Bjork, J., Dougherty, D., & Moeller, G. (1999) Symptomatology of depression and anxiety in female "social drinkers". American Journal of Drug and Alcohol Abuse, 25, 173-179.

A good theory implies testable predictions. These hypotheses give direction to research. Predicting behavior from correlation studies is one of the methods used by psychologists in their research. The basic purpose of the correlational research method is to discover natural relationships between two variables. This method was chosen by psychologists Bjork, Dougherty, and Moeller doing the research described in the article "Symptomatology of depression and anxiety in female 'social drinkers'". They hypothesized that women who use moderate amounts of alcohol will report greater symptomatology of anxiety and depression, and explored how moderate levels of alcohol use are correlated with differences in mood symptomatology.

The article informs the reader of previously done research, explains why exploring the relationship between alcohol use and depression in lower social-economic status women was of interest, and describes the procedure and results of this correlational study.

First of all, applicants were recruited using advertisements in the employment section of two newspapers (free and priced) for two studies. An uneducated and unemployed community sample was of interest since previous studies suggested that depression following alcohol use disorder was higher in women of lower economic status. This particular analysis was done to extend previous findings. Newspaper advertisements for the first study did not mention alcohol (general-recruited women group), but for the second study advertisements specified that female "social drinkers" were expected to respond (drinker-recruited women group). After a brief telephone interview, women who were not in current or past psychiatric treatment, did not have serious medical problems, and did not use illicit drugs or psychoactive medication, were invited to the laboratory. In order to be eligible to participate in the second study, the drinker-recruited women group, a woman had to report consuming at least five drinks per week.

During on-site interviews when participants completed a medical questionnaire the BDI (Beck Depression Inventory) and BAI (Beck Anxiety Inventory) scores were obtained and compared using the independent t test. One hundred seventy two females participated in this behavioral research.

In addition, based on their level of education the participants were stratified into 3 groups: sub-high school, high school, and college. Subjects also were divided into 3 other groups: "abstainers", "light use" (occasional use of alcohol or no more than 3 drinks per week), and "moderate—heavy use" (4 or more drinks per week). They all completed Beck Depression and Anxiety Inventories. Four separate analyses were conducted based on these divisions. The analyses explored differences in symptomatology of drinkers and abstainers, racial differences, and education and symptomatology.

As a result, the BDI and BAI scores are displayed in the table of depression and anxiety ratings as a function of alcohol use and demographic variables. The Chi Square test and two-way analyses of variance revealed that the women who reported use of alcohol were more symptomatic than the women reported to be abstinent. There were no significant effects of race or education level on either BDI or BAI scores.

This research suggested and supported the hypotheses that even light use of alcohol (up to 3 drinkers per week) is correlated to depression and anxiety. However, thinking critically, was a reader convinced that this research had produced the correct result? In order to evaluate it a reader should keep in mind that the correlation expresses the strength of relationship between one factor and another factor; however, it does not reveal cause and effect.

First, incorrectness could occur due to self-reported statistics of alcohol consumption and symptomatology. Since the applicants were paid for participation, they might have no interest in this research and give misleading responses, which might have affected the result. The participants were not given a structured clinical interview.

Second, we have to keep in mind that the women selected were mostly unemployed and uneducated. Earlier studies have showed that unemployed and uneducated samples have higher Beck Depression Inventory scores.

On the other hand, earlier studies show similar results. For instance, Grant and Harford reported higher rates of comorbidity of alcohol abuse and depression in women. IN addition, psychology Dejong Roy has found that depression following alcohol use was higher in subjects of lower economic status. This research extends and conforms to previous findings. An exception is an earlier research study that suggested that comorbid alcohol use and depression were higher in African-Americans. This study does not reflect those findings.

Although the causality issue is not resolved, there is evidence that comorbid depression and alcohol use occurs because chronic alcohol use produces depressive symptoms.

In conclusion, the article was very well organized and contained a lot of information needed for better understanding of the study. All of the testing methods used in this research were approved by the university's Institutional Review Board.

Professor's comments: Excellent! Good article, wonderful review. Full credit.