Election Nudge, one datapoint from Canada

For the last three weeks I had been working a provincial election as a special voting officer; enabling electors to vote at the returning office should they be unavailable for either the advanced or ordinary poll. As mentioned elsewhere; we did an entire provincial election in 28 days, the minimum statutory limit.

During this time I captured the amount of voters we helped out per day, which represents 1/6th the eligible voters; in order to see if there were any patterns or learnings to discover.

Here it is, as a crow-amsaa plot.

image

I’ve segmented the four different periods based on key differences:

  1. Initial Office Opening
  2. Notification Cards Sent
  3. Advanced Poll Accomplished
  4. Election Day.

In short, there was a huge jump in demand due to the voter cards being sent out; This is shown in the exponent being greater than two. In practicality, until the advanced poll; we were the only show in town to vote for this riding. And while I don’t complain about this, advanced polls did reduce our demand in the last week until election day.

Unsurprising, the act of sending a voter card could be argued as an effective and pretty cheap nudge. The unfortunate thing locally is we do not have mandatory voter registration, nor a folkregister style database to pull from to send this nudge out; and anecdotally, many of the younger voters had to be registered “day of” when they arrived at the poll.

2 Likes