Doubly robust (DR) estimators that combine regression adjustments and inverse probability weighting (IPW) are widely used in causal inference with observational data because they are claimed to be consistent when either the outcome or the treatment selection model is correctly specified (Scharfstein et al., 1999). This property of "double…

The Karlson-Holm-Breen (KHB) method has rapidly become popular as a way of separating the impact of confounding from rescaling when comparing conditional and unconditional parameter estimates in nonlinear probability models such as the logit and probit. In this note, we show that the same estimates can be obtained in a somewhat different way to…

Employing nested sequences of models is a common practice when exploring the extent to which one set of variables mediates the impact of another set. Such an analysis in the context of logistic regression models confronts two challenges: (1) direct comparisons of coefficients across models are generally biased due to the changes in scale that…

This article reviews important statistical methods for estimating the impact of interventions on outcomes in education settings, particularly programs that are implemented in field, rather than laboratory, settings. We begin by describing the causal inference challenge for evaluating program effects. Then four research designs are discussed that…

Network meta-analysis (NMA) is gaining popularity for comparing multiple treatments in a single analysis. Generalized linear mixed models provide a unifying framework for NMA, allow us to analyze datasets with dichotomous, continuous or count endpoints, and take into account multiarm trials, potential heterogeneity between trials and network…

Randomized controlled trials (RCTs) and quasi-experimental designs like regression discontinuity (RD) designs, instrumental variable (IV) designs, and matching and propensity score (PS) designs are frequently used for inferring causal effects. It is well known that the features of these designs facilitate the identification of a causal estimand…

Purpose: This paper aims to introduce matching in propensity score analysis (PSA) as an alternative statistical approach for researchers looking to make causal inferences using intact groups. Design/methodology/approach: An illustrative example demonstrated the varying results of analysis of variance, analysis of covariance and PSA on a heuristic…

When randomized experiments are infeasible, quasi-experimental designs can be exploited to evaluate causal treatment effects. The strongest quasi-experimental designs for causal inference are regression discontinuity designs, instrumental variable designs, matching and propensity score designs, and comparative interrupted time series designs. This…

The use of propensity scores as a method to promote causality in studies that cannot use random assignment has increased dramatically since its original publication in 1983. While the utility of these approaches is important, the concepts underlying their use are complex. The purpose of this article is to provide a basic tutorial for conducting…

High-performing teachers tend to seek out positions at more affluent or academically challenging schools, which tend to hire more experienced, effective educators. Consequently, low-income and minority students are more likely to attend schools with less experienced and less effective educators (see, for example, DeMonte & Hanna, 2014; Office…

Stochastic understanding of probability distribution undergirds development of conceptual connections between probability and statistics and supports development of a principled understanding of statistical inference. This study investigated the impact of an instructional course intervention designed to support development of stochastic…

Mediation is usually assessed by a regression-based or structural equation modeling (SEM) approach that we refer to as the classical approach. This approach relies on the assumption that there are no confounders that influence both the mediator, "M", and the outcome, "Y". This assumption holds if individuals are randomly…

Very little explanatory power is required in order for regressions to exhibit statistical significance. This article discusses some of the causes and implications. (Contains 2 tables.)

Students in statistics service courses are frequently exposed to dogmatic approaches for evaluating the role of randomization in statistical designs, and inferential data analysis in experimental, observational and survey studies. In order to provide an overview for understanding the inference process, in this work some key statistical concepts in…

Three versions of a probabilistic model adapted from the theory of information retrieval--a binary version, a version using the full value of the data, and a version using principal components--were tested and applied to data available from application forms to predict graduate school performance of library school students. (EM)