Statistics Term Project Report
Olga Chukhrai
Nov. 22, 2010
Research Question:
Is the number of hours spent at work per week related to the Grade Point Average
(GPA) of a student at SLCC?
Introduction - Purpose of the Study:
The following report is designed to evaluate the existence of a relationship
between hours spent at work and cumulative grades received which is given by the Grade
Point Average (GPA). This report will present data collected from various students at Salt
Lake Community College. A full evaluation of the data collected from 36 random
students will be presented. By analyzing this information it is possible to determine if
there is a correlation between work hours and GPA. This report will also discuss
underlying factors and possible problems with the study.
Study Design:
The purpose of our project was to evaluate if there is a relationship between hours
spent at work and Grade Point Average (GPA). We decided to question Salt Lake
Community College students with an anonymous survey which we distributed among our
group members. Each group member administered an equal number of surveys to random
students in their classes. The 3 question survey stated that answers were anonymous and
the survey taker was told this before they were given the survey. The survey asked the
subject to write the amount of hours they were working, their GPA, and to circle “yes” or
“no” if they thought work affected their GPA. Surveys were then collected and a
summary and analysis of the data was done.
Data, Statistics, and Graphs:
The data collected from the surveys was entered into Excel and organized. The
following is a summary table of the statistical analysis for each variable independently.
Variable 1 is the number of hours spent at work, and Variable 2 is GPA.
Figure 1
As can be seen from the above number summary, the surveyed group worked
anywhere from 0 - 49 hours per week, with an average of 31.1 hours and standard
deviation of 12.3 hours. The mode, or most common amount of hours worked in our
surveyed group was 40 hours per week which is considered full time. The mode was
determined by looking at the stem and leaf plot in Figure 2. The same group also had a
GPA ranging from 2.5 – 3.9, with a mean of 3.4 and standard deviation of 0.4 points. The
mode, or most common GPA among our subjects, was actually a 3.7 which is quite high
considering a 4.0 indicates straight A’s. The mode for GPA was determined from the
stem and leaf plot in Figure 3. Using the above data, the correlation coefficient was found
to be -0.11688754 which indicates a very slight negative correlation between the two
variables, much less than was expected. This indicates that as hours spent at work
increases, GPA decreases slightly.
Figure 2
Figure 3
Any direct correlation between Variable 1 (hours worked per week) and Variable 2
(GPA) can be seen in the scatter plot below.
Figure 4
Analysis, Difficulties and Surprises Encountered:
The collection of data for this research topic was not very difficult, but the
analysis did show some unexpected numbers. First of all, GPA was found to be quite
high, as were the hours worked. The correlation between the two variables was found to
be much less significant than initially thought and this brought up some interesting
questions. If this study was conducted at any other college or university would the results
be similar or is it possible that teachers at SLCC are more willing to understand and work
with students who are working full time? To address this question it would be necessary
to distribute the exact same survey to college students at the University of Utah and
BYU, and maybe even several institutions out of state to obtain a more diverse sample. It
would also be interesting to know what topics each of the respondents was studying to
compare difficulty levels with the GPA they obtained. Another underlying factor that was
not considered in our study was the role of children in the household and family issues
that may reflect on the student’s GPA in addition to the hours spent at work. After all,
children are very time consuming and may play a role in the college student’s school
success. Some other underlying factors that we did not control for include study habits
and IQ levels which also could have affected the data. A more concrete correlation
between the two variables may also have been determined if a larger sample size was
surveyed. While 36 students gives an idea of the relationship that might exist it certainly
may not speak for the entire population of students at all SLCC campuses. To obtain
more concrete data, a larger study population would need to be surveyed.
Lastly, while the first two questions of the survey are straightforward answers, the
third “yes” or “no” question turned out to be more open to interpretation. In later
discussion of the subject, people mentioned that they answered “yes” work hours affect
GPA but meaning in a positive way, that working more helps them prioritize and go
better in school. Others, including myself, felt that “yes” work hours affect GPA, but in a
negative way, with more hours spent at work meaning less time to spend on homework
and studying. While this did not affect our two variable directly, it could have been an
interesting topic to elaborate on.
Interpretation and Conclusions:
Over all our survey was administered in a random and unbiased way with simple
direct questions that were easy for people to understand and answer. Knowing what we
do now, it would have been beneficial to have a larger survey group that could have
included several different Salt Lake Community College campuses. For stronger data,
underlying factors such as degree programs, number of children in the household, and IQ
should be analyzed. While a correlation between the two variables of hours worked and
GPA was determined to exist, it was found to be much less significant that initially
hypothesized and may not represent the entire SLC College very well.
Appendix:
Histogram of variables 1 and 2:
Figure 5
Frequency Table Results for Variable 1 (Hours Worked):
var1 Frequency Relative Frequency
0
2
0.055555556
12
1
0.027777778
13
1
0.027777778
18
2
0.055555556
20
3
0.083333336
23
1
0.027777778
24
1
0.027777778
30
4
0.11111111
32
1
0.027777778
33
1
0.027777778
35
1
0.027777778
36
2
0.055555556
40
12
0.33333334
42
1
0.027777778
45
2
0.055555556
49
1
0.027777778
Figure 6
Frequency Table Results for Variable 2 (GPA):
var2 Frequency Relative Frequency
2.5
1
0.027777778
2.6
1
0.027777778
2.77
1
0.027777778
2.8
3
0.083333336
3
3
0.083333336
3.1
1
0.027777778
3.2
2
0.055555556
3.3
1
0.027777778
3.4
2
0.055555556
3.5
3
0.083333336
3.6
2
0.055555556
3.66
1
0.027777778
3.67
1
0.027777778
3.7
7
0.19444445
3.75
1
0.027777778
3.8
5
0.1388889
3.9
1
0.027777778
Figure 7