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 701–800 of 1,074
How to run an experimental auction: a review of recent advances
Canavari, M., Drichoutis, A. C., Lusk, J. L., & Nayga, R. M. (2019) · European Review of Agricultural Economics · doi:10.1093/erae/jbz038 · Cited by 140
“The key point is that by randomly assigning units to the treatments, the difference between the experimental and control units can be considered an unbiased estimate of the causal effect that the researcher is trying to isolate (Rubin, 1974).”
BackgroundMethodKocaj, A., Kuhl, P., Kroth, A. J., Pant, H. A., & Stanat, P. (2014) · KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie · doi:10.1007/s11577-014-0253-x · Cited by 147
Nasseh, K., Vujicic, M., & Glick, M. (2017) · Health Economics · doi:10.1002/hec.3316 · Cited by 120
“We analyzed the impact of a periodontal intervention on total all-cause healthcare costs (sum of total inpatient medical costs, outpatient medical costs, and prescription drug costs), total all-cause medical costs (sum of total inpatient and outpatient medical costs), and total type 2…”
Randomization Inference for Treatment Effect Variation
Ding, P., Feller, A., & Miratrix, L. W. (2015) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/rssb.12124 · Cited by 180
“Following the causal inference literature, we describe our approach by using the potential outcomes framework (Neyman, 1990; Rubin, 1974).”
Translating Causal Claims: Principles and Strategies for Policy‐Relevant Criminology
Sampson, R. J., Winship, C., & Knight, C. R. (2013) · Criminology & Public Policy · doi:10.1111/1745-9133.12027 · Cited by 166
A Simple Multivariate Test for Asymmetric Hypotheses
Clark, W. R., Gilligan, M. J., & Golder, M. (2006) · Political Analysis · doi:10.1093/pan/mpj018 · Cited by 120
Römer, D., Sznitman, S. R., DiClemente, R. J., Salazar, L. F., Vanable, P. A., Carey, M. P., Hennessy, M. H., Brown, L. K., Valois, R. F., STANTON, B. F., et al. (2009) · American Journal of Public Health · doi:10.2105/ajph.2008.155036 · Cited by 125
“To reduce this possible bias, we adjusted uncontrolled background factors by using a propensity-score approach to construct weights for individual observations.(42,43) We obtained the weights by taking the inverse of the estimated probability that an individual was in a media city.”
Entropy Balancing for Continuous TreatmentsPreprint
Tübbicke, S. (2020) · RePEc: Research Papers in Economics · Cited by 120
Short- and medium-term effects of informal care provision on female caregivers’ health
Schmitz, H., & Westphal, M. (2015) · Journal of Health Economics · doi:10.1016/j.jhealeco.2015.03.002 · Cited by 158
Causal inference for statistical fault localization
Baah, G. K., Podgurski, A., & Harrold, M. J. (2010) · International Symposium on Software Testing and Analysis · doi:10.1145/1831708.1831717 · Cited by 150
“2.1 Causal Graphs and Potential Outcomes The approach to fault localization that we present in this paper builds on Pearl s Structural Causal Model [22, 23] and on the potential outcome model of Neyman [21] and Rubin [27].”
MethodInfluentialSimtowe, F. P., Asfaw, S., & Abate, T. (2016) · Agricultural and Food Economics · doi:10.1186/s40100-016-0051-z · Cited by 124
“First proposed by Rubin (1974) the average treatment effect (ATE) parameter measures the effect or impact of a “treatment” on a person randomly selected in the population Wooldrige (2002) In the context of this study “treatment” corresponds to exposure to a technology and the ATE on the adoption…”
Practical Considerations for Questionable IVs
Clarke, D., & Matta, B. (2018) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x1801800308 · Cited by 136
“For example, Small (2007) proposes a case with as few as two instruments, and Kolesár et al.”
How to make causal inferences using texts
Egami, N., Fong, C., Grimmer, J., Roberts, M. E., & Stewart, B. (2022) · Science Advances · doi:10.1126/sciadv.abg2652 · Cited by 201
“Standard causal inference frameworks, such as potential outcomes ( 10 ) and directed acyclic graphs ( 11 ), assume that the treatment and outcome are known and do not depend on the data.”
Design and Analysis of Pen Studies in the Animal Sciences
St-Pierre, N. R. (2007) · Journal of Dairy Science · doi:10.3168/jds.2006-612 · Cited by 119
Angrist, J. D., & Kuersteiner, G. M. (2010) · The Review of Economics and Statistics · doi:10.1162/rest_a_00109 · Cited by 155
Cygan‐Rehm, K., Kuehnle, D., & Oberfichtner, M. (2017) · Health Economics · doi:10.1002/hec.3510 · Cited by 119
“Our empirical methodology departs from the potential outcomes framework (Rubin, 1974).”
Lingam: Non-Gaussian Methods for Estimating Causal Structures
Shimizu, S. (2014) · Behaviormetrika · doi:10.2333/bhmk.41.65 · Cited by 126
“A major framework for causal inference (Pearl, 2000) may be based on a combination of the counterfactual model of causation (Neyman, 1923; Rubin, 1974) and structural equation modeling (Bollen, 1989).”
BackgroundMethodInfluentialSTATISTICS AND CAUSAL INFERENCE*
Holland, P. W., Glymour, C., & Granger, C. W. J. (1985) · ETS Research Report Series · doi:10.1002/j.2330-8516.1985.tb00125.x · Cited by 138
Dunn, G., & Bentall, R. (2007) · Statistics in Medicine · doi:10.1002/sim.2891 · Cited by 118
“Indirect effects might also be investigated by methods based on the ideas of Frangakis and Rubin [45].”
MethodGirma, S., Gong, Y., Görg, H., & Lancheros, S. (2014) · Journal of International Economics · doi:10.1016/j.jinteco.2014.11.007 · Cited by 147
“In order to deal with selection and unobserved heterogeneity in this estimation we apply the conditional independence assumption (CIA) as in the propensity-score matching literature (Rubin, 1974).”
MethodEstrada, E., Ferrer, E., & Pardo, A. (2019) · Frontiers in Psychology · doi:10.3389/fpsyg.2018.02696 · Cited by 118
“Whether the observed change can be attributed to the intervention or not depends on a number of factors, including whether (a) a control group exists; (b) the study is experimental, quasi-experimental or observational; (c) relevant covariates and cofounds have been adequately controlled (Fisher, 1935; Rubin, 1974; Sha…”
BackgroundZubizarreta, J. R., Paredes, R., & Rosenbaum, P. R. (2014) · ScholarlyCommons (University of Pennsylvania) · Cited by 117
Super-Learning of an Optimal Dynamic Treatment Rule
Luedtke, A. R., & van der Laan, M. J. (2016) · The International Journal of Biostatistics · doi:10.1515/ijb-2015-0052 · Cited by 115
Real Time Pricing and Electricity Markets
Allcott, H., & Katz, L. S. (2009) · Cited by 117
Biewen, M., Fitzenberger, B., Osikominu, A., & Paul, M. (2014) · Journal of Labor Economics · doi:10.1086/677233 · Cited by 165
An evaluation of public employment programmes in the East German State of Sachsen-Anhalt
Eichler, M., & Lechner, M. (2002) · Labour Economics · doi:10.1016/s0927-5371(02)00039-8 · Cited by 114
“generalisation is achieved without the need to use instruments, because instruments that for example satisfy the strong assumptions of Angrist, Imbens, and Rubin (1996) are hard to find in general and not available for this particular application.”
Microeconometric Evaluation of the Active Labour Market Policy in Switzerland
Lechner, M., & Gerfin, M. (2000) · SSRN Electronic Journal · doi:10.2139/ssrn.233906 · Cited by 117
“This model is known as the Roy (1951) -Rubin (1974) model of potential outcomes and causal effects.”
Generic machine learning inference on heterogenous treatment effects in randomized experiments
Chernozhukov, V., Demirer, M., Duflo, E., & Fernández‐Val, I. (2018) · RePEc: Research Papers in Economics · doi:10.1920/wp.cem.2017.6117 · Cited by 227
“Let Y (1) and Y (0) be the potential outcomes in the treatment state 1 and the non-treatment state 0; see [46].”
Synthetic controls with staggered adoption.
Ben-Michael, E., Feller, A., & Rothstein, J. (2022) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/rssb.12448 · Cited by 120
“We adopt a potential outcomes framework to express causal quantities (Neyman, 1923; Rubin, 1974) and assume stable treatment and no interference between units (SUTVA; Rubin, 1980).”
MethodTesting Instrument Validity for LATE Identification Based on Inequality Moment Constraints
Huber, M., & Mellace, G. (2014) · The Review of Economics and Statistics · doi:10.1162/rest_a_00450 · Cited by 149
“However, this framework also provides testable implications for IV validity based on deriving bounds on (4)An equivalent result for the potential outcome distributions of the compliers under slightly stronger assumptions has been derived by Imbens and Rubin (1997).”
BackgroundWendling, T., Jung, K., Callahan, A., Schuler, A., Shah, N. H., & Gallego, B. (2018) · Statistics in Medicine · doi:10.1002/sim.7820 · Cited by 157
“We work within the potential outcomes framework.(8,9) We consider a set up where we have n patients, indexed by i=1,.”
BackgroundNetworks of Intergovernmental Organizations and Convergence in Domestic Economic Policies
Cao, X. (2009) · International Studies Quarterly · doi:10.1111/j.1468-2478.2009.00570.x · Cited by 143
On the use of propensity scores in principal causal effect estimation
Jo, B., & Stuart, E. A. (2009) · Statistics in Medicine · doi:10.1002/sim.3669 · Cited by 115
Impact of NERICA Adoption on Rice Yield: Evidence from West Africa
Diagne, A., Midingoyi, S.-K., & Kinkingninhoun-Mêdagbé, F. M. (2012) · Springer eBooks · doi:10.1007/978-94-007-5760-8_7 · Cited by 118
Martin, D., Glass, T. A., Bandeen‐Roche, K., Todd, A. C., Shi, W., & Schwartz, B. S. (2006) · American Journal of Epidemiology · doi:10.1093/aje/kwj060 · Cited by 126
“Under certain conditions, propensity scoring estimates the causal effects of exposures on health (29) in the potential-outcomes sense of Rubin (30).”
BackgroundAthey, S. C., Chetty, R., Imbens, G. W., & Kang, H. (2019) · National Bureau of Economic Research · doi:10.3386/w26463 · Cited by 230
Ethnographic Evidence, Heterogeneity, and Neighbourhood Effects After Moving to Opportunity
Small, M. L., & Feldman, J. (2011) · Springer eBooks · doi:10.1007/978-94-007-2309-2_3 · Cited by 157
Food and Cash Transfers: Evidence from Colombia
Attanasio, O. P., Battistin, E., & Mesnard, A. (2011) · The Economic Journal · doi:10.1111/j.1468-0297.2011.02473.x · Cited by 131
“The quantities tutmtw jjj ,, are potential outcomes that are observed depending on whether or not the programme is in operation (see Rubin, 1974).”
BackgroundCook, T. D., & Steiner, P. M. (2010) · Psychological Methods · doi:10.1037/a0018536 · Cited by 114
“Using X to indicate the covariate, Rubin (1977) wrote the following:”
BackgroundInfluentialSeafood prices reveal impacts of a major ecological disturbance
Smith, M. D., Øglend, A., Kirkpatrick, A. J., Asche, F., Bennear, L. S., Craig, J. K., & Nance, J. M. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1617948114 · Cited by 119
Design and Analysis of Experiments
Mattei, A., Mealli, F., & Nodehi, A. (2021) · Springer eBooks · doi:10.1007/978-3-319-57365-6_40-1 · Cited by 118
Cox, D. R., & Wermuth, N. (2004) · International Statistical Review · doi:10.1111/j.1751-5823.2004.tb00237.x · Cited by 112
“Rubin (1974), in an influential paper, adapted notions of causality from the design of experiments to observational studies via a representation similar to Fisher’s which, without the essential element of physical randomization, had been given by Neyman (1923).”
MethodCausal Effects of Perceived Immutable Characteristics
Greiner, D. J., & Rubin, D. B. (2011) · Review of Economics and Statistics · doi:10.1162/rest_a_00110 · Cited by 114
Self-citationStatistical power for regression discontinuity designs in education evaluations
Schochet, P. Z. (2009) · Journal of Educational and Behavioral Statistics · doi:10.3102/1076998609332748 · Cited by 90
“In what follows, the RD design is discussed in the context of the causal inference theory underlying RA designs (Holland, 1986; Imbens & Rubin, in press; Neyman, 1923/1990; Rubin, 1974; Schochet, 2007).”
MethodUntangling the Causal Effects of Sex on JudgingPreprint
Boyd, C. L., Epstein, L., & Martin, A. D. (2007) · SSRN Electronic Journal · doi:10.2139/ssrn.1001748 · Cited by 114
The Mediation Formula: A Guide to the Assessment of Causal Pathways in Nonlinear Models
Pearl, J. (2012) · Wiley Series in Probability and Statistics · doi:10.1002/9781119945710.ch12 · Cited by 114
“…1) and the symbiosis between graphical and counterfactual analysis that this interpretation engenders.11 In contrast, the structure-less approach of Rubin (1974) has spawned other definitions of direct effects, normally referred to as “principal-strata direct effect (PSDE)” (Frangakis and Rubin,…”
BackgroundSemiparametric Theory and Empirical Processes in Causal Inference
Kennedy, E. H. (2016) · ICSA book series in statistics · doi:10.1007/978-3-319-41259-7_8 · Cited by 241
“In this chapter we mostly rely on the potential outcome framework, due to Neyman [29] and Rubin [47, 48], but note that alternative frameworks based on structural equation models and graphs [31, 32], or decision theory [11] can also be useful.”
MethodCaliendo, M., Mahlstedt, R., & Mitnik, O. A. (2014) · Labour Economics · doi:10.1016/j.labeco.2017.02.001 · Cited by 173
“We base our discussion on the well known potential outcomes framework (Roy, 1951; Rubin, 1974) and focus on the usual parameter of interest in most evaluation studies, the average treatment effect on the treated (ATT): τ E Y D E Y D= ( = 1) − ( = 1).”
BackgroundConfounding in statistical mediation analysis: What it is and how to address it.
Valente, M. J., Pelham, W. E., Smyth, H., & MacKinnon, D. P. (2017) · Journal of Counseling Psychology · doi:10.1037/cou0000242 · Cited by 111
“The potential outcomes framework provides the mathematical notation for fixing relevant characteristics in randomized interventions and non-randomized studies (Rubin, 1974), which has been called revolutionary (Broadbent, 2015; Pearl, 2014).”
BackgroundLechner, M., & Miquel, R. (2009) · Empirical Economics · doi:10.1007/s00181-009-0297-3 · Cited by 128
How effective is targeted advertising?
Farahat, A., & Bailey, M. C. (2012) · Proceedings of the 21st international conference on World Wide Web · doi:10.1145/2187836.2187852 · Cited by 107
Establishing Equivalence: Methodological Progress in Group-matching Design and Analysis
Kover, S. T., & Atwood, A. K. (2013) · American Journal on Intellectual and Developmental Disabilities · doi:10.1352/1944-7558-118.1.3 · Cited by 129
“One approach to causality, the Rubin Causal Model, defines the causal effect— that is, the effect of a manipulable treatment—in terms of potential outcomes: what the outcome would have been for participants in the comparison group had they received the treatment and what the outcome would have been for those in the tr…”
BackgroundMethodResultInfluentialDevelopment from Representation? A Study of Quotas for the Scheduled Castes in India
Jensenius, F. R. (2015) · American Economic Journal Applied Economics · doi:10.1257/app.20140201 · Cited by 127
Understanding the Mechanism: Mediation Analysis in Randomized and Nonrandomized Studies
Mascha, E. J., Dalton, J. E., Kurz, A. M., & Saager, L. (2013) · Anesthesia & Analgesia · doi:10.1213/ane.0b013e3182a44cb9 · Cited by 128
CAUSAL INFERENCE, PATH ANALYSIS AND RECURSIVE STRUCTURAL EQUATIONS MODELS
Holland, P. W. (1988) · ETS Research Report Series · doi:10.1002/j.2330-8516.1988.tb00270.x · Cited by 112
Likelihood-Based Analysis of Causal Effects of Job-Training Programs Using Principal Stratification
Zhang, J. L., Rubin, D. B., & Mealli, F. (2009) · Journal of the American Statistical Association · doi:10.1198/jasa.2009.0012 · Cited by 143
“KEY WORDS: Monotonicity assumption; Principal stratification; Rubin Causal Model; Treatment effect on wages.”
BackgroundMethodSelf-citationA microeconometric evaluation of rehabilitation of long‐term sickness in Sweden
Frölich, M., Heshmati, A., & Lechner, M. (2004) · Journal of Applied Econometrics · doi:10.1002/jae.757 · Cited by 123
Little, R. J., Long, Q., & Lin, X. (2009) · Biometrics · doi:10.1111/j.1541-0420.2008.01066.x · Cited by 111
“…the definition of a causal effect of an active treatment as the difference in hypothetical outcomes under that treatment and a control treatment (Rubin, 1974, 1977, 1978); and principal stratification, where individuals are stratified according to the values of the posttreatment variable under…”
BackgroundFrom Bayesian Networks to Causal Networks
Pearl, J. (1995) · Springer eBooks · doi:10.1007/978-1-4899-1424-8_9 · Cited by 117
Zubizarreta, J. R., Paredes, R. D., & Rosenbaum, P. R. (2014) · The Annals of Applied Statistics · doi:10.1214/13-aoas713 · Cited by 109
“As in Neyman (1923) and Rubin (1974), each subject has two potential responses, rT ij if treated with Zij = 1, rCij if control with Zij = 0, so response Rij = ZijrT ij + (1 − Zij)rCij is observed from ij and the effect of the treatment on ij, namely, rT ij − rCij , is not observed.”
BackgroundCausality and Causal Modelling in the Social Sciences
Russo, F. (2008) · Springer eBooks · doi:10.1007/978-1-4020-8817-9 · Cited by 184
Benefit entitlement and unemployment duration
Lalive, R., & Zweimüller, J. (2003) · Journal of Public Economics · doi:10.1016/j.jpubeco.2003.10.002 · Cited by 126
“To what extent may the stable unit treatment value assumption (SUTVA) have been violated?”
InfluentialThe Selective Labels Problem: Evaluating Algorithmic Predictions in the Presence of Unobservables
Lakkaraju, H., Kleinberg, J., Leskovec, J., Ludwig, J., & Mullainathan, S. (2017) · Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/3097983.3098066 · Cited by 138
Freitag, M., & Stadelmann‐Steffen, I. (2010) · Electoral Studies · doi:10.1016/j.electstud.2010.04.009 · Cited by 123
“For this reason alone, linguistic differences should therefore be excluded from the model (Rubin, 1974: 699).”
BackgroundInput and output additionality of R&D subsidies
Czarnitzki, D., & Hussinger, K. (2017) · Applied Economics · doi:10.1080/00036846.2017.1361010 · Cited by 139
“(1)).7 Building on the conditional independence assumption (Rubin 1974, 1977), one can estimate the counterfactual situation by using a selected group of non-subsidized firms that have similar characteristics X: E R DCjX; S ¼ 1 ¼ E R DCjX; S ¼ 0 : (3) The construction of the control group depends…”
BackgroundScientist’s guide to developing explanatory statistical models using causal analysis principles
Grace, J. B., & Irvine, K. M. (2019) · Ecology · doi:10.1002/ecy.2962 · Cited by 126
Innovative Work Practices, Information Technologies, and Working Conditions: Evidence for France
Askenazy, P., & Caroli, È. (2010) · Industrial Relations A Journal of Economy and Society · doi:10.1111/j.1468-232x.2010.00616.x · Cited by 139
Modern Regression Discontinuity Analysis
Bloom, H. S. (2012) · Journal of Research on Educational Effectiveness · doi:10.1080/19345747.2011.578707 · Cited by 110
Principal Stratification -- Uses and Limitations
VanderWeele, T. J. (2011) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1329 · Cited by 131
“Stated brie y, ifX denotes some binary treatment and S some post-treatment variable and if we let Sx denote the potential outcome (Rubin, 1974) for each individual that we would have observed hadX , possibly contrary to fact, been x, then a principal stratum is simply a subgroup of individuals…”
BackgroundCompared to what? Variation in the impacts of early childhood education by alternative care type
Feller, A., Grindal, T., Miratrix, L. W., & Page, L. C. (2016) · The Annals of Applied Statistics · doi:10.1214/16-aoas910 · Cited by 151
Briscoe, F., & Kellogg, K. C. (2011) · American Sociological Review · doi:10.1177/0003122411401250 · Cited by 109
Double machine learning-based programme evaluation under unconfoundedness
Knaus, M. C. (2022) · Econometrics Journal · doi:10.1093/ectj/utac015 · Cited by 161
“We define the estimands of interest in the multiple treatment version of the potential outcomes framework (Rubin, 1974; Imbens, 2000; Lechner, 2001).”
MethodRecent Developments in Causal Inference and Machine Learning
Brand, J. E., Zhou, X., & Xie, Y. F. (2023) · Annual Review of Sociology · doi:10.1146/annurev-soc-030420-015345 · Cited by 174
“Rubin formalized and extended the potential-outcomes framework in a series of papers in statistics in the 1970s and 1980s (e.g., Rubin 1974, 1977, 1986).”
BackgroundThe Use of Propensity Scores for Nonrandomized Designs With Clustered Data
Thoemmes, F. J., & West, S. G. (2011) · Multivariate Behavioral Research · doi:10.1080/00273171.2011.569395 · Cited by 140
“Hong and Raudenbush (2003) defined causal effects in these designs within the framework of the potential outcomes model (eg, Rubin, 1974 , 1978 ).”
BackgroundCausal Inference in the Social Sciences
Sobel, M. E. (2000) · Journal of the American Statistical Association · doi:10.1080/01621459.2000.10474243 · Cited by 128
“Angrist, Imbens, and Rubin (1996). (For related material, see also Robins and Greenland 1992.”
BackgroundMethodCarlin, J. B. (2001) · Biostatistics · doi:10.1093/biostatistics/2.4.397 · Cited by 109
“Under suitable assumptions, such differences may sometimes be interpreted as causal effects (Rubin, 1974; Holland, 1986).”
BackgroundThe Link between R&D Subsidies, R&D Spending and Technological PerformancePreprint
Czarnitzki, D., & Hussinger, K. (2004) · SSRN Electronic Journal · doi:10.2139/ssrn.575362 · Cited by 117
Preoperative Placement of Inferior Vena Cava Filters and Outcomes After Gastric Bypass Surgery
Birkmeyer, N. J. O., Share, D., Baser, O., Carlin, A. M., Finks, J. F., Pesta, C. M., Genaw, J. A., & Birkmeyer, J. D. (2010) · Annals of Surgery · doi:10.1097/sla.0b013e3181e61e4f · Cited by 109
Evaluating online ad campaigns in a pipeline: causal models at scale
Chan, D. X., Ge, R., Gershony, O., Hesterberg, T., & Lambert, D. (2010) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/1835804.1835809 · Cited by 116
Testing Local Average Treatment Effect Assumptions
Mourifié, I., & Wan, Y. (2016) · The Review of Economics and Statistics · doi:10.1162/rest_a_00622 · Cited by 151
“We adopt Rubin’s (1974) potential outcome model. Let Y = Y1D + Y0(1 − D), where Y is the observed outcome taking values from the support Y , D ∈ {0, 1} is the observed treatment indicator, and (Y1, Y0) are potential outcomes. Let Z be the instrumental variable. For simplicity, we assume Z ∈ Z = {0, 1}, but our analysi…”
MethodMercatanti, A. (2013) · Australian & New Zealand Journal of Statistics · doi:10.1111/anzs.12025 · Cited by 199
“…effects are defined as comparisons of the quantities that are potentially observed for the same unit given different levels of exposure to the treatment (Rubin 1974, 1978), the exclusion restriction requires that the assignment to treatment does not have a direct causal effect on the outcome.”
BackgroundCausal Inference in the Health Sciences: A Conceptual Introduction
Pearl, J. (2001) · Health Services and Outcomes Research Methodology · doi:10.1023/a:1020315127304 · Cited by 132
Expressive Voting and Its Cost: Evidence From Runoffs With Two or Three Candidates
Pons, V., & Tricaud, C. (2018) · Econometrica · doi:10.3982/ecta15373 · Cited by 114
Distinguishing prognostic and predictive biomarkers: an information theoretic approach
Sechidis, K., Papangelou, K., Metcalfe, P., Svensson, D. J., Weatherall, J., & Brown, G. (2018) · Computer applications in the biosciences · doi:10.1093/bioinformatics/bty357 · Cited by 118
“The estimation of treatment effect (via first estimating f(X, ·)) motivates a set of methods that utilise the potential outcomes (or counterfactual) modelling (Rubin, 1974).”
MethodNagin, D. S., & Sampson, R. J. (2019) · Annual Review of Criminology · doi:10.1146/annurev-criminol-011518-024838 · Cited by 107
Teaching Statistical Inference for Causal Effects in Experiments and Observational Studies
Rubin, D. B. (2004) · Journal of Educational and Behavioral Statistics · doi:10.3102/10769986029003343 · Cited by 128
“Although Neyman’s potential outcome notation was standard in the context of randomized experiments, prior to Rubin (1974, 1977, 1978), the observed outcome notation was standard in the context of observational studies.”
BackgroundMethodSelf-citationCovariance adjustment versus gain scores--revisited.
Maris, E. (1998) · Psychological Methods · doi:10.1037//1082-989x.3.3.309 · Cited by 106
Keele, L. J., Titiunik, R., & Zubizarreta, J. R. (2014) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/rssa.12056 · Cited by 143
“We also calculated for each voter the median value of houses sold within a 500-m radius of her residence; we use this as a voter level measure of housing values.”
BackgroundMethodDeveloping multicomponent interventions using fractional factorial designs
Chakraborty, B., Collins, L. M., Strecher, V. J., & Murphy, S. A. (2009) · Statistics in Medicine · doi:10.1002/sim.3643 · Cited by 107
“As is well-known [33–35], findings that are not based on randomization are hard to replicate due to the likely presence of unknown confounders¶ (e.”
BackgroundCounterfactual graphical models for longitudinal mediation analysis with unobserved confounding
Shpitser, I. (2013) · Cognitive Science · doi:10.1111/cogs.12058 · Cited by 106
“…equation approach to mediation analysis is a special case of a more general framework based on potential outcome counterfactuals, developed by Neyman (1923) and Rubin (1974), and extended and linked to non-parametric structural equations and graphical models by J. Robins (1986), and Pearl (2000).”
BackgroundMethodThe Effects of Children’s Migration on Elderly Kin’s Health: A Counterfactual Approach
Kuhn, R., Everett, B. G., & Silvey, R. M. (2011) · Demography · doi:10.1007/s13524-010-0002-3 · Cited by 140
Forsman, H., Brännström, L., Vinnerljung, B., & Hjern, A. (2016) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2016.06.006 · Cited by 133
Interaction between dry and hot extremes at a global scale using a cascade modeling framework
Mukherjee, S., Mishra, A. K., Zscheischler, J., & Entekhabi, D. (2023) · Nature Communications · doi:10.1038/s41467-022-35748-7 · Cited by 127
A Theory of Statistical Inference for Matching Methods in Causal Research
Iacus, S. M., King, G., & Porro, G. (2018) · Political Analysis · doi:10.1017/pan.2018.29 · Cited by 124
“…by reducing the strength of the relationship between pre-treatment covariates and the treatment assignment variable, statistical methods applied to the matched subset have reduced model dependence, estimation error, and bias (Cochran and Rubin, 1973; Ho, Imai, King, and Stuart, 2007; Rubin, 1974).”
MethodAugmented Designs to Assess Immune Response in Vaccine Trials
Follmann, D. A. (2006) · Biometrics · doi:10.1111/j.1541-0420.2006.00569.x · Cited by 117
“Our approach to using these data is perhaps best described using counterfactual reasoning (Rubin, 1974, 1977, 1978; Halloran and Struchiner, 1995) and principal stratification (Frangakis and Rubin, 2002).”
Castaño‐Muñoz, J., Duart, J. M., & Vinuesa, T. S. (2013) · British Journal of Educational Technology · doi:10.1111/bjet.12007 · Cited by 113
“Given that an experimental design was not used, where randomisation—when it comes to designating the individuals of the treatment group—ideally manages to achieve two similar groups with comparable means, a decision was taken to use the Propensity Score Matching technique (Rosenbaum, 2002; Rosenbaum & Rubin, 1983; Rub…”
MethodSoullier, G., & Moustier, P. (2018) · Food Policy · doi:10.1016/j.foodpol.2018.07.004 · Cited by 131
“Tobit models found a stronger impact than PSM because they did not take the zerovalues into consideration.”
BackgroundMethodInfluentialModernizing the Bradford Hill criteria for assessing causal relationships in observational data
Cox, L. A. (2018) · Critical Reviews in Toxicology · doi:10.1080/10408444.2018.1518404 · Cited by 115
Zhang, Y., & Fricker, J. D. (2021) · Transport Policy · doi:10.1016/j.tranpol.2021.01.013 · Cited by 119
Hansford, H. J., Cashin, A. G., Jones, M. D., Swanson, S. A., Islam, N., Douglas, S. R. G., Rizzo, R. R. N., Devonshire, J. J., Williams, S. A., Dahabreh, I. J., et al. (2023) · JAMA Network Open · doi:10.1001/jamanetworkopen.2023.36023 · Cited by 130
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.