Printer Friendly

Social context, sensation seeking, and teen-age alcohol abuse.

Adolescent involvement with alcohol and prevalence of alcohol abuse are major public health problems in the United States. Sarvela et al|1~ found that 42% of high school seniors in one sample had driven a car at least once in the past six months after drinking. Also troubling is the lack of solid evidence for effectiveness of alcohol abuse prevention programs for this age group.|2-4~ In part, effectiveness of alcohol abuse prevention programming for young people may stem from an incomplete knowledge of the situational and motivational factors that contribute to drinking among adolescents. Both sources of influence -- social situations and internal motivations -- affect teen-age drinking. Interventions that attempt to prevent abuse must address these social and psychological factors.|5~

Thombs and Beck|6~ have employed a construct identified as "social context" to explain social psychological factors that influence teen-age alcohol use and abuse. Social context describes an individual's immediate social environment in which intrapersonal variables beliefs, expectancies, and mood) interact reciprocally with situational variables (time of day, location, and contact with peers) to shape behavior. Distinct social contexts may be identified by the way in which certain internal motivations tend to combine with complimentary social situations. For example, on weekend nights, teen-agers are more apt to drink alcohol to have fun and excitement, whereas drinking on a weekday, after school and work, would more likely be linked to stress relief.

Thus, these two elements of social context tend to change in ways consistent with one another. The social context model represents the process of "reciprocal determinism" as postulated in Social Learning Theory|7~ and applied specifically to teen-age drinking. That is, drinking behavior of youth is shaped by a reciprocal interaction of cognitive, behavioral, and environmental variables.

Several investigations of drinking among high school students used the Social Context of Drinking Scales to identify the distinct situational/motivational patterns of alcohol consumption.|5,6,8~ This line of research revealed that adolescents drink in one or more of the following contexts: Social Facilitation -- where drinking occurs in a party atmosphere, with other friends, and while parents are not present; Stress Control -- where drinking occurs alone, and is used to medicate emotions, depression, and to feel better about oneself; School-Defiance -- where drinking occurs at or around school, on school trips, and to defy school rules and adult authority; Peer Acceptance -- where drinking occurs to be part of a group, to feel grown up, and get someone's approval; and Parental Control -- where drinking occurs at home, with parents, and to celebrate a holiday.

Derived from factor analytic procedures, the Social Context of Drinking Scales are effective in discriminating between abusive and nonabusive drinkers on a variety of indices including binge drinking, drunkenness, and impaired driving. The scales also are useful in distinguishing among adolescent drinking patterns. Furthermore, the scales are internally consistent, not highly interrelated with one another, free of social desirability bias, and not highly redundant with frequency/quantity measures of alcohol use.|5,6,9~

In addition to situational and motivational factors of drinking situation, personality factors also are likely to influence the tendency toward abuse. Research|10-13~ has related the sensation seeking personality trait to various indicants of drug and alcohol involvement. The sensation seeking trait has been defined as, "the need for varied, novel and complex sensations and experiences and the willingness to take physical and social risks for the sake of such experiences."|14~ A high sensation seeker feels a heightened need for different experience, actively seeks thrill and adventure experiences, is disinhibited, and easily bored. Furthermore, individuals with high sensation seeking scores are more oriented to body sensations, extroverted, physically active, impulsive, nonconformist, and antisocial than those with lower scores.|14~ People with higher sensation seeking needs exhibit greater drug and alcohol consumption as well as being more likely to engage in high-risk activities such as drinking and driving.

This study examined the relationship between social context of drinking, sensation seeking, and alcohol abuse among adolescents. In particular, the investigation examined the relationship of these variables to three distinct approaches of measuring adolescent involvement with alcohol: drinking intensity (low, moderate, and high), alcohol-impaired driving, and riding with a driver who is alcohol-impaired. It was hypothesized that the drinking-specific state measures of social context would be superior to the general trait measures of sensation seeking at discriminating high-risk drinkers from low-risk drinkers.

Assessment of both social context and the sensation seeking trait is important because the two models suggest distinct approaches to preventing alcohol abuse in youth. Findings from previous social context research indicate that reducing adolescent alcohol abuse requires a multifaceted approach because teen-age drinking practices are related differentially to specific motivational contexts. For instance, heavy drinking is likely to occur in a context of social facilitation whereas problematic drinking, involving many negative consequences, tends to occur in contexts of stress relief and defiance of authority.|6~

Such findings indicate a comprehensive approach to reducing adolescent alcohol abuse would be complex and require multiple intervention strategies tailored to prominent drinking contexts of a variety of teen-age subgroups. In contrast, sensation seeking studies point to a single intervention strategy. Zuckerman|14~ noted that sensation seeking is a displaceable motive that can be satisfied by many different types of activities. Thus, teaching of nonalcoholic alternatives is endorsed as a viable intervention strategy. The need for stimulation, thrill, and excitement is not suppressed, but channeled toward less destructive activity such as sports, dance, and games.



During the first six months of 1993, an anonymous questionnaire was administered to students attending four rural central schools in western New York State. The questionnaire was completed in extended homeroom periods or in physical education classes. Efforts were made to survey all students in grades 7-12. Due to a variety of circumstances such as absence due to illness, skipping class, participation in vocational programs off campus, and inability to read the questionnaire items, an estimated 80% of students enrolled at the four schools (grades 7-12) actually responded to the questionnaire.

Human subjects approval was obtained from an institutional review board. Students were informed of the voluntary nature of the study, and assured that there was no way to connect them to their responses. To assist with privacy, optical scan answer sheets were used to record responses. School personnel had students place their completed answer sheets in large envelopes at the front of the class. None of the students openly refused to respond to the questionnaire.


Students completed a questionnaire in 20-30 minutes that consisted of three instruments and a variety of demographic and single-item measures of alcohol abuse. The first instrument, the Social Context of Drinking Scales, uses the question stem "How often do you drink alcohol:" which is followed by 32 items. Examples include: "at school or on school grounds," "to be part of a group," "at a friend's house when his or her parents are away," "to blow-off steam," and "on weekend nights." The items assess the social and motivational contexts of adolescent drinking that result from the joint influence of intrapersonal factors and the immediate social environment. Frequency of drinking in five contexts (Social Facilitation, Stress Control, School-Defiance, Peer Acceptance, and Parental Control) is measured by a four-point scale ("never" = 0, "seldom" = 1, "occasionally" = 2, or "frequently" = 3). Composite scores are derived by summing responses to items comprising each subscale. Previous research using high school samples from geographically distinct areas supports the validity of the social context model.|5,6~

The second instrument, the Sensation Seeking Scales (Form V), consists of 40 pairs of items. Using a forced-choice format, respondents select one statement from each pair that describes them best or with which they most agree. Examples include: "I like wild uninhibited parties" versus "I prefer quiet parties with good conversation," and "I would like to try parachute jumping" versus "I would never want to try parachute jumping." Endorsement of statements indicating a higher level of sensation seeking are scored as 1. Statements indicating a lower level of sensation seeking are scored as 0.

The sensation seeking instrument consists of four subscales: Thrill and Adventure Seeking, Experience Seeking, Disinhibition, and Boredom Susceptibility. These factors correlated with adolescent substance abuse, including measures of alcohol misuse.|15,16~ Baker and colleagues|17~ found a strong correlation between sensation seeking and scores on the MacAndrew alcoholism scale in a high school sample. In a longitudinal study of Israeli adolescents, Disinhibition scores strongly predicted heavy alcohol consumption in both boys and girls.|18~ Psychometric characteristics of the sensation seeking scale in high school populations are not clear. Teichman et al|16~ reported the Cronbach's alpha of a unidimensional version of the scale at .79. Pederson|18~ found the test-retest reliability (20 month) of three of the subscales as follows: Thrill/Adventure Seeking (.74), Disinhibition (.60), and Experience Seeking (.59).

The third instrument assessed intensity of alcohol use. It consisted of three items: "frequency of drinking" which uses a seven-point scale ranging from "less than once a month" (scored as 0), to "everyday" (scored as 7), "quantity of drinking" (on a typical occasion), which relies on a 10-point scale ranging from "less than one whole drink" (scored as 0) to "13 or more drinks" (scored as 9), and "frequency of drunkenness," which uses a 10-point scale ranging from "never" (scored as 0) to "everyday" (scored as 9). Composite intensity scores were derived by summing responses to the three items. Previous work using a sample of high school drinkers found alcohol use intensity a reliable measure (Cronbach's alpha = .79).|6~

Two items measured separate aspects of alcohol-impaired motor vehicle travel. Frequency of driving while impaired (DWI) and frequency of riding with an impaired driver (RWID), during the 12 months prior to completing the questionnaire, were measured on a 10-point scale ranging from "never" (scored as 0) to "26 or more times" (scored as 9). Both items describe impairment as "under the influence of alcohol." Copies of the survey instrument may be obtained from the first author.


A total of 1,228 adolescents responded to the questionnaire. Twenty-two answer sheets (1.79%) were withheld from scanning because of obviously frivolous or largely incomplete responses. A slight majority of the sample was female (52.4%). The mean age was almost 15. The proportion at each age was: 12 (14.5%), 13 (15.0%), 14 (15.5%), 15 (16.6%), 16 (17.5%), 18 (6.2%), 19 or older (1.0%). Racial status was not assessed, but the student body of each of the four school districts was at least 90% white.

Most of the sample (56.96%) had consumed alcohol in the preceding 12 months. Among this drinker sub-sample (n = 687), 43.4% drank less often than once a month, 11.7% drank once a month, 19.8% consumed TABULAR DATA OMITTED alcohol two-three times a month, 20.7% used it one-three times a week, and 4.5% drank four or more times a week. With respect to quantity of consumption (on a typical occasion), 35.1% of drinkers reported consuming one drink or less than one whole drink, 32.9% indicated they consumed two-four drinks, 17.7% drank five-eight drinks, and 14.1% drank nine or more drinks. About 39% of the drinkers reported they never become "drunk" when consuming alcohol and another 31% do so only once, or a few times a year.
Table 1

Reliability Coefficients for the Social Context, Sensation Seeking, and
Alcohol Use Intensity Scales

                                             Cronbach's Alpha   Test-Retest(*)

Social Context Scales

Social facilitation (11 items)                      .95               .76
Stress control (7 items)                            .91               .63
School defiance (6 items)                           .90               .39
Peer acceptance (5 items)                           .82               .45
Parental control (3 items)                          .65               .66

Sensation Seeking Scales

Thrill and adventure seeking (10 items)             .72               .69
Experience seeking (10 items)                       .52               .63
Disinhibition (10 items)                            .76               .81
Boredom susceptibility (10 items)                   .58               .66
Total sensation seeking scale (40 items)            .82               .82

Alcohol Use Intensity (3 items)                     .90               .55

* Test-Retest Interval = 14 days, n = 61 drinkers

In contrast, 12.9% of drinkers indicated they were "drunk" at least weekly. Most of the entire sample (85.7%) reported they had not driven a motor vehicle under the influence of alcohol during the previous 12 months. Among the drinker subsample, age 16 or older, 19.9% had driven while impaired at least once during the past 12 months. Examination of the measure assessing "riding with an intoxicated driver" revealed that 34% of the entire sample had engaged in this practice at least once in the previous 12 months.

Reliability Assessment of Instruments

Two methods assessed the scales' reliability among drinkers. Internal consistency was assessed using Cronbach's alpha. To assess the scales' temporal stability, the questionnaire was administered twice over a 14-day period to a subsample of 61 drinkers. Anonymous identifiers were used to link their "time 1" and "time 2" responses. Pearson correlation analyses assessed each subscale test-retest reliability.

Coefficient alphas for the social context, sensation seeking, and alcohol use intensity scales appear in Table 1. Reliability estimates, using Cronbach's alpha, ranged from .52 (Experience Seeking) to .95 (Social Facilitation) in the drinking subsample. Among the social context subscales, the Parental Control subscale possessed only marginal inter-item equivalence (alpha = .65). Among the sensation seeking subscales, neither Experience Seeking (alpha = .52) nor Boredom Susceptibility (.58) had acceptable levels of internal consistency. However, the total unidimensional sensation seeking scale possessed acceptable internal consistency (.82) as well as temporal stability (.82).

In general, the sensation seeking subscales had somewhat greater temporal stability than did the social context scales. Differences show that compared to the personality trait, social contexts of drinking tend to be more fluid and change readily with time. This finding is not surprising given that social context measures include assessment of the social environment of adolescent drinking which may be less stable than the sensation seeking trait. The coefficient alpha of alcohol use intensity was very good (.90), but its temporal stability was marginal (.55). Again, this finding probably reflects the relatively sporadic nature of adolescent alcohol use.

Findings from the reliability assessment indicate the social context scales generally possess superior internal consistency, but with the exception of the Social Facilitation subscale, have somewhat weak temporal stability. The sensation seeking subscales possess moderate to weak internal consistency, but somewhat better test-retest reliability. Because subsequent analyses assessed relationships between variables as they existed at one point in time rather than change over a time interval, internal consistency estimates were used as the basis for selecting subscale measures. Thus, all five social context scales were used in subsequent analyses. Given that only two of the four sensation seeking subscales possessed acceptable levels of internal consistency, the total sensation seeking scale was retained as the only measure of this trait.

A correlation analysis also was performed to determine the relationship between the five social context scales and sensation seeking scale. The highest correlation (.51) occurred between two social context subscales: Social Facilitation and Stress Control. All other correlations were below .50. The sensation seeking scale correlated at or below .27 with each of the social context scales. Therefore, the degree of interrelatedness among these variables indicate that multicollinearity was not an important feature of the data set.

Discriminant Function Analyses

Three separate discriminant function analyses were performed to test the ability of the social context of drinking scales and sensation seeking to distinguish between different types of drinkers. The first analysis used these measures to distinguish among different intensities of drinkers. Drinkers were divided into three categories based on their responses to the three alcohol use intensity items. The first category was defined as low-intensity drinkers (n = 180). These students' drinking frequency was once a month or less, their quantity of consumption was two drinks or less per occasion, and they reported that they were never drunk. Moderate-intensity drinkers (n = 341) exceeded the low-intensity drinkers on one or more of the alcohol use intensity items, but did not meet the criteria for a high-intensity drinker. High-intensity drinkers (n = 100) had a frequency of consumption that was at least once a week or more, their quantity of consumption was at least five or more drinks per occasion, and they reported being drunk at least once a month or more.

The second discriminant analysis distinguished alcohol-impaired drivers from nonimpaired drivers using a subsample of drinkers ages 16 or older. Those younger than 16 were excluded because they were unlikely to be licensed drivers. Impaired drivers were those reporting at least one occasion of driving under the influence in the preceding year, whereas nonimpaired drivers avoided the practice entirely in the same time period. Using the total sample, a third discriminant analysis separated TABULAR DATA OMITTED passenger types, that is, riders of impaired drivers from riders of nonimpaired drivers. Riders were classified with the same criteria used to group drivers. Number of cases used in each of the three discriminant analyses varied due to different selection rules and missing values.
Table 4

Classification Analysis of Three Levels of Drinking Intensity

Actual Group Membership            Predicted Group Membership

                                    Low        Moderate           High
                      N         Intensity      Intensity        Intensity

Low intensity        180       159 (88.3%)     21 (11.7%)        0 (0.00%)

Moderate intensity   341        75 (22.0%)    208 (61.0%)       58 (17.0%)

High intensity       100         3 (3.0%)      13 (13.0%)       84 (84.0%)

Percent of "grouped" cases correctly classified: 72.6%

Table 6

Classification Analyses of Drinking Drivers and Riders with Impaired Drivers

Actual Group Membership              Predicted Group Membership

                       N            Never DWI           Ever DWI

Never DWI             261          216 (82.8%)         45 (17.2%)

Ever DWI               65           21 (32.3%)         44 (67.7%)

Percent of "grouped" cases correctly classified: 79.75%

Never RWID            752          587 (78.1%)        165 (21.9%)

Ever RWID             372          143 (38.4%)        229 (61.6%)

Percent of "grouped" cases correctly classified: 72.60%

To assist in interpretation of the discriminant analyses, means for the social context and sensation seeking measures across each group are presented in Table 2. For all measures, the means increase with higher levels of drinking or high-risk behavior.

Results of the first discriminant analysis revealed that two statistically significant (p |is less than~ .0001) functions were extracted. The first function accounted for most of the variance (94.62%). An examination of the structure matrix (the pooled within-group correlations between the variables and the derived functions) shows that the first function was clearly dominated by the Social Facilitation factor, and to a lesser extent by the Stress Control factor. The other discriminating variables were of lesser importance (structure coefficients |is less than~ .44). An evaluation of the group centroids (points representing the typical position of each group on a scatterplot) show that Function 1 most clearly separated low-intensity drinkers from high-intensity drinkers. Thus, high-intensity drinkers are best distinguished from low-intensity drinkers by drinking in a context of social facilitation as well as stress control.

Function 2 accounted for a much smaller percentage of the variance (5.38%) and was dominated by School-Defiance. An examination of group centroids shows that this function most clearly separated the moderate-intensity drinkers from the high-intensity drinkers. Together, these results indicate a hierarchy of discriminations, such that high-intensity drinkers are separated from low-intensity drinkers by their frequency of drinking for enhancement of fun at social gatherings (Social Facilitation) as well as their need to reduce negative feelings (Stress Control), whereas the high-intensity drinkers can be effectively separated from the moderate-intensity drinkers on the basis of drinking on school grounds with the need to defy school and adult authority.

The importance of these discriminatory factors also was revealed in the accompanying classification analysis which showed that overall about 73% of these students can be classified correctly, and that among the low- and high-intensity drinkers, the classification accuracy reaches 88% and 84%, respectively. Sensation seeking does not appear to be of strong importance in accounting for discrimination among these different types of drinkers.

Two pairwise discriminant analyses distinguished drinkers who drove while impaired (DWI) from those who did not, and those who rode with an impaired driver (RWID) from those who did not. In each analysis, a statistically significant (p |is less than~ .0001) function was extracted. Drinking drivers were most clearly separated from nondrinking drivers by the School-Defiance and Stress Control measures (structure coefficients of .815 and .814, respectively). Social Facilitation also correlated strongly with the discriminant function (structure coefficient = .768), whereas Parental Control and Sensation Seeking had structure coefficients in a moderate range (.40 to .60). Peer Acceptance did not appear to be a particularly strong discriminator of the driver groups. The two passenger groups were best separated by the Social Facilitation and Stress Control measures (structure coefficients of .955 and .771, respectively). However, Peer Acceptance, Sensation Seeking, Parental Control, and School-Defiance also appeared important in distinguishing among the two rider groups. Results of the accompanying classification analyses show that overall these variables correctly classified about 80% of both the DWI groups, and about 73% of the RWID groups. Specifically, 68% of the drinking drivers and 62% of the riders were accurately identified.


Interpretation of findings must be qualified by two methodological limitations. A convenience sample of students from four different schools was used. This procedure introduces uncertainty about the representativeness of the sample. Second, the study relied on self-report data. Though anonymous self-report measures of alcohol abuse are generally considered valid,|19~ some degree of underreporting may have occurred.

Results confirmed the hypothesis that the drinking specific measures of social context were superior to the sensation seeking personality measure at being able to discriminate between different drinking patterns among adolescents. The sensation seeking measure was of marginal importance in discriminating among different intensity levels of teen-age drinking. The trait measure had somewhat greater ability to distinguish between types of drivers and passengers. These findings, combined with results from earlier investigations,|5,6~ substantiate the validity and usefulness of the Social Context of Drinking Scales at assessing different levels of alcohol involvement among adolescents.

Teen-agers prone to abuse alcohol not only display different patterns of alcohol intake but they show differences in where, when, and why they drink. These situational and motivational factors, or social contexts of drinking, are generally more important for distinguishing between abusing and nonabusing drinkers than mere measures of quantity and frequency.|9~ Thus, knowing where, when, and why an adolescent drinks is likely to be far more important for assessment as well as for prevention than knowing how much and how often they drink. This view was noted in previous research where heavy or high-intensity drinkers are best separated from high-consequence drinkers (those who experience many problems such as fights, arguments with friends, academic problems) by the Stress Control social context scale.|6~ While both groups reported similar levels of drinking intensity, high-consequence adolescents report greater levels of stress control drinking.

These results suggest that Social Facilitation, primarily, and Stress Control, secondarily, distinguish low-intensity from high-intensity drinkers, but use of alcohol in a context of School-Defiance further distinguishes drinkers among the higher-intensity groups (Moderates vs. Highs). This rebellious context also was important for discriminating alcohol-impaired drivers from nonimpaired drivers. Thus, adolescents who drink frequently to enhance sociability and have a good time at parties, to medicate against negative self-thought and mood, and to rebel against authority comprise a high-risk group prone to drink to excess, experience a significant number of alcohol-related problems, and drive while impaired by alcohol. The Social Context of Drinking Scales offer a reliable means of measuring these tendencies among adolescents, which enables them to be more easily identified and targeted for special intervention efforts.

These findings imply for program planning and prevention that heavy and problematic drinking are most likely to occur in a context of socialization, where adolescents have a strong need for interaction with peers to develop interpersonal competencies and have fun in an atmosphere free from parental supervision. Because of this tendency, and the literature which suggests adolescents experience higher levels of sensation seeking at this stage than at any other time in their life,|14~ it seems evident that most adolescents will actively pursue social and recreational opportunities to interact with each other in exciting and fun ways. These social needs are not inherently destructive and attempts should not be made to suppress them. Unfortunately, alcohol currently seems to be positioned very strongly as a means for accomplishing fun and excitement, especially among abusers.

School-based alcohol prevention efforts may be better served if they try to orchestrate fun, exciting, and stimulating party-like events for young people, which are free from alcohol, as well as overt parental supervision.|6~ High-intensity drinkers are using alcohol as a means of achieving a goal -- social facilitation. If that goal is satisfied through appealing activities such as after-prom parties, alcohol-free teen night clubs, or other events that offer exciting, novel activities and allow teens to interact with each other in an enhanced convivial atmosphere, then the need for alcohol as a means of satisfying social facilitation and sensation seeking needs should be diminished.

School-based efforts at alcohol-abuse prevention traditionally tend to be directed more at eliminating the means (alcohol) rather than emphasizing ways by which students can achieve the goal (fun, excitement, and social facilitation). Obviously, these issues extend to needs for stress relief and rebellion as well, and suggest alcohol abusers often need assistance with coping and social integration skills. Identifying such students in middle school, so that such training could be applied, may be more effective at reducing alcohol abuse in this population than reliance on student assistance programs which have primarily emphasized abstinence and treatment referral for the older adolescent.

Findings from this investigation also focus attention on two longstanding problems that thwarted development of comprehensive alcohol abuse prevention programs for youth. The first problem pertains to program goals. Should abstinence from alcohol for all teen-agers be sought? Or, should programs seek to reduce high-intensity drinking and alcohol-impaired traveling? Each goal suggests different intervention strategies. Findings presented here show discernable patterns to adolescent alcohol abuse and suggest that effective intervention requires specific program goals and objectives for different alcohol-related behaviors. Many prevention programs have lacked such clarity and direction.

The second problem concerns the degree to which the school, by itself, can prevent student alcohol abuse. Social context research demonstrated teen-age alcohol abuse is a complex phenomenon driven not only by motivational factors within the adolescent, but by the social environment of the community. Specifically, social context research documented that much drinking occurs away from school, on weekend nights, at unsupervised parties, and often involves motor vehicle travel. Given these conditions, it is difficult to imagine that classroom activities, student assistance programs, or other solely school-based interventions can possibly succeed. Schools will play a role in comprehensive alcohol abuse prevention programs, but coordinated community-based efforts are needed to effectively address such a complex problem.

Finally, results indicate that the Peer Acceptance factor was not particularly important in distinguishing among different levels of alcohol intensity or between impaired and nonimpaired drivers. Furthermore, the structure coefficient for Peer Acceptance was lower than those of Social Facilitation and Stress Control as a discriminator of passenger groups. This finding suggests peer pressure is not a major reason for alcohol abuse in adolescents. This view was supported by May|20~ who argues that peer pressure is a faulty premise for alcohol education and prevention because it is founded on the erroneous assumption that adolescents who misuse alcohol are in some way socially incompetent and ". . .unable to resist contamination by group influences . . ."

The need for peer acceptance and approval obviously exist for teen-agers, so drinking with peers for this reason is an identifiable context. However, this tendency does not seem to be a strong predictor of drinking intensity or abuse. Therefore, the peer resistance training component of many school-based alcohol abuse prevention programs (such as DARE) needs to be rethought. Results of this investigation suggest that, at least for issues concerning alcohol abuse prevention, the emphasis currently placed on acquisition of skills to instill immunity to peer pressure is misplaced.


1. Sarvela PD, Pape DJ, Odulana J, Bajracharya SM. Drinking, drug use, and driving among rural midwestern youths. J Sch Health. 1990;60:215-219.

2. Bagnell G. Educating Young Drivers. London, England: Tavistock/Routledge; 1991.

3. Mann RE, Vingilis ER, Leigh G, Anglin L, Blefgen H. School-based programs for the prevention of drinking and driving: Issues and results. Accident Anal Prey. 1986: 18:325-337.

4. Moskowitz J. The primary prevention of alcohol problems: A critical review of the research literature. J Stud Alcohol. 1989;50(1):54-87.

5. Beck KH, Thombs DL, Summons TG. The social context of drinking scales: Construct validation and relationship to indicants of abuse in an adolescent population. Addict Behav. 1993;18:159-169.

6. Thombs DL, Beck KH. The social context of four adolescent drinking patterns. Health Educ Res. In Press.

7. Bandura A. Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall; 1977.

8. Beck KH, Summons TG. The social context of drinking among high school drinking drivers. Am J Drug Alcohol Abuse. 1987;13: 181-198.

9. Beck KH, Summons TG, Thombs DL. A factor analytic study of social context of drinking in a high school population. Psychol Addict Behav. 1991;5:66-77.

10. Archambault D. Adolescence: A physiological, cultural, and psychological no man's land. In: Lawson GW, Lawson AW, eds. Alcoholism and Substance Abuse in Special Populations. Rockville, Md: Aspen; 1989:223-245.

11. McMillen DL, Pang MG, Wells-Parker E, Anderson BJ. Alcohol, personality traits, and high risk driving: A comparison of young drinking driving groups. Addict Behav. 1992;17:525-532.

12. Segal B, Huba GJ, Singer JL. Drugs, Daydreaming and Personality: A Study of College Youth. Hillsdale, NJ: Lawrence Earlbaum; 1980.

13. Webb JA, Baer PE, Francis DJ, Caid CD. Relationship among social and intrapersonal risk, alcohol expectancies, and alcohol usage among early adolescents. Addict Behav. 1993;18:127-134.

14. Zuckerman M. Sensation Seeking: Beyond the Optimal Level of Arousal. Hillsdale, NJ: Lawrence Erlbaum; 1979;10.

15. Arnett J. Drunk driving, sensation seeking, and egocentrism among adolescents. Person Ind Diff. 1990;11:541-546.

16. Teichman M, Barnea Z, Rahav G. Sensation seeking, state and trait anxiety, and depressive mood in adolescent substance users. Int J Addict. 1989;24:87-99.

17. Baker K, Beer J, Beer J. Self-esteem, alcoholism, sensation seeking, grade point average, and differential aptitude test scores of high school students in honor society. Psychol Rep. 1991;69:1147-1150.

18. Pederson W. Mental health, sensation seeking and drug use patterns: A longitudinal study. Brit J Addict. 1991;86:195-204.

19. Johnston LD, O'Malley PM. Issues of validity and population coverage in student surveys of drug use. In: Rouse BA, Dozel NJ, Richards LG, eds. Self-Report Methods of Estimating Drug Use: Meeting Current Challenges of Validity. Washington, DC: US Government Printing Office; 1985;31-54; publication 85-1402.

20. May C. Resistance to peer group pressure: An inadequate basis for alcohol education. Health Educ Res. 1993;8(2):159-165.

Dennis L. Thombs, PhD, Assistant Professor; Colleen A. Mahoney, PhD, Assistant Professor; Michael D. Bromley, MSEd, Graduate Student; and Karen M. Bezon, MSEd, Graduate Student, Dept. of Health Science, State University of New York, College at Brockport, Brockport, NY 14420; and Kenneth H. Beck, PhD, Associate Professor, Dept. of Health Education, University of Maryland, College Park, MD 20742. This article was submitted October 14, 1993, and revised and accepted for publication January 3, 1994.
COPYRIGHT 1994 American School Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Thombs, Dennis L.; Beck, Kenneth H.; Mahoney, Colleen A.; Bromley, Michael D.; Bezon, Karen M.
Publication:Journal of School Health
Date:Feb 1, 1994
Previous Article:A longitudinal analysis of adolescent smoking and its correlates.
Next Article:The Human Genome Initiative: implications for the comprehensive school health program.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters