I don’t mind that these tools commissioned a biased poll and then paraded it around as evidence of a sweeping revolution. I don’t think anyone who matters pays any attention to them, so there’s a limit to the damage they can do. Since the election they’ve gotten almost no press, which among other things has been denying me a hook for my “MJ Rosenberg got teased during Hebrew school so now he wants Obama to beat up the cool Jewish kids” post.
What I kind of resent is that I felt a professional obligation to waste five minutes of my life and check what kind of asinine stupid polling trick they used to rig their results. Obnoxious:
Q.33 (IF SUPPORT ACTIVE ROLE) (SPLIT A) Would you support or oppose the United States playing an active role in helping the parties to resolve the Arab-Israeli conflict if it meant the United States publicly stating its disagreements with both the Israelis and the Arabs?
Q.34 (IF SUPPORT ACTIVE ROLE) (SPLIT B) Would you support or oppose the United States playing an active role in helping the parties to resolve the Arab-Israeli conflict if it meant the United States publicly stating its disagreements with Israel?
Q.35 (IF SUPPORT ACTIVE ROLE) (SPLIT A) Would you support or oppose the United States playing an active role in helping the parties to resolve the Arab-Israeli conflict if it meant the United States exerting pressure on both the Israelis and Arabs to make the compromises necessary to achieve peace?
Q.36 (IF SUPPORT ACTIVE ROLE) (SPLIT B) Would you support or oppose the United States playing an active role in helping the parties to resolve the Arab-Israeli conflict if it meant the United States exerting pressure on Israel to make the compromises necessary to achieve peace?
“Split A” and “Split B” are randomized groups of 354 respondents, all of whom had previously indicated that the US should be engaged in the peace process. Why split them this way if you’re just going to ask the same question both times, except without the “… and Arabs” part? Especially if what you’re measuring is whether respondents support pressure on Israel, J-Street’s sine qua non?
Because they were hoping to get lucky and show that the same number of people want the US to pressure Israel regardless of how it approaches the Arabs.
Why not just measure that directly by asking the obvious third versions of the questions: “… the United States publicly stating its disagreements with the Arabs” and “… exerting pressure on Arabs”?
Because they didn’t know if they were going to get lucky. In fact, they might have gotten so unlucky that they’d get more “yes” answers to their Arabs-only version than to their balanced version. Think of how awkward that would’ve been.
Not that they’re actually measuring anything. The balanced versions don’t measure how many people are answering “yes” because they’re so enthusiastic about pressure on “the Arabs” that they’re willing to tolerate a little on Israel. The Israel-only versions don’t control for how many people assumed that the US would also naturally be pressuring the Palestinians. If the question had said something like “… pressure on Israel but not on Arabs” then at least the survey would be in the ballpark of merely badly designed. There would still be problems with the ordering of the parties and with priming and with positive vs. negative valance. But it’d be closer.
Those are points about the survey’s blundering stupidity though, and I don’t want to get into that because it would distract from the rank dishonesty
The support/oppose split for the balanced question was 86/14. The support/oppose split for the Israel-only version was 66/34. That’s a huge number of people being swayed specifically by how much pressure is getting put on Israel’s Arab enemies.
And that’s after those respondents had already said “yes” to a previous question about whether “the United States playing an active role,” even though they may have been thinking about US pressure on the Palestinians. To answer “no” in this question they would have to implicitly contradict themselves. Which respondents never want to do. Which is why we don’t write surveys this way.
So J-Street was actually pretty lucky to accidentally forget to ask an “Arabs-only” version of their question.
Example: I’m confident that everyone feels kind of embarrassed for MJ Rosenberg. He’s been mewing resentfully for weeks over the Freeman thing, wallowing in petulant fantasies of proxy revenge and writing diary entries filled with lines like “take that, neocons!” and “enjoy your weekend neocons… Obama is the guy [you] feared.” His delusions of relevance – apparently AIPAC staffers live in fear of his withering pen and his towering influence – are actually getting kind of uncomfortable.
Now imagine I took a survey where I split you into two groups. One group got asked “do you think that MJ Rosenberg is a moron for calling himself ‘pro-Israel’ but that he’s kind of a pathetic sop so we shouldn’t mock him?” The other group just got “do you think that MJ Rosenberg is a moron for calling himself ‘pro-Israel’?”
Since he is indeed a moron for calling himself “pro-Israel” I’d expect to get an overwhelming response to the first version. But since you’re all good people who are reluctant to speak ill even of his ilk, I’d expect a much more mellow ratio for the shorter version. The actual opinion – the one where everyone feels bad about the revelatory blustering of the parakeet journalist who got picked on as a kid – gets missed.
Now imagine I brazenly publish the results and announce “MR readers declare that MJ Rosenberg is a moron for calling himself ‘pro-Israel’.” That would be kind of like J-Street using this survey to insist that “American Jews support pressure on Israel.” Except in my scenario I didn’t have all their hackish priming tricks so what got reported would actually be true.
Anyway, if you have the patience make sure you check out the wording they used to discover – per their website – that “69 percent [of American Jews] support the U.S. working with a unified Hamas-Fatah Palestinian Authority government to achieve a peace agreement.” It’s like Greenpeace discovering mass support for a cap and trade system to “prevent the Earth from boiling,” minus the presence of at least a couple of legitimate experts who think that it might work.
References:
* Poll: Majority of US Jews back Obama’s active ME engagement [JPost]
* J Street Releases New Poll of American Jewish Community [J-Street]
* Pro-Carter Shill MJ Rosenberg: Why Aren’t Israelis More Excited About How Obama Is Going To Pressure Them? [MR]
* Freeman Withdraws Name [Rosenberg / TPM Cafe]
* Take That, Neocons: Obama’s Unprecedented Outreach to Iran [Rosenberg / TPM Cafe]
* Weeping Neocons: How Did This Happen to Us? [Rosenberg / TPM Cafe]
Previously:
* NJDC Tool Aaron Keyak Helpfully Illustrates How Liberal Activists Sneeringly Cocoon Themselves In Asinine Arguments And Dishonest Smears
* JStreet Tools Proudly Declare Their Inability To Distinguish Between Reality And What They’d Like Reality To Be
* Of Course: "Pro-Israel" J-Street Is More Anti-Israel Than Israel’s Sworn Enemies, Equates Israeli Self-Defense With Hamas Violence
Watcher’s Council Results
Well, the results are in on this week’s Watcher’s Council vote. There were some fine pieces submitted, and a couple of fine winners chosen. Council Submissions First place with 2 points! – Bookworm Room – Can it happen here? Second…