AP Statistics Project

May 2008

Kevin Morris

Callie Minor

Kelsey Floyd

Disclaimer

This study was done in an AP Statistics course with relatively small sample sizes. The validity of such studies must always be questioned. Please keep this in mind if you use or report the results of this study.

Abstract

Purpose: The purpose of this study is to determine if the three front-runners of the 2008 Presidential election (McCain, Clinton, and Obama) are all equally likely to actually become president in November. Many feel this Presidential election is evenly split between Obama, McCain, and Clinton and through this study, it will hopefully be apparent that this assumption is not valid. A Chi-squared Goodness of Fit test will be used to try and show that at least one candidate would not receive 33% of votes if all three candidates were able to go against one another.

Research question: In the 2008 US Presidential election, if the only candidates on the ballad were Obama, McCain, and Clinton, who would you elect to be President of the United States?

Hypothesis: At least one candidate would not receive 33% of the votes.

Population: A simple random sample of 71 Junior/Senior students at Tates Creek High School.

Test: Chi-Squared Goodness of Fit test, degrees of freedom = n-1 = 3-1 = 2, Significance level-alpha = 0.05

Methodology

Introduction

Some researchers have claimed that the American population is relatively evenly split between Obama, Clinton, and McCain for voting in the upcoming Presidential election. To try and prove this is not a valid claim, our AP Statistics group decided to perform a survey of students in Tates Creek High School. This survey asked students to choose who they would like to see in office if Obama, Clinton, and McCain were the only choices. Since Seniors and some Juniors are actually going to be able to vote in November’s Presidential election, class lists were obtained with each student’s name. Each student was labeled a different number. A simple random sample of 71 students was chosen by a random number generator on the calculator. Each student chosen was given a blank piece of white paper and asked to anonymously write the name of the candidate they would like to see in office of the three candidates given. Then, the student was to fold up their piece of paper and put it into a bag which contained the rest of the votes. By conducting the survey in such a way, we were able to reduce bias and each student was able to feel they could give their true opinion on who they thought should be the next President.

Note* - 79 students were randomly selected to participate in this survey; however, eight people were not present the day of sampling.

Analysis

Assumptions:

· All expected counts are at least one ü

· No more than 20% of expected counts are less than 5 ü

Degrees of freedom = n-1 = 3-1 = 2 significance level- alpha = .05

Null Hypothesis: 1/3 votes would go to Obama; 1/3 votes would go to Clinton; 1/3 votes would go to McCain

Alternative Hypothesis: At least one candidate would not receive 1/3 of the votes

X² test statistic =

X² test statistic 18.9577

Conclusion: Because our data yielded a X² value of 18.9577 > X^2* = 5.99146 (df=2), our data is statistically significant at the α= .05 level. We may therefore reject the null hypothesis (all candidates receive an equal proportion of votes), in favor of the alternative hypothesis (that at least one proportion was not equivalent to its expected proportion).

Keep in mind that a few of those sampled were under voting age and could possibly be unaware of the platform each candidate stands for. As result, some students may have voted what their parents opinions are instead of their own. Also, remember eight students of our population were unable to be surveyed and could have potentially skewed our results.