What it Takes to Make History Torbjorn Bjering Ho-Jung Hsiao Eric Griffin Chun-Hung Lin Gulsah Gunenc Gaoyuan Tian Laura Braeutigam Torbjorn Bjering Ho-Jung Hsiao Eric Griffin Chun-Hung Lin Gulsah Gunenc Gaoyuan Tian Laura Braeutigam
Table of Contents Introduction Brief Summary Descriptive Statistics Exploratory Data Analysis Statistical Analysis Conclusions Introduction Brief Summary Descriptive Statistics Exploratory Data Analysis Statistical Analysis Conclusions
Introduction A lot goes into winning an election. There are many factors that can lead a candidate to victory. A campaign is based on what is important to the population. A candidate’s image will help him appeal to many different aspects of society. A lot goes into winning an election. There are many factors that can lead a candidate to victory. A campaign is based on what is important to the population. A candidate’s image will help him appeal to many different aspects of society.
What are we studying? The 2008 Presidential election Obama vs. McCain How do different aspects of society influence the result of the election? Why do they affect the outcome? The 2008 Presidential election Obama vs. McCain How do different aspects of society influence the result of the election? Why do they affect the outcome?
Why are we studying the election? Monumental event in our nation’s history Something that affects all Americans Abundant accurate data Recently occurred Monumental event in our nation’s history Something that affects all Americans Abundant accurate data Recently occurred
How are we studying the election? Extracting data from U.S. Census Bureau 50 states plus District of Columbia DC is an outlier! Evaluating exploratory data LSM Regression Gender Age Financial Status Education Religion Race Extracting data from U.S. Census Bureau 50 states plus District of Columbia DC is an outlier! Evaluating exploratory data LSM Regression Gender Age Financial Status Education Religion Race
Brief Summary Obama won the election. Who support Obama? Women Younger voters Voters with higher income Highly educated voters Jewish voters Minority groups Obama won the election. Who support Obama? Women Younger voters Voters with higher income Highly educated voters Jewish voters Minority groups
Exploratory Data Analysis An approach to analyze data for the purpose of formulating hypotheses worth testing, complementing the tools of conventional statistics for testing hypotheses. Analyzing scatter diagrams to see if we can use linear regression An approach to analyze data for the purpose of formulating hypotheses worth testing, complementing the tools of conventional statistics for testing hypotheses. Analyzing scatter diagrams to see if we can use linear regression
Descriptive Statistics Dependent Variable Obama election percentage Independent Variable Minority Population Black Native American Hispanic Asian Dependent Variable Obama election percentage Independent Variable Minority Population Black Native American Hispanic Asian
Independent Variable Gender Age Religion Christian Jewish Education Persons with Bachelor's Degree or More Financial Status Unemployment Rate Personal Income Per Capita in Current Dollars Energy Consumption Per Capita Homeownership Rate Independent Variable Gender Age Religion Christian Jewish Education Persons with Bachelor's Degree or More Financial Status Unemployment Rate Personal Income Per Capita in Current Dollars Energy Consumption Per Capita Homeownership Rate Descriptive Statistics cont.
Statistical Analysis Analyzing collected data for the purposes of summarizing information to make it more usable and/or making generalizations about a population.
Obama vs. Minority Effect of minority groups are positive. Asian is more supportive to Obama than other groups. All coefficients except Hispanic and Native are significant. Minority = 100-white
Obama vs. Minority Blacks DC Minority (%)Blacks (%) Minority DC
Obama vs. Gender Women are more supportive to Obama than men.
Obama vs. Gender DC (Male %) (Female %) MaleFemale
Obama vs. Age All groups of ages have positive effect on Obama vote. The positive effect is decreasing with increase of age.
Obama vs. Religion Christian voters tend not to vote Obama, but not significant. Jewish voters are supportive to Obama.
Highly educated voters tend to vote Obama. Obama vs. Education Bachelor (%)
Obama vs. Unemployment Jobless voters seem to be supportive to Obama, but the coefficient is not significant. Unemployment (%)
Obama vs. Income Income $ Voters with higher income are supportive to Obama.
Obama vs. Financial Status Voters with higher income and less energy consumption are supportive to Obama. Coefficients on Homeowner and Unemployment are not significant.
Conclusions Surprising results Voters with higher income are supportive to Obama. Coefficient for African American voters is lower than expected High energy users and home owners were not supportive Surprising results Voters with higher income are supportive to Obama. Coefficient for African American voters is lower than expected High energy users and home owners were not supportive
More Conclusions Expected results Women Younger people Educated people Minorities Expected results Women Younger people Educated people Minorities