The problem with conclusion by observation

Post Reply New Topic

There is a scientific column in this weeks Newsweek that I found highly relevant to the decisions we all have to make for and with our AS children, and how we evaluate the effectiveness of those decisions. I think the points in this article are really important for us to keep in mind as we try to understand cause and effect. Not to say that we should stop using observation as a tool; but to say we need to remember it is only a tool, and not a conclusion.

The article's topic is science education, but much of the actual focus is on what observational conclusions actually mean in fact, and when they should be considered highly suspect.

Title: Wanted, BS Detectors
Link: http://www.newsweek.com/2010/10/28/what ... class.html

I'm actually going to sticky this for a while because I think it's an important perspective to remember.

Last edited by DW_a_mom on 15 Nov 2010, 1:30 pm, edited 2 times in total.

Thank you for this. I hope others will read and take it to heart.

Roll on those kind of science lessons!

This article/blog on the same subject is really good too: http://quackfiles.blogspot.com/2005/03/ ... teven.html

Totally agree about "observational studies", and about how unreliable our observation and interpretation can be.
.

Great article. Our oldest is starting University in March and will be majoring in Physics. She loves the sciences and always comes home telling us of new studies (especially since her little brother has been recently diagnosed PDD-NOS). This is a perfect article to pass on to her as well as embrace for ourselves. Thank you for sharing

I do feel critical thinking should be stressed more in elementary, jr high and high school science classes.

The author of the article gave many examples of alternate conclusions to various studies, but when a scientist, doctor, researcher, etc, reads a study in a journal, it's understood that the study is being submitted as researchers presenting their findings and offering one of many possible interpretations. This conclusion is only taken as a fact when presented to the general public.

It gets converted from "We found a statistically lower incident of heart disease in moderate drinkers than in the alcohol abstinent group," to "Moderate drinkers have a lower incident of heart disease than teetotalers"

The two are very different statements. It's very easy to tell when a scientific article in the news or a magazine was written by someone with a strong science background and someone without it.

I think I'll copy this into the recommended reading thread and then un-sticky. Just more consolidated that way.

Observational studies are like economics. It is not possible for any single economic theory to be true since there is never a situation where only two factors affect one another.

Interesting read, DW. Thanks for pointing it out. I particularly liked this:

I love asking that question to my son, and to kids in general...you can get the most surprising answers.