Visual Language Project
Newsletter: UPDATE 2000.4

Home

Published by MacroVU Press in support of the Visual Language Project at Stanford University, which is directed by Robert E. Horn.
The UPDATE summarizes the research and applications of visual language and information design.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

In this issue:

This is a SPECIAL ISSUE of the Visual Language Update with commentary by our project director, Robert E. Horn.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

PRESIDENTIAL ELECTION HINGES ON VISUAL INFORMATION DESIGN
Who would have thought the Presidential election would hinge on poor information design? Not most professional information designers, members of a rapidly growing, new profession whose job it is to make sure that the immense amounts of complex information we all receive is clear and easy to use. Practitioners and researchers in this new profession more often are found designing web pages than ballots. But the now-famous Palm Beach ballot is just the latest and most memorable example of poor information design that Americans endure from their government. After the first of the year we'll see another example--the IRS tax forms.

Practically everybody in America now acknowledges the Palm Beach "butterfly" ballot as an example of bad design. But you don't have to be an expert to see the ballot design flaws, so why do we need a profession to criticize bad design? The fact is we don't need a profession to critique bad messages. But we can call on the art and science of visual information design to create and test as many error-free ballots as necessary.

FEDERAL ELECTION COMMISSION REPORT
As far back as 1984, a panel of the National Clearinghouse of the Federal Election Commission pointed at the "need to look at related human engineering standards [for voting devices]; panelists stated that neither the manufacturer of voting systems nor most state and local election offices pay much attention to how the voter interacts with the various voting devices...Panelists stressed that standards should encompass such matters with an emphasis on ballot design and format." However, "subsequent federal guidelines produced for evaluating voting systems, while comprehensive in other areas, did not address the panel's concerns" about clear, readable, non-confusing ballots, wrote Susan King Roth, Chairperson of the department that covers visual communication design at The Ohio State University, and one of the few information design scientists who have studied the communication aspects of ballots. 

INFORMATION DESIGN JOURNAL RESEARCH
Exactly the problem that surfaced in Florida was described in one research study done by Roth and reported in the Information Design Journal. Subjects (voters) in that study indicated that the "electronic ballot was organized in a manner that caused confusion as to which button was associated with the corresponding candidate's name." Said one subject: "The square next to (candidate) Clinton's name was for the other candidate to the left. The square for Clinton was to the right." Sound familiar? It's precisely what many of the nineteen thousand confused voters in Palm Beach county have been saying.

For further information: <http://www.informationdesign.org/pubs/roth1998.html>

VOTERS CONFUSED BY WORDING
Other findings in that study showed that voters were confused by multiple negatives in the wording of referendum issues and that language was unclear. Another startling finding was that some of the reading areas for the voting machines where actually too high above the eye level for a substantial number of voters to even see the issues presented for them to decide!

RESEARCH ON PUNCH CARD BALLOTS
In a second study, Roth put the punch card voting format -- similar to the type of punch card used in Palm Beach and many other jurisdictions -- to the test. The error rate on the punch cards was approx. 15 per cent. Sound familiar? The elderly made more errors and took more time on the average than the other voters in the study. Said Roth's report: "Subjects stated verbally and on questionnaires that they were unsure which holes corresponded to numbers desired and had difficulty tracking progress on the card." The punch card format was "disliked by the majority of subjects" because it was difficult to read. As if looking into a crystal ball in a concluding remark in her study, Roth wrote, "does this imply that previous election results can be challenged by losing parties? Can public confidence in the voting process be maintained if problems with voting systems are published in the media?"

CANADIAN STUDY FINDS BUTTERFLY BALLOT IS FLAWED
Here's more data. Nature (408, 665-666; 2000) on December 7, 2000, published the results of a study by Robert Sinclair and colleagues of the University of Alberta which reported that Canadian shoppers made mistakes using the Butterflly ballot in a simulated voting experiment for a Canadian prime minister. Their results: Of 53 people using the double-column butterfly ballot, four made mistakes; three of these involved voting for the candidate in the wrong column. In a control group, a similar group of shoppers made no mistakes on a simpler, single-column ballot. While a limited study, the results were statistically significant.

For further information:
http://www.nature.com/nature/
http://www.abcnews.go.com/sections/science/DailyNews/butterflyballot001130.html

TEST THE BALLOT DESIGN
A couple of weeks ago when I asked a colleague, Australian David Sless, what he thought was the single most important thing we could all do to improve our communications with each other, he said, "test them." Indeed, a modest amount of testing, done by a professional can go a long way to clearing up these ballot messes. It would be much easier and less costly to do than to change the Constitution to replace the Electoral College with direct election. Given the difficulty of getting the small population states to change the Electoral College, it is eminently more practical.

On the other hand, even if we do get new voting machines, how the information is displayed on the machines is still an important factor. And, if the voting is done on a computer screen, we should have a (pop-up) dialogue box which appears that says, "Here is the candidate you chose. Is this what you want?" And, as Sless says, we should test it.

WHAT NEXT?
All of this may seem like common sense. And it is, because so much of information design is the attempt to design something that feels so easy and clear that one doesn't even think of it as having been designed. 

Can we reduce ballot reading errors to zero? Not likely. But we can reduce the amount of error significantly, and in the case of the Palm Beach "butterfly" ballots for President we should use all of the visual information design knowledge and skill available in the service of our democracy. It's possible. It's even nonpartisan. And it should be one of the top agenda items for Congress in the next session.

For further information: The Information Design Journal is the leading research journal in the field of visual information design.
http://www.benjamins.nl/jbp/journals/Idj_info.html

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
The VISUAL LANGUAGE UPDATE is free.

To subscribe/unsubscribe, email <info@macrovu.com> 
Apologies for any multiple recipients of this issue.

To read previous issues of this newsletter, go to <http://www.stanford.edu/~rhorn>

Editor and Publisher: Robert E. Horn, visiting scholar, Stanford University

Please share the Visual Language Update with your friends and colleagues; we ask only that you keep the fine print attached:
Entire contents copyright 2000 R. E. Horn. The Visual Language Project at Stanford University