rollcall: create an object of class rollcall

Description

Create a rollcall object, used for the analysis of legislative voting or, equivalently, item-response modeling of binary data produced by standardized tests, etc.

Usage

rollcall(data,
         yea=1, nay=0, missing=NA, notInLegis=9,
         legis.names=NULL, vote.names=NULL,
         legis.data=NULL, vote.data=NULL,
         desc=NULL, source=NULL)

Value

An object of class rollcall, a list with the following components:

votes: a matrix containing voting decisions, with rows corresponding to legislators (test subjects) and columns to roll call votes (test items). Legislators (test subjects) and items (or votes) have been labeled in the dimnames attribute of this matrix, using the legis.names and/or vote.names arguments, respectively.
codes: a list with named components yea, nay, notInLegis and missing, each component a numeric vector (possibly of length 1 and possibly NA), indicating how entries in the votes component of the rollcall object should be considered. This list simply gathers up the values in the yea, nay, notInLegis and missing arguments passed to the function.
n: numeric, number of legislators, equal to dim(votes)[1]
m: numeric, number of votes, equal to dim(votes)[2]
legis.data: user-supplied data on legislators/test-subjects.
vote.data: user-supplied data on rollcall votes/test-items.
desc: any user-supplied description. If no description was provided, defaults desc defaults to NULL.
source: any user-supplied source information (e.g., a url or a short-form reference). If no description is provided, source defaults to NULL.

Arguments

data: voting decisions (for roll calls), or test results (for IRT). Can be in one of two forms. First, data may be a matrix, with rows corresponding to legislators (subjects) and columns to roll calls (test items). data can also be a list with an element named votes containing the matrix described above.
yea: numeric, possibly a vector, code(s) for a Yea vote in the rollcall context, or a correct answer in the educational testing context. Default is 1.
nay: numeric, possibly a vector, code(s) for a Nay vote in the rollcall context, or an incorrect answer in the educational testing context. Default is 0.
missing: numeric or NA, possibly a vector, code(s) for missing data. Default is NA.
notInLegis: numeric or NA, possibly a vector, code(s) for the legislator not being in the legislature when a particular roll call was recorded (e.g., deceased, retired, yet to be elected).
legis.names: a vector of names of the legislators or individuals. If data is a list or data.frame and has a component named legis.names, then this will be used. Names will be generated if not supplied, or if there are fewer unique names supplied than legislators/subjects (rows of the roll call matrix).
vote.names: a vector of names or labels for the votes or items. If data is a list or data.frame and has a component named vote.names, then this will be used. Names will be generated if not supplied, or if there are fewer unique names supplied than votes/test-items (columns of the roll call matrix).
legis.data: a matrix or data.frame containing covariates specific to each legislator/test-taker; e.g., party affiliation, district-level covariates. If this object does not have the same number of rows as data, an error is returned.
vote.data: a matrix or data.frame containing covariates specific to each roll call vote or test item; e.g., a timestamp, the bill sponsor, descriptive text indicating the type of vote. If this object does not have the same number of row as the number of columns in data, an error is returned.
desc: character, a string providing an (optional) description of the data being used. If data is a list and contains an element named desc, then this will be used.
source: character, a string providing an (optional) description of where the roll call data originated (e.g., a URL or a short-form reference). Used in print and summary methods.

Details

See below for methods that operate on objects of class rollcall.

Examples

Run this code

## generate some fake roll call data
set.seed(314159265)
fakeData <- matrix(sample(x=c(0,1),size=5000,replace=TRUE),
                   50,100)
rc <- rollcall(fakeData)
is(rc,"rollcall")        ## TRUE
rc                       ## print the rollcall object on screen

data(sc9497)             ## Supreme Court example data
rc <- rollcall(data=sc9497$votes,
               legis.names=sc9497$legis.names,
               desc=sc9497$desc)
summary(rc,verbose=TRUE)                      

if (FALSE) {
## s107
## could use readKH for this
dat <- readLines("sen107kh.ord")
dat <- substring(dat,37)
mat <- matrix(NA,ncol=nchar(dat[1]),nrow=length(dat))
for(i in 1:103){
  mat[i,] <- as.numeric(unlist(strsplit(dat[i],
                                        split=character(0))))
}

s107 <- rollcall(mat,
                 yea=c(1,2,3),
                 nay=c(4,5,6),
                 missing=c(7,8,9),
                 notInLegis=0,
                 desc="107th U.S. Senate",
                 source="http://voteview.ucsd.edu")
summary(s107)
}