data_gen: Generate data from the one-parameter RSC model.
In peterhalpin/cirt: Psychometric models of small group collaborations

Description Usage Arguments Value

This is a wrapper for sim_RSC that saves the data generating parms and allows for multple response patterns per group. To generate data from a 2PL model, set theta1 = theta2 and w = 1/2.

1 2	data_gen(n_reps, u, parms, theta1, theta2, theta1_se = NULL, theta2_se = NULL, NA_pattern = NULL)

`n_reps`	integer indicating how many datasets to generate.
`u`	the logit of the weight parameter of the RSC model.
`parms`	a data.frame with columns `parms$alpha` and `parms$beta` corresponding to the discrimination and difficulty parameters of the 2PL model, respectively.
`theta1`	the latent trait for member 1.
`theta2`	the latent trait for member 2.
`theta1_se`	the standard error of the latent trait for member 1 (optional). If included, data generation samples `n_reps` values from `rnorm(theta1, theta1_se)`.
`theta2_se`	the standard error of the latent trait for member 2 (optional). If included, data generation samples `n_reps` values from `rnorm(theta2, theta2_se)`.
`NA_pattern`	an (optional) `length(u)` by `nrow(parms)` data.frame with `NA` entries denoting missing data. The missing values are preserved in the generated data.

A data.frame with length(u) rows containing an id variable for each pair and each sample, the data generating values of u, theta1, and theta2, and the simulated response patterns.

peterhalpin/cirt documentation built on May 25, 2019, 1:50 a.m.