Rwinsteps (version 1.0-1.1)

wdat: Winsteps Data File

Description

These functions read and write Winsteps data files, where formatting is specified in a command file object.

Usage

write.wdat(x, cmd, datfile = cmd$data, na = " ")
read.wdat(cmd, datfile = cmd$data, na = " ", ilabels = NULL)

Arguments

x

data frame of item responses

cmd

command file object

datfile

path to which the data file will be written, or from which it will be read, defaulting to the path supplied in the command file

na

character representing missing data, defaulting to a blank space

ilabels

vector of new item labels to be added to the item columns, overriding cmd$labels

Value

read.wdat returns a data frame

Details

When reading data, the person names should occupy 1 column and each item should occupy 1 column in the file. A warning is returned if any rows have fewer than the maximum number of columns and the person names are not in the last column, in which case the missing information pertains to items. ilabels will override the item labels in cmd$labels, if they exist. If both are NULL, generic item names are supplied.

When writing data, all unnamed columns are assumed to be item responses. If the command file contains item labels they will be used to extract and reorder the item responses from x and all other columns in x will be ignored, except for the person IDs which must have a column name of "name". The data are written to the filename datfile in fixed width format according to the item1, ni, name1, and namelen command file components.

Examples

# NOT RUN {
# Simulate data, with missings
set.seed(82911)
b <- seq(-3, 3, length = 15)
theta <- rnorm(100, 1)
rmat <- ifelse(rirf(b, theta)$p > runif(1500), 1, 0)
rmat[sample(1500, 30)] <- NA
rmat <- data.frame(rmat)

# Item and person labels
colnames(rmat) <- paste("i", 1:15, sep = "")
rmat$name <- paste("p", 1:100, sep = "")

# Create command file
tempdat <- tempfile()
cmd <- wcmd(title = "R2Winsteps Example", data = tempdat,
  item1 = 1, ni = 15, name1 = 16, namelen = 5,
  labels = paste("i", 1:15, sep = ""), hlines = "Y")

# Write to temp file and read back in
write.wdat(rmat, cmd, na = "N")
rmat2 <- read.wdat(cmd, na = "N")
# }