Aim of the study

This study assessed the associations between five alcohol consumption variables (length of time of last drinking occasion; number of drinks consumed during the last drinking occasion; frequency of alcohol consumption, heavy episodic drinking, and problem drinking as dependent variables) and three academic achievement indicators (importance of good grades, performance relative to peers, module mark as independent variables), whilst controlling for two socio-demographic variables (sex, age). Undergraduate students from University of Gloucestershire, UK completed a general health questionnaire, and their module grades were retrieved from the university computer systems. The specific objectives were to:

• Explore the alcohol consumption indicator/s (type of drinking) that were mostly associated with academic outcomes; and
• Explore the academic achievement measures that were mostly associated with alcohol consumption.

Data collection procedures

Ethical approval was provided by the university research ethics subcommittee after review of study design, tool, research material, and participant information sheet (included a letter of invitation highlighting that participation is voluntary). The population for this cross-sectional research design comprised undergraduate students enrolled for classes at the University of Gloucestershire, UK during the data collection period (2008-2009). Module teachers were also provided with information about the research (aims, objectives, voluntary nature of participation), and were approached for permission to collect data from students attending their module.

The study was a general student health survey[26] which also contained two questions about students’ views on their academic performance. The survey, usually administered toward the end of a lecture/class, required ≈ 15 min to complete. Each questionnaire had a participant information sheet attached. Students who remained in class to participate read the information sheet and if they wished to participate, removed and kept it for future reference. No monetary incentives nor course credit (as inducements) were provided to students for their participation. Once students completed the questionnaire, they brought it to the front of the room where it was placed in a large envelope. In the questionnaire, each participant provided her/his student registration number which was subsequently employed to import students’ actual grades that they accomplished from the university Strategic Information Technology System (SITS), and link the retrieved grades to students’ responses of the health survey. The students were assured that data protection and confidentiality were observed, and that their student registration numbers would be only used import students’ actual grades in an anonymous manner. The response rates were ≈ 80%, and for the analysis, we used data from 379 students.

Instrument

The Health and Wellbeing Questionnaire, as employed in many student health studies,[26, 27, 28, 29, 30, 31, 32] included socio-demographic information (e.g., sex and age), as well as self-reported health behavior data (e.g., five alcohol consumption indicators):

Length of time of the last (most recent) drinking occasion (1 item): “The last time you ‘partied’/socialized, how many hours did you drink alcohol?” Participants provided the number of hours.

Amount (number of drinks) of alcohol consumed during the last (most recent) drinking occasion (1 item): “The last time you ‘partied’/socialized, how many alcoholic drinks did you have? (including alcoholic drinks you possibly had before going out).” Participants provided the number of drinks.

Frequency of alcohol consumption (1 item): “Over the past 3 months how often did you drink alcohol, for example, beer?” (six response options: “never,” “once a week or less,” “once a week,” “a few times each week,” “every day,” and “a few times each day”).

Heavy episodic drinking (1 item): “Think back again over the last 30 days. How many times (if any) have you had five or more drinks on one occasion?” (A “drink” is a glass/bottle/can of beer [≈50 cl], a glass/bottle/can of cider [≈50 cl], two glasses/bottles of alcopops [≈50 cl], a glass of wine [≈15 cl], a glass of spirits [≈5 cl] or a mixed drink). Response options were “never,” “once,” “twice,” “3-5 times,” “6-9 times,” and “10 or more times.”

Problem drinking (4 items): Data was gathered using an alcoholism-screening CAGE test,[21] a brief screening instrument consisting of four short questions [“Have you ever felt you should Cut down on your drinking?; Have people Annoyed you by criticizing your drinking?; Have you ever felt bad or Guilty about your drinking?; Have you ever had a drink in the morning to get rid of a hangover?” (Eye opener)]. Each question is answered either “yes” or “no”. Two or three affirmative answers suggest problem drinking, while four positive responses indicate serious suspicion of alcohol dependence.

Educational achievement was measured using three indicators.[27] The health questionnaire contained two items on students’ educational achievement; the third indicator (module mark) was retrieved from university computer systems.

• Students’ internal reflection on their academic achievement (importance they attach to achieving good grades) - “How important is it for you to have good grades at university?” (4 response categories, 1 = “not at all important,” 4 = “very important”);
• Students’ subjective comparative appraisal of their overall academic performance (in comparison with their peers) “How do you rate your performance in comparison with your fellow students?” (5 response categories, 1 = “much worse,” 5 = “much better”); and
• An external objective teacher evaluation of students’ overall academic performance in terms of the final module marks (in percent) accomplished by students in their course assessment/s). Similar to other studies,[28] each student's registration number was employed to import the student's module mark/grade (as percentages) from the university SITS,[33] which was then linked to the student's questionnaire responses.[27]

Statistical analysis

Analyses were performed using SPSS 14.0 (significance level set at P < 0.05). For demographic and other characteristics, frequencies were calculated separately for males and females in order to provide precise estimates. This is because sex is related to drinking and sex differences in alcohol consumption among undergraduates have been reported.[34] In addition, sex had a significant, direct relationship with academic problems;[13] and prospective results of young adults suggested that there might be some important sex differences as to how a lower response or low sensitivity to alcohol is related to alcohol outcomes and future drinking problems.[35] Sex comparisons were undertaken using Chi-square (χ²) statistics for categorical variables and t-test statistics for continuous variables.

We then undertook five separate linear regression analyses to assess the relationship of each of the three academic achievement variables (importance of good grades, performance relative to peers, module mark), controlling for the demographic variables (age, sex), on each of the five alcohol consumption measures [length of time (duration) of last drinking occasion; amount (number of drinks) consumed during the last drinking occasion; frequency of consumption; heavy episodic drinking; and problem drinking]. For each regression analysis, in models 1-3, each of the three academic achievement variables was entered individually together with age and sex.

Alcohol consumption patterns

Male students spent more hours during the last (most recent) drinking occasion than females, and consumed more alcoholic drinks at that occasion [Table 1]. Most students (74%) consumed alcohol at least once week, and only 3% reported no alcohol consumption. Males consumed alcohol more often than females (P < 0.04). Half the sample (54%) testified heavy episodic drinking (≥5 drinks in a row) at least three times during the last month, and only 20% of respondents reported no heavy episodic drinking. The frequency of heavy episodic drinking was significantly higher among males than females (P < 0.001). About 31% of the sample reported two or three affirmative answers in CAGE suggesting problem drinking, while 2.4% indicated four positive responses (serious suspicion of alcohol dependence). Again, males had more positive answers in CAGE than females (P < 0.035).

Table 1. Demographic and other selected characteristics of sample

Characteristics of academic achievement

Two-thirds of the sample (64%) felt that having good grades at university was very important for them [Table 1], and ≈62% rated their own performance as being about the same as that of fellow students. The average module mark achieved by students was 55% (SD = 13). There were no significant differences between males and females across the three measures of academic achievement.

Are students’ duration of time of last drinking occasion or number of drinks consumed during last drinking occasion associated with academic achievement?

Neither the length (duration) of the last (most recent) drinking occasion nor the amount (number of drinks) consumed at the last occasion were significantly associated with any of the three academic achievement indicators (data not presented). However, age was consistently negatively associated with students’ length of time of last drinking occasion, and also with the number of drinks consumed during the last drinking occasion. Conversely, males were positively associated with these two alcohol consumption variables (data not presented).

Is students’ frequency of alcohol consumption associated with academic achievement?

When frequency of alcohol consumption was employed as the dependent variable in the regression analysis [Table 2], males were consistently positively associated with the frequency of alcohol consumption, while age was not significantly associated. Independent of age and sex, the models show that the less the importance of good grades was to students (model 1), the higher the frequency of alcohol consumption (P < 0.05). However, the associations between alcohol consumption frequency and the remaining two academic achievement variables (performance relative to peers-model 2; module mark-model 3) were not significant.

Table 2. Regression models for three academic achievement indicators on frequency of alcohol consumption^a as dependent variable

Is students’ heavy episodic drinking associated with academic achievement?

When heavy episodic drinking was selected as the dependent variable in the regression analysis [Table 3], age was consistently negatively associated, whilst males were positively associated with heavy episodic drinking. Controlling for age and sex, the less the students rated the importance of good grades for them (model 1) and the lower they felt their academic performance was relative to their peers (model 2), the higher the frequency of heavy episodic drinking. However, the negative association with the third indicator of academic achievement (module mark) was minimal (model 3) and not significant. In addition, for heavy episodic drinking, a comparison of model 2 with model 3 showed that the importance of good grades explained more of the variation of heavy episodic drinking than performance relative to peers.

Table 3. Regression models for three academic achievement indicators on heavy episodic drinking^a as dependent variable

Is students’ problem drinking associated with academic achievement?

When problem drinking (measured as positive “Yes” answers in CAGE) was employed as the dependent variable in the regression analysis, males were consistently positively associated with problem drinking, while age was not significantly associated [Table 4]. Controlling for age and sex, the less the students rated the importance of good grades (model 1), the higher the problem drinking (P < 0.05). The associations between problem drinking and performance relative to peers (model 2) and with module mark (model 3) were both not significant. Table 5 provides a summary of the relationships between alcohol consumption variables and academic achievement and demographic variables.

Table 4. Regression models for three academic achievement indicators on problem drinking^a as dependent variable

Table 5. Associations between alcohol consumption variables and academic achievement and demographic variables

Which alcohol consumption indicator/s (type of drinking) were mostly associated with academic outcomes?

As for objective one, Table 5 depicts an overview of the findings. We found relatively strong associations between heavy episodic drinking on the one hand, and the importance of good grades and the subjective academic performance relative to peers on the other, i.e., the more often students had ≥ 5 drinks on one occasion, the less they rated the importance of good grades was to them, and the lower they rated their own academic performance relative to their peers. Our current findings of negative significant relationships between heavy episodic drinking and two subjective academic achievement measures support previous research where more frequent heavy drinking was related to academic problems,[45] lower study hours,[46] and lower reported grades.[47] Our findings are also compatible with that heavy episodic alcohol use has been incriminated in acute in addition to the long-term health and social consequences.[48, 49] An important point is that in the current study, in order to determine heavy episodic drinking frequency, we employed a cutoff of consuming ≥ 5 drinks on one occasion. Such cut-off value is in line with others.[50] However, some researchers[18, 43] determined heavy alcohol use/heavy drinking days by computing a frequency based on the consumption of ≥5 consecutive drinks for men; or the consumption of ≥4 consecutive drinks for women. Hence, in terms of females, the current study employed a higher cut-off representing a more “relaxed” measure of heavy episodic drinking. Such cut-off could have resulted in a potential underestimation of the prevalence/number of heavy episodic drinking females; and a similar underestimation of the already significant associations between academic accomplishment variables and heavy episodic drinking.

In contrast to heavy episodic drinking, two of the other alcohol consumption measures we employed (frequency of alcohol consumption; and problem drinking estimated by number of affirmative answers in CAGE) showed fewer associations with academic outcomes. Frequency of alcohol consumption and problem drinking were both only associated with lower importance of achieving good grades, but not with ratings of one's academic performance relative to peers. The remaining two alcohol measures (length of time of last drinking occasion; amount consumed during the last drinking occasion) showed no association with the three academic achievement measures. This latter finding might be rather surprising given that the amount of drinks consumed at the last (most recent) drinking occasion could potentially be similar/parallel to heavy episodic drinking that measures the amount of alcohol consumed at one occasion. However, the questionnaire item asking only about the last drinking occasion reflects a more situational measure, in contrast to the heavy episodic drinking item that queried participant's drinking pattern over a period of time, and thus perhaps more likely to be associated with academic performance.

Direct comparisons of our findings with other studies are difficult, because of the diversity of approaches employed in the measurement of alcohol use/misuse (e.g., time period of recall; whether the number of drinks or actual alcohol units were employed in estimating the amount of alcohol consumed; cut-offs used for calculating heavy episodic drinking for men and for women, etc.). However, our associations between alcohol use and academic impairment were mostly demonstrated when measures of heavy episodic drinking or high risk drinking were used, as also shown in a review on the consequences of alcohol misuse in college populations.[15]

Which academic achievement measures were mostly associated with alcohol consumption?

For objective two, the current study found that alcohol consumption was related to impairments in subjectively reported academic achievement measures (importance of achieving good grades; one's academic achievement in comparison to one's peers). However, we did not find associations with the objective academic achievement indicator (actual module grade) [Table 5]. Very few studies had been able to link alcohol use to students’ achieved grades that are retrieved from records/university computer systems (i.e., objective measure). For instance, one study found that the amount of alcohol consumed was a negative predictor of the cumulative GPA retrieved from students records.[19] Similarly, research in the Netherlands found a significant association between alcohol dependency and failing the 1^st year as reported from student registries.[8] Other studies used self-reported GPA and found significant negative associations between heavy alcohol use and GPA,[46] while in some research the association disappeared when controlling was undertaken for other factors.[13, 18]

Several reasons are proposed for the lack of an association between alcohol consumption and module grade in the current study. Methodological limitations might account for this, including our limited sample size. In addition, the student's grade that was retrieved from the university SITS was that of a single module (the one that the student was attending when the questionnaire was completed) – this may not reflect the average (all year) final grades for any given student. It is also acknowledged that it is difficult to predict students’ grades generally, whether by using demographic and academic/educational variables;[51] or by using students’ educational satisfaction together with a range of health status (e.g., BMI, depressive symptoms, stress, etc.) and health behavior variables (e.g., health awareness/consciousness, self-rated health, physical activity, nutrition, etc.).[27] Hence to attempt to predict students’ grades specifically by alcohol consumption represents an even more formidable challenge.[13, 18] Many other factors (e.g., age, social support, etc.) might play crucial roles and quite often alcohol drinking loses its initial significance when controlling is undertaken for such other factors.[27] It is also likely that heavy alcohol use might not lead to lower grades directly, but indirectly through impairments in bodily/mental functions: A path diagram from a prospective study showed that amount of alcohol consumed was associated with sleepiness and negatively associated with sleep duration, which in turn had effects on students’ achieved cumulative GPA.[19] Likewise, a recent longitudinal study demonstrated that binge drinkers experienced socio-emotional problems over time which seemed to be a mechanism for the academic risk posed by drinking.[52]

Whether the links between alcohol use and academic achievement are causal relationships remains unclear from the current cross-sectional study. It could be that straight forward cause-effect relationship exists, or that such links might be moderated by sleep, given that alcohol use is related to sleep habits where such habits and resultant daytime sleepiness could mediate alcohol's possible influence on academic performance.[19] In addition, heavy alcohol use was associated with lower GPA both directly and indirectly through its association with fewer study hours.[44] Bidirectional relationships might also exist, where alcohol use and academic performance impairments mutually influence each other in a vicious circle. On the one hand, the level of academic performance in high school children predicts their drinking problems as young adults independent of individual/family confounders.[53] At university, it is also plausible that performing poorly in coursework could lead to alcohol use.[19] On the other hand, bidirectional effects are supported by a longitudinal study showing that missing classes or falling behind schoolwork was a result of drinking during college, but the same study also showed that high school GPA was negatively associated with heavy alcohol use.[18] Given the uncertain time precedence among the variables, causal interpretations of the current study's data should be cautious.

This study has limitations. The study was undertaken at one university in one country, and whether results would be similar for university students of a wider socio-economic/cultural background is an empirical question. However, even prospective studies of the association between drinking and academic performance have often relied on data collected at single universities.[8, 13, 18, 22] Besides the actual module grade (retrieved from the university SITS system), all data was self-reported (social desirability and sociability cannot be ruled out). For instance, we did not objectively measure drinking using biological measures (e.g., breath alcohol concentration).[16] However, self-reports from confidential and unidentified questionnaires is a commonly accepted procedure in substance use research of university students and provides data that are generally both valid and reliable.[54] The current survey was administered only once, but there might be a need to follow up undergraduates at “tighter” time intervals employing multiple measures of consumption within the context of a semester, as significant and abrupt changes occur frequently in the individual drinking trajectories of teenagers and young adults.[55, 56] We did not find relationships between some alcohol consumption indicators and academic performance, where modest and non-significant associations were exhibited, although low statistical power may explain some of the null associations.[57] We did not measure achievement orientation established prior to entering university, and this might affect drinking behavior in university. Whilst the alcohol indicators we employed assessed drinking during different time periods (e.g., “most recent,” “30 days,” “3 months.”), yet, the “objective” academic measure is for a module undertaken usually in an entire academic term. Different time frames might hinder the possible linkage between specific drinking incidents and specific academic behaviors. It could also be argued that the “objective” academic measure we employed might not be 100% “objective”: Teacher evaluations of academic performance need to consider faculty grading practices that might be biased by discipline-specific norms and instructor-level variables. The current survey employed the final module marks of students who were attending a variety of different modules, but academic performance (final module marks) were not weighted to account for course difficulty, or in terms of variations in the rigor of grading practices across academic departments and among individual faculty within departments.[58, 59, 60]

In order to delineate cause and effect, prospective study designs are needed to examine the relation between drinking and academic performance. Such designs would need to collect baseline data on the relevant variables beginning many years before college. Given that achievement orientation and alcohol use are characteristics that are both formed prior to university, it is possible that the former might cause the latter.

In addition, to facilitate more valid comparisons, future research would benefit from some standardization, were possible, to the measures employed to estimate the various “facets” of alcohol consumption and/or types of drinking patterns. Further, periodic/regular estimation of alcohol risk and harm (along with a range of health and wellbeing variables) in nationally representative samples of university students would contribute to an evidence base that is critical for this important young adult population, a point that seems currently lacking in the UK. Furthermore, the examination of the pathway/s linking academic/educational performance and alcohol use/dependency in young adults may help identify opportunities for preventive interventions.

Our findings point to that alcohol misuse, especially heavy episodic drinking is very likely to have negative consequences on academic performance at University. Such findings echo with similar studies undertaken in secondary schools, where drinking predicted deteriorating socio-emotional functioning, with negative consequences for adolescents’ academic grades by the end of high school.[52] The findings also echo with research of college students, where heavy alcohol consumption (and resultant unhealthy sleep habits) have been associated with a range of behavioral challenges, including poor academic performance (Buboltz et al., 2006; Perkins, 2002).[15, 61] Indeed, alcohol use disrupts learning and memory in adults (Zeigler et al., 2005).[62] Unsurprisingly, alcohol abuse is the single greatest public health hazard at American college and university campuses, and the alcohol drinking culture remains to be highly resistant to change.[63]

Therefore, alcohol policies on campus and educational and normative campaigns for students are highly relevant. Such programs and interventions should aim at prevention of alcohol abuse of college students and on college campuses as well as cessation programs. Hence universities might wish to pay attention to their alcohol policies and enforcement as these have been reported to be inversely associated with five outcomes related to student alcohol abuse or related consequences.[64] Likewise, campus and community coalitions/partnerships between campus leaders and senior college administrators and a range of community stakeholders can efficiently attend to the environment that may promote high-risk drinking, although few campuses work within such a structure.[65] For high-risk college students, brief intervention approaches with such students has been successful in reducing alcohol consumption and/or related consequences,[66] and although, many interventions for college students have traditionally been behavioral, pharmacological treatments could provide additional options.[67] Programs could involve creating motivation to change drinking; changing the drinker's expectancies about the effects of alcohol; clarifying norms through feedback on the drinker's alcohol use in comparison with other students; and making available cognitive-behavioral skills training, including how to monitor daily alcohol consumption and stress management.[67] The important point is that interventions that are selected need to be evidence-based.[67, 68]