Online Poll vs. Official Voting Match-up!

We've had numerous requests to comment on how the online poll that we ran during the lead up to the elections performed against official voting results. We were curious too!

The online poll ran for 18 days and over 750 people voted in every one of the three polls for CRD Director, Islands Trustees and School Board Trustees. During that same time, well over 12,000* people used the Exchange (* verified by Google Analytics). Our goal had been to have 500 people participate in the poll. Relative to the number of people that used the site, the participation in the poll was relatively low. Less than 10%. This is likely because we implemented the highest level of poll blocking that the service we use has, and also because some people simply just didn't want to participate.

We did get numersous questions during the poll itself about security and added this note as the poll ran to clarify how scientific online polling is:

NOTE: Salt Spring Exchange online election polls are not intended to be scientific and may not reflect actual results. Instead, they are a way for our readers to share their views with fellow site users. Respondents are not randomly selected, but select themselves to vote on a specific question. In addition, while Salt Spring Exchange online polls are designed to allow only one vote per user, savvy users may find ways around this limitation.

The simple reality is that this form of online polling isn't scientific. It isn't designed to be scientific. It is designed simply to be a self-selecting sample, that can demonstrate potential patterns and outcomes such as the results of the recent federal election demonstrate.

Vote for vote, there were some patchy results with the online poll and some bang on matches as well. When we talk about matches here we mean, a match is when the candidate in the online poll placed in the same order in the official 'preliminary' results.

CRD Director Voting

In the CRD Director poll, there were only three candidate matches and the winner was not predicted accurately. Our poll had Leslie Wallace winning, but in official results, Wayne Macintyre received the most votes. When we look back through our logs, we don't see any obvious vote fraud. Our only conclusion here is that Leslie's team was able to get the online vote out for her better than other candidates, but not at the same rate to other candidates during official voting.

Islands Trust Director Voting

In the votes of Islands Trustee, the online poll was a perfect match for every candidate in every order and the online poll reflected the official outcome exactly. Again, looking back through our logs, we didn't see any issues with the data and it turns out this was a perfect sample size to predict the final and actual outcome.

School Board Trustees Voting

Poll results here are even more curious, the polling accurately predicted each of the three winners, just not in the order of official votes. There were a number of mis-matches amongst the other candidates as well, including with Jean Elwell, who when she withdrew from the race we removed from our poll but her name was still on official voting ballots. She managed to still get 342 votes in official voting, further mis-aligning matches with our polling.

In closing, these online polls are just samples. They are not perfect science as the results clearly show, but in some cases can predict with near perfect results actual potential outcomes. The province of BC is exploring official online voting in the future and we may yet see going to the polls as simple as clicking a button.

The real reason we ran the poll was simple. We wanted to get people engaged in the voting process. People all around the world are fighting and dying for the opportunity to vote and choose their leadership and large portions of our community are complacent about the priveldege we have to elect our representatives and choose how we are governed. However broken the process might be, we're fortunate to live in a place, in country and in a time where we can freely choose who might represent how we govern ourselves.

We wanted to get as many people talking, thinking and ultimately voting for real and that's why we pushed as hard as did for people to participate. The polls got people talking and talking and talking all over the island and beyond and hopefully that helped get people up and out to polls to participate.

Thank you to those who participated in the polling, and the official voting, and in particular to all the candidates who took the personal interest, energy, money, time away from their families and friends to step up and want to help shape the future of Salt Spring Island,

Christopher Roy

November 23, 2011 8:27 AM

Community Comments