Report

I am comparing two brands of gum based on flavor. Gum brand A is Orbit and Gum brand B is Trident. Both brands are some type of mint flavor. A sample of people from my church are tested to see which gum brand has the better taste. My null hypothesis is that students will not say there is a difference between brand A and brand B. Each subject is given a piece of gum in random brand order. The subject is told whether it is brand A or brand B, but are not told the brand name. The subjects then chew the gum and rate it solely based on flavor. 0 being the worst possible and 10 being the best. Next, the subjects will eat how many pretzels they feel they need to eat and a drink of water in order to wash out the taste. The subjects will chew the other brand and rate it. Once again, the data will be collected in ratings solely based on taste. After I did the experiment, the data revealed that Brand B (Trident) is better than Brand A (Orbit). The mean rating for Orbit is 6.3 and the mean rating for Trident is 7.1. I did a 2 sample Z Test, and the p-value is .1226. Since the p-value is low, I fail to reject the null hypothesis. Therefore, people say that there is a difference between Brand A (Orbit) and Brand B (Trident). According to an experiment, Trident gum is better than Orbit gum. Seven out of ten people agree that Trident is better. The other 3 like the mint taste of Orbit better than the taste of Trident. The subjects were told to rate the taste of the gum from 0 to 10 with 10 being the best. Some of the problems I had were finding the time to get the test subjects to taste since my schedule is different than those chewing the gum. Also a possibility of bias could be because the subjects were different ages. Since the sample was taken from my church, some people were more elderly then others. Perhaps the younger test tasters have better taste buds the older ones. To improve the test, I could have taken a smaller sample just from my youth group that way the same generation would test the gum.