Last month, two advocacy groups, Independent Jewish Voices (IJV) and Canadians for Justice and Peace in the Middle East (CJPME), released the results of a poll that appears to show that the overwhelming majority of Canadians support sanctions against Israel over its continued occupation of Palestinian territories.
Based on this survey, IJV and CJPME conclude that the government’s recent objections to a boycott of Israel and its pro-Israel voting pattern at the United Nations are inconsistent with what most Canadians want. But is it true that Canada’s policy on Israel is out of step with public opinion?
While the survey was conducted by Ekos Research Associates, an experienced polling firm, and is said to consist of a representative sample of the population, it is not without methodological flaws that raise questions as to its findings.
First, the survey asked respondents about the “reasonableness” of sanctions against Israel, not whether respondents support sanctions against Israel. Suggesting that a proposition is reasonable is not the same as supporting it. In this way, the survey’s questions do not directly address the question of public support for a boycott of Israel. Nor do they address the government’s position on the Israeli-Palestinian conflict.
Second, the survey’s questions are ordered and worded in such a way that might bias the results in favour of positive answers. This is known as “priming.”
Depending on the objective of the survey, priming can serve an analytical purpose. However, if not handled properly, it can produce erroneous results.
‘the pollsters primed their respondents to answer positively the subsequent questions on whether Israel specifically should be sanctioned’
The Ekos survey shows evidence of priming in at least two areas. Respondents were first asked a general question about whether it is “reasonable” for Canada to sanction foreign countries that violate international law or human rights. The answer to this question was predictably high. Ninety-one per cent of respondents answered “yes.”
In doing so, the pollsters primed their respondents to answer positively the subsequent questions on whether Israel specifically should be sanctioned for violating international law. Not surprisingly, the response was high. As research in the field of social psychology has demonstrated, people seek to minimize hypocrisy and inconsistencies in their political and ethical beliefs.
Evidence of priming also appears in the wording of the questions specific to Israel. These questions reference UN Security Council resolutions and International Court of Justice decisions that decry Israeli violations of international law. In effect, these questions already contain a hint of the desired response. To suggest that Israel shouldn’t be sanctioned or boycotted goes against the precedent set by legitimate international institutions. Predictably, respondents again answered overwhelmingly in the affirmative.
Pollsters have widely available tools to avoid sources of potential bias. For example, Ekos could have used “control” groups in its survey, dividing its sample into two equally sized groups and asking one general questions about human rights and the other Israel-specific questions.
Similarly, Ekos could have eliminated any suggestive language about sanctions already imposed for violations of international law. Both would have avoided priming respondents for affirmative answers.
Finally, the survey touches on only one part of the campaign to boycott, divest and sanction Israel (BDS), and arguably its least controversial aspect. Not only do supporters of BDS call for a boycott of products made in the occupied territories or sanctions against the Israeli government, they also call for a boycott of Israeli civil society, including activists, academics, artists, and musicians, many of whom oppose the occupation. This strategy is especially contentious and something on which the survey results shed no light.
By not incorporating more robust techniques and asking more appropriate questions, Ekos has conducted a survey of questionable value. Indeed, we learn little from it regarding the population’s opinion on the Israeli-Palestinian conflict.
But more generally, dubious polls help push the profession further into disrepute. Pollsters keep getting it wrong – in the recent U.S. presidential election, Brexit, and other instances. If we want to better understand public opinion it’s crucial that we go beyond superficialities.
David Zarnett and Jamie Levin hold doctorates from the University of Toronto’s department of political science. Their work focuses on the international politics of the Israeli-Palestinian conflict.