Here in North Dakota, we’re gearing up for election season. Given our rich history of combining both direct democracy and representative democracy – where ordinary citizens vote directly on policy via ballot measures in addition to voting for candidates who serve us as political representatives – it seems like a good time to learn a little more about how ballot measures function in our state.
Since 2010, North Dakota has had 26 statewide ballot measures – 13 accepted and 13 rejected by voters. In statistical terms, this is a small sample size and so it is important not to draw very strong conclusions from these measures. However, it is fair to cautiously examine underlying trends and see what we can learn. With this in mind, I downloaded publicly available data from our secretary of state website and analyzed trends in North Dakotan direct democracy.
The most common type of measure is a legislatively proposed amendment to the state constitution (42 percent), followed by initiated statutory measures (31 percent), then initiated constitutional amendments (19 percent), and finally referenda to repeal an act of the legislature (8 percent).
Slicing the data another way, we see that a substantial portion of statewide ballot measures address potential constitutional amendments (62 percent) and a majority of all statewide measures are initiatives or referenda driven by citizens (58 percent).
Interestingly, I find that the rate at which voters abstain from voting on a particular measure is associated with the number of measures on the ballot. The more measures on a ballot, the higher the percentage of voters who do vote on a given measure. The fewer measures on a ballot, the higher the percentage of voters who abstain from voting on a given measure. Statistical models estimate that on average, a lone ballot measure will have about an 8 percent rate of voters abstaining. A ballot measure surrounded by seven other measures will have about a 3 percent rate of voters abstaining.
This means that North Dakotan voters don’t seem overwhelmed by measure-heavy elections, as a larger number of measures seems to encourage voters to vote on the ballot measures. Importantly, this trend cannot be explained by the fact that general elections and presidential elections tend to have more ballot measures. This is because the trend seems to hold even when controlling for whether the ballot measure is part of a general election, primary election, presidential election, or midterm election.
Another interesting finding is that the rate at which voters abstain from voting on a measure is nearly twice as high when the ballot measure originated from the Legislature (7.1 percent) rather than being initiated by citizens (3.6 percent).
Finally, as more voters abstain from voting on a ballot measure, the larger the share of “yes” votes that measure receives. Statistically speaking, if 2 percent of voters abstain from voting on a measure, it is predicted that the measure will fail with approximately 41 percent “yes” votes. If 11 percent of voters abstain from voting on a measure, it is predicted that a measure will pass with approximately 71 percent “yes” votes.
Of course, every ballot measure is unique. These trends still leave lots of room for unique factors to impact election outcomes.
Election day (Nov. 6) will be here before we know it, and early voting and absentee/mail ballot voting start soon. We’re voting on four ballot measures this election, all of them initiated by citizens and half of them proposing amendments to the state constitution. Be sure to participate in our state’s honorable tradition of direct democracy by weighing in on each one. Thank you, fellow North Dakotans, for doing so!
Ellie Shockley is a social scientist and education researcher. This column represents her personal views and not the views of any organization. She completed a doctorate at the University of Chicago and postdoctorate at Nebraska. She lives in Mandan.