Asymmetrical Opinion Structures and Attitudes toward Immigration in Canada.

This paper examines public opinion towards the restriction or expansion of immigration in Canada, using data from the 2000 Canadian Election Study (CES). The paper is largely modeled on Citrin, et al (1997)’s "Public Opinion Towards Immigration Reform: The Role of Economic Motivations," although we make a methodological improvement that increases both the confidence of our results and our understanding of what drives not only opposition to immigration, but also for more immigration. In addition, we find that the variables explaining restrictionist opinions in the United States do not hold cross-nationally. Our methodological improvement is simple. Rather than using an ordered probit, we choose to use a multinomial logit. Because our three choices are exhaustive, we do not have to consider the irrelevance of independent alternatives. Thus, we can harness the principal advantage of the multinomial logit; namely, the independent measurement of the effect of each variable on each outcome. The model we use includes approximately 1200 respondents. Our initial results find that both personal and national economic evaluations are the principal factors causing opposition to immigration. This last significant result contradicts the American findings of Citrin et al. according to which personal economic circumstances play little role in opinion formation. In addition to challenging their main finding in regards to restriction, we generate a number of new findings on the factors influencing support for increased immigration. Most importantly, we demonstrate that these factors are not entirely symmetrical with the factors influencing support for decrease in immigration levels. Indeed, we find that national economic evaluations have much effect on support for decreased immigration, whereas it does not bear on support for increased immigration. Therefore our results demonstrate the value of employing a multinomial logit instead of an ordered probit in understanding public opinion towards immigration expansion or restriction. Moreover, the differences in significance of predictor variables between support for increase and decrease demonstrates that there can be various simultaneous processes at work in opinion formation on a single issue. [ABSTRACT FROM AUTHOR]