mediation: R Package for Causal Mediation Analysis

Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions

This article provides researchers with a guide to properly construe and conduct analyses of conditional indirect effects, commonly known as moderated mediation effects. We disentangle conflicting d...

Introduction to Statistical Mediation Analysis

This volume introduces the statistical, methodological, and conceptual aspects of mediation analysis. Applications from health, social, and developmental psychology, sociology, communication, exercise science, and epidemiology are emphasized throughout. Single-mediator, multilevel, and longitudinal models are reviewed. The author's goal is to help the reader apply mediation analysis to their own data and understand its limitations. Each chapter features an overview, numerous worked examples, a summary, and exercises (with answers to the odd numbered questions). The accompanying CD contains outputs described in the book from SAS, SPSS, LISREL, EQS, MPLUS, and CALIS, and a program to simulate the model. The notation used is consistent with existing literature on mediation in psychology. The book opens with a review of the types of research questions the mediation model addresses. Part II describes the estimation of mediation effects including assumptions, statistical tests, and the construction of confidence limits. Advanced models including mediation in path analysis, longitudinal models, multilevel data, categorical variables, and mediation in the context of moderation are then described. The book closes with a discussion of the limits of mediation analysis, additional approaches to identifying mediating variables, and future directions. Introduction to Statistical Mediation Analysis is intended for researchers and advanced students in health, social, clinical, and developmental psychology as well as communication, public health, nursing, epidemiology, and sociology. Some exposure to a graduate level research methods or statistics course is assumed. The overview of mediation analysis and the guidelines for conducting a mediation analysis will be appreciated by all readers.

Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis

Abstract. Baron and Kenny’s procedure for determining if an independent variable affects a dependent variable through some mediator is so well known that i

Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models

Hypotheses involving mediation are common in the behavioral sciences. Mediation exists when a predictor affects a dependent variable indirectly through at least one intervening variable, or mediator. Methods to assess mediation involving multiple simultaneous mediators have received little attention in the methodological literature despite a clear need. We provide an overview of simple and multiple mediation and explore three approaches that can be used to investigate indirect processes, as well as methods for contrasting two or more mediators within a single model. We present an illustrative example, assessing and contrasting potential mediators of the relationship between the helpfulness of socialization agents and job satisfaction. We also provide SAS and SPSS macros, as well as Mplus and LISREL syntax, to facilitate the use of these methods in applications.

Estimating Causal Effects of Treatments in Randomized and Nonrandomized Studies

Presents a discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation. The objective was to specify the benefits of randomization in estimating causal effects of treatments. It is concluded that randomization should be employed whenever possible but that the use of carefully controlled nonrandomized data to estimate causal effects is a reasonable and necessary procedure in many cases.

Statistical Models Answer the Fundamental Clinical Question and Provide Clinical Trial Estimands – Statistical Thinking

Specific goals and estimation targets for randomized clinical trials have still not been well defined for general outcome variables. Proponents of causal inference calculus have claimed to define goals and estimands, but they have largely done so in a way that is not concordant with the most popular design, the parallel-group randomized trial. Causal inferential methods require the use of counterfactuals that are not informed by any data (outside of crossover studies) and make assumptions that are unverifiable, e.g., about the correlation structure of potential outcomes. Causal inferential structure also leads practitioners to act as if marginal treatment effect estimates are both helpful in decision making and transport to populations when in fact neither is true. Heterogeneity of participants within a treatment arm dictates heterogeneity of outcomes and heterogeneity of treatment effects when quantified on an absolute scale. Statistical models are best poised for estimation and causal inference that is specific to patient types. The increasing generality and robustness of statistical models bolsters the case. In this article I provide a succinct statement of the clinical goal of a parallel-group trial, and statistical estimands for it in the context of a general family of robust and efficient ordinal models that contain virtually all routinely used statistical models and tests as special cases.

The central role of the propensity score in observational studies for causal effects

Abstract. The propensity score is the conditional probability of assignment to a particular treatment given a vector of observed covariates. Both large and

The C-Word: Scientific Euphemisms Do Not Improve Causal Inference From Observational Data

Causal inference is a core task of science. However, authors and editors often refrain from explicitly acknowledging the causal goal of research projects; they refer to causal effect estimates as associational estimates. This commentary argues that using the term “causal” is necessary to improve the quality of observational research. Specifically, being explicit about the causal objective of a study reduces ambiguity in the scientific question, errors in the data analysis, and excesses in the interpretation of the results.

A Statistical Interrogation of “The Case for Causality, Part 1” by Rausch and Haidt – Matthew B. Jané

I don’t know anything about the literature on social media and mental health so my focus on this post is to interrogate the statistical approach taken by the article written by Zach Rausch and Jonathon Haidt (link here) and to some extent the original meta-analysis by Ferguson.

Why Most Published Research Findings Are False

Summary There is increasing concern that most current published research findings are false. The probability that a research claim is true may depend on study power and bias, the number of other studies on the same question, and, importantly, the ratio of true to no relationships among the relationships probed in each scientific field. In this framework, a research finding is less likely to be true when the studies conducted in a field are smaller; when effect sizes are smaller; when there is a greater number and lesser preselection of tested relationships; where there is greater flexibility in designs, definitions, outcomes, and analytical modes; when there is greater financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance. Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias. In this essay, I discuss the implications of these problems for the conduct and interpretation of research.

Recursive partitioning for heterogeneous causal effects

In this paper we propose methods for estimating heterogeneity in causal effects in experimental and observational studies and for conducting hypothesis tests about the magnitude of differences in treatment effects across subsets of the population. We provide a data-driven approach to partition the data into subpopulations that differ in the magnitude of their treatment effects. The approach enables the construction of valid confidence intervals for treatment effects, even with many covariates relative to the sample size, and without “sparsity” assumptions. We propose an “honest” approach to estimation, whereby one sample is used to construct the partition and another to estimate treatment effects for each subpopulation. Our approach builds on regression tree methods, modified to optimize for goodness of fit in treatment effects and to account for honest estimation. Our model selection criterion anticipates that bias will be eliminated by honest estimation and also accounts for the effect of making additional splits on the variance of treatment effect estimates within each subpopulation. We address the challenge that the “ground truth” for a causal effect is not observed for any individual unit, so that standard approaches to cross-validation must be modified. Through a simulation study, we show that for our preferred method honest estimation results in nominal coverage for 90% confidence intervals, whereas coverage ranges between 74% and 84% for nonhonest approaches. Honest estimation requires estimating the model with a smaller sample size; the cost in terms of mean squared error of treatment effects for our preferred method ranges between 7–22%.

Human Adults and LLMs as Scientists: Who Benefits from Active Exploration?

A long-standing finding in the causal learning literature is that adults struggle to identify conjunctive causal rules, where an effect requires the simultaneous presence of multiple causes, while...

How Field Experiments in Economics Can Complement Psychological Research on Judgment Biases

This review summarizes results of field experiments examining individual behaviors across several market settings—from open-air markets to rideshare markets to tax-compliance markets—where people sort themselves into market roles wherein they make consequential decisions. Using three distinct examples from my own research on the endowment effect, left-digit bias, and omission bias, I showcase how field experiments can help researchers understand mediators, heterogeneity, and causal moderation involved in judgment biases in the field. In this manner, the review highlights that economic field experiments can serve an invaluable intellectual role alongside traditional laboratory research.

“Influencing the influencers:” a field experimental approach to promoting effective mental health communication on TikTok

A substantial body of social scientific research considers the negative mental health consequences of social media use on TikTok. Fewer, however, consider the potentially positive impact that mental health content creators (“influencers”) on TikTok can have to improve health outcomes; including the degree to which the platform exposes users to evidence-based mental health communication. Our novel, influencer-led approach remedies this shortcoming by attempting to change TikTok creator content-producing behavior via a large, within-subject field experiment (N = 105 creators with a reach of over 16.9 million viewers; N = 3465 unique videos). Our randomly-assigned field intervention exposed influencers on the platform to either (a) asynchronous digital (.pdf) toolkits, or (b) both toolkits and synchronous virtual training sessions that aimed to promote effective evidence-based mental health communication (relative to a control condition, exposed to neither intervention). We find that creators treated with our asynchronous toolkits—and, in some cases, those also attending synchronous training sessions—were significantly more likely to (i) feature evidence-based mental health content in their videos and (ii) generate video content related to mental health issues. Moderation analyses further reveal that these effects are not limited to only those creators with followings under 2 million users. Importantly, we also document large system-level effects of exposure to our interventions; such that TikTok videos featuring evidence-based content received over half a million additional views in the post-intervention period in the study’s treatment groups, while treatment group mental health content (in general) received over three million additional views. We conclude by discussing how simple, cost-effective, and influencer-led interventions like ours can be deployed at scale to influence mental health content on TikTok.

Recognising, Anticipating, and Mitigating LLM Pollution of Online Behavioural Research

Online behavioural research faces an emerging threat as participants increasingly turn to large language models (LLMs) for advice, translation, or task delegation: LLM Pollution. We identify three interacting variants through which LLM Pollution threatens the validity and integrity of online behavioural research. First, Partial LLM Mediation occurs when participants make selective use of LLMs for specific aspects of a task, such as translation or wording support, leading researchers to (mis)interpret LLM-shaped outputs as human ones. Second, Full LLM Delegation arises when agentic LLMs complete studies with little to no human oversight, undermining the central premise of human-subject research at a more foundational level. Third, LLM Spillover signifies human participants altering their behaviour as they begin to anticipate LLM presence in online studies, even when none are involved. While Partial Mediation and Full Delegation form a continuum of increasing automation, LLM Spillover reflects second-order reactivity effects. Together, these variants interact and generate cascading distortions that compromise sample authenticity, introduce biases that are difficult to detect post hoc, and ultimately undermine the epistemic grounding of online research on human cognition and behaviour. Crucially, the threat of LLM Pollution is already co-evolving with advances in generative AI, creating an escalating methodological arms race. To address this, we propose a multi-layered response spanning researcher practices, platform accountability, and community efforts. As the challenge evolves, coordinated adaptation will be essential to safeguard methodological integrity and preserve the validity of online behavioural research.

Recognising, Anticipating, and Mitigating LLM Pollution of Online Behavioural Research

Online behavioural research faces an emerging threat as participants increasingly turn to large language models (LLMs) for advice, translation, or task delegation: LLM Pollution. We identify three interacting variants through which LLM Pollution threatens the validity and integrity of online behavioural research. First, Partial LLM Mediation occurs when participants make selective use of LLMs for specific aspects of a task, such as translation or wording support, leading researchers to (mis)interpret LLM-shaped outputs as human ones. Second, Full LLM Delegation arises when agentic LLMs complete studies with little to no human oversight, undermining the central premise of human-subject research at a more foundational level. Third, LLM Spillover signifies human participants altering their behaviour as they begin to anticipate LLM presence in online studies, even when none are involved. While Partial Mediation and Full Delegation form a continuum of increasing automation, LLM Spillover reflects second-order reactivity effects. Together, these variants interact and generate cascading distortions that compromise sample authenticity, introduce biases that are difficult to detect post hoc, and ultimately undermine the epistemic grounding of online research on human cognition and behaviour. Crucially, the threat of LLM Pollution is already co-evolving with advances in generative AI, creating an escalating methodological arms race. To address this, we propose a multi-layered response spanning researcher practices, platform accountability, and community efforts. As the challenge evolves, coordinated adaptation will be essential to safeguard methodological integrity and preserve the validity of online behavioural research.