Agreed, voting is not supposed to be a test of how well you can figure out how to read a ballot. You could argue that it’s a test of another sort; if you voted for Rick Scott, you failed.
And yet, I would also argue that it is up to you, the voter, to look carefully at your ballot before you turn it in. So I have a mixed opinion about this.
The sample ballot pictured looks pretty much like any other ballot I’ve ever seen, except for the 3 languages.
I wouldn’t have guessed that putting the Senate and House candidates in the column with the instructions would have been an especially big problem, and if people were just asserting that it was a problem out of the blue, I would have been skeptical.
But as the linked article points out, the guidelines (1) specifically forbid this and (2) state that empirical research shows it to be a source of problems. So (1) it looks like the designers used the guidelines as a manual on specifically how to not design a good ballot, and (2) we don’t need to rely on our gut feelings about the design, because someone has gathered actual data.
Whether those guidelines are right or not – whether it actually affected the result or not – if you do the specific thing you were told will cause missed votes, that sure makes it look like you intended to fuck with the election.
One of the things they hammer into you during a psychology degree is that survey design is incredibly complex and delicate. It is very, very easy to introduce a bias into your results with a seemingly trivial design feature.
Elections are surveys.
If I could mandate it, that ballot would copy the California ballot format; it’s drop-dead simple.
user research is important. that page looks huge. much longer than 11".
depending on how you fill it out, it’d be easy to have your arm cover that left side and,or hang off the edge of a table with the voting box hidden.
it’s easy to see everything on a computer screen, maybe not in real life.
I see another big problem with multiple ballots in the same sheet. Just watching the example in the article if you want to track who is voting in senate you could cast 1536 unique combinations of other votes, not to add the free name one could add.
I think that for paper ballot is mandatory that different cards are given for different voters.
That’s because the names at the top of the list attract more votes. (Due to whatever psychological reasons…)
We see this effect in British elections, which are conducted using simple paper ballots with the candidates listed in alphabetical order of their name.
In the US, the party in power is allowed to design the voting form, and place the candidates in whatever order they like.
Note that the Republican senatorial candidate was placed first on this form, although his name is lower in alphabetical order.
The fairest way to lay out the form would be to produce a number of versions listing the same candidates but in different orders, and distribute them randomly to voters so that overall, no candidate gets an unfair advantage by virtue of position on the form.
I voted in Broward and I had to go back and look for the Senate and Congressional races when I got to the front of ballot sheet 2 (page 3). Our ballot had 5 sheets in all.
If you had to read the complete ballot it would take 1/2 an hour so I had a crib sheet for all the state, city and county amendments. Instruction for the candidate choices was: “all democrats.” Not a check list of races.
And the bottom left corner is way down there. The ballots seemed over 14" tall. And the reason the undervotes are so valuable to Nelson is because the D to R ratio here is at least 75/25%.
Where I live the dollar value of bonds on a ballot measure was different in the English and Spanish text on the sample ballot. At least they fixed it on the actual ballot.
On a paper ballot, this actually makes sense to have each race be a separate ballot for recounting purposes. Collect the senate ballots in one box, the representative ballots in another, each individual office in their own box, and so on. In this case, it would mean 10 separate, smaller ballots, with a separate instruction sheet. Which since it isn’t a ballot, can also be given according to the voter’s preference.
The thing is, though, separating concerns like this can also lead to errors such as putting the wrong ballot in the wrong box, and just takes more time because it becomes more complicated handling multiple pieces of paper, so maybe it could be simplified to one ballot for federal elections, and one for state elections (and a third if county elections are on the same day). The main thing is that the instructions are on a separate piece of paper.
Think about it this way: if a ballot design is ambiguous enough to alter the way even 1 in every 500 people votes in (say) Palm Beach County then that’s the equivalent of 30,000 extra votes for one side. The outcome of the 2000 Presidential election came down to just 537 votes.
The state of Florida in particular needs to make every possible effort to employ the best available information design standards into their ballots.
(ETA: that math isn’t quite accurate since I was going by total population of the county but you get the idea)
Barchi-Hallaren 2020!
Technically the ballot was approved and put in place by a Democrat. Ballots are often created county by county to account for local races. And this ballot is from a heavily Democratic county that has Democratic Supervisor of Elections.
I don’t see the “design problem” here. It’s quite obvious that there’s two voting options in the lower left.
I mean … the grey headers for every single voting subject are quite clear.
Missing those two options in the lower left is a user error, not a ballot design error.
I’m not right wing either, btw. I’m left. I think this “problem” is hugely overstated.
IIRC my ballot here in Wisconsin had a couple items in the lower left as well. I sure didn’t miss them. From the results we had, I don’t think most other people missed them either.
The real question is if the statisticians can see the design problem in the election results.
Shouldn’t ballots be designed in such a way to reduce user error? If this or that design element contributes to (some) people making dumb mistakes, shouldn’t it be modified? What’s the downside?
That was your mistake.
------------------------------------------------------------------------------------------------------------------------------
“It’s quite obvious…”
Except, it’s not.
“Missing those two options in the lower left is a user error, not a ballot design error.”
A ballot that statistically incurs unnecessary user error is a ballot design error.
I think this “problem” is hugely overstated.
I do not think it is overstated. But in any case, It exists. It’s greater than 0. It’s an unnecessary impediment to voting, so it should be removed.
If you have a choice of 2 ballot designs to implement (which for the example are of equal cost)
and design A causes more error than design B, you implement design B.
IIRC my ballot here in Wisconsin had a couple items in the lower left as well. I sure didn’t miss them.
You need to think of people besides yourself and not blame others for not being as smart as yourself.
Fair nuf, and I hear what you’re saying.
Still tho, those grey headers for every voting category makes it pretty clear where and what they are, IMO.
This topic was automatically closed after 5 days. New replies are no longer allowed.
