That’s a Lot to Process! Pitfalls of Popular Path Models
2022/04/01 by Julia M. Rohrer, Paul Hünermund, Ruben C. Arslan +1 · 4 voices · 242 citations
Decision Sciences · Mathematics · Psychology · Social Sciences · #Advanced Causal Inference Techniques #Artificial intelligence #Causal inference #Causal model #Cognitive psychology #Computer science #Constraint (computer-aided design) #Econometrics #Economics #Epistemology #Evaluation and Performance Assessment #Independence (probability theory) #Inference #Machine learning #Mathematics #Mediation #Moderated mediation #Moderation #Path (computing) #Path analysis (statistics) #Process (computing) #Psychology #Qualitative Comparative Analysis Research #Social psychology #Social science #Sociology #Statistics
paper · pdf · doi:10.1177/25152459221095827
published in Advances in Methods and Practices in Psychological Science 5(2) (SAGE Publishing)
openalex publication_date 2022/04/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
Abstract
Path models to test claims about mediation and moderation are a staple of psychology. But applied researchers may sometimes not understand the underlying causal inference problems and thus endorse conclusions that rest on unrealistic assumptions. In this article, we aim to provide a clear explanation for the limited conditions under which standard procedures for mediation and moderation analysis can succeed. We discuss why reversing arrows or comparing model fit indices cannot tell us which model is the right one and how tests of conditional independence can at least tell us where our model goes wrong. Causal modeling practices in psychology are far from optimal but may be kept alive by domain norms that demand every article makes some novel claim about processes and boundary conditions. We end with a vision for a different research culture in which causal inference is pursued in a much slower, more deliberate, and collaborative manner.
Citations
Cited by
- Causal inference for psychologists who think that causal inference is not for them
- Elucidating the relevance of dissociation and emotion dysregulation for aggression in forensic and community samples
- Ideology shapes evaluation of history within the general population
- The ‘sour grapes’ of conspiracy theories: how does the emotion of ressentiment predict conspiracy beliefs?
- A Pre-Registered Examination of the Relationship Between Psychopathy, Boredom-Proneness, and University-Level Cheating
- Defending democracy: investigating the efficacy of elite democratic defence in a competitive information environment
- Belief strength and the attitudinal and behavioural correlates of conspiracy beliefs
- The role of partisan news and political parasocial relationships in shaping meta‐perceptions: Cross‐national evidence
- Rethinking measurement invariance causally
- Do they dislike us as much as we think? Positive contact as a potential rectifier of meta-attitude inaccuracy in conflictual intergroup settings
- Reducing the causal illusion: a question of motivation or of information?
- Through the lens of causal inference: Decisions and pitfalls of covariate selection
- The positivity assumption in causal mediation analyses? Checked!
- A social constructivist viewpoint of media effects: extending the social influence model of technology use to media effects
- Causal inference in ethnographic research: Refining explanations with abductive logic, strength of evidence assessments, and graphical models
- The Causal Cookbook: Recipes for Propensity Scores, G-Computation, and Doubly Robust Standardization
- These Are Not the Effects You Are Looking for: Causality and the Within-/Between-Persons Distinction in Longitudinal Data Analysis
- The combined well-being effects of social media activities: how self-affirmation can buffer against upward social comparisons on Instagram
- Thinking Clearly About Age, Period, and Cohort Effects
- The link between Empathy and Forgiveness: Replication and extensions Registered Report of McCullough et al. (1997)'s Study 1
- Social mobility beliefs and attitudes toward redistribution: Potential explanatory mechanisms
- Message-level Claims Require Message-level Data Analyses: Aligning Claims and Evidence in Communication Research
- Spite and Science‐Denial: Exploring the Role of Spitefulness in Conspiracy Ideation and COVID‐19 Conspiracy Beliefs
- Looking our limitations in the eye: A call for more thorough and honest reporting of study limitations
- From Ads to Action: Femvertising’s Impact on Brand Image and Purchase Probability Through Well-Being Dimensions
- The influence of minority stress-related experiences on mental wellbeing for trans/gender-diverse and cisgender youth: a comparative longitudinal analysis. [europepmc]
- Acceptance and Commitment Therapy Processes and Mediation: Challenges and How to Address Them. [europepmc]
- The regression trap: why regression analyses are not suitable for selecting determinants to target in behavior change interventions. [europepmc]
- Self-perceptions as mechanisms of achievement inequality: evidence across 70 countries. [europepmc]
- Informed participation? An investigation of the relationship between exposure to different news channels and participation mediated through actual and perceived knowledge. [europepmc]
- The significance of parental mentalizing for four-year-old children's solitary pretend play. [europepmc]
- Causal inference in ethnographic research: Refining explanations with abductive logic, strength of evidence assessments, and graphical models. [europepmc]
- Social media features, perceived group norms, and adolescents' active social media use matter for perceived friendship quality. [europepmc]
- Moral trade-offs reveal foundational representations that predict unique variance in political attitudes. [europepmc]
- From acceptance to change: The role of acceptance in the effectiveness of the Informative Process Model for conflict resolution. [europepmc]
- Self-compassion improves barrier self-efficacy and subsequently physical activity: A test of longitudinal mediation using a representative sample of the United Kingdom. [europepmc]
- Assessing the role of conspiracy beliefs in oncological treatment decisions: An experimental approach. [europepmc]
- Methods in causal inference. Part 1: causal diagrams and confounding. [europepmc]
- Power effects on interindividual and intergroup competition. [europepmc]
- Social signals of belonging: How the perceived ethnic-national background of friends affects ascriptions of belonging given to descendants of migrants. [europepmc]
- Reducing the causal illusion: a question of motivation or of information? [europepmc]
- Jurymen Seldom Rule Against a Person That They Like: The Relationship Between Emotions Towards a Defendant, the Understanding of Case Facts, and Juror Judgments in Civil Trials. [europepmc]
- Tutorial: Power analyses for interaction effects in cross-sectional regressions. [europepmc]
- Intersectional discrimination, psychological distress, and physical health symptoms among Latine sexual minority adults. [europepmc]
- Mourning ritual participation, subjective well-being and prosocial behaviour among the Luhya people of Kenya. [europepmc]
- From Perceived Economic Inequality to Support for Redistribution: The Role of Meritocracy Perception. [europepmc]
- The direct effect of drinking to cope on alcohol problems is not mediated by alcohol consumption: Invariance across gender and countries. [europepmc]
- The roles of individual differences in time perspective, promotion focus, and innovativeness: Testing technology acceptance model. [europepmc]
- The tendency to dehumanize, group malleability beliefs, and perceived threat from migrants in Hungary. [europepmc]
- Investing to gain others' trust: Cognitive abstraction increases prosocial behavior and trust received from others. [europepmc]
Discussions
- Another day, and another reference to these bangers. Contemporary causal inference has fundamentally changed the way I think about control variables and mediation analyses. These can show you why: do [bsky, 98 points, 5 comments]
- Tempted to conduct a cross-sectional mediation analysis? Here's a well-written paper by @dingdingpeng.bsky.social @phuenermund.bsky.social @the100.ci on why these designs are not so constructive of kn [bsky, 31 points, 2 comments]
- Of course, the management equivalent is to still teach Baron and Kenny (1986). I don't know what's worse. doi.org/10.1177/2515... [bsky, 6 points, 0 comments]
- Another reference that may be helpful about the first point is this article by @dingdingpeng.the100.ci @p-hunermund.com et al. [bsky, 2 points, 0 comments]
Related