All models are wrong, especially cross sectional mediation models.

Joined July 2020
3 Photos and videos
Cross Sectional Mediation Bot retweeted
Often times, one of the areas of peer review that can get contentious is the use of mediation analysis on cross-sectional data. A fun-but-insightful take on the commentary via Trafimow (2015): tandfonline.com/doi/abs/10.1…
3
9
Cross Sectional Mediation Bot retweeted
Replying to @ProfMattFox
ANCOVA and cross-sectional mediation
2
1
12
Cross Sectional Mediation Bot retweeted
It’s that PROCESS macro again. 99% sure that it's is one of the most misused statistical procedures in personality social psychology.
My intuition is that the effect of personalised ads tiny. But, I just read in the news that a study found people are persuaded by political ads that match their own personality traits, I checked out the paper and oh boy: journals.sagepub.com/doi/ful…
1
1
2

ALT GIF by Lush Interior Design

1

ALT Magic Shia Labeouf GIF

Replying to @RomainRaymondie
Indeed, in mediated moderation the coefficient associated with the interaction term of X & Mod must be reduced when controlling for the mediators. But what would you say is happening when this coefficient is not reduced?
Mediorated modiation 👇💯
Started reviewing a paper🥸 , it claims moderated mediation🧐 .....but reports mediated moderation 🙄
Cross Sectional Mediation Bot retweeted
I ran my last cross-sectional mediation analysis in 2013 and haven’t looked back 😃
2
1
2
Untenable models are not useful for anything 👇📈👮‍♀️
Replying to @ValerieJDougla1
Right? cross-sectional mediation & moderation are not mathematically equivalent, especially when using covariates which is.. always? I agree that cross-sectional mediation is applied too often, but it's great when developing theory (as long as the atemporal relt. is acknowledged)
Cross Sectional Mediation Bot retweeted
I'd LOVE to encourage a model of "choose your own adventure" style of manuscript. Who's in? * If you want to use PROCESS to test your mediation models, turn to p 73. * If you realize that your data is cross-sectional and can't support a mediation model, turn to p 15.
1
1
1
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
Replying to @ErinWestgate
Overall, high SES students seem to use public space (esp iconic public space) more, and using such space appears to promote feelings of belonging both experimentally and in cross-sectional mediation. 6/n
We can only hope 👇
I feel like we're on the cusp of accepting the limitations of concordant / cross sectional mediation and rejecting those who do shotgun SEM with a billion mediators.
1
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
JMIR Public Health: A Moderated Mediation Analysis of Condom Negotiation and Sexual Orientation on the Relationship Between Sexual Coercion and Condom Use in Chinese Young Women: Cross-Sectional Study dlvr.it/Rqv7qs
1
1
Cross Sectional Mediation Bot retweeted
Replying to @sTeamTraen
There are two major problems with mediation analysis as practiced: (1) most such analyses don’t have sufficient design to justify the causal ordering (e.g., cross-sectional, observational)—any number of alternative sequences are equally supported by the data; 1/
1
1
4
Cross Sectional Mediation Bot retweeted
Reviewer red flag: Authors test "mediation" in a CROSS-SECTIONAL study. The study design makes it hard to establish causality. Instead, try: "Potential mediators" "Intervening variables" Acknowledge limits of design Treat study as pilot data @OpenAcademics @PhDVoice
2
8
47
Truth 👇
No need for apologies! The character limit is constraining.Mediation can’t really be explicitly tested with a cross sectional study. You could argue you theoretically meet some conditions for causality, but here the authors probably have to rethink language and interpretation.
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
16 Feb 2021
Relationship Between Physical Activity, Depressive Symptoms and Low Back Pain Related Disability in Older Adults With Low Back Pain: A Cross-Sectional Mediation Analysis dlvr.it/RsrxPR
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
19 Feb 2021
Mediation Effect of Suicide-Related Social Media Use Behaviors on the Association Between Suicidal Ideation and Suicide Attempt: Cross-Sectional Questionnaire Study dlvr.it/Rt3lZX
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
JMIR Mental Health: Factors Influencing #depression and Mental Distress Related to COVID-19 Among University Students in China: Online Cross-sectional Mediation Study dlvr.it/RtDy4k
🚨 🚨 CROSS SECTIONAL MEDIATION ALERT 🚨 🚨
JMIR Mental Health: Factors Influencing #depression and Mental Distress Related to COVID-19 Among University Students in China: Online Cross-sectional Mediation Study dlvr.it/RtvpnY