Sunday, October 25, 2009

Can UX Swing an Election?

In 2000, the infamous "Butterfly Ballot" probably threw the election to George W. Bush, costing Americans more than a trillion dollars and who knows what else. Can it happen again? Unfortunately, the answer is yes. Right now, in King County, Washington, an abysmally designed ballot runs the risk of driving the state into the ground. The problem has two parts.

First, the State of Washington is unfortunate enough to have an idiot named Tim Eyman who makes his living on proposing initiatives. He doesn't do anything else. Initially, he denied that he was making money off the initiatives that he proposed, but enough information was made public that he had to admit it. When he proposed his first initiative, he apparently sold watches, but I think that business is long gone. His first initiative was a "$30 car tab initiative." Eyman picked that number out of thin air (literally, it came from Colorado, with no relationship to the cost of anything in Washington). Some of the people who voted for it who promptly lost their jobs after the election because of all the programs that were slashed, but they didn't see the connection. And Eyman, who said repeatedly that the initiative wasn't about him, went out and bought an expensive SUV right after the election. Amazingly, Eyman managed to garner a following of people who'll vote for anything that will supposedly cut their taxes without regard to what it actually means. This year, Eyman's initiative, I-1033, is one more poorly thought-out proposal. Pick the worst state budget in ten years, in the midst of a recession, and force the state to stick with that budget, essentially, forever.

So what's this have to do with UX? Well, Eyman is only thinking about himself. What provocative initiative can he propose to guarantee that he's still paid a salary? He doesn't care about the experience of the average Washingtonian. I don't even think he cares if the initiative passes. If it fails, he's got another one to propose to pay him through the next election. This is no way to design a government that works for all the people. With good design, you need to think about the whole range of people that will be affected -- and, funny thing about it, that's what our whole legislative process is designed for. Gadflies and devil's advocates can be extremely useful and have a long, rich history in this country. If Tim Eyman actually cared about the state, he would work with our government rather than against it, but, alas, there's no profit in that.

On to the second issue.

King County royally screwed up the ballot, literally marginalizing I-1033. Take a look:

The first column of the ballot contains the instructions that nobody reads, and the top of the second column looks like the start of the ballot, not the middle. This means that some King County voters, perhaps many, will start in the second column and miss voting in I-1033.

The naysayers argue that it's trivial and that, even it's not, it doesn't matter because both people voting for and against the initiative will miss it. On the trivial argument, go back to the top and think about the Butterfly Ballot and what an effect it had. On the second argument, that would be valid if every ballot in the state had the exact same problem, but only King County has the problem. And it is not the case that voters in King County and the rest of the state vote for and against initiatives in approximately equal percentages. In fact, King County voters are more likely to vote against Eyman's initiatives. Combine that with the fact that King County is the most populous county in the state and you definitely have the possibility of a fraudulent election. Despite a large push by the No On 1033 campaign to alert voters to the problem, I am sure that we will see significantly fewer people voting on I-1033 than on other ballot items.

In summary, design really matters and bad design hurts. The only question here is whether the bad design will be big enough to swing the election.


Steve Zemke said...

You are right about the problem of the ballot design affecting the election. Proponents say that both people voting for it and against it will miss it so all things are equal.. That's a false argument because it is only correct if the vote is split 50% for and 50% against.

If the King County vote is 1/3 for and 2/3 against, then more no votes are being lost and this could affect the final statewide vote.

This is critical because right now the election is viewed as very close and this potential loss of no votes could give Eyman a victory that was affected by King County's messed up ballot design.

Mason Boswell said...

Is it constitutions you don't believe in or just democracy? Tim Eyman follows a process provided by the Washington State Constitution and wins only if a majority of the people are convinced that his initiative is a good idea. That's pure democracy at work. Agree with him or not, it would appear your beef is with the people, not Eyman. Your post started with a good premise, but I'm so distracted by the political vitreol that it's lost on me.

Roy Leban said...

@Mason: Thanks for your comments.

My beef is with Eyman, not the people, but I do have a strong disdain for the initiative process as well. Eyman is following a process provided by the State Constitution and wins only if he convinces a majority of the people TO VOTE FOR IT. He's not really interested in convincing them the initiative is a good idea, because that would require actually explaining things. I remember after the car tabs initiative, people were quoted in the paper "I voted for it but nobody told me I would lose my job." Well, actually, they did, but Eyman and company just said "the government doesn't need money -- they'll figure it out" and, of course, the majority of our government spending goes to pay people, so cutting money means cutting jobs. Rather than offer solutions, they just say "let's take money away and trust that everything will all work out."

And that relates to why I don't like the initiative process. Our country was founded on, built on compromise. It was fundamental to the constitution and fundamental to our government. Three branches of government, not one. And the legislative branch is divided in half, elected separately. Every state has copied this model although some have only one legislature. But initiatives flaunt compromise -- they bring up a problem and then they provide a single, usually skewed solution. The people proposing an initiative have no reason to compromise, to ensure that the solution is good for all the people. They just need a problem that's important enough that 51% of the people will vote to fix it. If it were up to me, I would split every initiative in half -- you can vote on the problem and you can vote on the particular solution. I would bet far fewer people would vote for a particular solution than the problem.

Back to Eyman. He actually understands this very well, which is why he doesn't propose solutions. He proposes problems. He has no idea how government can spend less money, but he's happy to try and ensure they have less to spend. He has no idea how to make government more efficient, but he would be happy to add extra efficiency requirements. He has no idea how to make car tabs fairer, but he's happy to just eliminate them.

Read Tim Eyman's argument for I-1033 (it's been printed in a lot of places) and it boils down to "the government doesn't need as much money as they think they do." That's not a solution -- it's wishful thinking. Eyman has proved his mettle in finding problems, particularly things that the people are frustrated with. If he would spend his energies trying to help find solutions, instead of proposing uncompromising non-solutions, we would all be better off. But, like I said, there's no profit for Eyman in there, so he's got no incentive.

Finally, taking this back to UX, it is very, very easy to find problems. In fact, with a bit of training, it's almost impossible not to find them. But it is much, much harder to propose viable solutions. The purpose of this blog is to help people with that harder problem and that's why I decided to address this issue. If you're working on a product and you want to improve it, spend your energy trying to find solutions, not just complaining.

Mason Boswell said...

Fair enough, you disagree with the initiative process as is your right. However, I think you give too little credit to the people when you imply that Eyman is convincing them to do something they would not do if they were more informed. This state is full of smart people and voter turnout is so low in all elections that I doubt that those that do vote take it lightly. Moreover, part of what you accuse Eyman of, incomplete solutions that resemble new problems to you, is brought on by the current judicial misinterpretation of the initiative process. Everytime Eyman has tried to do something organized and comprehensive, a court overturns it for addressing more than one issue.

We live in a country where the three branches of government currently operate mostly in collusion, with questionable loyalty to the people. It seems at times that special interests have bought and paid for our representatives through the current overemphasis on getting reelected into indefinite terms of power, particularly in the legislature. I am thankful that in Washington we have an additional check on government through the initiative process to try to inform government that they are going in the wrong direction and must answer to the people. I see Eyman's initiatives as exactly that, a reminder to the government that the people are unhappy.

I'll get off this topic since this is a UX thread. Your comments on UX are something people of all political leanings can agree with. We are a democracy and yet clearly ballot design and even other ballot technology (such as the counting process) show almost a complete lack of understanding of improvements to user interfaces and technology over the last 20 years. It is sad that so little is invested in solutions to these problems, even after such large fiascos as Florida in 2000 and the Gregoire/Rossi governor's race here. Both major parties have lost to such problems, yet neither seems interested in addressing it. I guess if it doesn't get a politician reelected or make political contributions flow in, they just don't care.