[EM] range ballots chew up slots; "unsupported" range voting claims

[Abd ul-Rahman Lomax]

At 09:05 PM 8/18/2005, Warren Smith wrote:
>As was recently pointed out, it is correct that with range ballots
>run on ordinary plurality voting machines, slots (e.g. "levers" on
>NY-style machines) get "chewed up" 10 times faster than
>with plain plurality voting.  Assuming 10 levels.
>With L levels, L times faster.
>
>Consequently if enough elections or large enough elections, more
>machines would be needed, or you'd have to have fewer levels.
>
>So, not so fun.

I'll point out that Range covers a whole range of levels, starting with two 
(i.e., Approval), which requires one lever per candidate. Going to three 
raises the number of required options by one. If only one lever pull is 
allowed per candidate, then the number of levers required is L-1, not L, so 
the ratio is (L-1) not L.

However, if more than one level can be pulled per candidate, then for N 
levers, the number of levels that can be represented is L = 2^N.

Now, with many levers, it could easily be argued (and it will be argued) 
that voting would become too complex for some voters. The instructions 
might look like this:

For each candidate, rate the candidate. Your rating is the sum of the 
values of each lever. To give the top rating to a candidate, press all 
levers for that candidate.

And then the levers might be labelled 4, 2, 1, which would provide 8 levels 
(0-7).

I think, however, that the biggest bang for the buck would be 4 levels. And 
the instructions would be:

For each candidate, rate the candidate. To give a candidate top rating, 
press both levers.

And the levers would be labelled:

Acceptable, Better. (and press both levers for Best)

I disagree with Mr. Smith about the way in which no presses would be 
interpreted; for quite a number of reasons, some substantial and some 
political, I oppose averaging only expressed votes, but would instead 
interpret no presses as a rating of zero, i.e., the lowest rating. But 
there is another possible interpretation, which I have proposed on the 
Range list, which would allow the use of expressed-vote-averaging. However, 
another lever is required and the combination of presses would become quite 
confusing:

Oppose, Approve. or, for rating:

Opposed, Acceptable, Better. This, then would make averaging meaningful, 
but still takes an extra lever. (The Oppose or zero rating should not be 
used in combination, or, I guarantee it, far too many voters will be confused.)

>On the other hand with say, optical scan machines, you'd "chew up
>those little pen-fill blobs on the piece of paper" which seems much
>less of a limitation. Ditto with many kinds of punch card machines.

It is important to note that "optical scan machines" are the machines, not 
the ballots. The ballots can be quite human-readible. I have elsewhere 
noted that "optical scan" for vote-counting is easily accomplished with 
cheap and common equipment (the scanners could be borrowed and the 
computers could be junk), and if there is any suspicion of vote-rigging, 
counting the ballots by hand would be trivial. (Small towns do it all the 
time, and a large town election is really a bunch of precinct elections, 
with the precincts being, effectively, small towns. I have never understood 
the rush to automatic vote-counting, and I'm a technophile!)

>So I am not saying range voting is wonderful to run on
>all plurality machines.  I *AM* saying, it can be done on
>every plurality machine in the USA, with varying levels
>of convenience or inconvenience depending on the machine type
>and the election.

And depending, of course, on the granularity. With granularity four, i.e., 
two levers, as I think I have shown, very substantial improvement could be 
made. The same ballot could be used Approval (with additional information 
provided but not actually used to determine the winner -- in which case it 
might say "Approved" instead of "Acceptable" -- and Range. The Range values 
might indeed be binary, though that is not absolutely obvious. The highest 
Rating (both presses) might be considered 100, the Better Rating 75, and 
the lowest 50. Or 100, 50, 25, which would be the true binary values. In 
the latter case, the levers might read:

Not the Worst, Better (and Best is both levers)

The instruction would explicitly tell the voter that no vote is considered 
a rating of zero.

>I do not especially recommend running range elections in this style.
>I would much prefer it if there were voting machines specifically designed
>for range voting.  However, because range voting CAN be done on
>plurality machines as a stopgap measure, that makes it a lot
>more adoptible than many other forms of voting, for
>example IRV, which CANNOT be done on many kinds of plurality machines.

I question this assertion. It depends on the number of candidates. Yes, 
combination presses would be much more difficult to use in IRV, but Mr. 
Smith and the others who have written about the use of these machines for 
Range have assumed one lever per rating value. In IRV, levers equal to the 
number of candidates (minus one, but at least one net) is required for each 
candidate -- last place being indicated by no press. So for four 
candidates, three levers per candidate are needed.

I've said this before, so that the assertion continues to be made about IRV 
and voting machines does disturb me. Indeed, IRV and Range require, 
actually, the same kind of ballots, if you want granularity sufficient to 
give a unique rating to each candidate. That is, you could vote in the IRV 
fashion on any Range ballot. The only way that Range may handle more 
candidates than IRV on a ballot is by requiring equal rankings for some 
candidates, if the voter wants to express an opinion on all candidates.

But the counting methods are different, so how voters would vote would 
presumably be different.

>Finally, it has been claimed that I make a lot of "unsupported statements"
>about range voting.  (Which itself was an unsupported statement...)

No longer. See the above.

>If a list of such statements is brought to my
>attention, I will try to back them up.  In fact I have already done
>so on various occasions and the CRV web site also backs me up
>a good deal.  But anyhow, whatever statement you find insufficiently
>supported, query me on and I'll try to get back
>to EM on that statement.  I believe everything I say is supported,
>...  but I am not perfect.   Anyhow please let me know.  Thank you.

I trust, indeed, Mr. Smith's sincerity. He really does believe what he 
writes. However, as he said, he is not perfect, and often his arguments 
seem to have failed to consider this or that aspect of a question. However, 
it is also true that his statements are generally supported by some 
argument somewhere that is reasonably accessible. I'd suggest that if 
anyone sees a statement by Mr. Smith and thinks it is "unsupported," it 
might be more cautious, rather than asserting something which really may be 
narrow and only true within a constricted context (i.e., the immediate 
thread), to simply request support or references....

This does *not* mean that his statements have been proven, and often he 
writes as if they have been. I've told him that it is a bad habit, and I 
specially get to do this because I am a pot calling the kettle black. At 
least I have often been accused of arrogance in what I write, as if I know 
while others do not....

And maybe sometimes I do and definitely sometimes I don't....