subscales function

Function to create subscales based on a design matrix

Function to create subscales based on a design matrix

This convenience function is provided to facilitate extracting subscales from a single set of item responses.

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.

Returns

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)
  • Maintainer: John T. Willse
  • License: GPL (>= 2)
  • Last published: 2018-09-12

Useful links