rel_info

Compute marginal reliability for a unidimensional IRT model using either
MLE-based or EAP-based information, via Gaussian quadrature over a
standard normal ability distribution.

Provides functions for computing test reliability and conditional
standard error of measurement (CSEM) based on the methods described in
the Reliability in Educational Measurement chapter of the 5th edition of
"Educational Measurement" by Lee and Harris (2025, ISBN:9780197654965).

Huan Liu

emreliability

Test Reliability and CSEM in Educational Measurement

Won-Chan Lee

Min Liang

rel_info function

<dl><dt>ip</dt>
<dd>A data frame or matrix of item parameters with columns in the order
<code>b</code>, <code>a</code>, <code>c</code>, where <code>a</code> is on the <code>D = 1.702</code>
metric. If only 1 or 2 columns are supplied, the <code>info()</code> function is
expected to treat them as 1PL/2PL accordingly.</dd>
<dt>est</dt>
<dd>A character string specifying the ability estimation method:
<code>"MLE"</code> for maximum likelihood or <code>"EAP"</code> for empirical Bayes.</dd></dl>

Arguments

Marginal Reliability of a Unidimensional IRT Model — rel_info

<dl>

<dt>ip</dt>
<dd>A data frame or matrix of item parameters with columns in the order
<code>b</code>, <code>a</code>, <code>c</code>, where <code>a</code> is on the <code>D = 1.702</code>
metric. If only 1 or 2 columns are supplied, the <code>info()</code> function is
expected to treat them as 1PL/2PL accordingly.</dd>


<dt>est</dt>
<dd>A character string specifying the ability estimation method:
<code>"MLE"</code> for maximum likelihood or <code>"EAP"</code> for empirical Bayes.</dd>

</dl>

rel_info: Marginal Reliability of a Unidimensional IRT Model

Description

Usage

Value

Arguments

Details

Examples