Tell us you don't know how to do #math without saying it.

#FakeNews

In an amusing twist of #SelectionBias, the death rates were 11% lower than even pre #COVID normal life expectancy for vaccinated people. Probably because people who take #vaccines tend to be more informed about #health #risk.

The #death rate for unvaccinated people was 93% higher than the pre-COVID ones. The risk profile of people who don't have the skills to "do their own #research" is frankly #insane.

More in comments:

3/
Also, we wrote a non-technical introduction to #selectionbias due to conditioning on a #collider
https://bmj.com/content/381/bmj.p1135

As Jonathan Sterne and colleagues say, if you understand “Among successful actors, being physically attractive is inversely related to being a good actor”, you understand collider bias.
https://www.nature.com/articles/s41467-020-19478-2

@der_raDDler oh! 😱 Das ist ja wie hier bei den viel zu lauten Fahrradstraßengegnern, bei deren Umfrage sich letztendlich nur rd. 10 % dagegen, d. h. im Klartext für die Beibehaltung der Autoparkplätze ausgesprochen haben. Die anderen haben bspw. den Fragebogen gar nicht zurückgeschickt.

Ist ein großes Problem.
#FalseBalance bzw. #Stichprobenverzerrung aka #SelectionBias (soweit ich das richtig verstehe).

But to be clear: I am not against clever puns in the titles. I simply believe that people who poke at #editors complaining about humorous titles don't understand that this is #SelectionBias at work. The titles you see published are those that survived editorial work. if you could see all titles the editors see, you'd understand how bad sense of humour of many researchers is, and that giving you paper a silly title is a form of intellectual torture.

https://youtu.be/u9dQwsSc5NU #astronomy #BlackSwans #probability #SelectionBias

https://youtu.be/uZRDONE4zng #astronomy #selectionbias
#fermiparadox

@gbhnews good example of #SelectionBias

Some company: "We want your feedback!"

Also some company: sends link to a multi-question mandatory-response questionnaire

No, you don't want my feedback then, do you. If you wanted my feedback, you wouldn't constrain it in that way. #SelectionBias

Defining and evaluating the #HawthorneEffect in primary care: systematic review & meta-analysis https://pubmed.ncbi.nlm.nih.gov/36425097/

Aware or unconscious #behaviorchange in a study environment, related to the interaction of: #selectionbias; commitment and congruence bias; conformity and #socialdesirabilitybias; and, observation and measurement #bias." It overlaps with the #placeboeffect, and regression to the mean. Shifts Rx fx by avg >40% in binary outcomes, less in well-designed studies. #psychology #RCTs

@summer To “make something of yourself”, you need several things:
A little ability.
A little effort.
A tiny opportunity.
One or two connections.
But mainly A HUGE AMOUNT OF LUCK.
All these coaches and speakers who say, “keep trying, never give up,and you will achieve your dreams” are living in fantasy land. They conveniently ignore all the people who keep trying, never give up, but still don’t achieve their dreams because they are denied by an uncaring world. #SelectionBias

Interested in selection mechanisms/bias and their representation as DAGs (directed acyclic graphs)?

I can wholeheartedly recommend this article from Louisa Smith entitled "Selection Mechanisms and Their Consequences: Understanding and Addressing Selection Bias"

https://www.louisahsmith.com/publications/selection-mechanisms-and-their-consequences-understanding-and-addressing-selection-bias/selection-mechanisms-and-their-consequences-understanding-and-addressing-selection-bias.pdf

#EpiVerse #Epidemiology #CausalInference #SelectionBias #datascience

4/
The key point:

Explicit #TargetTrial emulation prevents mistakes in observational data analyses that we wouldn't make in analyses of randomized trials.

That we can do something with the data doesn't imply that the result is causally interpretable.

More here:
https://www.nejm.org/doi/10.1056/NEJMp2113319

If you're looking for a gentle introduction to #causaldiagrams, #selectionbias, and colliders
enjoy our free course
"Causal Diagrams: Draw Your Assumptions Before Your Conclusions"
https://www.edx.org/course/causal-diagrams

#xkcd: #SelectionBias

https://xkcd.com/2618/

@klimavoracht
Jeder is käuflich, näh? Obwohl, bei wie vielen wurde das eigentlich überhaupt versucht? 🤔

#SelectionBias

?

Znam wielu ludzi, którym szczepionka zaszkodziła, więc wiem z pierwszej ręki, że szczepionki szkodzą.
#selectionBias #falusyLogiczne

"When making decisions, an agent needs to condition on all the evidence it has observed"🤦‍♂️ #causaltwitter #causality #colliderbias #selectionbias