This primer systematizes the emerging literature on causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction to building and optimizing custom deep learning models and shows how to adapt them to estimate/predict heterogeneous treatment effects. It also discusses ongoing work to…

Sensitivity analysis, a statistical method crucial for validating inferences across disciplines, quantifies the conditions that could alter conclusions (Razavi et al., 2021). One line of work is rooted in linear models and foregrounds the sensitivity of inferences to the strength of omitted variables (Cinelli & Hazlett, 2019; Frank, 2000). A…

This is a practical guide for applying the Impact Threshold for a Confounding Variable and the Robustness of Inference to Replacement using the konfound packages in Stata and R as well as the R-shiny app. It includes motivation worked examples, and tutorials.

We are sometimes forced to use the Interrupted Time Series (ITS) design as an identification strategy for potential policy change, such as when we only have a single treated unit and cannot obtain comparable controls. For example, with recent county- and state-wide criminal justice reform efforts, where judicial bodies have changed bail setting…

Recent critiques of physics education research (PER) studies have revoiced the critical issues when drawing causal inferences from observational data where no intervention is present. In response to a call for a "causal reasoning primer" in PER, this paper discusses some of the fundamental issues in statistical causal inference. In…

This article provides an overview of different computational options for inference following normal theory maximum likelihood (ML) estimation in structural equation modeling (SEM) with incomplete normal and nonnormal data. Complete data are covered as a special case. These computational options include whether the information matrix is observed or…

In this article, we propose a variational Bayes (VB) inference method for the deterministic input noisy AND gate model of cognitive diagnostic assessment. The proposed method, which applies the iterative algorithm for optimization, is derived based on the optimal variational posteriors of the model parameters. The proposed VB inference enables…

We describe an activity that introduces students to population modelling, enables them to use estimates obtained from a sample to infer back to the population, and understands how the findings are translatable via penguins and their poo!

Too often, statistical inference and probability are treated in schools like they are unrelated. In this paper, we describe how we supported students to learn about the role of probability in making inferences with variable data by building models of real world events and using them to simulate repeated samples.

Students often believe that statistical significance is the only determinant of whether a quantitative result is "important." In this paper, I review traditional null hypothesis statistical testing to identify what questions inferential statistics can and cannot answer, including statistical significance, effect size and direction,…

In this article, we extend Liao's test for across-group comparisons of the fixed effects from the generalized linear model to the fixed and random effects of the generalized linear mixed model (GLMM). Using as our basis the Wald statistic, we developed an asymptotic test statistic for across-group comparisons of these effects. The test can be…

People use models every day without even realizing it. Models create a structure for predictions (inferences) that can be used or adapted as situations change. A model is a relational system that highlights aspects of a phenomenon that the modeler deems important and diminishes the rest. Statistical models capture variability of data, enabling…

The number of articles showing dissatisfaction with the null hypothesis statistical testing (NHST) framework has been progressively increasing over the years. Alternatives to NHST have been proposed and the Bayesian approach seems to have achieved the highest amount of visibility. In this last part of the special issue, a few alternative…

Researchers in many disciplines work with ranking data. This data type is unique in that it is often deterministic in nature (the ranks of items "k"-1 determine the rank of item "k"), and the difference in a pair of rank scores separated by "k" units is equivalent regardless of the actual values of the two ranks in…

Nonnormality of data presents unique challenges for researchers who wish to carry out structural equation modeling. The subsequent SPSS syntax program computes bootstrap-adjusted fit indices (comparative fit index, Tucker-Lewis index, incremental fit index, and root mean square error of approximation) that adjust for nonnormality, along with the…