Year: 2014
Type: article
Abstract: This article explores the use of longitudinal regression as a tool for identifying scoring inaccuracies. Student progression patterns, as evaluated through longitudinal regressions, typically are more stable from year to year than are scale score distributions and statistics, which require representative samples to conduct credibility checks. Historical data from a large‐scale K‐12 testing program were used to evaluate the usefulness of several proposed longitudinal procedures. Results showed that the use of longitudinal regression in quality control was effective in detecting scoring errors, especially when the scoring errors were non‐negligible.
Source: ETS Research Report Series
DOI: 10.1002/ets2.12032
Authors: Ying Lu, Wendy M. Yen
Language: English
Field: Decision Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: bronze
OpenAlex ID: W1518304353
