Wednesday, December 12, 2007

Meaningful Data Presentation in Science Writing

In class today, we discussed the use of imagery such as "a milligram of aspirin in an Olympic-sized swimming pool." My objection to the use of this kind of imagery is that it tricks the reader into thinking that he has a sense of magnitude of something, even though he is provided with no real information.

The aspirin quote was originally a way for the developers of the contaminant detector to describe the precision of their device in layman's terms. I don't think the original statement was meant to have any particular connotation.

When the article quoted the scientists, it was implying that the Department of Homeland Security is wasting our tax dollars by building unnecessarily precise detectors. The problem with this argument is that without a reference, there is no way of knowing whether explosives are dangerous at a concentration of one aspirin per swimming pool. The level of precision of the detector may or may not be required to keep us safe. The article doesn't really say; it's like an impossible dimensional analysis problem from high school chemistry:

If I have 10 Klein bottles full of nitrogen-based explosive at a concentration of 3 milligrams of aspirin per Olympic-sized swimming pool, and I can fill the swimming pool with Klein bottles in 0.7 fortnights at a rate of 1/3 Klein bottle per half minute, how many trains can I blow up? Please truncate number of trains to nearest integer.

Wow, I hope that made you cringe.

When I was reading one of my hippie blogs today (yes, I am vegetarian, do yoga, drive a hybrid car, grow my own hydroponic herbs AND use CFL lights), I found a slightly more reasonable use of this kind of comparison. I consider this article pretty awesome because it suggests eating your Chia Pet. Here's the link:

http://www.meatfreeblog.com/news/eat-your-chia-pet-it-might-save-your-life/

"The study found that the seed used in Chia Pets...is basically a “Super Grain,” with multiple potential health benefits...Findings included...reduced blood pressure by an average of six points . Dr. Amir Hanna, a diabetes specialist who reviewed the study, noted that, “That’s a very important reduction in blood pressure. Actually, some pills don’t lower blood pressure that much.”

The research team calculates that 3.5oz of Salba [Chia Pet grain] contain:

· the same amount of omega-3’s as 28 ounces of salmon

· as much calcium as 3 cups of milk

· as much iron as five cups of raw spinach"

I like how the data is presented here because it's both factual and meaningful. If the author had just said, "it lowers blood pressure 6 points and contains such-and-such number of milligrams of omega-3's," that would really mean nothing to me. I'm not a doctor, and I don't even pretend to play one on TV. I do have a good sense for how much calcium is in a cup of milk, however. Even better, I have enough information to find the actual amount of calcium in 3.5 oz of Salba.

I also like the way Dr. Amir Hanna said that the grain lowered blood pressure more than some pills. I can infer that "some pills" refers to blood-pressure lowering medications (maybe not top of the line ones) that are either currently prescribed or were prescribed in the recent past. I don't need to know the scientific name for the pills to be able to use them as a metric for gauging the effectiveness of Salba at lowering blood pressure.

2 comments:

Professor Mazzolini said...

Awesome post, pixelfishfood. Numbers or comparisons are only meaningful in context, and it's all too easy to let some number or another slide *just because* it's a number (or representation of a number).

If anyone's ever interested, there's a fantastic chapter in a recent book on the basics of science (_The Canon_, by Natalie Angier) on quantitative thinking. I can lend it to you.

Juliet said...

Thanks!

I would like to take a look at the book. If you bring it to class, I can skim/read it over winter break.

By the way, the answer to the dimensional analysis problem is 0, because Klein bottles have 0 volume by definition.