SL Ideas #5: Gender-o-Matic
So I was reading an amusing Herald article by noted SL personality and fashion designer Aimee Weber. It was a comic look a gender issues and romance in Second Life. Of primary concern was the RL gender of an apparently female avatar. Aimee's (facetious) answer was a simple quiz. She showed three items typically only associated with the female RL experience, and challenges the players of apparently female avatars to identify each from three multiple choice answers.

In the grand online tradition of creativity, I propose ripping off Aimee's excellent idea and running with it. (Okay, I'd give credit, which as far as I can tell is something entirely new and unheard of in SL. I've always been a trailblazer.) I envision a little booth with curtains, like an old style voting booth or a changing room at a department store. The user enters, closes the curtain, and activates the questionnaire. The machine then shows a series of ten to fifteen images, each with three possible answers shown. (Probably the simplest method would be to include the question and answers in each image. Instant messaging could also be used, although the inherent one second delays between each IM could be irritating. Says and whispers should be avoided, to prevent excessive spam and to avoid giving away the questions.) The machine would tally the answers and spit out a result. Perhaps something like: Certified Female, Probably Female, Unknown, Probably Male, and Certified Male. Maybe there would be T-shirts involved. A pink "Certified Female" or a blue "Certified Male" tee could be amusing. Perhaps a green T-shirt with a big question mark- and no explanatory text at all- would be amusing, too.

There are a few hitches to be worked out. The first, of course, is the quiz design itself. Ideally, the pictures would be half-and-half: half male-centric items, half female-centric. Aimee's examples are ideal for women. While many males (myself included) probably recognized them, they would all feel a little hinky about admitting it. I find I'm at a loss for similar items for males. Perhaps athletic supporters and fragments of porno mag covers? Cleverly creating two plausible (or one plausible and one amusing) wrong choices would be a bit challenging, as well.

The scoring could go two ways. I envision a general scoring range system, say from -7 to +7 (for a 14 question quiz). You would assign a point value for each of the three possible answers to each of the questions. In the first scheme, a "male" answer would be worth +1 point, a "female" answer worth -1 point, and a "neutral/weird" answer worth 0 points. The score would be tallied at the end, with number ranges being assigned to final scores. Say -7 to -3 is female, +3 to +7 is male, and -2 to +2 is unknown/indeterminate gender. (No judgments as to the author's opinions on the relative value of a given gender should be inferred from this scheme.) Or, if the five categories mentioned above are to be used, -7 to -5, -4 to -2, -1 to +1, +2 to +4, and +5 to +7.

As an aside, it occurs to me that the five category scoring scheme might be more useful if a nominal fee (say $5L to $10L) is charged for the test. Some players might be willing to run the test again in order to try for the coveted "Certified Male" or "Certified Female" tees, as opposed to the more ambiguous "Probably..." shirts. But, if the service is to be offered for free, the three category system would be more satisfying for the user. I think it'd be funny as heck to install these on tiny 4x4 plots (in small, tasteful, clearly non-griefing kiosks) next to the more popular dance clubs.

But back to the scoring. In the second scheme, a score would only be assigned to the correct answer. A +1 or -1 would be applied, depending on whether the item shown is male-centric or female-centric. The final tally would be the same as above. Here's a sample question:

What color is this?

A. Fuchsia B. Purple C. Kool-Aid Grape

The correct answer is Fuchsia. The underlying assumption is that no guy would know this, since most men operate off of a 3-bit color palette (or, in non-techie terms, the cheap box of crayons). Under the first scoring scheme, an answer of A would receive -1 point, an answer of B would receive +1 point, and an answer of C would receive 0 points. Under the second, A would receive -1 point, while B and C would be worth 0 points. The correct answer to a male-centric question under the second scheme would receive +1 point.

As you can see, the quiz is wildly inaccurate and all in fun. It's not intended to be a serious diagnostic tool. You may notice that, with either scoring scheme, a knowledgeable person of either gender can come out with a perfect 0 indeterminate score. To add a little more accuracy, perhaps the quiz taker should be instructed to pretend that a group of friends of his professed gender are watching his answers. A male among males would be less likely to admit to knowing what a pantyhose sizing chart looked like, while a female among females would probably be less likely to admit to recognizing a jockstrap or the "P" from a Playboy magazine cover. However, this kind of mental pre-programming may encourage meta-gaming the system to get the gender the user wants. Does this matter? If this were a serious psychological analysis, yes. But in this case? Probably not.

Oh, and I whipped up a logo for the Gender-o-Matic, too. Ain't it spiffy?


Posted by Moriash Moreau at 1:40 PM