Causal inference in statistics: An overview
Open accessThe DOI matches its Crossref record; 136 of the 136 references it cites were found in a registry or an index; 2 of the 72 sources its claims cite were checked for support.
Abstract
This review presents empirical researchers with recent advances in causal inference, and stresses the paradigmatic shifts that must be undertaken in moving from traditional statistical analysis to causal analysis of multivariate data. Special emphasis is placed on the assumptions that underly all causal inferences, the languages used in formulating those assumptions, the conditional nature of all causal and counterfactual claims, and the methods that have been developed for the assessment of such claims. These advances are illustrated using a general theory of causation based on the Structural Causal Model (SCM) described in Pearl (2000a), which subsumes and unifies other approaches to causation, and provides a coherent mathematical foundation for the analysis of causes and counterfactuals. In particular, the paper surveys the development of mathematical tools for inferring (from a combination of data and assumptions) answers to three types of causal queries: (1) queries about the effects of potential interventions, (also called “causal effects” or “policy evaluation”) (2) queries about probabilities of counterfactuals, (including assessment of “regret,” “attribution” or “causes of effects”) and (3) queries about direct and indirect effects (also known as “mediation”). Finally, the paper defines the formal and conceptual relationships between the structural and potential-outcome frameworks and presents tools for a symbiotic analysis that uses the strong features of both.
Overview
CoChat Summary
TL;DR
This survey argues that causal inference requires more than standard statistical association: it needs a formal language for causal assumptions and a mathematical machinery for combining those assumptions with data. Pearl presents the Structural Causal Model as a unifying framework that subsumes graphical, potential-outcome, and structural-equation approaches.
Key findings
- The paper distinguishes associational concepts (e.g., correlation, regression, conditional independence) from causal concepts (e.g., intervention, effect, confounding, attribution), arguing that causal claims cannot be defined from observed distributions alone.
- In Markovian structural models, the post-intervention distribution is identified by truncated factorization; for example, the paper shows that \(E(Y \mid do(x_0)) = E(Y \mid X=x_0)\) under the stated assumptions.
- The back-door criterion provides a graphical rule for choosing admissible adjustment sets; the paper states that if a set \(S\) blocks all back-door paths from \(X\) to \(Y\), then stratifying on \(S\) removes confounding bias.
- For noncompliance in clinical trials, the paper says treatment effects are generally nonidentifiable, but bounds can be derived; Balke and Pearl (1997) obtained closed-form bounds for 16 equivalent classes in the binary case, and the paper notes that these bounds can yield “significant and sometimes accurate” information.
Methods
- This is a conceptual review/survey, not an empirical study; it synthesizes advances in causal analysis “from the unifying perspective” of the Structural Causal Model.
- It develops the formalism using structural equations, causal diagrams, do-operator interventions, and counterfactual notation, and then maps these ideas to the potential-outcome framework.
- The analysis is mainly theoretical and illustrative, using examples such as mediation, instrumental variables, covariate adjustment/back-door identification, and noncompliance in clinical trials.
Limitations
- The paper states that counterfactual probabilities conditioned on outcomes “can in general [not] be identified from observational or even experimental studies” without additional conditions.
- It states that natural direct effects “cannot in general be identified or estimated, even with the help of ideal, controlled experiments” unless extra assumptions hold.
- Evident from the text: this is a survey and framework paper, so it does not report a new dataset, sample size, or empirical effect estimate.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 1 Not valid · 53 Not checked · 133 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Note also that the truncated factorization formula is not restricted to interventions on a single variable; it is applicable to simultaneous or sequential interventions such as those invoked in the analysis of time varying treatment with time varying confounders (Robins, 1986; Arjas and Parner, 2004).”
What the sources say
- Supportedcites Robins, 1986 source
Section F describes treatments assigned at successive times, with assignment depending on past measured covariates. The G-Computation Algorithm multiplies conditional probabilities along treatment and covariate-history paths and sums across those paths to calculate outcomes under a hypothetical treatment. This supports the cited point that the calculation extends beyond an intervention on one variable, although the paper calls it the G-computation algorithm rather than the truncated factorization formula. (LLM, conf=0.87)
- Not checked — no full text of the sourcecites Arjas and Parner, 2004 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The selection criterion can be applied systematically to diagrams of any size and shape, thus freeing analysts from judging whether ' X is conditionally ignorable given S ,' a formidable mental task required in the potential-response framework (Rosenbaum and Rubin, 1983).”
What the sources say
- Supportedcites Rosenbaum and Rubin, 1983 source
The citation supports the sentence’s reference to conditional ignorability in the potential-response framework. Section 1-2 defines strongly ignorable treatment assignment using the independence of treatment assignment from potential responses given covariates. The diagrammatic selection criterion is the citing authors’ own contrast, not a method attributed to Rosenbaum and Rubin. (LLM, conf=0.88)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A few but challenging algebraic steps would reveal that one can perform the summation over Z 2 to obtain”
- Also stated in: § 3.3. Coping with unmeasured confounders · plain
“which means that we need only adjust for Z 1 and Z 3 without ever measuring Z 2 .”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 3.3. Coping with unmeasured confounders · plain
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The set { Z 3 } , however, is not sufficient for adjustment because, as explained above, it does not block the path X ← W 1 ← Z 1 → Z 3 ← Z 2 → W 2 → Y .”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“This expression coincides with Robins' (1987) G -computation formula, which was derived from a more complicated set of (counterfactual) assumptions.”
What the sources say
- Not checked — no full text of the sourcecites Robins, 1987 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
114 more claims
Create a free account to see all 189 claims and to verify the 7 references still open.
- Probably not a substantial claim. 39
- It may state two separate claims as one. 39
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“which is known as the truncated factorization formula , see for example, Corollary 1 in Pearl (2009).”
From Discrimination-Free Insurance Pricing (2022) · cites this paper as Pearl (2009)
In the citing paper
““observational data” (such as rfMRI) is in general not a robust and safe tool for inferring causality, compared with “interventional” studies (such as task-fMRI experiments) [63].”
From Functional connectomics from resting-state fMRI (2013) · cites this paper as [63]
Citations
Cites
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241 · cited 3×
“The selection criterion can be applied systematically to diagrams of any size and shape, thus freeing analysts from judging whether “X is conditionally ignorable given S,” a formidable mental task required in the potential-response framework (Rosenbaum and Rubin, 1983).”
BackgroundMethod(2001) · RePEc: Research Papers in Economics · Cited by 27,577
Structural Equations with Latent Variables.
Clogg, C. C., & Bollen, K. A. (1991) · Contemporary Sociology · doi:10.2307/2072165 · Cited by 17,442
Structural Equations with Latent Variables
Bollen, K. A. (1989) · Wiley series in probability and statistics · doi:10.1002/9781118619179 · Cited by 17,277 · cited 3×
“…identification has received considerable attention in econometrics (Hurwicz, 1950; Marschak, 1950; Koopmans, 1953) and social science (Duncan, 1975; Bollen, 1989), usually in linear parametric settings, were it reduces to asking whether some model parameter, β, has a unique solution in terms of…”
Causality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653
Self-citationMediation in experimental and nonexperimental studies: New procedures and recommendations.
Shrout, P. E., & Bolger, N. (2002) · Psychological Methods · doi:10.1037/1082-989x.7.4.422 · Cited by 11,335 · cited 2×
“…) = ∑ z E(Y |x, z)[P (z|x′)− P (z|x)] (52) This expression deserves the label Mediation Formula, due to its pivotal role in mediation analysis (Imai et al., 2008), which has been a thorny issue in several sciences (Shrout and Bolger, 2002; MacKinnon et al., 2007; Mortensen et al., 2009).”
BackgroundEstimating causal effects of treatments in randomized and nonrandomized studies
Rubin, D. B. (1974) · Journal of Educational Psychology · doi:10.1037/h0037350 · Cited by 9,436 · cited 4×
“…equation models (SEM) used in economics and social science (Goldberger, 1973; Duncan, 1975), the potentialoutcome framework of Neyman (1923) and Rubin (1974), and the graphical models developed for probabilistic reasoning and causal analysis (Pearl, 1988; Lauritzen, 1996; Spirtes et al., 2000;…”
BackgroundMethodInfluentialMacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2007) · Annual Review of Psychology · doi:10.1146/annurev.psych.58.110405.085542 · Cited by 5,688 · cited 2×
“…) = ∑ z E(Y |x, z)[P (z|x′)− P (z|x)] (52) This expression deserves the label Mediation Formula, due to its pivotal role in mediation analysis (Imai et al., 2008), which has been a thorny issue in several sciences (Shrout and Bolger, 2002; MacKinnon et al., 2007; Mortensen et al., 2009).”
BackgroundDiscrete Multivariate Analysis: Theory and Practice.
Beniger, J. R., Bishop, Y. M. M., Feinberg, S. E., Holland, P. W., Light, R. J., & Mosteller, F. (1975) · Contemporary Sociology · doi:10.2307/2063625 · Cited by 5,108
Recent Developments in the Econometrics of Program Evaluation
Imbens, G. W., & Wooldridge, J. M. (2009) · Journal of Economic Literature · doi:10.1257/jel.47.1.5 · Cited by 4,926
Research works printed but not linked: 11
- Glymour, M. and Greenland, S. (2008). Causal diagrams. In, Modern Epidemiology (K. Rothman, S. Greenland and T. Lash, eds.), 3rd ed. Lippincott Williams & Wilkins, Philadelphia, PA, 183–209.
- Hurwicz, L. (1950). Generalization of the concept of identification. In, Statistical Inference in Dynamic Economic Models (T. Koopmans, ed.). Cowles Commission, Monograph 10, Wiley, New York, 245–257.
- Koopmans, T. (1953). Identification problems in econometric model construction. In, Studies in Econometric Method (W. Hood and T. Koopmans, eds.). Wiley, New York, 27–48.
- Marschak, J. (1950). Statistical inference in economics. In, Statistical Inference in Dynamic Economic Models (T. Koopmans, ed.). Wiley, New York, 1–50. Cowles Commission for Research in Economics, Monograph 10.
- Pearl, J. (1988)., Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, San Mateo, CA.
- Pearl, J. (2001). Direct and indirect effects. In, Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence. Morgan Kaufmann, San Francisco, CA, 411–420.
- Robins, J. (1989). The analysis of randomized and non-randomized aids treatment trials using a new approach to causal inference in longitudinal studies. In, Health Service Research Methodology: A Focus on AIDS (L. Sechrest, H. Freeman and A. Mulley, eds.). NCHSR, U.S. Public Health Service, Washington, D.C., 113–159.
- Wermuth, N. (1992). On block-recursive regression equations., Brazilian Journal of Probability and Statistics (with discussion) 6 1–56.
- Cole, P. (1997). Causality in epidemiology, health policy, and law., Journal of Marketing Research 27 10279–10285.
- Pearson, K., Lee, A. and Bramley-Moore, L. (1899). Genetic (reproductive) selection: Inheritance of fertility in man., Philosophical Transactions of the Royal Society A 73 534–539.
- Robertson, D. (1997). The common sense of cause in fact., Texas Law Review 75 1765–1800.
Cited by
Items 801–900 of 2,354
Causal chambers as a real-world physical testbed for AI methodology
Gamella, J. L., Peters, J., & Bühlmann, P. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-024-00964-x · Cited by 96
BAIT: Large Language Model Backdoor Scanning by Inverting Attack Target
Shen, G., Cheng, S., Zhang, Z., Tao, G., Zhang, K., Guo, H., Yan, L., Jin, X., An, S., Ma, S., et al. (2025) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp61157.2025.00103 · Cited by 63
Accelerated brain ageing during the COVID-19 pandemic
Mohammadi‐Nejad, A., Craig, M., Cox, E. F., Chen, X., Jenkins, G. R., Francis, S., Sotiropoulos, S. N., & Auer, D. P. (2025) · Nature Communications · doi:10.1038/s41467-025-61033-4 · Cited by 22
“We answer in detail each of their comments below.”
Supporting shared hypothesis testing in the biomedical domain
Agibetov, A., Jiménez-Ruiz, E., Ondrésik, M., Solimando, A., Banerjee, I., Guerrini, G., Catalano, C. E., Oliveira, J. M., Patanè, G., Reis, R. L., et al. (2018) · Journal of Biomedical Semantics · doi:10.1186/s13326-018-0177-x · Cited by 11
“Our work is perhaps similar in spirit to that of Pearl et al. [45, 46], where the authors advocate for a paradigmatic shift that must be undertaken in moving from traditional statistical analysis to causal analysis of multivariate data [45, 46].”
ResultTransporting stochastic direct and indirect effects to new populations
Rudolph, K. E., Levy, J., & van der Laan, M. J. (2020) · Biometrics · doi:10.1111/biom.13274 · Cited by 11
“The SEM first generates a random draw of a vector 𝑈 of unknown measurements (Pearl, 2009), where 𝑈 = (𝑈𝑆,𝑈𝑊 ,𝑈𝐴, 𝑈𝑍,𝑈𝑀,𝑈𝑌 ) ∼ 𝑃𝑈 .”
The Five-Factor Model of Personality: Consensus and Controversy
McCrae, R. R. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108264822.013 · Cited by 37
Vanhatalo, J. P., Hartmann, M., & Veneranta, L. (2020) · Jukuri (Natural Resources Institute Finland (Luke)) · Cited by 11
Toward practical causal epidemiology
Cox, L. A. (2021) · Global Epidemiology · doi:10.1016/j.gloepi.2021.100065 · Cited by 12
Wong, M. C., & Dowd, M. (2021) · Ecosphere · doi:10.1002/ecs2.3794 · Cited by 12
“1 and using structural equation modeling or path analysis, might also be a viable alternative (Pearl 2009).”
MethodEstimating causal effects with optimization-based methods: A review and empirical comparison
Cousineau, M., Verter, V., Murphy, S. A., & Pineau, J. (2022) · European Journal of Operational Research · doi:10.1016/j.ejor.2022.01.046 · Cited by 18
“These patients have a 60% response rate and patients with low social support ( X = 0 ) have a 40% response rate.”
Redondo, I., Pérez‐Rodríguez, L., Monclús, R., et al. (2022) · Oecologia · doi:10.1007/s00442-022-05180-1 · Cited by 11
“Conditional independence claims were evaluated by directed separation tests to determine the goodness of fit of each model (Pearl 2009).”
MethodSpatio-temporal dynamics of bluefin tuna (Thunnus thynnus) in US waters of the northwest Atlantic
Hansell, A. C., Becker, S. L., Cadrin, S. X., Lauretta, M. V., Walter, J. F., & Kerr, L. A. (2022) · Fisheries Research · doi:10.1016/j.fishres.2022.106460 · Cited by 12
Examining User Heterogeneity in Digital Experiments
Somanchi, S., Abbasi, A., Kelley, K., Dobolyi, D. G., & Yuan, T. T. (2023) · ACM Transactions on Information Systems · doi:10.1145/3578931 · Cited by 13
“These techniques are not directly applicable for inferring user HTEs, because the latter necessitates considering diferences in how users react in treatment versus control settings [22, 59, 67].”
BackgroundWhat is an exposure-response curve?
Cox, L. A. (2023) · Global Epidemiology · doi:10.1016/j.gloepi.2023.100114 · Cited by 15
Selecting robust features for machine-learning applications using multidata causal discovery
Ganesh, S. S., Beucler, T., Tam, F. I., Gomez, M. S., Runge, J., & Gerhardus, A. (2023) · Environmental Data Science · doi:10.1017/eds.2023.21 · Cited by 23
The NIST data science initiative
Dorr, B. J., Greenberg, C. S., Fontana, P., Przybocki, M. A., Bras, M. L., Ploehn, C., Aulov, O., Michel, M., Golden, E. J., & Chang, W. L. (2015) · IEEE International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa.2015.7344805 · Cited by 11
“, inferring causality [24], [29]? Knowledge assimilation and reasoning refers to”
BackgroundDebt and communications technology diffusion: Retrospective evidence
Majumdar, S. K. (2015) · Research Policy · doi:10.1016/j.respol.2015.10.004 · Cited by 11
Is Public Service Motivation Changeable? Integrative Modeling with Goal-Setting Theory
Han, Y. (2016) · International Journal of Public Administration · doi:10.1080/01900692.2016.1262871 · Cited by 17
“Four criteria are required for ensuring the causal relationship: the cause and effect must be (1) distinct entities that (2) co-vary with (3) the cause preceding the effect in time while (4) alternative explanations for co-variation and temporal precedence are ruled out (Pearl, 2009).”
BackgroundNeugebauer, R. S., Schmittdiel, J. A., Adams, A. S., Grant, R. W., & van der Laan, M. J. (2017) · Journal of Causal Inference · doi:10.1515/jci-2016-0015 · Cited by 17
“These two counterfactual equalities are developed based on the nonparametric structural equation model (NPSEM) framework [27, 28, 29, 30] and applicable to the evaluation of the joint effect of an adaptive treatment strategy and either a static or general stochastic clinical monitoring intervention under the NDE assum…”
Wang, Y., Fan, C., Zheng, Y., & Li, C. (2017) · Nucleic Acids Research · doi:10.1093/nar/gkx086 · Cited by 12
Chaudhari, A. M., Sha, Z., & Panchal, J. H. (2018) · Journal of Mechanical Design · doi:10.1115/1.4040166 · Cited by 13
A Normative Theory for Causal Inference and Bayes Factor Computation in Neural Circuits
Zhang, W., Wu, S., Doiron, B., & Lee, T. S. (2019) · neural information processing systems · Cited by 15
OrganITE: Optimal transplant donor organ offering using an individual treatment effect
Berrevoets, J., Jordon, J., Bica, I., Gimson, A. E., & van der Schaar, M. (2020) · Lirias · Cited by 49
Statistical Perspective on Functional and Causal Neural Connectomics: A Comparative Study
Biswas, R., & Shlizerman, E. (2022) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2022.817962 · Cited by 18
“We proceed to delineate causation from statistical associations by surveying existing approaches and describe relation of each approach to above statistical properties [58, 63, 64].”
Hatami, R. (2019) · Water Air & Soil Pollution · doi:10.1007/s11270-019-4150-9 · Cited by 12
“Causal modelling analyses (e.g. SEM) use a combination of statistical analysis and the theory of causation summarised in a causal diagram, while including the potential confounding variables and correcting the confounding bias (Greenland et al. 1999; Pearl 2009; Hoyle 2011).”
BackgroundMethodInfluentialNandy, P., Maathuis, M. H., & Richardson, T. S. (2017) · The Annals of Statistics · doi:10.1214/16-aos1462 · Cited by 11
“Equation (3) immediately follows from Pearl’s backdoor criterion [25] if all error variables are Gaussian [20].”
BackgroundMethodInfluentialKincaid, D. L., Babalola, S. O., & Figueroa, M. E. (2014) · JAIDS Journal of Acquired Immune Deficiency Syndromes · doi:10.1097/qai.0000000000000242 · Cited by 11
Some models and methods for the analysis of observational data
Ferreira, J. A. (2015) · Statistics Surveys · doi:10.1214/15-ss110 · Cited by 9
Mediation Analysis: Inferring Causal Processes in Marketing from Experiments
Pieters, R. G. M. (2017) · International series in quantitative marketing · doi:10.1007/978-3-319-53469-5_8 · Cited by 16
Some node ordering methods for the K2 algorithm
Aghdam, R., Tabar, V. R., & Pezeshk, H. (2018) · Computational Intelligence · doi:10.1111/coin.12182 · Cited by 11
“14 , 33 In this work, we are fascinated with the structure learning problem and using the score-based searching algorithms for mentioned data sets.”
MethodOrder-independent constraint-based causal structure learningPreprint
Colombo, D., & Maathuis, M. H. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1211.3295 · Cited by 30
“A DAG can be interpreted causally in the following way (Pearl, 2000, 2009; Spirtes et al., 2000):”
BackgroundQuantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden ConfoundingPreprint
Jesson, A., Mindermann, S., Gal, Y., & Shalit, U. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.04850 · Cited by 70
Structural Counterfactuals: A Brief Introduction
Pearl, J. (2013) · Defense Technical Information Center (DTIC) · doi:10.21236/ada580574 · Cited by 11
Self-citationDynamical Causal Graph Neural Network for EEG Emotion Recognition
Xiao, Y., Zheng, W., & Zhao, G. (2025) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2025.3582740 · Cited by 20
“A causal topology graph G = { X, E } is defined an asymmetric, directed graph, where each directed edge denotes causal connectivity [51], [52]: where To find the optimal causal topology graph G , an effective approach is to use Structural Causal Models (SCMs) to analyze the joint distribution P X…”
The issue of statistical power for overall model fit in evaluating structural equation models
Hermida, R., Luchman, J. N., Nicolaides, V., & Wilcox, C. (2015) · RePEc: Research Papers in Economics · Cited by 10
Detecting dynamical causality by intersection cardinal concavity
Tao, P., Wang, Q., Shi, J., Hao, X., Liu, X., Min, B., Zhang, Y., Li, C., Cui, H., & Chen, L. (2023) · Fundamental Research · doi:10.1016/j.fmre.2023.01.007 · Cited by 13
Inappropriate nighttime light reduces living comfort
Li, C., & Managi, S. (2023) · Environmental Pollution · doi:10.1016/j.envpol.2023.122173 · Cited by 12
Barkhordarian, A. (2024) · Environmental Research Letters · doi:10.1088/1748-9326/ad715a · Cited by 11
“Essentially, the FAR metric quantifies the extent to which GHG forcing is a necessary cause of the extreme events [3, 34].”
Weak reciprocal relationships between productivity and plant biodiversity in managed grasslands
Andraczek, K., Dee, L. E., Weigelt, A., Hinderling, J., Prati, D., Provost, G. L., Manning, P., Wirth, C., & van der Plas, F. (2024) · Journal of Ecology · doi:10.1111/1365-2745.14400 · Cited by 15
“…longitudinal data including repeated samples of 150 plots from 58 sites across the growing season at two time points (i.e. spring and summer) and over 2 years, informed by directed acyclic graphs to visualize causal relationships (Figure S4; Arif & MacNeil, 2022, Laubach et al., 2021, Pearl, 2009).”
De-Confounded Data-Free Knowledge Distillation for Handling Distribution Shifts
Wang, Y., Yang, D., Chen, Z., Liu, Y., Liul, S., Zhang, W., Zhang, L., & Qi, L. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01199 · Cited by 19
“To tackle this issue, we introduce the backdoor adjustment [35] to construct causal intervention P ( S | do ( X )) and block the back-door path between X and S via Z .”
BackgroundMethodInfluentialNarrative Creativity: An Introduction to How and Why
Fletcher, A., & Benveniste, M. (2025) · Cambridge University Press eBooks · doi:10.1017/9781009614801 · Cited by 17
A framework for assessing the trustworthiness of scientific research findings
Nosek, B. A., Allison, D. B., Jamieson, K. H., Mcnutt, M. K., Nielsen, A. B., & Wolf, S. M. (2026) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2536736123 · Cited by 18
Xiang, Y., & Tarasawa, B. (2015) · Journal of School Choice · doi:10.1080/15582159.2015.1028862 · Cited by 9
“Propensity score is an analytical technique that reduces selection bias by matching groups based on key covariates (Austin, 2011; Oakes, & Johnson, 2006; Pearl, 2009; Rosenbaum & Rubin, 1983; Stuart, 2008).”
BackgroundTowards a Framework for Observational Causality from Time Series: When Shannon Meets Turing
Sigtermans, D. (2020) · Entropy · doi:10.3390/e22040426 · Cited by 12
“It closely relates our approach to Structural Causal Models [16].”
MethodExplanation Framework for Intrusion Detection
Burkart, N., Franz, M., & Huber, M. F. (2020) · Technologien für die intelligente Automation · doi:10.1007/978-3-662-62746-4_9 · Cited by 12
“Depending on the requirements of the application, other approaches like those proposed by Pearl et al. [16] ought to be pursued in parallel.”
MethodCadiou, S., Basagaña, X., González, J. R., Lepeule, J., Vrijheid, M., Siroux, V., & Slama, R. (2021) · Environment International · doi:10.1016/j.envint.2021.106509 · Cited by 14
“We used the theory of structural causal modelling (Pearl, 2009, 1995) to identify in which causal structures a causal association could be expected to be identified using the oriented Meet-in-the-Middle design in the simpler situation of three unidimensional variables (e.g. one exposure, one CpG,…”
BackgroundMethodKrämer, M. D., van Scheppingen, M. A., Chopik, W. J., & Richter, D. (2022) · European Journal of Personality · doi:10.1177/08902070221118443 · Cited by 15
“However, structural causal modeling requires advanced knowledge of the causal structures underlying the involved variables (Pearl, 2009).”
BackgroundNiemantsverdriet, M. S. A., de Hond, T. A. P., Hoefer, I. E., van Solinge, W. W., Bellomo, D., Oosterheert, J. J., Kaasjager, K. A. H., & Haitjema, S. (2022) · BMC Emergency Medicine · doi:10.1186/s12873-022-00764-9 · Cited by 11
“Future studies should explore causal inference to evaluate the causal relationship to determine the dependence of the found variables with sepsis [40].”
BackgroundEnhancing Event Causality Identification with Counterfactual Reasoning
Mu, F., & Li, W. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-short.83 · Cited by 13
“Relying heavily on those ambiguous signals may make an ECI model learn the spurious correlation (Pearl, 2009) To solve this issue, it is intuitively to explicitly estimate the influence of context keywords and event pairs in training, so that we can mitigate those biases in inference.”
BackgroundLinking physical violence to women’s mobility in Chile
Contreras, H., Candia, C., Troncoso, R., Ferres, L., Bravo, L., Lepri, B., & Rodríguez‐Sickert, C. (2023) · EPJ Data Science · doi:10.1140/epjds/s13688-023-00430-5 · Cited by 12
“This criterion suggests that the confounding factor Z must block or intercept every pathway between X and Y that comprises an arrow directing towards X [24, 43, 44, 56].”
BackgroundAn empirical study of counterfactual visualization to support visual causal inference
Wang, A. Z., Borland, D., & Gotz, D. (2024) · Information Visualization · doi:10.1177/14738716241229437 · Cited by 18
“4,5 One approach that has gained popularity is the use of counterfactual reasoning , 6,7 a fundamental tool in statistical causal inference that uses hypothetical scenarios to investigate causal relationships.”
BackgroundInfluentialNexus: Correlation Discovery over Collections of Spatio-Temporal Tabular Data
Gong, Y., Galhotra, S., & Fernandez, R. C. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3654957 · Cited by 18
Network inference using steady-state data and Goldbeter–Koshland kinetics
Oates, C. J., Hennessy, B. T., Lu, Y., Mills, G. B., & Mukherjee, S. (2013) · Bioinformatics · doi:10.1093/bioinformatics/btt057 · Cited by 10
A Statistical Basis for Fuzzy Engineering Economics
Nguyen, H. T., Sriboonchitta, S., & Wu, B. (2015) · International Journal of Fuzzy Systems · doi:10.1007/s40815-015-0010-y · Cited by 10
Gómez-Gallego, M., Gómez-García, J., & Ato-Lozano, E. (2015) · Aging & Mental Health · doi:10.1080/13607863.2015.1093603 · Cited by 12
“Following Pearl (2009), identifying a causal mechanism in this approach is also formulated as a decomposition of the averaged total effect (ATE) into indirect (average causal mediation effect, ACME) and direct (average direct effect, ADE) effects.”
MethodA new data science research program: evaluation, metrology, standards, and community outreach
Dorr, B. J., Greenberg, C. S., Fontana, P., Przybocki, M. A., Bras, M. L., Ploehn, C., Aulov, O., Michel, M., Golden, E. J., & Chang, W. L. (2016) · International Journal of Data Science and Analytics · doi:10.1007/s41060-016-0016-z · Cited by 10
“, inferring causality [97,102]? Knowledge assimilation and reasoning refers to understanding new data in terms of information available in an already existing dataset and applying the necessary processing to reflect the expert’s view of the domain.”
BackgroundNeuroimaging with functional near infrared spectroscopy: From formation to interpretation
Herrera-Vega, J., Treviño-Palacios, C. G., & Orihuela‐Espina, F. (2017) · Infrared Physics & Technology · doi:10.1016/j.infrared.2017.06.011 · Cited by 11
Cox, L. A. (2020) · Critical Reviews in Toxicology · doi:10.1080/10408444.2020.1787329 · Cited by 9
“The concept “held fixed” is a causal, rather than a statistical, construct (Pearl 2009).”
BackgroundMethodInfluentialNovak, A. R., Impellizzeri, F. M., Garvey, C., et al. (2021) · Journal of Sports Sciences · doi:10.1080/02640414.2021.1943169 · Cited by 10
“priate causal inferences approaches and experimental studies (Pearl, 2009).”
BackgroundCETransformer: Casual Effect Estimation via Transformer Based Representation Learning
Guo, Z., Zheng, S., Liu, Z., Yan, K., & Zhu, Z. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-88013-2_43 · Cited by 18
Vike, N. L., Bari, S., Stetsiv, K., Walter, A. E., Newman, S. D., Kawata, K., Bazarian, J. J., Martinovich, Z., Nauman, E. A., Talavage, T. M., et al. (2021) · iScience · doi:10.1016/j.isci.2021.103483 · Cited by 12
“We followed a Pearl-MacKinnon viewpoint (Lockhart et al., 2011; Mackinnon, 2008; MacKinnon et al., 2007, 2012; Pearl, 2009, 2012, 2014), although the MacArthur viewpoint (Cole and Maxwell, 2003; Gollob and Reichardt, 1987; Kraemer et al., 2002, 2008; Maxwell and Cole, 2007) is also important, and…”
MethodDevelopment of Personality across the Life Span
Atherton, O. E., Donnellan, M. B., & Robins, R. W. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108264822.017 · Cited by 19
Personalized Public Policy Analysis in Social Sciences Using Causal-Graphical Normalizing Flows
Balgi, S., Peña, J. M., & Daoud, A. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i11.21437 · Cited by 29
“…from the benefits of ACE estimates with small bias and variance compared to standard statistical methods, c-GNF provides the unique benefit of counterfactual inference using ‘The First Law of Causal Inference’ (Pearl 2009a,b; Pearl and Mackenzie 2018), which is virtually non-existent in IPW/RWR/GCOM.”
InfluentialStatistical perspective on functional and causal neural connectomics: The Time-Aware PC algorithm
Biswas, R., & Shlizerman, E. (2022) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1010653 · Cited by 15
“, τ} as a completed partially directed acyclic graph (CPDAG), defined as the graph union of DAGs that satisfy DMP with respect to the subsample, and denoted by G t [36].”
BackgroundJensen, J. A., Wakefield, L. T., & Walkup, B. R. (2022) · Journal of Sport Management · doi:10.1123/jsm.2021-0292 · Cited by 10
Spieker, A. J., Delaney, J. A., & McClelland, R. L. (2015) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.3876 · Cited by 10
Kim, M. H., Banerjee, S., Zhao, Y., Wang, F., Zhang, Y., Zhu, Y., Deferio, J. J., Evans, L., Park, S. M., & Pathak, J. (2018) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2018.09.016 · Cited by 9
“Furthermore, causal inference needs to be evaluated based on potential causal pathways, which requires hypothesis-specific adjustment methods for potential confounders [65,66].”
BackgroundNetwork inference using steady-state data and Goldbeter–koshland kinetics
Oates, C. J., Hennessy, B. T., Lu, Y., Mills, G. B., & Mukherjee, S. (2012) · Bioinformatics · doi:10.1093/bioinformatics/bts459 · Cited by 17
West, S. G., & Koch, T. (2014) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2014.856700 · Cited by 15
“SEM researchers wishing an initial accessible introduction to many of the key ideas presented by Pearl would do well to consult Bollen and Pearl (2013), followed by more challenging, but accessible papers by Pearl (2009, 2012).”
BackgroundBeware of the Simulated DAG! Causal Discovery Benchmarks May Be Easy To GamePreprint
Reisach, A. G., Seiler, C., & Weichwald, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.13647 · Cited by 209
“Causal models not only describe the observational joint distribution of variables but also formalize predictions under interventions and counterfactuals [Spirtes et al., 2000, Pearl, 2009, Peters et al., 2017].”
BackgroundInference With Cross-Lagged Effects - Problems in TimePreprint
Driver, C. (2022) · doi:10.31219/osf.io/xdf72 · Cited by 10
Hashemi, M., Vattikonda, A. N., Jha, J., Ší́p, V., Woodman, M., Bartoloméi, F., & Jirsa, V. (2022) · medRxiv · doi:10.1101/2022.06.02.22275860 · Cited by 10
“This approach allows us to infer the probability of past and future events, as well as the dynamics of beliefs under changing conditions for generating novel hypotheses where new experiments are prohibitively difficult or impossible to perform (Pearl, 2009b).”
BackgroundFang, J. S., Wu, W., Chen, Y., Zhang, N., Wu, Y., & Zhang, M. (2025) · BMC Geriatrics · doi:10.1186/s12877-025-06273-1 · Cited by 13
“Grounded in foundational causal theory, Bayesian Network analysis offers a well-established methodology not for proving deterministic causation, but for modeling directional pathways as systems of probabilistic dependency [27].”
Chaos, coherence, and turbulence
Jiménez, J. (2025) · Physical Review Fluids · doi:10.1103/lkvl-dlcn · Cited by 14
Priorities and Diversities in Language and Thought
Camp, E. (2020) · Philosophical studies series · doi:10.1007/978-3-030-47641-0_3 · Cited by 9
Betchkal, D., Beeco, J. A., Anderson, S., Peterson, B. A., & Joyce, D. (2023) · Journal of Environmental Management · doi:10.1016/j.jenvman.2023.119201 · Cited by 12
Fire traps in the wet subtropics: New perspectives from Hong Kong
Chan, A. H. Y., & Coomes, D. A. (2024) · Journal of Applied Ecology · doi:10.1111/1365-2664.14575 · Cited by 11
Methodological Aspects of Pupillometry
Laeng, B., & Mathôt, S. (2024) · Springer eBooks · doi:10.1007/978-3-031-54896-3_13 · Cited by 9
The application of machine learning in clinical microbiology and infectious diseases
Cheng, X., Zhao, L., Ye, C.-S., Xu, K.-C., & Xu, K. (2025) · Frontiers in Cellular and Infection Microbiology · doi:10.3389/fcimb.2025.1545646 · Cited by 20
“…data leakage include proper data splitting (Kaufman et al., 2012), pipelines for preprocessing (Fabian et al., 2011), time-aware validation (Bergmeir and Ben ı ́ tez, 2012), causal feature analysis (Pearl, 2009), potential leak characteristics identi fi cation (Lundberg and Lee, 2017) and so on.”
Cai, S., Peng, H., Liu, R., & Chen, P. (2025) · Communications Physics · doi:10.1038/s42005-025-02170-6 · Cited by 14
A Survey on Bias in Machine Learning Research
Mikołajczyk, A., & Grochowski, M. (2025) · IEEE Access · doi:10.1109/access.2025.3649998 · Cited by 12
“In statistics, a confounder (also known as a confounding variable, confounding factor, extraneous determinant, or lurk-ing variable) is a variable that influences both the dependent variable (i.e., disease) and independent variable (the studied factor), causing a spurious association [90, 114].”
Orman, L. V. (2013) · IEEE Technology and Society Magazine · doi:10.1109/mts.2013.2259334 · Cited by 9
“Without such insight, calculating the risk to children from divorce would be too general to be useful, or even fundamentally incorrect [25].”
Multiple Test In Silico Weight-of-Evidence for Toxicological Endpoints
Aldenberg, T., & Jaworska, J. S. (2010) · Royal Society of Chemistry eBooks · doi:10.1039/9781849732093-00558 · Cited by 9
Counterfactual Representation Augmentation for Cross-Domain Sentiment Analysis
Wang, K., & Wan, X. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2022.3158843 · Cited by 15
“…Product of Experts (PoE) [72] method [73], [74], etc. Causal Inference is the process of identifying the relationship between causes and its effect (i.e., causality), which has been extensively studied in statistics [51], [75], social sciences [76], medical sciences [77], and econometrics [78].”
BackgroundMethodOrlova, E. V. (2023) · Mathematics · doi:10.3390/math11040863 · Cited by 10
“…in the network variables, except for the consequence vertices of 5 x We will use the concept of intervention ( do operator) introduced by J. Pearl [1] in order to obtain a joint probability , is completely contained in its direct vertices. distribution of network variables when We will use the…”
BackgroundMethodInfluentialFluid hunter motivation in Central Africa: Effects on behaviour, bushmeat and income
Froese, G., Mbélé, A. E., Beirne, C., Bazza, B., N’noh, S. D., Ebeba, J., Edzidzie, J., Koto, S. E., Metandou, J. L., Mossindji, C., et al. (2023) · People and Nature · doi:10.1002/pan3.10502 · Cited by 11
“…deemed important from the literature and our knowledge of the study site, and statistically sensible to include together based on causal theory (Pearl, 2009)— as well as the potentially important factors of weather (rainy or not during the hunt), hunting period (night, day or both) and wet…”
BackgroundEthics framework for predictive clinical AI model updating
Pruski, M. (2023) · Ethics and Information Technology · doi:10.1007/s10676-023-09721-x · Cited by 14
“…conclusions when using this approach, and as such it is not always possible to use this approach in a valid way (for more detail on causal inference the reader can consult the following texts: Pearl, 2009 ; Pearl et al., 2016 ; Hernán and Robins n.d.; Kleinberg & Hripcsak, 2011 ; Scheines, 1997 ).”
MethodCasgevy: Innovative Medicinal Products Require Innovative Approaches to Regulatory Assessment
Ghazaly, E. A., & Johnston, J. D. (2024) · Pharmaceutics · doi:10.3390/pharmaceutics16070906 · Cited by 11
“Graphical models display what one knows about the world; counterfactual argument articulates what one wants to know; and structural equations knit (i) and (ii) together [13].”
BackgroundA causal reasoning approach to DSA situational awareness and decision-making
Martin, T., & Chang, K. (2013) · Cited by 9
Vock, D. M., & Boehm, L. (2018) · Journal of Quantitative Analysis in Sports · doi:10.1515/jqas-2016-0029 · Cited by 14
Axiomatizing the Bayesian Paradigm in Parallel Small Worlds
French, S. (2020) · Operations Research · doi:10.1287/opre.2019.1896 · Cited by 9
“…of subjective probabilities conditioned on counterfactuals and causal decision decision theory (Gibbard and Harper 1988, Meek and Glymour 1994, Pearl 2009), but note that the ‘what if’ or ‘counterfactual’ conditioning implicit in scenario-focused thinking is much more informal than those much…”
BackgroundZhang, L., Rodrigues, L. O., Narain, N. R., & Akmaev, V. R. (2019) · Journal of Computational Biology · doi:10.1089/cmb.2019.0210 · Cited by 13
“As Pearl pointed out, causal and statistical inferences have fundamental differences since they focus on causation and association, respectively (Pearl, 2009a).”
BackgroundMethodA Wiener Causality Defined by Divergence
Chen, J., Feng, J., & Lu, W. (2020) · Neural Processing Letters · doi:10.1007/s11063-019-10187-6 · Cited by 10
Model Compression for Domain Adaptation through Causal Effect Estimation
Rotman, G., Feder, A., & Reichart, R. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00431 · Cited by 9
“In causal inference we typically care about estimating the effect that an intervention on a variable of interest would have on an outcome (Pearl, 2009).”
BackgroundInteractive Granular Computing Model for Intelligent Systems
Dutta, S., & Skowron, A. (2021) · IFIP advances in information and communication technology · doi:10.1007/978-3-030-74826-5_4 · Cited by 9
Evolutionary Personality Psychology
Lewis, D. M. G., Al-Shawaf, L., & Buss, D. M. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108264822.022 · Cited by 24
Time and Causality across the Sciences
(2019) · Cambridge University Press eBooks · doi:10.1017/9781108592703 · Cited by 12
Tackling Multipath and Biased Training Data for IMU-Assisted BLE Proximity Detection
He, T., Tan, J., Zhuo, W., et al. (2022) · doi:10.1109/infocom48880.2022.9796716 · Cited by 9
“Note that, the association absence between any two nodes assumes their prior independence; for instance, carriage state and distance are independent being RSSI unknown, but they are correlated when the RSSI is given (according to d-separation [33]).”
BackgroundUnderstanding Queries by Conditional Instances
Gilad, A., Miao, Z., Roy, S., & Yang, J. (2022) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3514221.3517898 · Cited by 9
“Another [40] quantifies the responsibility of each input tuples to the query answer, an idea inspired by causality [45].”
On the Relationship between ANOVA Main Effects and Average Treatment Effects
Graefe, L., Hahn, S., & Mayer, A. (2022) · Multivariate Behavioral Research · doi:10.1080/00273171.2022.2068122 · Cited by 16
“In contrast to the somewhat obscured definitions of main effects in the ANOVA literature, the causal inference literature (e.g., Neyman, 1923/1990; Pearl, 2009; Rubin, 2005; Steyer et al., 2000) provides us with a clear and unambiguous definition of an average treatment effect.”
Result
Metadata
- Authors
- Pearl, J.
- Affiliations
- University of California, Los Angeles (US)
- Corresponding author
- Pearl, Judea
- Venue
- Statistics Surveys
- Publisher
- Institute of Mathematical Statistics
- Published
- 2009-01-01
- Volume
- 3
- Issue
- none
- Pages
- 96-146
- ISSN
- 1935-7516 (ISSN-L)
- OpenAlex
- W2143117649
- Cited by
- 2,336
- References
- 135
- Citations by year
- 2026: 184 · 2025: 273 · 2024: 271 · 2023: 245 · 2022: 215 · 2021: 208 · 2020: 243 · 2019: 140 · 2018: 96 · 2017: 77 · 2016: 82 · 2015: 82 · 2014: 63 · 2013: 65 · 2012: 46
- FWCI
- 25.47
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Statistics Surveys (publishedVersion, cc-by)
- Project Euclid (Cornell University) (submittedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- Department of Statistics, UCLA
- Funders
- Office of Naval Research
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Philosophy of Biology and Scientific Explanation
- Keywords
- causal inference, structural causal models, counterfactual reasoning, causal assumptions, causal effects, policy evaluation, intervention effects, regret, causal attribution, direct and indirect effects, mediation analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.