Garbage Out

The thing I love about computers is that they do exactly what you tell them to do.

The thing I hate about computers is that they do exactly what you tell them to do.

My dissertation dataset is much the largest set of numbers I've ever tried to wring any kind of meaning from, and the most basic number-crunching is proving to be a larger task than I'd anticipated. I came back from our vacation feeling relieved, because I'd been able to learn enough about multilevel modeling to take a stab at building models with my data. (It is revealing that one of the highlights of our vacation for me was the afternoon spent alone with a stats book and a sleeping baby, in which a couple of light bulbs finally flickered on. Thanks to my husband, who made it possible by wrangling four boys through a water park that sounds like his personal version of purgatory.)

I'm not going to be building models for a while. First I have to get my numbers to line up and stop giving each other noogies.

Last week I spent an embarrassing amount of time just getting some descriptive statistics into a usable format. Today I discovered that the format may be usable but the statistics are garbage.

Somehow when I said to R "take the year 2 tallies for low-frequency vocabulary and divide them by the year 2 utterance counts" -- somehow I actually said, "Please serve me up a heaping bowl of number salad." And R said, "Happy to oblige, and how about some exotic phthiraptera eggs on top?"

I cannot figure out what went awry. The numbers all look plausible but they are all wrong. And because computers do exactly what you tell them to do, it must be my own fault somehow. Argh and blast and FIE.

Comments

[Rob]

Argh and blast and fie indeed.

What did you get from R, and what do you get doing the same thing with a pencil?

[Jamie]

AHA! I was cutting and pasting to show you my number salad when I saw that it's not actually number salad. The values are correct, but they've been re-ordered. So my descriptives are not garbage after all -- I just have to figure out why R scrambled the values. (Because I inadvertently told it to, I know, but that's not much consolation.)

[Rob]

Excellent! What order are they in now?