Statistic Project

I. Introduction

A. RQ 1: Do people with different flavor preferences have different perceptions about the prevalence of Starburst flavor distributions in the two-packs?

1. The two variables are people with different flavor preferences and the prevalence of Starburst flavor distributions in the two-packs. In this case, the explanatory variable is people with different flavor preferences and the prevalence of Starburst flavor distributions in the two-packs is the response variable. And both people with different flavor preferences and the prevalence of Starburst flavor distributions in the two-packs are categorical variables.

2. The scope of inquiry for this problem is all the people who took part in the survey. However, since our sample only consisted of students in STAT 4210, we do not have a very large and diverse sample.

3. Effect of interest is university students since all students in STAT 4210 are university students.

B. RQ 2: Is the proportion of two-packs that contain only yellow Starburst candies higher than would be expected than by random chance?

1. The two variables are proportion of two-packs that contain only yellow Starburst candies and proportion of two-packs that contain not only yellow Starburst. Both of them are categorical variables.

2. The scope of inquiry for this problem is that two-pack of Starburst candies sent to students in STAT 4210 class.

3. Effect of interest is the proportion of two-packs of Starburst candies.

C. RQ 3: Is the claim on the internet, that the four flavors are uniformly distributed, supportable?

1. The four variables are flavor orange, flavor pink, flavor yellow and flavor red. All of them are categorical variables.

2. All two-packs of Starburst candies in the internet.

3. Effect interest is all two-packs of Starburst candies.

II. Method

A. RQ 1:

1. H0: Flavor preference is independent of perception about flavor prevalence. Ha: Flavor preference is not independent of perception about flavor prevalence.

2. Chi square tests of independence will be used to evaluate the first research question.

3. We can follow-up tests of independence or goodness of fit by looking at the chi square contribution of each cell or category, to see which group has larger contributions. Similarly, we can look at the deviation (observed-expected) for each cell or category to see which group is farther away than is expected under the null hypothesis.

4.