In markdly/conquestr: Conquest Version 2 and R

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

conquestr

ConQuest generalised item response modelling software (Wu, Adams, Wilson, & Haldane, 2007) is used for a wide variety of purposes including the analysis of educational assessment data.

conquestr is an R package whose first goal was to make it easier to import ConQuest item analysis files into R. This functionality does not appear to be available in other R packages as far as we are aware.

Other resources

Several other R packages already integrate with ConQuest to various degrees. For example:

See TAM if you'd like to do analysis similar to ConQuest within R.

The R2conquest function in the sirt package provides functionality for using ConQuest from within R.

The WrightMap package also contains various features for working with ConQuest and other IRT packages.

Installation

You can install conquestr from github with:

# install.packages("devtools")
devtools::install_github("markdly/conquestr")

Alternatively, the latest updates are made and tested using the develop branch which can be installed with

# install.packages("devtools")
devtools::install_github("markdly/conquestr#develop")

Example

First load the relevant packages

library(conquestr)

Take a ConQuest version 2 item analysis file. In ConQuest, item analysis text files are generated by the ConQuest itanal command (e.g. itanal >> my_item_analysis_file.itn;). An example item analysis file is included with conquestr out of the box:

cq_example()

Use the example file to try out the other functions. The main one is cq_itanal:

# get the path to the example itanal file that comes with conquestr
fname <- cq_example(display = FALSE)

# import the itanal as a tibble
df <- cq_itanal(fname)

Once imported, the original text imported for each 'item' can be found in the data column.

df$data[1]

Item level details are provided in separate columns:

head(dplyr::select(df, -c(data, resp_stat)))

The details for each response category is contained in the list column resp_stat. The function cq_resp_stats() makes viewing these easier:

head(cq_resp_stats(df))

Other examples

Plausible values files

fname <- cq_example(display = FALSE, example_name = "ex1_10.pv") 
pv <- cq_pv(fname, np = 10)  
head(pv)

By default the pv files are imported in long format. To convert them to a more typical 'wide' format one way this can be done is:

library(dplyr)
library(tidyr)
pv_wide <- pv %>% select(recid, field, val) %>% spread(field, val)
head(pv_wide)

References

Wu, M., Adams, R., Wilson, M., & Haldane, S. (2007). ConQuest Version 2: Generalised Item Response Modelling Software.

markdly/conquestr documentation built on May 29, 2019, 5:40 a.m.