Python Data Analysis Project: US Presidential Election Contributions

This tutorial demonstrates a practical data analysis project using Python to explore US voter donation data for the 2020 presidential election.

Preparation: The required datasets (weball20.txt, ccl.txt, itcont_2020_20200722_20200820.txt) are downloaded from the FEC, and necessary Python packages (pandas, matplotlib, wordcloud) are installed.

<code># install packages
pip install wordcloud
pip install matplotlib
pip install pandas</code>

Data loading and merging: Candidates, committee linkage, and individual contribution files are read with pandas (specifying separators and column names) and merged on CAND_ID and CMTE_ID to create a combined table linking donors to candidates.

<code>import pandas as pd
candidates = pd.read_csv("weball20.txt", sep='|', names=[
    'CAND_ID','CAND_NAME','CAND_ICI','PTY_CD','CAND_PTY_AFFILIATION','TTL_RECEIPTS',
    'TRANS_FROM_AUTH','TTL_DISB','TRANS_TO_AUTH','COH_BOP','COH_COP','CAND_CONTRIB',
    'CAND_LOANS','OTHER_LOANS','CAND_LOAN_REPAY','OTHER_LOAN_REPAY','DEBTS_OWED_BY',
    'TTL_INDIV_CONTRIB','CAND_OFFICE_ST','CAND_OFFICE_DISTRICT','SPEC_ELECTION','PRIM_ELECTION',
    'RUN_ELECTION','GEN_ELECTION','GEN_ELECTION_PRECENT','OTHER_POL_CMTE_CONTRIB','POL_PTY_CONTRIB',
    'CVG_END_DT','INDIV_REFUNDS','CMTE_REFUNDS'])

ccl = pd.read_csv("ccl.txt", sep='|', names=[
    'CAND_ID','CAND_ELECTION_YR','FEC_ELECTION_YR','CMTE_ID','CMTE_TP','CMTE_DSGN','LINKAGE_ID'])

ccl = pd.merge(ccl, candidates)

itcont = pd.read_csv("itcont_2020_20200722_20200820.txt", sep='|', names=[
    'CMTE_ID','AMNDT_IND','RPT_TP','TRANSACTION_PGI','IMAGE_NUM','TRANSACTION_TP','ENTITY_TP',
    'NAME','CITY','STATE','ZIP_CODE','EMPLOYER','OCCUPATION','TRANSACTION_DT','TRANSACTION_AMT',
    'OTHER_ID','TRAN_ID','FILE_NUM','MEMO_CD','MEMO_TEXT','SUB_ID'])

c_itcont = pd.merge(ccl, itcont)
</code>

Data cleaning: Missing values in STATE, EMPLOYER, OCCUPATION are filled with "NOT PROVIDED", and the transaction date column is converted to a readable string format.

<code>c_itcont['STATE'].fillna('NOT PROVIDED', inplace=True)
c_itcont['EMPLOYER'].fillna('NOT PROVIDED', inplace=True)
c_itcont['OCCUPATION'].fillna('NOT PROVIDED', inplace=True)

c_itcont['TRANSACTION_DT'] = c_itcont['TRANSACTION_DT'].astype(str)
# Convert from YYYYMMDD integer to YYYY-MM-DD string (example conversion)
c_itcont['TRANSACTION_DT'] = [i[3:7] + i[0] + i[1:3] for i in c_itcont['TRANSACTION_DT']]
</code>

Exploratory analysis: The dataset contains 756,205 rows and 8 columns; basic statistics are displayed with .shape, .info(), and .describe().

Key analyses include:

Total donation amount by party (DEM, REP, etc.)

Total donation amount by candidate (BIDEN, TRUMP, etc.)

Top occupations by donation amount and by donor count

Top states by donation amount and by donor count

Visualization: Bar charts for state donation totals and donor counts, a pie chart for Biden’s state‑wise donation share, and a word cloud of Biden’s donors are generated using matplotlib and wordcloud.

<code># Example: bar chart of top states by donation amount
st_amt = c_itcont.groupby('STATE').sum().sort_values('TRANSACTION_AMT', ascending=False)[:10]
st_amt.plot(kind='bar')

# Example: pie chart of Biden’s donations by state
biden = c_itcont[c_itcont['CAND_NAME'] == 'BIDEN, JOSEPH R JR']
biden_state = biden.groupby('STATE').sum().sort_values('TRANSACTION_AMT', ascending=False)[:10]
biden_state.plot.pie(figsize=(10,10), autopct='%0.2f%%')
</code>

All code, data sources, and results are available at the competition page https://tianchi.aliyun.com/competition/entrance/531837/introduction.