The Cincinnati Beacon - Proportional Representation: The Dean v. Jeff Berding

Monday, September 29, 2008

Proportional Representation: The Dean v. Jeff Berding

Posted by The Dean of Cincinnati

Proportional Representation: The Dean v. Jeff Berding from Cincinnati Beacon on Vimeo.

• Share This Article!
Listen to this article

Help The Cincinnati Beacon Grow! Participate in Social Networking!

Socializer!

Members

Tell us what you think!

Anonymous comments are allowed, but you can create an account above to stamp your name and to avoid typing the anti-spam code.

If you are not familiar with our rules for leaving comments, click here! The Cincinnati Beacon is not responsible for the contents of any comments. Comments do not represent the views of the moderators of The Cincinnati Beacon.

Bearman says:
29 Sep 2008 at 12:51 pm | #

As much as you intend to get the message out about how easy PR is, the more I learn about it the more confusing it is. Sure it’s easy to vote but it seems more confusing on how my vote is counted.

When I initially heard about it, I assumed that when you voted under PR in rank order that your order had influence throughout your ballot. ie in a vote for 9 system, my first choice would get 9 points, my 2nd 8 points and so on. That way at the end the people with the most points would win. People who I add on as my 8-9 choice won’t get the benefits of counting like my first choice as they now do under the 9x.

So I read the charter amendment and looked up what amounts to Single Transferable Vote form of PR which this appears to be (See here.)

Feel free to correct me if I am wrong in my example below.

Assume there are 4 candidates running for 2 seats and 20 people vote. That would mean that anyone with 11 votes wins. Here is how they come out (first number is how many people voted them their first choice and second number is how many people voted that candidate their second choice)

Berding: 4,2
Bortz: 1, 12
Qualls: 12, 6
Ghiz 3, 2

Under the proposed plan Qualls would win the first seat b/c she got 11 votes. Her 12th vote would go to Bortz based on that particular ballot showing him as the 2nd choice of that voter. So of the remaining you have first choices at:

Berding: 4
Bortz: 2
Ghiz:3

Since none has the 11 votes needed to win, the person with the least amount of votes is eliminated - therefore Bortz is eliminated.

So under the proposed system the guy who got 65% of voters first and second choice is eliminated and the two who got 30% and 25%, are still electable???

It would seem that if PR is the answer, Single NON transferable votes or Cumulative Voting would be better tactics.

says:
29 Sep 2008 at 03:16 pm | #

Under the proposed plan Qualls would win the first seat b/c she got 11 votes. Her 12th vote would go to Bortz based on that particular ballot showing him as the 2nd choice of that voter.

This is an error, which I addressed in the video, and I think it messes up your computations. You have made it more complicated than it is. Also, your numbers are not specific.

Remember the name: “Proportional Representation.” Let’s say it takes ten votes to win. Qualls gets 12 votes. Half of her supporters put Bortz down for number two, and half put Ghiz down.

Since she needs ten votes, she has two “transferable” votes. Given the proportion of 2nd choice votes to Bortz and Ghiz—each would get half of her surplus, or one vote each.

Bearman says:
29 Sep 2008 at 03:54 pm | #

Thanks for the clarification. However, my concern still stands even if I do the math your way. Assume 100% of the Quall supporters pick Bortz as their second choice. That means he gets 100% of her surplus votes or exactly one vote.

He still has ends up with the least amount of first choice votes and gets removed from the list.

says:
29 Sep 2008 at 04:08 pm | #

No, you need to be more specific with your hypothetical.

In your numbers, I don’t know what individual ballots look like. Your conclusion is complicating the issue.

If you would like to pose a more defined hypothetical, I can use your example.

One is not necessarily removed from the list for receiving the least amount of first choice votes.

Nate Livingston says:
29 Sep 2008 at 05:26 pm | #

I watched the video and have one question. You chastise Berding for his comments about a voter suing if his 2nd place vote isn’t counted, then do a good job of explaining how proportional representation works. But you still don’t answer the question.

If Roxanne Qualls needs 10,000 1st place votes to win and gets 20,000 votes who determines which of the 10,000 2nd place votes get counted? Or let me say it another way. The 10,000 votes for Qualls that got her elected also have a 2nd choice on them. The 10,000 extra votes that Qualls doesn’t need have 2nd place votes on them. So, 20,000 people voted for Qualls with their 1st place vote and they all had 2nd place choices. Do all 20,000 of the 2nd place votes get counted or not? And if only 10,000 of the 2nd place votes get counted, how is that decided? The point I think Berding was trying to make, and its a point I agree with, is that I don’t want my 1st place vote to count for Qualls if my 2nd place vote could’ve been used to elect someone else and 10,000 other people’s 2nd place vote is being used. That’s not fair and it can’t be legal.

Somewhere in the stuff I’ve read on PR the word “random” is used. If the system randomly selects which 2nd place votes get counted and distributed, that’s a problem.

cincysuz says:
29 Sep 2008 at 06:21 pm | #

The explanations of how this will potentially work are unending. This might discourage rather than encourage voting. If it’s this complicated, more people may not bother. Is that the intent? To discourage a large turnout? Or maybe I should ask, “do you see this as a possible consequence?”

says:
29 Sep 2008 at 07:41 pm | #

If Qualls wins, then the total of 2nd place votes on all her ballots will be tallied, and a proportion figured. That proportion will be used to figure how many of her surplus get distributed.

Now, the question becomes how to tally the surplus.

Back in the day, before spreadsheets and software, they used a random sample because they were hand counting everything.

The current initiative uses this same language—however, City Council would fully have it in their power, should the issue pass, to immediately move into a fractional system and not a random sample. In other words, if a random sample isn’t good enough, City Council can switch to a computer solution that mathematically figures the issue perfectly.

Now, I did not write the petition. I would personally have gone the purely mathematic route. But with the way it is, should the BOE or City not get the technology in time, a hand count would be possible using the random sample, like Cincinnati used to do.

If anyone likes PR, but not that idea—just remember: Berding and Council can vote to make it the mathematically pure way if they desire. That’s why pretending otherwise is dishonest.

Like I said in the video—we can, and should, have a real discussion about what is best for our community, and I welcome it. But whatever your position on PR, Jeff Berding is using lies and misinformation and fear. He is unfit for office.

says:
29 Sep 2008 at 07:43 pm | #

cincysuz, you are right. People who don’t like PR may well engage discussions to potentially affect that goal. But the bottom line is this: PR lets you rank your candidate, and your vote is transferable based on preference. It makes it more likely that a diversity of opinion will be represented on Council—which means the Nate Livingstons of Cincinnati, the cincysuzes of Cincinnati, and even The Beaconites, all get their fair share—in a matter of speaking.

Bearman says:
29 Sep 2008 at 08:35 pm | #

Thought I hit send on this earlier but think it didn’t go through:

You asked for more specifics so consider this scenario:

Voters 1-10 vote Qualls first and Bortz Second
Voters 11-12 vote Qualls first and Berding Second
Voter 13 votes Bortz first and Qualls second
Voters 14-15 vote Berding first and Qualls second
Voters 16-17 vote Berding first and Ghiz Second
Voters 18-19 vote Ghiz first and Qualls second
Voter 20 votes Ghiz first and Bortz second.

I messed up my second voter count in my first post so the totals based on the immediate above would be:

Qualls 12, 5
Bortz, 1, 11
Berding, 4, 2
Ghiz 3, 2

One is not necessarily removed from the list for receiving the least amount of first choice votes.

Based on the PR initiative paragraph “l” seems to indicate what I referred to.

says:
29 Sep 2008 at 09:17 pm | #

The key here is that the victory threshold necessary to win a seat isn’t 10 votes. The victory threshold is the fewest number of votes that only the winning number of candidates can get. So if you’re electing one seat, it’s 50%(one-half) plus one. But if it’s two seats, it’s lower.

In this example there are 20 voters and two seats. The fewest number of votes that only the winning number of candidates can get is 7. So if Qualls gets 12 votes, she only needs 7 votes—her remaining 5 votes move to those voters’ second choices. If they all preferred Bortz, he would be at 5 votes as the count moved forward.

says:
30 Sep 2008 at 04:23 am | #

Paragraph “l” says one is defeated AFTER the transfer of ALL surpluses.

says:
30 Sep 2008 at 04:27 am | #

Now, I will admit that though I understand the concept, I’ve not tried to count votes before. But as best I can guess, in your scenario, it would end up like this:

Does Voter 13 have a third and fourth choice?

says:
30 Sep 2008 at 05:04 am | #

This of course is an odd example—only two seats being elected by PR doesn’t make it very proportional. You also need to know more than first and second rankings to know how this count would go.

For a flash animation explaining PR pretty clearly, check out:
http://www.citizensassembly.bc.ca/public/extra/animations.xml

cincysuz says:
30 Sep 2008 at 05:56 am | #

I’m not really trying to be obstinate on this one but Dean your #12 answer worries me. If you’re guessing how this works, that’s a problem. I want to know exactly to whom my vote is going. Not weighted against other factors. That’s the first concern. Then, will someone please play devil’s advocate and address “worse case scenario” when the inevitable happens and this method is, in all fairness, utilized by any number of groups/people that would otherwise be relegated to obscurity? This may potentially be the worse thing that could happen to black candidates, who are almost the majority. I can see a well-organized COAST effort taking over city council and successfully defunding social programs throughout the city. That would be my worse case scenario.

Bearman says:
30 Sep 2008 at 06:59 am | #

PR Fan you are absolutely right.

See this is the problem that I am facing. I actually tried to read the charter and am this confused by the counting process. I agree with both of you that 9x isn’t the way to go but I am not sure this is the answer either.

Single Non Transferable Vote: Vote for only your favorite candidate. Top vote getters win. (could increse this to say 3 without the issues of 9x) The benefit of this type of PR is that the BOE could execute this without any cost to change the way votes are counted.

Cumulative Voting: If there are 9 seats, each voter gets 9 votes. They can spread them over 9 candidates, give all 9 to one candidate or spread them out to several. Again determining the winners is merely a matter of counting the votes.

says:
30 Sep 2008 at 02:36 pm | #

Bearman, as mentioned above, the problem is that you want a hyper specific explanation, but your hypotheticals are different than what may become the reality.

I understand, and have explained, how the votes will be counted. But I don’t know what to say for a set of twenty votes where people have made two choices.

If your issue is with the language, please provide citations that you find confusing, and let’s talk about those. I think I can help with that; I cannot, however, help with unrealistic hypothetical counting exercises. (I do not mean to sound sarcastic when I type that!)

Bearman says:
30 Sep 2008 at 03:34 pm | #

Dean...I think in the end I like the idea of PR just not this version. (see #15 for alternatives) It’s easy to vote using this method, but not so easy to figure out how my vote will be counted. I think I personally have a better grasp on it but not sure the average voter will want to take the time to figure it out.

PR fan says:
30 Sep 2008 at 09:41 pm | #

This really is a better form of PR than cumulative voting and single non-transferable vote. That’s because it’s __liberating__ for the voter. They can vote just as they’d like to, without calculation. With the other systems, they have to be tactical.

Here’s a piece that compares PR (here called “choice voting") and cumulative voting.
http://www.fairvote.org/?page=533

Did you check out that flash animation posted above that walks through how it works? It’s really very logical—and for the voter, the smart thing to do is rank your favorite first, your second favorite second and so on.

Bearman says:
01 Oct 2008 at 07:20 am | #

PR...question for you.

On the video you linked to it stated that if a candidate has a surplus in the first round that his/her second preference on all ballots will be counted for distribution of the surplus. Isn’t this different (and ultimately fairer and less confusing) from the way the issue is described on our ballot?

What about my initial thought of what PR was. If 9 seats, 1st choice gets 9 points, second 8, etc. Top 9 point earners win. Gives the value of ranking and all of my choices have a factor in winning the race.

PR fan says:
01 Oct 2008 at 11:05 am | #

That approach to handling surplus ballots is permitted in Issue 8—e.g, it could be adopted by a new law without another change to the charter—but you’d only want to do it if recording voters’ rankings on optical scan ballots. The 1925-1955 method is the one spelled out in Issue 8 and is the best way for hand counts. The 2009 PR count in Cincinnati (if Issue 8 wins) will likely be done on optical scan, but in case it’s done by hand, the 1925-1955 rules are totally defensible and what’s still done today in Cambridge, Mass.

Your idea of a point system approach (9 points for first, 8 points for second, etc), doesn’t ensure fair representation—51% of like-minded voters might not be sure of winning a majority of seats.. It also creates an incentive to “bullet vote”, as a lot of people figure out now how to do in the 9x system. Right now, if you really like someone and want to make sure that one wins, you really should vote for just one person—with 9x, your other votes count equally for other people. With the point system, it’s close to the same—e.g., 9 points isn’t that many more than 8 points—and some savvy voters will vote tactically.

That’s one great thing about PR. You don’t have to vote tactically. Candidates can run without fear and voters can vote their real beliefs without fear. Check out a commentary about a similar ranked choice system used in San Francisco and how it’s accommodating voters having more choices. See:

ttp://www.sfbg.com/entry.php?entry_id=7202&catid=4&volume_id=398&issue_id=399&volume_num=43&issue_num=01

Huh??? says:
01 Oct 2008 at 12:35 pm | #

I hate to say it, but after hearing your explanation, Dean, it makes me less likely to vote for it. Not because you do a bad job (you actually do a good job of rebutting his points), but because the system is so convoluted and confusing.

The more it’s explained, the worse it sounds.

Monica says:
01 Oct 2008 at 02:55 pm | #

How many people own a house and did so by signing a mortgage? Of those people how many actually understood every single aspect of the mortgage? My guess would be very few. My point is simple or at least I think that it is The notion that PR is too confusing to be understood and should therefore fail at the ballot because of that is ridiculous. This is what I understand about PR. I believe that it will provide a mechanism for true public servants to be elected. I believe that we the people are not being represented by the folks currently in office and further that the current 9X system does not provide the means for us to be properly represented. If we continue as we have we will continue to have council members who represent the interests of corporate Cincinnati, of the FOP, of every and anybody except the people of Cincinnati. It is not a matter of having a council with a democratic majority because even when we had that the needs of the people were overlooked and trampled upon. It is not a matter of having Black council members because clearly simply being Black has not been enough to ensure that Black council members would act in the best interest of the Black community or at the very least not act in a manner that is detrimental to the Black community. Having said all of that I think that the call for change that has been heard across this nation should not begin and end on the national stage but should also have a place in the city of Cincinnati.

says:
01 Oct 2008 at 03:27 pm | #

Amen Monica!

cincysuz says:
01 Oct 2008 at 04:37 pm | #

Let me ask something that may be obvious. Something that probably everyone understands but myself. Something that will open me to great ridicule but I’ll risk it as I’m sure there are others scratching their heads on this. How are the actual mechanics of counting the votes handled? I mean handling the ballot not how the division ends up. Right now the voting is straightforward. An x by a name counts as a vote by that person. That ballot is retired. All those x’s are tallied and the one with the most x’s wins.

If votes have a certain weight based on how other people vote, and how candidates are positioned, what are the mechanics? How is it known that your candidate is already number one so your vote goes to the next person? Is it measured by a computer program? What about paper ballots? Does a ballot have to be revisited and reranked? Are votes juggled as numbers change? I don’t get it.

PR fan says:
01 Oct 2008 at 05:26 pm | #

And just to pile on, PR is really easy for the voter. With a good ballot design and instructions (and there are plenty of good examples out there), ranking candidates 1-2-3 is a snap. And more importantly, the psychology of it is even easier, as a tactical voter really should do exactly what a non-tactical voter would do—just rank your favorite first, second favorite second and so on, stopping when you are indifferent to the rest.

In contrast, the 9x system creates a lot of tactical incentives to not vote the way you want to—tensions between bullet voting for your favorite and trying to vote for at least five and so on. I suspect that after PR is passed and used, voters will say they found the new system easier than 9. And they will have fair results.

I suspect if you picked 100 randomly selected city voters and had them really look at what’s done now and then really look at PR, you’d have more than 90 say “let’s go to PR.” In fact, that’s what British Columbia did a few years ago in Canada, spending millions to make sure the participants had sufficient information, and this “citizens assembly” voted something like 153 to 7 to go to Cincinnati-style PR over its winner-take-all system. It then went on the ballot and 58% of the general public voted for it. See a website about it all here:
http://www.citizensassembly.bc.ca/public

cincysuz says:
03 Oct 2008 at 07:27 pm | #

Well? I’m waiting. All of you proponents that understand this backwards and forwards, how about an answer to my #24? While I was waiting, I thought I’d google some other cities that use PR and see what they think. Maybe I’m off my research game but I can’t seem to find any. So that’s a second question, where is this working, or even a practice, in a city in the continental U.S.? I see it in Ireland, Scotland, Candada and Iran.

says:
03 Oct 2008 at 08:19 pm | #

cincysuz, here’s a good link if you want to do more research on PR. There are other cities that use different types of PR, but Cambridge still uses the same system. PR can be counted by hand or by computer. We used it for 30 years and stopped using it in 1957.

Hot News!

Contact Us

Events

Monday, September 29, 2008