In causal mediation studies that decompose an average treatment effect into indirect and direct effects, examples of post-treatment confounding are abundant. In the presence of treatment-by-mediator interactions, past research has generally considered it infeasible to adjust for a post-treatment confounder of the mediator-outcome relationship due…

For binary experimental data, we discuss randomization-based inferential procedures that do not need to invoke any modeling assumptions. We also introduce methods for likelihood and Bayesian inference based solely on the physical randomization without any hypothetical super population assumptions about the potential outcomes. These estimators have…

In this commentary from Issue 14, n3, authors John Cadogan and Nick Lee applaud the paper by Aguirre-Urreta, Rönkkö, and Marakas "Measurement: Interdisciplinary Research and Perspectives", 14(3), 75-97 (2016), since their explanations and simulations work toward demystifying causal indicator models, which are often used by scholars…

Inferring causal effects of treatments is a central goal in many disciplines. The potential outcomes framework is a main statistical approach to causal inference, in which a causal effect is defined as a comparison of the potential outcomes of the same units under different treatment conditions. Because for each unit at most one of the potential…

One of the central assumptions of the causal-indicator literature is that all causal indicators must be included in the research model and that the exclusion of one or more relevant causal indicators would have severe negative consequences by altering the meaning of the latent variable. In this research we show that the omission of a relevant…

There is unexamined potential for developing and testing rival causal explanations in the type of data that participant observation is best suited to create: descriptions of in situ social interaction crafted from the participants' perspectives. By intensively examining a single ethnography, we can see how multiple predictions can be derived from…

Regression discontinuity design (RD) has been widely used to produce reliable causal estimates. Researchers have validated the accuracy of RD design using within study comparisons (Cook, Shadish & Wong, 2008; Cook & Steiner, 2010; Shadish et al, 2011). Within study comparisons examines the validity of a quasi-experiment by comparing its…

The sharp regression discontinuity design (RDD) has three key weaknesses compared to the randomized clinical trial (RCT). It has lower statistical power, it is more dependent on statistical modeling assumptions, and its treatment effect estimates are limited to the narrow subpopulation of cases immediately around the cutoff, which is rarely of…

Illustrates problems facing researchers trying to demonstrate causal relationships between types of nonparental care and differences between groups of Swedish children. Argues that efforts must be made to validate and interpret differences that are found. Indicates ways to avoid misinterpretation of differences that are attributable to…