No Time to Study Professor Fang Fang BUS 304 Final Paper 5/14/08 By. Nick Hall Andrew Kalfayan Michael Laow Joey Soitonu

Transcription

1 No Time to Study Professor Fang Fang BUS 304 Final Paper 5/14/08 By. Nick Hall Andrew Kalfayan Michael Laow Joey Soitonu 1

2 Table of Contents OBJECTIVE OF THE SURVEY 4 HYPOTHESIS 4 THE SURVEY 5 SURVEY SCOPE 6 CREATING THE SURVEY 6 MAJORS 7 CLASS STANDING 8 AGE SCOPE 9 GPA RANGE 10 GENDER DISTRIBUTION 11 DATA ANALYSIS 12 EMPLOYED VS. UNEMPLOYED 12 STRENGTH AND VARIATION BETWEEN VARIABLES AND GPA 13 REGRESSION 15 WILLING TO COMPROMISE 16 SCHEDULING 17 OTHER FINDINGS 18 SLEEPING PATTERNS 18 STUDYING HABITS (HOURS STUDIED) 19 2

3 LOGICAL EXPLANATION 19 EVIDENCE 20 INTANGIBLES 21 ERRORS IN THE SURVEY 22 WORKS CITED 23 3

4 Objective of the survey Everyday college students have to balance their lives between school and work. With today s cost of living, especially in Southern California, individuals have to work more hours to maintain their lifestyles. Focusing on California State University San Marcos, students battle a multitude of expenses to attend. The increasing tuition costs, due to budget cuts, rising gas prices, and inflated text book costs, is pushing students to work more hours to accommodate their expenditures. It seems as though students are working more hours to be educated. However, a common complaint from college students is the lack of preparation time for classes. Could the lack of time due to increasing work hours be the cause of a lower academic performance? The purpose of this survey is to provide students a better understanding of how to manage their lifestyles in order to maximize success in the University setting. The data collected can provide future students with the correlation between the effects of working during one s college career and their overall performance. Therefore, this survey attempts to ask the question: Does the amount of work hours taken on by students affect their grades? Hypothesis As stated we believe that there is a correlation between the employed and unemployed students and their overall G.P.A. We feel that the more hours being worked by students correlates to a lower G.P.A attained. We also hypostasize that their may be a correlation between how one schedules their time. With this we believe that if an individual were to schedule their work around school they will have a higher G.P.A than those who schedule school around work. If a student is not taking a class due to the fact that they have to work a shift then they may prolong their college career. 4

5 The Survey No Time to Study Male Female Age: Class Standing : Freshman Sophomore Junior Senior GPA : How many hours do you work in one week? Do you schedule your class schedule around work? Do you schedule your work around your class schedule? How many hours do you study outside of class per week? What is your major? On average, how many hours of sleep do you get per night? How many units are you currently taking? Would you be willing to compromise work hours for study hours? 5

6 Survey Scope The scope of this survey was focused on Southern California Cal State Universities. These campuses include San Diego State University and California State University San Marcos. Within both Southern California Universities we received surveys ranging from many different majors. Although the majority seemed to come from the College of Business at California State University San Marcos, we received surveys from the College of Arts and Sciences, and College of Literature and Writing. The surveys we received varied from Accounting, Finance, Management, Marketing, Kinesiology, Communications, Human Development, Psychology, Biology, Criminology, and liberal studies. Given that our objective was focused on work hours and its effect on GPA the differences in each major was not an issue. Creating the Survey When our group first came together we convened in order to draft the survey. We spent a few hours determining what questions will provide us the most relevant and precise information to help us statistically analyze the data. Our survey was compromised of both closed ended questions and open ended questions. Although the only closed ended questions focused on class level standing and gender of the surveyed. Since we wanted the most precise and relevant information we developed the majority of the survey to be open ended questions which are questions that allow respondents the freedom to respond with any answer they felt answered the question. The questions ranged from their personal study hours, work hours, course load determined through units, and their preference when scheduling class and work. 6

7 Majors We surveyed a multitude of students across an array of majors. Since we wanted to broaden our survey results outside of California State University San Marcos we surveyed students from San Diego State University. We received surveys ranging from all different majors. As portrayed in the graph below, the majors range from Accounting to SSM. Due to the fact that the majority of the sureyeyed were taken from the College of Business in Markstein Hall the majority of majors are related to business. We can see that Accounting, Finance, Management, and Marketing majors are the substantial leading majority of students surveyed. Due to the fact that our mission is to determine the correlation between work load and G.P.A the difference in all the majors plays no significant role. 7

8 Class Standing The results that came from the survey had some interesting information that we learned. This first chart explains the distribution of different classes of the people we surveyed. The majority of the people that we surveyed were Juniors. The Juniors were at 60% compared to the other classes. We think that this was because the classes in which we handed out our surveys came from classes where the majority of the people where in their third year of school. The senior class was the second largest on the pie graph and they accounted for 33% of the people surveyed. This was because a majority of the surveys were given out to classmates in the business buildings which are done with their business core classes and are either a Junior or Senior. The students that were the minority of the surveys were the Sophomores and Freshman. The Sophomores were at 4% and the Freshman were at 3%, so as you can see there was not much of a difference between the two lower classes. The reasoning behind this is because the only classes that we would have with sophomores and freshman are our lower or upper GE classes. Out of our group we did not have many GE classes in which we handed out surveys which show why there was such a little percentage of both of the lower classes. We do not think that by having a lower amount of lower classes was going to hurt our results because we are all students and some of us have to work during our school career while some of us don t. 8

9 Age Scope On our survey we asked what age the students were and we did this because age might have to do with were you have to work or not. We used a bar chart to show the inforation that we got on age. The range on the age was large and it consisted with people from 18 to 47 years of age. The graph starts off small and then goes up really quickly around the early to mid twentys. We think that the graph looks this way because when you graduate from high school you are most likely 18 years of age which is the age most freshmen are at CSUSM. Since they were the least amount of people surved it makes sense why the age is shown so low on the bar graph. It was no suprize to us to see that the age group with the most people was between because that is the typical age of the Jouniors, which was the largest class that we surved. It was intersting to see the wide range of people that we surved that were 26 and older. There was actully a lot more than we thought we would get since our sample size was not a huge amount of people. By looking at this graph you can assume that age will have to do with weather you are woring or not because it will also show most likely if you are still living at home with you parants whom are supporting you. 9

10 GPA Range This graph shows the results of the GPA range of the students that we surved. This historgram shows that the GPA range is from 2.0 to 4.0. From looking at the results, the most common GPA range was between 3.0 to 3.24 which is a little bit above average. Knowing that the students that did not work had better GPAs than those who did work we can assume that a majority of them are in the 3.0 to 4.0 range. Even though the diffrence between those who worked and did not work was not that much there is still a diffrence which can be used as evidence for saying that students who do not work get beter grades. On a positve note, the majority of the people that we interviewed had good GPA scores. Though GPA scores are important to some, people study hard regardles s of having work or not. However, while others who work or do not work don t study that much and still maintain decent grades shows us that some people are just good at taking in information and good at taking tests. 10

11 After looking at the histogram of GPA, we can now look at the average, standard deviation, and variance. Out of the 141 students that have been surveyed, including working and non working students, there is a 3.14 GPA average. With a 3.14 GPA, a B level performance, we felt that our study was a good representation of the general population. I know that most students want to have an A average but considering that a lot of them work a lot and would take a B. A B is above average and is what we were expecting in the results. After we got the average we calculated the standard deviation which was Then from that we got the variance, which is Gender Distribution There was an interesting Gender Distribution out of the 141 students that were surveyed. The first thing that we noticed was that there were 60 male students and 81 female students. We have heard before that there are more females going to college than males. When we conducted are surveys we got the same results as there were more women surveyed than men. When you look at the graph the percentage difference is 57% female and 43% male. So, over 50% of the students that were surveyed were female. That is a pretty big difference and would be an interesting subject to look into more deeply. Given the age discrepancy in our study, this could mean that at present times, females are not just staying at home and taking care of the house and the children. A lot of them are going to school, working, and taking care of their children. 11

12 Data Analysis We began our analysis with correlations between all independent variables and the GPA of each student surveyed. Our attempt to find significant correlations had failed given that we hadn t separated the employed from the unemployed students. Our correlation calculation for hours worked to GPA and hours studied to GPA proved to be extremely insignificant. However, with feedback from outside sources (Professor Fang), we understood that in order to get a more visible difference from our sample, we would need to compare our employed student sample with our unemployed student sample. We constructed our data by means of correlation, r- squares, scatter plots, hypothesis testing, regression analysis, and the differences in sample means. Employed vs. Unemployed After separating the employed from the unemployed students, we found that employed students have, on average, a higher GPA than unemployed students. With a sample size of 28, unemployed students maintain an average GPA of 3.327, while the employed sample of 113 had an average GPA of The confidence intervals were calculated with confidence levels of 95%. Unemployed students GPA had a half width of.193 (3.13~3.52), while employed students had a half width of.088 (3.01~3.18). By calculating the standard deviation of each sample group, a t test for the difference in the two means was prepared. With the null hypothesis stating a difference less than 0, and a level of significance of.05, our t value proved to be greater than the upper critical value. Therefore, we rejected the null hypothesis and concluded that there was a difference greater than 0 between the two samples. This test helped us validate the purpose behind our survey. We had found a significant difference between our two sample groups. 12

13 t Test for Differences in Two Means Employed vs. Unemployed Data Hypothesized Difference 0 Level of Significance 0.05 Population 1 Sample Sample Size 28 Sample Mean Sample Standard Deviation 0.32 Population 2 Sample Sample Size 113 Sample Mean 3.09 Sample Standard Deviation 0.6 Intermediate Calculations Population 1 Sample Degrees of Freedom 27 Population 2 Sample Degrees of Freedom 112 Total Degrees of Freedom 139 Pooled Variance Difference in Sample Means t Test Statistic Upper-Tail Test Upper Critical Value p-value Reject the null hypothesis Strength and Variation between Variables and GPA Tests to calculate correlations were done using correlation equations and scatter plots. All correlations were done using the employed student sample in order to distinguish any significant relationships causing the decline in GPA. The scatter plots shown below illustrate the lack of correlation between the variables being tested. The data is greatly disbursed resulting in little or no correlation between the variables. The more hours worked, slept, or time spent studying didn t correlate with one s GPA. All r-squares, or the variation in GPA given the variation in the 13

14 independent value, are between.001 and Therefore, the variation of hours worked, hours studied, and sleep hours has little to no effect on the variation of one s GPA if currently employed. 14

15 Once again using the employed sample data, a correlation grid test was prepared comparing all variables. With a cutoff r-value of , which was calculated using a level of significance of.05, all values in yellow prove to have a somewhat significant correlation. Referring to the correlation grid below, hours studied to GPA, and sleep to work hours had the greatest correlations. The grid shows that as hours studied increases, so does GPA; however, no conclusions were made on these relatively insignificant correlation percentages. GPA work hours/week hours studied outside of class sleep units GPA 1 work hours/week hours studied outside of class sleep units Regression A regression analysis was conducted to construct a simple linear regression equation given the independent variables. Work hours per week, hours studied outside of class per week, sleep hours per night, and units per semester were all included in the regression model. However, work 15

16 hours per week and sleep hours per night had high p-values, meaning their variation had virtually no effect on employed students GPA. Nevertheless, the significant F value of.0324 (<.05) confirms that the linear equation has significance. However, by removing the two high p-value variables, the simple linear regression model of y= (hours studied) +.026(units), is more significant. In the end, the r square value of.092 reminds you of the lack of strength the independent variables have on GPA. Once again, no conclusions were made on this data as the analysis and its output confirmed the lack of necessary significance needed. SUMMARY OUTPUT Regression Statistics Multiple R R Square Adjusted R Square Standard Error Observations 113 ANOVA df SS MS F Significance F Regression Residual Total Coefficients Standard Error t Stat P value Lower 95% Upper 95%Lower 95.0%Upper 95.0% Intercept E work hours/week hours studied outside of class sleep units Willing to Compromise The purpose of asking employed students if they were willing to compromise work hours for study hours was to see if people felt like they needed more time to help their academic performance. The data resulted with 78 students willing to compromise work hours and 35 students believing that they either had sufficient time to study or needed the money rather than the grade. Not surprisingly, the 78 students had a lower average GPA at compared to the 35 students who answered no with a GPA of However, when testing for a difference 16

17 between the two means, the null hypothesis was not rejected, concluding that there was a difference less than 0. t Test for Differences in Two Means for Compromise t Test Statistic Upper-Tail Test Upper Critical Value p-value Do not reject the null hypothesis Scheduling Students who schedule their class schedule around their work hours (51 students) have, on average, a GPA. On the other hand, students who schedule their works hours around their class schedule (62students) claim to have an average GPA of The difference in the two means is clearly distinguishable; however, when running a t-test the standard deviation of the samples decreases the difference. The null hypothesis of a difference less than zero is not rejected once again, which essentially means that there is no significant difference between the different schedules employed students make. t Test for Differences in Two Means for scheduling t Test Statistic Upper-Tail Test Upper Critical Value p-value Do not reject the null hypothesis 17

18 Other Findings There are many different factors that affect a college student s overall academic performance. Some of the major factors that can contribute to lower academic performance are, whether or not the student is employed, the amount of sleep they get per night, as well as the amount of hours they spend studying. All of these variables have a role in how well a student does in school, and some have more of an impact than others. Unfortunately, the data gathered from this study wasn t significant enough to establish an understanding as to why there is a difference between employed and unemployed students GPA. However, comparing two sets of means can sometimes be as logical as any statistical findings. The observation of sleeping hours and hours studied could pose as a possible explanation of the difference between employed and unemployed students. Sleeping Patterns When analyzing our data, we found there to be no direct correlation between hours slept and whether the students surveyed were employed or unemployed. However, there were some surprising results between the two variables. On average, employed students slept hours a night, whereas unemployed students slept about hours a night. The difference between the hours slept for employed and unemployed students is 42.8 minutes a night. Basically, students who weren t employed had close to ¾ of an hour extra slept than students who were employed. This may not seem like a significant difference from a statistical standpoint, but from a logical view, having an extra 45minutes of sleep can have a major impact on students. For example, if an unemployed and employed student forgot to set their alarm the 18

19 night before, the unemployed student has an extra 45 minute buffer to wake up and still make it to class on time. Studying Habits (Hours Studied) Another interesting find when analyzing our data was that students who were employed studied, on average, more than unemployed students. Employed students studied an average of hours, where unemployed students studied hours per week. The difference between the variables is minutes per week more that employed students studied. It was very surprising to see that employed students studied more than unemployed students. Before analyzing the data we all thought that unemployed students would study more hours than employed students, just from the simple fact that they have more free time than students working. Logical Explanation We concluded that with employed students studying longer, but sleeping around 45 minutes less per night, we believe that their studying is being done at the expense of their sleep. Unemployed students have the privilege of a higher probability of studying during the day while employed students are working. Studies have shown that studying during the day proves to be more effective than studying late at night. Studies have also shown that not getting enough sleep can have a negative impact on a student s overall performance at school. Employed students are studying longer and sleeping less which results in having a lower GPA than unemployed students. In our study we found that there was a difference of GPA for unemployed students. The difference between a 2.77 GPA and a 3.0 GPA can be the difference between receiving a scholarship and not receiving a scholarship. 19

20 Evidence In a report from the NASPA Journal, 2006, titled The Relationship of University Students Sleep Habits and Academic Motivation, by Kellah Edens, it states, Recent studies have found that college students are sleeping less number of hours per night during the week (NSF, 2005) and that insufficient sleep is reported as one of the three health impediments to academic performance (ACHA, 2005). Sleep is one of the most important factors in students as well as the general populations, overall performance. It can affect how well a student does on a test, or how well a business professional does on their presentation. In the article, College students performance suffers from lack of sleep, by Tali Yahalom, USA Today, it talks about some of the effects on students and student athletes when they don t get enough sleep. The article says that in A study at St. Lawrence University in Canton, N.Y All-nighters are not an effective way to succeed in school. Psychology Professor Pamela Thatcher who conducted the study says, You can t do your best work when you re sleep deprived. Obviously, sleep is an important factor in a student s ability to perform, but it also seems that study habits as well as the time a student studies also affects their overall GPA and academic performance. Although employed students study more, their overall GPA is lower than unemployed students who study less. This can be attributed to many other factors besides employment. Some of those factors can be the time of day a student studies, their access to certain resources, their living situation, etc. All of these factors play a role in the amount of time a student studies and more importantly the quality of study. An important question that was asked on our survey is whether or not the student schedules school around work or work around school. What makes this an important factor when analyzing a student s academic performance, is whether or not school is their number one priority. Those students, who schedule work around school, will 20

21 more than likely take time to study during the day, rather than those who place school around work. The problem with survey questions are that they are completely subjective and rely completely on the honesty of the person being surveyed. Also there are many factors that can be attributed to a student placing work before school, such as they have too, in order to pay for school, their mortgage, etc. Intangibles From the results of our sample, we concluded that there is not a positive or negative correlation between two variables and the factors that affect the two s overall GPA. With such a small sample size of one hundred and forty two students, it wasn t possible to find a significant correlation between the two variables. One of the biggest problems when conducting a survey is that they surveyor is at the mercy of the one being surveyed. The data collected cannot be looked at as completely accurate because it s solely based on the honor system. Who s to know whether or not the one being surveyed is being truthful? The larger the sample size, the more accurate the results will be, but there still may not be a significant correlation between unemployed and employed students, hours studied, and hours of sleep. Out of all the different factors that can affect a student s academic performance, one factor that seems immeasurable is a student s desire and or attitude towards their academics. A students attitude can have a major impact on their performance at school. Some students are just happy to barely get through, while some students will do whatever it takes to get that A. It s difficult to measure student s ulterior motives when it comes to performance at school and this can play a significant factor measuring their academic performance. In the end, with the data that we collected, we concluded that students who are unemployed have a significantly 21

22 higher GPA than employed students; therefore, the best option to maintaining a higher GPA is to, if possible, try and not work at all during the school semester. Errors in the survey The information that we gathered from our survey did prove to be very helpful. However, we feel that we could have asked a more precise question concerning the time of day students study. The original question stated: How many hours do you study outside of class per week? Our alteration would ask how many hours do you study outside of class per week and what time of day do you usually study. We would have also liked to have asked more open ended questions concerning the lifestyles of the surveyed. These questions would include if the student was dependent or not? This would help us to understand if they have a need to work more hours. We would also like to ask a question concerning the student s marital status. If a student is married they may have to work more hours to support the family. If a student does not work we would still like to see if there is something that may have a correlation with a lower G.P.A. This can be through extra curricular activities. 22

23 Works Cited Edens, K. (2006). The Relationship of University Students' Sleep Habits. NASPA Journal, 43(3), Retrieved May 3, 2008, from Yahalom, T. (2007, Sep. 16). College Students' Performance Suffers from Lack of Sleep. Retrieved May 2, 2008, from 23