[Sigia-l] ballot usability redux

[Whitney Quesenbery]

At 11:13 PM 8/10/2003 -0400, James Spahr wrote:
>I just could not believe that random order would be the preferred method 
>of ordering names. I'd really love to see the history of that spectacular 
>decision, and the arguments against alphabetical listings -- I can see why 
>some people might think alphabetical listings would put some canidates at 
>a disadvantage -- but that really would be assuming your users are morons ...
>
>I wonder if any studies were done.

In most places, ballot order is based on a mix of rules that include 
consideration of the current political party in power, party name (for 
ballots that allow for slates) and random selection.

It may sound crazy, but, these rules were intended to handle 2-6 
candidates, not hundreds.

There have been some studies done, and ballot position does matter in a 
"normal" ballot, which is why there is often an element of chance.

On a humorous side, alpha order would leave open the concept of name 
changes to affect ballot position. "We're naming him Aaaaa in case he ever 
runs for Governor."

Without calling voters names, or suggesting that we have a responsibility 
to do anything but ensure that they are created with the best possible 
design and accommodate any level of literacy and any type of disability, an 
elections official once told me delicately "not to underestimate the 
creativity some people can show in finding new ways to invalidate their 
ballot."

There have been ballots designed for countries with large numbers of 
non-literate people. One is Brazil - there is a demo of their ballot 
available from http://www.tse.gov.br/eleicoes/eleicoes2000/divnet/Urna2000.exe


Whitney Quesenbery
Whitney Interactive Design, LLC
w. www.WQusability.com
e. whitneyq at wqusability.com
p. 908-638-5467

UPA  - www.upassoc.org
STC Usability SIG: www.stcsig.org/usability