Introduction/Purpose:
The goal of this survey was to test if playing sports affected a student’s GPA at Georgia College. We had three quantitative variables: hours spent studying, GPA, and hours spent playing a sport. Our qualitative variables included major, and what sports occupied most of the subject’s time, either intramurals or a varsity sport. We wanted to compare GPA cohesively between intramural sports and varsity sports. We thought that the more intramurals or varsity sports you played, the lower the GPA. We also assumed that the more hours spent studying, the higher the GPA. Therefore, our prediction was that varsity athletes spent more time playing their sport than intramural athletes and those that did not play a sport at all and thus would have less time to make better grades.
Methods:
In order to collect the appropriate data, we asked random people to fill out a survey. On this survey, we asked 5 questions:
1) How many hours a week do you spend studying and doing homework?
2) What is your current GPA?
3) How many hours do you spend playing an intramural and/or varsity sport?
4) What is your major?
5) Which sport occupies most of your time?
In order to compare the categorical variable, we used a pie chart. To compare the quantitative variables, we used a scatter plot with a regression line.
Conclusions/ Interpretations:
After collecting data from 40 random volunteers we began to construct pie charts and scatter plots. With the construction of our first pie chart we found that out of 40 students, 52.5% played intramural sports, 15% played a varsity sport and 32.5% played neither an intramural or varsity sport. Our second pie chart was composed of the 40 students major’s. We found that 55% of the students had an arts and science major, 7.5% had a business major, 10% had an education major, 17.5% pursued a health science major and the remaining 10% were undecided.
After our pie charts were finished we commenced to constructing our scatter plots. Our first scatter plot was composed of how many hours each student studied per week compared to their GPA. With this kind of comparison one would expect a strong positive correlation; what we observed was a weak positive correlation of 0.073 with several outliers. Some students studied for 1214 hours a week but only had a 2.8 or 2.9 GPA compared to students who studied for anywhere from 310 hours a week and had a 4.0 GPA. In order that we may view a stronger correlation in the future a scatter plot with a large data base would better reflect a positive correlation. The second scatter plot we assembled, which was composed of GPA compared to hours spent playing a sport per week, had a positive correlation of 0.1815. The third scatter plot we created included a comparison of hours spent playing a sport per week and hours spent studying each week. With this scatter plot we predicted a negative correlation; we observed a weak negative correlation of 0.1309. This kind of correlation was expected because as time spent playing sports increases and time spent studying increases the time available to do the other is reduced drastically.
We expected to observe stronger correlations when we built our scatter plots. In order to be able to observe stronger correlations a larger data set would more adequately represent this. Other factors such as exercise instead of simply sports would have strongly affected our results. Many students that we interviewed asked us if exercise counted as a sport, we of course answered no because it did not relate to our survey or the questions we were trying to answer.
In this survey and experiment, there are many lurking variables. The first lurking variable could be that students had jobs, thus not allowing them to participate in sports very little or even at all. Another lurking variable could be that some students don’t need to study because they are naturally smart, thus allowing them to have a high GPA. There are many other lurking variable that need to be taken into account when interpreting the data.





Summary statistics:

Simple linear regression results:
Dependent Variable: GPA Independent Variable: hours studied GPA = 3.4970074 + 0.00514193 hours studied Sample size: 40 R (correlation coefficient) = 0.073 Rsq = 0.0053218515 Estimate of error standard deviation: 0.3745673 Parameter estimates:
Analysis of variance table for regression model:

Simple linear regression results:
Dependent Variable: Hours of Sports Independent Variable: hours studied Hours of Sports = 3.679197  0.11481868 hours studied Sample size: 40 R (correlation coefficient) = 0.1309 Rsq = 0.01712654 Estimate of error standard deviation: 4.6346855 Parameter estimates:
Analysis of variance table for regression model:

Simple linear regression results:
Dependent Variable: Hours of Sports Independent Variable: GPA Hours of Sports = 5.2190986 + 2.259816 GPA Sample size: 40 R (correlation coefficient) = 0.1815 Rsq = 0.032959476 Estimate of error standard deviation: 4.5972047 Parameter estimates:
Analysis of variance table for regression model:

Correlation matrix:

Already a member? Sign in.