clean_data: Clean Bawsalah data for analysis
In saudiwin/bawsala: Analysis of Tunisian Voting Data

Description Usage Arguments

Clean Bawsalah data for analysis

clean_data(keep_legis = 1, use_subset = FALSE,
  subset_party = c("Bloc Al Horra", "Mouvement Nidaa Tounes"),
  use_both = FALSE, refleg = "ARP_Bochra Belhaj Hamida", legis = 1,
  use_vb = FALSE, use_nas = FALSE, to_run = 3, sample_it = FALSE,
  sample_amt = 50)

`keep_legis`	Number of legislator votes required to keep them in the dataset
`use_subset`	TRUE/FALSE Use only a subset of the dataset?
`use_both`	TRUE/FALSE Should both the ANC and ARP datasets be combined?
`refleg`	Character string name of reference legislator
`legis`	Which legislator to use if use_both==FALSE. Options are 1 for ARP and 2 for ANC
`use_vb`	TRUE/FALSE Whether to use variational bayesian inference from the Rstan package. It is quicker but less accurate.
`use_nas`	TRUE/FALSE For ordinal data, whether absences should be coded as a separate category (TRUE) or coded as NA (FALSE)
`to_run`	Which of the datasets to use in the analysis. Put 1 for binary yes/no, 2 for binary yes/no v. abstain, 3 for ordinal
`sample_it`	Whether to use a sample of the dataset for analysis. Useful for testing models.
`sample_amt`	Integer number of legislators to sample from dataset
`if`	use_subset==TRUE, a vector of party bloc names with which to subset the data