Maarten van Smeden Profile picture
Feb 6, 2021 6 tweets 1 min read Read on X
Statistical terms: what they really mean

Multicolinearity— they all look the same
Heteroscedasticity— the variation varies
Attenuation— being too modest
Overfitting— too good to be true
Confounding— nothing is what it seems
P-value— it’s complicated
Sensitivity analysis— tried a bunch of stuff
Post-hoc— main analysis not sexy enough
Multivariate— oops, meant to say multivariable
Normality— a very rare shape for data
Dichotomized— data was tortured
Extrapolation— just guessing
Linear regression— line through data points
t-test— linear regression
correlation— linear regression
ANOVA— linear regression
ANCOVA— linear regression
Chi-square test— logistic regression
Deep learning— bunch of regressions
Advanced stuff:
Non-convergence— computer says no
Heywood case— science fiction
Bootstrap standard errors— could not do the math
Robust standard errors— pretending to be cautious
Shrinkage— regularization
Regularization— couldn’t get large enough dataset
Validated— we did a test (it might have failed)
Interaction— the variation varies
Heterogeneity— the variation varies
Risk factor— observed a correlation
Meta-analysis— calculated a weighted average
Collider— mass murderer of interpretable statistical analyses
Power calculation— effect size that matched budget
Exploratory— playing around
Replicated— did it again
Missing data— holes in the dataset
Measurement error— observe A, make conclusions about B
Stepwise regression— no idea what I am doing

• • •

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

Keep Current with Maarten van Smeden

Maarten van Smeden 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 @MaartenvSmeden

May 9, 2023
People often ask what it takes to develop a successful (clinical) prediction model

Here are TEN important things to avoid
1) Make sure you do not talk to domain experts. They only slow things down
2) Do not clearly define the intended use for the prediction model. It will make your model look boring
Read 11 tweets
Dec 25, 2022
This is my *top 10* favorite methods papers of 2022

Appearing in a single thread and in random order
Disclaimer: this top 10 is personal opinion. I am biased towards explanatory methods and statistics articles relevant to health research

Shameless plugs alert. 3 papers I co-authored but did not lead made the top 10
#1 When designing an individually randomized trial, what covariates should one plan to adjust for? This paper gives excellent practical guidance

Open access 👉…
Read 13 tweets
Apr 11, 2022
Yeah, that is not how it works
Apparently I need to update our myths about measurement error paper…
And if you really want to know about the effects of measurement error on results (conditional or marginal effect), this fabulous tool by @lindanab1 will help you…
Read 4 tweets
Mar 17, 2022
Medical Research Bingo Image
🧵with an explanation for each 👇
(posting this primarily as a note to self)
Unclear analysis aims - it often helps to think and clarify whether the (ultimate) goal is to predict, explain or describe. Thinking about analyses is much easier once this is clarified……… ImageImageImage
Read 19 tweets
Mar 12, 2022
The main problem with badly designed medical prediction models is not research waste or hampering scientific progress. It is the risk that someone takes the model seriously and use it to inform medical decision which can ruin someone’s life
There is no “hypothesis generating” or “too small, but maybe useful for a meta-analysis”. It’s building a tool, probably more that it is a scientific endeavor
Perhaps many of the published prediction models are not developed with the actual intention to eventually be used in (clinical) practice. We can and should do a lot better in flagging not-for-actual use clinical prediction models. And perhaps not publishing the fancy R-shiny app?
Read 5 tweets
Dec 16, 2021
This is my *top 10* favorite methods papers of 2021
Disclaimer: this top 10 is just personal opinion. I’m biased towards explanatory methods and statistics articles relevant to health research, particularly those relating to prediction models.

Shameless plugs alert. Two papers I co-authored (but did not lead) made the top 10
#1: (non-)collapsibility is one of these unintuitive phenomena that can confuse you for the rest of your career. This paper does an excellent job explaining

Open access 👉…
Read 15 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

Don't want to be a Premium member but still want to support us?

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

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!