mokken (version 1.2)

search.normal: Automated Item Selection Algorithm for Mokken Scale Analysis

Description

Returns a vector with as many elements as there are items, indicating the scale an item belongs to

Usage

search.normal(X, lowerbound = 0.3, alpha = 0.05)

Arguments

X

matrix or data frame of numeric data containing the responses of nrow(X) respondents to ncol(X) items. Missing values are not allowed

lowerbound

numeric scaling criterium; 0 <= lowerbound < 1

alpha

Type I error level

Value

• An indicator vector of length J. Each entry refers to an item. Items with same integer belong to the same Mokken scale. A zero indicates an unscalable item. If n is the largest integer, then n Mokken scales were found.

Details

The number of Mokken scales cannot exceed ncol(X)/2. Procedure may be slow for large data sets.

References

Mokken, R. J. (1971) A Theory and Procedure of Scale Analysis. Berlin, Germany: De Gruyter. Molenaar, I.W. and Sijtsma, K. (2000) User's Manual MSP5 for Windows [Software manual]. Groningen, The Netherlands: IEC ProGAMMA. Sijtsma, K, and Molenaar, I. W. (2002) Introduction to nonparametric item response theory. Thousand Oaks, CA: Sage. Van der Ark, L. A. (2007). Mokken scale analysis in R. Journal of Statistical Software. http://www.jstatsoft.org

See Also

coefH, check.monotonicity, check.pmatrix, check.restscore

Examples

data(acl)
# Partition all 212 items into mokken scales (may take some time).
scale <- search.normal(acl)      

# investigate monotonicity for all items in the first scale.
monotonicity.list <- check.monotonicity(acl[,scale==1])

# summary of the results
summary(monotonicity.list)