Résumés
Résumé
La méthode du score de propension devient de plus en plus populaire pour estimer les effets causaux d’un programme d’intervention. Si les applications empiriques de cette méthode sont encore rares dans les recherches en éducation, des exemples de son utilisation se trouvent aisément dans d’autres disciplines. Cependant, sa mise en place soulève plusieurs questions. L’objectif de cet article est de fournir des éléments de réponses guidant le chercheur et l’évaluateur du domaine de l’éducation pour l’estimation et l’utilisation du score de propension. Les différentes étapes de son application sont présentées pas à pas : évaluation du biais de sélection, construction du score de propension et mesure de sa qualité, et choix des stratégies d’utilisation du score dans l’estimation des effets d’un traitement. Les questions méthodologiques soulevées sont discutées à chaque étape. Pour faciliter la compréhension, un exemple d’une expérimentation en maternelle illustre la méthode.
Mots-clés :
- score de propension,
- appariement sur score de propension,
- pondération inverse sur les probabilités d’être traité,
- doubles différences,
- biais de sélection
Abstract
Propensity score methods become increasingly popular for estimating the causal effects of an intervention. If empirical applications of this method are still unusual in education researches, examples can be found easily in other fields. However, its implementation raises several questions. The aim of this paper is to provide guidance to educational researchers regarding the estimation and the utilization of the propensity score. The various stages of its application are presented step by step: evaluation of the selection bias, construction of the propensity score and measurement of its quality, and decision on the strategies to use in the estimation of treatment effects. Methodological issues are discussed at each stage. To facilitate the understanding, an example of an experiment in kindergarten illustrates the method.
Keywords:
- propensity score,
- propensity score matching,
- inverse probability of treatment weighting,
- difference in differences,
- selection bias
Resumo
O método da pontuação de propensão está tornar-se cada vez mais popular para estimar os efeitos causais de um programa de intervenção. Embora as aplicações empíricas deste método ainda sejam raras na investigação em educação, exemplos do seu uso podem ser facilmente encontrados noutras disciplinas. No entanto, a sua implementação levanta diversas questões. O objetivo deste artigo é fornecer algumas respostas para orientar o investigador e o avaliador do domínio da educação para a estimativa e uso da pontuação de propensão. As diferentes etapas da sua aplicação são apresentadas passo a passo: avaliação do viés de seleção, construção da pontuação de propensão e medição da sua qualidade e escolha das estratégias para o uso da pontuação na estimativa dos efeitos do tratamento. As questões metodológicas são discutidas em cada etapa. Para facilitar o entendimento, um exemplo de uma experiência no jardim-de-infância ilustra o método.
Palavras chaves:
- pontuação de propensão,
- aparelhamento da pontuação de propensão,
- ponderação inversa sobre as probabilidades de ser tratado,
- diferenças duplas,
- viés de seleção
Parties annexes
Références
- Abadie, A., & Imbens, G. W. (2005). Large sample properties of matching estimators for average treatment effects. Econometrica, 74(1), 235-267.
- Altman, D. G., & Bland, J. M. (1999). Treatment allocation in controlled trials: Why randomise? British Medical Journal, 318(7192), 1209.
- Altman, D. G., & Doré, C. J. (1990). Randomisation and baseline comparisons in clinical trials. The Lancet, 335(8682), 149-153.
- Banerjee, V. D., & Duflo, E. (2009). The experimental approach to development economics. Annual Review of Economics, 1(1), 151-178.
- Baser, O. (2006). Too much ado about propensity score models? Comparing methods of propensity score matching. Value in Health, 9(6), 377-385.
- Becker, S. O., & Ichino, A. (2002). Estimation of average treatment effects based on propensity scores. The Stata Journal, 2(4), 358-377.
- Bénabou, R., Kramarz, F., & Prost, C. (2004). Zones d’éducation prioritaire : quels moyens pour quels résultats ? Économie et Statistique, 380(1), 3-34.
- Berger, V. W. (2005). Quantifying the magnitude of baseline covariate imbalances resulting from selection bias in randomized clinical trials. Biometrical Journal, 47(2), 119-127.
- Berger, V. W., & Exner, D. V. (1999). Detecting selection bias in randomized clinical trials. Controlled Clinical Trials, 20(4), 319-327.
- Bjerk, D. (2009). How much can we trust causal interpretations of fixed-effects estimators in the context of criminality? Journal of Quantitative Criminology, 25(4), 391-417.
- Brodaty, T., Crépon, B., & Fougère, D. (2007). Les méthodes microéconométriques d’évaluation et leurs applications aux politiques actives de l’emploi. Économie et Prévision, 177, 91-118.
- Bryson, A., Dorsett, R., & Purdon, S. (2002). The use of propensity score matching in the evaluation of active labour market policies. Department for Work and Pensions (No. 4) Working paper.
- Caliendo, M., & Kopeinig, S. (2008). Some practical guidance for the implementation of propensity score matching. Journal of Economic Surveys, 22(1), 31-72.
- Cochran, W. G., & Chambers, S. P. (1965). The planning of observational studies of human populations. Journal of the Royal Statistical Society. Series A (General), 128(2), 234-266.
- Dehejia, R. (2005). Practical propensity score matching: A reply to Smith and Todd. Journal of Econometrics, 125(1), 355-364.
- Dehejia, R. H., & Wahba, S. (1999). Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs. Journal of the American Statistical Association, 94(448), 1053-1062.
- Dehejia, R. H., & Wahba, S. (2002). Propensity score-matching methods for nonexperimental causal studies. Review of Economics and Statistics, 84(1), 151-161.
- Dumont, M., Leclerc, D., Massé, L., Potvin, P., & McKinnon, S. (2009). Programme de gestion du stress des adolescents comme levier de résilience. In N. Nader-Grosbois (Eds.), Résilience, régulation et qualité de vie : Concepts, évaluation et intervention (pp. 301-305). Louvain-la-Neuve, Belgique : Presses universitaires de Louvain.
- Fortin, L. (2012). Trait d’union. Québec, Canada : Centre de transfert pour la réussite éducative du Québec (CTREQ).
- Fortin, L., Marcotte, D., Potvin, P., Royer, É., & Joly, J. (2006). Typology of students at risk of dropping out of school: Description by personal, family and school factors. European Journal of Psychology of Education, 21(4), 363-383.
- Fortin, L., & Potvin, P. (2007). Logiciel de dépistage du décrochage scolaire. Québec, Canada : Centre de transfert pour la réussite éducative du Québec (CTREQ).
- Fortin, L., Royer, É., Potvin, P., Marcotte, D., & Yergeau, É. (2004). La prédiction du risque de décrochage scolaire au secondaire : facteurs personnels, familiaux et scolaires. Revue canadienne des sciences du comportement, 36(3), 219-231.
- Grimes, D. A., & Schulz, K. F. (2002). Bias and causal associations in observational research. Lancet, 359(9302), 248-252.
- Harrison, G. P., Gruber, A. J., Hudson, J. I., Huestis, M. A., & Yurgelun-Todd, D. (2002). Cognitive measures in long-term cannabis users. The Journal of Clinical Pharmacology, 42(11), 41-47.
- Heckman, J. J., Ichimura, H., & Todd, P. E. (1997). Matching as an econometric evaluation estimator: Evidence from evaluating a job training programme. The Review of Economic Studies, 64(4), 605-654.
- Heckman, J. J., Ichimura, H., & Todd, P. E. (1998). Matching as an econometric evaluation estimator. The Review of Economic Studies, 65(24), 261-294.
- Heckman, J. J., LaLonde, R. J., & Smith, J. A. (1999). The economics and econometrics of active labor market programs. Handbook of Labor Economics, 3, 1865-2097.
- Hirano, K., Imbens, G. W., & Ridder, G. (2003). Efficient estimation of average treatment effects using the estimated propensity score. Econometrica, 71(4), 1161-1189.
- Imbens, G. W. (2004). Nonparametric estimation of average treatment effects under exogeneity: A review. Review of Economics and Statistics, 86(1), 4-29.
- Imbens, G. W., & Wooldridge, J. M. (2009). Recent developments in the econometrics of program evaluation. Journal of Economic Literature, 47(1), 5-86.
- Janosz, M. (2012). QES-Web. Québec, Canada : Centre de transfert pour la réussite éducative du Québec (CTREQ).
- Lechner, M. (2002). Program heterogeneity and propensity score matching: An application to the evaluation of active labor market policies. Review of Economics and Statistics, 84(2), 205-220.
- Lecocq, A. (2012). Genèse et évolution des compétences des élèves à la fin de l’école maternelle : éléments d’analyse à partir de données de panel et d’une expérimentation musicale (Thèse de doctorat non-publiée). Université de Bourgogne, Dijon, France.
- Leuven, E., & Sianesi, B. (2012). PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Boston, MA: Boston College Department of Economics.
- Marcotte, D. (2006). Pare-Chocs, programme d’intervention auprès d’adolescents dépressifs. Québec, Canada : Septembre éditeur.
- Marcotte, D., & Baron, P. (1993). L’efficacité d’une stratégie d’intervention émotivo-rationnelle auprès d’adolescents dépressifs du milieu scolaire. Revue canadienne de counseling et de psychothérapie, 27(2), 77-92.
- Murray, C., & Malmgren, K. (2005). Implementing a teacher-student relationship program in a high-poverty urban school : Effects on social, emotional, and academic adjustment and lessons learned. Journal of School Psychology, 43(2), 137-152.
- Robins, J. M., Hernán, M. Á., & Brumback B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11(5), 550-560.
- Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41-55.
- Rosenbaum, P. R., & Rubin, D. B. (1984). Reducing bias in observational studies using subclassification on the propensity score. Journal of the American Statistical Association, 79(387), 516-524.
- Rosenbaum, P. R., & Rubin, D. B. (1985a). The bias due to incomplete matching. Biometrics, 41(1), 103-116.
- Rosenbaum, P. R., & Rubin, D. B. (1985b). Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. The American Statistician, 39(1), 33-38.
- Rubin, D. B. (1997). Estimating causal effects from large data sets using propensity scores. Annals of Internal Medicine, 127(8), 757-763.
- Rubin, D. B., Stuart, E. A., & Zanutto, E. L. (2004). A potential outcomes view of value-added assessment in education. Journal of Educational and Behavioral Statistics, 29(1), 103-116.
- Schmidt, C. M., & Augurzky, B. (2001). The propensity score: A means to an end (No. 271). Extrait de http://anon-ftp.iza.org/dp271.pdf
- Smith, J. A., & Todd, P. E. (2001). Reconciling conflicting evidence on the performance of propensity-score matching methods. The American Economic Review, 91(2), 112-118.
- Smith, J. A., & Todd, P. E. (2005). Does matching overcome LaLonde’s critique of nonexperimental estimators? Journal of Econometrics, 125(1), 305-353.
- Steiner, P. M., Cook, T. D., Shadish, W. R., & Clark, M. H. (2010). The importance of covariate selection in controlling for selection bias in observational studies. Psychological Methods, 15(3), 250-267.
- Winship, C., & Mare, R. D. (1992). Models for sample selection bias. Annual Review of Sociology, 18, 327-350.
- Wood, L., Egger, M., Gluud, L. L., Schulz, K. F., Jüni, P., Altman, D. G., & Sterne, J. A. (2008). Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: Meta-epidemiological study. British Medical Journal, 336(7644), 601-605.