Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 392 - Bayesian Analysis of Complex, Structured Health and Social Data
Type: Contributed
Date/Time: Wednesday, August 10, 2022 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #322430
Title: Bayesian Analysis of Interrater and Intrarater Reliability with Multilevel Data
Author(s): Nour Hawila* and Arthur Berg
Companies: Penn State University - College of Medicine and Penn State University - College of Medicine
Keywords: interrater reliability; multilevel data; Bayesian model; Stan; kappa

We investigate and generalize three different published models for analyzing interrater and intrarater reliability in the presence of multilevel data. We derive generalized formulas for the kappa coefficient measure of agreement under these three models. Utilizing Bayesian models implemented in Stan, new estimates of interrater and intrarater reliability are constructed. Marginal correlations in the Bayesian models can also provide alternative measures of reliability assessment. Comparisons of the kappa estimates and marginal correlations across the different models are presented. Extensive simulations demonstrate the properties of the different measures of agreement according to different model choices. An R package is developed to easily fit the Stan models and estimate the various reliability measures in the presence of multilevel data.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program