Quinnipiac’s "Unbelievable" Poll on NYC Bike Lanes
"The majority of New Yorkers support bike lanes. According to the most recent Quinnipiac poll, 54 percent of New York City voters say more bike lanes are good 'because it's greener and healthier for people to ride their bicycles,' while 39 percent say bike lanes are bad 'because it leaves less room for cars which increases traffic.'" Given that a recent article in the Wall Street Journal reports that the mayor and his administration may have been spinning safety data to support the bike lane policy, you might think that Mayor Bloomberg himself had conducted the Quinnipiac poll, or commissioned it, or at the very least influenced its results in some way or another. How else could you account for the finding that a majority of New York City voters actually said increasing bike lanes is good because it’s “healthier for people to ride their bicycles”? Is that the first thing New Yorkers really think about when asked about bike lanes – how healthy it is for the bike riders? If so, kudos to them, for thinking of others. I live in New Hampshire, and when I asked a whole bunch of friends (admittedly not a representative sample of anything, not even of all my friends) whether they would want an expansion of bike lanes around our area, not one person said “yes” because they felt it was healthier for other people to ride bikes. Well, as it turns out, most New Yorkers probably wouldn’t give that reason either, if they were allowed to express their opinions in their own words. It was Quinnipiac that actually formulated the response, which in legal terms might be called “leading the witness.” Stacking the Deck Even before asking the loaded question, however, the Quinnipiac University Poll stacked the deck in favor of bike lanes. They did so by first informing respondents that “there has been an expansion of bicycle lanes in New York City.” How does that stack the deck? For almost all issues, even the most contentious, there is always a significant segment of the population that is not engaged – people who couldn’t care less one way or the other what happens. But once the pollster pretends as though everybody is informed (and makes that appear to be the case, by actually informing the respondents in the sample), and then asks all respondents to opine about the issue, the sample is fatally contaminated. It no longer represents the general public, many of whom are simply uninformed or otherwise unengaged in the issue. A more objective approach would have been to ask whether the respondents even knew of the expansion. After all, Bloomberg’s senior advisor Wolfson noted that in the past four years, the city has added 255 miles of bike lanes, while the city has over 6,000 miles of streets. This would suggest that probably a lot of residents were not even aware of the expansion – perhaps especially people who travel mostly by subway, or others who are generally clueless about anything. How many city residents are not engaged in the issue? Based on polling I’ve done and seen on jillions of other issues over the years, I would estimate about a third to a half of New York City voters don’t really care one way or the other, either because they don’t see bike lanes as affecting them, or because they simply haven’t been paying attention to the issue. (Of course, this is only a guess. What we need is for the pollster to ask the question, rather than taint the sample by giving respondents information.) Nevertheless, Quinnipiac asked all sample respondents – regardless of their knowledge or engagement in the issue – the following question: As you may know, there has been an expansion of bicycle lanes in New York City. Which comes closer to your point of view: A) This is a good thing because it's greener and healthier for people to ride their bicycle [sic], or B) This is a bad thing because it leaves less room for cars which increases traffic. There are two major problems with this question. First, it is a “forced choice” question, which means that there is no explicit option for a person to say they have “no opinion.” This format is used by pollsters when they want to suppress the percentage of people who are classified as “unsure.” (One reason pollsters do this is that they feel it’s not newsworthy if, say, 30% to 40% or more of the public has no opinion.) How do people with “no opinion” choose an option when an interviewer pressures them to come up with one? Typically, they succumb to the phenomenon of “response acquiescence” – a term which means they usually respond in a positive, rather than negative, manner. If that happened here, and I’d be surprised if it didn’t, that would mean an overestimate of the percentage who support the expansion of bike lanes. Second, the question could have asked whether the respondents supported or opposed the policy (or had no opinion). But instead, the pollster put words into the respondents’ mouths, by giving reasons for each choice. That’s how Wolfson could say that a majority of New York City voters believed bike lanes were “healthier for people to ride their bicycles,” though it’s highly unlikely many people would have made that statement on their own. The net result of the question phrasing, I believe, is to lead the respondents toward selecting option A, because it sounds good to have a policy that is “green” and “healthy” – even though there is the minor inconvenience of increased traffic. Of course, I could be wrong on this – it’s a matter of opinion which of the two alternatives sounds more positive to someone who otherwise doesn’t know anything about the issue. But that’s the point – rather than ask a tendentious question, Quinnipiac researchers could have asked an objective one. They did not. So, can we believe the finding that a majority of New York City voters support the Mayor’s policy? Not so, I would argue, if the only source is the Quinnipiac poll. It’s possible that even a majority don’t have an opinion one way or the other – though a good poll would let us know. |
In response to 
No comments:
Post a Comment