Abstract
Before commencing with the study, I decided to do some background research on this age-old debate. I searched Google for background information on Coke vs. Pepsi and some previous studies. All of the links I found were for a study done by Dr. Read Montague at the Baylor College of Medicine. The researchers wanted to know how the brain responds to cola and what (if any) impact it has on the brain. They carried out several taste tests where each subject first tested drinks in cups. Then, the subjects were loaded into an MRI machine, where they were given tiny squirts of Coke or Pepsi while a machine monitored brain activity. One drink was labeled and the other was anonymous. The researchers found that people were equally likely to choose Coke or Pepsi as the favorite when the study was anonymous. However, when only one drink was labeled, they always tended to choose the labeled drink for Coke but not Pepsi. The researchers found that a part of the brain that responds to rewards lights up on the brain scan, but when one drink was labeled Coke or Pepsi, other parts of the brain lit up, including sections that dealt with memory and cognitive control. The conclusion is that your brain does not respond just to flavor but also memories.
I obtained the samples from the master list of all the students. I took the list, and through Minitab, separated the list into four lists of students from each grade. From those lists, I took a stratified random sample of 30 students from each grade. I used Minitab to get a list of 30 students from each grade. I obtained the list by taking a sample from the columns. The sample that I chose was 30 students from each grade. I wasn’t really monitoring the amount of males and amount of females that the study had, but it was pretty even. There were 62 total males and 58 total females. There were 60 total underclassmen (30 freshmen & 30 sophomores), and there were 60 total upperclassmen (30 juniors and 30 seniors). There were 18 males and 12 females for the freshmen, 15 males and 15 females for sophomores, 7 males and 23 females for juniors, and 18 males and 12 females for seniors. I am going by grade and not by grade and gender due to the skewed results. The periods the students were in was somewhat normally distributed since 60% of the students (72/120) were in 4/5, 6/7, or 8/9. 24 out of 120 students were in each one of those periods. 45 out of the other 48 students were in 1, 2, 3, 10, or 11 and three (all senior males) were disqualified because they went to full-time Polaris. There are a total of 117 students in the sample pool. Of these 117 students, 95 responded. I alternated Coke and Pepsi on the surveys to prevent selection bias. 32 out of 95 people preferred Pepsi, and 63 out of 95 preferred Coke.
I ran several significance tests from the data. The first was a two proportion z-test to see if the proportion of females that prefer Pepsi was different than the proportion of males that preferred Pepsi. I ran the test and got a p-value (p1 - p2) of .949. The second test was a chi-squared test of the observed and expected values of the proportions of people in each grade that preferred Pepsi were all equal. The test results for that chi-square test came out with a p-value of 0.626. The third test was a 1-proportion z-test to see if the sample proportion of the people that prefer Pepsi is equal or not equal to the proportion of people that prefer Coke. I got a p-value of .004, so I can reject the null hypothesis and say that the proportion of people that prefer Coke is greater than the amount of people that prefer Pepsi.
The most prevalent weakness in my study was non-response bias. Of the 120 people that I had selected from the sample, 3 were excluded because they went to Polaris full-time. That left the amount of surveys that actually went out to 117. Of the 117 surveys that went out, 95 responded. Counting the three Polaris kids that were excluded from the survey, less than 80% of those sampled responded to the survey. Those 25 responses, even if they all had said they preferred Pepsi, would not have changed the outcome of the study since 62 out of the 95 people preferred Coca-Cola. The results of the significance test might have been different, as we may have ended up failing to reject the null hypothesis. None of the surveys that I had were voided because of writing in a response that was not on the sheet. There was some selection bias in the study, especially because there were 23 junior girls in the original sample of the classes, but as far as the other classes (freshmen, sophomores and seniors), there was almost no bias toward one particular gender. I could have also asked for reasons why people prefer Pepsi or Coke so that I could see their reasoning for choosing one or the other, whether it is because of taste, cost, or something else along those lines. In a simple study such as Coke vs. Pepsi, there is almost no bias. There are no real apparent confounding variables in the study that are apparent.
I could extrapolate the results to about all teenagers in the United States. I got a pretty diverse sample of teenagers in most areas. There is no known demographic appeal for whether or not people of certain backgrounds prefer one or the other, so as such, I think that my findings could be extrapolated to all teenagers in the U.S. As I had discussed earlier, Dr. Read Montague's studies in the field with his fMRI study in whether or not why people choose Coke or Pepsi is truly the most definitive study available in the field right now. As a result, there is not too much work that needs to be done in the field, but I would like to see if one group of people has a particular appeal by income level (whether or not rich or poor people tend to prefer one type or another), gender (whether or not males or females tend to prefer one or another since my results were semi inconclusive), and by race and ethnic background. That would determine whether or not Pepsi or Coke is more popular in different parts of the world. Any of these studies would be valuable marketing tools for Coke and Pepsi.
I ran three different hypothesis tests, and as such, I am going to make three different conclusions. I failed to reject the first null hypothesis that the proportion of females that preferred Pepsi was the same as the proportion of the males that preferred Pepsi since my p-value > α. Therefore, there is not enough significant information to prove that the amount of males that prefer Pepsi is significantly different than the amount of females that prefer Pepsi. I failed to reject the second null hypothesis that the proportion of people that prefer Pepsi is different by grade since the distribution is in line with the expected values. Therefore, there is not enough significant information to prove that the distribution of people that prefer Pepsi by grade is significantly skewed toward one grade or another. I rejected the third null hypothesis at the .05 level because our p-value < α. Therefore, there is enough significant evidence to prove that the true mean proportion of people that prefer Pepsi is not the same as people that prefer Coke.
The evidence that I have presented hear in this study shows that Pepsi is not the drink of a new generation as they have claimed to be. They have obviously failed on their marketing plans and need to restructure their marketing plans. Pepsi and Coke need to look at who their product is most popular with and restructure their marketing plans to go after those people that they already have in their core group. However, these results are not just for taste. Sales play an even larger factor, and just because more people prefer the taste of Coke does not exactly mean that those people tend to drink Coke more. I did not look up sales of the two drinks during my background research, but they are an important factor to whether or not more people tend to drink Pepsi or Coke.