You are on page 1of 7

1

Drinking and Exercise Behavior


By: Kace McKenna Winters
Introduction
When choosing the variables to explore in this lab assignment, I tried to think of some
behaviors that normal college students did that I took an interest in. The hypothesis that I chose
was: As the amount of drinking increases, then the amount of exercise will decrease. The amount
of alcohol consumed was treated as the predictor variable, and the amount of exercise was
treated as the outcome variable. In addition to the hypothesis, I also expect to see a greater
correlation within the Sample 2 population. Upon initial thought, I assumed that since binge and
heavy drinking had the adverse side effects contributing to poor behaviors that it would also
cause a decrease in exercise behavior1 as high risk behavior increased. The amount of drinking
was thought to take both time away from healthy behaviors, and the idea of being and staying
healthy considering how poorly the body reacts to alcohol.
Methods
This project used two population samples. Sample one came from the responses of The
Pennsylvania State Universitys BBH 411W sections during the Spring 2016 semester. Students
generated questions based off of their own hypotheses, and these questions were then emailed to
students for them to answer. The questions along with the results were then exported into an
Excel spreadsheet. Sample twos questions came from the same plot of questions, or some
variation of the questions from the students generating them, only they were sent to The
Pennsylvania State Universitys BBH 310 sections during the Spring 2016 semester. Sample one
contained more, older students; the students in BBH 411W are required to take BBH 310 as a
prerequisite.

For the variables in my project, sample one asked, How many drinks do you have per
week? and How many times a week do you exercise? The scale was 0, 1-2, 3-4, 5-6, 7-8, 910, 10+. For the second sample, I changed the question for drinking a little in terms of the scale.
The variables remained the same, only I asked for any number of drinks per week since the first
sample had so many zeros and ten or more values. Being that both of these variables are
quantitative, I chose to do a Pearsons Correlation with an alpha of 0.10, to make sure that I am
catching a difference if it truly does exist. All statistics were found using SPSS.
Results
For both the predictor and outcome variables in sample one, there were 149 responses,
and 1 no response. The predictor variable had a mean of 5.362 drinks per week, and a standard
deviation of 3.9856. The outcome variable had a mean of 2.826 times exercised per week, and a
standard deviation of 1.8708.
For sample two, there were 88 responses. The predictor variable had a mean of 3.0000
drinks per week, and a standard deviation of 1.81944. The outcome variable had a mean of
5.3864 times exercised per week, and a standard deviation of 5.53657.
The Correlation Coefficient for the Pearson Correlation in sample one was 0.110, with a
P-value of 0.182. R = 0.110 suggests a very weak positive correlation with the line of best fit at r2
= .012. The P-value is larger than 0.10, so there is no significance. The amount of drinking had
only a very weak correlation with the amount of times exercised, but this correlation was not
significant. The test statistic was 1.34182185075 with a degrees of freedom of 147.
The Correlation Coefficient for the Pearson Correlation in sample two was 0.211, with a
P-value of 0.048. R = 0.211 suggests a weak negative correlation. The P-value is smaller than
0.10, and therefore the results are significant. The amount of drinking seemed to be higher when

the amount of exercise was less. The test statistic was 2.00180194895 with a degrees of freedom
of 86.
Figure 1: Sample 1 Data (Scatterplot)

As drinking increased, so did the amount of times exercised, apparently opposite association from the hypothesis.
The results indicated a very weak correlation and no significance among 149 student responses from The

Pennsylvania State Universitys BBH 411W sections.


Figure 2: Sample 2 Data (Scatterplot)

4
As drinking increased, exercise decreased, supporting the hypothesis. The results indicated a weak correlation and
they were significant among the 88 student responses from The Pennsylvania State Universitys BBH 310 sections.

Discussion
The two samples actually gave two different results; which is what I had predicted.
Sample one showed no significance between the amount of drinks consumed and the amount
exercised. This sample did not support the hypothesis. There was a small correlation between the
variables, but it was very weak. This finding is supported by another study by Paul L, Grubaugh
A, Frueh B, Ellis C, and Egede L from their article Associations between binge and heavy
drinking and health behaviors in a nationally representative sample in Addictive Behaviors.
The results from this study also showed an in increase in physical activity among those who
drank more than others. This is something that I found completely baffling2.
Sample 2, on the other hand, supports the hypothesis. As drinking increased, exercise
decreased. There was a weak correlation, but it was significant. These findings were extremely
surprising considering the average amount of drinking is actually less in sample 2. The amount
of exercise is also greater in sample 2. This difference probably came from the change in
measurement, as people could list as many drinks as possible. Some individuals had almost 30
drinks in one week.
I think that the main differences and limitations between these two samples is the age, the
measurement, and people just not telling the truth. As stated before, BBH 310 is needed for BBH
411W; thus, meaning that the students are younger and tend to behave more inappropriately by
engaging in binge drinking behaviors. The second limitation lies in the fact that I chose to open
up my question to the 310 students to answer as many drinks as possible. This could have
obviously skewed the results. The last major limitation that I recognize is that some people may
not have told the truth when it comes to exercise. About ten percent of the responses indicated

that students exercised more times per week than days in a week. This just is not feasible. I also
found it extremely frustrating that my specific questions were not found, and this could have
jumbled up my statistics more than imaginable. Overall, as drinking increased, it did not affect
the amount of times an individual exercised. In fact, there may be evidence to suggest an
opposite association.

References
Miller J, Naimi T, Brewer R, Jones S. Binge Drinking and Associated Health Risk Behaviors
Among High School Students. PEDIATRICS. 2007;119(1):76-85. doi:10.1542/peds.2006-1517.
Paul L, Grubaugh A, Frueh B, Ellis C, Egede L. Associations between binge and heavy drinking
and health behaviors in a nationally representative sample. Addictive Behaviors.
2011;36(12):1240-1245. doi:10.1016/j.addbeh.2011.07.034.

1 Miller J, Naimi T, Brewer R, Jones S. Binge Drinking and Associated Health Risk
Behaviors Among High School Students. PEDIATRICS. 2007;119(1):76-85.
doi:10.1542/peds.2006-1517.
2 Paul L, Grubaugh A, Frueh B, Ellis C, Egede L. Associations between binge and heavy
drinking and health behaviors in a nationally representative sample. Addictive Behaviors.
2011;36(12):1240-1245. doi:10.1016/j.addbeh.2011.07.034.

You might also like