AddictiveBehaviors,Vol. 15, pp. 119-128, Printed in the USA. All rights reserved.
1990 Copyright
0306-460300 $3.00 + .OO Q 1990 Pergamon Press plc
BEHAVIORAL SELF-CONTROL STRATEGIES FOR DELIBERATELY LIMITING DRINKING AMONG COLLEGE STUDENTS CHUDLEY University
E. WERCH of Arkansas
All&net - This study examined the use of behavioral self-control strategies, across specific procedures and sex, and the relationship between self-control and alcohol consumption and problem variables. A total of 456 randomly selected college students participated in a survey of campus substance use. Behavioral strategies most commonly used to deliberately limit drinking among college students were those related to limiting driving/riding when drinking. controlling time and food stimuli, and awareness of internal and external cues to control drinking. Strategies related to self-reinforcement and punishment, certain alternatives to alcohol use, and specific rate control techniques were least used by drinkers. Females were more likely than males to use all but one of the behavioral strategies. The degree of self-control practiced was associated with alcohol consumption, driving/riding after drinking, improvement in limiting drinking, health beliefs related to alcohol problems, and perceived effectiveness of behavioral strategies to limit drinking. Specific self-control strategies were found to significantly predict alcohol-related variables, with confining drinking to certain times of the week and refusing unwanted drinks the most consistent predictors.
Various studies have been conducted examining self-initiated attempts to control smoking behavior (Baer, Foreyt, & Wright, 1977; Curry & Marlatt, 1985; DiClemente & Prochaska, 1982; Dropkin, 1984; Pederson & Lefcoe, 1976; Petri & Richards, 1977; Perri, Richards, & Schulteis, 1977; Prochaska & DiClemente, 1983; Rosen & Shipley, 1983; Watts & Kunkle, 1985), with a fewer number of recent studies assessing self-change attempts to limit alcohol use (Perri, 1985; Werch & Gorman, 1988). These studies are based upon two assumptions. First, that a considerable number of people appear to be successful in controlling and quitting addictive substances on their own, even in the face of evidence that formalized treatment programs commonly result in modest success. Second, that studying naturally occurring episodes of self-change can contribute to our understanding of change processes which in turn can stimulate the development of hypotheses leading to advanced theory and clinical applications. In one of the few recent studies of self-initiated attempts to control alcohol use, Perri (1985) examined the self-change efforts of adults coping with problem drinking over the course of a 6-8 month period. He found that successful drinkers, defined as those who were abstinent or who were not engaged in problem drinking, applied a wider array of behavioral self-control strategies in their coping efforts than did unsuccessful drinkers. In particular, successful drinkers were more likely to implement stimulus control procedures and use alternative behaviors to take the place of drinking, compared with unsuccessful drinkers. Werch and Gorman (1988) examined the relations between self-control strategies, alcohol problems, and alcohol consumption patterns of college students. They found that external (e.g., number of drinks consumed) and internal (e.g., physiological sensations) self-control strategies significantly correlated with alcohol problems, and the quantity and frequency of alcohol consumption. Students experiencing alcohol problems at least once during the past
Current affiliation: University of North Florida. Requests for reprints should be sent to Chudley E. Werch, Ph.D., Assistant Professor, University of North Florida College of Health, Center for Alcohol and Drug Studies, 4567 St. Johns Bluff Road, Jacksonville, PL 32216. 119
120
CHUDLEY
E. WERCH
year were as likely to employ self-control strategies as nonproblem drinkers with two exceptions: (a) problem drinkers were more likely to apply self-reinforcement and punishment strategies, and (b) problem drinkers were more likely to practice strategies to control time and food factors related to drinking. In addition, this study found that self-control strategies were applied more frequently across greater alcohol consumption patterns with moderate drinkers typically practicing the greatest amount and variety of strategies. Research to date on self-change approaches to alcohol use has been scarce, particularly when compared to that conducted on smoking cessation. The present study answered the following three questions: (a) Are there differences as to how frequently behavioral self-control strategies are employed across specific strategies and sex? (b) What is the relationship between the degree of self-initiated self-control and measures of alcohol consumption, alcohol-related problems, alcohol beliefs, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? and (c) Which specific behavioral self-control strategies have the greatest impact on alcohol consumption, alcohol-related problems, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? METHOD
Subjects Subjects were randomly sampled from all those enrolled at a southern university by generating a computer listing of 800 student names with addresses. Seventeen of these students could not be located. A total of 456 college students participated in this study by completing a mailed survey of their alcohol and other drug use. Subject demographic data indicated that they were equally distributed on sex with 5 1 .O% males. The mean age was 25 years (SD = 8.51). The majority of the students were single (67.0%) with 27.0% married,and 5.4% divorced. Students were enrolled in Business or other majors (34.8%), Arts and Sciences (31.7%), Education (16.3%), Engineering (14.3%), and Law (2.8%). Most were classified as Seniors (23.5%), then Graduate or other classification (22.8%), Freshman (20.2%), Juniors (18.7%), and Sophomores (14.8%). The majority were White (89.3%) followed by Oriental (6.1%), and Black (3.9%). Religious preferences included Protestants allowing alcohol use (38.1%), Protestants not allowng alcohol use (27.5%), none or other religion (22.0%), Catholic (12.2%), and Jewish (0.2%). The majority of students resided in rented apartments or houses (48.5%) followed by campus residence (22.3%), other residences (13.1%), Fraternity or Sorority housing (9.4%), and at home with parents (6.8%). Of the total sample, 79.9% drank alcohol beverages at least once a year or more. The mean number of drinking days in the last month was 4.86 (SD = 6.46), and the average number of drinks consumed equaled 3 .Ol (SD = 3.73). Procedure Randomly selected students were mailed a copy of the Student Substance Use Survey and an accompanying cover letter describing the study. A follow-up letter and a second copy of the survey were sent to those who did not respond to the earlier request. A third letter was sent to those subjects who did not respond to the two previous requests. The Student Substance Use Survey was a 6-page instrument designed to determine college student demographics, alcohol consumption patterns, frequency of alcohol-related problems, ratings of improvement concerning alcohol use, alcohol-related health beliefs, frequency of use of behavioral self-control strategies to limit drinking, ratings of effectiveness of behavioral strategies to limit alcohol use, and other drug-related measures. The survey was pretested on a sample of university students and revised to eliminate ambiguous items.
Behavioralself-control
121
Corrected split-half reliability determined from a sample of college students was found to be high, r = .85. Information on alcohol consumption consisted of items to determine the exact number of days drinking in the last month and the exact number of drinks usually consumed, with one drink being equal to one 12-0~ beer, one 4-0~ table wine, or one 1-oz 86-proof liquor. An additional item asked how often subjects drank alcohol (beer, wine or liquor), using a five-point response of: (a) never drank or used to drink but quit, (b) once or more a year, but less than once a month, (c) once or more a month, but less than once a week, (d) once or more a week, but less than once a day, (e) every day. Items measuring the frequency of alcohol problems asked how often students experienced each of 11 alcohol-related events over the past six months. These items were rated on a five-point scale of “ 1: Never” to “5: Always” and included the following events related to drinking: interpersonal conflicts, legal problems, health problems, financial problems, illness events, school conflicts, sexual problems, aggression, feelings of depression, fatigue or tiredness, and work problems. Ratings of improvement concerning alcohol use included two items. One item asked how much improvement subjects experienced in limiting their drinking over the past six months. The other item asked how much improvement was experienced in reducing problems due to drinking over the last six months. Both items were scored on a five-point scale of “ 1: None at all or never drank” to “5: Complete improvement.” Information on alcohol-related health beliefs was determined from four items representing the four psychological variables providing constructs for the Health Belief Model (HBM) (Becker, 1974). These four health belief variables, found useful in predicting health behavior, included perceived susceptibility, severity, benetits, and barriers related to alcohol use behavior and problems. Each of these four items was scored on a five-point scale of “ 1: Strongly agree” to “5: Strongly disagree.” The frequency of use of behavioral self-control strategies was determined by asking students how often they used certain strategies to deliberately limit their drinking, over the past six months. Each of these items were scored on a five-point scale from “1: Never” to “5: always.” Behavioral strategies were selected from a previous study (Werch & Gorman, 1986) of a factor analysis of 37 external self-control behaviors, developed originally from self-help program manuals (Miller & Munoz, 1982; Vogler & Bartz, 1982). For the purpose of survey brevity, the present study employed 14 of the original 37 strategies by selecting the two highest loadng strategies from each of the seven factors identified in the original analysis. These seven factors were identified as: (a) Rate Control items pertaining to controlling drinking rate and setting drinking limits; (b) Self-reinforcement and Punishment - items describing rewards and punishment for attempts at maintaining drinking limits; (c) Alternatives - items related to substituting alternative means of coping in place of drinking; (d) Avoidance - items concerning avoidance techniques specifically intended either to limit alcohol consumption or unwanted consequences; (e) Limiting Driving and Cash - items related to controlling driving and money when drinking; (f) Controlling Time and Food-items concerning limiting drinking to certain times of the day and week as well as eating before drinking; and (g) Awareness - items related to controlling the number of drinks by an awareness of internal (physiological) or external (number of drinks) factors. Test-retest reliability for the original 37 behavioral strategies has been found to be very high (r = .96) based on results from a group of uni.versity students (Werch & Gorman, 1986). Ratings of the effectiveness of behavioral strategies to limit alcohol use consisted of two items. One item asked how effective the above metnioned strategies were for helping students to limit their drinking over the last six months. The second item asked how effective
122
CHUDLEY
Table 1. Percentage
E. WERCH
of males and females frequency self-control strategies
of use of specific behavioral
Frequency of use of strategies Seldom Sometimes Often
Strategy/Sex
Never
Select drinks I drink slowly** Males Females
48.43 25.00
10.69 12.20
18.24 28.05
11.32 16.46
11.32 18.29
Set time limits before another* Males Females
63.52 48.15
16.35 14.81
10.06 16.67
3.77 11.11
6.29 9.26
Reward myself for limiting Males Females
79.87 70.00
11.95 13.13
5.03 10.63
1.26 3.13
1.89 3.13
Punish myself for not limiting Males Females
83.65 80.00
11.32 8.75
2.52 8.13
1.89 1.25
0.63 1.88
Substitute other menas for a “high” Males Females
70.44 65.82
4.40 10.13
8.81 11.39
11.32 7.59
5.03 5.06
Substitute other means for socializing Males Females
40.88 31.93
14.47 12.05
24.53 27.11
12.58 18.67
7.55 10.24
Avoid drinking with heavy drinkers* Males Females
43.48 26.95
11.80 11.98
16.15 19.16
18.63 22.16
9.94 19.76
Avoid drinking with those pressuring* Males Females
48.13 30.25
9.38 8.64
9.38 18.52
15.00 14.81
18.13 27.78
Drink less when driving Males Females
16.35 11.95
1.89 2.52
8.18 7.55
22.64 15.72
50.94 62.26
Don’t drive when drinking Males Females
18.24 10.37
5.66 3.05
11.32 14.02
28.30 27.44
36.48 45.12
Eat before drinking Males Females
11.25 9.15
5.63 1.83
22.50 22.56
32.50 36.59
28.13 29.88
Confine drinking Males Females
to certain times 15.00 13.50
9.38 6.13
10.00 15.34
36.25 35.58
29.38 29.45
Refuse unwanted Males Females
drinks 15.63 8.43
1.25 3.01
6.25 9.64
30.63 25.90
46.25 53.01
21.38 13.58
6.29 6.17
19.50 16.05
27.67 25.93
25.16 38.27
Use body sensations Males Females
Always
to slow down
*p 5 .Ol. **p 5 ,001.
these strategies were for preventing problems due to drinking over the past six months. Both items were scored on a five-point scale of “ 1: Not at all or never used” to “5: Extremely effective. ”
Behavioral self-control
123
RESULTS
The first question raised in this study was: Are there differences as to how frequently behavioral self-control strategies are employed across specific strategies and sex? As Table 1 shows, the answer is yes. With regard to specific strategies, those which were used (at least sometimes) by the greatest percentage of drinkers included: (a) eating before drinking, 86%, (b) refusing unwanted drinks, 85%; (c) drink less when driving, 83%; (d) arrange not to drive when drinking, 81%; (e) confine drinking to certain times of the week, 78%; and (f) use body sensations as cues to slow down drinking, 76%. Those strategies which were used by the least proportion of drinkers included: (a) punish oneself for failing to limit drinking, 8%; (b) reward oneself for limiting drinking, 12%; (c) substituting other means for achieving a “high”, 24%; and (d) setting time limits before having another drink, 28%. Differences were also found across sex, with a greater proportion of females than males employing all but one of the strategies studied (i.e., substitute other means for achieving a “high”). A significantly greater proportion of females used the following four strategies: (a) select drinks which are drunk more slowly, x2 = 19.80, 4 df, p = .OOl; (b) set time limits before having another drink, x2 = 12.82, 4 df, p = .Ol; (c) avoid drinking with heavy drinkers, x * = 12.60, 4 df, p = .Ol; and (d) avoid drinking with those who pressure one to drink, x * = 14.70, 4 df, p = .005. The second question raised in this study was: What is the relationship between the degree of self-initiated self-control and the following alcohol variables: alcohol consumption, alcohol-related problems, alcohol beliefs, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? For purposes of analysis, all drinkers were categorized as practicing either a low, moderate, or high degree of self-control. The degree of self-control was based upon the frequency of employing self-control strategies, totaled across the 14 behavioral self-control items. Self-control categories and collateral scores included: (a) low (seldom) control 5 28; (b) moderate (sometimes) control 2 29 and < 42; and (c) high (often) control L 43. Table 2 shows that nearly all alcohol-related variables were related to the degree of self-control (ANOVA tests, ps 5 .OS), with the exception of total alcohol problems and the health belief associated with overcoming barriers to limiting drinking. Alcohol consumption measures, including the number of drinking days last month and number of drinks per drinking occasion, differed significantly across self-control. The greatest mean alcohol measures were reported by those in the moderate control category, with no pairwise differences between low and high control subjects. Although no differences were found across self-control for total alcohol problems, the measures related to driving after drinking and riding with a drinking driver did differ significantly across self-control. As with alcohol consumption, measures of drinking and driving/riding were greatest for the moderate control subjects. Results for measures of perceived improvement in drinking were somewhat different. The greatest mean scores for improvement in drinking occurred among the high control subjects, although no pairwise differences were found between the moderate and high control subjects. Three of the four health beliefs were found to differ significantly across self-control, with the belief related to overcoming barriers to successfully limit drinking not associated with self-control. Beliefs related to the perceived susceptibility and severity of alcohol problems, and the benefits of limiting alcohol consumption, were most agreed to by the subjects in the high control category. Lastly, both measures of the effectiveness of behavioral strategies to limit alcohol use differed significantly across control, with the perceived effectiveness increasing across greater control efforts. The third question raised in this study was: Which specific behavioral self-control
CHUDLEYE. WERCH
124
Table 2. Mean scores of alcohol-relatedvariables by degree of behavioral self-control
LOW
Degree of self-control Moderate High
control N = 59
control N = 142
F
P
M
M
M
value
value
5.19
1.89
5.31
6.25
.0022
Number of drinks per occasion
3.62
4.52
3.19
4.00
.0191
Total alcohol problems score
Alcohol
variables Number of drinking last month
control N=l33
days
14.77
15.80
14.90
1.45
.2351
Frequency of driving after drinking
1.17
2.16
1.66
10.65
.ooOl
Frequency of riding with a drinking driver
2.06
2.32
1.83
11.11
.OOOl
Improvement drinking
1.66
2.46
2.51
10.19
.OOOl
Improvement in reducing drinking problems
1.50
2.30
2.38
9.47
.OOOl
Belief that serious problems can result from drinking
1.72
1.68
1.46
3.38
.0351
Belief that I could develop alcohol-related problems
2.91
2.19
2.50
2.94
.0543
Belief that there are benefits from limiting drinking
1.96
1.75
1.48
6.70
.0014
Belief that I can learn to limit my drinking
1.63
1.60
1.48
1.29
.2757
Effectiveness of behavioral strategies to limit drinking
2.15
3.34
3.87
51.09
.OOOl
Effectiveness of behavioral strategies to prevent problems
2.00
3.24
3.78
40.94
.OOOl
are significantly C .05).
different
in limiting
Mean scores connected with an underline (least-squares means pairwise comparisons
from one another
strategies have the greatest impact on the alcohol variables studied? Table 3 shows the summary of stepwise multiple regression analyses on the major alcohol-related variables. Two behavioral strategies were found to significantly predict the number of drinking days in the last month, and four strategies predicted the number of drinks per drinking occasion. Substituting other means for socializing was the most significant predictor of the number of drinking days. Avoiding drinking with those who pressure, confining drinking to certain times of the week, and refusing unwanted drinks were the best predictors of the number of drinks consumed. Six strategies were found to significantly predict total alcohol problems,
Behavioral self-control
125
with the most powerful predictors being: confining drinking to certain times of the week, rewarding oneself for limiting drinking, and refusing unwanted drinks. Three strategies significantly predicted improvement in limiting drinking and one strategy predicted improvement in reducing alcohol problems. For both measures of improvement, the behavioral strategy of confining drinking to certain times of the week was the most significant predictor. Lastly, four strategies significantly predicted the effectiveness of behavioral strategies to limit drinkingand five predicted the effectiveness of strategies to prevent alcohol problems. The best predictors of effectiveness of strategies to limit drinking were: refusing unwanted drinks, setting time limits before having another drink, and avoiding drinking with heavy drinkers. The most significant predictors of the effectiveness of strategies to prevent alcohol problems were: refusing unwanted drinks, avoiding drinking with heavy drinkers, and selecting drinks which are drunk slowly. DISCUSSION
This study found that behavioral self-control strategies are deliberately employed with varying levels of frequency, across specific strategies and sex, by college students. The degree of self-control practiced by subjects was associated with alcohol consumption, driving and riding after drinking, improvement in limiting drinking, health beliefs related to alcohol problems, and perceived effectiveness of behavioral strategies to limit drinking. Specific self-control strategies were found to significantly predict measures of alcohol consumption, problems, improvement, and effectiveness of behavioral strategies. Confining drinking to certain times of the week was the strategy most consistently predicting measures of alcohol use and problems, while refusing unwanted drinks was the strategy best predicting the perceived effectiveness of behavioral strategies to limit drinking. A surprisingly large percentage of college students was found using those behavioral self-control strategies related to limiting driving/riding when drinking, controlling time and food stimuli, and awareness of internal and external cues to control drinking. These data support previous research (Werch & Gorman, 1988) suggesting that college students deliberately employ behavioral strategies to limit alcohol consumption. Strategies related to self-reinforcement and punishment, certain alternatives to alcohol use (i.e., substituting other means for achieving a “high”), and specific rate control techniques (i.e., setting time limits before having another drink), however, were used by very few subjects. Females were found to be more likely to use behavioral self-control strategies than men, particularly those strategies related to rate control and avoidance. These data have implications for the development and testing of programs for alcohol abuse prevention and intervention, grounded within a self-control model. For example, innovative alcohol abuse interventions emphasizing education and training in those self-control strategies which are least likely to be employed without external intervention could be tested for their effects, particularly with college-aged males. This study also found that the degree of self-control practiced was associated with a host of alcohol-related variables. The number of drinking days per month, the number of drinks per drinking occasion, and the frequency of driving and riding after drinking were related to self-control. Specifically, those individuals who seldom or often employed self-control strategies generally reported significantly lower alcohol consumption and drinking-driving events than those who sometimes used self-cntrol strategies. Although self-control was not significantly associated with alcohol problems, there was also a trend for individuals who seldom or often used self-control to report experiencing fewer alcohol problems. These data suggest that the relationship between self-control and alcohol consumption and problem measures is nonlinear, with those students using few self-control strategies, and those
126
CHUDLEY
Table 3. Summary
E. WERCH
of stepwise multiple regression analyses of alcohol-related behavioral self-control strategies
Alcohol variables/ behavioral strategy
variables by
R2
F value
Number of drinking days last month Substitute other means for soctahzmg Eat before drinking Avoid drinking with those pressuring Substitute other means for a “high”
0.02 0.03 0.04 0.05
6.44 4.24 3.52 3.17
.0117 .0403 .0616 .0760
Number of drinks per occasion Avoid drmkmg with those pressunng Confine drinking to certain times Set time limits before another Eat before drinking Refuse unwanted drinks
0.04 0.07 0.08 0.09 0.11
12.61 10.19 4.50 2.88 6.24
.0004 .OQ16 .0346 .0907 .0130
Total Alcohol Problems Score Refuse unwanted drmks Confine drinking to certain times Set time limits before another Reward myself for limiting Avoid drinking with those pressuring Eat before drinking Use body sensations to slow down
0.02 0.06 0.08 0.10 0.11 0.12 0.13
6.37 14.37 5.02 7.98 3.98 3.27 3.90
.0121 .0002 .0258 .0050 .0467 .0715 .0490
Improvement in limiting drinking Confine drmkmg to certam times Set time limits before another Avoid drinking with heavy drinkers Punish myself for not limiting
0.06 0.07 0.09 0.10
20.39 3.36 4.99 4.23
.OOOl .0675 .0262 .0405
Improvement in reducing drinking problems Confme drmkme to certain times Punish myself fir not limiting Set time limits before another Don’t drive when drinking Eat before drinking
0.07 0.08 0.09 0.09 0.10
22.74 2.81 2.67 2.11 2.35
.OOOl .0948
Effectiveness of behavioral strategies to limit drinking Refuse unwanted drmks Set time limits before another Avoid drinking with heavy drinkers Confine drinking to certain times Drink less when driving
0.26 0.33 0.35 0.37 0.37
107.73 29.89 10.85 5.33 3.55
.OOOl .OOOl .OOll .0216 .0604
Effectiveness of behavioral strategies to prevent problems Refuse unwanted drinks Avoid drinking with heavy drinkers Select drinks I drink slowly Confine drinking to certain times Use body sensations to slow down
0.19 0.25 0.27 0.29 0.30
70.16 24.67 8.27 6.35 3.85
BOO1 .OOOl .0043 .0122 .0507
p value
.1028 .1473 .I259
continually employing self-control experiencing more success in limiting alcohol use. What factors might have accounted for both low and high self-control individuals reporting comparable drinking-related data? Recent discussion on the role of behavioral models in explaining adherence to health behavior change (Cameron & Best, 1987; Leventhal & Cameron, 1987; Marlatt, 1988) suggest that a whole host of variables may play a role in the initiation, compliance, and maintenance of behavior, in addition to the specific operations for self-control which this study examined. For example, factors such as cognitive processes like
Behavioral
self-control
127
self-efficacy and outcome expectancies, social influence variables such as social support for behavioral change, and specific stages of habit acquisition and change, may have served as mediators of the alcohol measures in this study. The indication for alcohol abuse prevention/intervention programming for college-aged populations is that these efforts must be broad-based and target multiple alcohol-related factors including cognitive, social, individual, biological, and environmental variables in order to offer the best chance of modifying alcohol consumption and problems. Increasing use of self-control procedures was related to greater reported improvement in limiting drinking and reducing alcohol problems, with those individuals seldom practicing self-control reporting the least improvements. Similarly, greater self-control was linked to greater perceived effectiveness of behavioral strategies to limit drinking and prevent alcohol problems. These data further support the notion that behavioral self-control strategies may play a significant role in limiting alcohol consumption among certain college students. Three of the four health beliefs studied were also found to be related to the degree of self-control applied, with those individuals often practicing self-control reporting the strongest alcohol-related health beliefs. In particular, high self-control individuals were more likely to believe that: (a) a serious problem could result from drinking; (b) they could develop an alcohol-related problem; and (c) there are many benefits to limiting the amount of alcohol used. These data are in accordance with the view points of some researchers (Becker, 1974; Janz & Becker, 1984) who have stated that health beliefs are important determinants of health practices and should be utilized in health planning. As an example, needs assessments of college student alcohol-related health beliefs could be collected to target specific needs which might in turn increase self-control behavior. Various self-control strategies were also found to significantly predict alcohol-related variables. Strategies related to confining drinking to certain times of the week and avoiding drinking with those who pressure one to drink were the best predictors of alcohol consumption, problems, and improvement in limiting drinking. Strategies concerning refusing unwanted drinks, setting time limits before having another drink, and avoiding drinking with heavy drinkers were the most powerful predictors of perceived effectiveness of behavioral strategies to limit drinking and prevent alcohol problems. These aforementioned strategies should be given priority in any future testing of the potential role of behavioral self-control in preventing or initiating alcohol use among college students. Since this study presented a cross-sectional analysis examining behavioral self-control strategies and alcohol-related variables, the direction of the relationships cannot be definitively stated. Longitudinal studies are needed to clarify specific directional relationships between self-control and alcohol consumption measures. Continued developmental studies examining self-initiated self-change events may eventually lead to the development of a refined understanding of behavioral change processes, and the development of more effective alcohol interventions. The present study has laid the groundwork for the development and testing of alcohol abuse interventions for college students grounded within a self-control model.
REFERENCES Baer, P.E., Foreyt, J.P., & Wright, S. (1977). Self-directed termination of excessive cigarette use among untreated smokers. Journal of Behavioral Therapy and Experimental Psychiatry, 8, 71-74. Becker, M.H. (Ed.). (1974). The health belief model and personal health behavior. Thorofare, NJ: Charles B. Slack, Inc. Cameron, R., & Best, J.A. (1987). Promoting adherence to health behavior change interventions: Recent findings from behavioral research. Patient Education and Counseling, 10,139-154.
128
CHUDLEY
E. WERCH
Curry, S.G., & Marlatt. G.A. (1985). Unaided quitters’ strategies for coping with temptations to smoke. In S. Shiffman & T.A. Wills (Eds.), Coping and subsfance use (pp. 243-265). New York: Academic Press. DiClemente, C.C., & Prochaska, J.O. (1982). Self-change and therapy change of smoking behavior: A comparison of processes of change in cessation and maintenance. Addictive Behaviors, 7, 133-142. Dropkin, D. (1984). Smokers, stopped smokers, and mulridimensional health locus ofcontrol. Unpublished doctoral dissertation, Hofstra University, Hempstead, NY (University Microfilms International No. 84- 18831). Janz. N.K., & Becker, M.A. (1984). The health belief model: A decade later. Healrh Education Quarrerly, 11, 147. Leventhal, H., & Cameron, L. (1987). Behavioral theories and the problem of compliance. Patient Education and Counseling, 10, 117-138. Marlatt, G.A. (1988). Research on behavioral strategies for the prevention of alcohol problems. Conremporar~ Drug Problems, 15, 3145. Miller, W.R., & Munoz, R.F. (1982). How to control your drinking. Albuquerque, NM: University of New Mexico Press. Pederson, L.L., & Lefcoe, N.M. (1976). Psychological and behavioral comparison of ex-smokers and smokers. Journal
of Chronic
Diseases,
29, 431434.
Peni,
M.G. (1985). Self-change strategies for the control of smoking, obesity, and problem drinking. In S. Shiffman & T.A. Wills (Ms.), Coping and substance use (pp. 295-318). New York: Academic Press. Petri, M.G., & Richards, G.S. (1977). An investigation of naturally occuring episodes of self-controlled behaviors. Journal
of Consulting
Psychology,
24,
178-183.
Petri, M.G., Richards, S.C., & Schulteis, K.R. (1977). Behavioral self-control and smoking reduction: A study of self-initiated attempts to reduce smoking. Behavior Therapy, 8, 360-365. Prochaska, J.O., & DiClemente, C.C. (1983). Staes and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390-395. Rosen, T.J.. & Shipley, R.H. (1983). A state analysis of self-initiated smoking reductions. Addictive Behaviors, 8. 263-272.
Vogler, R.E., & Bartz, W.R. (1982). The better way to drink. New York: Simon & Schuster. Watts, P.R., & Kunkle, R.B. (1985). Going it alone: Cigarette smokers who quit smoking by themselves. Sigma Gamman,
The Era
17, 10-12.
Werch, C.E., & Gorman, D.R. (1986). Factor analysis of internal and external self-control practices for alcohol consumption. Psychological Reports, 59, 1207-1213. Werch, C.W., & Gorman, D.R. (1988). Relationship between self-control and alcohol consumption patterns and problems of college students. Journal of Srudies on Alcohol, 49, 30-37.
1990 Copyright
0306-460300 $3.00 + .OO Q 1990 Pergamon Press plc
BEHAVIORAL SELF-CONTROL STRATEGIES FOR DELIBERATELY LIMITING DRINKING AMONG COLLEGE STUDENTS CHUDLEY University
E. WERCH of Arkansas
All&net - This study examined the use of behavioral self-control strategies, across specific procedures and sex, and the relationship between self-control and alcohol consumption and problem variables. A total of 456 randomly selected college students participated in a survey of campus substance use. Behavioral strategies most commonly used to deliberately limit drinking among college students were those related to limiting driving/riding when drinking. controlling time and food stimuli, and awareness of internal and external cues to control drinking. Strategies related to self-reinforcement and punishment, certain alternatives to alcohol use, and specific rate control techniques were least used by drinkers. Females were more likely than males to use all but one of the behavioral strategies. The degree of self-control practiced was associated with alcohol consumption, driving/riding after drinking, improvement in limiting drinking, health beliefs related to alcohol problems, and perceived effectiveness of behavioral strategies to limit drinking. Specific self-control strategies were found to significantly predict alcohol-related variables, with confining drinking to certain times of the week and refusing unwanted drinks the most consistent predictors.
Various studies have been conducted examining self-initiated attempts to control smoking behavior (Baer, Foreyt, & Wright, 1977; Curry & Marlatt, 1985; DiClemente & Prochaska, 1982; Dropkin, 1984; Pederson & Lefcoe, 1976; Petri & Richards, 1977; Perri, Richards, & Schulteis, 1977; Prochaska & DiClemente, 1983; Rosen & Shipley, 1983; Watts & Kunkle, 1985), with a fewer number of recent studies assessing self-change attempts to limit alcohol use (Perri, 1985; Werch & Gorman, 1988). These studies are based upon two assumptions. First, that a considerable number of people appear to be successful in controlling and quitting addictive substances on their own, even in the face of evidence that formalized treatment programs commonly result in modest success. Second, that studying naturally occurring episodes of self-change can contribute to our understanding of change processes which in turn can stimulate the development of hypotheses leading to advanced theory and clinical applications. In one of the few recent studies of self-initiated attempts to control alcohol use, Perri (1985) examined the self-change efforts of adults coping with problem drinking over the course of a 6-8 month period. He found that successful drinkers, defined as those who were abstinent or who were not engaged in problem drinking, applied a wider array of behavioral self-control strategies in their coping efforts than did unsuccessful drinkers. In particular, successful drinkers were more likely to implement stimulus control procedures and use alternative behaviors to take the place of drinking, compared with unsuccessful drinkers. Werch and Gorman (1988) examined the relations between self-control strategies, alcohol problems, and alcohol consumption patterns of college students. They found that external (e.g., number of drinks consumed) and internal (e.g., physiological sensations) self-control strategies significantly correlated with alcohol problems, and the quantity and frequency of alcohol consumption. Students experiencing alcohol problems at least once during the past
Current affiliation: University of North Florida. Requests for reprints should be sent to Chudley E. Werch, Ph.D., Assistant Professor, University of North Florida College of Health, Center for Alcohol and Drug Studies, 4567 St. Johns Bluff Road, Jacksonville, PL 32216. 119
120
CHUDLEY
E. WERCH
year were as likely to employ self-control strategies as nonproblem drinkers with two exceptions: (a) problem drinkers were more likely to apply self-reinforcement and punishment strategies, and (b) problem drinkers were more likely to practice strategies to control time and food factors related to drinking. In addition, this study found that self-control strategies were applied more frequently across greater alcohol consumption patterns with moderate drinkers typically practicing the greatest amount and variety of strategies. Research to date on self-change approaches to alcohol use has been scarce, particularly when compared to that conducted on smoking cessation. The present study answered the following three questions: (a) Are there differences as to how frequently behavioral self-control strategies are employed across specific strategies and sex? (b) What is the relationship between the degree of self-initiated self-control and measures of alcohol consumption, alcohol-related problems, alcohol beliefs, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? and (c) Which specific behavioral self-control strategies have the greatest impact on alcohol consumption, alcohol-related problems, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? METHOD
Subjects Subjects were randomly sampled from all those enrolled at a southern university by generating a computer listing of 800 student names with addresses. Seventeen of these students could not be located. A total of 456 college students participated in this study by completing a mailed survey of their alcohol and other drug use. Subject demographic data indicated that they were equally distributed on sex with 5 1 .O% males. The mean age was 25 years (SD = 8.51). The majority of the students were single (67.0%) with 27.0% married,and 5.4% divorced. Students were enrolled in Business or other majors (34.8%), Arts and Sciences (31.7%), Education (16.3%), Engineering (14.3%), and Law (2.8%). Most were classified as Seniors (23.5%), then Graduate or other classification (22.8%), Freshman (20.2%), Juniors (18.7%), and Sophomores (14.8%). The majority were White (89.3%) followed by Oriental (6.1%), and Black (3.9%). Religious preferences included Protestants allowing alcohol use (38.1%), Protestants not allowng alcohol use (27.5%), none or other religion (22.0%), Catholic (12.2%), and Jewish (0.2%). The majority of students resided in rented apartments or houses (48.5%) followed by campus residence (22.3%), other residences (13.1%), Fraternity or Sorority housing (9.4%), and at home with parents (6.8%). Of the total sample, 79.9% drank alcohol beverages at least once a year or more. The mean number of drinking days in the last month was 4.86 (SD = 6.46), and the average number of drinks consumed equaled 3 .Ol (SD = 3.73). Procedure Randomly selected students were mailed a copy of the Student Substance Use Survey and an accompanying cover letter describing the study. A follow-up letter and a second copy of the survey were sent to those who did not respond to the earlier request. A third letter was sent to those subjects who did not respond to the two previous requests. The Student Substance Use Survey was a 6-page instrument designed to determine college student demographics, alcohol consumption patterns, frequency of alcohol-related problems, ratings of improvement concerning alcohol use, alcohol-related health beliefs, frequency of use of behavioral self-control strategies to limit drinking, ratings of effectiveness of behavioral strategies to limit alcohol use, and other drug-related measures. The survey was pretested on a sample of university students and revised to eliminate ambiguous items.
Behavioralself-control
121
Corrected split-half reliability determined from a sample of college students was found to be high, r = .85. Information on alcohol consumption consisted of items to determine the exact number of days drinking in the last month and the exact number of drinks usually consumed, with one drink being equal to one 12-0~ beer, one 4-0~ table wine, or one 1-oz 86-proof liquor. An additional item asked how often subjects drank alcohol (beer, wine or liquor), using a five-point response of: (a) never drank or used to drink but quit, (b) once or more a year, but less than once a month, (c) once or more a month, but less than once a week, (d) once or more a week, but less than once a day, (e) every day. Items measuring the frequency of alcohol problems asked how often students experienced each of 11 alcohol-related events over the past six months. These items were rated on a five-point scale of “ 1: Never” to “5: Always” and included the following events related to drinking: interpersonal conflicts, legal problems, health problems, financial problems, illness events, school conflicts, sexual problems, aggression, feelings of depression, fatigue or tiredness, and work problems. Ratings of improvement concerning alcohol use included two items. One item asked how much improvement subjects experienced in limiting their drinking over the past six months. The other item asked how much improvement was experienced in reducing problems due to drinking over the last six months. Both items were scored on a five-point scale of “ 1: None at all or never drank” to “5: Complete improvement.” Information on alcohol-related health beliefs was determined from four items representing the four psychological variables providing constructs for the Health Belief Model (HBM) (Becker, 1974). These four health belief variables, found useful in predicting health behavior, included perceived susceptibility, severity, benetits, and barriers related to alcohol use behavior and problems. Each of these four items was scored on a five-point scale of “ 1: Strongly agree” to “5: Strongly disagree.” The frequency of use of behavioral self-control strategies was determined by asking students how often they used certain strategies to deliberately limit their drinking, over the past six months. Each of these items were scored on a five-point scale from “1: Never” to “5: always.” Behavioral strategies were selected from a previous study (Werch & Gorman, 1986) of a factor analysis of 37 external self-control behaviors, developed originally from self-help program manuals (Miller & Munoz, 1982; Vogler & Bartz, 1982). For the purpose of survey brevity, the present study employed 14 of the original 37 strategies by selecting the two highest loadng strategies from each of the seven factors identified in the original analysis. These seven factors were identified as: (a) Rate Control items pertaining to controlling drinking rate and setting drinking limits; (b) Self-reinforcement and Punishment - items describing rewards and punishment for attempts at maintaining drinking limits; (c) Alternatives - items related to substituting alternative means of coping in place of drinking; (d) Avoidance - items concerning avoidance techniques specifically intended either to limit alcohol consumption or unwanted consequences; (e) Limiting Driving and Cash - items related to controlling driving and money when drinking; (f) Controlling Time and Food-items concerning limiting drinking to certain times of the day and week as well as eating before drinking; and (g) Awareness - items related to controlling the number of drinks by an awareness of internal (physiological) or external (number of drinks) factors. Test-retest reliability for the original 37 behavioral strategies has been found to be very high (r = .96) based on results from a group of uni.versity students (Werch & Gorman, 1986). Ratings of the effectiveness of behavioral strategies to limit alcohol use consisted of two items. One item asked how effective the above metnioned strategies were for helping students to limit their drinking over the last six months. The second item asked how effective
122
CHUDLEY
Table 1. Percentage
E. WERCH
of males and females frequency self-control strategies
of use of specific behavioral
Frequency of use of strategies Seldom Sometimes Often
Strategy/Sex
Never
Select drinks I drink slowly** Males Females
48.43 25.00
10.69 12.20
18.24 28.05
11.32 16.46
11.32 18.29
Set time limits before another* Males Females
63.52 48.15
16.35 14.81
10.06 16.67
3.77 11.11
6.29 9.26
Reward myself for limiting Males Females
79.87 70.00
11.95 13.13
5.03 10.63
1.26 3.13
1.89 3.13
Punish myself for not limiting Males Females
83.65 80.00
11.32 8.75
2.52 8.13
1.89 1.25
0.63 1.88
Substitute other menas for a “high” Males Females
70.44 65.82
4.40 10.13
8.81 11.39
11.32 7.59
5.03 5.06
Substitute other means for socializing Males Females
40.88 31.93
14.47 12.05
24.53 27.11
12.58 18.67
7.55 10.24
Avoid drinking with heavy drinkers* Males Females
43.48 26.95
11.80 11.98
16.15 19.16
18.63 22.16
9.94 19.76
Avoid drinking with those pressuring* Males Females
48.13 30.25
9.38 8.64
9.38 18.52
15.00 14.81
18.13 27.78
Drink less when driving Males Females
16.35 11.95
1.89 2.52
8.18 7.55
22.64 15.72
50.94 62.26
Don’t drive when drinking Males Females
18.24 10.37
5.66 3.05
11.32 14.02
28.30 27.44
36.48 45.12
Eat before drinking Males Females
11.25 9.15
5.63 1.83
22.50 22.56
32.50 36.59
28.13 29.88
Confine drinking Males Females
to certain times 15.00 13.50
9.38 6.13
10.00 15.34
36.25 35.58
29.38 29.45
Refuse unwanted Males Females
drinks 15.63 8.43
1.25 3.01
6.25 9.64
30.63 25.90
46.25 53.01
21.38 13.58
6.29 6.17
19.50 16.05
27.67 25.93
25.16 38.27
Use body sensations Males Females
Always
to slow down
*p 5 .Ol. **p 5 ,001.
these strategies were for preventing problems due to drinking over the past six months. Both items were scored on a five-point scale of “ 1: Not at all or never used” to “5: Extremely effective. ”
Behavioral self-control
123
RESULTS
The first question raised in this study was: Are there differences as to how frequently behavioral self-control strategies are employed across specific strategies and sex? As Table 1 shows, the answer is yes. With regard to specific strategies, those which were used (at least sometimes) by the greatest percentage of drinkers included: (a) eating before drinking, 86%, (b) refusing unwanted drinks, 85%; (c) drink less when driving, 83%; (d) arrange not to drive when drinking, 81%; (e) confine drinking to certain times of the week, 78%; and (f) use body sensations as cues to slow down drinking, 76%. Those strategies which were used by the least proportion of drinkers included: (a) punish oneself for failing to limit drinking, 8%; (b) reward oneself for limiting drinking, 12%; (c) substituting other means for achieving a “high”, 24%; and (d) setting time limits before having another drink, 28%. Differences were also found across sex, with a greater proportion of females than males employing all but one of the strategies studied (i.e., substitute other means for achieving a “high”). A significantly greater proportion of females used the following four strategies: (a) select drinks which are drunk more slowly, x2 = 19.80, 4 df, p = .OOl; (b) set time limits before having another drink, x2 = 12.82, 4 df, p = .Ol; (c) avoid drinking with heavy drinkers, x * = 12.60, 4 df, p = .Ol; and (d) avoid drinking with those who pressure one to drink, x * = 14.70, 4 df, p = .005. The second question raised in this study was: What is the relationship between the degree of self-initiated self-control and the following alcohol variables: alcohol consumption, alcohol-related problems, alcohol beliefs, improvement in drinking, and effectiveness of behavioral strategies to improve drinking? For purposes of analysis, all drinkers were categorized as practicing either a low, moderate, or high degree of self-control. The degree of self-control was based upon the frequency of employing self-control strategies, totaled across the 14 behavioral self-control items. Self-control categories and collateral scores included: (a) low (seldom) control 5 28; (b) moderate (sometimes) control 2 29 and < 42; and (c) high (often) control L 43. Table 2 shows that nearly all alcohol-related variables were related to the degree of self-control (ANOVA tests, ps 5 .OS), with the exception of total alcohol problems and the health belief associated with overcoming barriers to limiting drinking. Alcohol consumption measures, including the number of drinking days last month and number of drinks per drinking occasion, differed significantly across self-control. The greatest mean alcohol measures were reported by those in the moderate control category, with no pairwise differences between low and high control subjects. Although no differences were found across self-control for total alcohol problems, the measures related to driving after drinking and riding with a drinking driver did differ significantly across self-control. As with alcohol consumption, measures of drinking and driving/riding were greatest for the moderate control subjects. Results for measures of perceived improvement in drinking were somewhat different. The greatest mean scores for improvement in drinking occurred among the high control subjects, although no pairwise differences were found between the moderate and high control subjects. Three of the four health beliefs were found to differ significantly across self-control, with the belief related to overcoming barriers to successfully limit drinking not associated with self-control. Beliefs related to the perceived susceptibility and severity of alcohol problems, and the benefits of limiting alcohol consumption, were most agreed to by the subjects in the high control category. Lastly, both measures of the effectiveness of behavioral strategies to limit alcohol use differed significantly across control, with the perceived effectiveness increasing across greater control efforts. The third question raised in this study was: Which specific behavioral self-control
CHUDLEYE. WERCH
124
Table 2. Mean scores of alcohol-relatedvariables by degree of behavioral self-control
LOW
Degree of self-control Moderate High
control N = 59
control N = 142
F
P
M
M
M
value
value
5.19
1.89
5.31
6.25
.0022
Number of drinks per occasion
3.62
4.52
3.19
4.00
.0191
Total alcohol problems score
Alcohol
variables Number of drinking last month
control N=l33
days
14.77
15.80
14.90
1.45
.2351
Frequency of driving after drinking
1.17
2.16
1.66
10.65
.ooOl
Frequency of riding with a drinking driver
2.06
2.32
1.83
11.11
.OOOl
Improvement drinking
1.66
2.46
2.51
10.19
.OOOl
Improvement in reducing drinking problems
1.50
2.30
2.38
9.47
.OOOl
Belief that serious problems can result from drinking
1.72
1.68
1.46
3.38
.0351
Belief that I could develop alcohol-related problems
2.91
2.19
2.50
2.94
.0543
Belief that there are benefits from limiting drinking
1.96
1.75
1.48
6.70
.0014
Belief that I can learn to limit my drinking
1.63
1.60
1.48
1.29
.2757
Effectiveness of behavioral strategies to limit drinking
2.15
3.34
3.87
51.09
.OOOl
Effectiveness of behavioral strategies to prevent problems
2.00
3.24
3.78
40.94
.OOOl
are significantly C .05).
different
in limiting
Mean scores connected with an underline (least-squares means pairwise comparisons
from one another
strategies have the greatest impact on the alcohol variables studied? Table 3 shows the summary of stepwise multiple regression analyses on the major alcohol-related variables. Two behavioral strategies were found to significantly predict the number of drinking days in the last month, and four strategies predicted the number of drinks per drinking occasion. Substituting other means for socializing was the most significant predictor of the number of drinking days. Avoiding drinking with those who pressure, confining drinking to certain times of the week, and refusing unwanted drinks were the best predictors of the number of drinks consumed. Six strategies were found to significantly predict total alcohol problems,
Behavioral self-control
125
with the most powerful predictors being: confining drinking to certain times of the week, rewarding oneself for limiting drinking, and refusing unwanted drinks. Three strategies significantly predicted improvement in limiting drinking and one strategy predicted improvement in reducing alcohol problems. For both measures of improvement, the behavioral strategy of confining drinking to certain times of the week was the most significant predictor. Lastly, four strategies significantly predicted the effectiveness of behavioral strategies to limit drinkingand five predicted the effectiveness of strategies to prevent alcohol problems. The best predictors of effectiveness of strategies to limit drinking were: refusing unwanted drinks, setting time limits before having another drink, and avoiding drinking with heavy drinkers. The most significant predictors of the effectiveness of strategies to prevent alcohol problems were: refusing unwanted drinks, avoiding drinking with heavy drinkers, and selecting drinks which are drunk slowly. DISCUSSION
This study found that behavioral self-control strategies are deliberately employed with varying levels of frequency, across specific strategies and sex, by college students. The degree of self-control practiced by subjects was associated with alcohol consumption, driving and riding after drinking, improvement in limiting drinking, health beliefs related to alcohol problems, and perceived effectiveness of behavioral strategies to limit drinking. Specific self-control strategies were found to significantly predict measures of alcohol consumption, problems, improvement, and effectiveness of behavioral strategies. Confining drinking to certain times of the week was the strategy most consistently predicting measures of alcohol use and problems, while refusing unwanted drinks was the strategy best predicting the perceived effectiveness of behavioral strategies to limit drinking. A surprisingly large percentage of college students was found using those behavioral self-control strategies related to limiting driving/riding when drinking, controlling time and food stimuli, and awareness of internal and external cues to control drinking. These data support previous research (Werch & Gorman, 1988) suggesting that college students deliberately employ behavioral strategies to limit alcohol consumption. Strategies related to self-reinforcement and punishment, certain alternatives to alcohol use (i.e., substituting other means for achieving a “high”), and specific rate control techniques (i.e., setting time limits before having another drink), however, were used by very few subjects. Females were found to be more likely to use behavioral self-control strategies than men, particularly those strategies related to rate control and avoidance. These data have implications for the development and testing of programs for alcohol abuse prevention and intervention, grounded within a self-control model. For example, innovative alcohol abuse interventions emphasizing education and training in those self-control strategies which are least likely to be employed without external intervention could be tested for their effects, particularly with college-aged males. This study also found that the degree of self-control practiced was associated with a host of alcohol-related variables. The number of drinking days per month, the number of drinks per drinking occasion, and the frequency of driving and riding after drinking were related to self-control. Specifically, those individuals who seldom or often employed self-control strategies generally reported significantly lower alcohol consumption and drinking-driving events than those who sometimes used self-cntrol strategies. Although self-control was not significantly associated with alcohol problems, there was also a trend for individuals who seldom or often used self-control to report experiencing fewer alcohol problems. These data suggest that the relationship between self-control and alcohol consumption and problem measures is nonlinear, with those students using few self-control strategies, and those
126
CHUDLEY
Table 3. Summary
E. WERCH
of stepwise multiple regression analyses of alcohol-related behavioral self-control strategies
Alcohol variables/ behavioral strategy
variables by
R2
F value
Number of drinking days last month Substitute other means for soctahzmg Eat before drinking Avoid drinking with those pressuring Substitute other means for a “high”
0.02 0.03 0.04 0.05
6.44 4.24 3.52 3.17
.0117 .0403 .0616 .0760
Number of drinks per occasion Avoid drmkmg with those pressunng Confine drinking to certain times Set time limits before another Eat before drinking Refuse unwanted drinks
0.04 0.07 0.08 0.09 0.11
12.61 10.19 4.50 2.88 6.24
.0004 .OQ16 .0346 .0907 .0130
Total Alcohol Problems Score Refuse unwanted drmks Confine drinking to certain times Set time limits before another Reward myself for limiting Avoid drinking with those pressuring Eat before drinking Use body sensations to slow down
0.02 0.06 0.08 0.10 0.11 0.12 0.13
6.37 14.37 5.02 7.98 3.98 3.27 3.90
.0121 .0002 .0258 .0050 .0467 .0715 .0490
Improvement in limiting drinking Confine drmkmg to certam times Set time limits before another Avoid drinking with heavy drinkers Punish myself for not limiting
0.06 0.07 0.09 0.10
20.39 3.36 4.99 4.23
.OOOl .0675 .0262 .0405
Improvement in reducing drinking problems Confme drmkme to certain times Punish myself fir not limiting Set time limits before another Don’t drive when drinking Eat before drinking
0.07 0.08 0.09 0.09 0.10
22.74 2.81 2.67 2.11 2.35
.OOOl .0948
Effectiveness of behavioral strategies to limit drinking Refuse unwanted drmks Set time limits before another Avoid drinking with heavy drinkers Confine drinking to certain times Drink less when driving
0.26 0.33 0.35 0.37 0.37
107.73 29.89 10.85 5.33 3.55
.OOOl .OOOl .OOll .0216 .0604
Effectiveness of behavioral strategies to prevent problems Refuse unwanted drinks Avoid drinking with heavy drinkers Select drinks I drink slowly Confine drinking to certain times Use body sensations to slow down
0.19 0.25 0.27 0.29 0.30
70.16 24.67 8.27 6.35 3.85
BOO1 .OOOl .0043 .0122 .0507
p value
.1028 .1473 .I259
continually employing self-control experiencing more success in limiting alcohol use. What factors might have accounted for both low and high self-control individuals reporting comparable drinking-related data? Recent discussion on the role of behavioral models in explaining adherence to health behavior change (Cameron & Best, 1987; Leventhal & Cameron, 1987; Marlatt, 1988) suggest that a whole host of variables may play a role in the initiation, compliance, and maintenance of behavior, in addition to the specific operations for self-control which this study examined. For example, factors such as cognitive processes like
Behavioral
self-control
127
self-efficacy and outcome expectancies, social influence variables such as social support for behavioral change, and specific stages of habit acquisition and change, may have served as mediators of the alcohol measures in this study. The indication for alcohol abuse prevention/intervention programming for college-aged populations is that these efforts must be broad-based and target multiple alcohol-related factors including cognitive, social, individual, biological, and environmental variables in order to offer the best chance of modifying alcohol consumption and problems. Increasing use of self-control procedures was related to greater reported improvement in limiting drinking and reducing alcohol problems, with those individuals seldom practicing self-control reporting the least improvements. Similarly, greater self-control was linked to greater perceived effectiveness of behavioral strategies to limit drinking and prevent alcohol problems. These data further support the notion that behavioral self-control strategies may play a significant role in limiting alcohol consumption among certain college students. Three of the four health beliefs studied were also found to be related to the degree of self-control applied, with those individuals often practicing self-control reporting the strongest alcohol-related health beliefs. In particular, high self-control individuals were more likely to believe that: (a) a serious problem could result from drinking; (b) they could develop an alcohol-related problem; and (c) there are many benefits to limiting the amount of alcohol used. These data are in accordance with the view points of some researchers (Becker, 1974; Janz & Becker, 1984) who have stated that health beliefs are important determinants of health practices and should be utilized in health planning. As an example, needs assessments of college student alcohol-related health beliefs could be collected to target specific needs which might in turn increase self-control behavior. Various self-control strategies were also found to significantly predict alcohol-related variables. Strategies related to confining drinking to certain times of the week and avoiding drinking with those who pressure one to drink were the best predictors of alcohol consumption, problems, and improvement in limiting drinking. Strategies concerning refusing unwanted drinks, setting time limits before having another drink, and avoiding drinking with heavy drinkers were the most powerful predictors of perceived effectiveness of behavioral strategies to limit drinking and prevent alcohol problems. These aforementioned strategies should be given priority in any future testing of the potential role of behavioral self-control in preventing or initiating alcohol use among college students. Since this study presented a cross-sectional analysis examining behavioral self-control strategies and alcohol-related variables, the direction of the relationships cannot be definitively stated. Longitudinal studies are needed to clarify specific directional relationships between self-control and alcohol consumption measures. Continued developmental studies examining self-initiated self-change events may eventually lead to the development of a refined understanding of behavioral change processes, and the development of more effective alcohol interventions. The present study has laid the groundwork for the development and testing of alcohol abuse interventions for college students grounded within a self-control model.
REFERENCES Baer, P.E., Foreyt, J.P., & Wright, S. (1977). Self-directed termination of excessive cigarette use among untreated smokers. Journal of Behavioral Therapy and Experimental Psychiatry, 8, 71-74. Becker, M.H. (Ed.). (1974). The health belief model and personal health behavior. Thorofare, NJ: Charles B. Slack, Inc. Cameron, R., & Best, J.A. (1987). Promoting adherence to health behavior change interventions: Recent findings from behavioral research. Patient Education and Counseling, 10,139-154.
128
CHUDLEY
E. WERCH
Curry, S.G., & Marlatt. G.A. (1985). Unaided quitters’ strategies for coping with temptations to smoke. In S. Shiffman & T.A. Wills (Eds.), Coping and subsfance use (pp. 243-265). New York: Academic Press. DiClemente, C.C., & Prochaska, J.O. (1982). Self-change and therapy change of smoking behavior: A comparison of processes of change in cessation and maintenance. Addictive Behaviors, 7, 133-142. Dropkin, D. (1984). Smokers, stopped smokers, and mulridimensional health locus ofcontrol. Unpublished doctoral dissertation, Hofstra University, Hempstead, NY (University Microfilms International No. 84- 18831). Janz. N.K., & Becker, M.A. (1984). The health belief model: A decade later. Healrh Education Quarrerly, 11, 147. Leventhal, H., & Cameron, L. (1987). Behavioral theories and the problem of compliance. Patient Education and Counseling, 10, 117-138. Marlatt, G.A. (1988). Research on behavioral strategies for the prevention of alcohol problems. Conremporar~ Drug Problems, 15, 3145. Miller, W.R., & Munoz, R.F. (1982). How to control your drinking. Albuquerque, NM: University of New Mexico Press. Pederson, L.L., & Lefcoe, N.M. (1976). Psychological and behavioral comparison of ex-smokers and smokers. Journal
of Chronic
Diseases,
29, 431434.
Peni,
M.G. (1985). Self-change strategies for the control of smoking, obesity, and problem drinking. In S. Shiffman & T.A. Wills (Ms.), Coping and substance use (pp. 295-318). New York: Academic Press. Petri, M.G., & Richards, G.S. (1977). An investigation of naturally occuring episodes of self-controlled behaviors. Journal
of Consulting
Psychology,
24,
178-183.
Petri, M.G., Richards, S.C., & Schulteis, K.R. (1977). Behavioral self-control and smoking reduction: A study of self-initiated attempts to reduce smoking. Behavior Therapy, 8, 360-365. Prochaska, J.O., & DiClemente, C.C. (1983). Staes and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390-395. Rosen, T.J.. & Shipley, R.H. (1983). A state analysis of self-initiated smoking reductions. Addictive Behaviors, 8. 263-272.
Vogler, R.E., & Bartz, W.R. (1982). The better way to drink. New York: Simon & Schuster. Watts, P.R., & Kunkle, R.B. (1985). Going it alone: Cigarette smokers who quit smoking by themselves. Sigma Gamman,
The Era
17, 10-12.
Werch, C.E., & Gorman, D.R. (1986). Factor analysis of internal and external self-control practices for alcohol consumption. Psychological Reports, 59, 1207-1213. Werch, C.W., & Gorman, D.R. (1988). Relationship between self-control and alcohol consumption patterns and problems of college students. Journal of Srudies on Alcohol, 49, 30-37.
