Half-Assed Study Makes Greeks At Colorado State Look Like Assholes And Alcoholics

Whiskey_Ginger12 years ago

A study was conducted by the ACHA (national college health research group) asserting that, in all measurable metrics other than finance, Greeks are worse off than non-Greeks. This would be pretty big news in most outlets that are keen on burning the idea of Greek life out of the American conscience, if it wasn’t such an inherently flawed article. I’m going to get a little technical before I launch into this mess, so bear with me.

Let’s list the problems:

1. It isn’t peer reviewed, and I can’t find any previous publications by the primary investigator cited on Google Scholar or any of the high-impact public health journals. That doesn’t mean they don’t exist, but it does mean most of the work she has done has not been considered worth looking at. It’s like she’s being published in The New York Post, as opposed to The New York Times, for those not familiar with the idea of impact factor. Surveys don’t need to be peer reviewed, but are taken much more seriously if they are worked into peer-reviewed works. This is why sociology is the red-headed step child of the sciences, because so little of this kind of work is peer-reviewed with any scrutiny.

2. The sample return rate was not much over 25%. A sample size, by the way, of only just over 2,000 returned surveys. Survey data is considered statistically non-viable below 25%, and this barely broke the threshold for non-Greeks, and didn’t for Greeks (18%). There aren’t error bars or margins of error shown on the study, and you might note that the Greek values are often off by no more than 3-5%, well within the theoretical margins of error on a study like this. Basically, it means this data is a shit sandwich on shit-wheat and rye. It’s only serviceable if your sample size is reliably high enough and all the respondees responded truthfully and didn’t just say they were Greek to fuck with the results. Basically, if this was sent to a journal review panel in an article, it would come back in the mail as a mound of shredded paper.

3. The questions are intentionally misleading, or seem to be, as the figures given don’t seem to list the questions used (another big mistake). The question asks what factors impact academic performance, not how positively or negatively, and with no extent given. Basically, the data is all being conflated with Greek = bad when the data isn’t good enough to support those conclusions.

Okay, now that the technical shit is out of the way, let’s burn the ideas behind this piece to the ground. The survey that is the crux of the argument, which is described above, omits the pro-Greek results in the figure given. We’re much better with contraceptive use, for example. The author of the actual article does a decent job of not getting out the crucifixion equipment and nailing us to a cross, so that’s nice, but the data behind this is still incredibly bad.

“I am so passionate about Greek life and the positive aspects of being in a Greek organization,” Zabriskie said. “It is terribly frustrating when people shut that down and assume they understand.”

Zabriskie said that when she tells other students she’s a member of Greek Life, she is continually met with a negative reaction — typically of disgust.

“It is so frustrating for me because every time I discuss my chapter, I have to constantly defend it and try to explain why I am even in one,” Zabriskie said.

That’s just unfortunate. We shouldn’t have to defend our affiliations because of poorly constructed survey data being used as a weapon against in the media. A lot of this shoddy sociological research is done when it comes to Greeks vs. non-Greeks, because a lot of said sociologists seem to have an axe to grind. I get it. You’re bitter, or something. That’s fine. That doesn’t mean you get to besmirch an entire institution with statistics twisted to fit your narrative.

Let’s put this whole thing in perspective: if I were to do a research project with this kind of sample size, these kinds of questions, drop figures like the ones cited, then make some kind of argument based on it all, I would fail the class. I would be laughed out of the professor’s office.

A fall 2011 survey from the National College Health Association, CSU Health Network and Division of Student Affairs compared responses from students in Greek Life with those from non-Greek students. When asked which factors impacted individual academic performance, Greeks cited alcohol use, anxiety, ADD or ADHD, depression, drug use, eating disorder or problems, participation in extracurricular activities, relationship difficulties, roommate difficulties, sexual assault and sleep difficulties more than non-Greek students.

The only factor that was less problematic for Greeks was finances.

Greeks also cited academics, personal appearances and personal health issues as being “traumatic or very difficult to handle” more than non-Greek students. The survey reported, for example, that 9.6 percent of non-Greek men said their personal appearance was very problematic –– 21.1 percent of Greek men said it was an issue for them.

This is where I realized the study was bullshit, with no reservations. No guy in Greek life would say their appearance issues were “traumatic” for them. None of us are such colossal pussies that we would be seriously distressed by how we look. Might we actually put more into how we dress for class? Sure. Freak out and have a serious traumatic event over it? Hell no.

So, in this case, I don’t blame the author for the article with the shitty premise. She’s just reporting the “facts,” which seem pretty loose in this case, considering the awful statistical practices used by the ACHA survey. They should be ashamed of themselves, but I’m sure they’re too busy gloating over sticking it to those nasty Greeks.

[via The Rocky Mountain Collegian]