Control variables and causal inference

A question of balance

Dados Bibliográficos

ID12594669
AutoresRichard York (0000-0002-0890-5803, University of Oregon, autor correspondente)
Ano2018
Volume21
Fascículo6
Páginas675-684
Data de publicação2018-05-02
Peer ReviewedSim
Open AccessNão
TipoARTICLE
PeriódicoInternational Journal of Social Research Methodology (JOURNAL)
Identificadores do periódicoISSN: 1364-5579 • E-ISSN: 1464-5300
EditoraTaylor & Francis (PUBLISHER • GB)
DOI10.1080/13645579.2018.1468730
OpenAlexW2799820001
IdiomaCA
Citações recebidas19
Referências citadas10

A common motivation for adding control variables to statistical models is to reduce the potential for spurious findings when analyzing non-experimental data and to thereby allow for more reliable causal inferences. However, as I show here, unless all potential confounding factors are included in an analysis (which is unlikely to be achievable with most real-world data-sets), adding control variables to a model in many circumstances can make estimated effects of the variable(s) of interest to the researcher on the dependent variable less accurate. Due to this fact, in some circumstances omitting control variables, even those that affect the dependent variable and are correlated with the variable(s) of interest, may allow for more accurate estimates of the effect(s) of the variable(s) of interest

Affect (linguistics · Causal inference · Causal model · Confounding · Control (management · Control variable · Econometrics · Inference · Instrumental variable · Omitted-variable bias · Spurious relationship · Statistical inference · Statistics · Variable (mathematics · Variables · Advanced Causal Inference Techniques · Bayesian Modeling and Causal Inference · Computer Science · Mathematics · Psychology · Statistical Methods and Inference · Artificial Intelligence