Some interesting research on supposedly random and reliable social statistics on the Daily Kos

Here: A bit over two weeks ago, a group of statistic wizards (Mark Grebner, Michael Weissman, and Jonathan Weissman) approached me with a disturbing premise -- they had been poring over the crosstabs of the weekly Research 2000 polling we had been running, and were concerned that the numbers weren't legit.

I immediately began cooperating with their investigation, which concluded late last week. Daily Kos furnished the researchers with all available and relevant information in our possession, and we made every attempt to obtain R2K's cooperation -- which, as I detail in my reaction post here -- was not forthcoming. The investigators' report is below, but its conclusion speaks volumes:

We do not know exactly how the weekly R2K results were created, but we are confident they could not accurately describe random polls.


This cannot be a design inference, right?

Interesting design inference concerning a historic photo

The cloud patterns in two photos taken during the Spanish Civil War are identical, according to a column by George Will:

In a slightly less dramatic photo of another falling soldier, taken by Capa at the same time - the cloud configuration is the same as in "Falling Soldier" - the soldier falls on the same spot.

The interesting thing is, don't blame photoshop; these pix were created in 1936.

Will goes on to note, quite properly, that photographer Capa had an honourable career as a war photographer (a highly dangerous profession), which came to an abrupt end when he stepped on a land mine. But it seems likely now that he manipulated an iconic photo.

For the evidence base I'll go with the cloud patterns, whose configuration is the same most likely because they had been captured once on film. A design inference.

Added: Commenter Voice Coil at 1 below gives a more expert account of how the photo was probably staged/faked. Note: When I make a design inference, I don't necessarily know how it was done; it's the statistical unlikelihood of chance that prompts investigation. Reuters bought doctored stuff for quite a while before the penny dropped, as have other news outlets.

Why did Capa do it? First, people have only recently begun to be fairly certain that he did it, and he has been dead a long time, so I can only hazard a guess. Will says he was a man of the Left, so I surmise he probably wanted to help his own side by producing an unforgettable photo. But, as every photographer knows, unforgettable photos cannot be produced just when needed. Lots of people hated Franco, as he did, so there probably was no great incentive to doubt or fact-checking.

Further to a friend's comment on how intelligent design is applied to crime detection ...

I've been rummaging my files for this design inference, and finally found it:

Excerpt from a recent Canadian Broadcasting Corporation report

The CBC report (October 25, 2006), which aired Wednesday night, says of the roughly 60,000 lottery ticket sellers in Ontario, retailers won nearly 200 times in the past seven years, with an average prize of $500,000.

A statistician with the University of Toronto called those numbers a statistical anomaly, saying there is a "one in a trillion, trillion, trillion, trillion" chance of that many retailers winning. Dr. Jeffrey Rosenthal said the number of wins should be closer to 57. In fact, an investigation was held and a number of persons were later charged.

Meanwhile, Michael Josem (for whom see here) writes to disclaim any association with intelligent design, and to insist that the sheer number of stars in the universe makes the origin of life more plausible than poker cheating.

Huh? I think Josem better stick to ratting out poker cheats, and go Josem! His note is appended, below the book ads.

Meanwhile, a legend in his own lab wrote me to announce that he is "a scientist, thus trained in evolutionary thinking, but as a christian wondered whether intelligent design had anything to offer." Apparently, he decided to try to find out not by reading Michael Behe's Edge of Evolution or Bill Dembski's No Free Lunch or Mike Gene's Design Matrix or anything challenging like that.

No, he decided to hang out at this news and culture blog, so he could decide there was nothing in it without engaging in any serious science thinking. I wonder what he really does for a living?

Just for the record, I am a journalist, not a scientist, and this is a news and culture blog. Years ago, I realized that most of what you'd hear in legacy media about the intelligent design controversy is not written by anyone who follows ID as a beat. So I decided to just start following it as a beat.

I am now about 1000 news stories behind, but fortunately, the first seven hundred have been reclassified as "recent archaeology." Meanwhile, many thanks to generous donors at the PayPal button!







From Michael Josem:

I saw your blog post about chance, online poker and intelligent design.

I think that your argument that this case of online poker cheating has any relevance to your argument about intelligent design is incorrect.

My claim with the poker data was simply to highlight one of two possibilities - either this poker player's run of hot luck was one of the luckiest streaks of luck in the history of the planet, or that there was cheating happening here. That is clearly sufficient proof for further investigation, which resulted in the cheating being uncovered.

The same principle can be applied to the creation of life - either it was incredibly lucky, or there was some intelligent design by a greater power. Neither option appears to be disproved by the data.

Now, in the online poker example, there are only a few million players, and thus, a freak result is still incredibly unlikely. By contrast, there are many, many, many, more stars in the universe than there are online poker players. This would mean that such an event (even at 15 standard deviations) is more likely to happen over the whole universe of possible events.

I hope that you'll be able to correct your article with this clarification.

Regards,
Michael Josem