Marc Ambinder has been having a good back-and-forth between Republican pollster Mike Murphy and labor pollster Guy Molyneux on the Employee Free Choice Act. The discussion is revealing of how structuring a poll can influence response, and how it's not really worthwhile to write about polls without seeing their actual questions and the cross-tabs. It also shows that the conservative message on the legislation is dependent on misdirection. Consider Murphy's comment here:
"I think Guy misses the big point here; the part of EFCA that really counts is the likely elimination of the secret ballot in most future union organizing elections. So, as a union member myself (WGA-West, AFTRA), I'll make this offer: If the AFL-CIO/Hart Research team re- tests the same ballot question they have released to the media, but adds the critical phrase "which would eliminate the secret ballot workers now utilize in most union organizing elections" to their question and then release the findings from this more accurate question to the media, I'll chip in $5,000 toward the cost of the conducting this more poll."
Oh, sure, except that the Employee Free Choice Act doesn't eliminate the secret ballot. It's categorically false. Marc notes as much here. Molyneux makes that point here. I would even recommend that folks check out the National Labor Relations Act, and particularly this provision, which is unmodified by EFCA:
(e) Secret ballot; limitation of elections (1) Upon the filing with the Board, by 30 per centum or more of the employees in a bargaining unit covered by an agreement between their employer and a labor organization made pursuant to section 158 (a)(3) of this title, of a petition alleging they desire that such authority be rescinded, the Board shall take a secret ballot of the employees in such unit and certify the results thereof to such labor organization and to the employer.
That means, of course, that if only 30 percent of workers feel intimidated by union organizers -- a feeling that is not in the organizers' interest, incidentally -- they can demand a secret ballot. No doubt their employer would help organize that petition. But to the main point: Both sides are tweaking polling language to their advantage, but the preferred labor language -- "[a]llows employees to have a union once a majority of employees in a workplace sign authorization cards indicating they want to form a union" -- has the advantage of being essentially true if a bit simplified. The clause Murphy could add to the question truthfully is, "which would eliminate the option for employers to coerce employees into secret ballot election weighted in the employer's favor after a majority of workers have already expressed their preferences for a union." But I imagine that wouldn't poll so well, either.
The fact is that there is no really "private" option for labor organizing: With the so-called secret ballot, an unfair election process begins with one side having nearly every structural advantage; voting is done under the supervision of management after weeks of intense one-on-one meetings, and it's naive to think that the company in question doesn't know how most of their employees are voting. On the other hand, under majority sign-up, union organizers have a good idea of how people vote but once the cards are delivered to the National Labor Relations Board, employers generally don't see them. Co-workers probably know which way their colleagues are leaning no matter which mechanism is used. Neither process is perfect, and I would err on the side of giving workers more choice, not less.
-- Tim Fernholz
