Möbus, Claus and Göricke, G. and Kröh, Peer (1983) Statistical Analysis of Single-Case Experimental Designs: Conditional Equivalence of the General-Linear-Model Approach of GLASS, WILLSON & GOTTMAN with the Intervention Model of BOX & TIAO. EDP in Medicine and Biology, 14 (3-4). pp. 98-108. ISSN 0300-8282

[img]
Preview
- Published Version

Volltext (2848Kb)
Official URL: http://www.qucosa.de/fileadmin/data/qucosa/documen...

Abstract

In single case diaqnostics the researcher has often to evaluate the influence of his intervention. If the data are measured at the level of interval-scales impact assessment can be pursued by the GLASS, WILLSON & GOTTMAN (1975) approach, which can be regarded as a mixture of the general linear model and time series modelling of the error component, or by the intervention model of BOX & TIAO (1975), which is a descendant of BOX & JENKINS transfer model (1970). In clinical research only the former method is used for impact assessment. The latter model is up to now not widely known. Even in statistical literature no unified treatment of both approaches can be found. We want to show that both are equivalent under certain conditions. But inspite of this, researchers are advised to abandon the GLASS, WILLSON & GOTTMAN method in favor of the intervention model because of its greater elegance and practicability. This means that the danger of misspecifying the intervention effects is far more negligible in the intervention model than in the GLASS, WILLSON & GOTTMAN method.

Item Type: Article
Uncontrolled Keywords: single case diagnostics, intervervention analysis, General Linear Model, GLASS-WILLSON & GOTTMAN, BOX & TIAO, BOX & JENKINS transfer model, intervention model
Subjects: Generalities, computers, information > Computer science, internet
Philosophy and psychology > Psychology
Divisions: School of Computing Science, Business Administration, Economics and Law > Department of Computing Science
Date Deposited: 15 Sep 2015 11:36
Last Modified: 15 Sep 2015 11:36
URI: https://oops.uni-oldenburg.de/id/eprint/2494
URN: urn:nbn:de:gbv:715-oops-25755
DOI:
Nutzungslizenz:

Actions (login required)

View Item View Item

Document Downloads

More statistics for this item...