Voting Shenanigans at the Northfield News

NN Poll ClippingAs a full-time geek, I couldn't help but think that the most interesting story that the Northfield News carried on Saturday was the small print under their current poll:

Note: As of noon Friday, 115 people had voted, 55 percent favoring the library and 23 percent the Ice Arena/Rec Center.

What an odd little note.  Right above that is the total votes that the poll received: 365.  Since the paper goes to bed on Friday nights, that means that there were 250 votes on Friday afternoon.  The fact that the paper felt the need to add some small print meant that there was something up.  Let's investigate...

Northfield News Poll - Vote Tallies

This is a chart that shows polling as it stood Friday at noon and Friday at the close.  There is a huge, huge spike in votes for the hockey rink.  Okay, okay, this looks like pretty clear cut evidence for ballot stuffing on the part of the Pro-Ice Rink group.  Let's take a look at how it should have turned out if the voting trends shown prior to Friday afternoon had continued.

Northfield News Poll - Predicted vs. Actual

Ah, interesting.  The predicted totals are light blue, the actual totals are a wine-purple color.  You can see that the hockey rink final totals were 700%+ more than what they should have been.  Also, the library votes show a touch of last-minute voting, but only to the tune of 10% or so. 

So we can draw some conclusions here - the ice rink option benefited from a huge last minute push designed to skew the polling.  However, the library option may not be entirely innocent either - they showed a small jump from predicted totals also, but it was much smaller than the ice rink's.

So fess up!  Was it just one person or a concerted effort?  Who stuffed the ballot box?

    NN Poll Clipping


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

after school

I'm not a ice rink fan but maybe everyone got home after school to vote?

Good eye, Adam.

You're right, something definitely looks funny. I wonder if anyone at the Nfld News knows how to access and interpret the data captured by the server (IP addresses, etc.)

Statistical significance

Cynthia may have a point about people voting after school on Friday, excepting that there was no school on Friday. This would fall into the very highly statistically significant result category. The odds of that happening are, probably, along the order of .0000001% . Of course, this is web voting, and I think that the only thing you have to do to rig a lot of web tallies is to vote, clear your browser of the appropriate cookies and vote again, repeat as often as necessary. (That's my guess anyway. Anyone with expert opinion on my hunch?) Also, someone with access to a large bank of computers - like computer labs at Carleton or St. Olaf - could vote on each computer. A pain, yes, but if you're trying to shape public opinion ahead of significant meetings about Northfield's priorities for the future, I'm sure some interested parties would think it's worth the effort. Seems like it would be a fun story for the News to cover about itself. There have been a couple other polls in the past that have seemed a bit dubious as well. Tracy, I think hits on the poll would be traceable, but again, I'm not a big computer guy.
- Brendon Etter
A Play A Day

Good thinking

Of course, this is web voting, and I think that the only thing you have to do to rig a lot of web tallies is to vote, clear your browser of the appropriate cookies and vote again, repeat as often as necessary. (That's my guess anyway. Anyone with expert opinion on my hunch?)

There's two simple methods for tracking votes on a website: via cookies or via IP address. Tracking cookies will thwart 99% of the population from voting multiple times. Tracking IP addresses will thwart 99.9% of the population.

It looks like the NN uses cookie-based. I was able to vote three times (stuffing!) in the 'rental issue' poll currently on the front page by voting, dumping cookies, and voting again.

A dedicated programmer could easily script up a program that could vote X number of times with no problem.

Two words on public voting: Sanjaya Malakar

Look, online polls are fun, but notoriously unreliable as an indicator of public opinion. Just note the uproar over the phone voting on "American Idol." It has been so manipulated that the show producers are considering changing the format to give the judges control to restore the integrity of the show (I can't believe I just used integrity and American Idol in the same sentence.)
Gallup made its name because it uses strict criteria to get honest opinion samples -- which takes time and money. With free polls, well, you get what you pay for.
The idea of pitting the ice arena against the library is just awful. And who will vote for a new police station? It's like asking kids to vote on whether they want ice cream or broccoli and then building a school cafeteria menu based on the results.
The City Council brought on this divisive discussion by abdicating its responsibility to have a clear capital improvement plan that considers all the needs of the community and builds concensus on a timeline for meeting them. Now we will get pushed into a new arena on the basis of all this "public support." I suppose we could run a poll asking whether people would like the city to construct a nghtclub or a waterpark, but that wouldn't make either good public policy.

Look, online polls are fun, but

Look, online polls are fun, but notoriously unreliable as an indicator of public opinion.

 TIME magazine had a ballot on the front cover last year: a picture of Hillary Clinton with two checkboxes, "Love Her" and "Hate Her." Inside the issue was a request for people to cut out the ballot, fill in a checkbox and mail it to the magazine.

True, true

Obviously, online and self-selected polls are not very accurate depictions of public opinion, but it's disheartening to think how much weight many people give to them - as can be evidenced by attempts such as this to manipulate them.

Of course, if the chicanery hadn't been as obvious (as it is in this case), public opinion still could have been influenced somewhat without the News and Mr. Gurno sniffing out the malfeasance. Can't we be more subtle in our pedestrian treachery, people? Folks just get greedy and end up hurting their own causes by looking dishonest and manipulative.

I haven't really made up my mind on the ice arena issue, but things like this can really negatively effect my opinion. I would hope the good people supporting the new arena proposal would distance themselves from this sort of nonsense.

- Brendon Etter
A Play A Day