The Scientific Method for Dummies

There's a... let's call it a "syndrome" that our staffers here at the Antisocial Network see on a regular basis. All you'd have to do is scroll through our blog listing searching on the tag "scientific illiteracy liberal arts" to see examples: it's people who went through college studiously avoiding science courses suddenly convinced that they know enough about science to write about it. Yeah, sure; and the latest proof is a Kimberley McGee post at Sciencing.com; the one titled, "Why Should We Make Multiple Trials of an Experiment?"

That's a valid question, one often raised by kiddos tackling their first science-fair experiments. The answer is pretty simple: we want to make certain that the results we got were not influenced by some random variable. Say, we want to know how far Jenny can throw a baseball: if she only throws it once, who knows whether her distance was helped (or hindered) by a sudden gust of wind. If she throws it several times, though, and we take the average of all the distances, "the law of averages" suggests that we're going to get an answer that's close to the truth.

McGee doesn't quite get it, though. What she says (in her journalism-major cutesy way) is,

"...how do you know for sure that your idea will hold up based on one experiment? A multitude of tests can narrow the chance that your original idea simply doesn’t hold water."

Not quite, Kimberley: there, you're talking about what a scientist calls "reproducibility." That's not the actual reason for multiple repetitions – repeatability – though it is a feature of the scientific method.

Not content to stop there (or actually stuck with having to pump out another couple hundred words), McGee prattles on. That's where she runs into her real problem: she found herself a cute little article about controlled experiments (it's in her references). That's when Kimberley decided to explain some more terminology:

"There are two types of variables when running tests: independent and dependent. An experiment with two groups, such as using water on one set of plants and nothing on a second set, has independent and dependent variables. The group that receives water, in this example, is the independent variable because it does not depend on happenstance. The scientist applies the water by choice. The dependent variable is the response that is measured in an experiment to show if the treatment had any affect."

That's Kim's bastardization of the actual verbiage in her reference, which says,

"The factor that is different between the control and experimental groups (in this case, the amount of water) is known as the independent variable. This variable is independent because it does not depend on what happens in the experiment. Instead, it is something that the experimenter applies or chooses him/herself. "

Unfortunately, this version of the post is newly rewritten by a member of Leaf Group's cleanup team. In our research we noticed that the original¹ is more accurate: for instance, eHow contributor Clayton Yuetter actually used the word "reproducibility"... McGee not only didn't understand the significance of multiple trials, she conflated the difference between control and experimental groups with the difference between dependent and independent variables. Perhaps if she'd taken a few science classes on her way to her BA in Journalism she wouldn't have ended up winning yet another award: the Dumbass of the Day.

¹ Although Leaf Group sent it to the rewrite team, you can find Yuetter's original version through a search at archive.org. The original URL was   ehow.com/info_12070735_should-make-multiple-trials-experiment.html

copyright © 2018-2022 scmrak

SI - EXPERIMENTS