What to methods to use?

[coffeeconsumer]

It's been 3 years or so since I graduated with psych degree, and I remember doing a lot of advanced stats, which I mostly hated until my last prof, who was just an amazing teacher. He started with explaining the general linear model and how everything derives from that, and everything fell into place, and I started to really like stats.

Anyway, that was then, this is now. I haven't had to use stats until now. I'm thinking of doing a personal project and need some help with what direction to go into.

What I'm trying to accomplish is something similar to what google does. Not relating to search engines, or anything like that, just the concept. The idea is to take many different variables, say 10 or 20, and produce a score on a scale, from 1 to 100 or from 1 to 10. These variables can be binary, intervals or continuous.

From what I remember and some basic searching, I'll have to be using something along the lines of ANOVA, MANOVA and/or repeated measures.

Any thoughts on either one of those? recommendations?

Thank you!

[tstats]

The idea is to take many different variables, say 10 or 20, and produce a score on a scale, from 1 to 100 or from 1 to 10. These variables can be binary, intervals or continuous.

From what I remember and some basic searching, I'll have to be using something along the lines of ANOVA, MANOVA and/or repeated measures.

to get a score only, ANOVA etc. are not needed. simply calculate a weighted average of all variables and then scale to whatever you like - 1 to 10 or 100