democratic aggregation of utility

[Anthony Simmons]

>> From: Forest Simmons
>> Subject: Re: democratic aggregation of utility

>> Good point about the invariance of the median under monotone
>> transformations.  Until we are sure of the best monotone
>> transforms for aggregation of utility, we could do worse
>> than using medians and other quantiles, as in non-
>> parametric statistical methods.

I wonder if it's really possible to add utilities
meaningfully.  The example you give, of a living wage,
doesn't seem to be amenable to such a thing.  What would we
actually measure in order to assign real numbers to physical
examples?  Stick electrodes in people's brains?  Personally,
I think the real-space model that is so ubiquitous in the
social sciences (e.g. factor analysis) is more of a metaphor.

>> Forest

>> On Thu, 3 May 2001, Anthony Simmons wrote:

>> > Nice thing about median, it isn't affected by running the
>> > scale through a monotonic transform.  For instance, suppose
>> > your income is the exact median.  Then if we look at the logs
>> > of the incomes, yours will be the median of that too.