By Maddy Reilly, Kasey Brisard, and Gabby DePhillips
Our Research Question is “on average do students at Georgia College prefer Sprite or
Sierra Mist?” Our hypothesis we were testing was there would be a preference for
Sprite among Georgia College students to Sierra Mist because of our location near
the Atlanta Coke factory (coke manufactures sprite).
We expected the mean difference would not be equal to zero.
Our population of interest was GC students and we selected
a sample of 30 students on campus to participate in our hypothesis
The Research variables were the difference between Sprite and Sierra Mist; we
obtained flavor ratings on each of the sodas on a scale of 1-10, 10 being the best taste and
1 being the worst.
We each gathered 10 people to test. Labeled cups A and B and gave them a blind
taste test asking them to rate the flavor on a scale of 1-10, 10 being the best tasting, how each
soda tasted. We then transferred this data to statcrunch.
To avoid bias we all sampled different people. We believe our sample is
representative because we all surveyed in different sites. Maddy surveyed people
at a party, Gabby surveyed people in Herty, and Kasey surveyed people around her
In the study design our group used a blind taste test with the matched pairs of two
brands of soda. The difference was computed by sprite minus sierra mist.
The confidence level is 95%. The point estimate is the difference in favor ratings between the sample mean of spite and sierra mist which is -0.16667. 2.045 is the degree of freedom at this confidence interval with a sample size of 30. The margin of error is 0.03408 and the confidence interval is (-0.20007, -0.13258). We are 95% confident that the population mean difference between the flavor rating of sprite and sierra mist will fall between -0.20007 and -0.13258.
The significance test was a success. At the 95% confidence level, there is sufficient statistical evidence to show that the mean difference in taste preference between Sprite and Sierra Mist is other than zero. The assumed value, 0, for the difference in preference for Sprite and Sierra Mist does not fall within the confidence interval of -.13258 and -.20007. Our hypothesis was incorrect; Georgia College students do not prefer Sprite to Sierra Mist. Instead, results show that Georgia College students prefer Sierra Mist .13 to .2 more than Sprite on a scale of 1 to 10. Though there is a difference between the preference for the two sodas, it is a small difference. Our sample is not large enough to be fully representative of the Ga College student population. Through completing this project, we learned how to apply significance tests in real life.