You know a course will be off to a good start when the intro reads "The world is awash in bullshit…". "Calling bullshit in the age of big data" is taught by Carl Bergstrom (biologist) and Jevin West (information scientist), two professors at the University of Washington. In case you didn’t get the memo, the course kicked off in Spring 2017 and is fully available online for free [1]. You can continue at self-paced tempo, as the lectures are divided into five or six parts of 3-12 min each.
 |
| Source: Twitter, used with permission |
Combating New-School Bullshit
Allow me to elaborate a bit on why the course is totally worth your time. It begins with an introduction to bullshit, its philosophy, interesting examples and measures to refute it. Hilariously, during one point in the first lecture the professors called bullshit on their own bullshit by butchering the claims that they saw explosive growth in the course website’s visibility once it was up-and-running. This was all in a bid to explain "new-school bullshit" whereby people try to overwhelm others by fancy graphs. Aside from explaining statistical traps and trickery and big data gimmicks, the course offers useful ways to debunk bullshit even for the lay audience. Beware though: certain parts such as replication crisis and predatory publishing etc., may sound like a (very interesting) repetition of themes from the lab.
Ducks and Glass Slippers
There were certain parts that made me laugh long after I had finished the course and was reviewing my notes. First, the concept of data "ducks", coined by Edward Tufte. A duck is any graphic so heavily decorated with design elements that it distracts the reader with its visuals rather than communicating meaningful information. Another one was ''glass slippers'' (inspired by Disney’s Cinderella), which refers to forcefully trying to mold your data to a standard model.
Importantly, the professors defended science while exposing all the bullshit that comes with practicing it. They argued that science is not bullshit even amid a big reproducibility crisis because most of the hypotheses researchers come up with are at least plausible. Moreover, there are multiple ways to deviate from the null hypothesis. Science and replication are cumulative and conflicting results help to drive the field forward by compelling researchers to think of alternative explanations.
For those of you who just had a feel-good rush after reading that science is not completely a recipe for disaster, let’s go to the sobering part. The professors laid bare the factors that motivate scientists. And no, they don’t talk about the quest for the unknown or saving the humanity (get real!). Their premise is that just like all the other fields, people in science respond to incentives. Scientists are often driven to research for recognition [2], prestige and, most importantly, grant money and job security. Furthermore, Goodhart’s law (1975) [3] and Campbell’s law (1979) [4] also state that when a type of measure (number of citations in elite journals, for instance) becomes a target for researchers, it is no longer a good measure.
Practice fact-checking online
Lastly, much to my delight, the professors addressed the hotly debated issue of fake news as well. Money is the driving force behind fake news. Creating fake news stories is one thing but why do people spread them? Many factors contribute to this fast spread of bullshit, such as laziness to do some serious fact-checking, self-validation and sharing links just to increase social visibility. Social media is a free resource that gives everyone the opportunity to create and disseminate bullshit. However, it becomes a problem of even greater magnitude when reputable media also become an accomplice. Their aim is to take the lead in breaking the news and triggering people, by any means necessary, towards click-bait.
How to Call Out Someone on Their BS Effectively
So where does that leave us? As scientists, we have an even greater responsibility not to get carried away by the cancerous trend of sharing for the sake of social visibility. In order to efficiently call out a skeptical claim on its BS, collect all relevant information, double-check your facts, and run them by someone who knows more about the subject matter. If you err, admit your mistake, and admit it fast. Refute bullshit by tackling the core facts, spend less time explaining the dubious ones, and keep your argument simple – or else your approach may backfire.
Fun, engaging and equal parts thought-provoking, "Calling bullshit" is a timely course for young would-be scientists as well as lay people. It is neither too long nor too predictable content-wise. You can follow the course on Twitter here: https://twitter.com/callin_bull
[1] http://bit.ly/2vVpa7v
[2] http://bit.ly/2vVmuXn
[3] http://bit.ly/2vVcIoa
[4] http://bit.ly/2vVnQRX
by Zara Khan, MSc Student MedNeuro
This article originally appeared September 2017 in CNS Volume 10, Issue 3, Spirituality in Science
No comments:
Post a Comment