get_item_index_max_information: Item index maximum information

In Karel-Kroeze/ShadowCAT: Multidimensional Computer Adaptive Testing with the Shadow Testing Procedure

Description

Get the index of the item with maximum information from the available items. For multi dimensional models, items that only load on dimensions for which the variance target has already been reached, will not be returned.

Usage

get_item_index_max_information(available, item_information, estimate, stop_test,
  alpha, number_answers)

Arguments

available	Vector with indices of available items.
item_information	Vector with summarized information of each yet available item, with zeros for administered items (as returned by get_summarized_information with pad = TRUE).
estimate	Vector containing the theta estimate, with its covariance matrix as an attribute.
stop_test	List indicating rules for when to terminate the test. Should be a list of the form list(target = ..., max_n = ..., min_n = ..., cutoffs = ...), where target is a vector indicating the maximum acceptable variance per dimension; NULL means no variance target, max_n is the test length at which the test should be terminated (even if the target has not been reached yet), min_n is the minimum test length; NULL means no mimimum test length, and cutoffs is a matrix containing cut off values per dimension (columns) and test iteration (rows). First row contains cut off values for when no items have been administered yet, second row for when one item has been administered, etc. If estimate + 3SE < cutoff for each dimension at a certain iteration, test terminates; NULL means no cut off values.
alpha	Matrix of alpha parameters, one column per dimension, one row per item. Row names should contain the item keys. Note that so called within-dimensional models still use an alpha matrix, they simply have only one non-zero loading per item.
number_answers	The number of answers given thus far (length of administered).

Value

Index of item with maximum information.

Karel-Kroeze/ShadowCAT documentation built on May 7, 2019, 12:28 p.m.