I think the idea of having some examples of relevant, high quality, punchy data journalism is great. Such examples are few and far between. Unfortunately, what I clicked on looked, to me, pretty disappointing (journalism, not software). Example: looking at their top link
It took me a while to figure out what exactly are they showing: the plots/colors look good, the software works cool, but it took me a couple of minutes to figure out what this is a plot of (area, at population density at your chosen location, needed to house X million people).
Maybe all they need are better titles / verbal descriptions, but as is it does not cut it for me as best practices of data journalism (interesting, correct, clear, grabs me by the balls). My 2c.
Perhaps you can help in discerning what is and isn't journalism since they stated that contributors are welcome. That first link turned me off as well.
Neat examples. Still a dearth in statistical aptitude to cross before "data driven" journalism becomes more than nieche market or circus, but every step counts.
Data journalism is such a broad umbrella that it's hard to agree how mainstream or niche it is. I think David Leonhardt (former editor of the NYT's Upshot) had it absolutely right [0] when he said "data journalism" is as dumb and meaningless a phrase as "word journalism". Though perhaps the fact that journalists are OK with calling it "data journalism" as a means to differentiate it from whatever normal journalism is, is a testament to how niche the concept of using data is in journalism.
That's a No True Scotsman way of looking at it. Data journalism is about making a justifiable, persuasive argument with evidence, and for some arguments you don't need p-values everywhere.
Granted, some news sources tend to use data journalism as a linkbait strategy ("In One Chart!") to make arguments which are hard to impeach, which is a separate issue.
I didn't mean it in terms of p values. Just an ability to reason from data. Statistics is not statistical inference. What does the graph actually show? Is it confounded? How was the data collected?
Everyone likes a bit of graph porn because it's associated to rigour. But clothing not maketh the man and all that.
http://projects.aljazeera.com/2013/syrias-refugees/
It took me a while to figure out what exactly are they showing: the plots/colors look good, the software works cool, but it took me a couple of minutes to figure out what this is a plot of (area, at population density at your chosen location, needed to house X million people).
Maybe all they need are better titles / verbal descriptions, but as is it does not cut it for me as best practices of data journalism (interesting, correct, clear, grabs me by the balls). My 2c.