Shadow-methods function

Run adaptive test assembly

Shadow is a test assembly function for performing adaptive test assembly based on the generalized shadow-test framework.

Shadow(
  config,
  constraints = NULL,
  true_theta = NULL,
  data = NULL,
  prior = NULL,
  prior_par = NULL,
  exclude = NULL,
  include_items_for_estimation = NULL,
  force_solver = FALSE,
  session = NULL,
  seed = NULL,
  cumulative_usage_matrix = NULL
)

## S4 method for signature 'config_Shadow'
Shadow(
  config,
  constraints = NULL,
  true_theta = NULL,
  data = NULL,
  prior = NULL,
  prior_par = NULL,
  exclude = NULL,
  include_items_for_estimation = NULL,
  force_solver = FALSE,
  session = NULL,
  seed = NULL,
  cumulative_usage_matrix = NULL
)

Arguments

• config: a config_Shadow object. Use createShadowTestConfig for this.

• constraints: a constraints object representing test specifications. Use loadConstraints for this.

• true_theta: (optional) true theta values to use in simulation. Either true_theta or data must be supplied.

• data: (optional) a matrix containing item response data to use in simulation. Either true_theta or data must be supplied.

• prior: (optional) density at each config@theta_grid to use as prior. Must be a length-nq vector or a nj * nq matrix. This overrides prior_dist and prior_par in the config. prior and prior_par cannot be used simultaneously.

• prior_par: (optional) normal distribution parameters c(mean, sd) to use as prior. Must be a length-nq vector or a nj * nq matrix. This overrides prior_dist and prior_par in the config. prior and prior_par cannot be used simultaneously.

• exclude: (optional) a list containing item names in $i and set names in $s to exclude from selection for each participant. The length of the list must be equal to the number of participants.

• include_items_for_estimation: (optional) an examinee-wise list containing:

  • administered_item_pool items to include in theta estimation as item_pool object.
  • administered_item_resp item responses to include in theta estimation.

• force_solver: if TRUE, do not check whether the solver is one of recommended solvers for complex problems (set-based assembly, partitioning). (default = FALSE)

• session: (optional) used to communicate with Shiny app TestDesign.

• seed: (optional) used to perform data generation internally.

• cumulative_usage_matrix: (optional) a nj by (ni + ns) matrix containing the number of times the item/stimulus was administered previously to each participant. Stimuli representations are appended to the right side of the matrix.

Returns

Shadow returns an output_Shadow_all object containing assembly results.

Examples

config <- createShadowTestConfig()
true_theta <- rnorm(1)
solution <- Shadow(config, constraints_science, true_theta)
solution@output

References

van der Linden, W. J., Reese, L. M. (1998). A model for optimal constrained adaptive testing. Applied Psychological Measurement, 22, 259-270.

van der Linden, W. J. (1998). Optimal assembly of psychological and educational tests. Applied Psychological Measurement, 22, 195-211.

van der Linden, W. J. (2000). Optimal assembly of tests with item sets. Applied Psychological Measurement, 24, 225-240.

van der Linden, W. J. (2005). Linear models for optimal test design.

Springer Science & Business Media.