[Apology: I wrote of Phil Weiss's headline, 'I wouldn’t look too closely at his headline, by the way, as the ratio of those opposing expansion to those supporting it is actually 3:2. ' The proportion of the entire sample is 3:2, but what he wrote was 'Influential Jews are against settlement expansion 3 to 1' [my emphasis], which it is.]
When J Street released the results of its most recent poll last week, Richard Silverstein welcomed the survey’s findings on his Tikun Olam blog, headlining his post, ‘J Street Poll: American Jews Believe Gaza War Failed, Support Hamas in Palestinian Government’. Similarly, Phil Weiss, entitled the post on his Mondoweiss blog, ‘First the good news: Influential Jews are against settlement expansion 3 to 1’. While I’m sure there’s a place for a glass half full approach, I don’t think they’ve read the results closely enough and may have misrepresented the outcomes in crucial respects.
To begin with, the
Gerstein Agne Strategic Communications designed the questionnaire for this survey of 800 self-identified adult American Jews, conducted February 28-March 8, 2009. The survey has a margin of error of +/- 3.5 percent; the margin of error in the split samples is +/- 4.9 percent. Gerstein Agne contracted the research company YouGovPolimetrix to administer the survey by email invitation to its web-based panel, which is regularly updated and consists of 1.2 million Americans.
An internet poll removes the looming problem of ‘wireless households’ – those that don’t have a landline connection – but at the same time, it restricts the population just to those who use the internet.
Advances in technology and sophisticated web-based panel techniques have greatly helped researchers seeking to gain a trustworthy understanding of small populations, such as American Jews, and web-based panels are a rapidly growing method across numerous audiences that are difficult to reach by traditional land line telephone surveys.
I must confess, I’m dubious about this methodology. It seems to me that it would be impossible to calculate the probability that those responding display the same attitudes as those entirely outside the sample frame with no chance of selection with the same confidence that you can in a traditional random sample where each sampling unit has an equal chance of selection. Furthermore, Gerstein is not explicit about how YouGovPolimetrix identified Jewish respondents or constructed their panel, nor what proportion of those selected actually participated. Omissions like this do little to ameliorate my doubts. If the pollsters have developed some method of imputing the responses of those outside the population sampled and overcoming the bias that arises from non response, they’re not telling us how it works.
In their code for ‘Conducting market and opinion research using the internet’, ESOMAR World Research ‘the world organisation for enabling better research into markets, consumers and societies’, insist,
Users of research and the general public must not be in any way misled about the reliability and validity of Internet research findings. It is therefore essential that the researcher:
a. follows scientifically sound sampling methods consistent with the purpose of the research;
b. publishes a clear statement of the sample universe definition used in a given survey, the research approach adopted, the response rate achieved and the method of calculating this where possible;
As evidence of the poll’s reliability, Gerstein asserts,
It is important to note that the demographics (such as denomination, synagogue attendance, age, region) and political measures (party identification) in this survey reflect those in other surveys of American Jews, including the 2007 Pew Forum on Religion and Public Life, the 2008 American Jewish Committee Annual Survey, and the 2000-2001 National Jewish Population Survey.
The age profile of the population polled in March does not match the sample in the July iteration of the
The AJC’s Annual Survey of American Jews, conducted in September last year, which does not report age, found 39% were ‘Slightly liberal’, ‘Liberal’, or ‘Extremely liberal’, while only 33% of the J Street poll respondents said they were ‘Liberal’. If we include those identifying as ‘Progressive’ in the
The proportion identifying their denomination as ‘Reform’ in the AJC poll was 30%, but 34% in this poll, while ‘Conservatives’ comprised 29% and 25%, respectively.
Some of these discrepancies are within the margin of error, but they do raise doubts about how representative the sample was. Still, I’ll assume they know what they’re doing and will take their results, including stated ‘margin of error’, at face value.
I never bothered trying to analyse the results of the July poll largely because the questions were so unwieldy that I didn’t think it would reveal much of interest. Granted, in an internet poll, where respondents can review the wording until they’re sure they understand the intent of the question, there may be scope for asking more complex questions. Still, demanding close reading of long and complicated questions exacerbates ‘respondent load’ and is generally considered bad practice. But more importantly, such questions make it difficult – sometimes impossible – to determine the respondent’s intent. Although the questions in this month’s iteration still leave a great deal to be desired, they are a big improvement over July’s.
Phil starts out,
First the good news. 60 percent of American Jews are against expanding the settlements and the same number say the
Assuming these results really do reflect American Jewish opinion, it’s mildly encouraging that 60% are against settlement expansion, but it’s hard to get excited when 40% support expanding them, and they weren’t asked whether they supported dismantling them, which I think would be more informative. As Phil writes,
There's evidence of some obdurate attitudes among
The 72% who oppose expansion are a proportion of the 44% who said they gave money to political campaigns. Far be it from me to criticise anyone for reading into survey questions, but since we don’t know whether they donated five cents or 5 million dollars, I couldn’t leap to the conclusion that this population is uniformly ‘influential’.
Speaking of sub populations, Gerstein reports that ‘Orthodox Jews [who] strongly support settlements (80 percent support)’. What’s interesting about this is that only 8% of the sample – 64 respondents – claimed to be Orthodox. Ordinarily, I wouldn’t expect anyone to mention such a statistic, as the confidence that such a small sample represents the population it’s supposed to represent would be extremely low.
57% believe that in George Mitchell’s role as Israel-Palestine envoy he should be an honest broker, rather than an
Only 25% agreed with the second of the two options offered to half the sample in Question 44:
Bear in mind that the margin of error claimed for questions to a split sample like this is ±4.9%. What I found most interesting about this question was that 8% replied ‘Both’, and 10%, ‘Neither’. The two options offered do not exhaust the possibilities – they could have asked whether Mitchell ought to side with the aggrieved party. By omitting possible responses, the pollsters introduce an additional layer of bias into the survey, which is unconscionable. But while it doesn’t mitigate the bias, 10% at least had the opportunity to specify, ‘Neither’. Still, we don’t know whether these respondents answered as they did because they thought the US negotiator should side with the Palestinians, or because they didn’t agree that the US should be involved in negotiations (12% ‘oppose the United States playing an active role in helping the parties to resolve the Arab-Israeli conflict’), or that it should be George Mitchell.
Although Gerstein’s question departs from standard practice in failing to provide an exhaustive set of possibilities, it does adhere to the practice of ensuring that the possibilities offered are mutually exclusive. Clearly Mitchell can’t ‘act as a fair and impartial broker’ at the same time as he sides with
In the same vein, Silverstein reports that
Jews are willing to see Pres. Obama crack heads, if necessary, to achieve those ends. 86% are in favor of an active
That is not how I would read the results. First of all, apart from Q66 on reducing military aid, none of the questions say anything about ‘strong criticism and pressure’. In the context of US-Israel relations, of course, a harsh word might be perceived as such.
No fewer than nine of the 41 questions ((Q32-36, 39, 43, 44, 46) not counting responses categories as separate questions) canvassing views on political issues are specifically about ‘the
The 86% who said they’d tolerate ‘stating disagreements with the parties’ were a proportion of half of the 88% of respondents who had already agreed that the
In Gerstein’s analysis,
Not surprisingly, support for
Bearing in mind that these two options were presented to separate halves of the 88% supporting
Similarly, the 64% who would allow pressure on a refractory
The ‘almost half’ (49%) who ‘support reducing Israeli military aid if it is such a party’ – reducing, mind you, not eliminating – contrast with the ‘more than half’ who oppose it. Note that 59% supported reducing ‘humanitarian aid for the Palestinians if they block the agreement from being reached’ [my emphasis] and 75% support the blockade of
Q.62 Eight years ago, Israeli, Palestinian, and American negotiators came very close to reaching a final status peace agreement, but ultimately fell short.
The details of that agreement include: a demilitarized Palestinian state in the West Bank and Gaza; internationally recognized borders that include some land swaps allowing for most Jewish settlers in the West Bank to be inside Israel while the Palestinians get comparable land areas in return; Palestinian neighborhoods in Jerusalem become part of the new Palestinian state while Israel retains control of Jewish neighborhoods and the Western Wall in Jerusalem; international forces to monitor the new Palestinian state and border crossings; and financial compensation for Palestinian refugees while allowing some refugees to return to Israel if they meet specific family reunification criteria and the Israeli government approves.
Silverstein describes this as ‘a future Israeli-Palestinian agreement along the lines of the Geneva Accords’, which is true enough, but in the context of the wording about ‘eight years ago’, Phil describes it more accurately as ‘a peace deal along the Clinton parameters’. According to all accounts, the famous generous offer of eight years ago was nowhere near reaching agreement. Indeed, even abu Ammar could never have sold its take it or leave it provisions to Palestinians. Whichever version of the two state ‘solution’ Gerstein’s description resembles most closely, there is no question that it can or could deliver a just peace. It strips the refugees of their right to return and leaves the rump Palestinian state at the mercy of their predatory neighbour, protected only by ‘international forces’ who have been so effective in protecting
Land swaps are part of every proposal for partition of Palestine, in recognition of the ‘facts on the ground’ that Israel has created over the last four decades with the intent of establishing a ‘matrix of control’ over the Palestinians living in the West Bank and ultimately annexing the whole area. As I’ve argued somewhere before, to countenance land swaps is to provide retrospective legitimation for the whole settlement project, sending the unambiguous message that under ‘international law’ if you hold out long enough, you can get away with anything.
Furthermore, when Gerstein asks about ‘comparable land areas in return’, I’m pretty confident he doesn’t intend, and respondents don’t expect, that for every Israeli road transecting the West Bank, there will be a Palestinian bypass road interrupting Israel’s territorial contiguity; for every strategic hilltop settlement in the West Bank, there will a Palestinian outpost in ‘Israel proper’; for every precious aquifer annexed to Israel…well, you know what I mean.
It would be nice to comfort myself with the knowledge that 24% of respondents rejected such a ‘peace plan’, but I suspect most of those who did so had nothing like a just solution in mind.
Silverstein is disappointed that 75% approved (47% ‘strogly’) ‘of the recent military action that
I suppose it’s worth reiterating that to ask whether ‘Israel's response to Hamas' attacks was disproportionate’ is a trick question, as it invites the respondent to presuppose that Israel’s ‘military action’ was a response to Hamas rockets, when in reality the rockets were the response to Israel’s 4 November incursion. In any case, 69% agreeing with this statement means that at least 45% of respondents approved the massacre even though they thought it was disproportionate. Similarly, at least 31% approve even though it ‘creates more terrorism’, and 40% even though they disapprove of collective punishment. Small consolation, indeed. In a response to a comment to the post, Silverstein is also heartened that
there is a rise of 4% or so regarding issues related to treatment of the Gazans, lifting the siege, and openness to talking to a Palestinian gov’t including Hamas. It’s not an earth-shattering change, but is noteworthy nonetheless.
But since the questions he refers to were asked of a split sample, a movement of 4% is well within the claimed margin of error, and unlikely to be noteworthy at all.
In this connection, Silverstein reckons, ‘Unfortunately, there doesn’t seem to be any polling of Jewish support for the 2006
Reporting on the survey results, Gerstein effuses,
The survey also probed deeply into Jewish perspectives of this winter’s military action in
When I read this kind of self aggrandisement, I’m embarrassed for the author, but I suppose that’s what you have to do if you want to make it on
Phil Weiss reckons,
These Jews are for peace. 72 percent are for the
But as I’ve written, with ongoing military and moral support for
Phil writes, ‘
Leaving aside the racist crap about ‘the Jewish homeland’,
On the whole,
At the same time, it reveals that 100% of American Jews, at least those polled – whether they supported or opposed the
On the other hand, while 72% approve ‘of the way Barack Obama is handling the Arab-Israeli conflict’, 76% said they thought he supported
What this shows is that a huge majority of US Jews can somehow reconcile themselves to the basic assumptions that underlie support for
- That Jews are a ‘people’ for the purposes of exercising the ‘right to self determination’
- That to achieve that right, Jews were entitled to engage in terrorism and ethnic cleansing to achieve the desired Jewish majority
- That in exercising that right, it is acceptable to privilege Jews in terms of land tenure, national symbols, public holidays, language, etc.
- That it was ok for
to annex territory beyond that stipulated in the UN partition resolution (181) by force in 1948. Israel
And that’s the case for those who support
- Refugee rights are negotiable, and may only be exercised to the extent permitted by
- It’s fair, reasonable, and viable to leave an unarmed Palestinian state at the mercy of the most heavily armed state in the region
- Acquisition of territory by force in 1967 and consolidated through cynical settlement ‘facts on the ground’ is legitimate.
In those respects,
But it turns out that only 39% preferred negotiation with
But then, J Street isn’t all that dovish itself. In answer to the cynically worded FAQ, ‘Was Israel justified in attacking Hamas?’,
It does differ from AIPAC, however, on tactical issues. Where AIPAC has nothing but praise for
The more relevant question is whether
Similarly, while they seem to oppose Israel’s siege of Gaza, or at least its severity, this too is on strictly practical grounds tied to their perception of what best serves Israel’s ‘security goals’.
‘The ceasefire’, by the way, ‘was 6 months but began to unravel in November’!
There is a real difference between AIPAC and