Is there a correlation between a student's level of involvement on Georgia College's campus and the number of weekends they leave the Milledgeville area on the weekends?
By:
Lindsey Harrison
Mary Ashley Gibson
Kelsey Richardson
Kelsey Stone
Katie Garrett
Introduction
As Georgia College student's we thought it would be interesting if see if there is a correlation between the level of involvement a student has and how often they leave the Milledgeville community on the weekends. Georgia College is often referred to as a "suitcasecollege" meaning students frequently leave the campus on the weekends. These results could be beneficial to various organizations on campus such as CAB, Students Affairs, and Georgia College Admissions Office.
The population for our survey was Georgia College students. Our sample size was 75 students ranging from freshman to seniors at Georgia College. We believe there is a moderately negative correlation because the more involved a student is on campus the less they will go home; however, there are other variables that could influence this greatly. Our two main research variables are level of involvement on campus and the frequency a student leaves the Milledgeville community on the weekends. Our demographic variables are class year, gender, place of residence in Milledgeville, and how far away a student's family lives.
Methods
We collected data by conducting a random sample survey at Georgia College's fountain. Our survey was on paper with a series of circled answers. The fountain is a central location on campus which attempted to reduce bias. Our survey was composed of two main variables of interest and several demographic variables. The construct was designed to measure the overall involvement per student on Georgia College's campus. The student's were asked to estimate the amount of time per week they devote to orgainizations including honor societies, greek life, student organizations, intramurals, religious groups, and job / internships. A descriptive summary for our two main variables is listed below.
Results
Two Main Variable Results
Summary statistics:

Frequency table results for # Weekends Gone:


Frequency table results for GC Involvement:



Simple linear regression results:
Dependent Variable: # Weekends Gone Independent Variable: GC Involvement # Weekends Gone = 2.9187787  0.123664126 GC Involvement Sample size: 75 R (correlation coefficient) = 0.1964 Rsq = 0.03855579 Estimate of error standard deviation: 1.1699039 Parameter estimates:
Analysis of variance table for regression model:

Our regression equation for our two main variables is y=2.9190.124x where y is the number of weekends a student leaves Milledgeville and x is the level of involvement of campus determined from our construct. The slope is 0.124. The interpretation of this is that with every point increase on our construct in campus involvement a student has, the number of weekends that students will go home decreases by 0.124. The yinterept for our regression equation is 2.919. However, this does not have any relevance to our particular survey because the lowest anybody can score on from our campus involvement construct is a 6. From our regression equation if a person is not involved at all, and therefore scores a 6, then the estimated number of weekends they will go home is about 2.165.
The strength of our regression equation is determined by the correlation coefficient. Our value r=0.196 shows that there is a negative relation between campus involvement and weekends gone from Milledgeville, but because the number is so small there is hardly any correlation at all. We conclude that there is an extremely weak negative correlation. Our value of rsquared=0.039 means that about 3.9% of the data is explained by our regression equation.
When exploring our data grouped by our demographic variables we noticed a strong difference between females and males, whose scatter plot is shown below. The regression equation for the females is given by y=3.6440.191x with r=0.269 and rsquared=0.072. The regression equation for the males is given by y=1.8780.029x with r=0.061 and rsquared=0.004. There is a stronger correlation between level of GC involvement and number of weekends gone for females than males because the correlation coefficient is closer to 1 for the females than for the males.

Demographic Variable Results





Conclusion
When planning this research study we expected there to be a strong negative correlation between the level of involvement on GC campus and the number of weekends a student typically leaves the Milledgeville area a month. What we actually discovered is that although there is a negative correlation it is an extremely weak one. When we grouped the scatter plot by gender the correlation was slightly better for females than our original scatter plot, but it would still be considered a weak correlation. Our population goal was to represent all Georgia College students; however, when looking at our demographic variables in graphical form it is easy to notice that not all class years were represented fairly. We have more sophomores and hardly any seniors at all. One mistake we made was to underestimate the potential for lurking variables. There are countless reasons for a student to leave the Milledgeville area on the weekends that we did not take into account. The student's home life is one lurking variable or if he/she has a significant other who goes to a different school. These are all factors that could influence a student's decision to leave Milledgeville which were not considered in our study. For our study to be more beneficial we could ask why and/or where they are leaving to on the weekends.
Already a member? Sign in.