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 801–900 of 1,074
Attributing Effects to Treatment in Matched Observational Studies
Rosenbaum, P. R. (2002) · Journal of the American Statistical Association · doi:10.1198/016214502753479329 · Cited by 105
“Each subject has two potential binary responses, rTij and rCij , where rTij would be observed if the subject were exposed to treatment and rCij would be observed if not (Neyman 1923; Rubin 1974).”
BackgroundA review of causal estimation of effects in mediation analyses
Ten Have, T. R., & Joffe, M. M. (2012) · Statistical Methods in Medical Research · doi:10.1177/0962280210391076 · Cited by 103
“,(8) van der Laan and Petersen(1) identified the average natural direct effect in (9) with assumptions (14), (15), (17), and an another identifying assumption:”
BackgroundMethodInfluentialLMest : An R Package for Latent Markov Models for Longitudinal Categorical Data
Bartolucci, F., Pandolfi, S., & Pennoni, F. (2017) · Journal of Statistical Software · doi:10.18637/jss.v081.i04 · Cited by 115
“We also plan to include estimation methods which are alternative to pure maximum likelihood estimation, as the three-step method proposed by Bartolucci, Montanari, and Pandolfi (2015) as well as the estimation procedure proposed by Bartolucci, Pennoni, and Vittadini (2016) to allow the model to be suitable in a potent…”
MethodReynolds, C. L., & DesJardins, S. L. (2009) · Higher Education: Handbook of Theory and Research · doi:10.1007/978-1-4020-9628-0_2 · Cited by 104
“Owing to Neyman (1923), Rubin (1974) , and Holland (1986), the counterfactual framework is that an individual has two potential outcomes: one that is observed and another that is unobserved.”
BackgroundInstrumental Variables for Logistic Regression: An Illustration
Foster, E. M. (1997) · Social Science Research · doi:10.1006/ssre.1997.0606 · Cited by 114
Alpha, beta, and gamma change in evaluation research: A structural equation approach.
Millsap, R. E., & Hartog, S. B. (1988) · Journal of Applied Psychology · doi:10.1037/0021-9010.73.3.574 · Cited by 120
Agency Problems in Public Firms: Evidence from Corporate Jets in Leveraged Buyouts
Edgerton, J. (2012) · The Journal of Finance · doi:10.1111/j.1540-6261.2012.01784.x · Cited by 115
“I discuss the likely relationship between excessive jet use and selection into PE ownership in a simple model that incorporates elements of the heterogeneous treatment effects framework of Rubin [1974] or Angrist, Imbens, and Rubin [1996]. Under the assumption that private equity ownership eliminates inefficient jet u…”
BackgroundMethodInfluentialAssessing inflation targeting through intervention analysis
Angeriz, Á., & Arestis, P. (2007) · Oxford Economic Papers · doi:10.1093/oep/gpm047 · Cited by 104
“For example, the programme evaluation literature pioneered by Rubin (1974) recommends a solution to this problem by providing a ‘counterfactual’ framework for estimating treatment effects across multiple individuals (see, also, Angrist et al., 1996).”
MethodAid and Growth: Have We Come Full Circle?Preprint
Arndt, C., Jones, S., & Tarp, F. (2009) · SSRN Electronic Journal · doi:10.2139/ssrn.1489392 · Cited by 129
Mobile Phone Coverage and Producer Markets: Evidence from West Africa
Aker, J. C., & Fafchamps, M. (2014) · The World Bank Open Knowledge Repository (World Bank) · doi:10.1596/1813-9450-6986 · Cited by 114
Empirical and counterfactual conditions for sufficient cause interactions
Vanderweele, T. J., & Robins, J. M. (2008) · Biometrika · doi:10.1093/biomet/asm090 · Cited by 103
“In both philosophy and statistics the work is associated with counterfactuals or potential outcomes (Hume, 1748; Neyman, 1923; Lewis, 1973a, 1973b; Rubin, 1974, 1978; Robins, 1986, 1987).”
BackgroundPimentel, S. D., Kelz, R. R., Silber, J. H., & Rosenbaum, P. R. (2015) · Journal of the American Statistical Association · doi:10.1080/01621459.2014.997879 · Cited by 104
“…would exhibit binary response rT ij if treated with Zij = 1 or binary response rCij if control with Zij = 0, so the observed response from ij is Rij = Zij rT ij + (1 − Zij ) rCij and the effect of the treatment on ij , namely θij = rT ij − rCij , is not observed; see Neyman (1923) and Rubin (1974).”
BackgroundEx post impacts of improved maize varieties on poverty in rural Ethiopia
Zeng, D., Alwang, J., Norton, G. W., Shiferaw, B. A., Jaleta, M., & Yirga, C. (2015) · Agricultural Economics · doi:10.1111/agec.12178 · Cited by 106
Personality traits and the evaluation of start-up subsidies
Caliendo, M., Künn, S., & Weißenberger, M. (2015) · European Economic Review · doi:10.1016/j.euroecorev.2015.11.008 · Cited by 132
“To illustrate the idea behind PS matching, we use the well known potential outcome framework (Roy, 1951; Rubin, 1974).”
MethodCausal discovery and the problem of psychological interventions
Eronen, M. I. (2020) · New Ideas in Psychology · doi:10.1016/j.newideapsych.2020.100785 · Cited by 102
“Similar considerations apply to Rubin’s causal model (RCM; Holland, 1986; Rubin, 1974, 2005), which is probably the most influential 10 Would it be possible for a causal relationship to hold at the population level, but not for any individual in the population?”
MethodA Causal Dirichlet Mixture Model for Causal Inference from Observational Data
Lin, A., Lü, J., Xuan, J., Zhu, F., & Zhang, G. (2020) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3379500 · Cited by 102
Probabilities Of Causation: Three Counterfactual Interpretations And Their Identification
Pearl, J. (1999) · Synthese · doi:10.1023/a:1005233831499 · Cited by 207
“19.2.5 Relation to Neyman–Rubin Model Several concepts defined in Section 19.2.1 bear similarity to concepts in the potential-outcome model used by Neyman (1923) and Rubin (1974) in the statis tical analysis of treatment effects.”
InfluentialThe effect of variable health insurance deductibles on the demand for physician visits
Schellhorn, M. (2001) · Health Economics · doi:10.1002/hec.630 · Cited by 103
Assessing mediation using marginal structural models in the presence of confounding and moderation.
Coffman, D. L., & Zhong, W. (2012) · Psychological Methods · doi:10.1037/a0029311 · Cited by 102
“Our proposed approach will be based on the potential outcomes framework (Holland, 1986; Rubin, 1974, 2005) and marginal structural models (MSMs; Robins, Hernan, & Brumback, 2000).”
Semiparametric Bayes analysis of longitudinal data treatment models
Chib, S., & Hamilton, B. H. (2002) · Journal of Econometrics · doi:10.1016/s0304-4076(02)00122-7 · Cited by 120
“…is to specify a joint model of the treatment and the potential outcomes (i.e., the outcomes, or counter-factuals, that would have been observed for levels of the treatment not received as in Rubin (1974, 1978)) and then to compare the observed outcome with the potential outcomes for each subject.”
ResultFitzenberger, B., Osikominu, A., & Völter, R. (2006) · SSRN Electronic Journal · doi:10.2139/ssrn.902814 · Cited by 102
Causality, Mediation and Time: A Dynamic Viewpoint
Aalen, O. O., Røysland, K., Gran, J. M., & Ledergerber, B. (2012) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.2011.01030.x · Cited by 126
“A very fruitful theory has been based on this framework which was originally developed, in somewhat different directions, by Donald B. Rubin and James M. Robins; see for example Rubin (1974) and Robins (1986).”
BackgroundMethodOptimal Food Price Stabilization in a Small Open Developing Country1
Gouel, C., & Jean, S. (2013) · The World Bank Economic Review · doi:10.1093/wber/lht018 · Cited by 126
Challenges to validity in single‐group interrupted time series analysis
Linden, A. (2016) · Journal of Evaluation in Clinical Practice · doi:10.1111/jep.12638 · Cited by 113
“Despite its widespread use, the single‐group ITSA design remains a vastly inferior evaluation approach to those utilizing a comparable control group to serve as the counterfactual—a fundamental element of the potential outcomes framework.(13,14) With a comparable control group, factors other than the intervention that…”
MethodText and Causal Inference: A Review of Using Text to Remove Confounding from Causal Estimates
Keith, K. A., Jensen, D. D., & O’Connor, B. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.474 · Cited by 172
Individual Therapy: New Dawn or False Dawn?
Senn, S. J. (2001) · Drug Information Journal · doi:10.1177/009286150103500443 · Cited by 108
“This is one possible simple statistical model for what happens in a randomized clinical trial and also corresponds to the Rubin causal model (41,42).”
MethodOn the International Effects of Inflation Targeting
Lin, S. (2010) · The Review of Economics and Statistics · doi:10.1162/rest.2009.11553 · Cited by 100
Principal Stratification -- a Goal or a Tool?
Pearl, J. (2011) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1322 · Cited by 132
The role of Science Parks: a puzzle of growth, innovation and R&D investments
Lamperti, F., Mavilia, R., & Castellini, S. (2015) · The Journal of Technology Transfer · doi:10.1007/s10961-015-9455-2 · Cited by 140
“3 Average treatment effects As first formalized in Rubin (1974) [51], the estimation of causal effects, whether from a randomized experiment or a non-experimental study like ours, is inherently a comparison of potential outcomes.”
BackgroundMallard, S. R., Howe, A. S., & Houghton, L. A. (2016) · American Journal of Clinical Nutrition · doi:10.3945/ajcn.116.136879 · Cited by 119
The Dozen Things Experimental Economists Should Do (More of)
Czibor, E., Jimenez‐Gomez, D., & List, J. A. (2019) · Southern Economic Journal · doi:10.1002/soej.12392 · Cited by 118
“Subjects can still opt in (d = 1) or out (d = 0) of the training course, and in this case Z = D. 4Our framework follows the tradition of Rubin (1974), which can traced back to the work of Jerzy Neyman; see also Freedman (2006).”
On the Specification of Propensity Scores, With Applications to the Analysis of Trade Policies
Millimet, D. L., & Tchernis, R. (2009) · Journal of Business and Economic Statistics · doi:10.1198/jbes.2009.06045 · Cited by 119
“One such assumption is unconfoundedness or selection on observables (Rubin 1974; Heckman and Robb 1985).”
Generalizing from Unrepresentative Experiments: A Stratified Propensity Score Approach
O'Muircheartaigh, C., & Hedges, L. V. (2014) · Journal of the Royal Statistical Society Series C: Applied Statistics · doi:10.1111/rssc.12037 · Cited by 99
“Rubin (1974, 1977, 1978, 1980, 1986) has provided a framework for defining causal effects and demonstrating how they can be estimated in experiments and other research designs.”
BackgroundStudies of Labor Market Intermediation
Autor, D. (2009) · University of Chicago Press eBooks · doi:10.7208/chicago/9780226032900.001.0001 · Cited by 104
“The causal effect of a treatment on an outcome is defi ned as the difference between two potential outcomes (Rubin 1974; Heckman 1990).”
Imbens, G. W. (2019) · National Bureau of Economic Research · doi:10.3386/w26104 · Cited by 119
Getting back into the labor market: the effects of start-up subsidies for unemployed females
Caliendo, M., & Künn, S. (2015) · Journal of Population Economics · doi:10.1007/s00148-015-0540-5 · Cited by 114
“5.1 Identification and estimation of causal effects In order to estimate causal effects, we base our analysis on the potential outcome framework, also known as the Roy (1951)–Rubin (1974) model.”
InfluentialHuber, M., Lechner, M., & Steinmayr, A. (2014) · Empirical Economics · doi:10.1007/s00181-014-0847-1 · Cited by 126
“Using the potential outcome framework advocated by Rubin (1974), among many others, we let Y (1) and Y (0) denote the potential outcomes under treatment and non-treatment, respectively.”
MethodGingerich, D. W. (2010) · Political Analysis · doi:10.1093/pan/mpq010 · Cited by 98
“4.1 Framework The Neyman-Rubin-Holland potential outcomes framework provides our point of departure (Rubin 1974; Holland 1986).”
BackgroundInfluentialTemporary Work Agencies in Italy: A Springboard Toward Permanent Employment?
Ichino, A., Mealli, F., & Nannicini, T. (2005) · Giornale degli Economisti · Cited by 144
A Multidimensional Approach to Measuring Fuel Poverty
Charlier, D., & Legendre, B. (2019) · The Energy Journal · doi:10.5547/01956574.40.2.bleg · Cited by 102
“In that case, comparing the difference in the outcome variable between treated and un - treated individuals provides the impact of the treatment ( Rubin, 1974).”
BackgroundSensitivity Analysis for Inverse Probability Weighting Estimators via the Percentile Bootstrap
Zhao, Q., Small, D. S., & Bhattacharya, B. B. (2019) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/rssb.12327 · Cited by 185
“For the missing data problem, Rubin (1976) used the term “missing at random” (MAR) to describe data that are missing for reasons related to completely observed variables in the data set:”
MethodCapturing interactions, correlations, fits, and transactions: A Person-Environment Relations Model
Rauthmann, J. F. (2021) · The Handbook of Personality Dynamics and Processes · doi:10.1016/b978-0-12-813995-0.00018-2 · Cited by 86
Causal Inference in Urban and Regional Economics
Baum-Snow, N. B., & Ferreira, F. V. (2015) · Handbook of regional and urban economics · doi:10.1016/b978-0-444-59517-1.00001-5 · Cited by 160
“Analysis of this simple case is a straightforward point of departure as it is well understood in the statistics literature going back to the classic treatment of Rubin (1974), and discussed extensively in Holland (1986), and in the economics literature going back to Roy (1951).”
Causally motivated attribution for online advertising
Dalessandro, B., Perlich, C., Stitelman, O., & Provost, F. J. (2012) · doi:10.1145/2351356.2351363 · Cited by 131
The effects of job displacement on job quality: Findings from the Wisconsin Longitudinal Study
Brand, J. E. (2006) · Research in Social Stratification and Mobility · doi:10.1016/j.rssm.2006.03.001 · Cited by 126
Race and Hypertension: Science and Nescience
Cooper, R. S., & Kaufman, J. S. (1998) · Hypertension · doi:10.1161/01.hyp.32.5.813 · Cited by 98
Cerulli, G. (2014) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x1401400301 · Cited by 117
“This section draws on the substantial literature on econometrics of program evaluation, such as Rubin (1974), Angrist (1991), Angrist, Imbens, and Rubin (1996), Heckman, LaLonde, and Smith (1999), Wooldridge (2010), and Cattaneo (2010).”
Heshmati, A., & Lööf, H. (2005) · RePEc: Research Papers in Economics · doi:10.22004/ag.econ.11862 · Cited by 97
Imbens, G. W., Rubin, D. B., & Sacerdote, B. (1999) · National Bureau of Economic Research · doi:10.3386/w7001 · Cited by 106
Self-citationRural Roads and Intermediated Trade: Regression Discontinuity Evidence from Sierra LeonePreprint
Casaburi, L., Glennerster, R., & Suri, T. (2012) · SSRN Electronic Journal · doi:10.2139/ssrn.2161643 · Cited by 120
Quasi-experimental study designs series—paper 6: risk of bias assessment
Waddington, H., Aloe, A. M., Becker, B. J., Djimeu, E. W., Hombrados, J. G., Tugwell, P., Wells, G., & Reeves, B. (2017) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2017.02.015 · Cited by 97
“QE studies are, however, potentially at higher risk of bias than their experimental counterparts [6,7], with perhaps the most critical biases for causal”
BackgroundUplift Modeling for preventing student dropout in higher education
Olaya, D., Vásquez, J., Maldonado, S., Miranda, J., & Verbeke, W. (2020) · Decision Support Systems · doi:10.1016/j.dss.2020.113320 · Cited by 153
“Formally, uplift models predict the ITE in terms of the potential outcomes framework [36].”
Causal assumptions and causal inference in ecological experiments
Kimmel, K. A., Dee, L. E., Avolio, M. L., & Ferraro, P. J. (2021) · Trends in Ecology & Evolution · doi:10.1016/j.tree.2021.08.008 · Cited by 112
“Our inability to observe the same study unit under both treated and control conditions is the Fundamental Prob-lem of Causal Inference [17,18,24] and implies that we cannot estimate unit-level treatment effects.”
Functional Causal Mediation Analysis With an Application to Brain Connectivity
Lindquist, M. A. (2012) · Journal of the American Statistical Association · doi:10.1080/01621459.2012.695640 · Cited by 97
“In this work, we adopt the potential outcomes framework for causal inference (Rubin, 1974) to show that under certain assumptions it is possible to obtain a valid estimate of the average causal effects of the mediator from the parameters of the lfSEM.”
MethodQuality Meets Quantity: Case Studies, Conditional Probability, and Counterfactuals
Sekhon, J. S. (2004) · Perspectives on Politics · doi:10.1017/s1537592704040150 · Cited by 97
“74 Holland 1986; Rubin 1990; Rubin 1978; Rubin 1974; Splawa-Neyman 1990 [1923].”
BackgroundHolland, P. W., & Rubin, D. B. (1982) · ETS Research Report Series · doi:10.1002/j.2333-8504.1982.tb01321.x · Cited by 97
Self-citationQuasi-Oracle Estimation of Heterogeneous Treatment EffectsPreprint
Nie, X., & Wager, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.04912 · Cited by 97
“We formalize our problem in terms of the potential outcomes framework (Neyman, 1923; Rubin, 1974).”
MethodBounds on Direct Effects in the Presence of Confounded Intermediate Variables
Cai, Z., Kuroki, M., Pearl, J., & Tian, J. (2008) · Biometrics · doi:10.1111/j.1541-0420.2007.00949.x · Cited by 95
“Rubin (1974) , equation (1) can be also written as”
BackgroundMethodOn Granger causality and the effect of interventions in time series
Eichler, M., & Didelez, V. (2009) · Lifetime Data Analysis · doi:10.1007/s10985-009-9143-3 · Cited by 123
“Rubin (1974, 1978) stresses that the various treatments to be compared need to be well defined in the sense that it must be clear how an individual or object under investigation can potentially receive any of the treatments.”
BackgroundInfluentialPropensity Score Methods for Creating Covariate Balance in Observational Studies
Pattanayak, C. W., Rubin, D. B., & Zell, E. R. (2011) · Revista Española de Cardiología (English Edition) · doi:10.1016/j.rec.2011.06.008 · Cited by 111
“one potential outcome can be observed for each patient, because each patient is assigned to either active treatment or control, but not both.(8,9) Therefore, causal inference is a missing data problem: the goal is to fill in the missing potential outcomes, estimating what would have happened to each patient had he bee…”
BackgroundSelf-citationAmbrosius, C., & Cuecuecha, A. (2013) · World Development · doi:10.1016/j.worlddev.2013.01.032 · Cited by 124
“Under the assumption of ‘ignorability of treatment assignment’ or ‘selection on observables’, the distribution across the treatment and control groups is therefore random with respect to outcomes, conditional on these confounding covariates (Rubin, 1974; cp. Gelman & Hill, 2006, p. 183ff).”
BackgroundA finite selection model for experimental design of the health insurance study
Morris, C. N. (1979) · Journal of Econometrics · doi:10.1016/0304-4076(79)90053-8 · Cited by 96
On Causation in Demography: Issues and Illustrations
Bhrolcháin, M. N., & Dyson, T. (2007) · Population and Development Review · doi:10.1111/j.1728-4457.2007.00157.x · Cited by 110
Causal inference methods to study nonrandomized, preexisting development interventions
Arnold, B. F., Khush, R. S., Ramaswamy, P., London, A. G., Rajkumar, P., Ramaprabha, P., Durairaj, N., Hubbard, A. E., Balakrishnan, K., & Colford, J. M. (2010) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1008944107 · Cited by 96
Persson, M. (2013) · Political Behavior · doi:10.1007/s11109-013-9254-0 · Cited by 113
Psychosocial Interventions for School Refusal Behavior in Children and Adolescents
Pina, A. A., Zerr, A. A., Gonzales, N. A., & Ortiz, C. D. (2009) · Child Development Perspectives · doi:10.1111/j.1750-8606.2008.00070.x · Cited by 94
“Two evaluators coded the 67 articles as either singlecase experimental design studies (Barlow & Hersen, 1976) or group-design studies (studies comparing two or more conditions using randomized or nonrandomized procedures; Rubin, 1974).”
MethodEconometrics for Evaluations: An Introduction to Recent Developments
Cobb‐Clark, D. A., & Crossley, T. F. (2003) · Economic Record · doi:10.1111/j.1475-4932.2003.00148.x · Cited by 119
Employment effects of the provision of specific professional skills and techniques in Germany
Fitzenberger, B., & Speckesser, S. (2006) · Empirical Economics · doi:10.1007/s00181-006-0088-z · Cited by 120
“Therefore, the evaluation literature uses extensively the result of Rosenbaum and Rubin (1983) that the CIA in equation (2) implies that participants and nonparticipants become comparable in expectation when conditioning on the treatment probability P (X) (propensity score) as a function of the observable characterist…”
MethodRegression Discontinuity Applications with Rounding Errors in the Running Variable
Dong, Y. (2015) · Journal of Applied Econometrics · doi:10.1002/jae.2369 · Cited by 94
“Follow Rubin (1974) to let Y.1/ and Y.0/ denote an individual’s potential outcomes of interest from being treated or not, respectively.”
BackgroundHigh-dimensional regression adjustments in randomized experiments
Wager, S., Du, W., Taylor, J., & Tibshirani, R. J. (2016) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1614732113 · Cited by 140
“We frame our analysis in terms of the Neyman–Rubin potential outcomes model [23, 24].”
MethodA Note on the Common Support Problem in Applied Evaluation Studies
Lechner, M. (2008) · Alexandria (UniSG) (University of St.Gallen) · Cited by 95
The Challenge of Summer Bridge Programs
Sablan, J. R. (2014) · American Behavioral Scientist · doi:10.1177/0002764213515234 · Cited by 95
“An experimental design in SBP research is important to highlight because this design is rarely used even though selection bias issues would affect the conclusiveness of determining program effectiveness (Rubin, 1974).”
MethodCausal identification of single-cell experimental perturbation effects with CINEMA-OT
Dong, M., Wang, B., Wei, J., de O. Fonseca, A. H., Perry, C. J., Frey, A., Ouerghi, F., Foxman, E. F., Ishizuka, J. J., Dhodapkar, R. M., et al. (2023) · Nature Methods · doi:10.1038/s41592-023-02040-5 · Cited by 94
“To perform causal single-cell perturbation effect inference, we have adopted the potential outcome causal framework [22, 26].”
Advances in Statistical Methods for Substance Abuse Prevention Research
MacKinnon, D. P., & Lockwood, C. M. (2003) · Prevention Science · doi:10.1023/a:1024649822872 · Cited by 107
“One of these models is Rubin's causal model (Rubin, 1974).”
BackgroundAsymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback
Saito, Y. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401114 · Cited by 131
“Most existing off-policy estimators utilize the Direct Method (DM) or the IPS estimation technique [6, 15, 31, 32].”
BackgroundMethodInfluentialFerron, J. M., Farmer, J. L., & Owens, C. M. (2010) · Behavior Research Methods · doi:10.3758/brm.42.4.930 · Cited by 94
“Need for Confidence Intervals of Individual Treatment Effects An individual treatment effect (T ) is generally defined as the difference between what would be observed for the individual under the treatment condition (Yt) and what would be observed for that individual under the control condition (Yc); therefore, the t…”
BackgroundEvaluating Markers for Selecting a Patient's Treatment
Song, X., & Pepe, M. S. (2004) · Biometrics · doi:10.1111/j.0006-341x.2004.00242.x · Cited by 94
“Using the notion of potential responses (Rubin, 1974, 1978; Holland, 1986), for t = 0, 1, we define Dt as the response if a subject receives treatment T = t.”
BackgroundDistinguishing Between Cross- and Cluster-Level Mediation Processes in the Cluster Randomized Trial
Pituch, K. A., & Stapleton, L. M. (2012) · Sociological Methods & Research · doi:10.1177/0049124112460380 · Cited by 94
On the Structure of Analyst Research Portfolios and Forecast Accuracy
Kini, O., Mian, S. L., Rebello, M. J., & Venkateswaran, A. (2009) · Journal of Accounting Research · doi:10.1111/j.1475-679x.2009.00338.x · Cited by 102
“oriented research making it more likely that analysts will cover multiple countries if employed by a large broker (see, for example, Rubino (2003)).”
Cost and Benefit of Apprenticeship Training - A Comparison of Germany and SwitzerlandPreprint
Dionisius, R., Mühlemann, S., Pfeifer, H. U., Walden, G., Wenzelmann, F., & Wolter, S. C. (2008) · SSRN Electronic Journal · doi:10.2139/ssrn.1126594 · Cited by 94
Double/debiased machine learning for difference-in-differences models
Chang, N.-C. (2020) · The Econometrics Journal · doi:10.1093/ectj/utaa001 · Cited by 93
“With potential outcome notations (Rubin, 1974), we have , where Y 0 i ( t ) is the outcome that individual i would attain at time t in the absence of the treatment, and Y 1 i ( t ) represents the outcome that individual i would attain at time t if exposed to the treatment.”
BackgroundTchetgen Tchetgen, E. J., & VanderWeele, T. J. (2014) · Epidemiology · doi:10.1097/ede.0000000000000054 · Cited by 93
Explaining Variance; Or, Stuck in a Moment We Can't Get Out Of
Braumoeller, B. F. (2006) · Political Analysis · doi:10.1093/pan/mpj009 · Cited by 107
Study Designs for Extending Causal Inferences From a Randomized Trial to a Target Population
Dahabreh, I. J., Haneuse, S., Robins, J. M., Robertson, S. E., Buchanan, A. L., Stuart, E. A., & Hernán, M. Á. (2020) · American Journal of Epidemiology · doi:10.1093/aje/kwaa270 · Cited by 142
“1 Causal quantities of interest In order to define causal quantities, let Y a be the potential (counterfactual) outcome under intervention to set treatment to a [13,14].”
BackgroundKino, S., Hsu, Y.-T., Shiba, K., Chien, Y.-S., Mita, C., Kawachi, I., & Daoud, A. (2021) · SSM - Population Health · doi:10.1016/j.ssmph.2021.100836 · Cited by 91
“Under the potential-outcome framework, causal effect is quantified by comparing two potential outcomes for an individual had they taken the treatment ( Y 1 ) or not taken it ( Y 0 ) (Hern ´ an & Robins, 2020; Rubin, 1974).”
MethodControlled Direct and Mediated Effects: Definition, Identification and Bounds
VANDERWEELE, T. J. (2011) · Scandinavian Journal of Statistics · doi:10.1111/j.1467-9469.2010.00722.x · Cited by 93
“We assume a counterfactual or potential outcomes framework (Neyman, 1923; Rubin, 1974, 1978).”
BackgroundExperimentation in Political Science
Morton, R. B., & Williams, K. C. (2009) · The Oxford Handbook of Political Methodology · doi:10.1093/oxfordhb/9780199286546.003.0014 · Cited by 87
“In order to make a causal inference about the effect of the treatment on voter behavior, researchers who are searching for facts like Gerber and Green typically take a Rubin Causal Model approach, hereafter RCM, to establishing causality [see Rubin (1974) and Holland (1988)].”
BackgroundMatching and Propensity Scores
Steiner, P. M., & Cook, D. (2013) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780199934874.013.0013 · Cited by 126
“Since the outcome we actually observe for unit i depends on the treatment status, we can define the observed outcome as i i i i i Z Y Z Y Y 1 0 ) 1 ( ( Rubin, 1974).”
BackgroundWhat Makes a Good Neighbor? Race, Place, and Norms of Political Participation
Anoll, A. P. (2018) · American Political Science Review · doi:10.1017/s0003055418000175 · Cited by 101
Sjölander, A., Frisell, T., & Öberg, A. S. (2021) · Annual Review of Statistics and Its Application · doi:10.1146/annurev-statistics-040120-024521 · Cited by 106
Simple relations between principal stratification and direct and indirect effects
VanderWeele, T. J. (2008) · Statistics & Probability Letters · doi:10.1016/j.spl.2008.05.029 · Cited by 106
“We assume a potential outcomes framework Rubin (1974, 1978, 1990).”
BackgroundDouble propensity-score adjustment: A solution to design bias or bias due to incomplete matching
Austin, P. C. (2017) · Statistical Methods in Medical Research · doi:10.1177/0962280214543508 · Cited by 92
“In a setting with two possible treatments, the potential outcomes framework assumes that the ith subject has a pair of potential outcomes: Yi(0) and Yi(1), the outcomes under the control and the active treatment, respectively.(11) However, each subject receives only one of the two treatments.”
To Apply or Not to Apply: FAFSA Completion and Financial Aid Gaps
Kofoed, M. (2016) · Research in Higher Education · doi:10.1007/s11162-016-9418-y · Cited by 110
“The basic idea of propensity matching is provided by the Roy (1951); Rubin 2011) and Rudin (1974) models.”
Confirmatory Program Evaluation: A Method for Strengthening Causal Inference
Reynolds, A. (1998) · American Journal of Evaluation · doi:10.1016/s1098-2140(99)80195-7 · Cited by 92
Tipton, E. (2013) · Evaluation Review · doi:10.1177/0193841x13516324 · Cited by 131
“In practice, this dual randomization procedure is generally infeasible (Bloom 2005; Rubin 1974; Shadish, Cook, and Campbell 2002).”
BackgroundHainmueller, J., Hofmann, B., Krug, G., & Wolf, K. (2016) · Scandinavian Journal of Economics · doi:10.1111/sjoe.12166 · Cited by 110
When parties go abroad: Explaining patterns of extraterritorial voting
Burgess, K., & Tyburski, M. D. (2020) · Electoral Studies · doi:10.1016/j.electstud.2020.102169 · Cited by 100
Mendelian randomization: causal inference leveraging genetic data
Chen, L. G., Tubbs, J. D., Liu, Z., Thach, T., & Sham, P. C. (2024) · Psychological Medicine · doi:10.1017/s0033291724000321 · Cited by 100
“Counterfactual) framework (Rubin, 1974; Splawa-Neyman, Dabrowska, & Speed, 1990), the Campbell Causal Model (Campbell, 1957), the Sufficient-Component Cause model (Rothman, 1976), and the Causal Graph (aka.”
MethodA Practical Asymptotic Variance Estimator for Two-Step Semiparametric Estimators
Ackerberg, D. A., Chen, X., & Hahn, J. (2012) · The Review of Economics and Statistics · doi:10.1162/rest_a_00251 · Cited by 116
“Thus, y1i − y0i can be called the treatment effect for the ith individual (see, e.g., Rubin, 1974).”
BackgroundAnalyzing Two-Stage Experiments in the Presence of Interference
Basse, G., & Feller, A. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2017.1323641 · Cited by 142
“We use the potential outcomes framework to describe the problem (Neyman 1923; Rubin 1974).”
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.