Of course, I was completely drunk at the time.
Sounds like a night out with Hunter S. and the Hells Angels.
As I noted, the key aspect about the Cristianini study is that it was able to validate a methodology - that an algorithm could achieve what a person trained in content analysis would achieve given the same textual data and the same query. I also quoted Dr. Cotter's caveats about readability and looked at Professor Lim's careful use of content analysis as data to be backed up with careful historical and textual analysis. So I don't think the piece simply leaps like a happy bunny over the obstacles in all of this. What it does not shy from is pointing out that such big data enables new kinds of critical questions to be asked of our alpha numeric world; the meta narratives - contra Lyotard - need to be taken seriously!
But what else did they involve? Vodka? Extreme sports? How does over caffeination by itself cause more than anxiety?
A good point Ralph, but I believe people are going to have a hard time defending prejudice when the possibility of answering certain kinds of questions with much greater precision increases. Of course, big data or big crit kinda works against the interests of punditry, so I can see a certain economically motivated unwillingness by, say, cable news, to endorse it. More worrying is that the public will react against technocratic determination. Data panics over vaccination or gmos are rarely scientific in a meaningful sense, but rather expressions of political and moral rebellion. That can only get worse.
Actually, my mistake, I first wrote about this on STATS.org. I've referenced it several times since in various places. http://www.stats.org/stories/2010/ny_school_milk_swindle_feb3_10.html
Thanks for this - yes, I wrote about this issue in much more detail for Forbes way back when. A couple of interesting points: kids don't really like low or skimmed milk without some flavoring, cos fat is where the flavor is; second, this policy was hailed by New York Public Schools as having an impact on obesity simply on the basis of the number of calories removed from the canteens. No kid actually had their weight measured.
There are two problems with taxing foods. The first is that it assumes that a tax sends a signal to the consumer that this is something they shouldn't consume; but this assumes we think rationally about what we eat and are not driven by emotion. The second problem is that if weight reduction is the goal of a tax, rather than, say, revenue for health interventions, then you are going to have to impose a broad and very heavy tax in order to curb substitution (that is, the switch to other items that are equally calorific). Even proponents of soda taxes are now talking not of one cent per ounce, but two or three. Given that one cent has proven politically unpalatable, it's hard to see how the public would have the appetite for much higher tax levels (it seems that the attempt to portray soda as the same as tobacco is an attempt to warm the public up for tobacco sized taxes). Finally - as with Monty Python's Spanish inquisition there's always an extra point - it is the poor who will pay a disproportionately high tax burden for these policies because they spend a far higher proportion of their income on food. One interesting point, which Adam Drewnowski discovered when he applied big data techniques to obesity and consumption in Seattle: the poor had access to the cheapest fresh fruit and veg in the city; they just didn't buy them.
Are you the Vasanti S. Malik who is a research fellow at the Harvard School of Public Health's Department of Nutrition and a co-author with Frank Hu on several papers?
Thanks for your comments. Unfortunately, these studies were embargoed until after the debate had taken place - which is a shame as it didn't allow the debaters to incorporate or criticize them. I would note that the AP's version of Allison's statement does not quite reflect the full statement he issued, I believe, during the conference upon release of the papers. I emailed him to ask for clarification, and it is, I think, important to note the qualification:
"Regarding the new papers in NEJM, I am very pleased to see these two new randomized controlled trials published. These are the important kinds of studies I have been calling for so that we may make our judgments on a sound scientific basis. Prior to these two randomized studies, there was not sufficient evidence to conclude that reducing sugar-sweetened beverage (SSB) consumption would reduce weight or total body fat. These two new studies, especially that of Ebbeling et al., now provide some tentative evidence that reducing SSB consumption may help to reduce weight for at least one year among Hispanic overweight or obese children and adolescents who habitually consume SSBs. The task for future research is to now see if these limited apparent beneficial effects can be increased in magnitude and duration, and generalized to other populations and circumstances beyond those included in these new studies."
I would note that the authors of these very interesting studies do note some potentially problematic limitations, in particular with Ebbeling et al, what exactly happened after year one with either the control or the intervention group that led to the loss of effect; the Dutch study appears to have had an attrition rate that may or may not affect the statistical conclusions, and it also appears not to have controlled for physical activity or screen time. It also depends on the baseline consumption being one SSB per day. The gene study is very interesting but its authors also note a long list of potential limitations and a lack of clarity as to what the mechanisms between SSB consumption and genetic disposition actually means.