performance: Performance Analysis Calculation

Description

Performance Analysis calculates election outcomes of past contests given hypothetical voting district(s). This analysis has been used to determine if a Gingles III violation occurs due to how a district map is drawn. It can also be used to demonstrate that a more equitable alternative map exists. This function assumes RPV so it should only be used with contests where RPV has been established.

Usage

performance(
  data = NULL,
  cands = "",
  candidate = "",
  preferred_candidate = "",
  total = "",
  contest = "",
  year = "",
  election_type = "",
  map = "",
  jurisdiction = "",
  includeTotal = FALSE
)

Value

data.frame of Performance Analysis results by candidate

Arguments

data: A data.frame object containing precinct-level election results for contests of interest. It must include candidate vote counts and contest total votes fields and must be subsetted to the relevant precincts. This data.frame will likely be the output of a "Split Precinct Analysis".
cands: A character vector of the candidate vote counts field names from data that are relevant to the given year and contest being analyzed.
candidate: A character vector of candidate names. The names must be listed in the same order cands. The values will appear in the output data.frame exactly as they are written in this argument.
preferred_candidate: A character vector of preferred racial groups associated with the candidates. The values must be listed in the correct order with respect to the cands/candidate arguments. The values will appear in the output data.frame exactly as they are written in this argument.
total: A character vector of the the contest total vote count field names from data.
contest: The name of the contest being analyzed
year: The year of the contest being analyzed
election_type: The election type the contest being analyzed (e.g "General" or "Primary")
map: String containing the name of the district map being analyzed (e.g "remedial" or "adopted"). This is an optional field that defaults to blank.
jurisdiction: String containing the name of the jurisdiction being analyzed (i.e a district number or "County"). Be sure that data is subsetted only to this jurisdiction.
includeTotal: Boolean indicating if a total number of votes row should be appended to the output data.frame

Author

Rachel Carroll <rachelcarroll4@gmail.com>

Loren Collingwood <lcollingwood@unm.edu>

Examples

Run this code


library(eiExpand)
data(south_carolina)

# Get sample election data
D5_election <- south_carolina %>%
   dplyr::filter(District == 5)

# Run performance Analysis on 2018 Governor contest
perf_results <- performance(
  data = D5_election,
  cands = c("R_mcmaster", "D_smith"), 
  candidate = c("McMaster (R)", "Smith (D)"), # formatted candidate names
  preferred_candidate = c("White", "Black"), # race preference of candidates respectively
  total = "total_gov",
  contest = "Governor",
  year = 2018,	
  election_type = "General",	
  jurisdiction = "District 5"
)

Run the code above in your browser using DataLab