Modeling competence development in the presence of selection bias.

Autor/inn/en	Zinn, Sabine; Gnambs, Timo
Titel	Modeling competence development in the presence of selection bias.
Quelle	In: Behavior research methods, 50 (2018) 6, S. 2426-2441 PDF als Volltext kostenfreie Datei Link als defekt meldenVerfügbarkeit
Beigaben	Literaturangaben
Sprache	englisch
Dokumenttyp	online; Zeitschriftenaufsatz
ISSN	0743-3808; 1554-3528
DOI	10.3758/s13428-018-1021-z
Schlagwörter	Längsschnittuntersuchung; Vergleichsuntersuchung; Vorurteil; Mathematische Kompetenz; Analyse; Kompetenzentwicklung; Statistische Methode; NEPS (National Educational Panel Study)
Abstract	A major challenge for representative longitudinal studies is panel attrition, because some respondents refuse to continue participating across all measurement waves. Depending on the nature of this selection process, statistical inferences based on the observed sample can be biased. Therefore, statistical analyses need to consider a missing-data mechanism. Because each missing-data model hinges on frequently untestable assumptions, sensitivity analyses are indispensable to gauging the robustness of statistical inferences. This article highlights contemporary approaches for applied researchers to acknowledge missing data in longitudinal, multilevel modeling and shows how sensitivity analyses can guide their interpretation. Using a representative sample of N = 13,417 German students, the development of mathematical competence across three years was examined by contrasting seven missing-data models, including listwise deletion, full-information maximum likelihood estimation, inverse probability weighting, multiple imputation, selection models, and pattern mixture models. These analyses identified strong selection effects related to various individual and context factors. Comparative analyses revealed that inverse probability weighting performed rather poorly in growth curve modeling. Moreover, school-specific effects should be acknowledged in missing-data models for educational data. Finally, we demonstrated how sensitivity analyses can be used to gauge the robustness of the identified effects. (Orig.).
Erfasst von	DIPF \| Leibniz-Institut für Bildungsforschung und Bildungsinformation, Frankfurt am Main
Update	2021/2