How Specific is Alcohol-Specific Self-Control? A Longitudinal Study of the Mediating Role of Alcohol-Specific Self-Control in the Relation between General Self-Control and Adolescent Alcohol Use

doi:10.21203/rs.3.rs-60040/v1

Download PDF

Research article

How Specific is Alcohol-Specific Self-Control? A Longitudinal Study of the Mediating Role of Alcohol-Specific Self-Control in the Relation between General Self-Control and Adolescent Alcohol Use

https://doi.org/10.21203/rs.3.rs-60040/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Although accumulating studies indicate that alcohol-specific self-control can be useful in predicting adolescent alcohol use, little is known about its specificity. This longitudinal study aimed to advance our understanding of domain-specific self-control by examining whether alcohol-specific self-control mediates the effect of general self-control on adolescent alcohol use or has generalizing effects by also mediating the effect of general self-control on other behavior requiring self-control (adolescent digital media use and smoking).

Methods: Data from 906 adolescents aged 11-14 years who were enrolled in the Dutch study Prevention of Alcohol Use in Students were used. Data were collected using online questionnaires at four annual measurements.

Results: Structural equation modelling revealed that higher alcohol-specific self-control fully mediated the effect of higher general self-control on alcohol use. Alcohol-specific self-control did not mediate the effect of higher general self-control on digital media use, but did partially mediate the effect of higher general self-control on smoking.

Conclusions: These results suggest that alcohol-specific self-control is domain-specific, but not necessarily substance-specific. The domain-specificity of alcohol-specific self-control provides evidence for its theoretical relevance for the explanation of adolescent alcohol use. It also suggests leverage points for intervention programs focusing on improving alcohol-specific self-control to reduce adolescent alcohol use.

Trial registration. The trial protocol (NTR649) was approved by the Medical Ethical Committee.

Health Economics & Outcomes Research

Health Policy

adolescents

alcohol use

general self-control

alcohol-specific self-control

domain-specific

One of the most commonly used drugs among adolescents is alcohol. In 2015, 45% of Dutch adolescents aged 12–16 years had consumed alcohol at least once in the past month, 24% had been drunk or tipsy at least once and 18% had been involved in binge drinking (five or more alcoholic drinks in one session). Among 16-year-olds, these percentages are notably higher; respectively 76%, 55% and 44% [1]. Adolescent alcohol use is associated with alcohol problems in adulthood [2] and problem behavior in middle and high school, such as early school leaving, violent and delinquent behavior, sexual risk behavior and co-morbid substance use [3–6]. Additionally, drinking at an early age can have damaging effects on adolescent brain development, especially on learning and attention abilities [7]. Because of these negative outcomes, it is important to prevent and reduce adolescent alcohol use. Since self-control has a direct influence on adolescent alcohol use (e.g.,[8]) and given the possibility to improve this individual characteristic [9], alcohol-specific self-control may be an important factor to target in preventing and reducing adolescent alcohol use.

Self-control is considered individual characteristic that reflects the ability to control and adjust thoughts, emotions and behavioral tendencies, and refrain from acting on them in order to achieve a goal or conform to standards [10, 11]. It includes capacities such as emotion regulation, thought suppression, temptation resistance and behavior modification [10]. A theory that identifies self-control as the key concept in understanding risk behavior among adolescents is the self-control theory by Gottfredson and Hirschi [12]. This theory assumes that individuals with low levels of self-control are more likely to engage in risk behavior, such as alcohol use, because they have a greater likelihood to react to stimuli in the environment that make them overstep their norms [13]. Particularly in adolescence, high self-control can play an important role in preventing engagement in and reducing risk behavior [14, 15].

In the literature a distinction is made between general self-control, which is assumed to affect behaviors across different domains [10, 16], and domain-specific self-control, such as alcohol-specific self-control, which refers to the ability to refrain from drinking alcohol [17–19]. A number of studies suggest that both general self-control and alcohol-specific self-control are related to adolescent alcohol use (e.g., [18, 20, 21]). However, to our knowledge, it has not yet been investigated whether the relationship between general self-control and adolescent alcohol use is mediated by alcohol-specific self-control, and whether this mediating role is domain-specific.

Domain-specific self-control implies that an individual can have high self-control in one domain (e.g., substance use) and at the same time low self-control in another domain (e.g., sexual risk behavior, digital media use; [16]). However, little is known about the extent to which this domain-specific self-control is actually specific for a specific domain. Research suggests that the level of self-control in one domain can be predictive for the level of self-control in another domain [9, 16]. This research thereby implies that distinguishing between general self-control and domain-specific self-control may not be meaningful. Furthermore, alcohol-specific self-control implies that it is specifically related to alcohol use and not to the use of other substances or addictive behaviors. This study examines the specificity of alcohol-specific self-control by investigating whether it mediates the effect of general self-control on adolescent alcohol use or generalizes to also mediate the effect of general self-control on adolescent digital media use and/or smoking. The answer to this question has important implications for the use of alcohol-specific self-control in the explanation of adolescent alcohol use and for the content and efficiency of alcohol-related interventions.

General Self-Control and Adolescent Alcohol Use

A wide range of empirical studies examined the relationship between general self-control and adolescent alcohol use and revealed that high general self-control is an important protective factor for several alcohol-related outcomes [8, 18, 20–31]. Longitudinal studies have demonstrated the effects of general self-control on adolescents’ alcohol use. For example, Kwon [27] found that the level of general self-control at age 11 predicted the frequency of alcohol use at age 15. Moreover, in a sample of Dutch adolescents, higher general self-control at age 13 predicted a lower frequency and quantity of alcohol consumption one year later [26]. Hence, the literature converges to suggest that a higher level of self-control is associated with lower levels of alcohol use among adolescents.

Alcohol-Specific Self-Control and Adolescent Alcohol Use

Next to general self-control, previous research also provides evidence for a relationship between alcohol-specific self-control and adolescent alcohol use. Though most studies were conducted among college students (e.g. [31, 32], among high school students too higher alcohol-specific self-control was associated with less alcohol use [17, 21]. Moreover, a longitudinal study found that alcohol-specific self-control predicted the frequency and quantity of drinking among high school students one year later [33]. Thus, although longitudinal evidence among high school students is scarce, existing findings indicate that a higher level of alcohol-specific self-control is associated with less adolescent alcohol use.

General Self-Control, Alcohol-Specific Self-Control and Adolescent Alcohol Use

Though both general and alcohol-specific self-control relate to adolescents’ drinking, alcohol-specific self-control may be a proximal factor that accounts for the more distal influence of general self-control on adolescent alcohol use. To our knowledge, the mediating role of alcohol-specific self-control in the relationship between general self-control and adolescent alcohol use has not yet been examined. However, Lindgren et al. [18] investigated the unique influence of both general and alcohol-specific self-control on multiple alcohol outcomes among college students. Both types of self-control were related to more favorable outcomes, but the association with alcohol-specific self-control was stronger than the association with general self-control. This indicates that alcohol-specific self-control may be a more proximal factor and general self-control a more distal factor of alcohol use. Additionally, Lindgren and colleagues found that general self-control was positively related to alcohol-specific self-control [18]. However, due to the cross-sectional design of their study, no conclusion could be drawn about the direction of the effect. Adolescents with higher general self-control have less difficulty in resisting temptations in general, because they are better able to foresee the long-term costs of their behavior [10]. Thus, we expect that adolescents with higher general self-control have higher alcohol-specific self-control, because they can resist the temptation of drinking, which in turn, leads to less alcohol consumption [33]. So, we hypothesize that the effect of general self-control on adolescent alcohol use can (partially) be explained by alcohol-specific self-control.

Specificity of Alcohol-Specific Self-Control

Little research has been conducted on the specificity of alcohol-specific self-control. One study examined whether alcohol-specific self-control is related to alcohol use and other behaviors such as smoking, caffeine consumption and exercise behavior [19]. The findings showed that alcohol-specific self-control did not predict these other behaviors, but it did predict the quantity of alcohol consumption. This supports the idea that alcohol-specific self-control is domain-specific as well as substance-specific. However, cross-sectional data were used and the participants in the study were first-year psychology students (mean age 20 years; [19]. Therefore, it is not possible to draw conclusions about the direction of the effect and the findings may not generalize to adolescents in general. Based on the existing study, we expected that alcohol-specific self-control is specifically related to adolescent alcohol use and not to other behaviors such as adolescent digital media use and smoking.

Current Study

In this 4-wave study among 906 adolescents aged 11–14 years, we aimed to examine whether the mediating role of alcohol-specific self-control in the relation between general self-control and adolescent alcohol use is domain- and substance-specific. Based on the self-control theory and previous empirical research, we hypothesized that (1) the effect of a higher level of general self-control on less adolescent alcohol use is mediated by a higher level of alcohol-specific self-control, and that (2) the mediating role of alcohol-specific self-control in the effect of general self-control on adolescent alcohol use is domain-specific and relates to adolescent alcohol use, but does not generalize to other behaviors that require self-control, specifically digital media use and smoking.

Procedure

The longitudinal data used in this study are part of a cluster randomized trial of a Dutch alcohol prevention program for adolescents called Prevention of Alcohol Use in Students (PAS; [34]). In this randomized controlled trial study, 80 schools were randomly selected from a list of all Dutch public secondary schools and invited to participate. Schools were randomly assigned to one of the four conditions: three intervention conditions and one control condition which did not receive any intervention. In the current study, only data from the control condition were used to ensure that the data were not affected by possible intervention effects. All 7th grade students were involved in the study. Data of the students were obtained using online questionnaires which were available on a secured website. Parents of the students received a letter of consent in order to give them the possibility to refuse their children’s participation (0.01% refusal). The data were collected in classrooms during school time by trained research assistants in 2006 (T1); 2007 (T2); 2008 (T3); and 2009 (T4). The study protocol was assessed and approved by the Medical Ethical Committee. Data can be requested by contacting the corresponding author.

Participants

At T1, 906 adolescents between the age of 11 and 14 years (M = 12.19, SD = 0.51) from four different secondary schools participated in the study. Of these adolescents, 52.5% were boys and 95.5% were born in the Netherlands. Sixty comma two percent enrolled in lower secondary vocational education and 39.7% in higher general secondary and pre-university education. Religious background of the adolescents was primarily non-religious (46.1%) and Roman Catholic (34.5%). A total of 865 adolescents (95.47%) at T2, 802 adolescents (88.52%) at T3 and 778 adolescents (85.87%) at T4 completed the follow-up assessments. Reasons for attrition were no permission or adolescents were not present at the day of data collection. Attrition analyses showed no differences at T1 between adolescents who completed the follow-up assessment at T2 and adolescents who did not with respect to gender, age, educational level, general self-control and alcohol use. Nonresponding adolescents at T3 and T4 significantly differed from responding adolescents at these follow-up assessments in being older (T3: t(121) = -2.26, p = .025; T4: t(159) = -3.09, p = .002), being in lower education (T3: F(1,904) = 19.09, p < .001; T4: F(1, 904) = 20.26, p < .001), having lower general self-control (T3: t(901) = 2.87, p = .004; T4: t(901) = 2.38, p = .018), and smoking more frequently at T1 (T3: t(108) = -2.41, p = .018; T4: t(132) = -3.11, p = .002). No differences were found for gender and alcohol use at T1, and alcohol-specific self-control at T2.

Measures

Alcohol use was the outcome variable in this study. By using the Quantity-Frequency measure, we computed the average number of glasses consumed in a week [34]. We asked adolescents how many days they usually consume alcohol during the week (Monday to Thursday) and during the weekend (Friday to Sunday) to assess the frequency of alcohol use. We measured the quantity of alcohol use by asking how many glasses of alcohol the adolescent usually drinks on a weekday and on a weekend day. Thereafter, we computed the average weekly alcohol use by calculating the product of the number of days and the number of glasses for weekdays and weekend days separately, after which these two products were added up. Higher scores indicated higher levels of alcohol use.

Digital media use was one of the control outcome variables and was measured by asking how many hours adolescents spent doing the following activities on a weekday and on a weekend day: watching television (films and DVD’s included), playing computer games or video games (PlayStation, Xbox, GameCube, etc.) and using the Internet for chatting, messaging, surfing the Web, etc. Response options ranged from 1 (never) to 9 (7 hours or more). The sum of the 6 items were computed. Higher scores reflected higher levels of digital media use.

Smoking was the other control outcome variable and was measured by asking the question: “How often do you currently smoke?”. Possible responses were 1 (I do not smoke), 2 (less than once a week), 3 (at least once a week, but not every day), 4 (every day). Higher scores indicated more frequent smoking.

General self-control is the ability to control and adjust thoughts, emotions and behaviors [10]. It was measured by using the brief Self-Control Scale developed and tested by Tangney, Baumeister and Boone [15] which consists of 13 items. Response options ranged from 1 (never) to 5 (very often). Example items are “I am good at resisting temptation", “I often act without thinking through all the alternatives” and “I say inappropriate things”. The mean score of the 13 items was computed. Several items were reversely scored, so that higher scores reflected higher general self-control. Cronbach’s alpha was .72 at T1 and .77 at T2.

Alcohol-specific self-control refers to the ability to refrain from drinking alcohol [17] and was measured by two items: “Are you able to resist drinking alcohol when your friends are offering you a glass of alcohol?” and “Are you able to resist drinking alcohol when all your friends are drinking?”. The mean score of these two items rated on a 4-point scale from 1 (definitely not) to 4 (definitely) was used, with higher scores indicating higher alcohol-specific self-control. Pearson correlation was .80 at T2 and .83 at T3.

Gender, age and educational level were included as control variables, because these factors are related to alcohol use [1, 35, 36]. Adolescents reported their gender (boy = 0, girl = 1) and date of birth. Educational level was known, because the data were collected at schools and adolescents were in classes representing a specific level of education (lower secondary vocational education = 0, higher general secondary and pre-university education = 1).

Data Analysis

Missing data were estimated in Mplus using the Full Information Maximum Likelihood with Robust Standard Errors default setting, allowing information of all participants to be used for analysis [37]. Descriptive statistics and correlations were obtained. We used structural equation modeling to test the effect of general self-control on adolescent alcohol use, digital media use and smoking and the extent to which these associations are mediated by alcohol-specific self-control. In all models, we controlled for gender, age, level of education and outcome at previous wave.

Alcohol-specific self-control was not included in the questionnaire at T1 and adolescent digital media use was not included in the questionnaire at T1 and T4. Therefore, the following measurement moments were used for the different analyses. First, the direct effect of general self-control at T2 on adolescent alcohol use/smoking at T4 was tested. Second, the effect of general self-control at T2 on the mediating variable alcohol-specific self-control at T3 was analyzed. Third, the effect of alcohol-specific self-control at T3 on adolescent alcohol use/digital media use/smoking at T4 was tested while controlling for the effect of general self-control at T2. Finally, the model indirect command in Mplus was used to examine whether the mediated effect was statistically significant. In the analyses with digital media use as outcome measure, general self-control at T1, alcohol-specific self-control at T2 and adolescent digital media use at T3 were used. Adolescent digital media use at T2 was included as control variable. The MLR estimator is robust to non-normality of the outcome variables and is therefore used in the analysis. In line with Nieminen, Lehtiniemi, Vähäkangas, Huusko, and Rautio [38], we used the standardized βs as effect size indices, whereby β < 0.2 was considered a small, 0.2 < β < 0.5 a moderate, and β > 0.5 a strong effect.

Descriptive Results

Table 1 depicts all descriptive data. The average amount of alcohol adolescents consumed at T2 was 1.9 glasses per week (SD = 8.43). At T4 the average amount was substantially higher: 6 glasses per week (SD = 13.10). On the contrary, general self-control and alcohol-specific self-control were quite stable over time; general self-control: M = 3.6 (SD = 0.51) at T1 and M = 3.5 (SD = 0.57) at T2; alcohol-specific self-control: M = 3.2 (SD = 0.88) at T2 and M = 3.2 (SD = 0.91) at T3. Adolescents spent, on average, 21.2 hours (SD = 10.05) in total per week watching television, playing computer games or video games and using the Internet at T2 and 21.8 hours (SD = 10.29) at T3. With regard to smoking, at T2 89.9% of the adolescents reported not smoking, 5.1% at least once a week, but not every day, 2.2% less than once a week and 2,8% reported to smoke every day. At T4 these percentages were 77%, 5.8%, 5.6% and 11.5% respectively.

Table 1

Descriptive Statistics of Research Variables
	n(%)	M(SD)	Min	Max
Gender (boys)	476(52.5%)
Educational level (lower)	546(60.3%)
Age T1		12.2(.51)	11	14
General self-control T1		3.6(.51)	2	4.92
General self-control T2		3.5(.57)	1.23	5
Alcohol-specific self-control T2		3.2(.88)	1	4
Alcohol-specific self-control T3		3.2(.91)	1	4
Alcohol use T2		1.9(8.43)	0	104
Alcohol use T4		6.0(13.10)	0	104
Digital media use T2		21.2(10.05)	0	48
Digital media use T3		21.8(10.29)	0	48
SmokingT2		1.2(.60)	1	4
SmokingT4		1.5(1.03)	1	4
Note. n = number of participants; M = mean; SD = standard deviation.

Table 2 presents the Pearson correlations between all variables of interest. Both types of self-control were significantly negatively correlated with alcohol use, digital media use and smoking, indicating that higher general and alcohol-specific self-control were related to less drinking, digital media use and smoking. In addition, a higher level of general self-control was significantly related to a higher level of alcohol-specific self-control.

Table 2

Correlations between Alcohol Use, Digital Media Use, Smoking, Alcohol-Specific Self-Control, General Self-Control and Control Variables
Variable	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Gender (ref = boys)	1.00
2. Age	− .05	1.00
3. Educational level (ref = lower)	− .03	− .23**	1.00
4. General self-control T1	.04	− .08*	.17**	1.00
5. General self-control T2	.08*	− .07*	.13**	.54**	1.00
6. Alcohol-specific self- control T2	.06	− .01	.08*	.20**	.36**	1.00
7. Alcohol-specific self-control T3	− .00	− .02	.09*	.17**	.30**	.37**	1.00
8. Alcohol use T2	− .08*	.13**	− .14**	− .18**	− .29**	− .30**	− .20**	1.00
9. Alcohol use T4	− .17*	.07	− .09*	− .15**	− .21**	− .27**	− .35**	.42**	1.00
10. Digital media use T2	− .12**	.03	− .16**	− .20**	− .23**	− .13**	− .09*	.20**	.16**	1.00
11. Digital media use T3	− .15**	.02	− .17**	− .18**	− .18**	− .09**	− .09**	.13**	.19**	.44**	1.00
12. Smoking T2	.00	.07*	− .11**	− .23**	− .28**	− .23**	− .20**	.46**	.27**	.16**	.08*	1.00
13. Smoking T4	.02	.03	− .21**	− .22**	− .22**	− .18**	− .20**	.37**	.40**	.16**	.16**	.41**	1.00
Note. Spearman correlation was used for ordinal and continuous variables. Pearson Correlation was used for dichotomous variables.
* p < .05. ** p < .01.

Mediating role of alcohol-specific self-control on Adolescent Alcohol Use

First, we tested the direct effect of general self-control at T2 on alcohol use at T4. Results showed that general self-control significantly predicted future alcohol use while controlling for gender, age, educational level and adolescent alcohol use at T2 (see Table 3). Adolescents who reported higher levels of general self-control consumed less alcohol on a weekly basis two years later (β = − .101, p = .028). The model explained 16.8% of the variance in adolescent alcohol use.

To test the indirect effect of general self-control on adolescent alcohol use via alcohol-specific self-control, we examined the effect of general self-control at T2 on alcohol-specific self-control at T3. As shown in Table 3 and Fig. 1, general self-control was a significant predictor of alcohol-specific self-control while controlling for gender, age, educational level and alcohol-specific self-control at T2. Higher levels of general self-control predicted higher levels of alcohol-specific self-control (β = .226, p < .001). The model explained 18.8% of the variance in alcohol-specific self-control.

Next, we tested the full mediation model (Table 3 and Fig. 1). According to Hu and Bentler’s [39] cut-off criteria for fit indexes to evaluate model fit, the mediation model showed a good fit to the data: CFI = .99, RMSEA = .031, [χ]² = 3.762(2), p = .152. We found a significant negative effect of alcohol-specific self-control at T3 on alcohol use at T4 while controlling for the effect of general self-control at T2 (β = − .200, p < .001). In this step, the effect of general self-control at T2 on alcohol use at T4 was no longer significant (β = − .043, p = .297). A significant total indirect effect of general self-control on use via alcohol-specific self-control was found, indirect = − .045, p = .001). The model explained 19.4% of the variance in adolescent alcohol use.

Table 3

Structural Equation Model of the Effects of General Self-Control and Alcohol-Specific Self-Control on Adolescent Alcohol Use
	Step 1 (X on Y): Alcohol use T4		Step 2 (X on M): Alcohol-specific self-control T3		Step 3 (X on Y via M): Alcohol use T4
Variable	β	SE	β	SE	β	SE
Gender (ref = boys)	− .14***	.03	− .03	.03	− .15***	.03
Age	.03	.04	− .01	.04	.03	.04
Educational level (ref = lower)	− .03	.03	.03	.03	− .03	.03
Alcohol use T2	.32*	.13			.29*	.13
Alcohol-specific self-control T2			.29***	.05
General self-control T2	− .10*	.05	.23***	.04	− .04	.04
Alcohol-specific self-control T3					− .20***	.05
Note. β = standardized coefficient; SE = standard error.
* p < .05. p < .01. * p < .001.

Mediating role of alcohol-specific self-control on Adolescent Digital Media Use

As Table 4 shows, a significant negative effect of general self-control at T1 on digital media use at T3 was found. That is, adolescents with higher general self-control reported less digital media use two years later (β = − .089, p = .012). The model explained 22% of the variance in adolescent digital media use.

We examined the effect of general self-control at T1 on alcohol-specific self-control at T2 (Table 4). A higher level of general self-control at T1 significantly predicted a higher level of alcohol-specific self-control at T2 while controlling for age, gender and educational level (β = .201, p < .001). The model explained 4.8% of the variance in alcohol-specific self-control.

The full mediation model showed an acceptable/good fit to the data, CFI = .96, RMSEA = .098, [χ]² = 9.694(1), p < .05. We found no significant effect of alcohol-specific self-control at T2 on digital media use at T3 (β = − .016, p = .686). Also, the model indirect command showed no significant indirect effect of general self-control on digital media use via alcohol-specific self-control, indirect = − .003, p = .693) (see Table 4). The model explained 21.9% of the variance in adolescent digital media use.

Table 4

Structural Equation Model of the Effects of General Self-Control and Alcohol-Specific Self-Control on Adolescent Digital Media Use
	Step 1 (X on Y): Digital media use T3		Step 2 (X on M): Alcohol-specific self-control T2		Step 3 (X on Y via M): Digital media use T3
Variable	β	SE	β	SE	β	SE
Gender (ref = boys)	− .12***	.03	.05	.03	− .12***	.03
Age	− .06	.03	.01	.04	− .06	.03
Educational level (ref = lower)	− .11***	.03	.05	.03	− .11***	.03
Digital media use T2	.38***	.04			.38***	.04
General self-control T1	− .09*	.04	.20***	.03	− .09*	.04
Alcohol-specific self-control T2					− .02	.04
Note. β = standardized coefficient; SE = standard error.
* p < .05. p < .01. * p ≤ .001.

Mediating role of alcohol-specific self-control on Adolescent smoking

Analyses showed a significant negative effect of general self-control at T2 on smoking at T4 while controlling for gender, age, educational level and adolescent smoking at T2 (see Table 5). Adolescents with higher general self-control smoked less two years later (β = − .137, p = .001). The model explained 27% of the variance in adolescent smoking.

The mediation effect of general self-control on smoking through alcohol-specific self-control was tested next. The mediation model showed a good fit to the data: CFI = .98, RMSEA = .063, [χ]² = 9.163(2), p = .010. As shown in Table 5 and Fig. 2, we found a significant negative effect of alcohol-specific self-control at T3 on smoking at T4 while controlling for the effect of general self-control at T2 (β = − .104, p = .014). The effect of general self-control at T2 on smoking at T4 was still significant (β = − .107, p = .013). The model indirect command showed a significant total indirect effect of general self-control on smoking via alcohol-specific self-control, indirect = − .023, p = .034). The model explained 27% of the variance in adolescent smoking.

Table 5

Structural Equation Model of the Effects of General Self-Control and Alcohol-Specific Self-Control on Adolescent Smoking
	Step 1 (X on Y): Smoking T4		Step 2 (X on M): Alcohol-specific self-control T3		Step 3 (X on Y via M): Smoking T4
Variable	β	SE	β	SE	β	SE
Gender (ref = boys)	.03	.03	− .03	.03	− .03	.03
Age	.00	.04	− .01	.04	.01	.04
Educational level (ref = lower)	− .14***	.03	.03	.03	− .13***	.03
Smoking T2	.42***	.05			.41***	.05
Alcohol-specific self-control T2			.29***	.05
General self-control T2	− .14**	.04	.23***	.04	− .11*	.04
Alcohol-specific self-control T3					− .10*	.04
Note. β = standardized coefficient; SE = standard error.
* p < .05. p < .01. * p < .001.

As far as we know, this is the first study that longitudinally investigated whether the effect of general self-control on adolescent alcohol use is mediated by alcohol-specific self-control and to what extent this mediating role is domain- and substance-specific. In order to answer the question whether the role of alcohol-specific self-control is domain- and substance-specific, we examined if its mediating effect only applies to the effect of general self-control on adolescent alcohol use or extends to the effect of general self-control on adolescent digital media use and/or smoking. Results from a longitudinal study among a large group of adolescents showed that the effect of general self-control on alcohol use was fully mediated by alcohol-specific self-control. Furthermore, the effect of general self-control on smoking was partially mediated by alcohol-specific self-control, but alcohol-specific self-control did not mediate the effect of general self-control on digital media use. These results imply that general self-control predicts alcohol-specific self-control, which seems to be domain-specific, but not substance-specific.

Consistent with previous research (e.g., [23, 27, 29, 40]) and the self-control theory [12], high general self-control seemed to be a protective factor of adolescent risk behavior, because it predicted less alcohol use, smoking and digital media use in adolescence two years later. Consistent with our first hypothesis, the effect of general self-control on alcohol use was completely explained by the level of alcohol-specific self-control. That is, adolescents with a greater ability to control and adjust thoughts, emotions and behavioral tendencies in general were more able to resist the temptations associated with alcohol use, which in turn predicted less alcohol consumption. The fact that alcohol-specific self-control fully mediated the effect of general self-control on alcohol use indicates that alcohol-specific self-control is a more proximal predictor of weekly alcohol use of adolescents than general self-control. Our findings contribute to Lindgren et al.[18] by showing that the association between alcohol consumption and alcohol-specific self-control was stronger than the association between alcohol consumption and general self-control. Possibly that the role of general self-control in adolescents’ alcohol use is dependent on whether it concerns the level of alcohol use or the onset of alcohol use. Some research suggests that general self-control might be a better predictor for the onset of adolescent alcohol use than alcohol-specific self-control. Tipton and Worthington [41] suggested that domain-specific self-control is a better predictor of an individual’s behavior in a clearly defined and familiar situation, and that general self-control is a better predictor of someone’s behavior in situations which are more ambiguous and less familiar to the individual. Adolescents’ first time exposure to peers who use substances may be an example of a more ambiguous and unfamiliar situation. Research shows that alcohol onset is associated with exposure to peers who use alcohol [42-44]. Adolescents who already drink may be more familiar with situations in which their friends are drinking or in which they are offered a drink by a friend. Therefore, it is possible that alcohol-specific self-control is a better predictor of the level of adolescent alcohol use and general self-control a better predictor of the onset of adolescent alcohol use. Our finding that alcohol-specific self-control is a more proximal predictor of the amount of adolescent alcohol use than general self-control supports this suggestion. However, it remains for future studies to examine whether alcohol-specific self-control also acts like a mediator of the effect of general self-control on the onset of alcohol use.

Our results partly confirmed our second hypothesis. Alcohol-specific self-control did not mediate the effect of higher general self-control on less digital media use, but did partially mediate the effect of higher general self-control on less smoking. This provides evidence that alcohol-specific self-control is domain-specific in that it affected substance use; alcohol use as well as cigarette use. This is in contrast with Oei and Burrow [19] who found in their cross-sectional study that alcohol-specific self-control is specifically related to alcohol use and not to other substance use. An explanation for our finding that the mediating role of alcohol-specific self-control extended to smoking is that drinking alcohol and smoking are relatively strongly correlated (r_s =.46 at T2 and r_s=.40 at T4). The link between alcohol use and smoking among adolescents has been shown in various studies (e.g., [45, 46]. Based on our findings, strengthening adolescents’ ability to refrain from drinking alcohol does not have beneficial effects on their ability to resist temptations in other domains (e.g., digital media use), but it may affect their ability to resist other substances the use of which is strongly associated with alcohol use, such as cigarettes. This suggests that it is possible for adolescents to transfer their self-control skills to resist alcohol to situations which involve temptations of other substances. Nevertheless, it remains to be examined whether it is more effective to improve self-control skills for substances in general or to focus on skills addressing each specific substance separately.

Since the results of the current study showed that the ability to refrain from drinking alcohol plays an important role in predicting the quantity of adolescent alcohol use, it is important to know which factors affect alcohol-specific self-control. In this study, we found a significant positive effect of general self-control on alcohol-specific self-control, but the model only explained 18.8% of the variance in alcohol-specific self-control. This indicates the influence of other factors in addition to the influence of general self-control on alcohol-specific self-control. For example, Spoth, Yoo, Kahn and Redmond [47] showed that environmental factors such as peers having prosocial norms and parent-child attachment can affect adolescent alcohol refusal skills. A close examination of the predictors of the ability to refrain from drinking alcohol would be helpful in identifying effective means of improving this ability for intervention purposes.

Strengths and Limitations

The present study is based on a large sample of adolescents and longitudinal data. Due to the longitudinal design it was possible to establish directionality. Furthermore, sophisticated analyses were performed using SEM. Despite these strengths, there are several limitations that should be mentioned. First, alcohol-specific self-control was measured by only two items. Further research should use a validated scale with more items. Second, adolescents’ self-reports were used which could have led to bias, especially related to substance use. Nevertheless, self-reports were found to be a reliable and valid method to measure alcohol use among adolescents [48] and other approaches such as using diary reports are hardly feasible in large studies. Third, because of the fact that not all variables of interest were measured at all four time points, the measurement moments that were used for the analysis with digital media use differed. Ideally, analyses were based on the same sample, but priority was given to optimizing the internal validity of the mediational analysis with alcohol use instead of using the same measurement moments.

In the current study we gained more insight into the relationship between general and alcohol-specific self-control and the level of alcohol use among adolescents, and the extent to which alcohol-specific self-control is domain- and substance-specific. The results showed that alcohol-specific self-control fully mediated the effect of general self-control on alcohol use. Due to its critical role, alcohol-specific self-control rather than general self-control may be a target to efficiently reduce the level of alcohol consumption among adolescents through intervention efforts. Interventions can make use of this knowledge by offering training aimed at improving alcohol resistance skills. Interventions aimed at delaying the onset of adolescent alcohol use may, in contrast, target general self-control. Future studies should examine whether general self-control or alcohol-specific self-control is a more proximal predictor of the onset of adolescent alcohol use in order to confirm this assumption. The finding that alcohol-specific self-control is domain-, but not substance-specific implies that strengthening adolescents’ ability to refrain from drinking alcohol does have beneficial effects on their ability to resist other substances, but not on their ability to resist temptations in other domains (e.g. digital media use). This has implications for the design of interventions. Since high general self-control can play an important role in reducing multiple adolescent risk behaviors via its effect on domain-specific self-control, interventions aimed at reducing adolescent risk behavior in general may focus on improving general self-control.

PAS = Prevention of Alcohol use in Students.

Ethics approval and consent to participate

The trial protocol (NTR649) was approved by the Medical Ethical Committee at Utrecht University. Passive informed consent was obtained; parents were given the opportunity to opt-out from participation of their children in the study.

Consent for publication

Not applicable

Availability of data and materials

The dataset used and analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

The data collection of this study was supported by the Dutch Health Care Research Organization (Z.O.N.-M.W) under Grant 6220, 0021. The funding body had no role in the design of the study and collection, nor in the analysis, and interpretation of data and in writing the manuscript.

Authors' contributions

SG conducted the literature search, analyzed and interpreted the data and drafted the manuscript. IK and CF contributed in writing the manuscript. All authors read and approved the final manuscript

Acknowledgements

Not applicable.

Van Dorsselaer S, Tuithof M, Verdurmen J, Spit M, Van Laar M, Monshouwer K. Jeugd en riskant gedrag 2015. Kerngegevens uit het Peilstationsonderzoek Scholieren. Utrecht: Trimbos-instituut. 2016. Received from http://trimbos.nl
DeWit DJ, Adlaf EM, Offord DR, & Ogborne AC. Age at first alcohol use: A risk factor for the development of alcohol disorders. Am J Psychiat. 2000;157:745-750. http://dx.doi.org/1176/appi.ajp.157.5.745
Ellickson PL, Tucker JS, & Klein DJ. Ten-year prospective study of public health problems associated with early drinking. Pediatrics. 2003;111:949-955. http://dx.doi.org/10.1542/peds.111.5.949
Green KM, Musci RJ, Johnson RM, Matson PA, Reboussin BA, & Ialongo Outcomes associated with adolescent marijuana and alcohol use among urban young adults: A prospective study. Addictive Beh. 2016;53:155-160. http://dx.doi.org/10.1016/ j.addbeh.2015.10.014
Kebede D, Alem A, Mitike G, Enquselassie F, Berhane F, Abebe Y, Ayele R, Lemma W, Assefa T, Gebremichael T. Khat and alcohol use and risky sex behaviour among in-school and out-of-school youth in Ethiopia. BMC Public Health. 2005;5:109. http://dx.doi.org/10.1186/1471-2458-5-109
Komro KA, Tobler AL, Maldonado-Molina MM, Perry CL. Effects of alcohol use initiation patterns on high-risk behaviors among urban, low-income, young adolescents. Prev Sc. 2010;11:14-23. http://dx.doi.org/1007/s11121-009-0144-y
Bava S, & Tapert SF. Adolescents brain development and the risk for alcohol and other drug problems. Neuropsychol Rev.2010;20:398-413. http://dx.doi.org/1007/s11065-010-9146-6
Griffin KW, Botvin GJ, Epstein JA, Doyle MM, & Diaz T. (2000). Psychosocial and behavioral factors in early adolescence as predictors of heavy drinking among high school seniors. J St Alcohol. 2000;61:603-606. http://dx.doi.org/10.15288/jsa.2000.61.603
Strayhorn JM. Self-control: Theory and research. J Am Ac Child Adol Psychiat. 2002;41:7-16. http://dx.doi.org/10.1097/00004583-200201000-00006
De Ridder DTD, Lensvelt-Mulders G, Finkenauer C, Stok M, & Baumeister RF. Taking stock of self-control: A meta-analysis of how self-control affects a wide range of behaviors. Person Soc Psychol Rev. 2012;16:76. http://dx.doi.org/10.1177/1088868311418749
D'Lima GM, Pearson MR, & Kelley ML. Protective behavioral strategies as a mediator and moderator of the relationship between self-regulation and alcohol-related consequences in first-year college students. Psychol Addictive Beh. 2012;26:330. http://dx.doi.org/10.1037/a0026942
Gottfredson MR, & Hirschi A general theory of crime. Stanford CA: Stanford University Press; 2010.
Visser L, de Winter AF, Veenstra R, Verhulst FC, Reijneveld SA. Alcohol use and abuse in young adulthood: Do self-control and parents' perceptions of friends during adolescence modify peer influence? The TRAILS study. Addictive Beh. 2013;38:2841-2846. http://dx.doi.org/10.1016/j.addbeh.2013.08.013
Crone EA, & Dahl RE. Understanding adolescence as a period of social–affective engagement and goal flexibility. Nature Rev. 2012;13:636-650, doi:10.1038/nrn3313.
Tangney JP, Baumeister RF, Boone AL. High self‐control predicts good adjustment, less pathology, better grades, and interpersonal success. J Pers. 2004;72:271-324. http://dx.doi.org/10.1111/j.0022-3506.2004.00263.x
Tsukayama E, Duckworth AL, Kim B. Resisting everything except temptation: Evidence and an explanation for domain‐specific impulsivity. Eur J Pers. 2012;26:318-334. http://dx.doi.org/10.1002/per.841
Jang SA, Rimal RN, Cho N. Normative influences and alcohol consumption: The role of drinking refusal self-efficacy. Health Comm. 2013;28:443-451. http://dx.doi.org/10.1080/10410236.2012.691455
Lindgren KP, Neighbors C, Westgate E, Salemink E. Self-control and implicit drinking identity as predictors of alcohol consumption, problems, and cravings. J St Alc Drugs. 2014;75:290-298. http://dx.doi.org/10.15288/jsad.2014.75.290
Oei TP, Burrow T. Alcohol expectancy and drinking refusal self-efficacy: A test of specificity theory. Addictive Beh. 2000;25:499-507. http://dx.doi.org/10.1016/S0306-4603(99)00044-1
Bogg T, Finn PR, & Monsey KE. A year in the college life: Evidence for the social investment hypothesis via trait self-control and alcohol consumption. J Res Pers. 2012 ;46:694-699. http://dx.doi.org/10.1016/j.jrp.2012.08.004
Yeh MY, Chiang IC, Huang SY. Gender differences in predictors of drinking behavior in adolescents. Addictive Beh. 2006;31:1929-1938. http://dx.doi.org/10.1016/j.addbeh.2005.12.01
Bräker AB, Göbel K, Scheithauer H, & Soellner R. Adolescent alcohol use patterns from 25 European countries. J Drug Issues. 2012 ;45:336-350. http://dx.doi.org/10.1177/0022042615589404
Cheung NW. Low self-control and co-occurrence of gambling with substance use and delinquency among Chinese adolescents. J Gambling Studies. 2014;30:105-124. http://dx.doi.org/10.1007/s10899-012-9351-8
Davies LE, Kuipers MA, Junger M, & Kunst AE. The role of self-control and cognitive functioning in educational inequalities in adolescent smoking and binge drinking. BMC Public Health. 2017;17:714. http://dx.doi.org/10.1186/s12889-017-4753-2
Innamorati M, Maniglio Psychosocial correlates of alcohol use and heavy episodic drinking among Italian adolescents: Data from the second international self‐reported delinquency study. Am J Addictions. 2015;24:507-514. http://dx.doi.org/10.1111/ajad.12239
Koning IM, Van den Eijnden RJ, Vollebergh WA. Alcohol-specific parenting adolescents' self-control, and alcohol use: A moderated mediation model. J St Alc Drugs. 2014;75:16-23. http://dx.doi.org/10.15288/jsad.2014.75.16
Kwon TY. Trajectory and predictors of early adolescent alcohol use. As Soc Work Policy Rev. 2013;7:117-134. http://dx.doi.org/10.1111/aswp.12012
Piquero AR, Gibson CL, Tibbetts SG. Does self‐control account for the relationship between binge drinking and alcohol‐related behaviours? Crim Beh Mental Health. 2002;12:135-154. http://dx.doi.org/10.1002/cbm.492
Quinn PD, Fromme K. Self-regulation as a protective factor against risky drinking and sexual behavior. Psychol Add Beh. 2010;24:376. http://dx.doi.org/10.1037/a0018547
Vazsonyi AT, Trejos-Castillo E, Huang L. Risky sexual behaviors, alcohol use, and drug use: A comparison of Eastern and Western European adolescents. J Adol Health. 2006;39:753-e1. http://dx.doi.org/10.1016/j.jadohealth.2006.05.008
Ehret PJ, Ghaidarov TM, & LaBrie JW. Can you say no? Examining the relationship between drinking refusal self-efficacy and protective behavioral strategy use on alcohol outcomes. Addictive Beh. 2012;38:1898-1904. http://dx.doi.org/10.1016/j.addbeh.2012.12.022
Oei TP, Jardim CL. Alcohol expectancies, drinking refusal self-efficacy and drinking behaviour in Asian and Australian students. Drug Alc Dep. 2007;87:281-287. http://dx.doi.org/10.1016/j.drugalcdep.2006.08.019
Connor JP, George SM, Gullo MJ, Kelly AB, & Young RM. A prospective study of alcohol expectancies and self-efficacy as predictors of young adolescent alcohol misuse. Alcohol Alcoholism. 2011;46:161-169. http://dx.doi.org/10.1093/alcalc/agr004
Koning IM, Vollebergh WAM, Smit F, Verdurmen JEE, Van den Eijnden RJJM, ter Bogt TFM, Stattin H, Engels RCME. Preventing heavy alcohol use in adolescents (PAS): Cluster randomized trial of a parent and student intervention offered separately and simultaneously. Addiction. 2009;104:1669-1678. http://dx.doi.org/1111/j.1360-0443.2009.02677.x
Johnson W, Kyvik KO, Mortensen EL, Skytthe A, Batty GD, Deary IJ. Does education confer a culture of healthy behavior? Smoking and drinking patterns in Danish twins. Am J Epid. 2010;173:55-63. http://dx.doi.org/1093/aje/kwq333
Webb JA, Bray JH, Getz JG, Adams G. Gender, perceived parental monitoring, and behavioral adjustment: Influences on adolescent alcohol use. Am J Orthopsychiat. 2002;72:392. http://dx.doi.org/10.1037//0002-72.3.392
Muthén LK, Muthén BO. Mplus User’s Guide. Seventh Edition. Los Angeles, CA: Muthén & Muthén. 1998-2015.
Nieminen P, Lehtiniemi H, Vähäkangas K, Huusko A, Rautio A. Standardised regression coefficient as an effect size index in summarising findings in epidemiological studies. Epidem Biostat Public Health 2013;10:1–15.
Hu LT, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Eq Mod: A Multidisc J. 1999;6:1-55. http://dx.doi.org/1080/10705519909540118
Nakhaie MR, Silverman RA, LaGrange TC. Self-control and social control: An examination of gender, ethnicity, class and delinquency. Can J Sociol/Cahiers canadiens de sociologie. 2000:35-59. http://dx.doi.org/10.2307/3341910
Tipton RM, Worthington EL. The measurement of generalized self-efficacy: A study of construct validity. J Pers Assess. 1984;48:545-548. http://dx.doi.org/1207/s15327752jpa4805_14
Dupre D, Miller N, Gold M, & Rospenda K. Initiation and progression of alcohol, marijuana, and cocaine use among adolescent abusers. The Am J Addictions. 1995;4:43-48. http://dx.doi.org/10.1111/j.1521-0391.1995.tb00257.x
Kosterman R, Hawkins JD, Guo J, Catalano RF, Abbott RD. The dynamics of alcohol and marijuana initiation: Patterns and predictors of first use in Am J Public Health. 2000;90:360. http://dx.doi.org/10.2105/ajph.90.3.360
Mundt MP. The impact of peer social networks on adolescent alcohol use initiation. Ac 2011;11:414-421. http://dx.doi.org/10.1016/j.acap.2011.05.005
Duncan SC, Duncan TE, & Hops H. Progressions of alcohol, cigarette, and marijuana use in J Beh Medicine. 1998;21:375-388. https://doi.org/10.1023/A:1018730814379
Jackson KM, Sher KJ, Cooper ML, Wood PK. Adolescent alcohol and tobacco use: onset persistence and trajectories of use across two samples. Addiction. 2002;97:517-5 https://doi.org/10.1046/j.1360-0443.2002.00082.x
Spoth R, Yoo S, Kahn JH, Redmond C. A model of the effects of protective parent and peer factors on young adolescent alcohol refusal skills. J Prim Prev. 1996;16: 373-394. http://dx.doi.org/10.1007/BF02411742
Koning IM, Harakeh Z, Engels RCME, Vollebergh WAM. A comparison of self‐reported alcohol use measures by early adolescents: Questionnaires versus diary. Journal of Subst Use. 2010;15:166–73. http://dx.doi.org/10.3109/14659890903013091

Download PDF

Review #1 received at journal
23 May, 2021
Reviewer #1 agreed at journal
13 May, 2021
Reviewers invited by journal
04 May, 2021
Editor assigned by journal
05 Oct, 2020
Submission checks completed at journal
04 Oct, 2020
Editor invited by journal
04 Oct, 2020

You are reading this latest preprint version

How Specific is Alcohol-Specific Self-Control? A Longitudinal Study of the Mediating Role of Alcohol-Specific Self-Control in the Relation between General Self-Control and Adolescent Alcohol Use

Status:

Version 1

Abstract

Figures

Background

General Self-Control and Adolescent Alcohol Use

Alcohol-Specific Self-Control and Adolescent Alcohol Use

General Self-Control, Alcohol-Specific Self-Control and Adolescent Alcohol Use

Specificity of Alcohol-Specific Self-Control

Current Study

Methods

Procedure

Participants

Measures

Data Analysis

Results

Descriptive Results

Mediating role of alcohol-specific self-control on Adolescent Alcohol Use

Mediating role of alcohol-specific self-control on Adolescent Digital Media Use

Mediating role of alcohol-specific self-control on Adolescent smoking

Discussion

Strengths and Limitations

Conclusions

Abbreviations

Declarations

Ethics approval and consent to participate

Consent for publication

Availability of data and materials

Competing interests

Funding

Authors' contributions

Acknowledgements

References

Status:

Version 1