Those random context-free "Please choose" polls I run once in awhile aren't quite as random as they look. There's actually a method to my madness. I don't normally explain anything about those polls, because the explanation would screw up the results, but I'll make an exception for the latest one.

That's the poll I posted on Thursday, the one whose exclusive choices were "love," "solitude," "sex," "friendship," and "intimacy."

It's an exact duplicate, down to the word choice and order, of a poll I posted last April. The only difference was that in the new one, I said that the responses were "semi-anonymous," or as anonymous as I could make them.

After the first poll, a few people commented on the fact that no one picked "sex" as their word of choice. At the time, H wondered out loud if people were too embarrassed to pick "sex" because they knew their names would appear with the response, and didn't want to appear shallow. Maybe if the poll was anonymous, the response would change.

So I waited nearly a year, and re-ran the poll with an extra layer of anonymity. The answer: nope, the extra level of anonymity doesn't matter a damn. The responses were nearly identical both times. There's no statistically valid difference between them.

Just in case you were wondering. Thanks for answering the polls, by the way. I'm always fascinated by the responses. (And they made wonderful sample data when I needed something to use for my stats homework.)