Showing papers by "Donald B. Rubin published in 2011"

Causal Effects of Perceived Immutable Characteristics

Abstract: Despite their ubiquity, observational studies to infer the causal effect of a so-called immutable characteristic, such as race or sex, have struggled for coherence, given the unavailability of a manipulation analogous to a “treatment” in a randomized experiment and the danger of posttreatment bias. We demonstrate that a shift in focus from actual traits to perceptions of them can address both of these issues while facilitating articulation of other critical concepts, particularly the timing of treatment assignment. We illustrate concepts by discussing the designs of various studies of the role of race in trial court death penalty decisions.

Propensity score methods for creating covariate balance in observational studies

Abstract: Randomization of treatment assignment in experiments generates treatment groups with approximately balanced baseline covariates. However, in observational studies, where treatment assignment is not random, patients in the active treatment and control groups often differ on crucial covariates that are related to outcomes. These covariate imbalances can lead to biased treatment effect estimates. The propensity score is the probability that a patient with particular baseline characteristics is assigned to active treatment rather than control. Though propensity scores are unknown in observational studies, by matching or subclassifying patients on estimated propensity scores, we can design observational studies that parallel randomized experiments, with approximate balance on observed covariates. Observational study designs based on estimated propensity scores can generate approximately unbiased treatment effect estimates. Critically, propensity score designs should be created without access to outcomes, mirroring the separation of study design and outcome analysis in randomized experiments. This paper describes the potential outcomes framework for causal inference and best practices for designing observational studies with propensity scores. We discuss the use of propensity scores in two studies assessing the effectiveness and risks of antifibrinolytic drugs during cardiac surgery.

Métodos de puntuación de propensión para crear una distribución equilibrada de las covariables en los estudios observacionales

Abstract: Resumen La asignacion aleatoria del tratamiento en los experimentos divide a los pacientes en grupos de tratamiento que estan aproximadamente equilibrados en cuanto a las covariables basales. Sin embargo, en los estudios observacionales, en los que la asignacion del tratamiento no es aleatoria, los pacientes de los grupos de tratamiento activo y de control difieren a menudo en covariables cruciales que estan relacionadas con las variables de respuesta. Estos desequilibrios en las covariables pueden conducir a estimaciones sesgadas del efecto del tratamiento. La puntuacion de propension (propensity score) es la probabilidad de que a un paciente con unas caracteristicas basales especificas se le asigne el tratamiento activo, y no el control. Aunque las puntuaciones de propension son desconocidas en los estudios observacionales, al parear o subclasificar a los pacientes segun las puntuaciones de propension estimadas, podemos disenar estudios observacionales que sean analogos a los experimentos aleatorios, con un equilibrio aproximado entre pacientes en cuanto a las covariables observadas. Los disenos de estudios observacionales basados en puntuaciones de propension estimadas pueden producir estimaciones aproximadamente insesgadas del efecto del tratamiento. Una cuestion crucial es que los disenos de puntuacion de propension deben crearse sin tener acceso a las respuestas, imitando la separacion entre el diseno del estudio y el analisis de las respuestas que es propia de los experimentos aleatorios. En este articulo se describen el marco conceptual de las respuestas potenciales para la inferencia causal y las mejores practicas para el diseno de estudios observacionales con puntuaciones de propension. Comentamos el uso de puntuaciones de propension en dos estudios en los que se evaluaron la efectividad y los riesgos de los farmacos antifibrinoliticos durante las cirugias cardiacas.

Credible Causal Inference for Empirical Legal Studies

Abstract: We review advances toward credible causal inference that have wide application for empirical legal studies. Our chief point is simple: Research design trumps methods of analysis. We explain matching and regression discontinuity approaches in intuitive (nontechnical) terms. To illustrate, we apply these to existing data on the impact of prison facilities on inmate misconduct, which we compare to experimental evidence. What unifies modern approaches to causal inference is the prioritization of research design to create—without reference to any outcome data—subsets of comparable units. Within those subsets, outcome differences may then be plausibly attributed to exposure to the treatment rather than control condition. Traditional methods of analysis play a small role in this venture. Credible causal inference in law turns on substantive legal, not mathematical, knowledge.

Bayesian Analysis of a Two-Group Randomized Encouragement Design

Abstract: Randomized encouragement designs, terminology established in the seminal article by Holland (1988) although used earlier (eg, Swinton, 1975), are the norm when dealing with human populations At least in much of the world today, thankfully, we cannot force anyone to take a randomly assigned treatment; rather, we can only encourage them to do so, typically after describing some details of what to expect under each of the treatment conditions prior to participation, so-called informed consent As a consequence, human experiments often face the complication of noncompliance with assigned treatment For example, the treatment group may be randomly assigned to be encouraged to study more, whereas the control group receives no extra encouragement In this example, hours of studying will be measured in both groups of the study – treatment and control – as will the primary outcome variable, final achievement on a test Not only is it of interest to study the effects of the encouragement on the amount of studying and on achievement, but it is also of interest to investigate the relationship between the amount of studying, an intermediate outcome variable, on the primary outcome variable in the treatment and control groups