[EM] The election methods trade-off paradox/impossibility theorems paradox.

[Kristofer Munsterhjelm]

On 07/01/2017 04:21 AM, fdpk69p6uq at snkmail.com wrote:

> Of course we should not let minorities rule over majorities, but we
> should not let majorities rule over minorities, either.  Majoritarianism
> is a form of tyranny, "might makes right".  Single-winner elections
> should choose the candidate who is the best representative of
> *everyone*, not just a stronger faction.

Has that actually happened? I often see Range advocates say that 
following the rule of majority leads to a tyranny of the majority, but 
I've not often seen that happen in reality.

For that matter, were tyranny of the majority true, democracy itself 
should degenerate into a system where the 51% plunders the 49%, but that 
rarely happens either. So something must be going on to keep the 
worst-case analysis from coming to fruition.

There have, of course, been times where a larger group has oppressed a 
smaller one, as I imagine RangeVoting's "kill the jews" example is 
supposed to bring to mind. But in both Nazi Germany and racist states in 
the US, it's not too much of a stretch to imagine that the people doing 
the oppressing would feel righteous enough about it to vote Approval-style.

> Well, psychologists do experiments on human opinion regularly.  When
> they use a ranked-choice system, they call it an "ipsative" measure.
> When a score-based system (the Likert scale), they call it a "normative"
> measure.  It's understood that rankings cannot meaningfully be compared
> between individuals; only ratings can.
>
>
> "In summary, one may state that scores originally obtained as ipsative
> measures may legitimately be employed only for purposes of
> intraindividual comparisons. Normative measures may be employed for
> either interindividual or intraindividual comparisons."
> http://psycnet.apa.org/psycinfo/1971-01501-001
>
> https://en.wikipedia.org/wiki/Likert_scale

But note also, from the same Wikipedia page:

"Whether individual Likert items can be considered as interval-level 
data, or whether they should be treated as ordered-categorical data is 
the subject of considerable disagreement in the literature,[10][11] with 
strong convictions on what are the most applicable methods. This 
disagreement can be traced back, in many respects, to the extent to 
which Likert items are interpreted as being ordinal data."

That is to say, it's not clear whether the Likert scale is a Range scale 
or an MJ scale.

> Well, after mathematically proving that all ranked-choice voting systems
> are flawed, Arrow became a fan of score voting, and evaluating voting
> systems by their actual outcomes.  :)
>
>     *CES: Do you have any particular preferences or ideas as far as how
>     voting methods should be evaluated in the future? Or, do you think
>     there are certain things we should look at in trying to figure out
>     what voting methods we should push?*
>
>     Dr. Arrow: Well, I’m a little inclined to think that score systems
>     where you categorize in maybe three or four classes probably (in
>     spite of what I said about manipulation) is probably the best. And
>     that is to look at the outcomes and see if everybody says, “well,
>     that seems intuitively a reasonable outcome given the inputs.” And
>     some of these studies have been made. In France, [Michel] Balinski
>     has done some studies of this kind which seem to give some support
>     to these scoring methods.
>
> https://electology.org/podcasts/2012-10-06_kenneth_arrow

The method Balinski was experimenting with was MJ, not Range, and "three 
or four classes" sounds more like MJ than Range. Unlike Range, MJ passes 
Majority. So it's not evident that this statement supports not letting 
the majority rule.