The Potential for Alcohol Abuse among First Year College Students

Marissa F. Heern

Abstract

Alcohol use and abuse is a continual battle in the college age population across the United States. This study explored which subgroups have the highest alcohol abuse potential. Participants (n=54) were first year students enrolled at a small Midwestern liberal arts college. A survey was compiled comprising of revised versions of the Type A Self-Rating Inventory (TASRI; Blumenthal, 1985), Your Personality � Is it Friend or Foe? (Forbes, 1979), Michigan Alcohol Screening Test (MAST; Selzer, 1971), Alcohol Use Disorders Identification Test (AUDIT; World Health Organization, 1992), and the CAGE questionnaire (Mayfield, McLeod, & Hall, 1974). Surveys were administered in 2 study skills classes, 2 basic first year English classes, and the first year honors English class. A score was comprised of all three alcohol measures to determine each participant�s abuse potential and then statistically compared to their personality and which class they were enrolled in. Males significantly reported a higher potential for alcohol abuse than females on the alcohol composite measure as well as the individual measures of the MAST and AUDIT. A significant difference was found between males and females on the Type A subscale. The alcohol measures were all significantly correlated with one another. Furthermore, low self-concept and introvert scores were significantly related. These findings are inconclusive as to whether an alcoholic personality exists. If high risk college students can be identified, then a preventive program can be used to address this issue.

            The 1980�s produced a plethora of research regarding Type A behavior patterns (TABP) with focus on its association with coronary heart disease. In the 1990�s some research began to appear regarding TABP and how it may be a link to an alcoholic personality. This research was not as common and still has many gaps that deserve attention. Another major area of research has been to look for the alcoholic personality in children of alcoholics. However, not as much research has been conducted with a general sample. Much research has already been devoted to alcohol consumption among the undergraduate population. However, again this research includes gaps.

            If an alcoholic personality could be found in the undergraduate population, a preventive program could possibly be developed to address this issue. Also, there are a few trial programs being started at the high school level around the country that are based on surveys used to help identify at-risk students. This idea could be used at the undergraduate collegiate level and may be another by product of findings of this study.

            Thus far, there have only been small amounts, if any, of research specifically looking at an undergraduate population and examining their Type A/Type B behavior pattern, high or low self-concept, and introvert/extrovert status. The current research also takes into account what level of class they are enrolled in as a first year student. High or low self concept is generally defined as how a person view�s oneself. This could include feelings of worth, beauty, or ability. Introverts are individuals that are not very outgoing; often they are shy or reserved. Extroverts on the other hand are individuals that are more interested in the outside world and their environment as opposed to an inward concentration on themselves.

            Type A behavior pattern is most often characterized by traits such as aggressiveness, competitiveness, lack of patience, muscle tension, accelerated rate of speech, time urgency, and criticalness of self (Day & Jreige, 2002, 109). Malatesta-Magai, Jonas, Shepard, and Culver (1992) further report that traits such as being hard-driving and hostile are often present in individuals categorized as Type A (551). Type B behavior pattern, on the other hand, can be thought of as having very few, if any of the characteristics of the Type A individual.

            The Jenkins Activity Survey (JAS) is one of the most common measures to assess TABP characteristics. Spence, Helmreich, and Pred (1987) developed two scales from the JAS based on psychometric analyses that they labeled Achievement Strivings (AS) and Impatience-Irritability (II). They believe that the Type A behavior pattern is too general of a term and should actually be broken down into components as they have suggested with the AS and II scales. Their findings suggested the two scales affected different aspects of a person�s life. For example, the AS component was found more often in Type A individuals who had high motivation. Furthermore in Spence, Helmreich, and Pred�s study, AS scores were significantly positively correlated with GPA; whereas, a II scores were significantly positively correlated with frequency of health complaints such as problems with headaches, sleep, digestion, and respiratory system.

            In 1989, Spence, Pred, and Helmreich published a continuation of this study where they specifically examined students� GPAs, SAT scores, and scores on the AS and II scales. Student GPA�s and SAT scores were significantly correlated, as well as AS scores and GPA�s. However, AS scores and SAT scores were not significantly correlated. Thus, the achievement striving aspect of the TABP appears to be a predictor of success at the undergraduate level.

            If Spence, Pred, and Helmreich�s findings have validity, then it could be assumed that individuals who are Type A would be more likely to be found in an Honors classroom at the undergraduate level as opposed to a Study Skills class. Furthermore, it could even be assumed that students found in a Study Skills class are more likely to demonstrate characteristics of Type B personality.

            Ward and Eisler (1987) found that individuals who exhibit TABP set too high goals for themselves; thus, they have a harder time reaching the goals. Due to this situation, they then suffered negative effects such as psychological distress associated with failure of reaching the goal they had set. This appears to support Spence, Pred, and Helmreich�s finding of two components to the Type A behavior. However, instead of the two components that go side-by- side, the two components could have a causal relationship, meaning that an individual may be in the Achievement Striving stage or scale at first, but then due to failing at a goal, falls more into the Impatience-Irritability scale.

            The existence of two components of TABP has also been investigated by Wright, Newman, Meyer, and May (1993). They explored the difference in how TABP manifested itself in women and men. There was no significant difference in scores on the JAS between male and female university faculty members. However, there was a significant difference on the hostility scale. Although, since hostility is normally thought of as a characteristic of Type A behavior it would be expected that statistical difference between males and females for the hostility and JAS scores would be similar. However, since they were not it can be inferred that �women may achieve their high Type A rating on the basis of nonhostile traits (e.g. time urgency or job involvement, etc.)� (p. 499).   

            Furthermore, Kopper (1993) found that �women obtained higher scores on indirect hostility, irritability, and dependency, and men obtained higher scores on assault and aggressiveness� (p. 232). Kopper�s study found no difference in how males and females express or suppress anger. It should be noted that Type A individuals scored higher in the areas of irritability, suspicion, guilt, and hostility. When this study is taken into account with the previous study, a conclusion may not be inferred because women score higher in one study and men scored higher in the other study on hostility.

            McCann, Woolfolk, Lehrer, and Schwarcz found varying results when they tried to break apart the TABP. They found that women scored higher on the subscales of Irritability, Guilt and Resentment. Men scored higher on Assaultiveness. Women demonstrated �higher correlations between Type A and the Guilt subscale, and men between Type A and the Suspiciousness subscale� (p. 355).

            Riska and Ettorre (1999) reported that the results of their qualitative study suggested that men �could use alcohol to restore their image of traditional masculinity and regain a sense of masculine dominance and control� (p. 760). McCreary, Newcomb, and Sadava (1999) reported in their study that males who hold traditional gender role ideas may negatively affect their health due to drinking more than their male counterparts who do not hold these gender role ideas. However, they did report finding that masculine gender role stress was not related to alcohol consumption as they previously thought.    

            Hussong (2003) provided a more in-depth look at alcohol use and abuse in regards to personality. Hussong references Martin and Sher (1994) and Trull and Sher (1994) stating that �extraversion did not differentiate individuals with versus without an alcoholism diagnosis� (p.142). Hussong also reports in her literature review that extraverted people were often encouraged by their peer group or social affiliations to consume alcohol. Her results suggest that young adults� alcohol use (not their problem use) was influenced by their own drinking motives as well as the motives of their friends which is supported by other studies� findings as well.

            Ferguson (2001) reported that neurotic-introverts tended to engage in inhibitory behaviors when stressed. These inhibitory behaviors included some alcohol and drug use. Ferguson also reported finding that those higher on the psychotisicm scale of the Eysenck model, were more likely to use alcohol. This is in agreement with previous research (p. 321).

            Loukas, Krull, Chassin, and Carle (2000) found that �neither young adult reports nor parental reports of extraversion significantly mediated the relationship between parental alcoholism and young adult diagnosis. Parental alcoholism was not predictive of extraversion nor was extraversion predictive of young adult alcohol diagnosis� (pp. 1163, 1165).       

            Kahler, Read, Wood, and Palfai (2003) explained that the male gender has always been thought of the gender that consumes more alcohol in college. They reported that basic knowledge such as thinking about the difference in body weight and metabolic ability explains why males are sometimes able to consume more alcohol. However, Kahler, Read, Wood, and Palfai continue by saying that males more often than females belong to an all male peer group which normally is more accepting of the alcohol consumption. Replicating numerous other studies, they found that male gender had a strong positive association with alcohol use. However, the results suggested that a male�s social environment choice has nothing to do with his alcohol consumption as was also found in Keefe and Newcomb�s 1996 publication (as cited in Kahler, Read, Wood, and Palfai, 2003).

            Friedman and Rosenman (1959) and Rosenman and Friedman (1961) (as cited in Folsom, Hughes, Buehler, Mittelmark, Jacobs, and Grimm, 1985) reported that men and women who were Type A individuals �drank more alcohol than their Type B counterparts� (p. 228). Koskenvuo and associates (1981; as cited in Folsom, Hughes, Buehler, Mittelmark, Jacobs, and Grimm, 1985) reported that Type A males and females did not drink wine or beer more frequently than Type B�s, but Type A�s did drink spirits more frequently. Folsom and associates examined in their study whether participants at high risk for coronary heart disease who were Type A�s drank more than Type B�s and then defined �drank� more clearly. They differentiated whether drank was frequency, quantity, or both. Their results indicated that Type A men drank more in the coronary heart disease prone population. Furthermore, the results indicate that they drank more due to more frequent drinking rather than drinks at each sitting. In addition, a hypothesis was made that since alcohol is known to cause hostility which is a trait of Type A behavior, it is possible that drinking causes Type A behavior rather than Type A behavior causing drinking.

            Loukas, Krull, Chassin, and Carle (2000) report that the analysis of their data suggests that individuals who are low in agreeableness predict a higher rate of alcohol use and abuse (p. 1169). Individuals low in agreeableness are characterized by hostility, aggression, self-centeredness, and indifference to other. Many would say that this is close to the definition of TABP. Thus, more support that those with characteristics of TABP may be at a higher risk for alcohol abuse.   

            Tomchin and Callahan (1996) explored and reported on the relationship between self-concept and coping strategies in a gifted adolescent sample. Participants were asked to rate how often they used a variety of coping methods. On a 1 (doesn�t apply or I don�t do it)-5 (used a great deal) scale, the sample only used negative tension reduction a mean of 2.13 which was next to least in the survey. Negative tension reduction is defined as activities such as alcohol and tobacco use. Tomchin and Callahan suggest that �Their advanced abilities may aid gifted adolescent in selecting from a repertoire of strategies that involve changing behaviors and beliefs to deal with troubling or stressful situations� (p. 26). Frydenberg (1993) on the other hand reports that gifted adolescents were more likely than their peers to use negative tension reduction strategies (as cited in Tomchin & Callahan). Terman (1921; as cited in Bronzaft and Hayes, 1983) reported that at a 35 year follow up with gifted children they had a lower rate of alcoholism than the general population. It should be noted that other studies using Terman�s research (Tomlinson-Keasey & Warren, 1987; Tomlinson-Keasey, Warren, & Elliott, 1986; as cited in Gust-Brey and Cross, 1999) identified alcohol abuse as one of seven factors used to predict suicide. However, due to the time the data was collected validity is questioned for the present time period.

            An inference could be made connecting the TABP expected in an Honors undergraduate with higher rate of alcohol use. However, other factors could come into play such as the individuals in the Honors class learning how to control the TABP so that it does not become excessive and negative. If this is the case, there would be no difference in what is expected regarding alcohol abuse potential between the Honors, Study Skills, and English 111 classes.    

            A great deal has been detailed thus far regarding Type A behavior pattern, class levels, and alcohol use and abuse. What needs to be explicated now is how these variables are measured. In this study, preexisting measures will mostly be used including the Your Personality � Is it Friend or Foe? (Forbes, 1979), the Type A Self-Rating Inventory (Blumenthal, Herman, O�Toole, Haney, Williams, and Barefoot, 1985), the Michigan Alcohol Screening Test (MAST; Selzer, 1971; as cited in Teitelbaum & Carey, 2000), Alcohol Use Disorders Identification Test (AUDIT; World Health Organization, 1992; as cited in Ash, 1999), and the CAGE questionnaire (Mayfield, McLeod, and Hall, 1974; as cited in Teitelbaum & Carey, 2000). A brief demographics questionnaire was also developed to gather information regarding to gender, age, academic major(s), ethnicity, academic year by credits, and intercollegiate athletic participation.

The TASRI measure consists of 38 characteristics that could be present in any person. Participants answer based on how true the characteristics are of them on a Likert-type scale of 1 (Almost Never)-7 (Almost always). In the present study, only 28 of the characteristics are included due to only 28 of the characteristics being used for scoring. A cutoff score for Type A or Type B behavior has not been established in the undergraduate population. Thus, the mean and standard deviation of the present results will be used to determine Type A or Type B classification.

The TASRI was compared with the Structured Interview (SI) and the JAS, both common Type A measures. To examinee how individual scores compared to the SI, Blumenthal, Herman, O�Toole, Haney, Williams, and Barefoot (1985) detail how they grouped a pilot test sample scores into quintiles and compared to the interviewer ratings from the SI. �The two behavior pattern classification systems were significantly related with a p value of less than 0.006� (p. 268). Sixty percent of the individuals classified as Type A according to the TASRI scored in the highest two quintiles compared to only 15% of the participants who were labeled Type B. When compared with the JAS, there was also a significant relationship �(r=0.58, p<0.0001)� (p.269). Blumenthal and his associates did begin to norm this measure with undergraduate students in their laboratory setting, but only had 96 participants. They did find a correlation of �0.47 (p<0.0001) with the student version of the JAS� (p. 270). One week test-retest reliability was found to be 0.88.

The TASRI was examined by Yarnold and Bryant (1994) attempting to develop a measurement model for the inventory. They found two subscales that seemed to comprise most of the measure�s common variance, hard-driving and extroverted. Furthermore, results indicated that women scored higher on the extrovert scale and men had a greater mean score on the hard-driving factor. Yarnold and Bryant also report that in the TABP literature it has been said that introverted Type As and Bs �are more illness prone than As and Bs who are extroverted and more socially gregarious� (p. 109). Thus, Yarnold and Bryant suggest that adding a measure of extrovert/introvert to a Type A measure may give more valid results concerning risk for health problems such as CHD.  

The CAGE questionnaire is a four question measure used to screen for alcohol problems in a primary care setting. �CAGE is a brief, verbally administered, easily remembered, and simple to score (each yes answer =1) (Knight, Goodman, Pulerwitz, & Durant, 2000, p. 949). According to Teitelbaum and Carey (2000), the CAGE �focuses on the behavioral consequences of alcohol abuse. Total score on the CAGE ranges from 0 to 4� (p. 402). Watson, Detra, Fox, Ewing, Gearhart, & DeMotts (1995), provide information on the four questions. The four questions examine recent cutting down on drinking, annoyance or anger resulting from criticism, feeling guilty about drinking, and if any eye opener behaviors are present such as needing a drink first thing in the morning so one may function.

Teitelbaum and Carey (2000) found that the �test-retest reliability coefficient� for the CAGE was r=.80� in the psychiatric sample and after adjusting for item length �retest reliability estimates of the CAGE increased from .80 to .96� (p. 402). Knight, Goodman, Pulerwitz, and Durant (2000) adapted the CAGE for adolescents. Their adapted version produced �high 1-week test-retest reliabilities (range r = .82-.90)� (p. 948). The adapted CAGE also had �acceptable internal consistency (Alpha=.60)� (Knight, et al., p. 948). According to Hays, Merz, and Nicholas (1995), the original CAGE questionnaire yielded an Alpha of .69 for internal consistency. This internal consistency is not great. However, there are only four questions so it is considered acceptable reliability.

Validity of the CAGE is straight forward. There is no question what the doctor or counselor wants to know when the CAGE is administered; thus, it has high face validity. However, the CAGE answers are �clearly measures of alcoholism and easily faked� especially due to social desirability (Watson, Detra, Fox, Ewing, Gearhart, & DeMotts, 1995, p. 677). Also, another negative aspect of the CAGE being used alone is that the CAGE deals �specifically with experiences in the past year (Watson, et al., 1995, p.683). One other important fact to note, Aertgeerts, Buntnix, Feverym, and Ansoms (2000) found no significant difference in results based on if the CAGE was administered orally or in writing. Reid, Tinetti, O�Connor, Kosten, and Concato (2003) found and recommended that the CAGE is good at recognizing adults with problematic alcohol use over the course of their life; however, may also fail to identify older adults with higher levels of cumulative alcohol use. Thus, the CAGE needs to be used in conjunction with other measures for older adults.

According to Teitelbaum and Carey (2000), the Michigan Alcohol Screening Test (MAST) is a self report measure that is composed of 25 questions. The scores can range from 0-53 and as assumed, a higher score means more probability of a problem with alcohol. Watson, Detra, Fox, Ewing, Gearhart, and DeMotts (2000) report that the 25 items �describe a history of alcohol-related problems. The items were weighted according to their abilities to differentiate alcohol treatment patients from controls� (p. 676). Sixty-six percent of the items comprising the MAST are phrased to cover the respondent�s entire lifetime. Hays, Merz, and Nicholas (1995) adapted the MAST and made a 13 item short version. The responses are all yes/no and equally weighted. Any response of �3 or greater is the clinical cut point for this [short version] measure� (Hays, et al., p. 277). Thus, answering six or more yes�s raises a red flag for an alcohol use problem on the original measure.

 Teitelbaum and Carey (2000) report that past research has produced test-retest coefficients in a psychiatric population of .97 for a one day retest period and .94 for a three day retest period. In their study, they found a test-retest reliability for a seven day period of �r=.95 for the MAST� (p. 402). Murdoch (2001) reports a 4.8 month test-retest reliability of .84. Hays et al. (1995) found the internal consistency reliability in their study to be r=.84 for the Short MAST. Murdoch reports that Hedlung and Vieweg reviewed six studies and found internal consistencies for the MAST to be �range from .83 to .95� (p. 753).                   

Similar to the CAGE, the MAST has high face validity and may also be easily faked. Due to the type of questions asked, the MAST and Short MAST are best suited to assess long-term drinking patterns (Hays et al., 1995). Validity for the MAST most likely lower due to there being more questions and thus more of a chance that questions are not worded properly to remain valid. The MAST appears to have high concurrent validity with the General Alcoholism Factor of the Alcohol Use Inventory (r=.83) (Skinner, 1979 as cited in Murdoch, 2001). Also, there is reasonable validity with the MacAndrew Alcoholism Scale (r=.31 to .46) (Friedrich & Loftsgard 1978a, b as cited in Murdoch). Murdoch emphasizes that the MAST does produce false positives and thus must be used as a screening tool along with other measures.

The Alcohol Use Disorders International Test, also known as the Alcohol Use Disorders Identification Test (AUDIT), is a 10 item self report measure published by the World Health Organization. It is the first measure of its kind to be developed internationally in six countries including �Australia, Bulgaria, Kenya, Mexico, Norway, [and] USA� (Ash, 1999, p.4). Hays, Merz, and Nicholas (1995) state that the ten items have �varying numbers of response options per item, ranging from 3 to 6. Responses are differentially weighted such that between 0 and 4 points are possible per item. A score of 8 or greater is suggestive of alcohol problems� (p. 277). Burke and Schneider (1999) report that scores on the AUDIT range from 0-40. According to the U.S. Preventive Services Task Force (USPSTF) (1996), the AUDIT integrates questions regarding the amount of alcohol drank, frequency of drinking, binging behaviors, and questions about consequences such as if anyone has been injured due to your drinking behaviors. The USPSTF, also cautions that a false-positive score may be given. Numerous college students receive higher than a 8; however, only 38% over an 8 met DSM criteria for abuse or dependence. Ash further reports that

the differences between the AUDIT and most other existing questionnaires include the following:  (a) it tries to identify problem drinkers at the �less severe end of the spectrum� rather than those with established dependence or alcoholism; (b) it emphasizes hazardous consumption and frequency of intoxication rather than drinking behavior itself and its consequences; and (c) it refers to alcohol experiences in the past years as well as over the patient�s lifetime, improving relevance to current drinking status, and it does not require the test-taker to identify himself or herself as a problem drinker. (p.5)

Hays et al. (1995) report the internal consistency reliability of the AUDIT as r=.83. In a different study, the one week test-retest reliabilities were examined for taking the AUDIT by means of paper and pencil, on the internet, and on the internet while taking a break in the middle for 1-48 hours. The paper and pencil reliability coefficient was r=.92, the web based coefficient was r=.92, and the web based with interruption coefficient was r=.95 (Miller et al., 2002). No significant differences were found based on how the AUDIT was administered.

            Validity statistics are harder to find for the AUDIT; however, the USPSTF (1996) reports that the AUDIT had high specificity (94%) and high sensitivity (92%) in the population that was tested. Validity for �current abuse/dependence were high (92% and 96%, respectively)� (USPSTF, par. 11) in an inner-city clinic population. However, among rural outpatients, it was worse with 61% sensitivity and 90% specificity (USPSTF). Reviews in the Mental Measurements Yearbook report that the instrument has high validity but does not specifically detail what the validity is.

            The present study specifically examined certain types of behavior characteristics, or personality traits, of a sample to try and determine if an alcoholic personality exists in an undergraduate population. Specifically, type A and Type B behavior, introvert and extrovert, and high and low self concept were examined. In addition, a comparison for the potential of alcohol abuse was made between Study Skills classes, English 111 classes, and an Honors English 111 class. Type A and Type B personality were determined by use of two measures the Type A Self-Rating Inventory (Blumenthal, Herman, O�Toole, Haney, Williams, and Barefoot, 1985) and Your Personality � Is it Friend or Foe? Questionnaire. The latter measure also determined if a participant should be placed in the introvert or extrovert group, as well as the high or low self concept. Finally, three self report screening measures were administered to participants to determine what their current level of alcohol abuse potential was.

            Based off of the previous research, the following hypotheses were devised:

            1)  First year students enrolled in Honors English, English 111, and Study Skills classes will show no significant difference in their alcohol abuse potential as they self-report drinking characteristics on the Michigan Alcohol Screening Test (MAST), Alcohol Use Disorders International Test (AUDIT), and the CAGE questionnaire.

            2)  If a student reports Type A characteristics as measured by the Type A Self-Rating Inventory and the Your Personality � Is it Friend or Foe? Type A subscale, then the student will report a higher alcohol abuse potential measured by the MAST, AUDIT, and CAGE,  than participants with Type B characteristics.

            3)  If participants are male, they will self-report higher rates of alcohol abuse potential as measured by the MAST, AUDIT, and CAGE than female participants.

            4)  Introverted participants with a low self-concept as measured by the Your Personality � Is it Friend or Foe?  questionnaire will have higher rates of alcohol abuse potential as measured by the MAST, AUDIT, and CAGE than high self-concept extroverts.

            5)  Students enrolled in the Honors English course at McKendree College will more often rate themselves Type A personalities on the Type A Self-Rating Inventory and the Your Personality � Is it Friend or Foe? Type A subscale whereas students in the Study Skills class will be more likely to rate themselves as Type B�s.

  Method

Participants

            The participants were undergraduate students enrolled in a small Midwestern private liberal arts college. Participants were recruited based on their enrollment in freshman English and study skills classes. Twenty-two percent of the participants were enrolled in a Honors English course (n=12), 32% in study skills course (n=17), and 46% were enrolled in English 111 courses (n=25). Fifty-four percent were male (n=29), 42% were female (n=23), and 4% no response (n=2). The mean age for all respondents was 18.54 years (SD = 0.64). Fifty-three percent were white (n=45), 11% black (n=6), 2% Cherokee (n=1), and 4% no response (n=2). One hundred percent of the sample were freshman students (n=54).

Measures

            Type A Self-Rating Inventory (TASRI; Blumenthal, Herman, O�Tooley, Haney, Williams, Jr., & Barefoot, 1985.)  The original measure consists of 38 characteristics that could be present in any person. Participants answer based on how true the characteristics are of them on a Likert-type scale of 1 (Almost never)-7 (Almost always). The TASRI was adapted for this study to consist of only 28 characteristics of a person (See Appendix A). The ten characteristics that are not used to compute the score were deleted from the originally published questionnaire to shorten the survey in order to achieve higher validity. Scores are based on only 28 of the characteristics. The cutoffs were determined based on the sample�s mean and standard deviation due to norms not being established for the undergraduate population (See Tables 1 & 2).

            Your Personality � Is it Friend or Foe? (Forbes, 1979). The original measure consists of four subscales. The measure was adapted to use three in the present study (See Appendix B):  Type A, Introvert, and Low Self-Concept. The participant was asked to self-report on a series of statements on a scale of 1(never)-5(always). These scales do not include specific cutoffs to determine the behavior being measured; thus, means and standard deviations were used to determine classification of traits (See Tables 1 & 2).

               Alcohol Use Disorders Identification Test (AUDIT; World Health Organization, 1992; as cited in Ash, 1999). The Alcohol Use Disorders International Test, also known as the Alcohol Use Disorders Identification Test (AUDIT), is a 10 item self report measure published by the World Health Organization. Scores may range from 0-40 and a score of 8 or more is indicative of possible alcohol abuse problems. The measure has questions ranging from amount of alcohol consumed to injuries due to alcohol use (See Table 1).

            Michigan Alcohol Screening Test (MAST; Selzer, 1971; as cited in Teitelbaum & Carey, 2000). The MAST consists of 25 yes or no forced choice questions that focus more on a history of alcohol use and abuse rather than current use status. All yes answers are worth one point and any score of 6 or greater is considered clinically significant (See Table 1).

            CAGE questionnaire (Mayfield, McLeod, and Hall, 1974; as cited in Teitelbaum & Carey, 2000). The CAGE measure consists of four questions that simply ask about cutting down, annoyance at those who voice concerns, feeling guilt, and needing an eye-opener. Each yes answer is scored as 1 point and a score of 2 or more is clinically significant (See Table 1).

            Alcohol Abuse Potential Scale. The Alcohol Abuse Potential Scale is a combination of the MAST, CAGE, and AUDIT (See Appendix C). Duplicate questions were deleted to produce a 32 questions measure used in the present study. Scoring was conducted by scoring the individual measure per the publishing instructions. The three scores were then summed for the alcohol potential scale (See Table 1).

Table 1

Descriptive Statistics for Measures

Table 2

Classification of Scores

Procedure

            The before mentioned measures were combined into a single questionnaire. A demographics section was added to gain data regarding gender, age, major, ethnicity, and intercollegiate athletic participation. The adapted Your Personality measure was placed in the order of survey question sections. Next, the TASRI was inserted using the adapted number of characteristics. Finally, the Alcohol Abuse Potential Scale was placed last due to the high face validity of the questions.

            Surveys were administered during class time in two study skills classes, two basic first year English classes, and the first year honors English class. Participants were verbally instructed that all answers would remain confidential and they were free to spread out to protect their response from classmates. Also, they were reminded that they had the right to not answer a question(s) if they chose to do so. Finally, they were instructed to place completed questionnaires in a folder placed at the front of the room. This collection procedure was to ensure anonymity by preventing the research from knowing the order in which the surveys were collected. Data was analyzed using SPSS 11.0, t-tests, ANOVA, correlations, and chi-square all with a significance level of .05.

Results

Primary Analyses

            First, results for hypothesis #1 were computed by use of an ANOVA. The results were not significant (F=.396, p=.675). T-tests were also run to individually examine the relationship between the course enrollment groups (Figure 1):  Honors vs. English 111 (t=-.027, p=.979), Study Skills vs. Honors (t=-.647, p=.523), Study Skills vs. English 111 (t=.852, p=.399).

Figure 1. Mean alcohol potential scores for the three courses sampled:  Honors (n=12), Study Skills (n=16), and English 111 (n=25). No significant differences were found; however, the difference between Study Skills and English 111 were the closest to being a significant difference.

            Second, the relationship between Type A and B behavior pattern and alcohol potential score were examined. A correlation was run to determine concurrent validity between the two Type A measures. Due to a low concurrent validity (r=.145, p=.322) the Type A measures were examined independently. In the TASRI, a non-significant relationship was found between individuals classified as Type A and Type B on the alcohol potential scale (t=.953, p=.363; TABP (n=7; M=9.43), TBBP (n=5; M=5.20))  On the Type A subscale a still non-significant relationship was found (t=.451, p=.659; TABP (n=6; 15.83), TBBP (n=9; 12.89)) (See Figure 2).

Figure 2. Mean alcohol abuse potential scores for Type A, Type B, and neither behavior pattern according to the Type A subscale and the TASRI.

            Third, males and females were compared on their alcohol potential scores ((t=2.443, p=.018) (See Figure 3).

Figure 3. Mean alcohol abuse potential scores for male (n=29) and female (n= 23) participants.   

            Fourth, due to sample size restraints, low self concept and introverts were examined independently of one another. High self-concept in comparison to low self-concept showed no significant difference (t=-1.163, p=.135) in their alcohol abuse potential scores. Introverts in comparison with extraverts also did not show a significant difference in alcohol abuse potential scores (t=-1.012, p=.326) (See Figure 4).

Figure 4. Alcohol abuse potential scores are shown based on the divisions of extravert(n=11)  and introvert (n=8) and low (n=8) and high (n=9) self-concept.

            Fifth, no significant difference was shown between Study Skills classes and the Honors class in relation to their Type A scores ( TASRI t=-.538, p=.596; Type A subscale (t=.642, p=.527) (See Figure 5). A high score indicates reporting of Type A behavior.

Figure 5. Mean Type A subscale and TASRI totals for courses surveyed:  Honors (n=12 ), Study Skills (n=15), English 111 (n=24).

Ancillary Analyses

            First, reliabilities were run on all scored scales to ensure valid results (See Table 3).

Table 3

Reliabilities of Scored Scales

            Second, alcohol potential scores were examined between individuals classified as neither Type A nor Type B and compared with Type B�s scores. On the TASRI there was a significant difference (t=1.809, p=.078). A significant difference was also found using the Type A subscale (t=-1.966, p=.073) (See Figure 2).

            Third, males and females significantly differed on several factors in this study. On the Type A subscale (t=-2.825, p=.007) females had a mean of 22.22 and males averaged 19.63. On the AUDIT (t=2.832, p=.007), males had a mean of 8.00 whereas females had a mean of 3.22. On the MAST (t=2.006, p=.050), males had a mean of 2.50 and females only 1.30.

 Figure 6. Significant gender differences on the MAST, AUDIT, and Type A subscales.

            Fourth, several significant correlations were found. There was significant correlation between low self-concept scores and introverted behavior (r=.382, p=.005). There were also very significant correlations between all the alcohol measure used:  CAGE and AUDIT (r=.754, p=.001), AUDIT and MAST (r=.740, p=.001), MAST and CAGE (r=.768, p=.001) (See Table 4).

Table 4

Significant Intercorrelations Between Measures            

*Correlation is significant at the 0.01 level.

Discussion

            Results support two of the five hypotheses:  males will report higher alcohol potential scores than females and there will be no difference in alcohol potential scores dependent on course enrollment. College aged males reporting higher levels of alcohol use and abuse has been supported in a majority of previous research. Yet, as Kahler, Read, Wood, and Palfai (2003) suggested, a male�s ability to metabolize alcohol should be accounted for when comparing drinking behaviors between genders. In addition, research has not specifically focused on comparing study skills, basic English, and honors English courses. However, research regarding gifted students use of alcohol conflicts. Tomchin and Callahan (1996) report that negative tension reduction strategies including alcohol are rarely used. However, Frydenberg (1993; as cited in Tomchin and Callahan) disagrees and reports that gifted students are more likely to use tension reduction strategies than their peers.

            Two of the hypotheses were not clinically significant at the .05 level; however, upon examination of their means, significance could be possible if a larger sample size had been used. In previous research there has been conflicting results as to whether Type A or Type B individuals drink more than the other. However, it appears as if in each of the previous studies, it was either Type A or Type B that drank more, they never drank the same amount which would indicate that a larger sample would be more likely to produce significance with this study. Results were in the prediction predicted with Type A individuals reporting higher alcohol abuse potential scores; however, they results were not significant. Furthermore, previous research also conflicts with whether introverts or extraverts have more problems with alcohol use and abuse. The previous research suggests that the reason for the extravert or introvert drinking needs to be taken into account when comparing the two. Extraverts reported higher alcohol abuse potential scores which was not in the direction predicted; however, originally introvert and low self-concept were to be examined but due to sample size, results were not obtainable.       

            The last hypothesis of more students in Honors being Type A than in Study Skills was not supported. This lack of support could be due to a design flaw in the experiment. Classification of Type A and Type B were based on the means and standard deviations of the individual scales. Thus, there is a fairly equal distribution of Type A and Type B individuals in addition to a small sample size. Therefore, a statistical difference may not be possible to find using the current method of classifying Type A and Type B individuals.       

            The biggest limitation in this study was the sample size. First year students were chosen as the sample due to both convenience and that they are normally �traditional� students entering college for the first time directly out of high school. This prevented extraneous variables such as age and �real world experiences� from interfering with the data. The small sample size limited both the strength of the analyses run as well as preventing definite implications due to many inconclusive results.

            Sample size was too small to determine if there is any predictive validity for problems related to alcohol use and abuse. Evidence suggests that personality measures such as  the TASRI should be investigated in the future as a measure to predict alcohol abuse problems in an undergraduate population. These results are inconclusive as to whether an alcoholic personality exists.
                       

References

Aertgeerts, B., Buntinx, F., Fevery, J., & Ansoms, S. (2000). Is there a difference between CAGE interviews and written CAGE questionnaires?  Alcoholism:  Clinical and Experimental Research, 24, 733-736.

Ash, P. (1999). Review of the Alcohol Use Disorders Identification Test. In Plake, B.S. & Impara, J.C. (Eds.), The Supplement to the thirteen mental measurement yearbook, pp. 4-5. Lincoln, NE:  Buros Institute of Mental Measurement.

Blumenthal, J. A., Herman, S., O�Toole, L. C., Haney, T. L., Williams, Jr., R. B., Barefoot, J. C. (1985). Development of a brief self-report measure of the Type A (coronary prone) behavior pattern. Journal of Psychosomatic Research, 29, 265-274.

Bronzaft, A. L. & Hayes, R. F. (1983). Family characteristics and life satisfaction of high academic achievers. Mankind Quarterly, 24, 37-59.

Burke, S. K. & Schneider F. D. (1999, January 15). Alcohol-related problems:  Recognition and intervention. Retrieved July 14, 2004, from American Academy of Family Physicians, http://www.aafp.org/afp/990115ap/361.html.

Day, A. L. & Jreige, S. (2002). Examining Type A behavior pattern to explain the relationship between job stressors and psychosocial outcomes. Journal of Occupational Health Psychology, 7, 109-120.

.

Ferguson, E. (2001). Personality and coping traits:  A joint factor analysis. British Journal of Health Psychology, 6, 311-325.

Folsom, A. R., Hughes, J. R., Buehler, J. F., Mittelmark, M. B., Jacobs, D. R., Jr., & Grimm, R. H., Jr. (1985). Do Type A men drink more frequently than Type B men?:  Findings in the Multiple Risk Factor Intervention Trial (MRFIT). Journal of Behavioral Medicine, 8, 227-235.

Forbes, R. (1979). The stress-prone personality. Life Stress, (pp. 47-48). Garden City, NY: Doubleday. Gust-Brey, K. & Cross, T. (1999). An examination of the literature base on the suicidal behaviors of gifted students. Roeper Review, 22.

Hays, R. D., Merz, J. F., & Nicholas, R. (1995). Response burden, reliability, and validity of the CAGE, Short MAST, and AUDIT alcohol screening measures. Behavior Research Methods, Instruments, and Computers, 27, 277-280.

Kahler, C. W., Read, J. R., Wood, M. D., & Palfai, T. P. (2003). Social environmental selection as a mediator of gender, ethnic, and personality effects on college student drinking. Psychology of Addictive Behaviors, 17, 226-234.

Knight, J. R., Goodman, E., Pulerwitz, T.,&  DuRant, R. H. (2000). Reliabilities of short substance abuse screening tests among adolescent medical patients. Pediatrics, 105, 949- 945.

Kopper, B. A. (1993). Role of gender, sex role identity, and Type A behavior in anger expression and mental health functioning. Journal of Counseling Psychology, 40, 232-237.

Loukas, A., Krull, J. L., Chassin, L., & Carle, A. C. (2000). The relation of personality to alcohol abuse/dependence in a high-risk sample. Journal of Personality, 68, 1153-75.

McCann, B. S., Woolfolk, R. L., Lehrer, P. M., Schwarcz, L. (1987). Gender difference in the relationship between hostility and the Type A Behavior Pattern. Journal of Personality Assessment, 51, 355-366.

McCreary, D. R. & Sadava, S. W. (1998). Stress, drinking, and the adverse consequences of drinking in two samples of young adults. Psychology of Addictive Behaviors, 12, 247-261.

.

Malatesta-Magai, M. Jones, R. Shepard, B., & Culver, L. C. (1992). Type A behavior pattern and emotion expression in younger and older adults. Psychology and Aging, 7, 551-561.  

Miller, E. T., Neal, D. J., Roberts, L. J., Baer, J. S., Cressler, S. O., Metrik, J., & Marlatt, G. A. (2002). Test-retest reliability of alcohol measures:  Is there a difference between internet-based assessment and traditional methods?  Psychology of Addictive Behaviors, 16, 56-63.

Murdoch, J. W. (2001)   Review of the Michigan Alcoholism Screening Test. In Plake, B.S. & Impara, J.C. (Eds.), The fourteenth mental measurement yearbook, pp. 753-754. Lincoln, NE:  Buros Institute of Mental Measurement.

Reid, M. C., Tinetti, M. E., O�Connor, P. G., Kosten, T. R., & Concato, J. (2003). Measuring alcohol consumption among older adults:  A comparison of available methods. The American Journal on Addictions, 12, 211-219.

Riska, E. & Ettorre, E. (1999). Mental distress:  Gender aspects of symptoms and coping. Acta Oncologiica, 38, 757-761.

Spence, J. T., Helmreich, R. L., & Pred, R. S. (1987). Impatience versus achievement strivings in the Type A pattern:  Differential effects on students� health and academic achievement. Journal of Applied Psychology, 72, 522-528.

Spence, J.T., Pred, R.S., & Helmreich, R. L. (1989). Achievement striving, scholastic aptitude, and academic performance:  A follow-up study to �Impatience versus achievement striving in the Type A pattern.�  Journal of Applied Psychology, 74, 176-178.

Teitelbaum,. L. M. & Carey, K. B. (2000). Temporal stability of alcohol screening measure in a psychiatric setting. Psychology of Addictive Behaviors, 14, 401-404.

Tomchin, E. M. & Callahan, C. M. (1996). Coping and self-concept. Journal of Secondary Gifted Education, 8, 16-27.

U.S. Preventive Services Task Force. (1996). Screening for problem drinking. Guide to clinical preventive services, 2^nd ed. Washington, DC: U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion.

Ward, C. H. & Eisler R. M. (1987). Type A behavior, achievement strivings, and a

dysfunctional self-evaluation system. Journal of Personality and Social Psychology, 53, 318-326.

Watson, C. G., Detra, E., Fox, K. L., Ewing, J. W., Gearhart, L. P., & DeMotts, J. R. (1995). Comparative concurrent validities of five alcoholism measure in a psychiatric hospital. Journal of Clinical Psychology, 51, 676-684.

Wright, L., Newman, R. A., Meyer, D. L., & May, K. (1993). Gender difference in negative affectivity among university faculty members. Professional Psychology:  Research and Practice, 24, 497-499.

Yarnold, P. R. & Bryant, F. B. (1994). A measurement model for the Type A Self-Rating Inventory. Journal of Personality Assessment, 62, 102-115.

Appendix A

Please rate yourself as to how you typically feel or react in each of the situations.

1=Never          2=Seldom          3=Sometimes          4=Frequently          5=Always

_____ 7. Meeting new acquaintances is very stressful for me.

_____ 8. My family or friends think I am hard driving and work too hard.

_____ 9. If given the chance, I prefer to work alone rather than in a group.

_____10. When a job or assignment is not clearly laid out for me, I begin to feel anxious.

_____11. A negative evaluation about my work makes me depressed for days.

_____12. I pride myself on accomplishing the most work in my department or class and being the first to meet quotas or get done with a project.

_____13. Having to make business decisions or school decisions such as where to go to grad school is particularly stressful for me.

_____14. My work is less productive when I have to interact with others.

_____15. I usually work with frequent deadlines and time pressures.

_____16. I tend to withdraw from people rather than confront them with problems.

_____17. If one method for getting the job or assignment done works, I am not likely to change it.

_____18. I need the praise of others to feel I am doing a good job.

_____19. Since I do not want to fail, I avoid risks.

_____20. I seldom feel good about myself.

_____21. I personally do not reveal things about myself.

_____22. I tend to become overly cautious and anxious in new situations.

_____23. I have a tendency to produce more and more work in less time.

_____24. If someone criticizes me, I begin to doubt myself.

_____25. I pride myself on being orderly, neat and punctual.

_____26. I do not like to go to parties or places where there is a large number of people.

_____27. I do a great deal of studying in a group setting with friends while hanging out at the same time.

_____28. I become particularly upset if I am contradicted.

Appendix B

Instructions:  You will be shown a number of adjectives in the boxes below. We would like you to use these words to describe yourself by indicating, on a scale of 1 to 7, how true of you these various characteristics are. Please give your own opinion of yourself. If you are not sure, put down the number that comes closest to what you think best describes you. Do not leave any blank spaces if you can avoid it.

   1                     2                           3                       4                        5                      6                    7 

 Almost        Usually not     Infrequently    Occasionally      Often            Usually          Almost 

  never                                                                                                                                         always

Appendix C

Please circle the answer that is correct for you. If you feel uncomfortable answering a certain question, you do not have to answer. However, please remember all answers are confidential.

11. Have you ever felt you should cut down on your drinking?

Yes                  No

12. Have people annoyed you by criticizing your drinking?

Yes                  No