I’ve talked about how you have three basic moves in web investigations:

• Check for previous work
• Go upstream
• Read laterally

These can be used on simple claims (“Bernie Sanders shouted ‘Death to America’ at a Communist rally”) to get an answer quickly. But the real reason I like this set of moves is that they can be combined and chained together for more complex investigations.

To show that, I recorded my screen for 50 minutes while I looked into the claim that millions may die of cancer due to the Fukushima reactor meltdown. As I went upstream I found there was no there there. There was literally no source to this information. About 15 minutes into the research I decided to focus on the more empirical claim that the rates of thyroid cancer in Fukushima Prefecture were hundreds of times above normal.

The thing I find when I do these investigations is it is just these moves, chained together over and over. You go upstream for a bit to find that one route is a dead end. You come back to your original document and find another route upstream. You get upstream there, but laterally reading shows you the site has no authority. You go to Google to see if Google can get you closer to the origin of the claim. You find counter-evidence to the claim. You go upstream to find the source of that counter evidence. You read laterally to assess the counter-evidence. And so on.

Here’s the video, sped up by a factor of three and re-narrated to make it (slightly) less boring:

You can look at the resulting page. It’s a really drafty writing job, but it’s a wiki, so feel free to sign up, log in, and make it better. 😉

There’s a lot of domain knowledge I have here that an average student might not. I helped develop statistical literacy guidelines and taught a n introductory class on statistical literacy and health for years, so I already know quite a bit about issues caused by global screening for cancer. I recognize the journal Science as a giant in the field, and gravitate to that link in the Google results because of that knowledge. But those issues aside, what is most interesting to me is that a complex investigation looks like many simple investigations chained together. When you see that in a literacy context, it’s usually good news.