On the empirical construction of implications between bi-valued test items

We describe a data-analytic method which allows one to derive implications between test items from observed response patterns to these items. This method is a further development of item tree analysis (J.F.J.v. Leeuwe, Nederlands Tijdschrift voor de Psychologie, 29 (1974) 475–484). A simulation stud...

Full description

Saved in:
Bibliographic Details
Published in:Mathematical social sciences 1999-11, Vol.38 (3), p.361-375
Main Author: Schrepp, Martin
Format: Article
Language:English
Subjects:
Data analysis
Empirical research
Improved method
Item tree analysis
Mathematical models
Methodology
Test items
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We describe a data-analytic method which allows one to derive implications between test items from observed response patterns to these items. This method is a further development of item tree analysis (J.F.J.v. Leeuwe, Nederlands Tijdschrift voor de Psychologie, 29 (1974) 475–484). A simulation study shows that the improvements have a positive effect on the ability of the method to reconstruct the valid implications from data. The improved method is able to reconstruct the underlying structure on the items with high accuracy. As a concrete example the implications in a set of letter series completion problems are analysed.
ISSN:0165-4896
1879-3118
DOI:10.1016/S0165-4896(99)00025-6