Sorting Students Into Houses

Hello House Leaders,

I was pleased with our Heads of House meeting today; we got a lot done. I'm excited that we're so close to starting up this whole thing!

The new ideas Jackie and Mel had for Sorting students into Houses were fantastic; just exactly what we needed. I was never really completely happy with the process as we had discussed it over the summer, but I think this new system will (eventually) be perfect, as it combines elements of the earliest discussions we had about Sorting based on characteristics in some way with the process that, till today, we were going to employ. Of course, we'll have to work out the kinks and smooth it out a bit, and I'm sure we'll learn some things from this year for next year. Let's all try to remember that we still have to view this as a work-in-progress.

Obviously, I think one of the most important facets will be the questions. Even though we're still going to employ a stochastic element in the Sorting process, having the starting foundation of the questions, I think, balances the ideal of uniformity with at least a nod to alignment via traits. But I completely agree with Jonathan that it's important that the questions not be too obvious. I mean, if we ask questions like, "Do you prefer medical/legal dramas, sci-fi shows, reality shows, or sitcoms," it will be too easy to figure out. On the other hand, I guess we could always put a few questions like that on the forms just to make them think they've figured it out, but then we only look at the real questions. What does everyone think of that?

Of course, once we have the results from the questions, there are still the details of how we're going to translate that. The rating scale we discussed could work, although I still think that's where we need to insert the random element; right after the ratings are determined and before the categorization. I understand JP's point about the potential for imbalance if we do it then; but like I said, I also think the problems with putting the random factor in before that step could potentially cause much larger problems; by analogy, it would be like starting a voyage with your bearing half a degree off. At first, it won't make much of a difference, but the longer you travel, the further off you get from where you meant to be! If we put in the random element too early, it basically defeats the purpose of the whole new process. (At least, though, we all agreed that we can't do the random component after the categorization.)

So we only have a week or two to figure this out...let's get some ideas flowing!