|
1. Introduction:
With academics, athletics, L.A.C.’s, and numerous other extracurricular activities, Kalamazoo College students
are often faced with a lot to do in a little amount of time. With this is mind, we decided to conduct a study to determine
how much sleep Kalamazoo College students are receiving. According to the American Psychological Association (APA), “sleep
is essential for a person’s health and wellbeing” and “in general, most healthy adults…
need an average of eight hours of sleep. ” With the large amount of activities that students at Kalamazoo College
are faced with, sleep deprivation is a very important issue. In our study we hope to determine how much sleep students are
getting and what they are doing during the night with their time (when they aren’t sleeping). The study is important
because, as mentioned earlier, sleep affects students’ health and wellbeing. One simple example of how sleep deprivation
can be dangerous to students is when they are driving. According to the National Highway Traffic Safety Administration, “drowsiness
and fatigue cause more than 100,000 traffic accidents each year--and young drivers are at the wheel in more than half of these
crashes. ” Also, not getting enough sleep can hinder students’ academic performances. Again, according
to the APA, “A 1998 survey of more than 3,000 high-school students… found that students who reported that
they were getting C's, D's and F's in school obtained about 25 minutes less sleep and went to bed about 40 minutes later than
students who reported they were getting A's and B's.” Also, to our surprise and “to researchers' surprise,
in the past two decades studies have shown that teen-agers require considerably more sleep to perform optimally than do younger
children or adults. Starting around the beginning of puberty and continuing into their early 20s, (studies) have shown adolescents
need about 9.2 hours of sleep each night, compared with the 7.5 to 8 hours that adults need. ” This information
is relevant to our study because all of our subjects are in their late teens or early 20’s. A University of Minnesota
sleep study led some school officials to push back early morning start times because of “medical research showing
adolescents have a natural sleep pattern that leads to a late-to-bed, late-to-rise cycle… From the onset of puberty
until late teen years, the brain chemical melatonin, which is responsible for sleepiness, is secreted from approximately 11
p.m. until approximately 8 a.m., nine hours later. ” The results of our study should be able to help Kalamazoo College
officials with the scheduling of school sanctioned activities and may lead to changes like this.
2. Population and Sample:
Our sample was selected randomly with the random number generator on a calculator and the newest student directory. We
numbered the students in alphabetical order so that each number from the calculator would correspond to one student. Our population
of interest was Kalamazoo College students, although we did omit seniors because most of them lived off campus. We chose 80
students to sample. After printing off 80 surveys, we took them to the dorm rooms of the randomly selected students. We asked
the students if they were willing to take a sleep survey for an economics class project and offered them a lollipop upon completion.
If the student was not there we taped the survey to their door with a lollipop and asked the student if they would re-tape
it to the door when finished. We handed out the surveys on Wednesday, (so information would be fresh in their mind) and collected
the taped surveys on Thursday night. Our sample size was 36 out of 80 surveys. The response rate was then 36/80 or 45 percent.
B Confidence Intervals for important variables:
I) Mean CI
These are the 95 percent confidence intervals of hours of sleep for each day of the week surveyed.
Friday:
(6.18, 7.58)
Saturday:
(6.87, 8.2)
Sunday:
(6.21, 7.37)
Monday:
(5.36, 6.79)
Tuesday:
(4.75, 6.34)
Notice how eight hours does not fall into any of these intervals. Weekdays (Mondays and Tuesdays) have the lowest amounts
of sleep while Friday and Saturday are the highest.
II) Proportion CI
These are the 95 percent confidence intervals of the proportion of students who did get at least eight hours of sleep
on each particular day.
Friday:
(.318, .342)
Saturday:
(.514, .544)
Sunday:
(.213, .232))
Monday:
(.24, .26)
Tuesday:
(.132, .145)
These statistics show that the day that the most percentage of students got at least eight hours of sleep on is Saturday,
in which 51 to 54 percent of students did so. The day that the smallest percentages of students received eight hours of sleep
was Tuesday. The confidence interval shows that only 13 to 14.5 percent of students will get at least hours of sleep according
to this study.
C Hypothesis testing of means or proportions:
These are one-tailed hypothesis tests of whether or not students get eight hours of sleep for each day surveyed. We hypothesized
that for each day of the week in the survey, students will get less than eight hours of sleep.
Friday:
(6.79, 6.97)
Saturday:
(7.44, 7.620)
Sunday:
(6.65, 6.81)
Monday:
(7.43, 7.63)
Tuesday:
(5.45, 5.66)
The reason we chose eight hours (instead of 9.2 hours) is because this is the amount of sleep recommended to adults by
the American Psychological Association. We realize that recently, “starting around the beginning of puberty and
continuing into their early 20s, (studies) have shown adolescents need about 9.2 hours of sleep each night. ” However,
we think the eight hours figure is safer to use and smarter to use since college students are preparing for the workforce
as adults. The results show that for no days of the week will eight hours fit into any of these confidence intervals. Therefore,
our hypothesis that students will get less than eight hours of sleep is correct.
D Hypothesis testing of the difference of means or proportions:
These are hypothesis tests between the differences in three different groups. We hypothesized I) that there will not be
a significant difference in the sleep habits of men and women,
II) that GPA’s of 3.0 and above will get significantly more sleep than GPA’s below 3.0.
and III) that students will get more sleep on a weekend day than on a weekday.
I) The null hypothesis here is that mean one minus mean two equals does not equal zero, while the alternative hypothesis
is that mean one minus mean two equals zero. To find the means of each sex we must first find the mean hours of sleep that
each subject had in their five days from the survey. After doing this and making a list we found that the mean hours of sleep
for men was 6.81 and the standard deviation was 1.05. The mean hours of sleep for women was 6.35 and the standard deviation
was .69. The difference in means between men and women is .46. From this data we can compute differences hypothesis test using
degrees of freedom of 16 because there were 17 males and this is the smaller of the two n values. This is a two-tailed test.
Confidence Interval: (-1.15, 1.15)
Since .46 (6.81-6.35) falls in this confidence interval we cannot reject the null hypothesis and we can be 95 percent
confident that there is little difference between the sleep habits of men and women at Kalamazoo College.
II) We will need to use similar procedures to determine if GPA’s of 3.0 and above get more sleep than GPA’s
below 3.0. For GPA’s above 3.0 the mean hours of sleep is 6.92 with standard deviation .94. For GPA’s
below 3.0 the mean hours of sleep is 6.33 with standard deviation 1.01. This is a one-tailed z-test with degrees of freedom
12. We hypothesized that students with higher GPA’s will get more sleep so our null hypothesis is that mean one
minus mean two equals zero, while the alternative hypothesis is that mean one is greater than mean two. The difference in
means is .59.
Confidence Interval: (-.2, .2)
Since .59 falls outside of this range we can reject the null hypothesis and be 95 percent certain that there is a significant
difference in the sleep habits of those with a GPA above 3.0 and those with a GPA lower than 3.0.
III) We decided to use Saturday for the weekend day because this is a day without classes and a day where there is no
classes the next day and we decided to use Monday as the weekday for the same reasons. The null hypothesis is that mean one
minus mean two equals zero and the alternative hypothesis is that mean one is greater than mean two. This is a one-tailed
t-test. The differences in means is 7.53-6.88 which equals .65.
Confidence Interval: (-.18, .18)
Since .65 falls outside of this interval we can be 95 percent confident that students get significantly more sleep on
weekend days than weekdays.
4. Conclusions:
Our statistical analysis shows that, on average students are getting nearly enough sleep. On school days, when students
should be getting the most sleep so that their brain functions the best, students are getting the least amount of sleep. The
recommended amounts of sleep range from eight hours a night for adults to 9.2 hours a night for college students. The most
sleep students got per night on average was Saturday nights with 7.5 – still not eight hours. The smallest average
amount of sleep was on Tuesday’s when students averaged an incredibly meager 5.5 hours of sleep.
From the results of our study and from our research we can see that students are not getting enough sleep due to a combination
of biological and social factors. We can also see that there is a correlation between what grades students get and how much
sleep they get, proving how important it is to get a good nights rest. While we think Kalamazoo College officials can learn
a lot about students’ nightly habits from our study, there are various limitations to it. For example, there is
a lot of rounding when taking data and we rely a lot on students’ memory. Also, we would have liked to include all
Kalamazoo College students in our population of interest, but since seniors mostly all live off campus they were excluded.
Further studies could expand this topic to different times of the year, all seven days of the week and could also look at
multiple colleges and universities. It would be interesting to see how Kalamazoo College compares to other schools. Hopefully,
our research will help lead school officials to a better understanding of students’ sleep habits and maybe, if necessary,
change schedules accordingly.
5. References:
http://www.apa.org/topics/whysleep.html
http://www.apa.org/monitor/oct01/sleepteen.html
http://www.apa.org/monitor/oct01/sleepteen.html
http://education.umn.edu/Pubs/ResearchWorks/sleep.html
http://www.apa.org/topics/whysleep.html
http://www.apa.org/monitor/oct01/sleepteen.html
Kalamazoo College Student Sleep Habits
Michael E. Leifeld and Nicholas Z. Vanden Heuvel
Quantitative Methods II
5-26-06
|
