| Verfügbarkeit |
Autoren | Pohl, Steffi; Carstensen, Claus H. |
Titel | NEPS technical report - scaling the data of the competence tests. |
URL | https://www.neps-data.de/Portals/0/Working Papers/WP_XIV.pdf |
Erscheinungsjahr | 2012 |
Seitenzahl | 31 S. |
Verlag | Bamberg: Otto-Friedrich-Universität |
Reihe | NEPS Working Paper. Band 14 |
Dokumenttyp | Monographie; Discussion Paper / Working Paper / Konferenzbeitrag; online |
Beigaben | Literaturangaben, Abbildungen, Tabellen, Anhang |
Sprache | englisch |
Forschungsschwerpunkt | Bildungspanel (NEPS) |
Schlagwörter | Bildungsforschung; Kompetenzmessung; Bildungsverlauf; Testentwicklung; Plausibilität; Item-Response-Theorie; Skalierung; Modellierung; Empirische Untersuchung; Quantitative Methode; Deutschland |
Abstract | The National Educational Panel Study (NEPS) aims at investigating the development of competences across the whole life span. Tests for assessing the different competences are developed in NEPS and response data is collected from study participants on different competence domains in different age cohorts. The data of the competence tests are scaled using models of Item Response Theory (IRT). In the Scientific Use File (SUF) competence data are provided for researcher in form of item responses, manifest scale scores, as well as plausible values that allow investigating latent relationships. This paper aims at achieving different purposes. First, at describing the scaling model used to estimate competence scores in NEPS. This includes aspects like dealing with different response formats and accounting for missing responses in the estimation, as well as describing the parameters that are estimated in the model. Second, describing the various analyses that are performed for checking the quality of the competence tests. This includes item fit measures, differential item functioning, test targeting, unidimensionality, and local item independence. And third, outlining different approaches on how the competence data provided in the SUF may be used for further analyses. While the sections on the scaling model and the quality check are written for researchers familiar with IRT, the section on how to use the competence scores provided in the SUF is written for substantive researchers interested in using competence scores to investigate research questions. ConQuest-syntax is provided for some analyses examples. (DIPF/Orig.) |
Förderkennzeichen | 01GJ0888 |