Exam I

Introduction

Purpose:

The following document provides an analysis of the shifting political climate in the Cincinnati region from 2015 to present day.

Background

The Federal Election Commission is an independent regulatory agency whose purpose is to enforce campaign finance law in United States federal elections. The agency provides transparency, fair enforcement, and administration of federal campaign finance laws by making data related to federal elections publicly available.

The Analysis

Packages for Analysis

To process the data in R, first install the Tidyverse package and the following libraries to use manipulate the data:

Package	Explanation
tidyverse	For all things tidy
DT	To display data using Data tables
lubridate	To standardize date formats in the source data

Data Preparation

Import the data

The data is first imported from http://asayanalytics.com/cinci_politics.

Cleaning the data

Before the data can be analyzed, we inspect the data for anomalies, errors, and missing information. We notice missing information in two columns:

Contributor Employer
Contributor Occupation

** One note on the missing fields under Last Name and Street: these were not addressed because they do not have an impact on the overall analysis of the data. Therefore, these fields were left as is.

##       contributor_last_name      contributor_first_name 
##                           6                           0 
##        contributor_street_1        contributor_employer 
##                          17                        1430 
##      contributor_occupation   contribution_receipt_date 
##                        1342                           0 
## contribution_receipt_amount   contributor_aggregate_ytd 
##                           0                           0 
##              committee_name              committee_type 
##                           0                           0 
## committee_party_affiliation 
##                           0

We also note inconsistency in the date format in the Contribution Receipt Date column:

##  [1] "14/10/2016" "27/09/2016" "03/11/2016" "13/10/2016" "26/02/2016"
##  [6] "28/12/2016" "14/09/2016" "07/03/2016" "30/04/2016" "27/10/2016"
## [11] "31/03/2016" "17/10/2016" "29/09/2016" "06/07/2015" "05/10/2016"
## [16] "16/09/2016" "22/09/2016" "30/09/2016" "12/10/2016" "29/01/2016"
## [21] "15/09/2016" "11/07/2016" "28/09/2016" "30/09/2016" "12/10/2016"
## [26] "03/10/2016" "10/09/2016" "20/10/2016" "15/09/2016" "28/09/2016"
## [31] "07/11/2016" "26/10/2016" "25/10/2016" "24/09/2016" "07/11/2016"
## [36] "14/09/2016" "13/09/2016" "06/11/2016" "24/10/2016" "06/09/2016"
## [41] "03/11/2016" "11/10/2016" "04/10/2016" "19/09/2016" "30/10/2016"
## [46] "23/09/2016" "27/10/2016" "21/10/2016" "29/09/2016" "20/09/2016"

To standardize the date format into Day-Month-Year for all observations, we use the lubridate package.

We are only interested in observations from contributors that either made a donation or received a refund. Therefore, we exclude any observations where there are $0 in donations or refunds from the Contribution Receipt Amount field.

Since our analysis covers the period of 2015 to present day, we want to also exclude observations that do not fall within that time period.

Consistency of Data

R recognizes blank fields as (N/A), but we also note values that were manually entered into the data as meaning ‘No information provided’. These include variations of N/A (i.e. N.A., NA, NA-Retired, etc.) and they should be standardized into common form. The chosen standardization for these values is ‘NA’.

Another inconsistency exists in the way individuals identified themselves as RETIRED. These include: (RET., RETIREE, RET) and for consistency, all observations are standardized as RETIRED.

Next, we standardize the (N/A) values recognized in R by imputing those fields with ‘NA’. This is consistent with the previous blank observations found in the Contribution Employer field. Similarly, the blank Occupation fields are also imputed with ‘NA’ for standardization.

We now see that Employer and Occupation have been updated and we no longer have missing values in these columns.

##       contributor_last_name      contributor_first_name 
##                           6                           0 
##        contributor_street_1        contributor_employer 
##                          17                           0 
##      contributor_occupation   contribution_receipt_date 
##                           0                           0 
## contribution_receipt_amount   contributor_aggregate_ytd 
##                           0                           0 
##              committee_name              committee_type 
##                           0                           0 
## committee_party_affiliation 
##                           0

Dummy Variables

The column Committe Type contains several variables to indicate the Committee to which the donation/refund is intended to support. The options include:

Presidential
Senate
House
Political Party
Political Action Committee (PAC)

Per the FEC, a committee or party must meet specific criteria to be considered a Qualified committee or party. The Qualified designation confers several benfits to the group, such as access for nominated and selected candidates to appear on State and National ballots as members of their respective party. The following table is a sample of the Qualified designations based on the Committee Type.

3.3

There are 57,375 observations in the original data file. After cleaning up the data, removing $0 contributions and records prior to 2015, there are 57,351 observations. Column Name Explanation ———– ———– contributor_last_name Last Name of contributor contributor_first_name First Name of contributor contributor_street_1 The street address of the contributor contributor_employer The individual’s reported employer contributor_occupation The individual’s occupation contribution_receipt_date The date the contribution was submitted contribution_receipt_amount The $ amount of the contribution contributor_aggregate_ytd The total amount of contributions by a particular individual committee_name The name of the committee committee_type The type of committte (House, Senate, Presidential, PAC, Party) committee_party_affiliation The political party name

3.4

CP3 %>% 
  select(contributor_last_name:Qualified_Committee) %>% 
  head(50)

## # A tibble: 50 x 12
##    contributor_las… contributor_fir… contributor_str… contributor_emp…
##    <chr>            <chr>            <chr>            <chr>           
##  1 LIRA             MONICA           3230 LONGMEADOW… NA              
##  2 WILMES           JOHN             4215 DELANEY ST  SELF-EMPLOYED   
##  3 JAPIKSE          CORNELIS         2507 EVERGREEN … RETIRED         
##  4 HILL             THOMAS           2811 QUEENSWOOD… ACOSTA SALES AN…
##  5 DAVIS            MICHAEL          7321 CLOUGH PIKE SELF            
##  6 KUES             EILEEN           573 EVANSWOOD PL NONE            
##  7 LIRA             MONICA           3230 LONGMEADOW… LIBRARY OF CONG…
##  8 ROSENTHAL BERLI… JENNIE           1846 KEYS CRESC… SELF-EMPLOYED   
##  9 ROSENTHAL BERLI… JENNIE           1846 KEYS CRESC… SELF-EMPLOYED   
## 10 PEPPER           FRANCES          233 OLIVER RD.   NOT EMPLOYED    
## # … with 40 more rows, and 8 more variables: contributor_occupation <chr>,
## #   contribution_receipt_date <date>, contribution_receipt_amount <dbl>,
## #   contributor_aggregate_ytd <dbl>, committee_name <chr>,
## #   committee_type <chr>, committee_party_affiliation <chr>,
## #   Qualified_Committee <chr>

3.5

SUMMARY STATISTICS

Simple Analysis and Trends

Visualization 1

Donations to Political Parties vary over time. Below, we see more contributions to the Republican party leading up to the 2016 election. By contrast, the Democratic party saw an increase leading up to the midterm elections in 2018. This makes sense given that a Republican won the federal election in 2016, and Democrats, a minority representation in the House in 2016, won a majority in the House in the 2018 elections.

Visualization 2

The amount of money raised by each party per election cycle

Visualizaiton 3

The following graphic shows the total donation amount given to each type of Committee. A qualified party (either Republican or Demorcratic) received the overwhelming amount of donations. The graphic also shows that the next two benefitting committees were the House and Senate as they have the next highest number of contribution amounts.

Directed Analysis

5.1

WHICH CINCINNATI RESIDENTS SPEND THE MOST MONEY ON POLITICS?

The following is a list of the residents who spent the most money on politics. The top 2 contributors donated over 500K dollars each. (To limit the number of observations shown, only contributions greater than $500 are included in the table.)

WHERE DOES IT GO?

The Republican party received the highest dollar amount of contributions. They received over 2.4 million more dollars than the next recipient, the Democratic party. The Green Party received the fewest donation amount with $3500 from Cinicinnati residents.

5.2

WHO MADE THE MOST INDIVIDUAL CONTRIBUTIONS AND WHY?

Listed below are the residents who made the most number of contributions. The first place individual donated 722 more times than the individual with the second most contributions. One theory might be that Ruchhoft, the highest contributing member, wanted to make sure all members of her party received a donation. The donations were likely smaller donations so that all committees were represented by their donations.

5.3

WHICH EMPLOYERS ARE THE MOST HEAVILY REPRESENTED IN THE DATA?

Individuals who were self-employed, retired, or unemployed, were among the most active contributing members to political campaigns. They occupied 9 of the top 10 slots for number of contributions. Among the listed employers with more than 1 employee, the University of Cincinnati, Cincinnati Children’s Hospital, and Proctor & Gamble were the top 3 employers heavily represented in the data.

5.4

HOW MANY INDIVIDUALS WHO WORK FOR XAVIER HAVE MADE AT LEAST ONE CONTRIBUTION AND WHICH EMPLOYEES HAVE CONTRIBUTED THE MOST?

There were 23 Xavier University employees who contributed to politics during our time period. Below is a complete list of the employees and how much each individual contributed to the political system.

#### HOW MUCH MONEY IN TOTAL HAVE XAVIER UNIVERSITY EMPLOYEES CONTRIBUTED?