Getting by on the day after
Tap, tap, tap… is this thing still on? It’s been a long time since I posted anything here, but it’s been a rough day, and I thought maybe firing up the old blog would be a cathartic release.
Like many of in America, I stayed up late last night and watched the results in total disbelief. No one saw this coming, how could we have missed this? I spent my Sunday canvassing voters in Philadelphia, I took my daughters to vote on Tuesday, and when she pressed the button to vote for the first female president, I was certain we would win.
As the results proved otherwise, the questions started popping up on Facebook and Twitter—what will I tell my children? What will I teach my students? And I watched as some really great answers came across my feeds—here’s a great piece on how to talk to kids, and @bowmananimal has one of the very best stats lessons I’ve ever seen focused specifically on how we could have missed this (seriously, if you teach stats, this is gold).
Once I figured out what I was going to say to my 6 year old daughter, I turned to my physics class, and struggled to think about what we were doing in there. How can we just be figuring out the energy momentum relation when it seems like so many more important things are happening around us.
Then I saw these tweets from Ben Lille, director of the awesome StoryCollider podcast
One clear thing about the polling failure: science/data ppl failed in the way they usually fail, by vastly underestimating the uncertainty
— Ben Lillie (@BenLillie) November 9, 2016
And I saw this post on the http://scienceblogs.com/startswithabang/2016/11/09/35381/
In our class, we talk often about how important it is that you tell the full story of every number that you write, and this summer, I saw someone write about how every number you write in a science class represents a measurement, and when you write a measurement, you want to communicate 2 things:
- What you’re measuring—this is why we include units
- An idea of the uncertainty of the measurement
I decided to put together these ideas into a short slideshow to share the idea of measurement uncertainty and systematic uncertainty, and how even when you’ve got poll with a large sample size and a very narrow measurement uncertainty, your poll might still be invalid if one of the assumptions that underly the poll (like that every demographic of voters will turn out in equal numbers). And this is true of experiments as well—If you read graduated cylinders from the top of the meniscus, then all of your volume measurements will be slightly too large—and that’s why scientists work incredibly hard to identify and correct for systematic errors.
Lastly, we talked about inspiring Sagan quote:
Science is not perfect, it’s often misused.
It’s only a tool, but it’s the best tool we have.
I tried to add some nuance to this quote by talking about how science is a tool used by humans, and we bring a lot of baggage to science. We have unconscious biases, and these biases who gets to be a scientist, what experiments we choose to conduct, and even the data that we take. And sometimes these biases, mixed in with racism and sexism lead to some really horrible things done in the name of science. Ben Lille’s Science Needs a New Ritual is a great 2 page primer on this.
And then we did some physics problems, because there can be great joy in thinking about ideas, because unlike many of the other challenges we are now facing, these problems had solutions.