# Consumption of Alcohol by Currently Enrolled College Students and Recent Graduates Essay

Kristen Petreyko Statistics: MA 105 Statistics Project Due: December 17, 2012 INTRODUCTION Being a recent graduate of the University of Massachusetts Amherst, I have found that a person’s whole lifestyle changes with the simple act of graduating college. One of the most obvious lifestyle modifications for many recent graduates is the decrease in the amount of partying, and the realization that it ends when college does. I conducted a statistical analysis on the consumption of alcohol for those still enrolled in undergraduate programs compared to the consumption of recent graduates.

My research objectives include the following: * Document with statistical evidence that UMass Amherst students currently enrolled in courses consume more alcohol than UMass Amherst recent graduates. * Use hypothesis testing and a confidence interval to test the claim “the mean of days per week and the mean of the amount of alcohol consumed each time one drinks will be greater for UMass students than those who’ve recently graduated” * Establish a connection between the amount of alcohol consumed each time one drinks and the number of days per week alcohol is consumed.

The null hypothesis is: the mean of days per week, and mean of amount of alcohol per sitting, is equal for those enrolled at UMass Amherst and recent UMass Alum. This null hypothesis, and all null hypotheses, indicates that there is no change in the two means. The alternative hypothesis which is also the claim in this case is, as stated above, “the mean of days per week and the mean of the amount of alcohol consumed each time one drinks will be greater for UMass students than those who’ve recently graduated from UMass. I chose this claim because from speaking to fellow recent graduates and witnessing first hand, I find that while in college there is more time to drink and it is a much more alcohol induced atmosphere. One major aspect is the fact that in college Thursday evenings are considered “the weekend” and a big night to go out partying. This increases the amount of days per week a college student parties and could partake in alcohol consumption. Time is also another strong factor in this claim. Majority of college students do not have jobs, or if they do they are part time and not taken seriously.

Recent graduates are busy working, finding a job, or even attending/applying to graduate schools. Age is a factor as well. College students are on average younger than recent graduates and can explain some of the immaturity and cause for excessive drinking. This claim takes all of these factors into consideration and will be either proven correct or incorrect based on the results found from a survey conducted. A seven question survey was used to collect the data needed to test this claim. The questionnaire surveyed the population of recently graduated UMass Amherst men and women and current UMass students.

I conducted the survey on the campus of UMass Amherst in October 2012 during UMass’s Alumni Weekend. I attended several parties and bars filled with current students and recent graduates and asked thirty current students, as well as, thirty recent graduates to fill out the survey. I made sure to inform them the information would be kept anonymous and also that it was a voluntary survey and they did not have to participate. I finished the collection of data by the end of the Alumni Weekend. It took me Saturday morning – Sunday morning to be able to successfully ask sixty different men and women at UMass to complete the survey.

The majority of the completed surveys were done at a bar Saturday afternoon where the recent graduates and current students collaborated to celebrate the alumnae. Overall, the collection of the data was not too difficult. Majority of the people were willing and interested in filling out the survey. I feel as though the sample would accurately represent the two populations. There is a fairly average number of males and females in both categories. One place that may be argued is that for the college students still enrolled, it might not accurately represent freshmen-sophomores.

The majority of the college students were seniors and juniors, although some freshmen and sophomores were surveyed. Statistically, a voluntary response sample, which is when the respondents chose to participate or not, is not considered a sample that cannot be used to make conclusions about a population so in conclusion, this sample should not be used to make a conclusion about the population of UMass students or UMass recent graduates. BODY OF THE PROJECT The hypothesis of what is expected is that as a whole UMass Amherst college tudents will consume more alcohol than UMass Amherst recent graduates. In particular, the mean of the days per week alcohol is consumption and the mean of the amount of alcohol drank each day one drinks will be greater for UMass Amherst students than the means of UMass Amherst recent graduates. I am interested in calculating if the average amount of alcohol each time one drinks is exceptionally higher for those in college than not. Also, if the days per week correspond to the amount drank and if the mean of days per week and mean of amount drank will differ for students and/or alumni.

RESULTS Descriptive Statistical Analysis Consumption of alcohol (days per week): | UMass Amherst Current Students (n=35)| UMass Amherst Recent Graduates (n=35)| Mean: Mean=| Mean=3. 357 days per week | Mean=2. 643 days per week | Median: Median=Center value when arranged in increasing order| Median=3. 5| Median=3| Midrange: Midrange=| Midrange=4| Midrange=2. 5| Variance:Variance=| Variance= . 920| Variance= . 5966| Standard Deviation: Standard Deviation=| Standard Deviation= . 959| Standard Deviation= . 7724| Range:Range= (Max. Value)-(Min.

Value)| Range=4| Range=3| Amount of Consumption of Alcohol Each Time One Drinks (drinks per day/night of drinking):| UMass Amherst Current Students| UMass Amherst Recent Graduates| Mean: Mean=| Mean=7. 714| Mean=5. 686| Median: Median=Center value when arranged in increasing order| Median=9| Median=6| Midrange: Midrange=| Midrange=6. 75| Midrange=5. 75| Variance:Variance=| Variance=4. 710| Variance=5. 237| Standard Deviation: Standard Deviation=| Standard Deviation=2. 170| Standard Deviation=2. 288| Range:Range= (Max. Value)-(Min. Value)| Range=6. 5| Range=8. 5|

Participants Demographics| UMass Amherst Current Students| UMass Amherst Recent Graduates| Mean of Ages of Participants Mean=| Mean= 20. 6 years old| Mean= 22. 9 years old| Ratio of males:females: | M:F= 14:16=7:8 | M:F= 13:17 | Percent of participants who answered “Yes” | To the question: “would you say you drink more now than you did prior to attending UMass”: 100% | to the question: “would you say you drink less now being graduated from UMass”: 100%| FREQUENCY DISTRIBUTIONS OF DAYS PER WEEK OF ALCOHOL CONSUMPTION Frequencies of UMass Amherst Current Student: Days per week of alcohol consumption 3| 5| 4| 3. | 2. 5| 2| 3| 3| 3| 2| 3. 5| 2. 5| 2. 5| 3| 4. 5| 6| 2. 5| 3. 5| 3. 5| 5| 2. 5| 3. 5| 5| 4| 3. 5| 2| 3| 4| 4| 4| 3. 5| 3| 2. 5| 2. 5| 2| ———-| Frequency Distribution of UMass Amherst Current Students: Days per week of alcohol consumption Days per Week of alcohol consumption| Frequency| 1-1. 5| 0| 2-2. 5| 11| 3-3. 5| 14| 4-4. 5| 6| 5-5. 5| 3| 6-6. 5| 1| *Normal Distribution Frequencies of UMass Amherst Recent Graduations: Days per week of alcohol consumption 2| 1. 5| 3| 2. 5| 2| | 1| 3| 2| 3| 4| 3. 5| 2| 3| 4| 3. 5| 3| 3| 3| 2| 1. 5| 2| 3. 5| 2| | 3| 3| 3| 2. 5| 2| 3. 5| 4| 2. 5| 2| 2. 5| 1. 5| Frequency Distribution of UMass Amherst Recent Graduations: Days per week of alcohol consumption Days per Week of Alcohol Consumption| Frequency| 1-1. 5| 4| 2-2. 5| 13| 3-3. 5| 15| 4-4. 5| 3| 5-5. 5| 0| 6-6. 5| 0| FREQUENCY DISTRIBUTION OF AMOUNT OF ALCOHOL CONSUMED EACH DAY/NIGHT ONE DRINKS Frequencies of UMass Current Students: Amount of alcohol consumed each time one drinks 9| 10| 10| 9| 3. 5| 9| 10| 6| 6| 10| 10| 9| 9| 6| 6| 6| 6| 9| 9| 10| 10| 6| 3. 5| 9| 9| 9| 9| 3. 5| 9| 3. 5| 6| 9| 6| 6| 9| |

Frequency Distribution of UMass Current Students: Amount of alcohol consumed each time one drinks Amount of drinks consumed each time one drinks| Frequency| 1-2. 5| 0| 3-4. 5| 4| 5-6. 5| 10| 7-8. 5| 0| 9-10. 5| 21| Frequencies of UMass Recent Graduates: Amount of alcohol consumed each time one drinks 1. 5| 6| 2. 5| 6| 10| | 3| 6| 6| 10| 10| 3. 5| 5| 6| 3. 5| 3. 5| 6| 6| 6| 9| 3. 5| 6| 9| 6| 5| 6| 9| 3. 5| 6| 3. 5| 6| 3. 5| 9| 3. 5| 6| 4| Frequency Distribution of UMass Recent Graduates: Amount of alcohol consumed each time one drinks Amount of drinks consumed each time one drinks| Frequency| 1-2. | 2| 3-4. 5| 10| 5-6. 5| 16| 7-8. 5| 0| 9-10. 5| 7| PIE CHART OF RECENT UMASS GRADUATE AGES PIE CHART OF CURRENT UMASS STUDENT AGES Testing Claim Hypothesis Test used to test the following claim: The mean of days per week that current UMass Amherst students consume alcohol is greater than the mean of days per week that UMass Amherst recent graduates consume alcohol. H0: 1 =2 Null Hypothesis H1: 1>2Alternative Hypothesis 1: Mean of days per week that students currently enrolled at UMass Amherst consume alcohol 2: Mean of days per week that recently graduated UMass Alum consume alcohol Procedure: Requirements are met: * 1 and 2 are unknown and not assumed to be equal * Two samples are independent and are simple random samples * n1 and n2> 30 * Use test statistics formula : Test statistic=3. 4311 * Using table A-3: And finding the degrees of freedom: df= Where A= and B= * df=65. 03 * critical value = 1. 669 * Right tailed test, so p-value: area to right of test statistic= 0. 0005 Conclusion of test: Because the test statistic falls within critical region, eject null hypothesis and this sample data supports the claim that the mean of days per week that alcohol is consumed by current UMass students is greater than the mean of days per week of alcohol consumption by recent UMass graduates. Confidence Interval to test claim: The mean of days per week that current UMass Amherst students consume alcohol is greater than the mean of days per week that UMass Amherst recent graduates consume alcohol. Procedure: * Construct a 90% confidence interval to test claim * Find the margin of error E= = . 3472366 * Confidence Interval = = 0. 3667634 < ( < 1. 61237 Conclusion of confidence interval: We are 90% confident that the limits of 0. 3667634 days per week and 1. 061237 days per week actually contain the difference between the two population means. Because the limits do not contain 0 it is very unlikely that the two population means are equal. Hypothesis Test to test claim about mean of the amount of alcohol consumed each time one drinks: Claim is as follows: the mean amount of alcohol consumed each time one drinks for UMass students will be greater than the mean amount of alcohol consumed each time one drinks for recently graduated UMass alumnae.

Procedure: H0: 1 =2 Null Hypothesis H1: 1>2Alternative Hypothesis 1: Mean amount of alcohol consumed each time one drinks of students currently enrolled at UMass Amherst 2: Mean amount of alcohol consumed each time one drinks of recently graduated UMass Alum * Requirements are met: * 1 and 2 are unknown and not assumed to be equal * Two samples are independent and are simple random samples * n1 and n2> 30 * Use test statistics formula : Test statistic=3. 8047 * Using table A-3: And finding the degrees of freedom: df= Where A= and B= df= 67. 81 Critical value= 1. 668

Right tailed test, so p-value: area to right of test statistic: 0. 0002 Conclusion of test: Because the test statistic falls within critical region, reject null hypothesis. The sample data supports the claim that the mean amount of alcohol consumed each time one drinks by current UMass students is greater than the mean amount of alcohol consumed each time one drinks by recent UMass graduates. Confidence Interval to test claim: The mean amount of alcohol consumed each time one drinks for current UMass Amherst students is greater than the mean amount of alcohol consumed each time one drinks for UMass Amherst recent graduates.

Procedure: * Construct a 90% confidence interval to test claim * Find the margin of error E= = . 888882 * Confidence Interval = = 1. 139118 < ( < 2. 916882 Conclusion of confidence interval: We are 90% confident that the limits of 1. 139118 amount of alcohol each time one drinks and 2. 916882 amount of alcohol consumed each time one drinks contain the difference between the two population means. Because the limits do not contain 0 it is very unlikely that the two population means are equal. Correlation Results:

The days per week of alcohol consumption and amount of alcohol per day of current UMass students were compared to determine whether or not there is a linear correlation between the two. Correlation coefficient: r= .0045411 Critical value r= . 3338443 P-value: . 97935 Fail to reject the null hypothesis, sample does not provide enough evidence to support a linear correlation CONCLUSION After the study was completed, the conclusion states that the initial hypothesis that UMass students currently enrolled consume alcohol more often (more days a week), and also in greater quantities when they do consume it.

The claim that the mean of days per week that current students consumed alcohol would be greater than the mean of days per week of recent graduates of UMass Amherst was proven to be correct via the hypothesis test and the 90% confidence interval. Also, the second part of the claim stating that the mean amount of alcohol consumed each time one drinks would be greater for current UMass students versus recent UMass graduates was proven to be correct by a hypothesis test and 90% confidence interval.

There seems The results that were concluded were as I expected and what I guessed would be the results. Based on my four years at UMass and now the few shorts months being graduated, I noticed a drastic change in the lifestyles of myself and my fellow graduates. I was very confident about the first part of the claim being proven correct but I was unsure if the amount of alcohol consumed each time one drinks changed that much for current students and recent graduates. The data that was collected could be used for further studies to be done.

For example, one could compare the results of the days per week and/or amount consumed in one day/night for males versus females and/or for seniors versus all other grades. I noticed a difference in the amount of days per week and how many drinks per day for those who were seniors compared to the others. Also, the majority of males had higher numbers for both questions. I would recommend further studies testing if gender and grade in college affect the numbers and also it would be interesting to conduct a study of high school students and compare their data to those in college, and also to recent graduates and then compare all three.