Determination of Alcohol Consumption Motivations in Turkish Sample

Abstract

The aim of the current study is to investigate alcohol motivations that drive Turkish people’s behavior when they consumed alcohol. In order to this, we investigated alcohol consumption intensity, presence of marital status and psychiatric disorders, and gender impact on alcohol motivations. Since role of some influencers on alcohol consumption still remain unknown (i.e, anhedonia), we developed 22 items containing Alcohol Motivation Scale (AMS) by testing internal consistency of questionnaire. Participants answered 5-point Likert scale (1= completely disagree to 5=completely agree) and data collected in response to answers of eligible subjects from the surveymonkey.com link which was spread in the webpages. This study included 142 individuals age ranging between 13 to 62 (46 males, 96 females) from different alcohol use intensity. Principal component analysis demonstrated 5 factors that drive Turkish people’s alcohol consumption behavior: (1) anhedonic depression, (2) self-worth, (3) fun, (4) familial and (5) conformity. Results suggests that although gender, marital status, and presence of psychiatric disease did not affect alcohol motivations, people who have heavy alcohol consumption also have higher AMS scores. In addition, only self-worth dimension significantly differed between males and females. Implications and limitations were discussed. In conclusion, 5 motivations drive Turkish people’s alcohol consumption behavior.

Introduction    

Previous papers indicate that high rates of heavy drinking and alcohol-related drinking problems occurred among university students throughout the world including Turkey (İlhan, Yıldırım, Demirbaş, & Doğan, 2008; Knight et al., 2002). Although this situation is not only limited to sample belonging to university students, a research suggests that overall population including adolescents and adults has been affected by alcohol-related problems (B. F. Grant & Dawson, 1997) indicating the importance of preventive strategies. Therefore, there has been increasing interest in identification of risk factors that motives drinking behavior in individuals regarding to problematic alcohol consumption. Several research investigated alcohol-use motives to understand why individuals consume alcohol (Cooper, 1994; Cooper, Russell, Skinner, & Windle, 1992), but these papers did not include some dimensions that motives drinking behavior and alcohol consumption behavior was only analyzed in specific sample (i.e, only in adolescents) rather than considering a whole population (sample aged over than 18). In addition, no previous paper demonstrated alcohol drinking motivations that play important role in Turkish sample.

According to World Heath Organization (WHO), heavy alcohol consumption is associated the 4% (2,5 million) of overall death. This number is greater than deaths from HIV/AIDS and tuberculosis and alcohol consumption is known to be risk for oral, liver and breast cancer as well as cardiovascular diseases and stroke (Kraus, Bloomfield, Augustin, & Reese, 2000). According to Turkish Statistical Institute (TUİK, 2012) reports, 79.9% of sample over age of 15 never used alcohol, the remaining 9.7% of sample occasionally uses alcohol but 10.4% of sample regularly consumed alcohol. Although, the indicated numbers seem to be relatively small, another research conducted by Turkish Public Health Institution suggest that 13% of participants actively consume alcohol and 7% of sample use alcohol in risky way (THSK, 2013) indicating the importance of the determination of motivational behaviors as well as finding out prevention strategies to overcome relevant health related risk factors mentioned.

Earlier models on alcohol drinking motives included two (social and coping) (Farber, Khavari, & Douglass, 1980) or three factors (social, enhancement and coping) (Cooper et al., 1992). These works were further improved in which four factor model was suggested which categorizing drinking motives into affiliative, conformity, enhancement and coping (Cooper, 1994). More recently, Grant et al. (2007) modified this model into five-factor model by distinguishing drinking to cope with anxiety and drinking to cope with depression (V. V Grant, Stewart, O’Connor, Blackwell, & Conrod, 2007).  However, in coping-depression dimension, they used non-clinical sample and they did not deeply investigated root consequences of alcohol motives. Including depression, the loss of the capacity to feel pleasure and motivational deficits (i.e, anhedonia) are seen in individuals (Loas, 1996), whether the inability to experience pleasure from daily activities motives alcohol consumption behavior or not is still unknown. On the other hand, alcohol use in college students has been associated with lower self-worth (Luhtanen & Crocker, 2005) and the familial risk of alcoholism was previously proven (Finn, Sharkansky, Brandt, & Turcotte, 2000). But according to our up to date knowledge, these factors has not been included into a validated scale which might be playing important role in alcohol consumption motivations.

According to Cooper (1994) social motives drive higher frequency of alcohol consumption whereas conformity (i.e, to fit in with admired group) motives have negative association with quantity and frequency of alcohol consumption. In the present study, including some dimensions that Cooper et al. (1994) published we developed a novel Alcohol Motivations Scale (AMS) considering previous influencers such as familial risk, self-worth and anhedonia that was not previously used in the development of a scale. In order to this, we initially prepared 22 items containing AMS pool with 5-point Likert scoring system (0= Strongly disagree to 5=Strongly agree). Our research questions were as follows: (1) Which motivations that drive alcohol consumption in Turkish population (2) To what degree are alcohol users’ motivations correlate with the alcohol consumption intensity, (3) How gender has an association with the alcohol motivations and (4) How other contributors such as presence of psychiatric disorders, marital status and gender have impact on total alcohol motivations?

We hypothesized that who have greater alcohol consumption will be having more motivational scores. In addition, based on previous work in Turkish survey, males will consume more alcohol than females (Buzrul, 2016). Finally, considering AMS center of our research, we aimed to develop a global scale that would be used to identify potential motivations of global population which will be important for the preventive strategies as well as reducing alcohol-related problems.

Methods

Subjects

Our sample consisted of Turkish people with different characteristics. Total sample size included 177 individuals (N=177) age ranging between 12 to 62 with 56 males and 121 females. After evaluation of alcohol consumption intensity, those who never consume alcohol for excluded from the study and our sample size included total 142 individuals (N=142) age ranging between 13 to 62 with 46 males and 96 females.

Experimental Procedure

We created a survey (www.surveymonkey.com) which including informed consent form at the beginning, demographics form (questions 1 to 5), alcohol consumption intensity (question 6) and Alcohol Motivation Scale (questions 7 to 28). Basically, we spread the link in the facebook pages of different groups as much as we could do (see survey sample at appendix A). Informed consent form comprises ethical statement that demonstrating voluntariness of participants. Therefore, we took consent of volunteers prior to experiment in which subjects confirmed voluntarily participate to the experiment. In addition, a statement indicating that anytime they could be withdrawn from the study were added to this form. Those who have gave their consent included in data analysis.

Research Design

Considering drinking motives, we initially performed a literature review from previous studies (Cooper, 1994; Finn et al., 2000; Luhtanen & Crocker, 2005) and created 22 items by taking into account those influencers that was shown to play an important role alcohol motivations.

Forms and Measures

a. Demographics Form

Demographics representing age, gender, educational and marital status as well as psychiatric disease presence were anonymously asked to participants and data were collected (see question numbers 1-5 in appendix A).

b. Alcohol Consumption Intensity

In order to evaluate alcohol consumption intensity, we asked to respondents a question: “How often do you consume alcohol?” (0= Never use, 1= One-two days per month, 2= One-two days per week and 3= Everyday). This question was used to assess the time spent for the alcohol consumption to measure alcohol consumption intensity in our sample (see question number 6 in Appendix A).

c. Alcohol Motivations Scale (AMS)

AMS 7-28 consisting of 22 questions that measures alcohol motivations which is 5 point Interval type scale (1= Strongly Disagree, 2=Disagree, 3= Neither agree or disagree, 4= Agree and 5=Strongly Agree). Data collected from this scale is used to analyze frequency of response, principal components and reliability statistics (also see appendix A). It is important to note that higher scores represent high motivation for alcohol consumption and no reverse scoring was applied while the scale was generated.

Statistical Analysis

Collected data were analyzed using SPSS program version 21. Histogram analysis, Q-Q plots and Shapiro Wilks test revealed that our data is non-normally distributed. Internal reliability tests indicating cronbach’s α value was evaluated to measure scale reliability. Descriptive analysis to measure frequency of data, KMO and Barlett test of sphericity to evaluate sampling adequacy, and principal component analysis to extract factor loading by items were performed. Multivariate analysis of variance (MANOVA) were used to evaluate gender affect among different motivations and Spearman rho correlation analysis was run to find association between alcohol consumption intensity and motivations that drive alcohol consumption behavior. Non-parametric Mann Whitney U test was used to compare group differences between AMS and gender, marital status and presence of psychiatric disease. p value smaller than .05 was accepted as significant at 95% confidence interval.

Results

Descriptive Statistics on Alcohol Consumption Intensity and Sample Characteristics

Table 1: Frequency and percent showing Alcohol Consumption Intensity

Alcohol consumption intensity frequency table was demonstrated in table 1. From the 19.8%  of respondents (N=35) did not consume alcohol, 48.6% (N=86) of them consume alcohol one-two days per month, 29.9% (N=53) of them consume alcohol one-two days per week and 1.7% of them (N=3) consume alcohol everyday.

Table 2: Frequency and percent showing characteristics of Turkish alcohol consumers that included data analysis

Characteristics of our sample who consumes alcohol was demonstrated in Table 2. Our respondents (N=142) was different educational background with the dominancy of university students (high school (N=8), bachelors (N=104), masters (N=24) and PhD (N=6). In addition, their marital status (N=141 due to missing data) was different (single N=107, married (N=6) and divorced (N=8)) in which majority of them were single. From those (N=142), only 14 of them told that they have current psychiatric disease whereas remaining 128 did not demonstrated psychiatric disease.

Internal Reliability Analysis

In order to test the scale reliability, we evaluated cronbach’s α from the respondents’ data to AMS scale who consumes some level of alcohol (N=142). α value of total 22 item was found as .873 before item deletion. Table 3 represents change in cronbach’s α after item removal. Items reducing cronbach’s α is Q7 and Q11 represented as green color. Since they reduce α values, removal of them increased α value from .873 to .879. However, a previous report suggest that cronbach’s α values between 0.70 to .95 is acceptable to conduct a research (Tavakol & Dennick, 2011). Therefore, in order not to reduce factors that we previously generated for questionnaire, we didn’t remove Q7 and Q11 from our scale, instead, we decided to use 22 items for our further principal component analysis. Although data is not shown here, removal of those questions lead to reduction in factors from 5 to 4 which could also reduce alcohol motivations that we did not want.

Table 3: 22 item containing AMS and cronbach’s α values if item deleted

Principal Component Analysis for AMC Factor Extraction

Factor extractions were done by exposing 22 items to principal component analysis using SPSS program. We analyzed items applying Varimax rotation since unrotated results may create inter-item bias (Osborne, 2015). Sampling adequacy was measured with Keiser-Meyer-Olkin’s (KMO) index and Barlett’s test of sphericity gave strongly acceptable results (KMO:.841 and p=.000). High factor loadings greater than .400 were included to scale, cross-loadings were omitted due to there is not closer loadings among components. Therefore, highest loadings were chosen (represented as yellow in table 4) as motivations and items were group based on this approach.

Results in Table 5 suggests that five motivations drive Turkish people alcohol consumption behavior. We named these motivations based on the item grouping which are: Anhedonic depression, self-worth, fun, familial and conformity.

Table 4: AMS Factor Extraction Model

Table 5: Cronbach’s α values of AMS motivation dimensions

In order to test subscale α values, we evaluated cronbach’s α representing internal reliability of factors. Although conformity motivation seems to be less reliable than other motivations, we have good α values indicating that our subscale has good consistency.

Factor 1: Anhedonic Depression

Items (12, 14, 15, 17, 19 and 24) loading on factor 1 showed that anhedonic depression motivates Turkish people’ alcohol consumption behavior. Anhedonia -a core symptom of depression- has been described as the loss of interest to rewarding and enjoyable activities. In anhedonia, there is motivational and hedonic impairments in individuals leading to a depressive symptomology (Chentsova-Dutton & Hanley, 2010). As indicated in table 6, Turkish sample use alcohol since they cannot enjoy the life, cannot have a pleasure from daily works, to get away from their problems, depressive, bad or irritable moods, and they use alcohol when their motivation dropped. These data indicate that alcohol consumption motivation has negative dimension which may associate with psychopathology.

Table 6: Items, mean and standard deviations for anhedonic depression motivation

Factor 2: Self-Worth

Items (16, 18, 25 and 26) loading on factor 2 showed that self-worth dimension influence on Turkish sample’ alcohol behavior. Self-worth has been defined as the sense of one’s own value as an individual. Therefore, according to table 7, Turkish sample consume alcohol to feel more precious, to be valued by others, to be taken care of. This data indicated that alcohol consumption has been driven by self-worth dimension which may reflects personality of individuals affects alcohol use.

Table 7: Items, mean and standard deviations for self-worth motivation

Factor 3: Fun

Items (7, 10, 11, 21 and 28) loading on factor 3 showed that Turkish sample use alcohol to get more fun. According to results represented in table 8, Turkish sample use alcohol because of they loved and find it fun, and the feeling they got makes them enjoyable and good.

Table 8: Items, mean and standard deviations for fun motivation

Factor 4: Familial                                                     

Items (9, 20, 22 and 27) loading on factor 4 showed that familial factors influence on Turkish sample alcohol consumption behavior.  As represented in table 9, alcohol consumption at home effects their alcohol consumption. Furthermore, they use alcohol since their parent and family members used it.

Table 9:  Items, mean and standard deviations for familial motivation

Factor 5: Conformity

Items (8, 13 and 23) loading on factor 5 showed that Turkish sample use alcohol in order to conform the group. As represented in table 10, friend cycle and groups they involved influence on their alcohol consumption behavior, furthermore, for instance, when they involved a celebrations and invitations, they consume more alcohol since alcohol use of others may influence this behavior.

Table 10: Items, mean and standard deviations for conformity motivation

We later tested the correlation between alcohol usage intensity and the dimensions we found in AMS (anhedonic depression, self-worth, fun, familial and conformity and AMS total scores). Results gathered from table 11 indicate that those who have intense alcohol consumption also have significant positive correlation in anhedonic depression, self-worth, fun and total AMS scores (see table 11). These results show that heavy alcohol users were having more anhedonic depression, more problems in their self-worth and having more fun with alcohol (p<.05, see table 11)

Table 11: Correlation Between Alcohol Usage Intensity and AMS dimensions

Further, we tested whether the found motivations significantly differed between males and females. In order to test our research question “How gender has an association with the alcohol motivations”, we conducted multivariate analysis of variance (MANOVA) test. As indicated table 12, males significantly have higher scores than females in self-worth motivation (p=.003) which indicating that they use more alcohol since they wanted to be valued by others. These results showed the pathologic dimension of alcohol use occurs more in males than females.

Table 12: Mean Differences in AMS dimensions among gender

Total AMS scores did not differ between males and females

In order to analyze alcohol motivation scores between males and females, a bar graph was drawn represented in figure 1. Mann Whitney U test showed that total AMS median ± interquartile range scores in males (N=45) compared to females (N=96) did not significantly differ indicating the similar pattern on alcohol consumption motivation was seen among gender (see figure 1, p=.133).

Figure 1: Scores (median ± interquartile range) of the AMS scale between males (N=45) and females (N=96). Note: n.s=not significant

Total AMS scores did not differ among marital status

In order to analyze alcohol motivation scores among marital status, we compared AMS median ± interquartile range scores in single, married and divorced individuals. Mann Whitney U test showed that no statistical differences between single (N=106) and married (N=26), single (N=107) and divorced (N=8) and married (N=26) and divorced (N=8) (p>0.05 in all groupings, see figure 2) indicating that alcohol motivations are not affected by marital status in our sample.

Figure 2: Median ± interquartile range scores of the AMS scale between single (N=107), married (N=26) and divorced (N=8) individuals. Note: n.s=not significant

Total AMS scores did not differ in the presence of psychiatric disease

We next assessed the differences of alcohol motivation scores in the presence of psychiatric disease and non-psychiatric disease condition. Mann Whitney U test showed that no statistical differences in individuals with psychiatric disease (N=13) and no psychiatric disease (N=128) (p=.056, see figure 3).

Figure 3: Median ± interquartile scores of the AMS scale in the presence of psychiatric disease (N=13) and non-psychiatric disease condition (N=128). Note: n.s: not significant

Discussion

Our paper for the first time analyzes alcohol consumption motivations in Turkish sample, our up to date knowledge. In order to gather deeper knowledge on Turkish people’s alcohol motivation behaviors, we developed 22 items AMS questionnaire and seek answers to following questions: (1) Which motivations that drive alcohol consumption in Turkish population (2) To what degree are alcohol users’ motivations correlate with the alcohol consumption intensity, (3) How gender role has relationship with the alcohol motivations and (4) How other contributors such as presence of psychiatric disorders, marital status and gender have impact on total alcohol motivations?

First of all, we collected data from 177 individuals, of those 142 participants reported that they consume alcohol. We built AMS scale based on answers of those respondents and majority of our sample (48.6%, N=66) reported that they use alcohol one-two days per month, the remaining 29.9% (N=53) reported that they use alcohol one-two days per week, 1.7% (N=3) of them was heavy users and 19.6% (N=35) did not use alcohol. This data indicates that our sample actively uses alcohol (see table 1).  Previous report from Buzrul (2016) propose a data in which 8% of respondents consume alcohol once or less per month, 3% of them 2-4 times per month, and 13% of total respondents report that they actively use alcohol in Turkey. In comparison to Buzrul’s data we have relatively more alcohol users. That is because we shared AMS questionnaire in internet groups in which we believe that they like to consume alcohol. Since we would like to investigate alcohol motivations, we required active alcohol users to investigate dimensions of alcohol consumption. In addition, in parallel to Buzrul’s (2016) results, we found that more males than females use more alcohol in Turkish sample (data not shown). We next assessed the scale reliability by calculating cronbach’s α which represents internal reliability of AMS questionnaire. As we found the cronbach’s α as .873, we further checked α values if item deleted. Deletion of Q7 and Q11 leading to increase in α value to .879 (see table 3) but removal of those questions results in reduction in the factor numbers that we extracted in principal component analysis. Therefore, we decided not to remove this questions from AMS since .873 is quite good value to conduct a research in accordance with the previous report (Tavakol & Dennick, 2011) and we used 22 items for AMS component extraction. We then subjected 22 items to principal component analysis and measured KMO and Barlett’s test of sphericity to understand how sampling adequacy have a fit to our data. As we found quite strong sampling adequacy (p=.000), we next named the components that we extracted in principal component analysis. In response to our first hypothetical research question “(1) Which motivations that drive alcohol consumption in Turkish population “, Turkish sample we used motivated by five dimensions which are: (1) Anhedonic depression, (2) Self-worth, (3) Fun, (4) Familial and (5) Conformity (see table 4).

We later assessed internal reliability of dimensions of AMS. Although conformity dimension seems to have less reliability than other dimensions (see table 5), anhedonic depression, self-worth, fun and familial components had quite strong α values.

Previous reports suggested that moderate to heavy alcohol consumption is leading to several diseases including coronary artery disease (Moore & Pearson, 1986) and physical health problems (Schmidt & Popham, 1975). Therefore, investigating alcohol motivations may further assist us to find out preventive strategies to overcome alcohol-associated problems.

As shown in Table 6, Turkish people were displaying anhedonic depression -motivational and hedonic impairments- that may be leading to depressive symptomology due to alcohol use. They use alcohol since they cannot enjoy and get pleasure daily life, have a decrease in their motivations and to cope with their depressive symptomology. In parallel to our results, a previous paper proposed data showing the association of alcohol problems in people with depression (Sullivan, Fiellin, & O’Connor, 2005). This especially important since the heavy depressive symptomology may result in suicide risk (Dumais et al., 2005). Therefore, if we prevent depressive symptomology, we can reduce alcohol consumption and prevent alcohol associated health risks.

As shown table 7, Turkish people were displaying self-worth related problems in their alcohol use.  For instance, they use alcohol to be valued by others, and to be taken care of. A previous report found that individuals who refrain from drinking have higher self-esteem in comparison to those who consume alcohol indicating the self-worth dimension might be impaired in individuals who use alcohol since self-esteem is an internal sense of worth (Luhtanen & Crocker, 2005). Therefore, self-worth impairments may increase to vulnerability to use alcohol indicating the importance of personal dimensions in alcohol use and further prevention strategies. In additions, males significantly showed higher self-worth related alcohol use than females (see table 12). Indeed, our data fit previous literature which is parallel to a meta-analysis investigating gender differences in self-esteem (Kling, Hyde, Showers, & Buswell, 1999). They explain these differences especially with the age. In our case, these differences might be arisen due to age differences in our sample between males and females, although level of self-worth between males and females depend on different sources (Kling et al., 1999).

As shown in Table 8, Turkish people use alcohol since they found to consume it “fun”. They use it because they love it, they find drinking enjoyable and they like the feeling. A previous research on alcohol use demonstrated that college students with harmful alcohol use had higher scores on the behavior approach system scale’s fun seeking subscales (Yen, Ko, Yen, Chen, & Chen, 2009). Since fun seeking in alcohol use has an association with harmful alcohol use, it is important to consider this dimension as it could be leading potential harm of individuals. Therefore, prevention strategies might also consider this dimension in order to prevent alcohol associated problems.

As shown in table 9, Turkish people use alcohol since they were affected by their parents. In parallel to our results, Finn et al. (2000) showed the role of familial alcoholism which is leading directly to alcohol problems. Therefore, it is important to consider the familial aspect of alcohol use which might be a targeting strategy in the development of prevention strategies.

As shown in table 10, Turkish people use alcohol to conform their friend cycle and groups.  According to Cooper (1994) conformity motives are negatively associated with heavy alcohol use. In addition, we did not find significant association between alcohol use intensity and conformity (see table 11) dimension. This might be possible due to conformity is not the pathological dimension of alcohol motivation, instead it is an activity for people who try to fit specific groups.

As shown in table 11, heavy alcohol users were having more anhedonic depression, more problems in their self-worth and use alcohol to have more fun. This might be possible since all these dimensions could express pathologic way of alcohol use. Therefore, those who have anhedonia, depression, self-worth impairments may express their feelings by using alcohol more heavily and desiring to get more fun from their alcohol consumption activity. As also expected, heavy alcohol consumption also associated with the higher AMS scores.

As shown in figure 1,2 and 3, total AMS scores did not affected by gender, marital status and presence of psychiatric disease. This indicates that males and females may utilize similar pattern in their alcohol consumption, so no significant change was observed (see figure 1). Although being divorced and having psychiatric disease have been shown to a risk factor for alcohol dependence (B. F. Grant & Dawson, 1997), they did not affect alcohol use motivations in our research. One possible explanation which is also a limitation in our research is that we did not use homogenous sample and we had also small sample size. Therefore, this could be affected our data. On the other hand, our research has several implications for researchers and psychologist as well: (1) Academic research may utilize AMS to understand potential alcohol use behaviors to develop prevention strategies but our scale requires to be validated with different sample belonging to different nations, (2) If psychologists could perceive the root causes of alcohol use applying AMS they can directly solve the problems associated with alcohol use.

In regard to our limitations mentioned, AMS requires to validate and tested more. In addition, bias due to personal responses might be affecting our data. Considering our research design, we tried to reach many people from different characteristics which indicates a deeper data analysis.

In conclusion, we reported that alcohol motivation scale including five dimensions which are anhedonic depression, self-worth, fun, familial and self-worth. Turkish people use these dimensions to drive their alcohol consumption behavior. For example, our sample displayed anhedonic depression that may be leading to depressive symptomology due to alcohol use. Self-worth related problems were also related to alcohol consumption indicating the importance of self-worth which is observed higher in males than females. Familial reasons play role in alcohol use in which family affects Turkish people’s alcohol use. Also, Turkish people use alcohol to conform a society and for fun. In addition, although gender, marital status, and presence of psychiatric disease did not affect alcohol motivations, people who have heavy alcohol consumption also have higher alcohol motivation scale scores indicating alcohol use intensity is affected by alcohol motivation.

In sum, we developed AMS indicating alcohol consumption motivations in Turkish sample.

Appendix A

References

Buzrul, S. (2016). Türkiye’de Alkollü İçki Tüketimi. Journal of Food and Health Science Buzrul, 2(3), 112–122.

Chentsova-Dutton, Y., & Hanley, K. (2010). The effects of anhedonia and depression on hedonic responses. Psychiatry Research, 179(2), 176–180.

Cooper, M. L. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6(2), 117.

Cooper, M. L., Russell, M., Skinner, J. B., & Windle, M. (1992). Development and validation of a three-dimensional measure of drinking motives. Psychological Assessment, 4(2), 123.

Dumais, A., Lesage, A. D., Alda, M., Rouleau, G., Dumont, M., Chawky, N., … Turecki, G. (2005). Risk factors for suicide completion in major depression: a case-control study of impulsive and aggressive behaviors in men. American Journal of Psychiatry, 162(11), 2116–2124.

Farber, P. D., Khavari, K. A., & Douglass, F. M. (1980). A factor analytic study of reasons for drinking: Empirical validation of positive and negative reinforcement dimensions. Journal of Consulting and Clinical Psychology, 48(6), 780.

Finn, P. R., Sharkansky, E. J., Brandt, K. M., & Turcotte, N. (2000). The effects of familial risk, personality, and expectancies on alcohol use and abuse. Journal of Abnormal Psychology, 109(1), 122.

Grant, B. F., & Dawson, D. A. (1997). Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: results from the National Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 9, 103–110.

Grant, V. V, Stewart, S. H., O’Connor, R. M., Blackwell, E., & Conrod, P. J. (2007). Psychometric evaluation of the five-factor Modified Drinking Motives Questionnaire—Revised in undergraduates. Addictive Behaviors, 32(11), 2611–2632.

İlhan, İ. Ö., Yıldırım, F., Demirbaş, H., & Doğan, Y. B. (2008). Alcohol use prevalence and sociodemographic correlates of alcohol use in a university student sample in Turkey. Social Psychiatry and Psychiatric Epidemiology, 43(7), 575–583.

Kling, K. C., Hyde, J. S., Showers, C. J., & Buswell, B. N. (1999). Gender differences in self-esteem: a meta-analysis. Psychological Bulletin, 125(4), 470.

Knight, J. R., Wechsler, H., Kuo, M., Seibring, M., Weitzman, E. R., & Schuckit, M. A. (2002). Alcohol abuse and dependence among US college students. Journal of Studies on Alcohol, 63(3), 263–270.

Kraus, L., Bloomfield, K., Augustin, R., & Reese, A. (2000). Prevalence of alcohol use and the association between onset of use and alcohol‐related problems in a general population sample in Germany. Addiction, 95(9), 1389–1401.

Loas, G. (1996). Vulnerability to depression: a model centered on anhedonia. Journal of Affective Disorders, 41(1), 39–53.

Luhtanen, R. K., & Crocker, J. (2005). Alcohol use in college students: effects of level of self-esteem, narcissism, and contingencies of self-worth. Psychology of Addictive Behaviors, 19(1), 99.

Moore, R. D., & Pearson, T. A. (1986). Moderate alcohol consumption and coronary artery disease. A review. Medicine, 65(4), 242–267.

Osborne, J. W. (2015). What is rotating in exploratory factor analysis. Practical Assessment, Research & Evaluation, 20(2), 1–7.

Schmidt, W., & Popham, R. E. (1975). Heavy alcohol consumption and physical health problems: a review of the epidemiological evidence. Drug & Alcohol Dependence, 1(1), 27–50.

Sullivan, L. E., Fiellin, D. A., & O’Connor, P. G. (2005). The prevalence and impact of alcohol problems in major depression: a systematic review. The American Journal of Medicine, 118(4), 330–341.

Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53.

Yen, J., Ko, C., Yen, C., Chen, C., & Chen, C. (2009). The association between harmful alcohol use and Internet addiction among college students: comparison of personality. Psychiatry and Clinical Neurosciences, 63(2), 218–224.