2020/02/18 by Lawrence T. DeCarlo, Xiaoliang Zhou
Decision Sciences · Mathematics · #Advanced Statistical Methods and Models #Reliability and Agreement in Measurement #Statistical Methods and Bayesian Inference
paper · doi:10.1111/jedm.12265
crossref issued 2020/02/18 · crossref published 2020/02/18 · crossref published-online 2020/02/18 · openalex publication_date 2020/02/18 · crossref created 2020/02/18 · crossref published-print 2021/03/01 · crossref deposited 2023/09/01 · openalex created_date 2025/10/10 · crossref indexed 2026/07/24 · openalex updated_date 2026/07/25
Abstract In signal detection rater models for constructed response (CR) scoring, it is assumed that raters discriminate equally well between different latent classes defined by the scoring rubric. An extended model that relaxes this assumption is introduced; the model recognizes that a rater may not discriminate equally well between some of the scoring classes. The extension recognizes a different type of rater effect and is shown to offer useful tests and diagnostic plots of the equal discrimination assumption, along with ways to assess rater accuracy and various rater effects. The approach is illustrated with an application to a large‐scale language test.