SSOAR Logo
    • Deutsch
    • English
  • Deutsch 
    • Deutsch
    • English
  • Einloggen
SSOAR ▼
  • Home
  • Über SSOAR
  • Leitlinien
  • Veröffentlichen auf SSOAR
  • Kooperieren mit SSOAR
    • Kooperationsmodelle
    • Ablieferungswege und Formate
    • Projekte
  • Kooperationspartner
    • Informationen zu Kooperationspartnern
  • Informationen
    • Möglichkeiten für den Grünen Weg
    • Vergabe von Nutzungslizenzen
    • Informationsmaterial zum Download
  • Betriebskonzept
Browsen und suchen Dokument hinzufügen OAI-PMH-Schnittstelle
JavaScript is disabled for your browser. Some features of this site may not work without it.

Download PDF
Volltext herunterladen

(378.1 KB)

Zitationshinweis

Bitte beziehen Sie sich beim Zitieren dieses Dokumentes immer auf folgenden Persistent Identifier (PID):
https://nbn-resolving.org/urn:nbn:de:0168-ssoar-101176-4

Export für Ihre Literaturverwaltung

Bibtex-Export
Endnote-Export

Statistiken anzeigen
Weiterempfehlen
  • Share via E-Mail E-Mail
  • Share via Facebook Facebook
  • Share via Bluesky Bluesky
  • Share via Reddit reddit
  • Share via Linkedin LinkedIn
  • Share via XING XING

Technical notes: Robust regression for at least ordinal outcomes

[Arbeitspapier]

Förster, Martin

Abstract

If the task is to conduct a regression analysis in order to examine the statistical associations between variables, many scientists think, of course, of OLS first. And it makes perfect sense: it’s probably the first regression technique you'll encounter - it's part of the basic repertoire of statistic... mehr

If the task is to conduct a regression analysis in order to examine the statistical associations between variables, many scientists think, of course, of OLS first. And it makes perfect sense: it’s probably the first regression technique you'll encounter - it's part of the basic repertoire of statistical analysis. Furthermore, OLS results are very informative. However, OLS needs a bunch of assumptions regarding the data at hand to be satisfied (Wooldridge 2010). If these assumptions are not met, results from OLS can be unstable, biased, or misleading. It is important to note that the assumptions of OLS are seldom fully met. Hence, to get stable results, we must apply robust regression techniques - that is, techniques which do not need some of the assumptions to be satisfied. I will discuss two scenarios where alternative regression techniques provide more robust results compared to OLS. Both are about handling certain characteristics of the dependent variable. First, we consider a scenario where the measurement level of the dependent variable is ordinal. In the second scenario, the outcome is metric, but its distribution is strongly skewed. Finally, it is outlined how robust inference statistics can be achieved for both scenarios. The techniques discussed are not only relevant but particularly so in the context of extended replications. Since replications aim to assess the stability of results, the application of alternative techniques aids in identifying methodological artifacts.... weniger

Klassifikation
Forschungsarten der Sozialforschung

Sprache Dokument
Englisch

Publikationsjahr
2025

Seitenangabe
8 S.

Schriftenreihe
Schriftenreihe für erweiterte Replikationen, Crowdsourcing und empirische Theorieüberprüfung, 4

DOI
https://doi.org/10.17605/OSF.IO/4HWXU

ISSN
2366-5041

Status
Veröffentlichungsversion; nicht begutachtet

Lizenz
Creative Commons - Namensnennung 4.0


GESIS LogoDFG LogoOpen Access Logo
Home  |  Impressum  |  Betriebskonzept  |  Datenschutzerklärung
© 2007 - 2025 Social Science Open Access Repository (SSOAR).
Based on DSpace, Copyright (c) 2002-2022, DuraSpace. All rights reserved.
 

 


GESIS LogoDFG LogoOpen Access Logo
Home  |  Impressum  |  Betriebskonzept  |  Datenschutzerklärung
© 2007 - 2025 Social Science Open Access Repository (SSOAR).
Based on DSpace, Copyright (c) 2002-2022, DuraSpace. All rights reserved.