STATS 212 Project
The data I collected was a Chi Square Independence test to determine whether or not if college males drank and if it had a direct affect on the amount they exercise. In my own opinion I don’t believe there is a direct correlation between the two because there are to other factors that play a greater role in determining exercise levels in males. Although this study came to mind during a conversation I has having with a few fraternity brothers. The conversation started out with how each guy was “buff” and “ripped” before coming to college, than transitioned into some guys thinking that drinking or not has an direct impact on a guys exercise level. The amount a guy exercises may be correlated with whether not he drank but then I thought to myself that drinking doesn’t determine a person or male in this case exercise habits, because I knew of a few cases where there were guys who drank and still went and exercised regularly. Drinking doesn’t play a direct role in exercise levels because there are to many other factors that play a part, but given my what I thought on the topic I felt that a good alternative hypothesis would be to test that drinking or not doesn’t have a direct affect on a males exercise levels, compared to the null hypotheses that it does have an effect on weather a guy drinks or not. I collected my data in four simple steps first I typed and printed out a two yes or no question survey containing the following questions “do you drink “ & how often do you exercise (regularly, sometimes, never) to help gather the data I needed. Second, I used a sample size of 56 males. The first 41 of my sample were all members of my fraternity and the other 15 were all guy on the floor of my dorm. Third, I collected the data from the members of my fraternity which was an easy task because all 41 of my sample was at our chapter meeting, had pasted out the survey I created containing the two questions after our chapter meeting was over, than recorded the data on my computer. Lastly, after collecting and recording my pervious data I got the data from all the guys on my floor which was a little more difficult because I wasn’t aware of the times guys were available, and I ran out of surveys, so I went door to door and asked the two questions to 15 guys and recorded the data in a notebook. After I had all the data a created a contingency table containing my results. After I made my contingency table the results are as followed. Out of the 56 people 22 males drink and 34 of them do not drink, breaking this down further out of the 22 males that drink 5 work out regularly, 8 never work out, and 9 do sometimes. With the 34 people that don’t drink 17 workout regularly, 6 never do, and 11 work out sometimes. Once the table was complete I conducted a Chi Squared test by first finding the expected values, the chi squared test statistic, and P-value. I had all of those calculations I could see if there was enough information to tell if my hypothesis correct or not. To summarize my results, the alternative hypothesis that the amount a college male exercised is not a direct correlation to if they drink or not is false, according to my data whether or not if a person was to drank showed a significant difference in the amount that they went and worked out. We know that when there is a small P-value it usually suggests that the observed data is inconsistent with the assumption that the null hypothesis is true, and that hypothesis must be rejected, which is true in this case. It is obvious that we reject the alternate hypothesis just by looking at the expected values as well you can see that with a random variable the evidence still shows that drinking or not and the amount a college male exercises are not independent variables.
Once I found my P-value of 0.0966 I…