Others mentioned additional matters worth pondering:Interesting, but I'd like to see scatter plots (because of https://t.co/mBlDYJQX1q Relationship may not be linear) pic.twitter.com/i8HGr8Feu4— Alberto Cairo (@albertocairo) March 15, 2016
Really cool, I agree. After that, NYT's Gregor Aisch replied:not to mention ecological fallacy, but still! Correlations! In the newspaper! Cool! https://t.co/eNe2qOv0OC— Robert Grant (@robertstats) March 15, 2016
I think that Gregor took the first comment as a complaint, although it was a mere suggestion: Wouldn't it be possible to make both data and even models available in a single place —perhaps as a sidebar— rather than forcing dorky readers to find them on their own, in case they wish to explore the evidence behind the story and the table? The article itself mentions those data and models, so why not disclosing them?@albertocairo what's holding you back? census data and election results are public..— gregor aisch (@driven_by_data) March 15, 2016
ProPublica and FiveThirtyEight, have been doing that on a regularly basis, and they often explain their methodology (see also this, by Illinois's Daily Herald. UPDATE: Or this by the NYT itself, via Gregor) I think that there's plenty of discussions to be had about how we journalists and news designers can be more transparent and mimic scientists a bit when doing analysis and data-driven stories.
UPDATE: Jeremy Singer-Vine, data editor at BuzzFeed, has told me that they publish the data and analyses for their projects here. Please also read his article ‘What we’ve learned about sharing our data analysis: Publishing reproducible data that’s genuinely useful.’
Below are my favorite paragraphs, which summarize the point I intended to make in this post better than I ever could: the key isn't to disclose your data for some major projects here and there; the key is to do it more systematically, disclosing it for most —if not all— of them:
At BuzzFeed News, our main motivation is simple: transparency. If an article includes our own calculations (and are beyond a grade-schooler’s pen-and-paper calculations), then you should be able to see—and potentially criticize—how we did it.
And that holds us accountable. Indeed, the very prospect of public scrutiny forces us to be as lucid and straightforward as possible. It discourages us from cutting corners. It lights a fire under our proverbial posteriors, and improves our work.