Description of Project:

The goal of this project is to analyze the GOP through sentiment data based on the first debate. We have polling data to see if the debate made a difference in polling numbers as well as expenditure data to see if expenditure is a confounding variable. Maybe expenditure has more results on polling that debate results. The code in this project can be used to analyze any twitter data to see if one candidate was more successful (subjectively) than another.

Author's Information:

Bryan Alcorn Nick Zoghb Todd Vogel Proneet De

• University of Califronia students, class project for Statistics 133

Organization of Subdirectories and Files:

The skeleton.R will build the subdirectiories and gather all of the data. The skeleton.R script will populate the raw data with the .csv files. Then the clean data script stored in code will put cleaned data in the data dirtectory. Finally, the data_alalysis script will produce graphs given the data so that the debate can be visually analyzed.

Specific Instructions or Comments:

There is a lot of potential with this project. The datasets contained a lot more information than we actually used. This is just preliminary analysis. There are other ways to quantify success in a debate, in our project, we analyzed success as number of times mentioned. This is valid under our model because we assume all press is good press, whether good sentiment or bad.

License to Share code:

This code can be shared as long as its creators are cited.