---
title: "PF 10 Analysis"
output: html_notebook
---
```{r}
library(eRm)
```
Next load the data
```{r load}
source("pf10examples.R")
```
## Dichotomous Scale
```{r pf10di.items}
pf10dich.items
```
```{r pf10di}
pf10dich.students
```
## Start Rasch model analysis
```{r Rasch}
## data in pf10examples.R
pf10.rm <- RM(pf10dich.students)
pf10.rm
```
Look at Item parameters
```{r item.par}
## This provides informationa about items
coef(pf10.rm)
```
## Build raw score to total score conversion.
This provides information about raw score to scale
score conversion
```{r scale}
person.parameter(pf10.rm)
plot(person.parameter(pf10.rm))
```
## Wright Map or Person/Item map
```{r wright}
plotPImap(pf10.rm)
```
## plot ICC curves for items
```{r ICC}
plotICC(pf10.rm,ask=FALSE)
```
## Look at test information
```{r test info}
plotINFO(pf10.rm)
```
# Polytomous data
* PCM -- partial credit model -- each item has same scale
* RSM -- rating scale model -- all items use common (Likert) scale
## Fit the model
```{r rsm}
## Note for trichotomous data, first column is gender.
pf10.rsm <- RSM(pf10trich.students[,-1]) ## Leave out first column
summary(pf10.rsm)
```
## Look at ICC curves
```{r ICC2}
plotICC(pf10.rsm,ask=FALSE)
```
## Look at Wright Map
```{r wright3}
plotPImap(pf10.rsm)
```
## Conversion table
```{r table.rsm}
person.parameter(pf10.rsm)
plot(person.parameter(pf10.rsm))
```
## Look at test information
```{r rsm.info}
plotINFO(pf10.rsm)
```
# Partial Credit Model
* PCM -- partial credit model -- each item has same scale
## Fit the model
```{r pcm}
## Note for trichotomous data, first column is gender.
pf10.pcm <- PCM(pf10trich.students[,-1]) ## Leave out first column
summary(pf10.pcm)
```
## Look at ICC curves
```{r ICC2 pcm}
plotICC(pf10.pcm,ask=FALSE)
```
## Look at Wright Map
```{r wright3 pcm}
plotPImap(pf10.pcm)
```
## Conversion table
```{r table.pcm}
person.parameter(pf10.pcm)
plot(person.parameter(pf10.pcm))
```
## Look at test information
```{r pcm.info}
plotINFO(pf10.pcm)
```