Here we go again… The good old “user error” excuse.
Technology fails, and instead of understanding what is causing the errors and fixing what’s wrong with the technology, we play the blame game.
Last week, there were quite a few voters in Texas – including Sophic’s hometown of Houston – claiming that the Hart eSlate machines were changing their votes in hotly contested races like the Senate race between O’Rourke and Cruz. In other cases, voters reported that the machines completely removed any selection for U.S. Senate.
Voters took to Twitter staying things like:
“I chose the ‘straight ticket’ option for the Democratic Party on the first screen. Once I toggled through the 16+ page ballot and reached the final screen to review my choices, I saw that my vote for Beto O’Rourke had been changed to a vote for Ted Cruz.”
The office of Texas Secretary of State Rolando Pablos blamed voters, using the infamous term “user error”, a spokeman for the office stated:
“The Hart eSlate machines are not malfunctioning, the problems being reported are a result of user error — usually voters hitting a button or using the selection wheel before the screen is finished rendering.”
Subsequently, votes were told things like: “Make sure to confirm that your summary page accurately reflects your choices BEFORE casting your ballot!” These instructions were meant to serve as the main “fix” for this problem. Apparently, it is now the user’s responsibility to make sure that the technology works correctly. Truth be told, the user is the last link in the chain of events to fail, and the errors are usually precipitated by multiple failures whether procedural or technological.
The actual root cause
The machine’s maker, Hart Intercivic, accepted no responsibility for design issues as they released a statement saying they “proudly stand behind our voting systems and our customers.” Then they praised election officials. Officials the passed the buck to the users. In doing so, they absolved Hart Intercivic of any accountability, and the company will continue to make and sell machines with confusing user interfaces and design flaws as a result.
Here is what actually happened:
Last week, when voters selected the straight ticket option on the eSlate machines, they expected that it would fill in all votes for the one party they were voting for. However, the user interface was incredibly confusing. When they pressed a button while the page was loading, a bug occurred and immediately switched the top race on the ballot to the other party. Some people noticed the design flaw and reported it, but there’s really no way to know how many people didn’t notice. There’s no way to tell how many people voted for someone they didn’t want in office.
When there’s an estimated 5 million voters across 82 counties in Texas counting on eSlate machine technology to vote, this is not acceptable. Sadly, this is not new. This isn’t the state’s (or the country’s) first issue with voting machines. In fact, Texas experienced issues with eSlate machines during the 2016 election. The same issue of candidates being deselected was occurring when voters tried to select a presidential candidate (the first race on the ballot). This year, the Senate race is the first one on the ballot. To fix the issue back then, the office placed signage around polling stations and held training seminars for election administrators. Again, they put the responsibility of bad design back on the users. They should know that training, procedures, labels, signage and other such things will never overcome a poor design.
And the research says…
In 2017, two Rice University researchers took an in-depth look at the usability of eSlate machines, citing a 2008 study of 1500 voters that ranked them the lowest out of 6 machines for ease of use. “There is evidence, both anecdotal and experimental suggesting that the eSlate is not particularly usable,” they wrote. “Counties are already spending a great deal of money on the eSlate and using the systems in elections despite potential usability issues that could lead to longer voter times… and mistakes made by voters while making selections on ballots.”
If we look back at history, we’ll see that user error has been blamed time and time again when it’s really the fault of bad design. News headlines often scream that user error is the cause of design fails that have a massive impact on public safety like data breaches, factory injuries, aviation accidents, and so on. Until products are designed with the human at the center, we can expect to see more stories like these in the news. The strategic implementation of human factors methodologies has never been more crucial, sometimes it can mean the future of a country