>> From: Richard Moore
>> Subject: Re: [EM] Thoughts on majority potential simulations
>> Anthony,
>> Some, but not all issues, are going to have the
>> correlation you are talking about. Given that there are
>> two issues that are really just facets of some third
>> issue, we can just let that third issue be one of the two
>> dimensions in my model. But surely there will be some
>> fourth issue that doesn't correlate with the first two.
I think what you're more likely to find is factors that are
correlated, but not perfectly. I once heard a sociologist
say that if you're doing sociology, always be sure to include
socio-economic class as one of the variables, because it's
correlated with everything. In an election for something
like legislative office, I'd expect everything to be
correlated with political party. Problem is, they won't be
colinear, and they won't be independent.
I have a couple of questions:
How will you deal with a situation in which there are, say,
ten variables, all correlated, but none colinear?
How will you deal with the fact that different voters will be
positioning candidates in different spaces, with different
scales and different dimensions?
I imagine a good way to deal with these issues would be to
ignore them. After all, if the model were to take these into
account fully, it would probably be necessary to construct a
smaller, simpler model to describe it.
