Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Dr. Scott Davies


Dr. Cyril Levitt


Historically, many studies of social inequality in Canada have tended to focus upon its ethnic dimension. Since the time of Porter (1965), most studies have been preoccupied with either supporting or refuting his Vertical Mosaic thesis. Research results are mixed. Some social scientists have found that, over the years, there has been a convergence in the earnings of ethnic groups; others have maintained that ethnic earnings inequality is persistent. Recently, proponents of the "visibility" thesis have argued that "colour" has replaced ethnicity in the structure of social inequality in Canada. Since the mid-1980s, the gender and nativity dimensions of ethnic inequality have been incorporated into the analysis of earnings differentials among ethnic/"visible"groups. There has been, however, a relative silence on the importance of class in this analysis. Even intersectionalist analyses that claim to interconnect class and gender with ethnicity are mostly qualitative, usually focus on small groups, often conflate "race" and class or examine only one class, gender and nativity group (immigrant women of "colour"). The present study is based upon the Public Use Microdata File on Individuals drawn from the 1996 Census of Canada. The class dimension is re-introduced and earnings differentials are examined, not only across, but also within ethnic groups, in terms of their class, gender and nativity segments. Evidence shows that the ethnic groups examined are not homogeneous or monolithic entities. They vary in terms of sex, nativity and class compositions. It is argued that despite the noticeable earnmgs differentials across ethnic groups, they have, with the exception of Jewish-descent respondents, similar class structures. The class and gender earnings differentials within them are greater than the ethnic earnings differentials among them. In addition, when class is introduced into the gender and nativity dimensions of ethnic earnings, the image of social inequality becomes more complex. The "visibility" thesis does not provide a completely accurate earnings picture. The earnings differentials within ethnic/"visible" groups are greater than those among them. The regression models show that human capital variables like schooling, labour market/production variables like class, weeks worked and industrial sectors, and the ascriptive variable sex, explain more of the variation in earnings than ethnicity or "visibility".

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Sociology Commons