blackbox
is a function that takes a matrix of survey data in which individuals
place themselves on continuous scales across multiple issues, and locates those
citizens in a spatial model of voting. Mathematically, this function generalizes
the singular value of a matrix to cases in which there is missing data in the
matrix. Scales generated using perceptual data (i.e. scales of legislator locations
using liberal-conservative rankings by survey respondents) should instead use
the blackbox_transpose
function in this package instead.
blackbox(data,missing=NULL,verbose=FALSE,dims=1,minscale)
matrix of numeric values containing the issue scale data. Respondents should be organized on rows, and stimuli on columns. It is helpful, though not necessary, to include row names and column names.
vector or matrix of numeric values, sets the missing values for the data. NA values are always treated as missing regardless of what is set here. Observations with missing data are discarded before analysis. If input is a vector, then the vector is assumed to contain the missing value codes for all the data. If the input is a matrix, it must be of dimension p x q, where p is the maximum number of missing values and q is the number of columns in the data. Each column of the inputted matrix then specifies the missing data values for the respective variables in data. If null (default), no missing values are in the data other than the standard NA value.
logical, indicates whether aldmck
should print out detailed
output when scaling the data.
integer, specifies the number of dimensions to be estimated.
integer, specifies the minimum number of responses a respondent needs needs to provide to be used in the scaling.
An object of class blackbox
.
vector of data frames of length dims. Each data frame presents results for estimates from that dimension (i.e. x$stimuli[[2]] presents results for dimension 2). Each row contains data on a separate stimulus, and each data frame includes the following variables:
N
Number of respondents who provided a response to this stimulus.
c
Stimulus intercept.
w1
Estimate of the stimulus weight on the first dimension. If viewing the
results for a higher dimension, higher dimension results will appear as w2, w3, etc.
R2
The percent variance explained for the stimulus. This increases as
more dimensions are estimated.
vector of data frames of length dims. Each data frame presents results for estimates from that dimension (i.e. x$stimuli[[2]] presents results for dimension 2). Individuals that are discarded from analysis due to the minscale constraint are NA'd out. Each row contains data on a separate stimulus, and each data frame includes the following variables:
c1
Estimate of the individual intercept on the first dimension. If viewing the
results for a higher dimension, higher dimension results will appear as c2, c3, etc.
A data frame of fit results, with elements listed as follows:
SSE
Sum of squared errors.
SSE.explained
Explained sum of squared error.
percent
Percentage of total variance explained.
SE
Standard error of the estimate, with formula provided on pg. 973 of the article cited below.
singular
Singluar value for the dimension.
Number of rows/stimuli.
Number of columns used in estimation. This may differ from the data set due to columns discarded due to the minscale constraint.
Total number of data entries.
Number of missing entries.
Sum of squares grand mean.
Number of dimensions estimated.
Keith Poole, Jeffrey Lewis, Howard Rosenthal, James Lo, Royce Carroll (2016) ``Recovering a Basic Space from Issue Scales in R.'' Journal of Statistical Software. 69(7), 1--21. doi:10.18637/jss.v069.i07
Keith T. Poole (1998) ``Recovering a Basic Space From a Set of Issue Scales.'' American Journal of Political Science. 42(3), 954-993.
'Issues1980', 'summary.blackbox', 'plot.blackbox'.
# NOT RUN {
### Loads issue scales from the 1980 NES.
data(Issues1980)
Issues1980[Issues1980[,"abortion1"]==7,"abortion1"] <- 8 #missing recode
Issues1980[Issues1980[,"abortion2"]==7,"abortion2"] <- 8 #missing recode
### This command conducts estimates, which we instead load using data()
# Issues1980_bb <- blackbox(Issues1980,missing=c(0,8,9),verbose=FALSE,dims=3,minscale=8)
data(Issues1980_bb)
summary(Issues1980_bb)
# }
Run the code above in your browser using DataLab