Thursday, October 29, 2009
Childbirth and Evolution
"One reason women die in childbirth has to do with anatomy, arising from two basic evolutionary trade-offs. The first is that once our ancient ancestors began to walk upright, too large a pelvis made upright walking and running inefficient and exhausting. A narrow pelvis permits fast running. That however makes childbirth exceedingly difficult. So the evolutionary adaptation is that women generally have medium-sized pelvises that permit moderately swift locomotion and allow them to survive childbirth - most of the time.
The other trade-off is head size. Beginning with our Cro-Magnon ancestors, human skull size expanded to accommodate more complex brains. Larger brains offer an evolutionary advantage once a child is born, but they increase the chance that a large-headed fetus will never emerge alive from the mother."
Thursday, October 8, 2009
Wednesday, October 7, 2009
Improving Self Reported BMI
Monday, March 16, 2009
Does the NCAA Selection Committee Use Regression Analysis?
Based on the slow involvement of statistics in other sports, my guess is no.
When college basketball experts talk about using "computers", I fear they are only talking comparing team's RPIs and Strength of Schedules via an eye test. The problem with these two measurements is that they are co-linear (a statistical way of saying too similar). All teams that have crappy RPIs are going to have crappy SOSs, like Penn State, since SOS is a part of RPI. Do we know how likely these measures predict success? I don't know, but there is a lot data from previous tournaments and it would be easy to determine. However, RPI and SOS would only explain a portion of the variability of the model. There are a lot of other variables that should be considered like record in the last 10 games, previous performance in tournaments, margin of victory, and potential injuries to key players. If these variables could help better predict a winning outcome (explain more of the variability in the model) then they should be included.
Yeah it's a human decision, and the committee does look at these variables and takes them into account. But there is a huge amount of bias in a purely human decision. It should be the human's decision to determine the variables in the regression analysis that makes the final decision. The committee may think that RPI is the most important variable and deserves the most weight. This may be true, but their assumption should at least be tested by regression.
During the selection show last night a committee member started to explain how they chose Arizona/Dayton over the other bubble teams. He emphasized that teams chosen were based on playing a competitive schedule throughout the year and winning quality games away from home. Of course Penn State had two huge road wins (MSU, Illinois) and Arizona didn't have any (and were 1-5 in their last six games). It's quite possible that the regression could have chosen the same 65 teams, but it would be interesting to see the analysis.
Thursday, February 26, 2009
Say Goodbye to Humanities?

It's not too surprising since the cost of education has dramatically increased, students have desired more return for their education. Now many don't have the luxury of studying less directly applicable areas. I also wonder if humanities degrees were artificially high in the 60s and 70s from all those hippies. haha
I agree in the article's point that Obama benefit the humanities field in multiple ways. He is a role model to many and he seems to greatly respect literature and philosophy. He also plans to make it more affordable to go to school through programs like the GI Bill and increased student aid. However, if the economy doesn't turn around, those points will probably be moot.
Wednesday, February 25, 2009
How Reliable is Health Advice?

Via the Health Blog:
Just 11% of more than 2,700 established heart recommendations are backed by high-quality testing, says a study in the current issue of JAMA.
The American College of Cardiology and American Heart Association jointly issue guidelines to doctors, WSJ explains. The ones thought to have the highest level of evidence are based on multiple randomized clinical trials. Those considered weakest are backed by expert opinion or case studies.
According to the JAMA paper there are three type of sources:
1. Sources based on multiple randomized clinical trials (high quality testing)
2. Sources based on a single randomized clinical trial or observational study (moderate)
3. Sources based on case studies, expert opinion - etc (poor)
As an epidemiologist, I will probably spend most of my time dealing with observational studies in the moderate category. While not "high quality", these are still important since many randomized control trials can not be conducted because of ethical, feasibility, and financial reasons. For example, we can't ethically conduct a randomized control trial on whether smoking causes lung cancer - since there is substantial evidence that we would inflecting harm in our control arm.
However, randomized clinical trials (RCTs) should be used whenever possible because they are removing many biases and confounding from the relationship of interest. A classic case of the benefit of RCTs is Hormone Replacement Theory (HRT) in women. The Nurses Health Study and others found a protective effect on mortality for HRT in several of their observational studies. Their evidence looked so convincing that hundreds of thousands of women went on the therapy. That was until randomized clinical trial results found that it can increase breast cancer and disease risk.
Getting back to the JAMA article, it's pretty scary when you think about. Heart Disease is America's biggest killer, and quite possibly the most widely studied disease. What this says is that there is a lot of work to do in the health field and that there will be a lot of reversals, ala HRT, of doctor opinion in future recommendations.
On a positive to note, I'm happy to say that we're headed in the right direction with this $1.1 billion dollar allocation to compartive effectiveness research in the stimulus package. Maybe Billy Beane had Mr. Obama's ear?

