The Post-Statistics World

A wonderful long form essay by William Davies, "How statistics lost their power – and why we should fear what comes next"

The declining authority of statistics – and the experts who analyse them – is at the heart of the crisis that has become known as “post-truth” politics. And in this uncertain new world, attitudes towards quantitative expertise have become increasingly divided. From one perspective, grounding politics in statistics is elitist, undemocratic and oblivious to people’s emotional investments in their community and nation. It is just one more way that privileged people in London, Washington DC or Brussels seek to impose their worldview on everybody else. From the opposite perspective, statistics are quite the opposite of elitist. They enable journalists, citizens and politicians to discuss society as a whole, not on the basis of anecdote, sentiment or prejudice, but in ways that can be validated. The alternative to quantitative expertise is less likely to be democracy than an unleashing of tabloid editors and demagogues to provide their own “truth” of what is going on across society. [...]

In many ways, the contemporary populist attack on “experts” is born out of the same resentment as the attack on elected representatives. In talking of society as a whole, in seeking to govern the economy as a whole, both politicians and technocrats are believed to have “lost touch” with how it feels to be a single citizen in particular.

Some thoughts:

Some clear-cut questions can be adjudicated, purely based on observation, measurement, and statistics. For example, "how far is the sun?", "what is the average life-span of a human?", etc. Other questions, especially those that arise from inherently complex systems, often resist simple interpretation of numbers. These include questions about nutrition, ecology, macroeconomics etc. It is difficult to interpret measurements and facts, without an underlying theory or story.

Such numbers need compelling narratives to hang on. In such cases, a single counter-example doesn't disprove a thesis: a chain smoker who lives to be 100, doesn't disprove the claim that smoking is bad for you. Likewise, good narratives need numbers to ground them (think any pseudo-scientific claim). 

Calculations and stories go together; it is not one or the other. It has to one and the other.

The Need for Narratives

Neal Koblitz writes in the Chronicle:

The common element in all of this is knowing how to tell a story. Contrary to popular misconceptions about science and technology, a good piece of technical work is not a disembodied sequence of formulas and calculations, but rather is part of a narrative that has a long plot line and a large cast of characters. [...] Story-telling is a fundamental part of being human, from the time we are little children.

I couldn't agree more. The ability to weave a compelling story through a presentation or journal article makes a truly memorable one stand out from the run-of-the-mill kind.

You should check out the rest of the opinion for why STEM majors need grounding in the humanities.

An Adventure in Active Learning

I decided that I finally wanted to give this flipped classroom thing a shot. When I found out that I was scheduled to teach a Matlab/Mathematica class, a couple of years ago, I figured it was ideally suited for this experiment.

I spent the previous summer thinking through the format, reimagining the material, and the pacing. The idea was simple.

There were excellent 15-minute videos on the Mathworks and Wolfram sites, which students would watch before class. They would take a quick, but super-easy, quiz at the start of every class to ensure that everyone watched the videos. We’d then spend class time doing actual modeling, programming, and discussing common pitfalls and misconceptions.

I was very excited at the start of the semester. If I were a student, I would have loved the class I put together. Without a doubt. That's what I thought!

The excitement drained away quickly.

About 1/3 of the class was super-engaged. They watched videos, they were active in class, and I felt that they truly got something out of the class. If all of my class did as well, I would have thought that the experiment was successful. Grudgingly though, I had to admit that this 1/3 would have mastered the subject, even if I did not show up to class.

The middle 1/3 tried to keep up. They were somewhat inconsistent. In some classes, they were very active, and in others, they struggled. Would they have done better in a traditional class? Who knows!

About 1/3 of the class did not watch the videos. Consistently. I exhorted them, unsuccessfully, to come see me after class, so that we could work on the gaps in their understanding. Class time was miserable for them. They hid behind the terminals, doing their homework, while the rest of the class was busy solving problems.

After about a month I realized I had completely lost them.

Overall, I felt terrible about the experiment. It failed.

I haven’t tried flipping my classroom, since.

I have tried to do several post-mortems to figure out what went wrong; if there was something I could have done differently to fix the problem I saw, fairly early on.

The whole process relied heavily on students doing their homework. I was open to them not getting all the material in the videos. We could discuss that stuff in class. That would have been wonderful.

But flipped classrooms are still relatively rare in my university. Their novelty meant that students hadn’t realized the importance of keeping up with assigned material.

This was an undergrad class. The distribution of “work ethic” is often wider than in grad classes. From my experience flipped classrooms can work well even with a wide distribution in ability, given relatively high appetite for hard work.

But, vice versa, is another story.