simEquiTable: Computes equivalence table based on simulated data

Description Usage Arguments Value Author(s) Examples

Description

Function provides the equivalence table for unidimensional models, specifying the individual competence level for each total score of the test.

Usage

1
2
simEquiTable  ( anchor, mRef, sdRef, addConst = 500, multConst = 100, 
cutScores , dir , n = 2000, conquest.folder )

Arguments

anchor

A data frame with anchor parameters on the logit scale, transformed to the metric of the reference population. The first column contains the names of all anchored items. The second column contains anchor parameters.

mRef

Scalar: mean of the reference population.

sdRef

Scalar: Standard deviation of the reference population.

addConst

Additive constant for parameter transformation.

multConst

Multiplicative constant for parameter transformation.

cutScores

Named list of one or two elements. "values" is a numeric vector of cut scores (increasing), "labels" is an optional character vector of cut score labels. Note that "labels" (if specified) has to be of one more length than "values".

dir

An already existsing directory with writing permission in which the (temporary) output will be written to.

n

Scalar: sample size for the simulated data.

conquest.folder

Character string with path and name of the ConQuest console, for example "c:/programme/conquest/console_Feb2007.exe".

Value

A list of two data frames, including the complete table and the reduced table with the following 5 columns.

Score

Students raw score

Estimate

Estimated individual WLE according to the raw score.

std.error

Standard error of the individual WLE estimate. This column is not included in the reduced table.

estBista

Transformed WLE

ks

competence level

Author(s)

Sebastian Weirich

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
## Not run: 
# create arbitrary anchor parameter
anchor <- data.frame ( item = paste("i",1:20,sep=""), 
          par = rnorm(20, mean = -.1, sd = 1.5))

# create arbitrary cut scores
# note that the number of labels (if specified) must equal the number of cuts + 1
cuts   <- list ( values = 330+0:4*75, labels = c("1a", "1b", 2:5) )

# create the equivalence table
ret <- simEquiTable( anchor = anchor, cutScores = cuts , mRef = -0.05, sdRef = 0.9, 
       dir = "c:/users/weirichs/test", conquest.folder = "N:/console_Feb2007.exe")
View(ret$short)       

## End(Not run)

eatModel documentation built on May 2, 2019, 6:49 p.m.