plotICC {eatModel}R Documentation

Plots item charactestic curves.

Description

Function provides item characteristic plots for each item. To date, only dichotomouls 1pl and 2pl models are supported.

Usage

plotICC  ( resultsObj, defineModelObj, item = NULL, personsPerGroup = 30, 
       pdfFolder = NULL, smooth = 20 )

Arguments

resultsObj

The object returned by getResults.

defineModelObj

The object returned by defineModel.

item

Optional: The item for which the ICC should be plotted. If NULL, ICCs of all items will be collected in a common pdf. The pdfFolder argument than must not be NULL.

personsPerGroup

Specifies the number of persons in each interval of the theta scale for dividing the persons in various groups according to mean EAP score.

pdfFolder

Optional: A folder with writing access for the pdf file. Necessary only if ICCs for more than one item should be plotted.

smooth

Optional: A parameter (integer vaklue) for smoothing the plot. If the number of examinees is high, the icc plot may become scratchy. smooth defines the maximum number of discret nodes across the theta scale for evaluating the icc. Higher values result in a less smooth icc.

Note

This function is beta! Use with care...

Author(s)

Sebastian Weirich

Examples

data(sciences)

# first reshape the data set into wide format
datW <- reshape2::dcast(sciences, id+grade+sex~variable, value.var="value")

# defining the model: specifying q matrix is not necessary
mod1 <- defineModel(dat=datW, items= -c(1:3), id="id", software = "tam")

# run the model
run1 <- runModel(mod1)

# get the results
res1 <- getResults(run1)

# plot for one item 
plotICC  ( resultsObj = res1, defineModelObj = mod1, item = "BioPro13")

[Package eatModel version 0.6.11 Index]