Premium
Calculating Conditional Reliability for Dynamic Measurement Model Capacity Estimates
Author(s) -
McNeish Daniel,
Dumas Denis
Publication year - 2018
Publication title -
journal of educational measurement
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.917
H-Index - 47
eISSN - 1745-3984
pISSN - 0022-0655
DOI - 10.1111/jedm.12195
Subject(s) - item response theory , reliability (semiconductor) , classical test theory , econometrics , computer science , test theory , validity , psychology , reliability engineering , statistics , machine learning , psychometrics , mathematics , engineering , power (physics) , physics , quantum mechanics
Dynamic measurement modeling (DMM) is a recent framework for measuring developing constructs whose manifestation occurs after an assessment is administered (e.g., learning capacity). Empirical studies have suggested that DMM may improve consequential validity of test scores because DMM learning capacity estimates were shown to be much less related to demographic factors like examinees’ socioeconomic status compared to traditional single‐administration item response theory (IRT)–based estimates. Though promotion of DMM has hinged on improved validity, no methods for computing reliability (a prerequisite for validity) have been advanced and DMM is sufficiently different from classical test theory (CTT) and IRT that known methods cannot be directly imported. This article advances one method for computing conditional reliability for DMM so that precision of the estimates can be assessed.