Sunday, April 05, 2009

A nice example of why you need to learn empirical skills

Conventional wisdom holds that breastfeeding is vastly superior to bottle feeding. The superiority of breast milk is supposedly supported by scientific evidence. Upon closer examination, though, the evidence does not appear that overwhelming. Why?
most of the claims about breast-feeding’s benefits lean on research conducted outside the lab: comparing one group of infants being breast-fed against another being breast-fed less, or not at all. Thousands of such studies have been published, linking breast-feeding with healthier, happier, smarter children. But they all share one glaring flaw.

An ideal study would randomly divide a group of mothers, tell one half to breast-feed and the other not to, and then measure the outcomes. But researchers cannot ethically tell mothers what to feed their babies. Instead they have to settle for “observational” studies. These simply look for differences in two populations, one breast-fed and one not. The problem is, breast-fed infants are typically brought up in very different families from those raised on the bottle. In the U.S., breast-feeding is on the rise—69 percent of mothers initiate the practice at the hospital, and 17 percent nurse exclusively for at least six months. But the numbers are much higher among women who are white, older, and educated; a woman who attended college, for instance, is roughly twice as likely to nurse for six months. Researchers try to factor out all these “confounding variables” that might affect the babies’ health and development. But they still can’t know if they’ve missed some critical factor. “Studies about the benefits of breast-feeding are extremely difficult and complex because of who breast-feeds and who doesn’t,” says Michael Kramer, a highly respected researcher at McGill University. “There have been claims that it prevents everything—cancer, diabetes. A reasonable person would be cautious about every new amazing discovery.”

The article continues to describe some very clever ways researchers have attempted to tease out the "true" effects of breastfeeding. Economists call these types of clever approaches "identification strategies", and a significant share of graduate training in labor economics consists of learning how to come up with these kinds of clever strategies.

In many case, I am misled from the result of the statistics. Because I usually look the result like the numbers without thinking the hidden conditions. I mean, in this case, I am driven the result "the breast-feeding is suporier than the bottle-feeding" from the numbers in common parenting magazine. I accept the result without thinking about the background of the babies' family and so on. The hidden condition can make the differences at the result, not the original condition breast-feeding vs. bottle-feeding. I need to study or learn more about empirical skills, because the knowledge about the empirical skills can provide how
I can read the result of research rightly. And through the studying, I can minimize the misleading result from other conditions.
Through this article I was instantly amazed to see how much drama has surrounded the idea of breast feeding when it has been at least semi proven that it in then end does not matter whether you breast feed or formula feed your baby, the author of this article even states that she her-self still uses a combination of breast feeding and formula feeding for her son. The amount of statistics that this author just through around without any back ground on how the studies were conducted left me feeling a little irritated, but I still feel that I could I got a decent grasp on what stats that were presented in this article have more truth too them, either do to the back ground in-which they were originated or do to the people who either conducted or influenced the statistic. I found this article to be very helpful, allowing me to become more educated on how statistics can be misused, vague and how the point of view on certain issues greatly depends on where and who the idea comes from.
This article was interesting in several different ways. I thought the author did a very nice job of mixing the social issues and conflicts with the perhaps dryer mathematical and scientific problems with many of the studies. If nothing else, the article was able to keep my attention well. I thought the section describing why some of the previous tests had found results that may not be true was very interesting. The author’s reasons were so logical and rational that it makes me wonder very much why the group running the observations in the first place did not think of them. This essay really is just a specific example in a much larger problem in general when conducting experiments that involve human observation. There are so many possible reasons for a symptom (High blood-pressure, for example) that it is nearly impossible to test for just one. I wonder if many studies have been done on identical twins? Perhaps that would give more accurate results. In any case, this article helped me to realize that I need to take the statistics that I read everyday with a grain of salt. It is hard to trust these numbers without knowing exactly how they were obtained.
Post a Comment

Subscribe to Post Comments [Atom]

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]