Responsible and irresponsible opinion poll reports

 When is a poll a "push poll"?

By Maurice Ostroff

With the surfeit of public opinion polls to which we are increasingly subjected, it is important to consider how to evaluate them. When we learn about a poll from media reports we are completely dependent not only on the credibility of the pollster, but also on the reporters'' understanding of the results as well as their ability to avoid injecting personal prejudices into their interpretations.
Obviously a poll has no value unless respondents understand clearly what they are being asked. Nor is there value in a "push poll" in which the pollster asks loaded questions which contain unjustified assumptions that "push" interviewees towards providing desired responses.
On October 23, Haaretz published an article by Gideon Levy under the headline "Survey: Most Israeli Jews would support apartheid regime in Israel", in which he presented his misinterpretation of a poll commissioned by the Israela Goldblum Fund.
Haaretz was irresponsible in publishing the article in view of the startling admission by the poll''s initiators in their press statement that questions were asked about apartheid, even though "it was not clear what these respondents understand by the term". Surely this crucial aspect should have alerted the editor to examine the poll''s credibility before rushing to print.
The duty of editors to establish the credibility of polls  before publishing their results is emphasized by the USA National Council on Public Polls (NCPP) which published a document titled "20 Questions A Journalist Should Ask About Poll Results.
The standard of journalism would no doubt be enhanced if every journalist and editor abided by the above mentioned NCCP''s guidelines as well as the "Esomar/Wapor Guide To Opinion Polls And Published Surveys" and the code of conduct issued by The European Society for Opinion and Market Research (ESOMAR) and the World Association for Public Opinion Research.
In addition to the flaws in the poll, Levy''s overtly biased interpretations are represented as facts, in flagrant violation of Article 11a of the code of ethics which emphasizes the importance of distinguishing between the results that emerge directly from the questions asked and any commentary or interpretation based on these results
The 503 interviewees comprised 60 Ultraorthodox Jews (12%), 57 Orthodox (12%), 143 Traditional (29%), 235 non-religious (47%) and 8 who did not respond. Although these percentages differ from their relative proportions in the general population no details were given of any compensating weighting procedures that may have been used as required by Article 5 of Esomar''s code of conduct.
Having ignored obvious flaws in the poll, Haartez compounded its irresponsibility by publishing Gideon Levy''s blatant misinterpretations as shown in the following examples.
a) The poll reveals the opposite of the misleading headline "Most Israeli Jews would support apartheid regime in Israel"
b) Levy''s statement "A majority also explicitly favors discrimination against the state''s Arab citizens" is directly contradicted among others by answers to question 6 which asked if respondents favor preventing Arabs from voting for the Knesset. Only 33% answered yes. 59% would oppose such a law and 8% don''t know. If the 8% who don''t know are excluded, the percentage of respondents who would oppose such a law increases to 64%.
c) Levy wrote "42 percent don''t want to live in the same building with Arabs and 42 percent don''t want their children in the same class with Arab children" ignoring the fact that a clear 53% overall majority and 68% of the secular respondents have no objection. An unbiased analyst would weigh the important factor that the average is skewed by the religious respondents who tend to live in insular religious neighborhoods because they would be unhappy to have Jewish or non-Jewish neighbors who do not observe the Sabbath. This applies wherever they live in the world and these preferences have nothing remotely concerned with South African apartheid.
d) Levy wrote "A sweeping 74 percent majority is in favor of separate roads for Israelis and Palestinians in the West Bank. A quarter - 24 percent - believe separate roads are "a good situation" and 50 percent believe they are "a necessary situation."
The above is far from a fair reflection of the poll''s results. "50 percent did not "believe they are a necessary situation". As will be seen from the details below Levy omitted to acknowledge they actually disapprove of the situation but believe that nothing can be done about it. He also omitted to add that a further 17% disapprove and believe the situation should be discontinued.
The facts are that question 17 states that in the territories there are roads on which only Israelis are allowed to travel and others only Palestinians. It asks "which of the following answers most closely equals your opinion?"  (The percentage answers are shown against each question and in brackets the adjusted percentage if the don''t knows are excluded)
It is a good situation 24%  (26%)
It is not good but nothing can be done about it 50%, (55%)
It is not good and it should be discontinued 17% (19%)
Don''t know 9%
It is obvious that contrary to Levy''s interpretation a sweeping majority of 67% (74% if don''t knows are excluded) disapprove of the situation, although some feel that nothing can be done about it. The poll failed by omitting to ask why they considered nothing can be done about it and under what conditions something could be done.
The sad part about this type of strident partiality about important issues is the counterproductive results. Yes, there are very necessary discussions in progress about what many consider to be the diminishing prospects of a peaceful two state solution and the demographic implications of a one state alternative.
Unfortunately, instead of engaging in serious debate about the fundamental issues and suggesting alternative solutions, many well-intentioned people have adopted the propagandistic language of the BDS movement in using the irrelevant apartheid bogey to bolster their arguments against annexation of areas in the West bank. They use this libelous canard even though, as admitted by the initiators of the poll, the respondents did not understand the meaning of apartheid and it is evident that neither do the initiators of the survey.
The objection to raising the apartheid canard could not be expressed better than by Professor Gideon Shimoni in an article titled The apartheid analogy: Lessons for Israel (Jerusalem Post Feb. 20, 2011):
"Apartheid, todays prime stigmatic code-word for racist evil, has become a potent weapon for delegitimizing and demonizing Israel, especially since it evokes the precedent of powerful external pressure in the form of boycott and sanctions as was applied against the apartheid regime in South Africa. Hence, in the propaganda war against Israel an equation is fabricated insidiously between the present State of Israel and the former apartheid state of South Africa.
This must be exposed as a malicious slander, and utterly refuted. It is also a crass abuse of the valuable lessons that might be learned from the odious apartheid experience of South Africa. There is no objective basis whatsoever for attributing to Israel the ideology, policies and praxis that were known as apartheid in South Africa.
Referring to the West Bank he added:
"Yet, no matter how morally deplorable, this is not apartheid: it simply is not the same phenomenon. If one is to draw lessons, Israels occupation regime is equally comparable to the situation in any number of other cases of post-war occupation or ethnic domination in deeply divided and conflict-ridden countries, not least of all in the Arab world".