A Multilevel Longitudinal Nested Logit Model for Measuring Changes in Correct Response and Error Types

Autor/inn/en	Suh, Youngsuk; Cho, Sun-Joo; Bottge, Brian A.
Titel	A Multilevel Longitudinal Nested Logit Model for Measuring Changes in Correct Response and Error Types
Quelle	42 (2018) 1, S.73-88 (16 Seiten)Infoseite zur Zeitschrift PDF als Volltext (1); PDF als Volltext kostenfreie Datei (2) Verfügbarkeit
Zusatzinformation	Weitere Informationen
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
Schlagwörter	Hierarchical Linear Modeling; Longitudinal Studies; Error Patterns; Change; Responses; Pretests Posttests; Bayesian Statistics; Models; Item Response Theory; Instructional Effectiveness + Suchen Sie Ihr Suchwort? Longitudinal study; Longitudinal method; Longitudinal methods; Längsschnittuntersuchung; Fehlertyp; Wandel; Analogiemodell; Item-Response-Theorie; Unterrichtserfolg
Abstract	This article presents a multilevel longitudinal nested logit model for analyzing correct response and error types in multilevel longitudinal intervention data collected under a pretest-posttest, cluster randomized trial design. The use of the model is illustrated with a real data analysis, including a model comparison study regarding model complexity and cluster bias. Two substantive research questions regarding the intervention effect on correct response probability and error patterns are investigated using the proposed model. The recovery of item parameters for the proposed model using two sample size conditions is examined via a simulation study. The accuracy of the parameter estimates is comparable with those found in previous studies for the same family of models, except for the intercept parameters of correct responses. Finally, the impact of ignoring cluster membership in the model on the parameter estimation is also studied by fitting a single-level model to multilevel data. Ignoring cluster membership in the model adversely affects the estimation of intercept parameters in correct and error responses. (As Provided).
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01