Total Drek

Or, the thoughts of several frustrated intellectuals on Sociology, Gaming, Science, Politics, Science Fiction, Religion, and whatever the hell else strikes their fancy. There is absolutely no reason why you should read this blog. None. Seriously. Go hit your back button. It's up in the upper left-hand corner of your browser... it says "Back." Don't say we didn't warn you.

Friday, September 12, 2008

Can it be? The end of the Lenski Saga?!

As many of you probably recall, I am a longtime follower of the unique brand of whatthefuckery provided by Andrew Schlafly's collaborative blog Conservapedia. As such, I have long been watching the bizarre saga of Schalfly's absurd effort to do battle with top notch scientist Richard Lenski. Most recently, Schlafly drug this insanity out by sending a letter to the Proceedings of the National Academy of the Sciences, otherwise known as PNAS. The letter alleged, among other things, that Lenski is an incompetent statistician and a fraud.

Well folks, PNAS has responded, and it's a beaut.

My first warning of the reply came from Conservapedia's mainpage, which doesn't mince words:

Or, in simple human terms:

PNAS refuses to address the 5 errors in the Lenski study identified by the Letter to PNAS, and hides behind anonymity in censoring the letter from the PNAS readership. See PNAS Response to Letter.

Needless to say, I was intrigued, and clicked the supplied link. What I found is, supposedly, the text of PNAS' response:

Or, in simple text:

The PNAS has refused to publish the Letter to PNAS. The PNAS has provided the following non-responsive explanation dated September 11, 2008.

Note how the PNAS hides behind anonymity to justify its failure to address the five errors identified in the Letter to PNAS:

"A member of the Editorial Board has evaluated the letter and concluded that PNAS cannot publish it for the following reasons:

From what I take to be the underlying issue from the numbered points, Mr. Schlafly's main concern has to do with the fact that one experiment failed to yield a statistically significant result, and this happened to be the experiment with the largest sample size. Every experiment has limited power to detect a difference of any given magnitude, and so in a series of experiments some may yield non-significant results even when the null hypothesis is false. The non-significant experiment may even be the one with the largest sample size. There is nothing exceptional in this--it is a matter of chance. Nevertheless, from a statistical point of view, it is proper to combine the results of independent experiments, as Blount et al. did correctly in their original paper. If the overall result is significant, as it is in this case, then the whole series of tests is regarded as significant. Mr. Schlafly seems to suggest that experiments differing in sample size cannot be combined in an overall analysis, and if this is what he is suggesting, he is wrong.

I think Letters published in PNAS should raise points that in themselves, or in conjunction with the authors' response, should be of wide interest to the readership of PNAS or should illuminate some obscure or subtle point. The issues raised by Mr. Schlafly are neither obscure nor subtle, but are part of everyday statistical analysis at a level too elementary to need rehearsal in the pages of PNAS.

Mr. Schlafly's final comment about release of data is uncalled for. My understanding is that the authors have made the relevant materials available on their web site. This seems to me to meet the requirement that "data collected with public funds belong in the public domain." If Mr. Schlafly believes that the disclosure is incomplete, that is an issue that needs to be argued with the original funding agency, not with the readers of PNAS."

This response can be discussed at Talk:PNAS Response to Letter [links original]

So, in essence, the reviewer makes five points. First, he concedes that one of Lenski's tests was non-significant but observes that even if the null hypothesis is false (i.e. your research hypothesis should be accepted) our statistics will sometimes fail to recognize it. This is known as "beta error" or "type two error" and is a well understood part of statistical analysis.

The second point is subsidiary to the first, wherein the reviewer observes that when multiple analyses are combined mathematically and the result is significant, then the entire series is generally regarded as significant. There is nothing inappropriate about this practice so long as it is done properly and, indeed, Schlafly failed to give any good reasons to believe otherwise.

Third, the reviewer comments that Schlafly is incorrect if he believes that samples of different sizes cannot be combined in a single analysis. Indeed, if this were true, techniques like the independent means t-test would be in trouble. Granted, what Lenski did is more complex, but the concept is sound.

Fourth, the reviewer basically indicates that for a letter to be publishable in PNAS it must have an interesting or meaningful point. Schlafly's letter does not and, indeed, the objections it raises show nothing more than Schlafly's lack of basic statistical expertise. Therefore, publishing Schlafly's letter would be akin to printing undergraduate statistics homework in Nature.

Finally, the reviewer indicates that the data are already available and that, if Schlafly disagrees, it is in any case not PNAS' job to address the situation.

In short, this is a brief but devastating rebuttal of the usual Schlafly insanity. Not that he, or any of his acolytes, show any signs of realizing it:

Again, in plain text:

In this day and age, scientists have their own agenda and have corrupted science. Just look at global warming or cloning or stem cells as proof. With that said, the only way to get the real truth is by suing in court. Unfortunately, scientists are bound to vast wealth and have the power to defend themselves vigorously. If ever a fund was set up to pay for a suit, I would contribute. It is a classic case whereby the truth be known, the truth will prevail. -- jp 22:14, 12 September 2008 (EDT)

Thanks, Jpatt. One additional beauty of the truth is that it remains the truth no how much some deny it. PNAS can deny its errors all it likes, but that doesn't change the fact they are errors.--Aschlafly 22:21, 12 September 2008 (EDT)

And because I'm a sucker for irony, I just had to get involved:

In regular language:

"...the only way to get the real truth is by suing in court." Which is why the Kitzmiller et al. vs. Dover Area School District et al. decision was so important. Science was able to vindicate itself against the claims of intelligent design creationism. As for PNAS: It's not like the reviewer is saying anything that commenters here hadn't told ASchlafly already. -Drek

Place your bets on how fast they ban me and revert my edit. If there's one thing Conservapedia can't stand, it's the bloody truth.

Good night, kiddies: my wife is growing impatient* and this particular endeavor is futile.



...and reverted... under two hours. I'm almost kind of proud.

* And if there's one thing you don't want,** it's for the woman you love to be cross with you.

** I leave it as an exercise for the reader to decide what the other things I do not want are.

Labels: , , , ,


Anonymous Dr. Pablito said...

Ha. Nice blogging. I loved the bit on the Conservapedia talk page about how we have to leave science to the courts. Seriously? Wow. Anyhoo, good jibe at them. Don't worry about being banned. Wrestling with a pig and all.

Saturday, September 13, 2008 11:44:00 PM  
Anonymous Argon said...

Andy's ego has got to be smarting to wipe your comments that quick. Exactly what was he expecting, a reward from PNAS for 'exposing' fraud with his 'five errors' that any bacteriologist who actually read the paper would recognize as BS?

Kudos for DinsdaleP for enabling Schlafly to shoot himself in the foot again. That was masterful.

Prediction: Pure bitching, perhaps a letter to a Congressman, and no attempt to actually reproduce the experiment.

Sunday, September 14, 2008 3:47:00 PM  
Blogger Practicing Idealist said...

This is a very nice post. Your wife, despite the crankiness, should be proud.

Monday, September 15, 2008 9:34:00 AM  

Post a Comment

<< Home

Site Meter