Apparently I struck a nerve with some on the left with my poll question on collective bargaining. Not only did Left-wing Cheerleader-in-Chief Chris Fitzsimon decry the results as "push polling", but the talking points were passed on to UNC Professor Andrew Perrin (who is basically a government-paid lobbyist for the labor unions) who accused us of the same on his blog.
Take a look at the definition of "push polling" by the polling community standards (and interestingly enough, it’s the one Perrin links to on his blog.
Basically, a "push poll" is something that is not really a poll, but calls large quantities of people (usually voters) and spreads false information in order to influence public opinion, i.e. "Would you be more or less likely to vote for John McCain if you knew that he had fathered 6 illegitimate children?"
Since we conducted an actual scientific telephone survey of a random sample of voters and published the results (that Fitzsimons likes to use when results turn out his way, btw), our poll does not fit the definition of a "push poll."
However, what our poll does do is it tests different messages, and that is perfectly legitimate. Part of the debate in the public policy arena is how issues are framed and many issues can be framed many ways. My question on collective bargaining framed the policy question in the terms of organized labor’s participation in elections. It was just another way to frame the debate on the issue of collective bargaining.
Unfortunately, Fitzsimon and Perrin are falling into the usual left-wing play book of using demonization and scare tactics to distract you from a poll result that they disagree with. It is sad that they have to resort to this to get their message across. If either would like to have a real debate on the perils of collective bargaining and the influence of big labor in our elections, I welcome it any time. (As an aside, isn’t it funny how Fitzsimon can on one had decry the influence of unregulated money in the political system and the need for clean elections, yet turn a blind eye to the millions of unregulated money spent by labor unions.)
I don’t understand Perrin’s comment here:
“Last week the John W. Pope Civitas Institute, which used to be a fairly interesting, principled Libertarian-oriented think tank on North Carolina Politics, but has recently veered toward toeing the Republican Party line, released a press release via their Red Clay Citizen blog claiming widespread opposition to allowing public employees their collective bargaining rights.”
I’d like to know when Civitas has ever been perceived as more libertarian, and when we became somehow perceived as less so. (BTW, this has nothing to do with allowing a “right” of collective bargaining, which is not inherently libertarian.) Allowing the government collectively to bargain is not libertarian at all. It’s just a recipe for dangerous, unholy alliances between politicians and bureaucrats that is fundamentally undemocratic. I guess the Civitas-qua-libertarian line was a non sequitur and had very little to do with this issue.
Chris and I have had a lengthy email correspondence on this, which I won’t bore you with here. Of course, there is no truth whatsoever to the claim that I am a “government-paid lobbyist for the labor unions,” but that is neither here nor there.
The issue is whether Civitas’s polling misled respondents into giving the answers Civitas wanted. The answer is unequivocally “yes”, as is obvious from the question wording itself. Thankfully, I have been polling this question since 2005 using *neutrally* worded questions, and my results are therefore both more trustworthy and dramatically different from those of Civitas. This is not about results I disagree with; it’s about accurately measuring the public’s beliefs without priming them ahead of time.
Finally, Chris gets the order of events wrong. I did not get “talking points” from Chris Fitzsimon or anyone else — certainly not three years ahead of time, when I began polling on this question! I had poll results that were relevant to a current policy debate, so put them out there. That the Pulse finds them useful is gravy.
I also have been polling collective bargaining since Feb. 2007 and here are the results:
Feb 2007:
Should state government employees be able to unionize and engage in collective bargaining in North Carolina?
Yes – 38%
No – 46%
April 2007:
The International Labor Organization, a United Nations Agency not recognized by the United States, has ruled that state employees in North Carolina should have the right to unionize, engage in collective bargaining and strike if necessary. Do you support or oppose this ruling?
Support 37%
Oppose 47%
Andy,
Your results are more “trustworthy” because you polled them with wording you feel is more neutral and get results you agree with? Please.
In the interest of full disclosure, how much have you or your Department at UNC received from unions or union-affiliated organizations to conduct your polling or for your research and advocacy of this issue?
Andy,
Perhaps as a follow up to your “neutrally” worded question on unionization and collective bargaining for state employees, have you ever asked respondents:
“Do you feel there is a potential conflict of interest if state employees could unionize and collectively bargain with the same politicians their union dues are used to endorse?”
Or maybe:
“Imagine this scenario: a state employee union donates millions of its members’ union dues to a gubernatorial candidate. This candidate is elected, and is now influencial in determining the level of pay raise those same state employees will receive. Do you think this would present a conflict of interest?”
Perhaps that could better gauge the mindset of the respondents and provide more insight into their answers.
Then, maybe ask the original “neutrally” worded question again and see if the results change.
As a sociologist, I would think you’d be interested in testing the public’s perception of such an arrangement.
1.) I think my results are more trustworthy because the question I asked was more neutrally worded, not because of the answer I got. In fact, as an individual citizen I would have preferred that far more support collective bargaining rights, but I am bound by the data. Readers can determine for themselves which of the questions more correctly approximates the question legislators are being asked to evaluate in HB1583.
2.) I have received no funding, either personal or professional, from any union or labor organization. I make no secret of my general belief, as a citizen, that all workers ought to have the right to organize into unions if they so choose, and I have been a member of several unions at different times in my life, all by choice. I do not have access to the full funding profile of the UNC sociology department, but I do not know of any union support for the department or any of my colleagues in it. If there is any such funding, it is certainly a trivial fraction of the department’s overall budget.
3.) Civitas clearly believes the principal rationale for collective bargaining is corruption; that explains the loaded question as well as the others proposed by Brian Balfour above. I see no evidence that this is the primary issue, hence I see no reason to foreground it above all others (including, principally, freedom of association and freedom to contract). There are 48 states where collective bargaining is *not* prohibited by statute — perhaps we could learn something by examining these states’ experiences?
4.) Chris asked me by email about the details of my polls. They are not RDD polls, as he surmised, but rather polls of registered voters in North Carolina. In each case, the collective-bargaining question was one of many questions on public policy and political issues in the state. I have been doing this–first as a part of the Carolina Poll, more recently on my own–as part of my ongoing research about the character of public opinion. This is the only question that deals with collective bargaining on the polls. All the Carolina Poll data should be available through the Carolina Poll, which is run by the UNC School of Journalism; the two recent polls will be released publicly once I and my graduate students have completed our academic research using them.
Andrew,
I am confused, if the school of Journalism is conducting the “Carolina Poll” who pays for “…more recently on my own–as part of my ongoing research about the character of public opinion.” Do you pay for the poll personally? That is going the extra mile for the cause if you do. You are to be applauded!
I have paid for part of it out of my own personal funds, part of it out of unrestricted research funds through my department. None has come from any union, corporation, or other lobbying entity.
“Civitas clearly believes the principal rationale for collective bargaining is corruption; that explains the loaded question as well as the others proposed by Brian Balfour above. I see no evidence that this is the primary issue,”
Why are my proposed questions “loaded?” Do you deny even the possibility of the scenerios asked about? Are you saying there is absolutely no potential conflict of interest at all? Are you really that naive?
Don’t you think it would be interesting to ask if your respondents think there is the potential for a confict of interest? How is that “loaded?” Are you that intellectually uncurious about a topic you obviously care so much about?
Brian, I wrote that the question on the existing poll was loaded. The questions you suggest are also loaded in the sense that they ask respondents to consider only one element of the question. Note that a loaded question isn’t necessarily a problem, it’s just disingenuous to use one, then claim that it represents public opinion toward the *policy* as opposed to toward the spin you gave it.
To my mind, the question of whether these are conflicts of interest is a technical question, not one of public opinion. Citizens may believe there are great conflicts of interest in cases where there are none, and vice versa. “Do you believe the sky is purple” is not a useful question for a poll since it’s a matter of fact, not opinion.
As for the material question of whether there are conflicts of interest: there are lots of interested parties who participate in state politics, including endorsements, donations, and public statements. Businesses do this all the time in the hope of securing more business for themselves, better infrastructure, better business climate, etc. Removing the government prohibition against public employees doing the same thing does not introduce any greater conflict of interest than those that are already present; one could even make the argument that it levels the playing field since other interests are already permitted to do so.
“Citizens may believe there are great conflicts of interest in cases where there are none, and vice versa.”
But you have no interest in gauging the public’s opinion in this case? How convenient.
If you are truly interested in getting the public’s opinion of state employees unionizing and collectively bargaining, is it not important for them to be informed? Should they not be made aware of the relationship between the state employee unions and politicians? Don’t you think providing more information about the situation you are polling on would provide more informative results?
“Removing the government prohibition against public employees doing the same thing does not introduce any greater conflict of interest than those that are already present”
Of course it does. The government is an agent of the people. Businesses, unless there is corruption or rent-seeking are not intimately tied to government. They are certainly not agents of the people.
There is no rational argument for the notion that government employees have some “right” collectively to bargain when they act as agents of the people. And while favor-seeking is a phenomenon of government-to-business collusion, you seem wildly optimistic that this would not happen among government employees and politicians. That’s downright pollyannish (and dangerous). And it’s the equivalent of arguing that referrees at sporting events should be able to negotiate with fans or teams. Politicians are stewards of the rule of law and bureaucrats execute or implement the law in some way. Unionizing them creates a level of incest that is an affront to the rule of law. Never mind that unions must yield to economic realities of business revenue. Governments have the power continuously to tax!
Brian’s retort is simply silly. It’s not “convenient” that I’m interested in measuring the public’s actually-existing attitudes toward collective bargaining instead of the imaginary hypotheticals he can dream up. If Civitas wants to poll on those, be my guest — but don’t claim that these leading questions reflect the public’s views on collective bargaining, which is precisely the claim Civitas made in its press release.
Max: the vast majority of public employees are not “bureaucrats,” and existing labor law prohibits managers from unionizing anyway. Let’s be clear about whom we’re talking about: teachers, bus drivers, trash collectors, librarians, technicians…. people who perform a service for pay. Allowing them freedom of association and freedom to contract — which 48 other states do — is not “an affront to the rule of law,” it’s simple fairness. When the government acts as an employer it ought to be subject to the same pressures as are private employers — one of these is that it ought not be artificially shielded from employees’ desire to organize.
Andrew,
You may be interested in changing the wikipedia entry for “push poll” to reflect your view of meaning of the term.
Currently, the entry for “push poll” includes the following: “The term is also sometimes used inaccurately to refer to legitimate polls which test political messages, some of which may be negative.”
Brian M.: we’ve been around this block a few times now. I didn’t say it was a push poll, I said it “verged” on one. I’ve also offered privately to Chris H. to retract that statement, and I will do so publicly here, since clearly it’s the name, and not the substantive critique, that has y’all in a huff.
The substantive point remains the same: that the poll’s point was to come up with the answer Civitas hoped for, not to measure citizen opinion accurately.
Hoped for answer? Here is the offending question:
“Should state employee and teachers’ unions be allowed to engage in collective bargaining with elected officials they might endorse in elections?”
I don’t think Chris’s aim here was to determine popular sentiment about unionizing government employees. He wanted to gauge sentiment about a potentially incestuous aspect of such. I don’t see anything wrong with this. Does anyone deny the claims in his question? If you have a problem with it being “informational” or “message tested”, then you will have a field day with PPP or Elon, both of whom commonly use this technique (sometimes with bald innaccuracies – as here:
http://redclaycitizen.typepad.com/redclay/2007/07/mark-ryan-bias-.html
It’s no different from a lefty question worded as follows:
Should companies be allowed to contribute to the campaigns of politicians whose future legislative actions would directly benefit said company or hurt the company’s competitors?
I’m no expert, but this doesn’t seem seem illegitimate, only context-dropped and incomplete – which is the nature of many poll questions. But the main point is: if yours is the litmus test for what constitutes verging on a push poll, then please turn your critical lenses to PPP and Elon, both of whom routinely employ these kinds of questions.
I don’t think the question itself is illegitimate – if it’s message testing, so be it. That’s why I offered to retract my mention of push polling. What it *is* is intentionally leading, as the very concept of message testing suggests. The question really is: “how can we frame collective bargaining to insure public opposition?”. This is not in itself illegitimate — but Civitas’s press release (http://www.jwpcivitasinstitute.org/media/press-releases/poll-nc-voters-oppose-collective-bargaining ) is quite misleading since it claims that voters oppose collective bargaining when the data show nothing of the sort.
As a way of cutting to the chase, Chris “Simple Simon” Fitzsimon is a moron, a hypocrite and a paid lapdog for the Goodmon family and their increasingly bizarre social agenda. Simple Simon has zero credibility.