Exploring the effects of omitted variable bias in physics education research

Omitted variable bias occurs in most statistical models. Whenever a confounding variable that is correlated with both dependent and independent variables is omitted from a statistical model, estimated effects of included variables are likely to be biased due to omitted variables. This issue is parti...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Cole Walsh, Martin M. Stein, Ryan Tapping, Emily M. Smith, N. G. Holmes
Formato:	article
Lenguaje:	EN
Publicado:	American Physical Society 2021
Materias:	Special aspects of education LC8-6691 Physics QC1-999
Acceso en línea:	https://doaj.org/article/3553fa02740a49ce869323f6437a550c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:3553fa02740a49ce869323f6437a550c
record_format	dspace
spelling	oai:doaj.org-article:3553fa02740a49ce869323f6437a550c2021-12-02T13:20:50ZExploring the effects of omitted variable bias in physics education research10.1103/PhysRevPhysEducRes.17.0101192469-9896https://doaj.org/article/3553fa02740a49ce869323f6437a550c2021-03-01T00:00:00Zhttp://doi.org/10.1103/PhysRevPhysEducRes.17.010119http://doi.org/10.1103/PhysRevPhysEducRes.17.010119https://doaj.org/toc/2469-9896Omitted variable bias occurs in most statistical models. Whenever a confounding variable that is correlated with both dependent and independent variables is omitted from a statistical model, estimated effects of included variables are likely to be biased due to omitted variables. This issue is particularly problematic in physics education research where many research studies are quasiexperimental or observational in nature due to ethical and logistical limitations. In this paper, we illustrate the mechanisms behind omitted variable bias in explanatory modeling using authentic data and analytical solutions. We demonstrate that omitting confounding variables that are strongly correlated with included variables and have large effects on the dependent variable can significantly bias estimated effects for included variables. We also find that controlling for variables that are uncorrelated with other variables or have no effect on the dependent variable does not appreciably bias estimated effects and may or may not affect the precision of those estimates. These results suggest that removing from explanatory models variables that are not “statistically significant” can have unintended consequences on model and variable interpretations. Our results underscore the importance of carefully considering why or why not to include a variable in a model, informed by both data and theory.Cole WalshMartin M. SteinRyan TappingEmily M. SmithN. G. HolmesAmerican Physical SocietyarticleSpecial aspects of educationLC8-6691PhysicsQC1-999ENPhysical Review Physics Education Research, Vol 17, Iss 1, p 010119 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Special aspects of education LC8-6691 Physics QC1-999
spellingShingle	Special aspects of education LC8-6691 Physics QC1-999 Cole Walsh Martin M. Stein Ryan Tapping Emily M. Smith N. G. Holmes Exploring the effects of omitted variable bias in physics education research
description	Omitted variable bias occurs in most statistical models. Whenever a confounding variable that is correlated with both dependent and independent variables is omitted from a statistical model, estimated effects of included variables are likely to be biased due to omitted variables. This issue is particularly problematic in physics education research where many research studies are quasiexperimental or observational in nature due to ethical and logistical limitations. In this paper, we illustrate the mechanisms behind omitted variable bias in explanatory modeling using authentic data and analytical solutions. We demonstrate that omitting confounding variables that are strongly correlated with included variables and have large effects on the dependent variable can significantly bias estimated effects for included variables. We also find that controlling for variables that are uncorrelated with other variables or have no effect on the dependent variable does not appreciably bias estimated effects and may or may not affect the precision of those estimates. These results suggest that removing from explanatory models variables that are not “statistically significant” can have unintended consequences on model and variable interpretations. Our results underscore the importance of carefully considering why or why not to include a variable in a model, informed by both data and theory.
format	article
author	Cole Walsh Martin M. Stein Ryan Tapping Emily M. Smith N. G. Holmes
author_facet	Cole Walsh Martin M. Stein Ryan Tapping Emily M. Smith N. G. Holmes
author_sort	Cole Walsh
title	Exploring the effects of omitted variable bias in physics education research
title_short	Exploring the effects of omitted variable bias in physics education research
title_full	Exploring the effects of omitted variable bias in physics education research
title_fullStr	Exploring the effects of omitted variable bias in physics education research
title_full_unstemmed	Exploring the effects of omitted variable bias in physics education research
title_sort	exploring the effects of omitted variable bias in physics education research
publisher	American Physical Society
publishDate	2021
url	https://doaj.org/article/3553fa02740a49ce869323f6437a550c
work_keys_str_mv	AT colewalsh exploringtheeffectsofomittedvariablebiasinphysicseducationresearch AT martinmstein exploringtheeffectsofomittedvariablebiasinphysicseducationresearch AT ryantapping exploringtheeffectsofomittedvariablebiasinphysicseducationresearch AT emilymsmith exploringtheeffectsofomittedvariablebiasinphysicseducationresearch AT ngholmes exploringtheeffectsofomittedvariablebiasinphysicseducationresearch
_version_	1718393202453512192

Exploring the effects of omitted variable bias in physics education research

Ejemplares similares