subscales {CTT} | R Documentation |
Function to create subscales based on a design matrix
Description
This convenience function is provided to facilitate extracting subscales from a single set of item responses.
Usage
subscales(items, scales, scale.names = NA, score.items = FALSE,
check.reliability = FALSE, key=NA)
Arguments
items |
The item response (scored or not) |
scales |
A design matrix, with items represented in rows and separate subscales represented in columns. An item may appear in more than one subscale. |
scale.names |
Optional vector of names for the subscales |
score.items |
If responses are not scored, they may be scored using score.items=TRUE (key must be provided) |
check.reliability |
If check.reliability=TRUE, the reliability for each subscale will be calculted |
key |
Optional key, required only if score.scales=TRUE. |
Details
This function provides an easy way to create new datasets from a single set of item responses. This function is also a front end for score and reliability, enabling the item responses to be partitioned into separate scales, scored, and reliability analyses performed using this one function.
Value
A list is returned. Results for each subscale (i.e., column in the scales matrix) are provided as sparate objects in that list.
score |
Each examinee's score on the associated subscale |
reliablity |
Reliability results (if requested) for the associated subscale |
scored |
The scored item responses (if required) for each respondent for the associated subscale |
Author(s)
John Willse, Zhan Shu
See Also
reliability, score
Examples
# Example data included with package
data(CTTdata)
data(CTTkey)
# design matrix
q <- matrix(c(1,0,
1,0,
1,0,
1,0,
1,0,
1,0,
1,0,
1,0,
1,0,
1,1,
0,1,
0,1,
0,1,
0,1,
0,1,
0,1,
0,1,
0,1,
0,1,
0,1), ncol=2, byrow=TRUE)
subscales(CTTdata,q,c("T1","T2"),TRUE,TRUE,CTTkey)