This article from FiveThirtyEight highlights how people of color made up only 28% of the general election candidates in the 2022 election, while as a group they make up 41% of the general population. Additionally, the article highlights how this disparity among candidate representation continues to be more stark within the Republican party, where only 19% of canidates were people of color, compared to 46% of candidates running within the Democratic Party.
Link to Article: https://fivethirtyeight.com/features/2022-candidates-race-data/
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──## ✓ ggplot2 3.3.5 ✓ purrr 0.3.4
## ✓ tibble 3.1.5 ✓ dplyr 1.0.7
## ✓ tidyr 1.1.4 ✓ stringr 1.4.0
## ✓ readr 2.0.2 ✓ forcats 0.5.1## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(janitor)##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testdem_candidates_raw = read.csv('https://raw.githubusercontent.com/fivethirtyeight/data/master/primary-project-2022/dem_candidates.csv')
rep_candidates_raw = read.csv('https://raw.githubusercontent.com/fivethirtyeight/data/master/primary-project-2022/rep_candidates.csv')
dem_candidates = as_tibble(dem_candidates_raw)
rep_candidates = as_tibble(rep_candidates_raw)dem_candidates = dem_candidates %>% mutate(Party = 'Democrat') #Add party affiliation to the dataframe
colnames(dem_candidates)## [1] "Candidate" "Gender" "Race.1"
## [4] "Race.2" "Race.3" "Incumbent"
## [7] "Incumbent.Challenger" "State" "Primary.Date"
## [10] "Office" "District" "Primary.Votes"
## [13] "Primary.." "Primary.Outcome" "Runoff.Votes"
## [16] "Runoff.." "Runoff.Outcome" "EMILY.s.List"
## [19] "Justice.Dems" "Indivisible" "PCCC"
## [22] "Our.Revolution" "Sunrise" "Sanders"
## [25] "AOC" "Party.Committee" "Party"dem_candidates = dem_candidates %>% select(c('Candidate','Party','Gender','Race.1', 'Race.2', 'Race.3','Incumbent', 'Incumbent.Challenger','Office', 'District','State','Primary.Outcome'))rep_candidates = rep_candidates %>% mutate(Party = 'Republican') #Add party affiliation to the dataframe
colnames(rep_candidates)## [1] "Candidate" "Gender" "Race.1"
## [4] "Race.2" "Race.3" "Incumbent"
## [7] "Incumbent.Challenger" "State" "Primary.Date"
## [10] "Office" "District" "Primary.Votes"
## [13] "Primary.." "Primary.Outcome" "Runoff.Votes"
## [16] "Runoff.." "Runoff.Outcome" "Trump"
## [19] "Trump.Date" "Club.for.Growth" "Party.Committee"
## [22] "Renew.America" "E.PAC" "VIEW.PAC"
## [25] "Maggie.s.List" "Winning.for.Women" "Party"rep_candidates = rep_candidates %>% select(c('Candidate','Party','Gender','Race.1', 'Race.2', 'Race.3','Incumbent', 'Incumbent.Challenger','Office', 'District','State','Primary.Outcome'))all_candidates = rbind(dem_candidates, rep_candidates)#Ensuring that there are no invalid repsonses in any of the fields
unique(all_candidates$Gender)## [1] "Male" "Female" "Nonbinary"unique(all_candidates$Race.1)## [1] "White" "Black"
## [3] "Asian (Indian)" "Latino"
## [5] "Latino (Mexican)" "Latino (Puerto Rican)"
## [7] "Unknown" "Black (Eritrean)"
## [9] "Asian (Bangladeshi)" "White "
## [11] "Latino (Peruvian)" "Middle Eastern (Saudi)"
## [13] "Latino (Ecuadorian)" "Black (Nigerian)"
## [15] "Native American (Saponi)" "White (Azerbaijani)"
## [17] "Middle Eastern (Palestinian)" "Latino (Venezuelan)"
## [19] "Native American (Comanche)" "Asian"
## [21] "Native American" "Asian (Korean)"
## [23] "White (Polish)" "Latino (Cuban)"
## [25] "Latino (Colombian)" "Asian (Japanese)"
## [27] "Latino (Salvadoran)" "Asian (Pakistani)"
## [29] "Middle Eastern (Armenian)" "Asian (Chinese)"
## [31] "Middle Eastern (Iranian)" "Latino (Guatemalan)"
## [33] "Asian (Taiwanese)" "Latino (Puerto Rican / Ecuadorian)"
## [35] "Asian (Filipino)" "Native American (Lakota)"
## [37] "Asian (Thai)" "Native American (Cherokee)"
## [39] "Native American (Creek / Yuchi)" "White (Albanian)"
## [41] "Latino (Dominican)" "White (Belgian)"
## [43] "Native American (Ho-Chunk)" "Latino (Chilean)"
## [45] "Black (Ethiopian)" "Black (Somali)"
## [47] "Black (Jamaican)" "Black (Liberian)"
## [49] "Asian (Laotian)" "Pacific Islander (Native Hawaiian)"
## [51] "Native American (Yupik)" "Middle Eastern (Iraqi)"
## [53] "Native American (Arapaho)" "Black (Haitian)"
## [55] "Asian " "Middle Eastern"
## [57] "Latino " "White (Kosovar)"
## [59] "White (Russian)" "Black (Sudanese)"
## [61] "Black (Gambian)" "Asian (Vietnamese)"
## [63] "Middle Eastern (Kuwaiti)" "Middle Eastern (Turkish)"
## [65] "Latino (Brazilian)" "Latino (Uruguayan)"
## [67] "Asian (Malaysian)" "Middle Eastern (Egyptian)"
## [69] "Latino (Nicaraguan)" "Middle Eastern (Afghani)"
## [71] "Latino (Cuban / Mexican)" "Middle Eastern (Lebanese)"
## [73] "Middle Eastern (Israeli)" "Latino (Salvadoran / Mexican)"
## [75] "White (Serbian)" "White (Italian)"
## [77] "Middle Eastern " "Asian (Vietnamese / Hmong)"
## [79] "Native American (Inupiaq)" "Latino (Honduran)"
## [81] "Middle Eastern (Syrian)"unique(all_candidates$Race.2)## [1] "Asian (Indian)" ""
## [3] "Black" "Latino (Mexican)"
## [5] "Middle Eastern (Iraqi)" "White"
## [7] "Asian (Pakistani)" "Latino"
## [9] "Native American (Pomo)" "Pacific Islander (Native Hawaiian)"
## [11] "Native American (Yupik)" "Latino (Cuban)"
## [13] "Unknown" "Asian (Japanese)"
## [15] "Asian" "Latino (Brazilian)"
## [17] "Native American (Cherokee)" "Asian (Filipino)"
## [19] "Native American (Chickasaw)" "Middle Eastern (Iranian)"
## [21] "Latino (Puerto Rican)"unique(all_candidates$Race.3)## [1] "" "Asian (Filipino)" "Asian"unique(all_candidates$Incumbent)## [1] "No" "Yes"unique(all_candidates$Incumbent.Challenger)## [1] "No" "Yes"unique(all_candidates$State)## [1] "Texas" "Indiana" "Ohio" "West Virginia"
## [5] "Nebraska" "Kentucky" "Oregon" "Idaho"
## [9] "North Carolina" "Pennsylvania" "Virginia" "Alabama"
## [13] "Arkansas" "Georgia" "New Mexico" "Mississippi"
## [17] "Montana" "Iowa" "South Dakota" "New Jersey"
## [21] "California" "Maine" "Nevada" "South Carolina"
## [25] "North Dakota" "Colorado" "Illinois" "New York"
## [29] "Oklahoma" "Utah" "Maryland" "Arizona"
## [33] "Kansas" "Michigan" "Missouri" "Washington"
## [37] "Tennessee" "Connecticut" "Minnesota" "Vermont"
## [41] "Wisconsin" "Hawaii" "Alaska" "Wyoming"
## [45] "Florida" "Massachusetts" "Delaware" "New Hampshire"
## [49] "Rhode Island"unique(all_candidates$Office)## [1] "Representative" "Governor"
## [3] "Senator" "Senator (unexpired term)"unique(all_candidates$District)## [1] "1" "2" "3" "4" "5" "7"
## [7] "8" "9" "10" "12" "13" "14"
## [13] "15" "16" "17" "18" "20" "21"
## [19] "22" "23" "24" "27" "28" "29"
## [25] "30" "32" "33" "34" "35" "36"
## [31] "37" "38" "N/A" "6" "11" "19"
## [37] "25" "26" "31" "39" "40" "41"
## [43] "42" "43" "44" "45" "46" "47"
## [49] "48" "49" "50" "51" "52" "At-Large"unique(all_candidates$Primary.Outcome)## [1] "Lost" "Made runoff" "Won"Clean up the candidates dataframe and expand the features, by completing the following steps:
#Split Race.1 column to only include Race and split the Ethnicity into a seperate column
all_candidates = all_candidates %>% separate(Race.1, into=c("Race","Ethnicity"), sep="\\(")## Warning: Expected 2 pieces. Missing pieces filled with `NA` in 2418 rows [1, 2,
## 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, ...].all_candidates = all_candidates %>% mutate(Race = str_trim(Race))
unique(all_candidates$Race)## [1] "White" "Black" "Asian" "Latino"
## [5] "Unknown" "Middle Eastern" "Native American" "Pacific Islander"#Remove the trailing paren from the Ethnicity string
all_candidates = all_candidates %>% mutate(Ethnicity = str_remove(Ethnicity,"\\)"))
unique(all_candidates$Ethnicity)## [1] NA "Indian"
## [3] "Mexican" "Puerto Rican"
## [5] "Eritrean" "Bangladeshi"
## [7] "Peruvian" "Saudi"
## [9] "Ecuadorian" "Nigerian"
## [11] "Saponi" "Azerbaijani"
## [13] "Palestinian" "Venezuelan"
## [15] "Comanche" "Korean"
## [17] "Polish" "Cuban"
## [19] "Colombian" "Japanese"
## [21] "Salvadoran" "Pakistani"
## [23] "Armenian" "Chinese"
## [25] "Iranian" "Guatemalan"
## [27] "Taiwanese" "Puerto Rican / Ecuadorian"
## [29] "Filipino" "Lakota"
## [31] "Thai" "Cherokee"
## [33] "Creek / Yuchi" "Albanian"
## [35] "Dominican" "Belgian"
## [37] "Ho-Chunk" "Chilean"
## [39] "Ethiopian" "Somali"
## [41] "Jamaican" "Liberian"
## [43] "Laotian" "Native Hawaiian"
## [45] "Yupik" "Iraqi"
## [47] "Arapaho" "Haitian"
## [49] "Kosovar" "Russian"
## [51] "Sudanese" "Gambian"
## [53] "Vietnamese" "Kuwaiti"
## [55] "Turkish" "Brazilian"
## [57] "Uruguayan" "Malaysian"
## [59] "Egyptian" "Nicaraguan"
## [61] "Afghani" "Cuban / Mexican"
## [63] "Lebanese" "Israeli"
## [65] "Salvadoran / Mexican" "Serbian"
## [67] "Italian" "Vietnamese / Hmong"
## [69] "Inupiaq" "Honduran"
## [71] "Syrian"#Create Simplified Race category
all_candidates = all_candidates %>% mutate(Race.Simplified = ifelse(Race == 'White','White','Non-White'))
all_candidates = all_candidates %>% clean_names()
colnames(all_candidates)## [1] "candidate" "party" "gender"
## [4] "race" "ethnicity" "race_2"
## [7] "race_3" "incumbent" "incumbent_challenger"
## [10] "office" "district" "state"
## [13] "primary_outcome" "race_simplified"Based on this initial cleaning of the data discovery, there’s a lot of additional ways to cut the data to better understand the demographic breakdown of the candidates and to identify other potential correlations - both in tabular form as well as through data visualizations. Some examples of this additional exploratory analysis includes:
An understanding of this data can be very useful to influence outreach efforts for drumming up political engagement by more people from underrpresented communities, in order to expand the bench of potential candidates in the future.