So perhaps we should think more carefully about the process by which anecdote is transformed into data and information. We might break it down into four rough steps... the collection of data or evidence.... organization.... explanation.... generalization.....
March 17, 2014
"The point is that data journalism isn’t just about using numbers as opposed to words."
"To be clear, our approach at FiveThirtyEight will be quantitative — there will be plenty of numbers at this site. But using numbers is neither necessary nor sufficient to produce good works of journalism...."
Subscribe to:
Post Comments (Atom)
Posts
23 comments:
Spinning with numbers.
Lies, damn lies, and statistics.
Categorization battles.
OK. But the question still comes up - what data to collect - as the story itself says. It talks about how data begins in anecdote - something is seen and then data is collected to turn that anecdote into something meaningful. So start with the wrong anecdote and you'll collect the wrong data. Look at the disputes on this blog about what is happening with Obamacare - the answer is in the data but the anecdotes are driving our mental data collection, the data we will or won't accept. Look at the search for the Malaysian plane - when the answer is found I bet it will turn out that the answer was there all along in data no one looked at. Look at Putin - why is everyone surprised? Don't we gather intelligence from phones etc? But the anecdote was that Russia was weak so all the data was ignored. Probably Russia is weak - just that Obama is extremely weak and Putin realized it while Democrats denied it. So my point is that I think that a traditional education gives you an unexcelled outline within which to ask questions of data. If you know history you know what will happen when a vain weak leader like Nero or Obama wants to do over a city or a society, when a political leader thinks he's an artist and concerns himself with the fashion in jeans.
Lenin, Hitler, Stalin, Roosevelt, and Mao would have had an easier time engineering their societies if they had people like Nate Silver working for them.
Lenin, Hitler, Stalin, Roosevelt, and Mao would have had an easier time engineering their societies if they had people like Nate Silver working for them.
Smart take.
I read Silver's book and thought it was the best general interest argument for the value of statistics and statistical thinking that I have ever read. The web site seems dull. In a world where the Daily Mail online is the leading online newspaper, as assessed by traffic, I am predicting, with 80% certainty, that this venture of Silver's fails.
Oh geez, I get props from Robert Cook and garage mahal this week.
I think I should hang myself from a doorknob.
Scott said...
Lenin, Hitler, Stalin, Roosevelt, and Mao would have had an easier time engineering their societies if they had people like Nate Silver working for them.
Why don't you expand on this and tell us what terrible things Silver has done?
Accurately forecasting the last presidential election does not count as a terrible thing.
Valid, clear statistics generally don't help to support librul arguments. For example, for years, the official wisdom was that there are 45-50 million uninsured. Well, where are they, so far fewer than 5 million have signed up for Obamacare and many of them were not uninsured. So should we start to question the official wisdom of 45-50 million uninsured?
Data is important, but clear, measurable goals are even better.
Hey, Werner von Braun didn't do anything bad either. He was just excellent at his technical specialty.
Gather round while I sing you of Wernher von Braun,
A man whose allegiance
Is ruled by expedience.
Call him a Nazi, he won't even frown.
"Ha, Nazi Schmazi," says Wernher von Braun.
Don't say that he's hypocritical,
Say rather that he's apolitical.
"Once the rockets are up, who cares where they come down?
That's not my department," says Wernher von Braun.
...
Scott said...
Hey, Werner von Braun didn't do anything bad either.
You still haven't explained how Silver is really Hitler's factotum.
Why don't you expand on this and tell us what terrible things Silver has done?
Poured cold water on the narrative that Romney was going to win.
"You still haven't explained how Silver is really Hitler's factotum."
As Alex Lester might say, "There goes the claxon of pedantry!"
If you're misusing the word "factotum" do I have to answer your question? Restate the question in English.
Here's a clue for you all: The walrus was Paul.
And if one can statistically divine the moods and actions of large groups of people, a skillful leader can oppress them much more easily.
MEGO
" AJ Lynch said...
Valid, clear statistics generally don't help to support librul arguments. For example, for years, the official wisdom was that there are 45-50 million uninsured. Well, where are they, so far fewer than 5 million have signed up for Obamacare and many of them were not uninsured. So should we start to question the official wisdom of 45-50 million uninsured?
3/17/14, 9:23 PM"
There's a difference between valid statistics and making shit up.
And if one can statistically divine the moods and actions of large groups of people, a skillful leader can oppress them much more easily.
Well thank the heavens that Hitler, Stalin, and Mao didn't have statistics. We really dodged a bullet there.
"So should we start to question the official wisdom of 45-50 million uninsured?"
Maybe. Or maybe we should question the assumption that all those people are uninsured but would buy insurance if they could.
Maybe most of them just don't care that much.
Aside from thinking innumerately, the biggest flaw is assuming everybody else thinks just like us.
One of the clear, overriding messages that comes from data analysts like Nate Silver (or Tom Tango on baseball) is how LIMITED statistical interpretations are. If the only thing you know about statistics is the problem of small sample sizes, you're LIGHT YEARS ahead of the average pundit.
I've skimmed some of the stories at 538, mostly in the sports section. It's a mixed lot. The hardest part of data journalism isn't assembling data. It's asking interesting questions. Asking interesting, counter-intuitive questions is what made Bill James great, not his statistical chops. There are thousands of sabremetricians with more statistical ability than Bill James. But James has a kind of fool's ability to see through convention. Convention actively irritates him. Nate Silver needs more of that.
At 538, Neil Paine's article on cornerback stats posed and answered an interesting question. But Bill Barnwell at Grantland is putting out an interesting data-journo-column every week. 538 has some catching up to do.
Asking interesting, counter-intuitive questions is what made Bill James great, not his statistical chops...Convention actively irritates him. Nate Silver needs more of that.
Given my background I should be a major Silver fanboy, but this is the thing that bothers me most about him.
As just one example, he can write 2,000 words on why the polling data says the Supreme Court should approve of gay marriage.
Meanwhile, the 2,000 words on why the polling data says Obamacare should be dumped are conspicuously absent.
Post a Comment