Monday, December 17, 2007

Weasel words

After Mrs. Gerbil determined that she writes like a man, we had ourselves a lively discussion about what, exactly, makes for manly vs. womanly (girly?) writing. When I analyzed several sections of my dissertation, I found that I write like a weak (girly?) man, or possibly a European man, or maybe a European metrosexual. But other writings of mine--especially letters of complaint--came back strongly male.

I freely admit: my nickname in late adolescence was "Manly Woman," on account of my death-grip handshake and my arm-wrestling prowess.

Mrs. Gerbil's theory was that scientific writing, especially in social sciences, contains more "weasel words" than in other areas. Social scientists such as yours truly are taught never to write "X is," but rather "X may be," or "the data suggest that X is" or "it appears safe to conclude that X." This is an unfortunate effect of the scientific method, whereby you can never actually prove anything, just disprove its opposite (i.e., the null hypothesis) within a reasonable margin of error (i.e., less than 5%).

But theologians such as my better half don't engage in a whole lot of hypothesis testing. They just come up with an interpretation, hopefully think about it for a while to make sure it's internally consistent, and present it. They don't need to do validation studies to see if their conclusions hold up under different conditions. I suppose it's all just a matter of faith for them.

So then this begged the question of how social scientists insult each other. (Trust me. It did.) If we social scientists really wanted to be true to our hypothesis-testing heritage, we might have to say things like

The data suggest that you suck.

It appears that you suck.

We can confidently reject the null hypothesis that you do not suck.

You suck (p < .05).

It appears safe to conclude that you suck. However, further research is recommended to determine what factors, if any, mediate or moderate your suckage.

And now, if you will excuse me, I must go calculate the Spearman's rho of your mom.


me said...


Lavender said...

I just had a nasty flashback to statistics class... and split my sides laughing also. It is true - we science types never give anything 100% certainty. ESPECIALLY in medicine. We are actually BANNED from saying we are completely sure or positive about anything.