Estimating causal effects of treatments in randomized and nonrandomized studies
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Presents a discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation. The objective was to specify the benefits of randomization in estimating causal effects of treatments. It is concluded that randomization should be employed whenever possible but that the use of carefully controlled nonrandomized data to estimate causal effects is a reasonable and necessary procedure in many cases.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“That work argues models may memorize variable names but lack knowledge of the underlying observational distributions ( P ( X,Y ) ) required to identify traps like confounding (Rubin, 1974).”
From Diagnosing and Mitigating Sycophancy and Skepticism in LLM Causal Judgment (2026) · cites this paper as Rubin, 1974
In the citing paper
“Specifically, the assumption that a subunit’s own output or outcomes are not affected by any other unit’s implementation assignment (what is called SUTVA (99)) is invalid, as interactions and synergistic effects underlie many of our implementation strategies.”
From An Overview of Research and Evaluation Designs for Dissemination and Implementation (2017) · cites this paper as [99]
Citations
Cites
Statistical Methods for Research Workers
(1945) · American Journal of Clinical Pathology · doi:10.1093/ajcp/15.11.552b · Cited by 10,880
(2001) · Empirical Direction in Design and Analysis · doi:10.4324/9781410600004-24 · Cited by 6,808
SCHEFFE, H. (1960) · Soil Science · doi:10.1097/00010694-196006000-00016 · Cited by 5,516
Testing Statistical Hypotheses
(2021) · Wiley Series in Probability and Statistics · doi:10.1002/9781119243830.ch12 · Cited by 5,106
Design and Analysis of Experiments
(2007) · Modern Engineering Statistics · doi:10.1002/9780470187548.ch12 · Cited by 1,424
Controlling Bias in Observational Studies: A Review
Cochran, W. G. (2006) · Matched Sampling for Causal Effects · doi:10.1017/cbo9780511810725.005 · Cited by 958 · cited 1×
“Recent work on methods of reducing bias in nonrandomized studies is summarized in Cochran and Rubin (1974).”
MethodThe University Group Diabetes Program. A further statistical analysis of the mortality findings
Cornfield, J. (1971) · JAMA · doi:10.1001/jama.217.12.1676 · Cited by 119
University Group Diabetes Program
Smelo, L. S. (1971) · JAMA: The Journal of the American Medical Association · doi:10.1001/jama.1971.03180260057027 · Cited by 115
Design and Analysis of Industrial Experiments.
E., R. T., & Davies, O. L. (1954) · OR · doi:10.2307/3007149 · Cited by 111
Statistics: A Guide to the Unknown.
Pine, V. R., Tanur, J. M., Mosteller, F. C., Kruskal, W., Link, R. F., Pieters, R. S., & Rising, G. R. (1974) · Contemporary Sociology A Journal of Reviews · doi:10.2307/2062882 · Cited by 89
Cited by
Items 601–700 of 1,074
Identification and Inference in Nonlinear Difference-in-Differences ModelsPreprint
Athey, S. C., & Imbens, G. W. (2002) · SSRN Electronic Journal · doi:10.2139/ssrn.311920 · Cited by 143
Rosenbaum, P. R. (2007) · Biometrics · doi:10.1111/j.1541-0420.2006.00717.x · Cited by 168
“The j th subject in set i would exhibit response rTij if this subject received treatment, Zij = 1, and would exhibit response rCij if this subject received control, Zij =0 , so the response actually observed from this subject is Rij = Zij rTij + (1 − Zij )rCij , and the effect of the treatment on this subject is rTij…”
Covariate selection for the nonparametric estimation of an average treatment effect
De Luna, X., Waernbaum, I., & Richardson, T. S. (2011) · Biometrika · doi:10.1093/biomet/asr041 · Cited by 142
Campbell’s and Rubin’s perspectives on causal inference.
West, S. G., & Thoemmes, F. (2010) · Psychological Methods · doi:10.1037/a0015917 · Cited by 141
“…randomized experiment, quasi-experiment, observational study, research design, causal inference In this article, we provide an introduction to Donald Campbell’s (Campbell, 1957; Shadish, Cook, & Campbell, 2002) and Donald Rubin’s (Holland, 1986; Rubin, 1974, 2005) perspectives on causal inference.”
BackgroundMethodInfluentialVandor, P., & Franke, N. (2016) · Journal of Business Venturing · doi:10.1016/j.jbusvent.2016.03.003 · Cited by 183
“Nonetheless, we included them as statistical controls in all subsequent analyses (Rubin, 1974).”
MethodImplications of Small Samples for Generalization: Adjustments and Rules of Thumb
Tipton, E., Hallberg, K., Hedges, L. V., & Chan, W. (2017) · Evaluation Review · doi:10.1177/0193841x16655665 · Cited by 140
“This is the case when the experimental sample is randomly drawn from the population, when Pr(Z 1⁄4 1) 1⁄4 1, or if the sample selection and treatment effect heterogeneity are independent (Imai, King, & Stuart, 2008; Rubin, 1974).”
BackgroundCausal Inference for Recommender Systems
Wang, Z., Liang, D., Charlin, L., & Blei, D. M. (2020) · ACM Conference on Recommender Systems (RecSys) · doi:10.1145/3383313.3412225 · Cited by 151
“) The pair (yui (0),yui (1)) is the potential outcomes notation in the Rubin causal model [10, 20, 21], where watching a movie is a “treatment” and a user’s rating of the movie is an “outcome.”
BackgroundGerfin, M., Lechner, M., & Steiger, H. (2004) · Labour Economics · doi:10.1016/j.labeco.2004.04.002 · Cited by 149
Journal De La Société Française De Statistique
Caussinus, H., & de Falguerolles, A. (2005) · Wiley eBooks · doi:10.1002/0471667196.ess7017 · Cited by 141
Why ask Why? Forward Causal Inference and Reverse Causal Questions
Gelman, A., & Imbens, G. (2013) · National Bureau of Economic Research · doi:10.3386/w19614 · Cited by 141
Causation and Explanation in Social Science
Brady, H. E. (2009) · The Oxford Handbook of Political Methodology · doi:10.1093/oxfordhb/9780199286546.003.0010 · Cited by 141
Ansolabehere, S. D., & Iyengar, S. (1994) · Political Communication · doi:10.1080/10584609.1994.9963048 · Cited by 141
Estimating causal effects for multivalued treatments: a comparison of approaches
Linden, A., Uysal, S. D., Ryan, A. M., & Adams, J. (2015) · Statistics in Medicine · doi:10.1002/sim.6768 · Cited by 165
Saunders, J., Parast, L., Babey, S. H., & Miles, J. V. (2017) · PLOS ONE · doi:10.1371/journal.pone.0179494 · Cited by 141
“He proposed a causal model to eliminate group differences on the back end, mimicking the conditions and covariate balance of a randomized controlled trial [45].”
MethodLöschel, A., Lutz, B. J., & Managi, S. (2019) · Resource and Energy Economics · doi:10.1016/j.reseneeco.2018.03.001 · Cited by 140
Chapter 59 Evaluating Anti-Poverty Programs
Ravallion, M. D. (2007) · Handbook of development economics · doi:10.1016/s1573-4471(07)04059-4 · Cited by 245
Statistical Models for Causation: What Inferential Leverage Do They Provide?
Freedman, D. A. (2006) · Evaluation Review · doi:10.1177/0193841x06293771 · Cited by 134
Dropping out of High School in the United States: An Observational Study
Rosenbaum, P. R. (1986) · Journal of Educational Statistics · doi:10.2307/1165073 · Cited by 140
War and Wages: The Strength of Instrumental Variables and Their Sensitivity to Unobserved Biases
Small, D. S., & Rosenbaum, P. R. (2008) · Journal of the American Statistical Association · doi:10.1198/016214507000001247 · Cited by 139
“Each subject ij has two potential responses: rT ij if encouraged, Zij = 1, or rCij if not encouraged, Zij = 0, so the effect of encouragement, rT ij − rCij , is not observed (see Neyman 1923; Rubin 1974).”
BackgroundAn Introduction to Causal Inference
SOBEL, M. E. (1996) · Sociological Methods & Research · doi:10.1177/0049124196024003004 · Cited by 138
Causal inference in randomized experiments with mediational processes.
Jo, B. (2008) · Psychological Methods · doi:10.1037/a0014207 · Cited by 139
“Treatment assignment is random (Holland, 1986; Rubin, 1974, 1978, 1980).”
BackgroundTargeted maximum likelihood estimation for a binary treatment: A tutorial
Luque-Fernández, M. Á., Schomaker, M., Rachet, B., & Schnitzer, M. E. (2018) · Statistics in Medicine · doi:10.1002/sim.7628 · Cited by 184
“We observe Y(1) only for those in the exposure group and Y(0) only for those in the unexposed group.(4,5) Common causal estimands of interest are the average treatment effect (ATE), defined as E[Y(1) − Y(0)], and the marginal odds ratio (MOR), defined as E[Y(1)] × {1 − E[Y(0)]}/{E(1 − E[Y(1)]) × E[Y(0)]}.”
BackgroundLocal Projections for Applied Economics
Jordà, Ò. (2023) · Annual Review of Economics · doi:10.1146/annurev-economics-082222-065846 · Cited by 209
The Formation of Voting Habits
Dinas, E. (2012) · Journal of Elections, Public Opinion & Parties · doi:10.1080/17457289.2012.718280 · Cited by 138
“Following the Neyman-Rubin causal framework, individuals are defined according to their treatment and treatment assignment states by using a potential outcomes terminology (Neyman 1923 1990; Rubin 1974).”
MethodCausal Inference in Statistics
Pearl, J., Glymour, M., & Jewell, N. P. (2016) · Cited by 137
Flexible treatment strategies in chronic disease: clinical and research implications
Lavori, P. W., Dawson, R., & Rush, A. J. (2000) · Biological Psychiatry · doi:10.1016/s0006-3223(00)00946-x · Cited by 137
Mediation analysis via potential outcomes models
Albert, J. M. (2008) · Statistics in Medicine · doi:10.1002/sim.3016 · Cited by 137
“While SEM has had limited use among biostatisticians, the approach of causal modeling based on the potential outcome [13] has gained some recent attention and interest—though not universal approval.”
BackgroundKang, J., & Schafer, J. L. (2008) · doi:10.48550/arxiv.0804.2958 · Cited by 138
“The notion of potential outcomes was introduced by Neyman (1923) for randomized experiments and by Rubin (1974a) for nonrandomized studies.”
Turning Unemployment into Self‐Employment: Effectiveness of Two Start‐Up Programmes*
Baumgartner, H. J., & Caliendo, M. (2008) · Oxford Bulletin of Economics and Statistics · doi:10.1111/j.1468-0084.2008.00505.x · Cited by 185
“Identifying average treatment effects We base our analysis on the potential outcome framework, also known as the Roy (1951)–Rubin (1974) model.”
MethodWu, T. N., Yang, S., & Tan, J. (2020) · Resources Policy · doi:10.1016/j.resourpol.2020.101715 · Cited by 137
Systemic Consequences of Incumbency Advantage in U.S. House Elections
King, G., & Gelman, A. (1991) · American Journal of Political Science · doi:10.2307/2111440 · Cited by 137
Neighborhood Effect Heterogeneity by Family Income and Developmental Period
Wodtke, G. T., Elwert, F., & Harding, D. J. (2016) · American Journal of Sociology · doi:10.1086/684137 · Cited by 171
Augmented Inverse Probability Weighting and the Double Robustness Property
Kurz, C. (2021) · Medical Decision Making · doi:10.1177/0272989x211027181 · Cited by 183
Uniform consistency in causal inference
Robins, J. M. (2003) · Biometrika · doi:10.1093/biomet/90.3.491 · Cited by 136
Graphical Models for Probabilistic and Causal Reasoning
Pearl, J. (1998) · Quantified Representation of Uncertainty and Imprecision · doi:10.1007/978-94-017-1735-9_12 · Cited by 126
Can Reputation Discipline the Gig Economy? Experimental Evidence from an Online Labor Market
Benson, A. M., Sojourner, A., & Umyarov, A. (2019) · Management Science · doi:10.1287/mnsc.2019.3303 · Cited by 185
Politics with the People: Building a Directly Representative Democracy
Neblo, M. A., Esterling, K., & Lazer, D. M. J. (2018) · Cambridge University Press eBooks · doi:10.1017/9781316338179 · Cited by 207
Gelman, A., & Pardoe, I. (2007) · Sociological Methodology · doi:10.1111/j.1467-9531.2007.00181.x · Cited by 162
“The comparison (1) corresponds to an expected causal effect under a counterfactual assumption (Neyman 1923; Rubin 1974, 1990), if it makes sense to consider the inputs u causally.”
BackgroundHow Generalizable Is Your Experiment? An Index for Comparing Experimental Samples and Populations
Tipton, E. (2014) · Journal of Educational and Behavioral Statistics · doi:10.3102/1076998614558486 · Cited by 134
“Interestingly, this rule of thumb corresponds roughly results from Rubin (2001), which suggest that balance is achieved when |d| .”
BackgroundMethodResultInfluentialFerguson, K. K., Chen, Y.-H., VanderWeele, T. J., McElrath, T. F., Meeker, J. D., & Mukherjee, B. (2017) · Environmental Health Perspectives · doi:10.1289/ehp282 · Cited by 133
“Application of a counterfactual framework (Rubin 1974) has further provided a strong theoretical basis for causal inference from mediation analysis by precisely defining the necessary assumptions.”
Dahabreh, I. J., Hayward, R., & Kent, D. M. (2016) · International Journal of Epidemiology · doi:10.1093/ije/dyw125 · Cited by 132
Nonparametric Causal Effects Based on Incremental Propensity Score Interventions
Kennedy, E. H. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2017.1422737 · Cited by 249
“In this article, we use potential outcomes (Rubin 1974), and so let YaT denote the outcome that would have been observed had the treatment sequence AT = aT been received.”
MethodDark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense
Zhu, Y., Gao, T., Fan, L., Huang, S., Edmonds, M., Liu, H., Gao, F., Zhang, C., Qi, S., Wu, Y. N., et al. (2020) · Engineering · doi:10.1016/j.eng.2020.01.011 · Cited by 132
“Rubin [146] laid the foundation for causal analysis in sta-”
BackgroundCarrubba, C. J., & Gabel, M. J. (2014) · Cited by 134
The employment effects of job-creation schemes in Germany: A microeconometric evaluation
Caliendo, M., Hujer, R., & Thomsen, S. L. (2008) · Advances in econometrics · doi:10.1016/s0731-9053(07)00013-8 · Cited by 159
Wang, L., & Tchetgen Tchetgen, E. (2018) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/rssb.12262 · Cited by 132
“e right panel gives a Single World Intervention Graph (Richardson and Robins, 2013) with a bi-directed arrow. An alternative definition of IV is based on the potential outcome framework (Neyman, 1923; Rubin, 1974), under which we assume D(z), the potential exposure if the instrument would take value zto be well-defined…”
BackgroundLearning representations for image-based profiling of perturbations
Moshkov, N., Bornholdt, M., Benoit, S., Smith, M., McQuin, C., Goodman, A., Senft, R. A., Han, Y., Babadi, M., Horvath, P., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-45999-1 · Cited by 131
“We use a causal framework to reason about the challenges of learning representations of cell morphology (e.g., confounding factors), which naturally fi ts in the context of perturbation experiments 29,30 , and serves as a tool to optimize the work fl ow and yield better performance (Fig.”
MethodFrom quality to innovation: Evidence from two French Employer Surveys
Peković, S., & Galia, F. (2009) · Technovation · doi:10.1016/j.technovation.2009.08.002 · Cited by 155
“Rubin (1974) defines the average treatment effect as the difference between what would be the health of an individual if he was treated and what it would be if he was not, in our case it presents the difference of impact of firm that receive quality and what the impact would be if firm did not…”
BackgroundAmplification of Sensitivity Analysis in Matched Observational Studies
Rosenbaum, P. R., & Silber, J. H. (2009) · Journal of the American Statistical Association · doi:10.1198/jasa.2009.tm08470 · Cited by 133
“Following Neyman (1923) and Rubin (1974)’s notation for treatment effects, each subject ij has two potential responses, (rTij, rCij), and this subject would exhibit response rCij if assigned to control, Zij = 0, or response rTij if assigned to treatment, Zij = 1, so the response actually observed…”
MethodCausal Inference in Retrospective Studies
Holland, P. W., & Rubin, D. B. (1988) · Evaluation Review · doi:10.1177/0193841x8801200301 · Cited by 132
Self-citationBrodaty, T., Crépon, B., & Fougère, D. (2001) · ZEW economic studies · doi:10.1007/978-3-642-57615-7_5 · Cited by 158
“This framework generalizes the modelling that Rubin (1974, 1977) introduced for the case of a unique treatment.”
MethodManaging Churn to Maximize Profits
Lemmens, A., & Гупта, С. (2020) · Marketing Science · doi:10.1287/mksc.2020.1229 · Cited by 185
Genetic markers as instrumental variables
von Hinke, S., Davey Smith, G., Lawlor, D. A., Propper, C., & Windmeijer, F. (2016) · Journal of Health Economics · doi:10.1016/j.jhealeco.2015.10.007 · Cited by 130
Legal Status and the Criminal Activity of Immigrants
Mastrobuoni, G., & Pinotti, P. (2015) · American Economic Journal Applied Economics · doi:10.1257/app.20140039 · Cited by 231
The Production of Knowledge: Enhancing Progress in Social Science
(2020) · Cambridge University Press eBooks · doi:10.1017/9781108762519 · Cited by 244
Programme Evaluation with Multiple Treatments
Frölich, M. (2004) · Journal of Economic Surveys · doi:10.1111/j.0950-0804.2004.00001.x · Cited by 171
“Such individual treatment effects (Rubin, 1974) can never be ascertained since only one of the potential outcomes Y i ;Y 1 i ; .”
BackgroundMethodInfluentialAbadie, A., Angrist, J. D., & Imbens, G. W. (1999) · SSRN Electronic Journal · doi:10.2139/ssrn.195733 · Cited by 132
“As in Rubin (1974, 1977) and our earlier work on instrumental variables estimation of causal effects, we define the causal effects of interest using potential outcomes and potential treatment status.”
Does outsourcing affect agricultural productivity of farmer households? Evidence from China
Deng, X., Xu, D., Zeng, M., & Qi, Y. (2020) · China Agricultural Economic Review · doi:10.1108/caer-12-2018-0236 · Cited by 131
Dee, L. E., Ferraro, P. J., Severen, C. N., Kimmel, K. A., Borer, E. T., Byrnes, J. E. K., Clark, A. T., Hautier, Y., Hector, A., Raynaud, X., et al. (2023) · Nature Communications · doi:10.1038/s41467-023-37194-5 · Cited by 204
A Mathematical Theory of Evidence turns 40
Shafer, G. R. (2016) · International Journal of Approximate Reasoning · doi:10.1016/j.ijar.2016.07.009 · Cited by 148
“In uential calls for explicitly causal probabilistic language came from the philosopher Pat Suppes in 1970 [133] and the statistician Don Rubin in 1974 [79].”
BackgroundEmployed 40 Hours or Not Employed 39: Lessons from the 1982 Mandatory Reduction of the Workweek
Crépon, B., & Kramarz, F. (2002) · Journal of Political Economy · doi:10.1086/342807 · Cited by 172
Shadish, W. R. (2010) · Psychological Methods · doi:10.1037/a0015916 · Cited by 146
“The exception, ironically, was the very first presentation of his work on causal inference, which appeared in the Journal of Educational Psychology (Rubin, 1974).”
BackgroundInfluentialThompson, A. J., & Pickett, J. T. (2019) · Journal of Quantitative Criminology · doi:10.1007/s10940-019-09436-7 · Cited by 139
Abel, U. R., & Koch, A. (1999) · Journal of Clinical Epidemiology · doi:10.1016/s0895-4356(99)00041-4 · Cited by 133
Kot, F. C. (2014) · Research in Higher Education · doi:10.1007/s11162-013-9325-4 · Cited by 172
Wong, V. C., Steiner, P. M., & Cook, T. D. (2012) · Journal of Educational and Behavioral Statistics · doi:10.3102/1076998611432172 · Cited by 158
“The first is to use potential outcomes notation (Holland, 1986; Rubin, 1974) to define the causal estimand tMRD for an MRDD with two assignment variables (M and R) and cutoffs.”
BackgroundKam, C. D., & Trussler, M. J. (2016) · Political Behavior · doi:10.1007/s11109-016-9379-z · Cited by 134
“…from scrambled sentence tasks to audio-visual clips—under the assumption of random assignment, the Neyman–Rubin causal framework (Neyman 1923; Rubin 1974) suggests that a simple difference of means test will indicate the magnitude and significance of the average treatment effect— the…”
MethodCausal Inference Methods for Combining Randomized Trials and Observational Studies: A Review
Colnet, B., Mayer, I., Chen, G., Dieng, A., Li, R., Varoquaux, G. P., Vert, J., Josse, J., & Yang, S. (2024) · Statistical Science · doi:10.1214/23-sts889 · Cited by 308
“outcomes (PO) framework (Neyman, 1923; Rubin, 1974), associated with the work by Donald Rubin, and the work on structural causal models (SCM) using directed acyclic graphs (DAGs), much of it associated with work by Judea Pearl (Pearl, 1995) and his collaborators.”
BackgroundMethodBayesian Inference for Causal Effects
Rubin, D. B. (2005) · Handbook of statistics · doi:10.1016/s0169-7161(05)25001-0 · Cited by 133
Self-citationGould, E. D., Lavy, V., & Paserman, M. D. (2004) · The Quarterly Journal of Economics · doi:10.1162/0033553041382193 · Cited by 127
“13 The counterfactual framework was pioneered by Rubin (1974) and since adopted by many in both statistics and econometrics, including Rosenbaum and Rubin (1983), Heckman (1992), Imbens and Angrist (1994), Angrist, 11 Conditional independence implies conditional mean independence: ( ) ( ) ( ) (…”
Analysis of Large Health Surveys: Accounting for the Sampling Design
Korn, E. L., & Graubard, B. I. (1995) · Journal of the Royal Statistical Society. Series A (Statistics in Society) · doi:10.2307/2983292 · Cited by 127
“) Such variables have been called concomitant observations (Cox (1958), pages 48-49), prior variables (Rubin, 1974) and attributes (Holland, 1986); we have called them exogenous variables (Korn and Graubard, 1988).”
BackgroundAdjusting for differential base rates: Lord's Paradox again.
Wainer, H. (1991) · Psychological Bulletin · doi:10.1037//0033-2909.109.1.147 · Cited by 127
“model ( Rubin, 1974,1977,1978,1980; Holland, 1986a, I986b)”
Experimental Evidence on Distributional Effects of Head Start
Bitler, M., Hoynes, H. W., & Domina, T. (2014) · National Bureau of Economic Research · doi:10.3386/w20434 · Cited by 183
Matching on the Estimated Propensity Score
Abadie, A., & Imbens, G. W. (2009) · National Bureau of Economic Research · doi:10.3386/w15301 · Cited by 192
Petras, H., Masyn, K. E., Buckley, J. A., Ialongo, N. S., & Kellam, S. (2011) · Journal of Educational Psychology · doi:10.1037/a0021545 · Cited by 126
Wan, E. Y. F., Fung, C. S. C., Jiao, F. F., Yu, E. Y. T., Chin, W. Y., Fong, D. Y. T., Wong, C. K. H., Chan, A. K. C., Chan, K. H. Y., Kwok, R. L. P., et al. (2018) · Diabetes Care · doi:10.2337/dc17-0426 · Cited by 126
To Adjust or Not to Adjust? Sensitivity Analysis of M -Bias and Butterfly-Bias
Ding, P., & Miratrix, L. W. (2015) · Journal of Causal Inference · doi:10.1515/jci-2013-0021 · Cited by 126
“The above can also be expressed in the potential outcomes framework [24, 25].”
BackgroundCausal Inference: A Missing Data Perspective
Ding, P., & Li, F. (2018) · Statistical Science · doi:10.1214/18-sts645 · Cited by 195
“A mainstream statistical framework for causal inference is the potential outcomes framework, under which each unit has a set of potential outcomes corresponding to all possible treatment levels (Neyman, 1923, 1935; Rubin, 1974, 1977, 1978).”
BackgroundElements of External Validity: Framework, Design, and Analysis
Egami, N., & Hartman, E. (2022) · American Political Science Review · doi:10.1017/s0003055422000880 · Cited by 188
“First, we formalize all four dimensions of external validity within the potential outcomes framework (Neyman 1923; Rubin 1974).”
Volunteering and Income – The Fallacy of the Good Samaritan?
Hackl, F., Halla, M., & Pruckner, G. J. (2007) · Kyklos · doi:10.1111/j.1467-6435.2007.00360.x · Cited by 185
Emmert‐Streib, F., Glazko, G. V., Altay, G., & Simoes, R. D. M. (2012) · Frontiers in Genetics · doi:10.3389/fgene.2012.00008 · Cited by 142
“Further, we content ourselves with approaches for observational data (Rubin, 1974; Rosenbaum, 2002) neglecting methods that utilize interventional or perturbational data.”
MethodA natural experiment of social network formation and dynamics
Phan, T. Q., & Airoldi, E. M. (2015) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1404770112 · Cited by 124
Doubly robust nonparametric inference on the average treatment effect
Benkeser, D. C., Carone, M., van der Laan, M. J., & Gilbert, P. B. (2017) · Biometrika · doi:10.1093/biomet/asx053 · Cited by 167
Propensity Score Matching and Subclassification in Observational Studies with Multi-Level Treatments
Yang, S., Imbens, G. W., Cui, Z. L., Faries, D. E., & KADZIOLA, Z. A. (2016) · Biometrics · doi:10.1111/biom.12505 · Cited by 154
“Following Rubin (1974) and RR83 we use the potential outcome set up, generalized to the case with more than two, unordered, levels for the treatment in Imbens (2000), Lechner (2001), Imai and Van Dyk (2004), and Cattaneo (2010).”
Heinsman, D. T., & Shadish, W. R. (1996) · Psychological Methods · doi:10.1037/1082-989x.1.2.154 · Cited by 125
“, Cook & Campbell, 1979) emphasize that randomized experiments may yield biased estimates if differential attrition occurs; similarly, matching on reliable covariates may aid nonrandomized experiments (Holland, 1986; Rubin, 1974).”
BackgroundWunsch, C., & Lechner, M. (2008) · Kyklos · doi:10.1111/j.1467-6435.2008.00396.x · Cited by 147
“…comparison state and the participation decision, then -conditional on these factors -participation and the outcomes, which the participants would have obtained in the comparison state, are independent, and the effects of interest are identified (Rubin, 1974; Imbens, 2000; Lechner, 2001, 2002a,b).”
Hujer, R., Caliendo, M., & Thomsen, S. L. (2004) · Research in Economics · doi:10.1016/j.rie.2004.07.001 · Cited by 170
“The standard model in the microeconometric evaluation literature is the potential outcome approach or Roy(1951)Rubin(1974)-model.”
MethodCausal analysis in control–impact ecological studies with observational data
Larsen, A., Meng, K. C., & Kendall, B. E. (2019) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13190 · Cited by 152
“It wasn't until the work of Rubin (1974) that the potential outcomes framework was considered for observational data settings.”
BackgroundEvaluating (weighted) dynamic treatment effects by double machine learning
Bodory, H., Huber, M., & Lafférs, L. (2022) · Econometrics Journal · doi:10.1093/ectj/utac018 · Cited by 170
Probabilistic Evaluation of Counterfactual Queries
Balke, A., & Pearl, J. (2022) · Probabilistic and Causal Inference · doi:10.1145/3501714.3501733 · Cited by 124
“We will refer to this domain-minimal variable as a response-function variable. rb is closely related to the potential response variables in Rubin's model of counterfactuals (Rubin 1974), which was introduced to facilitate causal inference in statistical analysis (Balke & Pearl 1993).”
Carlin, R. E., Love, G. J., & Zechmeister, E. J. (2013) · Political Research Quarterly · doi:10.1177/1065912913495592 · Cited by 166
“First by empirically approximating the Rubin Causal Model (Rubin 1974), it maps the logic of controlled experiments onto observational data and, thus, increases confidence that the observed relationship is causal.”
MethodPopulation heterogeneity and causal inference
Xie, Y. (2013) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1303102110 · Cited by 138
“E Y 1 −Y 0jpðXÞ =E Y 1jD= 1; pðXÞ −E Y 0 D= 0; p X : [12] 3.”
Dibba, L., Fialor, S. C., Diagne, A., & Nimoh, F. (2012) · Food Security · doi:10.1007/s12571-012-0180-5 · Cited by 125
“One class of methods identifies causal effects under the assumption of ‘ignorability’ or conditional independence (Rubin 1974; Rosenbaum and Rubin 1983).”
BackgroundHarnessing naturally randomized transcription to infer regulatory relationships among genes
Chen, L. S., Emmert-Streib, F., & Storey, J. D. (2007) · Genome Biology · doi:10.1186/gb-2007-8-10-r219 · Cited by 122
“Randomization is the 'gold standard' for inferring causality of one variable on another [18-20].”
BackgroundDunn, G., Emsley, R., Liu, H., Landau, S., Green, J., White, I., & Pickles, A. (2015) · Health Technology Assessment · doi:10.3310/hta19930 · Cited by 122
High-dimensional multivariate mediation with application to neuroimaging data
Chén, O. Y., Crainiceanu, C. M., Ogburn, E. L., Caffo, B. S., Wager, T. D., & Lindquist, M. A. (2017) · Biostatistics · doi:10.1093/biostatistics/kxx027 · Cited by 150
“Using potential outcomes notation (Rubin (1974)), let M ( x ) denote the value of the mediators if treatment X is set to x .”
BackgroundEcnomic consequences of workplace injuries and illnesses: Lost earnings and benefit adequacy
Boden, L. I., & Galizzi, M. (1999) · American Journal of Industrial Medicine · doi:10.1002/(sici)1097-0274(199911)36:5<532::aid-ajim5>3.0.co;2-9 · Cited by 122
Titus, M. A. (2006) · Research in Higher Education · doi:10.1007/s11162-006-9034-3 · Cited by 148
Endogenous selection or treatment model estimation
Lewbel, A. (2007) · Journal of Econometrics · doi:10.1016/j.jeconom.2006.11.004 · Cited by 134
“See, e.g., Horvitz and Thompson (1952), Heckman (1974, 1976, 1979), Rubin (1974), Koul, Susarla, and van Ryzin (1981), Lee (1982), and Rosenbaum and Rubin (1985).”
Network A/B Testing: From Sampling to Estimation
Gui, H., Xu, Y., Bhasin, A., & Han, J. (2015) · Proceedings of the 24th International Conference on World Wide Web · doi:10.1145/2736277.2741081 · Cited by 121
“Rubin causal model [23], a standard machinery of testing framework, is usually adopted in conducting and analyzing A/B tests.”
Method
Metadata
- Authors
- Rubin, D. B.
- Affiliations
- Educational Testing Service (US)
- Corresponding author
- Rubin, Donald B.
- ORCID
- Rubin, Donald B.
- Venue
- Journal of Educational Psychology
- Published
- 1974-10
- Volume
- 66
- Issue
- 5
- Pages
- 688-701
- ISSN
- 0022-0663 (ISSN-L)
- 1939-2176
- OpenAlex
- W2132917208
- Cited by
- 9,436
- References
- 10
- Citations by year
- 2026: 490 · 2025: 658 · 2024: 612 · 2023: 587 · 2022: 589 · 2021: 636 · 2020: 625 · 2019: 554 · 2018: 493 · 2017: 389 · 2016: 370 · 2015: 382 · 2014: 406 · 2013: 343 · 2012: 325
- FWCI
- 2.83
- Type
- journal article
- Language
- en
- Open access
- green
- nrs.harvard.edu
- Licence
- cc-by
- Locations
- Journal of Educational Psychology (publishedVersion)
- Digital Access to Scholarship at Harvard (DASH) (Harvard University) (submittedVersion, cc-by)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Education › Educational Research and Methods
- Keywords
- causal effect estimation, randomization, non-randomized studies, matching, random sampling, treatment effects, observational studies
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.
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.