The Charter, the voter, and the questions left unanswered

As the 2026 election cycle draws to a close, with much of the analysis now starting to cool off, there are still results worth looking at that could potentially answer other questions about the election.

Honolulu Hale – the City Hall for the City and County of Honolulu
PC: Joel Bradshaw, Public domain, via Wikimedia Commons

To start, a question emerges about the 20 proposed Honolulu City Charter amendments on the ballot: what percentage of blank votes did each amendment receive, showing voters’ willingness to make a choice when they did vote?

(For this analysis, the blank-vote percentages are calculated against the total number of ballots cast in Honolulu. All statistics come from the state office of elections.)

This is an interesting question, something which this piece – the 250th Politics Hawaii article – examines here.

The reason why this blogger feels that measuring blank votes is important goes back to the premise of an article written in the last election cycle – 2024 – in which the blank vote can, in some circumstances, be seen as the de facto “none of the above” vote that can potentially change the overall outcome of any ballot measure – candidate or charter amendment.

And with nearly two-thirds of Hawaiʻi’s registered voters in the City and County of Honolulu (Oʻahu), anything that would affect voting in that large block of voters could also affect the statewide total. However, for this analysis, weʻll look only at the voting statistics of the City and County of Honolulu on the 20 ballot questions.

With 254,486 Honolulu voters participating in the General Election, the analysis shows that as the electorate went into answering yes or no on each of the questions, the number of blank votes generally went up as the questions progressed from 1 through 20..

Analysis by PHwSF shows that when the 20 ballot measures are divided into four groups – 1 through 5; 6 through 10; 11 through 15; and 16 through 20 – the average percentage of blank votes in each group inched up.

QuestionsAverage blank rate
1–59.3%
6–109.7%
11–1511.2%
16–2012.6%
Hawaiʻi is a vote by mail state
PC: ArielGlenn, 2026. CC BY-SA 4.0


What is interesting about these numbers is that voters were leaving questions blank right from the start. On Question 1, there were 18,975 blank votes. That is nearly 19,000 people who voted in the election but chose not to answer the very first Charter question.

And the numbers generally went up from there. By Question 19, the number of blank votes had grown to 37,747, almost double that of Question 1. Were voters simply getting tired of answering all these questions?

But then something interesting happened. On Question 20, which dealt with impeachment, the number of blank votes dropped to 25,015. So perhaps it wasn’t just the number of questions, but what was being asked that made a difference in whether people decided to answer.

Still, when you have anywhere from 19,000 to nearly 38,000 voters leaving individual Charter questions unanswered, one has to wonder whether 20 questions were simply too many, or whether the subject matter really drove the desire of a voter to answer the question?

But looking more closely at some of the individual questions suggests there may have been something else going on.

Another factor worth considering is how familiar voters were with the questions before they received their ballots.

Take Question 5, which dealt with ranked-choice voting. There was an organized effort called Yes on 5, along with news coverage explaining what ranked-choice voting would mean for Honolulu elections. Despite being the fifth question on the ballot, only 8.9% of voters left it blank.

Then there was Question 20, dealing with impeachment of city officials. This was the last of the 20 Charter questions, yet only 9.9% of voters left it blank. Compare that with Question 19, dealing with administrative support for the Charter Commission, where nearly 15% left the question unanswered.

That is quite a difference for two questions appearing right next to each other.

How much information voters have about ballot questions may influence whether they choose to answer them. (Illustrative photograph from Oregon, 2008.)
PC: Mitchell Laurren-Ring, 2008. CC BY 2.0

So perhaps there was more to this than voters simply getting tired of answering questions. It could also be that voters were more willing to answer questions they had heard about, understood, or had an interest in.

And that brings in a whole other discussion about money and its influence in electoral politics, with certain questions seen as more important by larger entities (special interests) willing to spend money promoting their point of view through mass media (TV, radio, internet). Other questions received attention through news coverage and interviews, including those on Hawaiʻi Public Radio.

If that is the case, what does it say about placing 20 different questions before voters, some of which received considerable attention while others received relatively little?

Perhaps the lesson from this election isn’t that 20 questions were too many. Perhaps it is that the quality of the questions, how they were presented, how much attention they received, and how well voters understood them mattered just as much, if not more, than the number of questions on the ballot.

Something to think about.