Ethnicity and Class: Booming in Alberta & Saskatchewan

Feodor Snagovsky | University of Alberta | feodor.snagovsky@ualberta.ca

Michelle Maroto | University of Alberta | maroto@ualberta.ca

June 2, 2021

Introduction

Identity is central to how we see the world. How people see themselves affects how they see others, which in turn affects everything from friendship networks to political loyalties, attitudes, and behaviours. Two of the most common characteristics that people develop identities around are class and race/ethnicity. [1] Using data from the March 2021 Viewpoint Survey, this research brief provides an overview of class and racial/ethnic identity during a tumultuous time in Alberta and Saskatchewan. In the midst of a pandemic, both provinces were dealing with rising COVID-19 cases, high rates of unemployment, and continuing financial insecurity. As we show, class and racial/ethnic identities are important for Albertans and Saskatchewanians, and both affect how they see politics and the future of their communities.

Measuring Identity

How do we measure identity? A good place to start is by asking people where they think they fit in relation to different groups in society. We asked participants to indicate how they identify their social class, using the categories of poverty class, working class, lower-middle class, or upper-middle class/upper class, and to indicate the category that best represented their racial and ethnic heritage from a provided list.[2] Categories presented for racial/ethnic heritage were based on those provided in the Census. Figure 1 presents results related to social class identity and Figure 2 presents results related to race and ethnicity.

Most respondents in both provinces saw themselves as middle class. About one-third of respondents in both provinces indicated they were lower-middle class, and 29% in Alberta and 19% in Saskatchewan indicated that they were upper-middle class/upper class. It was not surprising so many people identified as middle class, given how sought-after the class position tends to be in Canadian society. For instance, the federal Liberals have talked about “the middle class and those working hard to join them” in various policy proposals, and have even gone so far as to appoint a Minister of Middle-Class Prosperity. However, there is no unique definition of who is and is not middle class, and Canada doesn’t have a statistical measure of the term. The federal Liberals aren’t alone, as the other major federal and provincial parties often talk about how policies affect middle-class Canadians while also avoiding a specific definition of who exactly they mean. Because so many people identify as middle-class, ambiguity has clear political benefits. Although "middle class" was the most common identity, identification with the "working class" was also strong in both provinces. In Alberta, 29% of respondents identified as working class, and in Saskatchewan the corresponding percentage was 36%. Fewer respondents identified as poverty class.

[1] While there are many definitions of both terms, generally speaking, race refers to social grouping based on perceived physical characteristics that groups have determined to be socially significant, while ethnicity refers to social grouping based on common history, culture, and other characteristics that often coincide with similar backgrounds. The Viewpoint survey asked about respondents’ racial and ethnic identities based on census categories; however, because of limitations due to sample size, we discuss differences for people based on race. We recognize the correspondence here may not be perfect.

[2] Respondents actually had a choice between upper-middle class and upper class. However, since very few respondents reported being upper class (8 individuals in Alberta and 5 individuals in Saskatchewan), we merged this category with upper-middle class for analysis.

Figure 2 shows the percentages of respondents who identified as white or as members of another racial or ethnic category. Most respondents in both provinces identified as white – approximately 80% in each. Although the percentages in Figure 2 are weighted to account for sample bias, our sample is still less diverse than the actual population of both provinces (as identified in the Census). This reveals an important limitation to survey research: racial/ethnic minority respondents – including Indigenous people – are very often underrepresented in survey research because they are harder to reach. Further, while the term “non-white” is far from ideal (for starters, in that it may be interpreted as endorsing the notion that “white” is the “default” racial category) – we use it here to guard against improper generalizations about specific racial and ethnic communities. Alternative terms, such as BIPOC, carry a different set of identity markers, which may not capture how respondents meant to be identified. In future research, we intend to explore new ways of measuring racial and ethnic identification that respect these important nuances.

Attachment to Social Group Identities

It is one thing to ask people which groups they identify with, but another to understand how strongly they feel about those identities. The Viewpoint survey aims to do so using respondents’ answers to three questions:

1. How important is your [ethnic group or social class] to your identity?

2. How strongly do you identify with other people from your [ethnic group or social class]?

3. Do you agree or disagree that what happens to people from your [ethnic group or social class] in this country will have something to do with what happens in your life?

The first measure is the most straightforward – respondents’ own subjective understanding of how much their group matters to them. The second is closely related, asking about respondents’ identities in relation to other members of their group. The final measure taps into a concept known as “linked fate” – a person’s belief that their future life prospects are tied to their membership in a specific group. Figure 3 compares the results of these three questions for race/ethnicity and social class across Alberta and Saskatchewan.

Overall, respondents in Alberta and Saskatchewan showed high levels of identification with and attachment to both race/ethnicity and social class. In both provinces, more respondents thought that their ethnic group – as opposed to their social class – was important to their identity. In Alberta, 50% of respondents thought race/ethnicity was an important part of their identity and 41% thought that social class was important. In Saskatchewan, 47% of respondents thought race/ethnicity was an important part of their identity and 37% thought that social class was important.

However, more respondents in both provinces felt a sense of linked fate with their class group than their ethnic group. Seventy percent of respondents in Alberta and 65% in Saskatchewan indicated that they believed what happened to people in their social class would affect their lives. The corresponding percentages for race/ethnicity were 63% and 59%.

There were interesting differences between Albertans and Saskatchewanians in how they saw themselves in relation to their racial/ethnic and class groups. In Alberta, more respondents strongly identified with their racial/ethnic group than with their social class. In Saskatchewan, the pattern was reversed; more respondents strongly identified with their social class than with their racial/ethnic group. However, in both provinces the percentages in each of these categories were very close.

As you might expect, strength of attachment also varied within identities. Figure 4 presents results showing variation by social class group, and Figure 5 presents results for race/ethnicity. Here, “strength of attachment” is measured using a combination of all three questions we discussed above.

As shown in Figure 4, 30% or more of respondents in all class groups reported strong levels of attachment. However, class attachment was strongest among ‘poverty class’ Albertans and working and poverty class Saskatchewanians. In other words, the salience of class identities tended to be stronger among those who saw themselves as occupying a less privileged place in society.

Regarding racial/ethnic attachment in Figure 5, non-white individuals in both provinces identified more strongly with their ethnic group than white respondents. Racial/ethnic minorities in Alberta showed greater attachment to their groups than those in Saskatchewan. However, white identity was also higher than expected according to conventional wisdom in both provinces. One-third of white respondents reported high attachment to their ethic group, and more than three-quarters reported either medium or high levels of attachment. We return to this point later in the brief.

Intersecting Identities

Our results thus far show that Albertans and Saskatchewanians have strong attachments to both their racial/ethnic and social class identities. However, group identities do not exist in a vacuum – they also intersect with one another. Figure 6 examines the intersections between racial/ethnic identity and social class identity. In Alberta, racial/ethnic differences in class identification were limited. Slightly more white than non-white respondents identified as poverty class and more non-white respondents identified as lower middle class. Differences in Saskatchewan were more apparent. More white respondents identified as poverty class and as upper middle class, while many more non-white respondents identified as working class.

Racial/ethnic and class identities also varied in relation to respondents’ income. Figure 7 shows self-identified class groups across household income, while Figure 8 shows the same for self-identified racial/ethnic groups. If household income was the only determinant of social class, Figure 7 would just be a series of solid-coloured bars. That is not what we see. For example, a quarter of those with a household income of $120,000 or more identified as lower-middle class, while 7% and 17% of Albertans and Saskatchewanians with those incomes, respectively, identified as working class. The overwhelming majorities of those with household incomes of $40,000 to $80,000 identified as either working class or lower-middle class. Working-class and lower-middle class identities were also very strongly represented in those making between $80,00 and $120,000. These results speak to how subjective and complex identities like social class are, and imply that many people see themselves as "average" or less well-off than the average – regardless of their actual material circumstances.

Income and racial/ethnic identity have a different relationship, compared to income and social class. However, Figure 8 still reveals marked differences between racial/ethnic groups in terms of income. In both provinces, more respondents with household incomes of under $40,000 are non-white than among any other income group. In Saskatchewan, the proportion of non-white respondents decreased linearly as income grew. Only one-tenth of Saskatchewanians with a household income of $120,000 or greater were non-white. In Alberta, the differences were less marked – decreasing substantially in the $40,000-$60,000 category before levelling off at about one-fifth thereafter.

Attitudinal Differences

Racial/ethnic and class identity has attitudinal and behavioural implications. Group identification is related to a range of attitudes, including how optimistic people are about the future. However, these identities do not always have the expected effects. Figures 9 and 10 show how much respondents agreed with the statement, “The province's best days are behind it.” Figure 9 breaks down the results by social class identity and Figure 10 breaks down the results by race/ethnicity.

In Saskatchewan, class identity tracks neatly onto views regarding the province's future (as shown in Figure 9). Poverty and working-class respondents were more likely to think their province’s best days are behind them, while lower and upper middle-class respondents were more likely to disagree with this statement. Disagreement was particularly high among upper middle-class respondents with 53% disagreeing with the statement and indicating more optimistic views. In Alberta, working class respondents tended to be the most pessimistic, with 50% agreeing that the province's best days were behind it. However, pessimism was high for all class groups in that province.

In terms of racial/ethnic differences, Figure 10 shows that levels of pessimism again varied across provinces. White respondents in Alberta tended to be somewhat more pessimistic than non-white respondents, but differences were limited across groups. A clearer pattern emerged in Saskatchewan, where non-white respondents tended to be more pessimistic. In that province, 22% of white respondents agreed that the province's best days were behind it, compared to 31% of non-white respondents.

Voting Behavior

Not surprisingly, both racial/ethnic and class identities have implications for how people see the political world. As we discuss in earlier briefs, our March 2021 survey showed the New Democratic Party (NDP) ahead in Alberta, while the Saskatchewan Party (SKP) was up in Saskatchewan. Figure 11 shows that vote intention varied across class groups and racial/ethnic groups.

In both provinces, pluralities of those who self-identified as either poverty-class or working-class intended to vote for the left-leaning NDP. While this was also the case for lower-middle class respondents in Alberta, large pluralities of lower- and upper-middle class respondents in Saskatchewan intended to vote for the right-leaning SKP. By contrast, upper-middle/upper class respondents in Alberta were divided between the NDP and the right- leaning United Conservative Party (UCP).

There were also interesting differences between white and non-white respondents. In both provinces, a plurality of non-white respondents intended to vote for the NDP. This was also the case for white respondents in Alberta, while white respondents in Saskatchewan preferred the SKP. The fact that many non-white and poverty-class and working-class respondents prefer the NDP is consistent with what we know about the voting behaviour of members of these groups, given left-leaning parties are often associated with placing greater emphasis on the priorities of these groups than do right-leaning parties.

Perceptions of Discrimination

The last set of attitudinal differences we focus on pertains to perceptions of discrimination. Many people recognize inequalities in how society treats members of different groups – privileging some groups while leaving others to bear the brunt of discrimination. As Figure 12 shows, racial/ethnic identity is an important predictor of which groups people see as worse-off in society.

There is a societal consensus regarding some targets of discrimination. For example, similar percentages of white and non-white respondents recognize that there is a “great deal” or a “lot” of discrimination in society against Indigenous people and Muslims. However, non-white respondents are more likely to perceive discrimination against minority groups – the differences being strongest for perceptions of discrimination against Black people and immigrants. In particular, 38% of non-white Albertans thought there was either a great deal or a lot of discrimination against Black people in society, while 29% of white Albertans thought the same, a difference of nine percentage points. The differences – as well as the overall levels – were less pronounced in Saskatchewan, where 29% and 24% of non-white and white respondents agreed with the same statement. Non-white respondents in both provinces were also more likely to agree that there was significant discrimination against immigrants, while white respondents were less convinced of this statement. Finally, white respondents in both provinces were more likely to suggest there was either a great deal or a lot of discrimination against white people. Overall, however, the number of respondents of all racial/ethnic identities who agreed with that statement were quite low.

Finally, we return to a finding we originally presented in Figure 5: a large number of white respondents, or at least more than conventional wisdom might suggest, strongly identify with their ethnic group. Although this idea has been discussed in much greater depth in the context of the United States and the policies and rhetoric of former president Trump, Canadians are not immune to the potentially toxic effects of white identity politics. As Figure 5 showed, approximately one-third of white Albertans and Saskatchewanians were strongly attached to their racial/ethnic group. At least regarding the question of discrimination, strong white identifiers feel differently than white respondents who identify less strongly with their racial/ethnic group. Approximately two-thirds of strong white identifiers in Alberta and three-quarters in Saskatchewan think there is at least a little discrimination in society against white people, and more than one-fifth of each group think there is a lot or a great deal of it.

Conclusion

Although Canadians sometimes like to think we live in a race-less and class-less society, our findings show that this is not the case. Large numbers of individuals strongly identify with their ethnic/racial group and social class, and both identities have important effects on political attitudes and voting behaviour. Moreover, the intersection of these two identities reveals important inequalities in society. Finally, we show that a group that we are only starting to understand – strong white identifiers – are not just an element of US politics, but are rather present in significant numbers in both Alberta and Saskatchewan.

Methodology of the Viewpoint Alberta Survey

The Viewpoint Alberta Survey was conducted between March 1 and 8, 2021. The survey was deployed online by the Leger. A copy of the survey questions can be found here: http://bit.ly/30VcYEY. Leger co-ordinates the survey with an online panel system that targets registered panelists that meet the demographic criteria for the survey. Survey data is based on 802 responses with a 17-minute average completion time. The Viewpoint Alberta Survey was led by co-principal investigators Loleen Berdahl and Jared Wesley.  It was funded in part by a Kule Research Cluster Grant and an Alberta-Saskatchewan Research Collaboration Grant from the Kule Institute for Advanced Study (KIAS) at the University of Alberta and the College of Arts and Science at the University of Saskatchewan.