You know what’s said about doing the same thing over and over again but expecting a different result—that’s insanity.
Local union officials might ponder that. The thought struck me Tuesday afternoon, when I noticed a sudden spike in responses to our RBJ Daily Report Snap Poll. Not only did the number of poll participants begin to rise rapidly, but the results started to shift dramatically.
More than three years of weekly polling has taught me those are the telltale signs of “ballot stuffing.” I did a quick check of the poll responses, and my suspicions were confirmed. Someone was trying to skew the poll results—and I knew who.
The fact that it happened was not a surprise. The topic of this week’s poll was reform of the Rochester city schools. An issue that has stirred intense debate for years, it generated even more heat recently when the Rochester Teachers Association, under longtime President Adam Urbanski, delivered a vote of no confidence in city schools Superintendent Jean-Claude Brizard.
Our poll essentially put the same question to RBJ Daily Report readers—but we asked them to weigh in not just on Brizard but also on Urbanski and the RTA.
We make no claim that RBJ Snap Poll results reflect the views of the entire community. While we have many readers who work in the public sector or for non-profits, most are in the private sector; many own or help manage businesses. And a good number do not live in the city, though many work there and have a stake in Rochester’s schools.
What we do want to be able to say, however, is that the poll results are credible. The only way to do that is to prevent stuffing.
Fortunately, we can. As I’ve noted here on several occasions, the software we use to conduct the polls enables us to identify and filter out responses stemming from a Daily Report reader forwarding the survey link to a group of non-readers. In addition, the software makes it possible for us to exclude multiple responses from a single Daily Report reader.
Why was I not surprised at this week’s stuffing? Whenever we do a poll on a particularly controversial topic, the odds of stuffing increase. But there’s another reason too: The local unions seem especially eager to try to skew the results. When we did a poll on the Employee Free Choice Act two years ago, the Rochester and Genesee Valley Area Labor Federation sent an e-mail providing a link to the poll and asking its members and supporters to vote their support for the EFCA. A copy of that e-mail found its way to my inbox.
I wasn’t on the mailing list this time. But with the poll’s trackable links, we were able to trace the stuffing back to an official with the New York State United Teachers union.
I’ll give the teachers union an E for effort: The poll responses totaled nearly 3,000—roughly twice as many as the previous high. But of those, more than 2,200 were stuffers. Unfiltered, the results showed 74 percent of respondents with no confidence in Brizard compared with 68 percent who do have confidence in Urbanski. With the ineligible respondents removed, the level of confidence in the superintendent rose to 74 percent—and the RTA chief fell to 7 percent.
To repeat what I’ve written before: We don’t think our readers want to see who can best mobilize support on a particular issue. Rather, we think they want to know where the business community honestly stands on key issues.
The vast majority of Snap Poll participants do not attempt to skew the tallies. To them, let me say thanks.
As for those who tried to manipulate the poll this week, I’ll remind them again about repeating a mistake but expecting a different result. You have to wonder if the need for that reminder also says something about what Brizard is trying to change in the city schools.
2/25/11 (c) 2011 Rochester Business Journal. To obtain permission to reprint this article, call 585-546-8303 or e-mail firstname.lastname@example.org.