And now a thread of things that must be true in order to validly interpret a standard mediation analysis.

These are assumptions you must defend with outside knowledge. They are not tested as part of the analysis, and often they are not testable even in principle.

Here goes...
1. The effect of M on X is exactly zero
2. The effect of Y on M is exactly zero
3. The effect of Y on X is exactly zero
4. Either there exists no variable correlated with X that affects M, or there are multiple such variables working in different directions that happen to exactly cancel each other out
5. Either there exists no variable correlated with M that affects Y, or there are multiple such variables working in different directions that happen to exactly cancel each other out
(5b. A special case of a variable correlated with M that affects Y would be another mediator. So that means you are assuming that the M you are studying is the only mediator between X and Y. Or else some kind of complicated canceling-out thing)
6. Either there exists no variable correlated with X that affects Y, or there are multiple such variables working in different directions that happen to exactly cancel each other out
7. Either the effects of X on M and M on Y do not vary (for every person, X affects that person's M to the exact same degree and likewise for M affecting Y) OR...
... the people for whom X affects M are the same people for whom M affects Y (fudged this a little to assume that "how much X affects M" is binary, it's a little more complicated if - more realistically - it's continuous)
There's lots more. This is just from a causal modeling perspective, doesn't get into the standard assumptions if you're doing OLS regression etc. Also people who know causal inference better than me will probably correct/complicate my list so check back for replies
And in the meantime read this…

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Sanjay Srivastava

Sanjay Srivastava Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @hardsci

21 Jul
If psychologists had to present and defend a DAG every time they did a mediation analysis, 10% of them would result in "sorry the effect we care about isn't identified" and the other 90% would be "sorry we don't understand this phenomenon well enough to draw a DAG"
(Apparently today is "be salty about mediation" day in my brain)
Real talk though, I think mediation analysis can sometimes be useful if it comes with a thorough, eyes-open discussion of all the things that must be true for it to be interpretable as mediation
Read 4 tweets
7 Apr
Hey psychology twitter. The journalist Jesse Singal is probably going to be showing up on your radar because he has a new book about social psychology. He has also written about trans people. I want to encourage you to read what trans people have said about that work
There is a lot of stuff out there and it can feel a little overwhelming. Here is one place to start, a collection of links to critiques of a very influential 2018 Atlantic article he wrote
This tweet and the thread it is a part of encapsulate some of what really sunk in the most to me about the criticism
Read 10 tweets
28 Jan
Read 7 tweets
27 Jan
Is who you follow on Twitter correlated with your mental health? Using machine learning, we find that the high-degree accounts that people follow - celebrities, public figures, etc. - can predict anxiety, depression, post-traumatic stress, & anger (out-of-sample R = .2)
Substantive upshot: A user's Twitter experience - what they see in their feed - is heavily affected by their decisions of who to follow. Those curation decisions, in the aggregate, are associated with mental health
Some notes and points of interest:

1. "Predict" is used in the machine-learning sense, think of this as a correlation - causality wasn't tested

2. The size of the correlation is neither huge nor neglible. We don't think it can be dismissed but it shouldn't be exaggerated either
Read 6 tweets
17 Dec 20
Were you taught that you cannot interpret a main effect in the presence of an interaction? That interactions "supersede" main effects? That you have to use hedging language like, "the effect of A depends on B"?

Then I've got a little provocation for you. Thread...
Imagine the following study: People with depression are randomly assigned to get either drugs or psychotherapy. In addition, they are asked which they believe is more effective: half say drugs, half say therapy. Outcome is functioning after treatment (0-100). So it's a 2x2
The PI asks two grad students to each run an ANOVA. (What can I say, she likes redundancy.) She wants to know if psychotherapy is more effective than drugs. Both grad students go off and do their thing, and report back at the next lab meeting
Read 15 tweets
10 Dec 20
I’m sorry but Bounty is just begging to be memed
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!