About that EKOS poll

A recent EKOS poll found "the incidence of [Canadians] thinking there are too many visible minorities is up significantly and no longer trails opposition to general immigration (as it has historically)."

Here is a picture that shows the question EKOS asked, and the long-term trend in Canadians' responses to this question (click on the picture to make it bigger):

The poll generated a certain amount of discussion (e.g. here and here). Andray Domise, writing in Macleans, observed:

And given that Canada‘s immigrant population is fairly broadly distributed between a range of countries (unlike the U.S., for example, where the plurality of immigrants come from Mexico), there is no real basis in fact for this “too many visible minorities” sentiment. Only the deeply wonkish would have a working knowledge of the ethnic makeup of Canada’s immigrant population, which effectively made EKOS’ poll a litmus test on Canadian racism.

Deeply wonkish? We're here, and we hear your call!

The 2016 long-form Census asked millions of Canadians about their ethnic origins, and their immigration history. Using the 2016 Census Public Use Microfile, I estimated the portion of immigrants living in Canada who are members of visible minorities by year of immigration (click on the picture for a bigger image).

From 1885, when the head tax on Chinese immigrants was introduced, until the immigration reforms of the late 1960s/early 1970s, Canada did not welcome visible minority immigrants. However that changed in the 1970s and 1980s. The portion of immigrants who are members of visible minority groups has been hovering around 80 percent for 25 years.

For those who are interested in a detailed breakdown of immigrant cohorts into various visible minority groups, here is a .pdf with a detailed breakdown: Download Composition of immigrant cohorts by visible minority status. As these are survey numbers, they are subject to some margin of error. Small year-to-year fluctations in the composition of immigrants mostly reflect sampling variation. However large fluctuations are meaningful, such as the increase in Chinese immigration in the late 1990s prior to the handover of Hong Kong, as are the long-term trends.

 

 

Comments

[Jim Rootham]

Side track: How robust is that polling? It's only one data point.

[Frances Woolley]

Jim, "The margin of error associated with the total sample is +/- 3.0 percentage points, 19 times out of 20."

[Bob Smith]

"Only the deeply wonkish would have a working knowledge of the ethnic makeup of Canada’s immigrant population"?

What cave does the author live in? Anyone who resides within driving distance of Canada's major urban areas- which both account for the bulk of Canada's population and attract the bulk of our migrants - would have a "working knowledge" of the ethnic makeup of Canada's immigrant population, because it has materially changed the demographics of those regions over the past 30-odd years (a fact reflected in your data). People might disagree with whether that's good or bad, but you have to be fairly oblivious not to have noticed it.

I'm also not sure the data confirms the "Canadian are racist hypothesis". That the percentage of Canadians who say that there are too many immigrants and that there are too many immigrants who are visible minority are more or less the same doesn't tell us very much (note the way I phrased that, I come back to this interpretation below). Presumably, one might expect that if Canada was filled with racists, the percentage who says that there are too many visible minority migrants would be greater than the number who say that there are too many migrants (some Canadians want immigrants, just from the "right" countries). Had polls been done 120 yeas ago, that's what public opinion - reflected in our expressly racist immigration policies - would have have reflected. Had you asked the question 120 years, few Canadians would have said we had too many immigrants (reflected in government policy to promote immigration) but most would have said there are too many visible minority (they wouldn't have used that term) immigrants (reflected in concerns over, inter alia, Chinese, Sikh or Japanese migration).

Moreover, a whole lot turns on how the second question is interpreted. Literally, it's asking you to consider the hypothetical of holding migration constant and should we reduce, increase, keep the same the number of visible minority migrants, i.e., should we change the percentage of migrants who are visible minorities. Now, interpreting the "too many" response as being an indicia of racism raises the question how we treat the "too few" - are 12% of Canadians racist against whites? It also raise the question of whether "visible minority" status might be a proxy for something else (culture? housing prices in Vancouver/Toronto? who knows). But setting that aside, that's actually a fairly complicated question - asked in the context of a robo-dialed phone call - which might be understood (and I say this, because this is how I first read it) as asking whether we should increase, decrease, or keep the same the number of visible minorities migrants, period (rather than as a percentage of all migrants) - which is how I first interpreted the answer, above, before going back and reading it more carefully - in which case, you would expect opposition to migration and opposition to migration of visible minorities to coincide (as they do).

It is telling that the poll notes that opposition to migration increased systematically when they switched to the IVR methodology in 2009 (presumably they were live calling previously - they adjusted the past number to reflect that "bias") which does raise the concern that what we're seeing is a function of their poll methodology - people are misunderstanding the question being asked (perhaps in the context of a live call, under the old methodology, they might be able to clarify the question). It's also suggestive that the gap that had previously been reported largely disappeared post-2009, after they changed methodology. It does raise questions of whether they're really comparing like with like.

[Yildo Y]

Anyone who resides within driving distance of Canada's major urban areas- which both account for the bulk of Canada's population and attract the bulk of our migrants - would have a "working knowledge" of the ethnic makeup of Canada's immigrant population, because it has materially changed the demographics of those regions over the past 30-odd years (a fact reflected in your data).

Yes and no. Eastern European immigrants like myself pass for white and for non-immigrant in the Canadian context, and are therefore invisible to a casual eyeball survey.

[Francis Graves]

A few comments . On the issue of 'bias' that is a mode effect introduced by the shift from live interviewer to IVR . We conducted random controlled experiments to calibrate the effect . We also showed the methods generated highly similar results once this social desirability effect was held constant. This suggests that the questions were understood in a semantically consistent fashion . I actually think the validity of the IVR method may be higher due to the relative absence of social desirability. We also explicitly noted this and adjusted for it , unlike many other reports on this topic which have spuriously claimed shifts which were artefacts of mode effects.
I am not clear on the issue of ethnic fluency but we have tested that in the past and it really doesn't make much difference to this research. This is based on perceptions not objective reality . Anecdotally we have found as strong positive association between exaggerated estimates of the size of ethnic groups and racial intolerance (we avoided the term racism in favour of racial discrimination). The 5 point shift from the last poll was statistically significant as was the larger ten point shift over the past decade. as was the disappearance of the gap between the two indicators over that period. The shifts are not noise but link to the data in an intuitively and theoretically plausible manner. The most striking finding was the widening gulf across partisanship which could not be explained by people misunderstanding the question (which is an imperfect measurement for sure). We note that this huge polarization is consistent with the American literature showing a similar widening gap on related measure across Democrats and Republicans in the US over the past couple of decades. If this was simply misunderstanding (and I am sure there is some of that ) we would not be seeing clear patters but rather random noise. These results also mirror more rigourous social distance experiment showing similar effect when source countries are randomly rotated and people are asked to rate the acceptability of different types of immigrants moving into their neighborhood. We are trying to locate these findings in the latest work of authors such as Pippa Norris and Ron Inglehart , Karen Stenner and Jonathan Haidt and others studying the rise of authoritarian populism. We will reporting the more in depth work in the near future. Thanks for your interest and comments Frank Graves

[Francis Graves]

One last comment on the issue of how respondents understood visible minority , which is clearly an increasingly fuzzy concept (we have 25 years of tracking so we are loathe to change the wording and disrupt the time series). We conducted another random experiment rotating 'not white' for visible minority and it produced the same result . The purpose was to have a test whether racial discrimination is contained in some opposition to immigration . There are better tests but this one clearly shows that there is, it may be rising and that is definitely polarizing dramatically across partisan lines. This merits noting . It also is , for the first time in our research , a strong predictor of voter mobility since the last election . This is new and concerning as immigration has never been a serious ballot booth issue in Canada . Thanks

[Frances Woolley]

Frank - thanks for those comments and clarifications.

I believe that we have the same mission: to inject some factual background into the public debate. As you mention in your comments, one of the key factors associated with racial intolerance is exaggerated estimates of the size of various ethnic groups. This is why it's so important to make data available.

To be honest, what's surprising to me is the degree of tolerance, rather than the degree of intolerance. I'm thinking, for example, of the number of older Canadians, who grew up in very different Canada, and yet are are totally accepting of the interracial marriages of their children and grandchildren. And the group of kids for whom the fact that one of their members is Muslim is simply a trivial constraint on their choice of pizza toppings.

[Bob Smith]

"Yes and no. Eastern European immigrants like myself pass for white and for non-immigrant in the Canadian context, and are therefore invisible to a casual eyeball survey."

Two observations. First, White immigrants are typically not "invisible" to a casual survey. They may be invisible to an "eyeball" survey, but would be obvious, for example, if they spoke English with a pronounced accent (which, obviously, might not always be true, but is hardly uncommon).

Second, given that non-whites make up 80% of immigrants - as they have for decades - even if a mere "eyeball" survey excluded "white" migrants, the observation that the overwhelming majority of immigrants are non-white remains correct.

"To be honest, what's surprising to me is the degree of tolerance, rather than the degree of intolerance. I'm thinking, for example, of the number of older Canadians, who grew up in very different Canada, and yet are are totally accepting of the interracial marriages of their children and grandchildren. And the group of kids for whom the fact that one of their members is Muslim is simply a trivial constraint on their choice of pizza toppings."

Frances, I agree with you 100%. To my mind, it's a testament to the tolerance and openness of Canadians that we have undergone a fairly radical (I don't intent that term to have a pejorative tone) change in the ethnic and cultural make-up of our country over the past 40 years without much in the way of serious opposition or complaint. Visible minorities have gone from 4.7% of Canada's population in 1981 to 22.3% as of 2016. Europe has seen far less radical demographic change and it has provoked the rise of mainstream far-right anti-immigrant parties which are non-existent in Canada (our FPTP system also serves us well in this regard), and of course that sort of change is simply unimaginable in countries like Japan or China (imagine the reaction in Japan if 20% of its population became Gai-Jin). Though there's precedent for this, since we evolved from a largely British/French country at the end of WWII to one that, while still largely white, had incorporated people from a much broader assortment of ethnic and cultural backgrounds over the subsequent 40 years. I don't think people realize just how unique our relatively harmonious multi-ethic/racial society is.