Re-design of Ballots

The background of this case study is building on cases where the design of the voting ballots made the voting process challenging.

The first case was the 2018 Broward County election in Florida, USA where there was a significant undervote in the senate race.

Here the reason was that the senate race was placed below the left-most column under the ballot instructions. Another issue is poorly filled out ballots, where voters either do not fill out the ballot correctly or they poorly mark the ballots, making them hard to scan.

In the 200 election it was said that the design cost Al Gore the election to George Bush with a 537 vote margin.

The main issue was that the names were not properly aligned with rows where voters would indicate the candidate they preferred.

The last ex are ballots from Sarasota County which exemplifies how communication can fail if it is not very clear. Here the point was that the ballot should say "Vote for One" in each category. However, voters understood it as only voting for one candidate in the whole ballot.

The first ballot was more confusing than the second ballot.

1

2

What can be done to improve communication between the government and its citizens?

In an article by the scientific american, Philip Kortum, a psychology professor at Rice University claims that there are four general classes of major voting issues we see in balloting.

We got ballot design issues.
We got instructional issues(telling people how to vote)
We’ve got issues related to language on the ballot.
The way people interact with voting systems in general.

The hypothesis used for this study is to incorporate the picture superiority effect

In this case study the psychological theory “picture superiority effect” will be used to try to improve issue number 2. The hypothesis states that:

“Sensory semantic theory aims to explain why we are able to better remember pictures over words or numbers. This theory, which was proposed by D.L. Nelson, further asserts that pictures present meaning in a more direct way. Thus, they are more efficiently understood and recalled. Such advantage is termed as picture superiority effect. Moreover, pictures’ unique visual attributes make them easier to remember.”

By using instructional pictures we can potentially help reduce the difficulties with understanding the instructions.

Hypothesis

H0: There picture will have no effect on the amount of mistakes when voters fill out ballots

H1:The ballot with the picture will yield fewer mistakes when voters fill out ballots.

Success Metrics

Low instance of incorrect voting
Easy to fill out

Control

Variant

Type 1

Qualitative usability testing

Field studies/user interview
Aiming for around 5 users

Duration: 10-15 min

Location: University of Oslo

Compensation: A coffee

Type 2

Quantitative usability testing

Survey: using https://www.qualtrics.com/uk/
Aiming for 60-70 users
Use hot spot feature to simulate voting
Use a chi-square test to see if there is grounds for rejecting the null hypothesis. Our null hypothesis is: there is no difference between incorrect voting for ballot 1 and 2.

Duration: A few days to gather enough answers

Location: Online

Compensation: set by the website

Process for the two methods

Results from qualitative research

Results from quantitative research

A suggestion of re-design based on the results

Based off the results this a proposed re-design of the voting ballot.

Some features included based of the results

Quantitative: evidence suggests that we can support the picture superiority effect hypothesis. Therefore it is helpful to include an instructional picture.
Qualitative: messy and no sections. In order to make the process easier for the user there are sections instructing the user where they should start. Furthermore, in order to make the text more readable I chose a different background colour to get a better contrast.

WCAG 2.0 level AA requires a contrast ratio of at least 4.5:1 for normal text and 3:1 for large text. WCAG 2.1 requires a contrast ratio of at least 3:1 for graphics and user interface components (such as form input borders). WCAG Level AAA requires a contrast ratio of at least 7:1 for normal text and 4.5:1 for large text.

Large text is defined as 14 point (typically 18.66px) and bold or larger, or 18 point (typically 24px) or larger.

PDF of case