Some of the most depressing stats I've ever read.

D. Freedman, Evaluation Review, December 2006:
"Experimental data are often analyzed by fitting regression models and their ilk. As shown in section 10, randomization does not justify such models. Experimental data should therefore be analyzed first by comparing rates or averages, according to the intention-to-treat principle. Only then should models be deployed. It must be emphasized that statistical adjustments to experimental data often depend substantially on assumptions, not just on randomization."

Comments

Post a Comment

Popular posts from this blog

The Adze Picket Co. (I Love the PCT)

Image
Me at the Border I spent the weekend at the ADZPCTKO with Nitro , Tatu Joe, Ducky (the Wise) , Tomato, BearCan't , Mattress, Love Barge, Nano, Sly, Disco , WeatherCarrot, Too Obtuse, Nacho, Tall Paul, Mags , and Wildflower (AKA Icognito Cheeto Bandito). I got a ride down with Firefly. I'm tempted to just end my post there, as it would seem like complete gibberish to many of you. But I won't. The ADZPCTKO is the Annual Day Zero Pacific Crest Trail Kickoff, held the last weekend in April (also known as NFL Draft weekend, where, miracle of miracles, the Redskins, despite Dan Snyder's best attempts to screw it up, actually managed to avoid throwing away multiple draft picks on Chad Johnson and instead traded for more picks. Hallelujah!) at Lake Morena campground, 20 miles north by trail from the Mexican border near Campo, CA. Conventional wisdom says that the weather window on the PCT is such that there's not much point to starting earlier thanks to snow in the S...
Read more

Hardrock 34:42

Image
It's Tuesday, I'm back in Berkeley, and I just ran my first Hardrock last Friday/Saturday in 34:42. I wanted to record thoughts immediately after the experience, because it was hard , and now that it's been a few days, I'm already forgetting the pain. [I don't want kids, so you'll have to insert your own reference to childbearing here if you so desire.] Start The race started at 6AM. I slept reasonably well the night before in my truck just behind the school gym where the race starts--Silverton is very low key, so it's easy to dirtbag inside the city limits. I did, however, have a bit of GI non-normality, but it didn't seem like a big deal. I started in shorts, T-shirt, and a windshirt. As expected, the windshirt soon became too warm, likely because of the clouds that were obviously going to bring rain sooner rather than later. First Climb I started maybe 30-40 deep in the field, going faster than I'd ever climbed in training, but not feel...
Read more

Should I run UTMB?

That's it, the title is the post. No, of course that's not the post, but I am wondering. My schedule has filled in a bit for the year. I'm now signed up for: Orcas 100 (Feb), Paiute Meadows 50K (May), Scout Mountain 100 (June), TRT 100 (July), and hope to run the fun runs Lupine Loopfest (Feb), Desolate Peaks (August), and Euchre Bar Massacre (October). But the big news is that I got selected in the UTMB lottery. I have until the 29th of January to confirm my registration. For the non-runners: UTMB is a 106-mile loop around Mont Blanc through France, Switzerland, and Italy, starting and ending in Chamonix. By all accounts it is spectacularly beautiful and very difficult, with about 33,000 feet of climbing. For most of my running life I haven't really cared about European races or followed European runners, but I got swept up in the spectacle of it last year. It was the race's 20th year, an American man finally won (Jim), and also finished second (Zack), and Courtney...
Read more