Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jun 1.
Published in final edited form as: Addict Behav. 2021 Feb 4;117:106855. doi: 10.1016/j.addbeh.2021.106855

Exploratory and Confirmatory Factor Analysis of the Parental Rules toward Adolescent Drinking Questionnaire: Two Factors are better than the Original One

Bradley M Trager 1, Ina M Koning 2, Rob Turrisi 3,4
PMCID: PMC8582330  NIHMSID: NIHMS1670135  PMID: 33621921

Abstract

The parental rules toward drinking questionnaire (PRQ; Van der Vorst et al., 2005) assesses strictness toward adolescent drinking situations. The aim of the current study was to address a gap in the literature on the psychometric testing and evaluation of the factor structure of the PRQ. The current sample consisted of Dutch adolescents (N = 2922) who participated in a randomized control trial with three intervention groups (parent, student, and parent + student) and a control. PRQ and frequency of alcohol use (past month and year) were measured at baseline (T1) and 12 months later (T2). Results from Exploratory and Confirmatory Factor Analyses revealed two reliable factors: (a) rules about normative drinking situations and (b) rules about non-normative drinking situations (both αs ≥ 0.88). Regression analyses conducted to examine the prospective effects of the interventions revealed that both parent conditions predicted increases in strictness toward normative drinking situations relative to the control condition, while only the parent + student condition affected the original PRQ (single factor). Further, the normative subscale predicted increases in drinking (past month and year), as did the original PRQ. Significant effects with the normative subscale indicate that rules toward these drinking situations are ones that account for the effects in the original PRQ, and that the original PRQ can mask effects. The results illustrate that the PRQ is multidimensional. The effects of the normative subscale suggest that intervention efforts should focus on preventing drinking situations that parents normally permit their adolescents to engage in.

Keywords: measure, parent, alcohol, rules, adolescence

1. Introduction

Alcohol is one of the most widely used substances by adolescents (Johnston et al., 2019; World Health Organization [WHO], 2018). Drinking during adolescence is a serious public health concern that has substantial behavioral, clinical, social, and economic ramifications (Substance Abuse and Mental Health Service Administration, 2017; U.S. Department of Human Health Services, 2016; WHO, 2018). Parents’ rules toward adolescent drinking are one strategy that can attenuate these risks (Koning et al., 2012; Sharmin et al., 2017; Verdurmen et al., 2014; Yap et al., 2017). Much of our knowledge on the effects of parental rules during adolescence can be attributed to the parental rules toward drinking questionnaire (PRQ; Van der Vorst et al., 2005; 2006). The PRQ is a 10-item measure that was designed to assess the strictness of parents toward adolescent drinking situations (Van der Vorst et al., 2005). Examination of the effects of the PRQ on alcohol use has contributed significantly to our understanding of the role that parent rules can play in preventing adolescent alcohol use (e.g., Koning et al., 2009; 2011; Mares et al., 2012; Percy et al., 2019; Van den Eijnden et al., 2011; Van der Vorst et al., 2007; Wadolowski et al., 2015; 2016). The PRQ has been found to be reliable and valid across several studies (e.g., Creemers et al., 2017; Koning & Vollebergh, 2016; Schelleman-Offermans et al., 2012), and is considered the most comprehensive measures of its kind (Bourdeau et al., 2012). Despite the attention that the PRQ has received in the literature, results from the original exploratory factor analysis used to determine the factor structure of the measure are unpublished (see Van der Vorst et al., 2005). Moreover, there are no published studies to confirm the single-factor structure of this measure.

Since the PRQ first appeared in Van der Vorst et al. (2005), evidence has emerged that supports the notion that parental rules toward alcohol use can be considered multidimensional. For example, Bourdeau and colleagues (2012) identified at least four types of rule settings in a qualitative study (i.e., contingency rules only, no-tolerance only, conflicting rules, call-me rules; Bourdeau et al., 2012). Not surprisingly, the most common types of parental rules reported in that study are those that are intended to limit exposure to alcohol and drinking situations: contingency rules only (~44%) and no tolerance rules only (~42%). Furthermore, less than 10% of the families surveyed in Bourdeau et al. (2012) indicated having no rules pertaining to drinking.

Additional support for a multidimensional model of parental rules stems from research on parental approval toward alcohol use, which has revealed drinking situations that parents of adolescents typically permit children to engage in (i.e., normative drinking situations). Normative situations include drinking on special occasions, during family meals, and with friends (Bourdeau et al., 2012; Livingston et al., 2010; McMorris et al., 2011). This set of situations illustrates the extent that parents typically prefer to limit adolescent alcohol use. Limiting alcohol use to normative drinking situations suggests parents believe relaxing rules prohibiting those situations may be beneficial to their offspring. We can assume this because parents are more likely to adopt strategies they believe will benefit their offspring’s well-being (see Geary & Flinn, 2001, for a review on human parental investment). The presence of these normative drinking situations that parents believe will promote well-being also suggests there are drinking situations that most parents would like to prohibit exposure to as well (i.e., non-normative drinking situations). Drinking situations can be classified as non-normative if they include patterns inconsistent with parents’ beliefs about safe and responsible drinking (e.g., in excess or frequently, drinking alone).

PRQ items reflect a variety of drinking situations (see Table 1), several of which are likely to map onto the two proposed dimensions: (a) normative (e.g., I am allowed to drink one glass of alcohol at home when my father or mother is at home) and (b) non-normative (e.g., I am allowed to drink multiple glasses of alcohol at home when my father and mother are absent). Thus, the aim of the current study was to explore whether the PRQ consisted of these two dimensions. In addition, examination of the factor structure of the PRQ here will address the dearth of research on the psychometric properties of the PRQ and on the measurement of parent alcohol rules (see Bourdeau et al., 2012).

Table 1.

Demographics and descriptive information for the sample

Total (N=2922) Sample A (n=1429) Sample B (n=1493)
T1 N (%) N (%) N (%) x2 df p
Sex
Girls 1406 (48.1%) 706 (49.4%) 700 (46.9%) 1.86 1 0.17
Boys 1516 (51.9%) 723 (50.6%) 793 (53.1%)
Intervention Condition
Parent 680 (23.3%) 331 (23.2%) 349 (23.4%) 3.82 3 0.28
Student 790 (27.0%) 406 (28.4%) 384 (25.7%)
Parent + Student 650 (22.3%) 319 (22.3%) 331 (22.2%)
Control 802 (27.4%) 373 (26.1%) 429 (28.7%)
T1 M(SD) M(SD) M(SD) t df p
Age 14.33 (0.49) 14.57 (0.49) 14.56 (0.49) −0.74 2920 0.46
PRQ
Original 40.65 (9.02) 40.47 (9.11) 40.83 (8.93) 1.09 2916 0.28
Normative 10.71 (3.18) 10.63 (3.18) 10.78 (3.18) 1.29 2916 0.20
Non-Normative 17.40 (3.65) 17.33 (3.71) 17.47 (3.56) 1.07 2916 0.28
Alcohol Use
Month 1.54 (2.93) 1.54 (2.89) 1.54 (2.96) 0.35 2920 0.97
Year 3.43 (4.27) 3.40 (4.22) 3.46 (4.32) 0.41 2920 0.68
Open in a new tab

1.1. Current Study

The current study explored the factor structure of the PRQ in a sample of Dutch adolescents who participated in a well-documented, cluster-randomized control trial (RCT) (see Koning et al., 2009; 2011; 2012; 2014). This study’s large sample size permitted us to divide the sample so we could conduct Exploratory and Confirmatory Factory Analysis (EFA and CFA) on independent groups of adolescents; the longitudinal design permitted us to examine test-retest reliability and evaluate the prospective effects of the intervention on the original PRQ and newly proposed PRQ subscales, as well the effects of the PRQ on alcohol use (past month and year). Examination of the prospective effects will help to determine the validity of any newly identified PRQ subscales, as well as reveal potential insights that the original PRQ does not. For example, intervention conditions (parent only intervention condition) that should have influenced the original PRQ but did not (Koning et al., 2009), may be significant with PRQ subscales. This is plausible given that combining all items from a multidimensional measure into a single scale can mask effects (e.g., Mershon & Gershon, 1988). Thus, PRQ subscales identified and examined here may provide additional insights that can further inform parent-based intervention (PBI) efforts.

2. Method

2.1. Design and Procedures

Participants were adolescents from 19 high schools in the Netherlands whose parents consented to take part in a clinical trial designed to assess the prospective effects of two types of alcohol prevention programs: student and/or parent component (see Koning et al., 2009, 2011, 2012, 2013, 2014). Schools were randomized to one of four conditions: (a) parent intervention; (b) student intervention; (c) parent + student intervention; and (d) control. Digital surveys were administered by trained research assistants to the students at four time points: September/October 2006, June/July 2007, June/July 2008, and June/July 2009. To assess mid- to later adolescence, data from the latter two time points were examined in this study (N=2922).

2.2. Participants

The overall sample for this study consisted of adolescent participants in the aforementioned RCT who completed questionnaires in 2008 (T1: N=2922; Mage=14.33, SDage=0.49) and 2009 (T2: n=2609; Mage=15.57, SDage=0.49). For the purpose of this study, the sample was then randomized into two groups (Sample A and B). Descriptive information and comparisons of these groups on key variables can be found in Table 1.

2.3. Measures

2.3.1. Parental Rules Toward Alcohol Use (T1 and T2)

Participants were asked to complete the 10-item measure on parental rules toward alcohol use (PRQ; Van der Vorst et al., 2005; 2006). Adolescent reports were used, since the way they experience parenting is a better predictor of their behavior than are parent reports on their own parenting (e.g., Chassin et al., 2005; Koning et al., 2010). All 10 PRQ items are listed in Table 1. Items were rated on a 5-point scale (1 = always, 2 = often, 3 = occasionally, 4 = seldom, 5 = never). Items were summed to create a composite variable for original PRQ at T1 and T2 (both αs > .90). Higher scores on this variable were indicative of stricter rule setting.

2.3.2. Frequency of Alcohol Use (T1 and T2)

Participants were asked to report the number of drinking occasions where they consumed at least one alcoholic beverage during the last month and year (O’Malley et al, 1983). Response options for both the month and year items were on a 14-point scale (0 = zero, 1 = 1 … 10 = 10, 11 = 11 to 19, 12 = 20 to 39, 14 = 40 or more).

2.4. Analytic Strategy

2.4.1. Preliminary Analyses

Prior to conducting analyses, the select random sample of cases function in SPSS version 26 was used to generate two independent samples (A and B) that were approximately equal in size (see Table 1). (Differences in sample size are a function of the SPSS randomization algorithm). Chi-square and independent samples t-tests were conducted to determine if Sample A and B differed by intervention condition, sex, and baseline age, alcohol use, and original PRQ.

2.4.2. Exploratory and Confirmatory Factor Analysis (Sample A)

EFA with varimax rotation was conducted in MPlus version 7 on Sample A to determine the initial factor structure of the PRQ items at T1. A combination of indicators was used to determine the number of factors to retain: (a) eigenvalues ⪆ 1 (Cliff, 1988), (b) parallel analysis (Horn, 1965), and (c) interpretability of factors (Gorsuch, 1983; Wood et al., 2015). Final factor structure was then determined using the following combination of criteria: (a) retain items with factor loadings ≥ 0.40; (b) drop all items with cross-loadings ≥ 0.30; (c) retain factors that had ≥ 3 items; and (d) internal consistency of multi-item were ≥ 0.70.

2.4.3. Confirmatory Factor Analysis

CFAs were then conducted on Sample B to verify the results of the EFA (multi-factor solution) at T1 and T2. Verification of the multi-factor solution was based on the interpretability of the factors and a combination of fit indices (i.e., CFI/TLI ≥ .90, RMSEA/SRMR < .10; Hu & Bentler, 1999; Kline, 2005). CFAs were also conducted on the original 1-factor solution at T1 and T2 to see if the fit indices from those models were comparatively better or worse than those for the new multi-factor solution.

2.4.4. Test-Retest Reliability

Intraclass correlations (ICC) were calculated to examine test-retest reliability for new PRQ subscales and the original PRQ (i.e., ICC between Factor 1 at T1 and Factor 1 at T2). Higher ICCs suggest greater test-retest reliability.

2.4.5. Prospective Effects of the Intervention on the Original PRQ and New PRQ Subscales

The following series of regression models were conducted to examine intervention effects on (a) the original PRQ and (b) new PRQ subscales. Dummy variables for the three intervention conditions (Parent, Student, and Parent + Student) were created and included in all of the regression models to assess the effects of each condition relative to the control (reference group). In the first model, T2 original PRQ was regressed onto the three intervention dummy variables, baseline original PRQ, alcohol use, age, and sex (male = 1, female = 0). Models were then conducted with new PRQ subscales replacing the original measure (e.g., PRQ Factor 1 at T2 was regressed onto the intervention dummy variables, baseline PRQ Factor 1 and Factor 2, alcohol use and age, and sex). The number of regression models conducted here depended on the number of factors identified.

2.4.6. Prospective Effects of the Original PRQ and New PRQ Subscales on Alcohol Use

Next, four regression models were conducted to examine the effects of the original PRQ and the new PRQ subscales on alcohol use (past month and year). These models were similar to those described above, with alcohol use (month and year) replacing original PRQ and PRQ subscales as the outcomes (i.e., T2 alcohol use past month/past year was regressed onto the intervention dummy variables, baseline original PRQ/PRQ subscales, alcohol use past month/year and age, and sex).

Bootstrapped asymmetrical 95% confidence intervals (CIs) were calculated (N=2000) for all of the aforementioned regression models. Results were considered to be not statistically significant (p>0.05) if the 95% CIs included zero.

3. Results

3.1. Preliminary Analyses

T1 descriptive information for Samples A and B, along with comparisons of the two groups on all relevant study variables, are illustrated in Table 1. Comparative analyses revealed no significant differences between Samples A and B on key variables of interest (e.g., PRQ, alcohol use). This suggests that the randomization process generated comparable samples.

3.2. Exploratory Factor Analysis (Sample A)

Results from the parallel analysis revealed that there may be as many as five factors. Factor loadings for the first three factor solutions are provided in Table 2. The 2-factor solution was the only multi-factor solution that would still include 3 items per factor after dropping items with cross-loadings > 0.30. Examination of the 2-factor solution revealed three items that needed to be dropped due to cross-loadings exceeding the criteria of 0.30 (see Table 2). A second EFA conducted on the 7 remaining items revealed that the 2-factor solution should be retained based on the following criteria: factor loadings were all > 0.39, cross-loadings were all < 0.30, at least 3 items per factor (see Table 3). Internal consistency for Factor 1 (α=0.93; 4 items) and Factor 2 (α=0.88; 3 items) were also well above the specified threshold for retaining factors (α ≥ 0.70).

Table 2.

Factor loadings for the first three factor solutions from EFA

Sample A
T1 - EFA (10-items)
| 1-Factor Solution | | 2-Factor Solution | | 3-Factor Solution |
Item Factor 1 Factor 1 Factor 2 Factor 1 Factor 2 Factor 3
1 I am allowed to drink one glass of alcohol at home when my father or mother is at home. 0.68 0.08 0.69 0.87 0.05 −0.05
2 I am allowed to drink one glass of alcohol at home when my father and mother are absent. 0.86 0.68 0.26 0.27 0.59 0.15
3 I am allowed to drink multiple glasses of alcohol at home when my father or mother is at home. 0.82 0.49 0.41 0.53 0.41 0.06
4 I am allowed to drink multiple glasses of alcohol at home when my father and mother are absent. 0.87 0.99 −0.03 0.03 0.92 0.08
5 I am allowed to drink as much alcohol as I want outside of the house. 0.84 0.68 0.23 0.03 0.51 0.41
6 I am allowed to drink alcohol at a party with friends. 0.78 0.08 0.82 0.45 −0.01 0.49
7 I am allowed to come home tipsy. 0.86 0.52 0.40 0.06 0.26 0.68
8 I am allowed to become tipsy when going out with friends. 0.84 0.37 0.54 0.05 0.07 0.87
9 I am allowed to drink alcohol on the weekend. 0.79 0.02 0.89 0.62 −0.02 0.37
10 I am allowed to drink alcohol on weekdays. 0.82 0.64 0.25 0.22 0.54 0.19
Open in a new tab

Table 3.

Results from two factor EFA (7 items) and from CFA models

| Sample A | | Sample B |
Items T1 - EFA (7 items) T2 - CFA (7 items) T2 - CFA (10 items) T1 - CFA (10 items) T1 - CFA (7 items) T2 - CFA (7 items)
Non-Normative Rules 2 Factor Solution 2 Factor Solution 1 Factor Solution 1 Factor Solution 2 Factor Solution 2 Factor Solution
2 I am allowed to drink one glass of alcohol at home when my father and mother are absent. 0.69 0.88 0.81 0.83 0.89 0.88
4 I am allowed to drink multiple glasses of alcohol at home when my father and mother are absent. 0.97 0.89 0.81 0.82 0.91 0.90
5 I am allowed to drink as much alcohol as I want outside of the house. 0.68 0.77 0.79 0.77 0.84 0.78
10 I am allowed to drink alcohol on weekdays. 0.62 0.72 0.70 0.71 0.79 0.74
Normative Rules
1 I am allowed to drink one glass of alcohol at home when my father or mother is at home. 0.66 0.74 0.72 0.70 0.77 0.75
6 I am allowed to drink alcohol at a party with friends. 0.74 0.85 0.83 0.78 0.84 0.89
9 I am allowed to drink alcohol on the weekend. 0.95 0.90 0.83 0.81 0.91 0.92
Dropped Items
3 I am allowed to drink multiple glasses of alcohol at home when my father or mother is at home. --- --- 0.80 0.80 --- ---
7 I am allowed to come home tipsy. --- --- 0.85 0.85 --- ---
8 I am allowed to become tipsy when going out with friends. --- --- 0.85 0.83 --- ---
CFI --- .94 .82 .83 .95 .94
TLI --- .90 .77 .78 .93 .90
SRMR --- .04 .06 .06 .03 .04
RMSEA, 90% CI [LB, UB] --- .15 [.14, .16] .21 [.20, .22] .22 [.21, .22] .14 [.13, .15] .15 [.14, .17]
𝛘2(df) Test of Model Fit --- 419.06(13) 2246.96(35) 2415.01(35) 388.45(13) 463.38(13)
Open in a new tab

Interpretation of these two correlated factors (r=0.61) is consistent with the expectations that the PRQ items, at the very least, reflect drinking situations that are non-normative (e.g., allowed to drink multiple glasses of alcohol at home when my father and mother are absent; Factor 1) and normative (e.g., allowed to drink one glass of alcohol at home when my father or mother is at home; Factor 2) for parents to permit during adolescence (see Table 2). Taken together, the 2-factor solution with 7 items was selected to be examined in the subsequent CFA models alongside the original PRQ. Findings were similar when analyses were conducted on the control group only.

3.3. Confirmatory Factor Analysis

Using Sample B, CFAs were first conducted to test the fit of the 2-factor, 7-item PRQ at T1 and T2 (see Table 3). Fit at both T1 and T2 were judged to be good based on CFI/TLI ≥ 0.90 and SRMR < .10. CFAs were then conducted to examine model fit for the original 10-item PRQ with one factor at T1 and T2. As illustrated in Table 3, none of the fit indices reached the minimum thresholds in these models. CFAs conducted using Sample A’s data from T2 also revealed a similar pattern of results for the both the 2-factor PRQ with 7 items and 1-factor PRQ with 10 items (see Table 3). Taken together, results suggest that the new 2-factor 7-item PRQ had better fit than the original 1-factor 10-item PRQ.

Samples A and B did not differ on any of the measured variables and were collapsed for all remaining analyses (see Table 1).

3.5. Test-Retest Reliability

ICCs calculated for the original PRQ (0.56) and non-normative rules (0.57) were judged to be fair/moderate, and was good/moderate for normative rules (0.61) (Cicchetti, 1994; Koo & Li, 2016).

3.6. Prospective Effects of the Intervention on the Original PRQ and New PRQ Subscales

3.6.1. Interventions Effects on Original PRQ

Scores on the original PRQ increased at T2 (indicating stricter rules) in the Parent + Student condition compared to the control condition (b=1.19, SE=0.50, 95% CI [0.21, 2.17]; see Table 4).

Table 4.

Results from regression models

Outcome (T2)
Predictor (T1) Original PRQ Alcohol Use Past Month Alcohol Use Past Year
PRQ Total b(SE) B 95% CI b(SE) B 95% CI b(SE) B 95% CI
Original 10-item 0.38(0.03) 0.35 [0.32, 0.43] −0.06(0.01) −0.14 [−0.08, −0.04] −0.07(0.01) −0.12 [−0.09, −0.04]
Intervention Condition
Parent Intervention 0.87(0.49) 0.04 [−0.09, 1.85] − 0.08(0.19) − 0.01 [−0.45, 0.31] −0.56(0.23) −0.05 [−1.01, −0.11]
Student Intervention 0.34(0.41) 0.02 [−0.44, 1.17] − 0.22(0.17) − 0.03 [−0.55, 0.14] − 0.31(0.21) − 0.03 [−0.73, 0.11]
Parent + Student 1.19(0.50) 0.05 [0.21, 2.17] −0.73(0.19) −0.08 [−1.09, −0.36] −1.03(0.24) −0.09 [−1.50, −0.57]
Control (ref)
Covariates
Sex 0.01(0.33) 0.0004 [−0.65, 0.68] 0.53(0.13) 0.07 [0.29, 0.79] 0.16(0.16) 0.02 [−0.15, 0.46]
Age −1.07(0.35) −0.06 [−1.77, −0.39] 0.52(0.14) 0.07 [0.23, 0.79] 0.41(0.17) 0.04 [0.08, 0.74]
Alcohol Use - Past Month −0.26(0.09) −0.08 [−0.43, −0.11] 0.47(0.04) 0.36 [0.40, 0.55] --- --- ---
Alcohol Use - Past Year --- --- --- --- --- --- 0.59(0.02) 0.50 [0.54, 0.63]
Outcome (T2)
Predictor (T1) Normative Non-Normative Alcohol Use Past Month Alcohol Use Past Year
PRQ Subscales b(SE) B 95% CI b(SE) B 95% CI b(SE) B 95% CI b(SE) B 95% CI
Normative 0.47(0.03) 0.46 [0.42, 0.53] 0.15{0.03) 0.13 [0.09, 0.22] −0.14(0.03) −0.12 [−0.20, −0.07] −0.21(0.04) −0.13 [−0.29, −0.13]
Non-Normative −0.11(0.03) −0.12 [−0.15, −0.05] 0.24(0.04) 0.23 [0.18, 0.32] − 0.03(0.03) − 0.03 [−0.09, 0.03] 0.02(0.04) 0.01 [−0.05, 0.08]
Intervention Condition
Parent Intervention 0.77(0.16) 0.10 [0.45, 1.08] − 0.05(0.20) − 0.01 [−0.44, 0.34] − 0.04(0.19) − 0.01 [−0.43, 0.32] −0.49(0.23) −0.04 [−0.92, −0.03]
Student Intervention 0.04(0.14) 0.01 [−0.25, 0.31] 0.06(0.17) 0.01 [−0.26, 0.38] - - [−0.54, 0.13] - - [−0.67, 0.13]
Parent + Student 0.77(0.16) 0.10 [0.45, 1.09] 0.06(0.20) 0.01 [-0.34, 0.46] −0.69(0.18) −0.08 [−1.07, −0.33] −0.95(0.24) −0.08 [−1.41, −0.47]
Control (ref)
Covariates
Sex 0.09(0.12) 0.01 [−0.14, 0.31] −0.39(0.14) −2.73 [−0.67, −0.11] 0.58(0.13) 0.08 [0.34, 0.84] 0.28(0.16) 0.03 [−0.04, 0.59]
Age −0.55(0.11) −0.08 [−0.78, −0.34] −0.30(0.14) −2.15 [−0.58, −0.02] 0.51(0.14) 0.06 [0.22, 0.78] 0.41(0.17) 0.04 [0.09, 0.75]
Alcohol Use Past Month −0.10(0.03) −0.09 [−0.15, −0.05] −0.12(0.03) −4.16 [−0.18, −0.05] 0.49(0.04) 0.37 [0.41, 0.56] --- --- ---
Alcohol Use Past Year --- --- --- --- --- --- --- --- --- 0.59(0.02) 0.50 [0.54, 0.63]
Open in a new tab

Notes:b = unstandardized coefficients; B = standardized coefficients; SE = standard errors; results were significant at p < .05 (bolded) if 95% confidence intervals (CIs) did not include zero; PRQ total = original parental rules questionnaire (10 items summed); Sex (males = 1; females = 0).

3.6.2. Interventions Effects on the Normative Subscale

Relative to the control condition, scores on the normative subscale increased at T2 (stricter) in both the Parent (b=0.77, SE=0.16, 95% CI [0.45, 1.08]) and the Parent + Student conditions relative to controls (b=0.77, SE=0.16, 95% CI [0.45, 1.09]; see Table 4).

3.6.3. Interventions Effects on the Non-Normative Subscale

As shown in Table 4, participants in the treatment conditions did not significantly differ from the controls on the non-normative subscale at T2.

3.7. Prospective Effects of the Original PRQ and New PRQ Subscales

3.7.1. Effects of the Original PRQ on Past Month and Year Drinking

Results for the model that assessed past month drinking as an outcome revealed that higher scores on T1 original PRQ (stricter) predicted a significant decrease in drinking at T2 (b=−0.06, SE=0.01, 95% CI [−0.08, −0.04]). A similar pattern of effects emerged when past year drinking was assessed as the outcome; higher scores on original PRQ at T1 predicted a significant decrease in drinking at T2 (b=−0.07, SE=0.01, 95% CI [−0.09, −0.04]).

3.7.1. Effects of PRQ subscales on Past Month and Year Drinking

Results for the model that examined past month drinking as an outcome revealed that higher scores on T1 normative PRQ (stricter) predicted a significant decrease in drinking at T2 (b=−0.14, SE=0.03, 95% CI [−0.20, −0.07]), but the effect was not significant for the non-normative subscale (b=−0.03, SE=0.03, 95% CI [−0.09, 0.03]). A similar pattern of effects emerged when past year drinking was assessed as the outcome; higher scores on normative PRQ at T1 predicted a significant decrease in drinking at T2 (b=−0.21, SE=0.04, 95% CI [−0.29, −0.13]), but the effect was not significant for the non-normative subscale (b=0.02, SE=0.04, 95% CI [−0.05, 0.08]).

4. Discussion

There has been an increase in research on the influence of parent alcohol rules toward underage drinking over the last decade. These studies have revealed that generally, stricter rules against underage drinking can be protective against subsequent alcohol use (e.g., Mares et al., 2012; Percy et al., 2019; Sharmin, et al., 2017; Van der Vorst et al., 2005; Yu, 2003). To extend this research, the current study explored the factor structure of the PRQ to determine whether or not the parent rules toward adolescent drinking construct is multidimensional (e.g., Bourdeau et al., 2012). Results from the EFA and CFA models revealed two related and reliable subscales: (a) rules toward normative and (b) rules toward non-normative drinking situations. As expected, the normative subscale consisted of rules toward drinking situations that map onto those parents typically relax during adolescence to socialize safe and responsible drinking (i.e., on special occasions, during family meals, with friends; Bourdeau et al., 2012; Livingston et al., 2010; McMorris et al., 2011). Conversely, drinking situations on the non-normative subscale can be interpreted as situations that parents would not permit as they would be counterproductive to socializing safe or responsible alcohol use (e.g., drinking in excess or frequently, alone). These findings are consistent with the notion that the relaxation of rules during adolescence (e.g., Mares et al., 2012) follows a predictable pattern of limiting drinking to situations parents are known to erroneously perceive as either beneficial or low risk (e.g., Livingston et al., 2010).

The normative and non-normative subscales were found to have good internal consistency (Nunally, 1994), with αs ≥ 0.88 at baseline and follow-up. Test-retest reliability of the subscales and original PRQ measured one year apart was somewhat lower than expected, but acceptable given that scores on this measure should change over time as parents relax their rules as adolescents age (e.g., Percy et al., 2019). Correlation between the normative and non-normative factors was somewhat high, but comparable to the strength of the associations between subscales in other well-established multidimensional measures from the alcohol literature (e.g., drinking motives; young adult alcohol consequences questionnaire; Grant et al., 2007; Read et al., 2007).

Findings from the CFAs revealed that the 2-factor solution with seven items performed better than the 1-factor solution with all 10 items (original PRQ) with respect to model fit. The fit for the 2-factor and 1-factor solutions were judged to be adequate and poor, respectively. Dropping items from the PRQ might account for some improvements in model fit that led to the 2-factor model outperforming the 1-factor model. Items that were dropped reflected behaviors that can be construed as non-normative drinking situations (i.e., drinking multiple alcohol beverages while parents are at home; coming home tipsy; becoming tipsy while out with friends). Results from the prospective analyses on the effects of the original PRQ and the new subscales on alcohol use suggest that rules toward the three dropped drinking situations, two of which reflect drunkenness, may not contribute as much to our understanding of parent rules for this young age group. A likely reason for this is that rules toward situations that involve drunkenness might be more relevant later on when parents have a need to prohibit those behaviors (e.g., a parent does not want their child who lives at home and is legally permitted to drink to come home drunk). Given this, researchers should refrain from omitting PRQ items that did not load onto the new PRQ factors in future studies, as some of those items could discriminate between individuals or groups not represented in the current sample.

Examination of the prospective effects of the original PRQ and subscales also revealed that the normative drinking situations were the ones that mattered the most, and are also likely to be driving the effects of the original PRQ on drinking. The notion that stricter rules toward normative drinking situations plays a more dominant role in preventing alcohol use is consistent with previous studies that demonstrate parental permissiveness toward similar situations during adolescence is more strongly a risk factor for onset of drinking than subsequent level of alcohol use (e.g., Koning et al., 2010).

Related to prevention, the current findings replicated prior research that revealed effects of the parent + student intervention on the original PRQ (see Koning et al., 2009; 2011), and extend this work by demonstrating effects that were only observed with the subscales. Specifically, the current findings illustrate that the effects of the combined PBI on parental rules were only significant on the normative subscale. This is important given that the normative drinking situations were also the ones that accounted for increases in alcohol use. The PRQ subscales also revealed effects for the parent intervention (without student intervention) on parental rules that were not significant with the original PRQ. Unidimensional measures can conceal clinically relevant effects (e.g., Mershon & Gershon, 1988), which is what appears to have occurred here. Taken together, these findings, coupled with the psychometric evidence and results on the effects of PRQ subscales on alcohol use presented above, supports the 2-factor PRQ over the original 1-factor solution. However, we recommend that researchers who measured PRQ in their prior studies test both the unidimensional 10-item scale and the new PRQ subscales to better to understand the generalizability of the current findings.

4.1. Implications and Future Directions

Findings from this study suggest that researchers interested in assessing parental rules toward adolescent drinking using the PRQ should evaluate the normative and non-normative subscales rather than combining all items into a single measure. Results indicate that totaling items can conceal important effects and reduce specific insights about parental rule setting that can be used to inform intervention efforts (e.g., target normative rules). Focusing on a more relevant subset of alcohol-specific rules in PBIs may help improve intervention effects. Targeting rules that reflect behaviors and situations adolescents are more likely to encounter in earlier drinking phases (i.e., normative) rather than ones they are less likely to encounter (i.e., non-normative) or ambiguous ones (e.g., zero-tolerance policy) may improve the efficacy of PBIs as they may better match parents’ needs. Moreover, the notion that different types of rules will be more relevant to preventing or reducing alcohol use in some groups and not others is consistent with previous findings in the literature (e.g., zero-tolerance is a less effective parenting strategy than harm-reduction for reducing alcohol use in drinkers relative to non-drinkers; Napper, 2019). Developing computer-based PBIs that can provide tailored feedback, such as the types of rules to stress to offspring based on their individual characteristics (drinker vs. nondrinker), may be one way to accomplish this. However, it is first important that studies be conducted to identify the types of rules that are optimal for preventing or reducing alcohol use in certain populations (e.g., drinkers and nondrinkers; Napper, 2019).

4.2. Limitations

Despite the many strengths of this study (e.g., large sample size, longitudinal design) there are a few limitations that should be considered. As is the case with most psychometric studies, the generalizability of the current findings may be limited by sample characteristics (Morgado et al., 2018). This sample consisted of adolescents living in the Netherlands. Parenting behaviors and normative beliefs toward adolescent alcohol use can vary between countries, societies, and cultures (e.g., individualism vs. collectivism; Huiberts et al., 2006; Kuntsche et al., 2017; MacKinnon et al., 2017; Wissink et al., 2006). Therefore, future research is required to investigate how the factor structure of the PRQ identified in this sample may generalize to other samples outside of the Netherlands. Future studies should also consider whether the PRQ subscales replicate in older age groups, such as underage adults (ages 18–20 in the U.S.). Changes in lifestyle and residence during this period (e.g., going to college) can engender new drinking situations for which rules can be assigned. Thus, new measures that include drinking situations and behaviors suited for underage adults may be required to identify the normative and non-normative dimensions (e.g., Trager, 2020).

Another limitation of this study was that the data used here were not collected for the purpose of conducting psychometric testing. As a result, there were no measures included that would have allowed for a more thorough examination of validity. For example, examining measures of parental permissiveness toward underage drinking (e.g., Abar et al., 2009; Wood et al., 2004) would have allowed for a more thorough investigation of convergent validity. Given this, future evaluations of the psychometric properties of this measure should consider examining the aforementioned constructs to further establish the validity of the PRQ and its subscales.

4.3. Conclusion

To summarize, this is the first study to provide an in-depth evaluation of the factor structure of the PRQ (Van der Vorst et al., 2005). Findings revealed that the PRQ consisted of two reliable subscales: parental rules toward normative drinking situations and non-normative drinking situations. Results illustrate that the effects of the original PRQ on alcohol use can be driven by rules toward the normative drinking situations, and that combining all of the PRQ items into a single measure can mask effects. Thus, findings suggest that PRQ subscales presented here should be evaluated prior to totaling items into a single measure. Furthermore, this study supports targeting parental rules toward normative drinking situations as an effective strategy for alcohol prevention with adolescents in earlier phases of drinking.

Highlights.

  • Psychometric evaluation of the Parental Rules toward Drinking Questionnaire (PRQ)

  • EFA and CFA revealed two reliable subscales: normative and non-normative rules

  • PRQ subscales were positively correlated with adolescent alcohol use

  • PRQ subscales revealed effects on drinking not shown found with total PRQ

  • Implications for prevention efforts are to prohibit normative drinking situations

Acknowledgements:

Support for this research was provided by the Dutch Health Care Research Organization (6220, 0021) and NIAAA (R34 AA026422 02). ). The Dutch Health Care Research Organization and NIAAA had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

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.

Author CRediT

Date were provided from a larger project that was designed and conducted by Ina Koning. Bradley Trager conducted the statistical analysis, and wrote the first draft of the manuscript. Ina Koning and Rob Turrisi revised subsequent versions of the manuscript. All authors contributed to and have approved the final manuscript.

References

  1. Abar C, Abar B, & Turrisi R. (2009). The impact of parental modeling and permissibility on alcohol use and experienced negative drinking consequences in college. Addictive Behaviors, 34(6–7), 542–547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bourdeau B, Miller B, Vanya M, Duke M, & Ames G. (2012). Defining alcohol-specific rules among parents of older adolescents: moving beyond no tolerance. Journal of Family Communication, 12(2), 111–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chassin L, Presson CC, Rose J, Sherman SJ, Davis MJ, & Gonzalez JL (2005). Parenting style and smoking-specific parenting practices as predictors of adolescent smoking onset. Journal of Pediatric Psychology, 30(4), 333–344. [DOI] [PubMed] [Google Scholar]
  4. Creemers HE, Spanakis P, Delforterie MJ, & Huizink AC (2017). Alcohol use of immigrant youths in The Netherlands: The roles of parents and peers across different ethnic backgrounds. Drug and Alcohol Review, 36(6), 761–768. [DOI] [PubMed] [Google Scholar]
  5. Cicchetti Domenic V. (1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment, 6(4), 284–290. [Google Scholar]
  6. Cliff N. (1988). The eigenvalues-greater-than-one rule and the reliability of components. Psychological Bulletin, 103(2), 276. [Google Scholar]
  7. Geary DC, & Flinn MV (2001). Evolution of human parental behavior and the human family. Parenting, 1(1–2), 5–61. [Google Scholar]
  8. Gorsuch RL (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. [Google Scholar]
  9. Grant VV, Stewart SH, O’Connor RM, Blackwell E, & Conrod PJ (2007). Psychometric evaluation of the five-factor Modified Drinking Motives Questionnaire—Revised in undergraduates. Addictive Behaviors, 32(11), 2611–2632. [DOI] [PubMed] [Google Scholar]
  10. Horn JL (1965). A rationale and test for the number of factors in factor analysis. Psychometrika, 30(2), 179–185. [DOI] [PubMed] [Google Scholar]
  11. Hu L, & Bentler P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. [Google Scholar]
  12. Huiberts A, Oosterwegel A, Vandervalk I, Vollebergh W, & Meeus W. (2006). Connectedness with parents and behavioural autonomy among Dutch and Moroccan adolescents. Ethnic and Racial Studies, 29(2), 315–330. [Google Scholar]
  13. Johnston LD, Miech RA, O’Malley PM, Bachman JG, Schulenberg JE, & Patrick ME (2019). Monitoring the Future National Survey Results on Drug Use, 1975–2018: Overview, Key Findings on Adolescent Drug Use. Institute for Social Research. [Google Scholar]
  14. Kline R. (2005). Principles and practices of structural equation modeling (2n ed.). New York: Guilford Press. [Google Scholar]
  15. Koning IM, Van den Eijnden RJ, Verdurmen JE, Engels RC, & Vollebergh WA (2011). Long-term effects of a parent and student intervention on alcohol use in adolescents: a cluster randomized controlled trial. American Journal of Preventive Medicine, 40(5), 541–547. [DOI] [PubMed] [Google Scholar]
  16. Koning IM, Van den Eijnden RJ, Verdurmen JE, Engels RC, & Vollebergh WA (2012). Developmental alcohol-specific parenting profiles in adolescence and their relationships with adolescents’ alcohol use. Journal of Youth and Adolescence, 41(11), 1502–1511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Koning IM, Van den Eijnden RJ, Verdurmen JE, Engels RC, & Vollebergh WA (2013). A cluster randomized trial on the effects of a parent and student intervention on alcohol use in adolescents four years after baseline; no evidence of catching-up behavior. Addictive Behaviors, 38(4), 2032–2039. [DOI] [PubMed] [Google Scholar]
  18. Koning IM, Van den Eijnden RJ, & Vollebergh WA (2014). Alcohol-specific parenting, adolescents’ self-control, and alcohol use: A moderated mediation model. Journal of Studies on Alcohol and Drugs, 75(1), 16–23. [DOI] [PubMed] [Google Scholar]
  19. Koning IM, & Vollebergh WA (2016). Secondary Effects of an Alcohol Prevention Program Targeting Students and/or Parents. Journal of Substance Abuse Treatment, 67, 55–60. [DOI] [PubMed] [Google Scholar]
  20. Koning IM, Vollebergh WA, Smit F, Verdurmen JE, Van Den Eijnden RJ, Ter Bogt TF, … & Engels RC (2009). Preventing heavy alcohol use in adolescents (PAS): cluster randomized trial of a parent and student intervention offered separately and simultaneously. Addiction, 104(10), 1669–1678. [DOI] [PubMed] [Google Scholar]
  21. Koo TK, & Li MY (2016). A guideline of selecting and reporting intraclass correlation coefficients for reliability research. Journal of Chiropractic Medicine, 15(2), 155–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kuntsche E, Sznitman SR, & Kuntsche S. (2017). Alcohol and other substance use in a cross-cultural perspective. Drug and Alcohol Review, 36(6), 717–720. [DOI] [PubMed] [Google Scholar]
  23. Livingston JA, Testa M, Hoffman JH, & Windle M. (2010). Can parents prevent heavy episodic drinking by allowing teens to drink at home?. Addictive Behaviors, 35(12), 1105–1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mackinnon SP, Couture ME, Cooper ML, Kuntsche E, O’Connor RM, Stewart SH, & DRINC Team. (2017). Cross-cultural comparisons of drinking motives in 10 countries: Data from the DRINC project. Drug and Alcohol Review, 36(6), 721–730. [DOI] [PubMed] [Google Scholar]
  25. Mares SH, Lichtwarck-Aschoff A, Burk WJ, Van der Vorst H, & Engels RC (2012). Parental alcohol-specific rules and alcohol use from early adolescence to young adulthood. Journal of Child Psychology and Psychiatry, 53(7), 798–805. [DOI] [PubMed] [Google Scholar]
  26. Mershon B, & Gorsuch RL (1988). Number of factors in the personality sphere: Does increase in factors increase predictability of real-life criteria? Journal of Personality and Social Psychology, 55, 675–680. [Google Scholar]
  27. McMorris BJ, Catalano RF, Kim MJ, Toumbourou JW, & Hemphill SA (2011). Influence of family factors and supervised alcohol use on adolescent alcohol use and harms: similarities between youth in different alcohol policy contexts. Journal of Studies on Alcohol and Drugs, 72(3), 418–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Morgado FF, Meireles JF, Neves CM, Amaral A, & Ferreira ME (2017). Scale development: ten main limitations and recommendations to improve future research practices. Psicologia: Reflexão e Crítica, 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Napper LE (2019). Harm-reduction and zero-tolerance maternal messages about college alcohol use. Addictive Behaviors, 89, 136–142. [DOI] [PubMed] [Google Scholar]
  30. Nunnally JC (1994). Psychometric theory 3E. Tata McGraw-hill education. [Google Scholar]
  31. O’Malley PM, Bachman JG, & Johnston LD (1983). Reliability and consistency in self-reports of drug use. International Journal of the Addictions, 18(6), 805–824. [DOI] [PubMed] [Google Scholar]
  32. Percy A, Agus A, Cole J, Doherty P, Foxcroft D, Harvey S, … & Sumnall H. (2019). Recanting of previous reports of alcohol consumption within a large-scale clustered randomised control trial. Prevention Science, 20(6), 844–851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Percy A, McKay MT, & Cole JC (2019). Interplay between sensation seeking and parental rules in the emergence of heavy episodic drinking. Journal of Research on Adolescence, 29(4), 814–821. [DOI] [PubMed] [Google Scholar]
  34. Read JP, Merrill JE, Kahler CW, & Strong DR (2007). Predicting functional outcomes among college drinkers: Reliability and predictive validity of the Young Adult Alcohol Consequences Questionnaire. Addictive Behaviors, 32(11), 2597–2610. [DOI] [PubMed] [Google Scholar]
  35. Schelleman-Offermans K, Knibbe RA, Kuntsche E, & Casswell S. (2012). Effects of a natural community intervention intensifying alcohol law enforcement combined with a restrictive alcohol policy on adolescent alcohol use. Journal of Adolescent Health, 51(6), 580–587. [DOI] [PubMed] [Google Scholar]
  36. Sharmin S, Kypri K, Khanam M, Wadolowski M, Bruno R, Attia J, … & Mattick RP (2017). Effects of parental alcohol rules on risky drinking and related problems in adolescence: Systematic review and meta-analysis. Drug and Alcohol Dependence, 178, 243–256. [DOI] [PubMed] [Google Scholar]
  37. Substance Abuse and Mental Health Services Administration [SAMHSA] (2017). Report to Congress on the Prevention and Reduction of Underage Drinking. Rockville, MD: U.S. Department of Health and Human Services. [Google Scholar]
  38. Trager BM (2020). Development and validation of the parental approval toward alcohol questionnaire: An in-depth examination of perceived parental approval in underage adults (Unpublished doctoral dissertation). The Pennsylvania State University, Pennsylvania, United States. [Google Scholar]
  39. U.S. Department of Health and Human Services (2016). The Surgeon General’s Report on Alcohol, Drugs, and Health. Rockville, MD: U.S. Department of Health and Human Services. [Google Scholar]
  40. Van Der Vorst H, Engels RC, Deković M, Meeus W, & Vermulst AA (2007). Alcohol-specific rules, personality and adolescents’ alcohol use: A longitudinal person–environment study. Addiction, 102(7), 1064–1075. [DOI] [PubMed] [Google Scholar]
  41. Van Der Vorst H, Engels RC, Meeus W, & Deković M. (2006). The impact of alcohol-specific rules, parental norms about early drinking and parental alcohol use on adolescents’ drinking behavior. Journal of Child Psychology and Psychiatry, 47(12), 1299–1306. [DOI] [PubMed] [Google Scholar]
  42. Van Der Vorst H, Engels RC, Meeus W, Deković M, & Van Leeuwe J. (2005). The role of alcohol-specific socialization in adolescents’ drinking behaviour. Addiction, 100(10), 1464–1476. [DOI] [PubMed] [Google Scholar]
  43. Van den Eijnden R, Van De Mheen D, Vet R, & Vermulst AD (2011). Alcohol-specific parenting and adolescents’ alcohol-related problems: The interacting role of alcohol availability at home and parental rules. Journal of Studies on Alcohol and Drugs, 72(3), 408–417. [DOI] [PubMed] [Google Scholar]
  44. Verdurmen JE, Koning IM, Vollebergh WA, Van den Eijnden RJ, & Engels RC (2014). Risk moderation of a parent and student preventive alcohol intervention by adolescent and family factors: A cluster randomized trial. Preventive Medicine, 60, 88–94. [DOI] [PubMed] [Google Scholar]
  45. Wadolowski M, Hutchinson D, Bruno R, Aiken A, Clare P, Slade T, … & Mattick RP (2015). Early adolescent alcohol use: are sipping and drinking distinct?. Alcoholism: Clinical and Experimental Research, 39(9), 1805–1813. [DOI] [PubMed] [Google Scholar]
  46. Wadolowski M, Hutchinson D, Bruno R, Aiken A, Najman JM, Kypri K, … & Mattick RP (2016). Parents who supply sips of alcohol in early adolescence: A prospective study of risk factors. Pediatrics, 137(3). [DOI] [PubMed] [Google Scholar]
  47. Wissink IB, Dekovic M, & Meijer AM (2006). Parenting behavior, quality of the parent-adolescent relationship, and adolescent functioning in four ethnic groups. The Journal of Early Adolescence, 26(2), 133–159. [Google Scholar]
  48. Wood MD, Read JP, Mitchell RE, & Brand NH (2004). Do parents still matter? Parent and peer influences on alcohol involvement among recent high school graduates. Psychology of Addictive Behaviors, 18(1), 19. [DOI] [PubMed] [Google Scholar]
  49. Wood ND, Akloubou Gnonhosou DC, & Bowling JW (2015). Combining parallel and exploratory factor analysis in identifying relationship scales in secondary data. Marriage & Family Review, 51(5), 385–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. World Health Organization [WHO] (2018) Global status report on alcohol and health 2018. Geneva: World Health Organization. [Google Scholar]
  51. Yap MB, Cheong TW, Zaravinos-Tsakos F, Lubman DI, & Jorm AF (2017). Modifiable parenting factors associated with adolescent alcohol misuse: A systematic review and meta-analysis of longitudinal studies. Addiction, 112(7), 1142–1162. [DOI] [PubMed] [Google Scholar]
  52. Yu J. (2003). The association between parental alcohol-related behaviors and children’s drinking. Drug and Alcohol Dependence, 69(3), 253–262. [DOI] [PubMed] [Google Scholar]

RESOURCES