic beverage to drink---more than just a few sips…” Lifetime alcohol use was coded dichotomously (0 = Never and 1 = 1 or more occasions).

2.4. Analytic plan

All analyses were conducted in R Version 4.0.3 (R Core Development Team, 2021). Descriptive analyses were conducted prior to inferential statistical analyses. To account for the multilevel daily data structure where days at Level 1 (daily-level) were nested within people at Level 2 (person-level), multilevel generalized linear mixed models with a random intercept for person-level were estimated to test each hypothesis. Because there were no hypotheses regarding individual differences in the effects of predictors, random slopes were not estimated in the multilevel models to reduce potential concerns in model estimation and over-fitting.

The current study examined daily-level associations where drinking cognitions (i.e., attitudes, prototypes, perceived vulnerability) were used to predict four outcomes: (1) descriptive norms - number of drinks participants think a friend will consume, which was estimated using a multilevel zero-inflated negative binomial model (count outcome with excessive zeros); (2) descriptive norms - percentage of friends expected to drink any alcohol, which was then dichotomized into 0 if 0 percentage and 1 if 1+ percentages and estimated using a multilevel logistic model (binary outcome); (3) injunctive norms - perceived friend approval of drinking, which was estimated using a multilevel linear model (continuous outcome); and (4) injunctive norms - perceived number of drinks participants think friends feel is acceptable to consume, which was then dichotomized into 0 if 0 drink and 1 if 1+ drinks and estimated using a multilevel logistic model (binary outcome). The reasons for dichotomizing descriptive norms (percentage of friends expected to drink alcohol) and injunctive norms (number of drinks participants perceive their friends think is acceptable to consume) outcomes are two-fold. First, the raw scale of the two outcomes were count variables regarding the number of drinks, which had excessive zeros with zero proportions > 50% in this adolescent sample. To account for such zero-inflation at raw scale, multilevel zero-inflated count models were conducted but had convergency issues likely due to the skewness of data. Second, only less than half (<50%) of the outcome values were positive counts, which has small variability. Dichotomizing outcomes can increase model robustness and interpretability with only limited information loss, which informed the decision to analyze them as binary variables.

In each model, the daily-level cognition constructs were centered within-person at Level 1. Daily-level cognition predictors were also included as person-level covariates at Level 2 using the person-mean of each construct, which was grand-mean centered. Age, birth sex, race, ethnicity, lifetime alcohol use at baseline, weekend/weekday, month, and burst number were included as covariates, where continuous variables were mean centered in all models (Brauer & Curtin, 2018).

3. Results

3.1. Descriptive information

See Table 1 for descriptive statistics on all key variables by person-level (N = 306 adolescents) and day-level (N = 9,174 days). Percentages were reported for demographic variables, while means, standard deviations, and ranges were reported for continuous variables.

Table 1.

Descriptive Statistics for Key Variables

Predictor N
(people or
days)		 M (SD) or
%		 Range
Person-Level
  Sex [n % male biological sex] 306 39% 0-1
  Age 306 16.01 (0.83) 15-17
  Race 304
    White, NH 115 38%
    Asian, NH 43 14%
    Black, NH 41 13%
    Hispanic 96 32%
    Other 9 3%
  Lifetime Drinking [n % drinkers] 306 49% 0-1
  Attitudes 305 0.58 (1.00) 0-8.29
  Prototype Favorability 305 1.61 (0.42) 0.07-3.04
  Prototype Similarity 305 0.65 (0.66) 0-2.86
  Perceived Vulnerability 305 1.87 (1.09) 0-4
  # Drinks of Friend 305 0.71 (1.03) 0-8.86
  % of Friends Drinking 303 6.67 (10.07) 0-75
  Perceived Friend Approval of Drinking 305 1.01 (0.72) 0-2.98
  Perceived Friend Acceptable # Drinks 305 1.20 (1.42) 0-8.86
Day-Level
  Attitudes 9174 0.52 (1.21) 0-15
  Prototype Favorability 9174 1.62 (0.55) 0-4
  Prototype Similarity 9174 0.65 (0.85) 0-4
  Perceived Vulnerability 9174 1.89 (1.40) 0-4
  # Drinks of Friend 9174 0.67 (1.40) 0-15
  % of Friends Drinking 9120 0.04 (0.04) 0-100
  Perceived Friend Approval of Drinking 9174 1.01 (0.93) 0-4
  Perceived Friend Acceptable # Drinks 9174 1.15 (1.85) 0-15
Open in a new tab

Note: NH = Non-Hispanic.

3.2. Multilevel models predicting daily-level drinking norms

We tested the effects of four predictors of drinking cognitions (i.e., attitudes, prototype favorability, prototype similarity, perceived vulnerability) on four perceived drinking norms outcomes (i.e., number of drinks for a friend, percentage of friends drinking, friend drinking approval, number of drinks acceptable by friends). The following results are summarized for the outcome “perceived number of drinks for a friend” and presented in Table 2. First, within-person variation in attitudes and prototype similarity were significantly associated with same-day perceived number of drinks for a friend in the count portion (top panel of Table 2) and the likelihood of perceiving friends will drink any alcohol in the logistic portion (bottom panel of Table 2). Specifically, on days when a person reported elevated attitudes and prototype similarity (i.e., higher than their own average), they also perceived that a friend would consume a higher number of drinks and had a higher likelihood of perceiving that a friend will have any drinks. Within-person variation in perceived vulnerability was not associated with same-day perceived number of drinks for a friend. Within-person covariates indicated that a higher perceived number of drinks for a friend were reported on weekend days, later calendar months, and later bursts in the count portion, and a higher likelihood of perceiving a friend will have any drinks was reported on days on weekend days and later bursts in the logistic portion.

Table 2.

Zero-Inflated Negative Binomial Multilevel Model Predicting # Drinks of Friend

Zero-Inflated Negative Binomial Regression Submodel (Count)
OR/RR 95% CI
LL, UL
Intercept 0.19*** 0.12, 0.31
Person-Level
Male Biological Sex 0.61* 0.39, 0.97
Age 1.17 0.94, 1.45
Race (White, NH - reference)
 Asian, NH 0.55 0.28, 1.08
 Black, NH 1.25 0.64, 2.45
 Hispanic 1.33 0.80, 2.21
 Another Race 0.74 0.21, 2.66
Lifetime Drinking 1.19 0.89, 1.58
Attitudes (GMC) 1.78*** 1.41, 2.24
Prototype Favorability (GMC) 1.37* 1.06, 1.77
Prototype Similarity (GMC) 1.32* 1.02, 1.71
Perceived Vulnerability (GMC) 1.18 0.93, 1.50
Daily-Level
Weekend 1.60*** 1.48, 1.74
Month 1.08*** 1.04, 1.12
Burst 1.13*** 1.08, 1.17
Attitudes (CWP) 1.09*** 1.06, 1.11
Prototype Favorability (CWP) 1.09 0.98, 1.22
Prototype Similarity (CWP) 1.13*** 1.07, 1.19
Perceived Vulnerability (CWP) 1.02 0.97, 1.06
Logistic Regression Submodel (Likelihood of Zero)
Intercept 0.15*** 0.07, 0.34
Person-Level
Male Biological Sex 0.33*** 0.19, 0.57
Age 0.88 0.73, 1.05
Race (White, NH - reference)
 Asian, NH 1.25 0.64, 2.46
 Black, NH 1.22 0.71, 2.11
 Hispanic 3.00*** 1.81, 4.98
 Another Race 2.47* 1.11, 5.47
Lifetime Drinking 1.10 0.86, 1.40
Attitudes (GMC) 0.07*** 0.02, 0.25
Prototype Favorability (GMC) 0.81* 0.67, 0.98
Prototype Similarity (GMC) 1.21 0.95, 1.54
Perceived Vulnerability (GMC) 0.52*** 0.41, 0.66
Daily-Level
Weekend 0.24*** 0.17, 0.34
Month 0.82* 0.69, 0.98
Burst 1.43*** 1.19, 1.71
Attitudes (CWP) 0.06*** 0.02, 0.22
Prototype Favorability (CWP) 0.36*** 0.23, 0.57
Prototype Similarity (CWP) 0.68* 0.50, 0.93
Perceived Vulnerability (CWP) 0.87 0.79, 1.10
Open in a new tab

Note: Daily-level predictors were centered within-person (CWP); Person-level predictors were grand-mean-centered (GMC); LL = lower limit; UL = upper limit; NH = Non-Hispanic; Number of people = 302; Number of days = 8,163.

*

p < .05

**

p < .01

***

p < .001

Between-person variation in attitudes, prototype favorability, and prototype similarity were significantly associated with perceived number of drinks for a friend in the count portion, while only attitudes and prototype favorability were associated with likelihood of perceiving friends will drink any alcohol in the logistic portion. Prototype similarity and perceived vulnerability was not associated with the likelihood of perceiving friends will drink in the logistic portion. Specifically, participants who reported a higher average level of attitudes, prototype favorability, and prototype similarity tended to report perceiving higher number of drinks for their friends. Participants who reported a higher average level of attitudes and prototype favorability tended to report higher likelihood of perceiving friends will drink any alcohol across the sampled days. Between-person covariates indicated that male participants perceived a lower number of drinks for a friend and higher likelihood of perceiving friends will have any drinks.

Second, we tested the effects of attitudes, prototype favorability, prototype similarity, and perceived vulnerability on drinking norms outcomes perceived percentage of friends drinking, friend approval of drinking, and friend acceptable number of drinks (Table 3). Within-person variation in attitudes, prototype favorability, prototype similarity, and perceived vulnerability were significantly positively associated with same-day: (1) perceived percentage of friends drinking, (2) friend approval of drinking, and (3) friend acceptable number of drinks. Within-person covariates indicated that a higher perceived percentage of friends drinking, friend approval of drinking, and friend acceptable number of drinks were reported on weekend days and later months in the study.

Table 3.

Logistic and Normal Regression Multilevel Models Predicting Descriptive and Injunctive Norms

Logistic Regression:
% Friends Drinking
N = 301
N (days) = 8,217		 Normal Regression:
Friend Approval of Drinking
N = 303
N (days) = 8,260		 Logistic Regression:
Friend Acceptable # of Drinks
N = 303
N (days) = 8,151
Odds
Ratio		 95% CI
LL, UL 		beta 95% CI
LL, UL 		Odds
Ratio		 95% CI
LL, UL
Intercept 0.16*** 0.07, 0.37 0.94*** 0.80, 1.08 0.50* 0.26, 0.98
Person-Level
Male Biological Sex 1.68 0.73, 3.83 −0.20** −0.34, −0.06 1.14 0.58, 2.22
Age 1.56* 1.06, 2.29 0.01 −0.06, 0.08 1.70** 1.24, 2.33
Race (White, NH - reference)
  Asian, NH 0.42 0.13, 1.39 −0.10 −0.31, 0.11 0.32* 0.12, 0.84
  Black, NH 1.02 0.29, 3.54 0.04 −0.17, 0.25 1.18 0.43, 3.24
  Hispanic 0.65 0.26, 1.60 0.04 −0.11, 0.20 1.74 0.83, 3.64
  Another Race 0.22 0.02, 2.01 0.05 −0.34, 0.45 0.51 0.09, 2.97
Lifetime Drinking 1.03 0.61, 1.73 0.08 −0.01, 0.17 0.85 0.55, 1.29
Attitudes (GMC) 4.59*** 2.70, 7.80 0.32** 0.24, 0.40 16.24*** 10.10, 26.10
Prototype Favorability (GMC) 1.32 0.85, 2.07 0.10* 0.02, 0.18 1.77** 1.23, 2.56
Prototype Similarity (GMC) 1.97** 1.23, 3.16 0.04 −0.04, 0.13 1.08 0.74, 1.58
Perceived Vulnerability (GMC) 1.67* 1.10, 2.53 0.04 −0.02, 0.11 2.26*** 1.62, 3.15
Daily-Level
Weekend 3.71*** 3.15, 4.37 0.17** 0.15, 0.20 3.18*** 2.70, 3.74
Month 1.20*** 1.11, 1.30 0.03*** 0.02, 0.05 1.18*** 1.09, 1.28
Burst 0.93 0.85, 1.01 −0.02* −0.03, −0.00 0.70*** 0.64, 0.77
Attitudes (CWP) 1.75*** 1.59, 1.93 0.15*** 0.13, 0.16 4.39*** 3.74, 5.15
Prototype Favorability (CWP) 2.13*** 1.69, 2.69 0.20** 0.16, 0.23 1.98*** 1.57, 2.49
Prototype Similarity (CWP) 1.55*** 1.35, 1.78 0.13** 0.10, 0.15 1.32*** 1.14, 1.53
Perceived Vulnerability (CWP) 1.10* 1.01, 1.20 0.02* 0.00, 0.03 1.10* 1.02, 1.20
Open in a new tab

Note: Daily-level predictors were centered within-person (CWP); Person-level predictors were grand-mean-centered (GMC); LL = lower limit; UL = upper limit; NH = Non-Hispanic

*

p < .05

**

p < .01

***

p < .001

Between-person variation in attitudes was significantly associated with higher perceived percentage of friends drinking, friend approval of drinking, and friend acceptable number of drinks (Table 3). Between-person variation in prototype favorability was significantly associated with more friend approval of drinking and friend acceptable number of drinks. Prototype similarity was significantly associated with higher perceived percentage of friends drinking. Lastly, perceived vulnerability was significantly associated with higher perceived percentage of friends drinking and friend acceptable number of drinks. However, not all findings supported hypotheses. Notably, prototype favorability was not associated with percentage of friends drinking. Prototype similarity was not associated with friend approval of drinking or friend acceptable number of drinks. Perceived vulnerability was not associated with friend approval of drinking. Between-person covariates indicated that: (1) older participants perceived a higher percentage of friends drinking and friend acceptable number of drinks, (2) females reported higher perceived friend drinking approval, and (3) Asian, non-Hispanic participants reported lower perceived friend acceptability of number of drinks compared to non-Hispanic white participants.

4. Discussion

Our results demonstrate that attitudes, prototypes, and perceived vulnerability varied from day-to-day within persons. These daily-level fluctuations (e.g., elevated cognitions on a given day) were associated with both perceived descriptive and injunctive drinking norms. Drinking attitudes were positively associated with both descriptive and injunctive drinking norms at the between- and within-person levels. Within-person effects showed that on days when individuals had more accepting attitudes toward drinking, their perceptions of friends’ drinking and approval of drinking were also higher. Our between-person findings were consistent with previous literature where individuals with more positive drinking attitudes report more positive normative perceptions of drinking (Ajzen, 1988; Bandura, 1986; Fishbein & Ajzen, 1975; Gerrard et al., 2008). Prototype favorability and similarity were consistently associated with descriptive and injunctive norms at the daily level. These findings align with other data at the global level indicating that prototypes are strongly associated with drinking outcomes (Gerrard et al., 2008). The current study results extend previous research by showing when adolescents are most at risk for increased descriptive and injunctive drinking norms at the daily level based on attitudes and prototypes.

Perceived vulnerability was not consistently associated with descriptive or injunctive norms at the daily level. Perceived vulnerability may be less relevant to peer drinking (i.e., normative perceptions) when compared to one’s own drinking, given that that one’s personal risk may not be directly associated with perceptions of others’ behavior. It may be that perceived vulnerability predicts personal drinks consumed more so than perceived peer drinking. Lastly, age was positively associated with descriptive norms (perceived percentage of friends drinking) and injunctive norms (perceived friend approval of drinking). These results suggest that older individuals (i.e., 17 vs. 15-year-olds) are more likely to perceive that more of their friends’ drink and that those friends are more approving of drinking. This provides evidence for the importance of including age as a covariate when studying adolescents as many developmental shifts are occurring. Drinking cognitions are often established prior to initiating drinking behavior (e.g., Kuntsche & Kuntsche, 2019; Smit et al., 2018), suggesting that examining whether cognitions are differentially associated with social drinking norms across adolescence is of importance.

4.1. Clinical implications and future directions

Findings could improve current approaches by delivering intervention messages to adolescents on days when they are at risk for higher normative perceptions. Specifically, content might be targeted for days when attitudes and prototypes are higher than usual, given that these fluctuations were associated with greater descriptive and injunctive norms. For example, personalized text messages could be sent when individuals are with peers who they perceive as heavy drinkers, with the goal of reducing the influence of normative perceptions that may lead to heightened risk.

Research has consistently shown that correcting normative misperceptions with interventions that include personalized normative feedback have been effective at reducing high-risk drinking among adolescents (Carey et al., 2007; Cronce & Larimer, 2011; Lewis & Neighbors, 2006; Miller et al., 2013; Lewis & Neighbors, 2007; Neighbors et al., 2010; NIAAA, 2019). However, these interventions do not consider windows of risk when interventions may be particularly necessary, such as weekend nights or days when normative perceptions and their cognitive antecedents are elevated. Based on the current study results, adolescents may be more receptive to personalized normative feedback on days when attitudes and prototypes are higher than average. With major technological shifts in how adolescents receive information, a personalized smartphone app intervention could be tested where content is delivered on days with higher-risk perceived attitudes and prototypes. Lastly, future research should examine differences between same-sex norms and general norms drinking outcomes among adolescents (i.e., “typical female your age” vs. “typical person your age”).

Highlights.

  • Cognitive and normative underpinnings of risky alcohol use in adolescents are important.

  • We examined daily-level drinking attitudes, prototypes, and perceived vulnerability.

  • Drinking attitudes were positively related to descriptive and injunctive norms.

  • Prototypes were positively related to descriptive and injunctive norms.

  • Perceived vulnerability may be less relevant to peer drinking.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest

All authors (Waldron, Lewis, Fairlie, Litt, Zhou, Bryant) declare they have no conflicts of interest.

References

  1. Ajzen I (1985). From intentions to actions: a theory of planned behavior. In Kuhi J &, Beckmann J (Eds.), Action-Control: From Cognition to Behavior (pp. 11–39), Heidelberg: Springer. [Google Scholar]
  2. Ajzen I (1988). Attitudes, personality, and behavior. Chicago: Dorsey Press. [Google Scholar]
  3. Armitage CJ, & Conner M (2001). Efficacy of the Theory of Planned Behaviour: A meta-analytic review. British Journal of Social Psychology, 40(4), 471–499. [DOI] [PubMed] [Google Scholar]
  4. Bandura A (1969). Social learning of moral judgments. Journal of Personality and Social Psychology, 11, 275–279. [DOI] [PubMed] [Google Scholar]
  5. Bandura A (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. [DOI] [PubMed] [Google Scholar]
  6. Bandura A (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice - Hall. [Google Scholar]
  7. 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(4), 1084–1104. [Google Scholar]
  8. Borsari B, & Carey KB (2001). Peer influences on college drinking: A review of the research. Journal of Substance Abuse, 13(4), 391–424. [DOI] [PubMed] [Google Scholar]
  9. Boyle HK, Merrill JE, & Carey KB (2020). Location-specific social norms and personal approval of alcohol use are associated with drinking behaviors in college students. Substance Use and Misuse, 55(10), 1650–1659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brauer M, & Curtin JJ (2018). Linear mixed-effects models and the analysis of nonindependent data: A unified framework to analyze categorical and continuous independent variables that vary within-subjects and/or within-items. Psychological Methods, 23(3), 389–411. [DOI] [PubMed] [Google Scholar]
  11. Carey KB, Merrill JE, Boyle HK, & Barnett NP (2020). Correcting exaggerated drinking norms with a mobile message delivery system: Selective prevention with heavy-drinking first-year college students. Psychology of Addictive Behaviors, 34(3), 454–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Carey KB, Scott-Sheldon LAJ, Carey MP, & DeMartini KS (2007). Individual-level interventions to reduce college student drinking: A Meta-analytic review. Addictive Behaviors, 32(11), 2469–2494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cialdini RB, Reno RR, & Kallgren CA (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58(6), 1015–1026. [Google Scholar]
  14. Cronce JM, & Larimer ME (2011). Individual-focused approaches to the prevention of college student drinking. Alcohol Research and Health, 34(2), 210–221. [PMC free article] [PubMed] [Google Scholar]
  15. Dotson KB, Dunn ME, & Bowers CA (2015). Stand-alone personalized normative feedback for college student drinkers: A meta-analytic review, 2004 to 2014. PLoS One, 10(10), e0139518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dumas TM, Davis JP, & Neighbors C (2019). How much does your peer group really drink? Examining the relative impact of overestimation, actual group drinking and perceived campus norms on university students' heavy alcohol use. Addictive Behaviors, 90, 409–414. [DOI] [PubMed] [Google Scholar]
  17. Fishbein M, & Ajzen I (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, Mass: Addison-Wesley Pub. [Google Scholar]
  18. Gerrard M, Gibbons FX, Houlihan AE, Stock ML, & Pomery EA (2008). A dual-process approach to health risk decision making: The prototype willingness model. Developmental Review, 28(1), 29–61. [Google Scholar]
  19. Gibbons FX, Gerrard M, Blanton H, & Russell DW (1998). Reasoned action and social reaction: Willingness and intention as independent predictors of health risk. Journal of Personality and Social Psychology, 74, 1164–1180. [DOI] [PubMed] [Google Scholar]
  20. Ingram JR, Patchin JW, Huebner BM, McCluskey JD, & Bynum TS (2007). Parents, friends, and serious delinquency: An examination of direct and indirect effects among at-risk early adolescents. Criminal Justice Review, 32(4), 380–400. [Google Scholar]
  21. Kuntsche E, & Kuntsche S (2019). Parental drinking and characteristics of family life as predictors of preschoolers' alcohol-related knowledge and norms. Addictive Behaviors, 88, 92–98. [DOI] [PubMed] [Google Scholar]
  22. Lewis MA, Litt DM, King KM, Fairlie AM, Waldron KA, Garcia TA, LoParco C, & Lee CM (2020). Examining the ecological validity of the prototype willingness model for adolescent and young adult alcohol use. Psychology of Addictive Behaviors, 34(2), 293–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lewis MA, Litt DM, & Neighbors C (2015). The Chicken or the Egg: Examining Temporal Precedence Among Attitudes, Injunctive Norms, and College Student Drinking. Journal of Studies on Alcohol and Drugs, 76(4), 594–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lewis MA, & Neighbors C (2006). Social norms approaches using descriptive drinking norms education: A review of the research on personalized normative feedback. Journal of American College Health, 54(4), 213–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lewis MA, & Neighbors C (2007). Optimizing personalized normative feedback: The use of gender-specific referents. Journal of Studies on Alcohol and Drugs, 68, 228–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Litt DM, & Lewis MA (2016). Examining a social reaction model in the prediction of adolescent alcohol use. Addictive Behaviors, 60, 160–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Litt DM, Waldron KA, Wallace EC, & Lewis MA (2019). Alcohol-specific social comparison as a moderator of the norms-behavior association for young adult alcohol use. Addictive Behaviors, 90, 92–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Litt DM, Lewis MA, Fairlie AM, & Head-Corliss M (2020). An examination of the relative associations of prototype favorability, similarity, and their interaction with alcohol and alcohol-related risky sexual cognitions and behavior. Emerging Adulthood, 8, 168–174. [Google Scholar]
  29. Merrill JE, Boyle HK, Barnett NP, & Carey KB (2018). Delivering normative feedback to heavy drinking college students. Addictive Behaviors, 83, 175–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Merrill JE, Miller MB, Balestrieri SG, & Carey KB (2016). Do my peers approve? Interest in injunctive norms feedback delivered online to college student drinkers. Addictive Behaviors, 58, 188–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Miller MB, Leffingwell T, Claborn K, Meier E, Walters S, & Neighbors C (2013). Personalized feedback interventions for college alcohol misuse: An update of Walters & Neighbors (2005). Psychology of Addictive Behaviors, 27(4), 909–920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. National Institute on Alcohol Abuse and Alcoholism [NIAAA]. (2019). Planning alcohol interventions using NIAAA’s College AIM: Alcohol Intervention Matrix. Rockville: National Institute on Alcohol Abuse and Alcoholism; NIH publication no. 15-AA-8017. [Google Scholar]
  33. Neighbors C, LaBrie JW, Hummer JF, Lewis MA, Lee CM, Desai S, Kilmer, & Larimer ME (2010). Group identification as a moderator of the relationship between perceived social norms and alcohol consumption. Psychology of Addictive Behaviors, 24(3), 522–528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Neighbors C, Lee CM, Lewis MA, Fossos N, & Larimer ME (2007). Are social norms the best predictor of outcomes among heavy-drinking college students? Journal of Studies on Alcohol and Drugs, 68(4), 556–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Neighbors C, Rodriguez LM, Rinker D, Gonzales RG, Agana M, Tackett JL, & Foster DW (2015). Efficacy of personalized normative feedback as a brief intervention for college student gambling: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83, 500–511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pedersen ER, Osilla KC, Miles JN, Tucker JS, Ewing BA, Shih RA, & D'Amico EJ (2017). The role of perceived injunctive alcohol norms in adolescent drinking behavior. Addictive Behaviors, 67, 1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Prince MA, Maisto SA, Rice SL, & Carey KB (2015). Development of a face-to-face injunctive norms brief motivational intervention for college drinkers and preliminary outcomes. Psychology of Addictive Behaviors, 29(4), 825–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. R Core Development Team. (2021). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. [Google Scholar]
  39. Reid AE, & Carey KB (2015). Interventions to reduce college student drinking: State of the evidence for mechanisms of behavior change. Clinical Psychology Review, 40, 213–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rivis A, & Sheeran P (2003). Descriptive norms as an additional predictor in the Theory of Planned Behaviour: A Meta-Analysis. Current Psychology, 22(3), 218–233. [Google Scholar]
  41. Simons-Morton B, Haynie D, Bible J, Liu D (2018). Prospective associations of actual and perceived descriptive norms with drinking among emerging adults. Substance Use and Misuse, 53(11), 1771–1781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Smit CR, de Leeuw RNH, Bevelander KE, Burk WJ, Buijs L, van Woudenberg TJ, & Buijzen M (2018). An integrated model of fruit, vegetable, and water intake in young adolescents. Health Psychology, 37(12), 1159–1167. [DOI] [PubMed] [Google Scholar]
  43. Thrul J, Lipperman-Kreda S, & Grube JW (2018). Do associations between drinking event characteristics and underage drinking differ by drinking location? Journal of Studies on Alcohol and Drugs, 79(3), 417–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ward RM, & Guo Y (2020). Examining the relationship between social norms, alcohol-induced blackouts, and intentions to blackout among college students. Alcohol, 86, 35–41. [DOI] [PubMed] [Google Scholar]

RESOURCES