extract.mirtCAT: Extract elements from the internal person, test, and design...
In philchalmers/mirtCAT: Computerized Adaptive Testing with Multidimensional Item Response Theory

extract.mirtCAT

R Documentation

Extract elements from the internal person, test, and design objects

Description

This function extracts elements, as well as builds a few convenient elements, from the three internal person, design, or test objects that are accessible through a customNextItem function definition (see mirtCAT for details).

Usage

extract.mirtCAT(x, what)

Arguments

`x`	either the `person`, `design`, or `test` object defined through a `customNextItem` definition
`what`	a character vector extracting the desired element (see the Details section)

Details

Depending on which object is supplied, the following elements can be extracted.

The 'person' argument

ID: a scalar value indicating the ID of the participant (generally only needed in Monte Carlo simulations)
responses: an integer vector indicating how items that have been responded to. Each element pertains to the associated item location (e.g., responses[100] is associated with the 100th item), and is NA if the item has not been responded to
raw_responses: of the same form as responses, pertaining to the observed responses in a character vector
items_in_bank: an integer vector indicating items which have not been administered yet and are also valid candidates for administration
items_answered: an integer vector indicating the order in which items have been responded to
thetas: the current ability/latent trait estimates given the previously administered items
thetas_SE: the current ability/latent trait standard error estimates given the previously administered items
thetas_history: history of the ability/latent trait estimates
thetas_SE_history: history of the latent trait standard error estimates
item_time: of the same form as items_answered, pertaining to the amount of time it took the participant to response to the item
demographics: a data.frame containing the (optional) prior survey information from the GUI interface
clientData: a list of useful information from shiny's session$clientData

The 'design' argument

items_not_scored: an integer vector indicating items which should be included but not scored in the test (these are experimental items)
min_items: minimum number of items to administer
max_items: maximum number of items to administer
max_time: maximum amount of time alloted to the GUI
met_SEM: logical vector indicating whether the SEM criteria has been met
met_delta_thetas: logical vector indicating whether the delta_thetas criteria has been met
met_classify: logical vector indicating whether the classify criteria has been met
exposure: exposure control elements of the same form as responses
content: content constraint information
content_prop: content proportions
test_properties: user-defined data.frame of test-based properties
person_properties: user-defined data.frame of person-based properties

The 'test' argument

mo: extract the defined model from the mirt package. Afterward, users can use the extract.mirt function to pull out a large number of internal elements for easy use

Author(s)

Phil Chalmers rphilip.chalmers@gmail.com

References

Chalmers, R., P. (2012). mirt: A Multidimensional Item Response Theory Package for the R Environment. Journal of Statistical Software, 48(6), 1-29. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.18637/jss.v048.i06")}

Chalmers, R. P. (2016). Generating Adaptive and Non-Adaptive Test Interfaces for Multidimensional Item Response Theory Applications. Journal of Statistical Software, 71(5), 1-39. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.18637/jss.v071.i05")}

Examples


## Not run: 
 #example test
set.seed(1234)
nitems <- 25
itemnames <- paste0('Item.', 1:nitems)
a <- matrix(rlnorm(nitems, .2, .3))
d <- matrix(rnorm(nitems))
dat <- simdata(a, d, 500, itemtype = 'dich')
colnames(dat) <- itemnames
mod <- mirt(dat, 1, verbose = FALSE, TOL = .01)

# simple math items
questions <- answers <- character(nitems)
choices <- matrix(NA, nitems, 5)
spacing <- floor(d - min(d)) + 1 #easier items have more variation in the options

for(i in 1:nitems){
  n1 <- sample(1:50, 1)
  n2 <- sample(51:100, 1)
  ans <- n1 + n2
  questions[i] <- paste0(n1, ' + ', n2, ' = ?')
  answers[i] <- as.character(ans)
  ch <- ans + sample(c(-5:-1, 1:5) * spacing[i,], 5)
  ch[sample(1:5, 1)] <- ans
  choices[i, ] <- as.character(ch)
}

df <- data.frame(Question=questions, Option=choices, 
  Type = 'radio', stringsAsFactors = FALSE)
df$Answer <- answers

pat <- generate_pattern(mod, Theta = 0, df)

#------------------------------------------------
# administer items in sequence
customNextItem <- function(person, design, test){
   # browser()
   items_left_2_choose_from <- extract.mirtCAT(person, 'items_in_bank')
   min(items_left_2_choose_from)
}

res <- mirtCAT(df, local_pattern=pat, 
  design = list(customNextItem=customNextItem))
summary(res)

#------------------------------------------------
# administer items in order, but stop after 10 items
customNextItem <- function(person, design, test){
   items_left_2_choose_from <- extract.mirtCAT(person, 'items_in_bank')
   items_answered <- extract.mirtCAT(person, 'items_answered')
   total <- sum(!is.na(items_answered))
   ret <- if(total < 10) min(items_left_2_choose_from)
     else return(NA)
   ret
}

res <- mirtCAT(df, local_pattern=pat, 
  design = list(customNextItem=customNextItem))
summary(res)

#------------------------------------------------
# using findNextItem() and stopping after 10 items

customNextItem <- function(person, design, test){
   items_answered <- extract.mirtCAT(person, 'items_answered')
   total <- sum(!is.na(items_answered))
   ret <- NA
   if(total < 10) 
     ret <- findNextItem(person=person, test=test, design=design, criteria = 'MI')
   ret
}

res <- mirtCAT(df, mod, local_pattern=pat, start_item = 'MI',
  design = list(customNextItem=customNextItem))
summary(res)

# equivalent to the following
res2 <- mirtCAT(df, mod, local_pattern=pat, start_item = 'MI', 
  criteria = 'MI', design = list(max_items = 10))
summary(res2)


## End(Not run)

philchalmers/mirtCAT documentation built on Oct. 24, 2023, 11:14 a.m.