Campbell's and Rubin's Perspectives on Causal Inference

Stephen West, Felix Thoemmes

Research output: Contribution to journalArticle

73 Citations (Scopus)

Abstract

Donald Campbell's approach to causal inference (D. T. Campbell, 1957; W. R. Shadish, T. D. Cook, & D. T. Campbell, 2002) is widely used in psychology and education, whereas Donald Rubin's causal model (P. W. Holland, 1986; D. B. Rubin, 1974, 2005) is widely used in economics, statistics, medicine, and public health. Campbell's approach focuses on the identification of threats to validity and the inclusion of design features that may prevent those threats from occurring or render them implausible. Rubin's approach focuses on the precise specification of both the possible outcomes for each participant and assumptions that are mathematically sufficient to estimate the causal effect. In this article, the authors compare the perspectives provided by the 2 approaches on randomized experiments, broken randomized experiments in which treatment nonadherence or attrition occurs, and observational studies in which participants are assigned to treatments on an unknown basis. The authors highlight dimensions on which the 2 approaches have different emphases, including the roles of constructs versus operations, threats to validity versus assumptions, methods of addressing threats to internal validity and violations of assumptions, direction versus magnitude of causal effects, role of measurement, and causal generalization. The authors conclude that investigators can benefit from drawing on the strengths of both approaches in designing research.

Original languageEnglish (US)
Pages (from-to)18-37
Number of pages20
JournalPsychological Methods
Volume15
Issue number1
DOIs
StatePublished - Mar 2010

Fingerprint

Observational Studies
Public Health
Economics
Research Personnel
Medicine
Psychology
Education
Research
Direction compound
Causal Inference
Threat
Causal
Randomized Experiments

Keywords

  • causal inference
  • observational study
  • quasi-experiment
  • randomized experiment
  • research design

ASJC Scopus subject areas

  • Psychology (miscellaneous)
  • History and Philosophy of Science

Cite this

Campbell's and Rubin's Perspectives on Causal Inference. / West, Stephen; Thoemmes, Felix.

In: Psychological Methods, Vol. 15, No. 1, 03.2010, p. 18-37.

Research output: Contribution to journalArticle

West, Stephen ; Thoemmes, Felix. / Campbell's and Rubin's Perspectives on Causal Inference. In: Psychological Methods. 2010 ; Vol. 15, No. 1. pp. 18-37.
@article{69238f6a4b0642f9a46c57c6832153f3,
title = "Campbell's and Rubin's Perspectives on Causal Inference",
abstract = "Donald Campbell's approach to causal inference (D. T. Campbell, 1957; W. R. Shadish, T. D. Cook, & D. T. Campbell, 2002) is widely used in psychology and education, whereas Donald Rubin's causal model (P. W. Holland, 1986; D. B. Rubin, 1974, 2005) is widely used in economics, statistics, medicine, and public health. Campbell's approach focuses on the identification of threats to validity and the inclusion of design features that may prevent those threats from occurring or render them implausible. Rubin's approach focuses on the precise specification of both the possible outcomes for each participant and assumptions that are mathematically sufficient to estimate the causal effect. In this article, the authors compare the perspectives provided by the 2 approaches on randomized experiments, broken randomized experiments in which treatment nonadherence or attrition occurs, and observational studies in which participants are assigned to treatments on an unknown basis. The authors highlight dimensions on which the 2 approaches have different emphases, including the roles of constructs versus operations, threats to validity versus assumptions, methods of addressing threats to internal validity and violations of assumptions, direction versus magnitude of causal effects, role of measurement, and causal generalization. The authors conclude that investigators can benefit from drawing on the strengths of both approaches in designing research.",
keywords = "causal inference, observational study, quasi-experiment, randomized experiment, research design",
author = "Stephen West and Felix Thoemmes",
year = "2010",
month = "3",
doi = "10.1037/a0015917",
language = "English (US)",
volume = "15",
pages = "18--37",
journal = "Psychological Methods",
issn = "1082-989X",
publisher = "American Psychological Association Inc.",
number = "1",

}

TY - JOUR

T1 - Campbell's and Rubin's Perspectives on Causal Inference

AU - West, Stephen

AU - Thoemmes, Felix

PY - 2010/3

Y1 - 2010/3

N2 - Donald Campbell's approach to causal inference (D. T. Campbell, 1957; W. R. Shadish, T. D. Cook, & D. T. Campbell, 2002) is widely used in psychology and education, whereas Donald Rubin's causal model (P. W. Holland, 1986; D. B. Rubin, 1974, 2005) is widely used in economics, statistics, medicine, and public health. Campbell's approach focuses on the identification of threats to validity and the inclusion of design features that may prevent those threats from occurring or render them implausible. Rubin's approach focuses on the precise specification of both the possible outcomes for each participant and assumptions that are mathematically sufficient to estimate the causal effect. In this article, the authors compare the perspectives provided by the 2 approaches on randomized experiments, broken randomized experiments in which treatment nonadherence or attrition occurs, and observational studies in which participants are assigned to treatments on an unknown basis. The authors highlight dimensions on which the 2 approaches have different emphases, including the roles of constructs versus operations, threats to validity versus assumptions, methods of addressing threats to internal validity and violations of assumptions, direction versus magnitude of causal effects, role of measurement, and causal generalization. The authors conclude that investigators can benefit from drawing on the strengths of both approaches in designing research.

AB - Donald Campbell's approach to causal inference (D. T. Campbell, 1957; W. R. Shadish, T. D. Cook, & D. T. Campbell, 2002) is widely used in psychology and education, whereas Donald Rubin's causal model (P. W. Holland, 1986; D. B. Rubin, 1974, 2005) is widely used in economics, statistics, medicine, and public health. Campbell's approach focuses on the identification of threats to validity and the inclusion of design features that may prevent those threats from occurring or render them implausible. Rubin's approach focuses on the precise specification of both the possible outcomes for each participant and assumptions that are mathematically sufficient to estimate the causal effect. In this article, the authors compare the perspectives provided by the 2 approaches on randomized experiments, broken randomized experiments in which treatment nonadherence or attrition occurs, and observational studies in which participants are assigned to treatments on an unknown basis. The authors highlight dimensions on which the 2 approaches have different emphases, including the roles of constructs versus operations, threats to validity versus assumptions, methods of addressing threats to internal validity and violations of assumptions, direction versus magnitude of causal effects, role of measurement, and causal generalization. The authors conclude that investigators can benefit from drawing on the strengths of both approaches in designing research.

KW - causal inference

KW - observational study

KW - quasi-experiment

KW - randomized experiment

KW - research design

UR - http://www.scopus.com/inward/record.url?scp=77949620091&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949620091&partnerID=8YFLogxK

U2 - 10.1037/a0015917

DO - 10.1037/a0015917

M3 - Article

C2 - 20230100

AN - SCOPUS:77949620091

VL - 15

SP - 18

EP - 37

JO - Psychological Methods

JF - Psychological Methods

SN - 1082-989X

IS - 1

ER -