Extract Item Difficulty Estimates (beta)

Returns a tidy table of estimated item difficulties with credible intervals and convergence diagnostics.

Usage

item_params(object, prob = 0.95)

Arguments

object

A fitted model from rasch_fit().

prob

Width of the credible interval (0 to 1). Default 0.95 gives a 95% credible interval.

Value

A data frame with columns:

item

Question name.

mean

Posterior mean — the "best guess" point estimate.

median

Posterior median — robust alternative to the mean.

sd

Posterior SD — how uncertain we are about this estimate.

q_lower, q_upper

Credible interval bounds.

rhat

Convergence diagnostic. Must be < 1.01.

ess_bulk

Effective sample size (center). Should be > 400.

ess_tail

Effective sample size (tails). Should be > 400.