Granger meets Rasch. Investigating Granger causation with multidimensional longitudinal item response.

Autor/inn/en	Koller, Ingrid; Carstensen, Claus H.; Wiedermann, Wolfgang; Eye, Alexander von
Titel	Granger meets Rasch. Investigating Granger causation with multidimensional longitudinal item response.
Quelle	Aus: Wiedermann, Wolfgang (Hrsg.); Eye, Alexander von (Hrsg.): Statistics and causality. Methods for applied empirical research. 1. Auflage. New York, NY: John Wiley & Sons (2016) S. 231-248 PDF als Volltext Link als defekt meldenVerfügbarkeit
Reihe	Wiley Series in Probability and Statistics
Beigaben	Literaturangaben
Sprache	englisch
Dokumenttyp	online; gedruckt; Sammelwerksbeitrag
ISBN	978-1-118-94704-3; 978-1-118-94705-0
DOI	10.1002/9781118947074.ch10
Schlagwörter	Rasch-Modell; Kompetenzmessung; Regressionsanalyse; Kausalität; Datenverarbeitung
Abstract	This chapter introduces the basics of Granger causation. It describes multidimensional longitudinal item response models and the investigation of Granger causation with multidimensional longitudinal item response models. The chapter provides an empirical example on the topic of science knowledge of children. It addresses potential limitations of the approach and outlines potential future research. The Rasch model (RM) is a special case of the multidimensional random coefficient multinomial logit model, which is a generalization of a wide class of Rasch type models, for example, the partial credit model or the linear logistic test model. The possibility to estimate the measurement model and to assess Granger causation may be highly valuable for psychometric research. Future research in the field of multidimensional item response models should examine how many dimensions can be considered in one step and which effect complex linking designs can have on parameter estimation and interpretation. (Orig.).
Erfasst von	DIPF \| Leibniz-Institut für Bildungsforschung und Bildungsinformation, Frankfurt am Main
Update	2020/2