Calling Bullshit in the Age of Big Data

Paul Grant

Well-known member
Sep 8, 2004
3,180
0
CT
The University of Washington is offering a course in critical thinking.

Here is the link to the syllabus:

http://callingbullshit.org/syllabus.html#Fake

The course will require the class, in the first week to read Harry Frankfurt's piece "On Bullshit" in the first week.

https://www.stoa.org.uk/topics/bullshit/pdf/on-bullshit.pdf

In the second week the course delves into how to spot bullshit with the help of Carl Sagan and his piece, "The Fine Art of Baloney Detection."

There's a great bit on 99.9% caffeine free cocoa a pretty apolitical item.

http://callingbullshit.org/case_studies/case_study_caffeine_free.html

Week three is about the ecology of bullshit and it specifically calls out TEDtalks as being a particularly upscale source for it.

The fourth week explores correlation and causality.

http://callingbullshit.org/case_studies/case_study_traffic_improvements.html

Week five is dedicated to statistical traps with a brilliant piece about the lifespan of musician in various genres.

http://callingbullshit.org/case_studies/case_study_musician_mortality.html

Week six is about data visualization and week seven Big Data.

Week eight is about publication bias and why most published science results are false (http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124) and how false facts become established (https://elifesciences.org/content/5/e21451).

Week nine focuses on predatory publishing and the misconduct of scientists whereas the last three weeks deal with the ethics of bullshit, spotting fake news and finally refuting bullshit.

I wish a course like this was a requirement for all college students. It would go a long way towards enhancing critical thinking among the population. Please read as many of the links as you time.
 

Tugela

Well-known member
May 21, 2007
4,765
564
Seattle
When I taught statistics in grad school (at the University of Washington, go Huskies!) one of the required books on the course list was "How to Lie with Statistics". This was a more practical resource than the standard textbooks filled with Bayes theorem and multivariate regression analysis. I always taught stats like it was the Force: the discipline itself is not inherently "good" or "bad", but its powers can be used for a range of purposes. Statistics can provide profound insights into the nature of how the world works, however they can also deceive in convincing and subtle ways. Understanding the tool allows you to identify how it is being used. Glad to see the UW continues to promote critical thinking.
 

Paul Grant

Well-known member
Sep 8, 2004
3,180
0
CT
How does that saying go? "Statistics - If You Torture The Numbers Long Enough, You Can Get Them To Say Anything."
 

Paul Grant

Well-known member
Sep 8, 2004
3,180
0
CT
"Another big winner is Cambridge Analytica. Its board member Steve Bannon, "

Yet another new tangent to follow. Thanks for the post Walt.
 

WaltNYC

Well-known member
Mar 3, 2010
710
138
NYC
I suspect the use of 'big data' will become more prevalent over time. Trump/Bannon had the "first mover's advantage" and many others will follow. Eventually it becomes an arms race and the overall effect is deadened.