pairwise (version 0.3.1)

andersentest.pers: Andersen's Likelihood Ratio Test for Object of class "pers"

Description

The Andersen likelihood ratio test is based on splitting the dataset into subgroups of persons. One can argue that it is a significance testable version of the more descriptive graphical model check - see grm.

Usage

andersentest.pers(pers_obj, split = "median", splitseed = "no", pot = NULL, zerocor = NULL)

Arguments

pers_obj

an object of class"pers" - see function pers.

split

Specifies the splitting criterion. Basically there are three different options available - each with several modes - which are controlled by passing the corresponding character expression to the argument.

1) Using the rawscore for splitting into subsamples with the following modes: split = "median" median raw score split - high score group and low score group; split = "mean" mean raw score split - high score group and low score group. Finaly split = "score" that is splitting daten into as many subsamples as there are raw score groups - discarding min and max (theoretical) score group - which matches the concept proposed by Andersen (1973).

2) Dividing the persons in daten into subsamples with equal size by random allocation with the following modes: split = "random" (which is equivalent to split = "random.2") divides persons into two subsamples with equal size. In general the number of desired subsamples must be expressed after the dot in the character expression - e.g. split = "random.6" divides persons into 6 subsamples (with equal size) by random allocation etc.

3) The third option is using a manifest variable as a splitting criterion. In this case a vector with the same length as number of cases in daten must be passed to the argument grouping the data into subsamples. This vector should be coded as "factor" or a "numeric" integer vector with min = 1.

splitseed

numeric, used for set.seed(splitseed) for random splitting - see argument split.

pot

optional argument, at default (pot=NULL) setting is read from pers_obj - see description for pair.

zerocor

optional argument, at default (zerocor=NULL) setting is read from pers_obj - see description for pair.

pot=pers_obj$pair$fuargs$pot, zerocor=pers_obj$pair$fuargs$zerocor

Value

A (list) object of class "andersentest.pers" ...

Details

Andersen (1973) proposed to split the dataset by [raw] score groups, which can be achieved setting the argument split = "score". However as pointed out by Rost (2004) there might be several different splitting criteria for testing subsample invariance of the raschmodel. Thus the argument split provides some other options for splitting the data - see description of arguments.

References

Andersen, E. B. (1973). A goodness of fit test for the rasch model. Psychometrika, 38(1), 123–140.

Rost, J. (2004). Lehrbuch Testtheorie - Testkonstruktion (2 nd Ed.) Huber: Bern.

Examples

data(bfiN) # loading example data set

data(bfi_cov) # loading covariates to bfiN data set
model <- pers(pair(bfiN,m=6))
andersentest.pers(model, split = bfi_cov$gender)
andersentest.pers(model, split = "random")
andersentest.pers(model, split = "median")
### unsing simulated data:
data("sim200x3")
model2 <- pers(pair(sim200x3))
andersentest.pers(model2, split = "median")