Monday, July 10, 2017

How to Learn About Science Like a Scientist



Politics these days can feel like a war between those who believe in science and those that don’t. Many conservatives are ignoring the scientific consensus on climate change, while President Trump lives in a word of "alternative facts," yelling "fake news!" at any scientific results he doesn't like. But conservatives aren't the only ones who have trouble with science. No matter what side they're coming from, media reports about science almost never the whole story, and sometimes they get it totally wrong.

 As a scientist, this has been a pet peeve of mine for a while, so I was thrilled to find1 this excellent piece on Last Week Tonight, where John Oliver breaks down this issue:





If you're curious, here's the FiveThirtyEight piece mentioned in the video, and here's another FiveThirtyEight article with an interactive "p hacking" app - where you can use real data to "prove" that either Democrats OR Republicans are better for the economy!

Still A Problem


John Oliver's piece was from 2016, but bad science reporting is still alive and well. Maybe you heard about this back in June:



Did scientists really create a black hole? Of course not. Here “black hole” is a metaphor, like when we say “my phone is dead” or “Steph Curry is a machine at the free throw line” The actual study just found that when you take out some electrons from a molecule, it sucks in electrons from elsewhere in the molecule. You know what else “sucks” things? Black holes. Also vacuum cleaners. But "scientists create miniature vacuum cleaner" doesn't have quite the same ring does it?

We Can Do Better


Good science reporting is hard, but it's possible.

Here's a what some people were saying about one of my own studies:


That's not what I found at all. But the actual story by Tyler Cowen linked to in this tweet does better. Cowen says that I found that "attitudes about economic inequality don’t correlate very well with the desire for government to address it.”

That's still not exactly right. I was only looking at how people’s views CHANGE over time. At any given point in time being concerned about inequality is correlated with a desire for the government to do something about it. But when Americans become more concerned about inequality I found no evidence that they also become more supportive of government action. That's a subtle point, but it's important. And at least Cowen was in the ballpark. He didn't say I created a black hole!

An even better description of my results is by Dan Kopf at Quartz. Dan does an excellent job telling the story of my data without oversimplifying it. This is partly because he actually contacted me while working on the article, to make sure he understood everything, and that he accurately described my results. If more reporters did this, media reports about science wouldn't be so terrible.

In fact, Dan might have even done a better job talking about my results than I did! Dan's Quartz piece isn't as detailed as my own post about my results, but he manages to tell the basic story in fewer words, and using less technical language. That's something we scientists have trouble with, partly because we're trained to always be super precise in our language. And "super precise" can often equal "long" and "boring." But we've also spent so much time learning how to do science, that we often don't have a lot of free time left over to learn how to talk about it. So we need journalists like Dan Kopf to help us communicate our results to the public.

Don't get fooled!

 

When you hear that "A new study has found X", don't just take the media's word for it. Go beyond the Tweet or Facebook post and read the actual story. If there's a link to a press release or report, read that too. And do some googling to see if the result has actually been replicated. This might sound boring and time-consuming - but so is science.




Notes:

1 Thanks to Adam Daroff for telling me about this piece.

Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)