How should public policy be directed in order to counter negative attitudes and discrimination towards immigrants and minorities?


Jim Frideres
Department of Sociology
University of Calgary


Prejudice and discrimination have long been concerns for both scholars and practitioners. However, over the years, little research has been carried out investigating the causes or consequences, in spite of the fundamental nature of such social behaviour. Both prejudice (attitudes) and discrimination (behaviour) have an important role to play in impacting the quality of life in Canadian society. At a different level of inquiry, there is a need to assess the costs of living in a society that exhibits evidence of prejudice and discrimination.

However, before embarking upon any major research programs, it is imperative that scholars and policy makers determine the nature and extent of prejudice and discrimination in our society. This means that a major pan-Canadian study needs to be undertaken to determine the extent to which these attitudes and behaviours exist in Canada. This would require a survey with a large random national sample and a commitment to carry out a similar survey each year for the foreseeable future. This would allow us to determine if polices and/or programs put into place are having an impact on the level of prejudice and discrimination evident in our society. In short, the first survey would benchmark the existing level and establish targets for the future. It will also allow us to identify areas or cohorts that exhibit high or low levels of prejudice or discrimination over time.

Researchers need to work with policy makers to clearly define what is meant by the two concepts (prejudice and discrimination) and to make sure that measurements reflect all dimensions of the concepts. The distinction between "objective" and "perceived" discrimination must also be ascertained in all research projects. For example, it has long been determined that discrimination (real and/or perceived) can be at the individual, structural (systemic), or cultural level. Discrimination can take place in private situations as well as in public and the different contexts, e.g., work place, educational, religious, or political institutions, will need to be investigated. Likewise, prejudice can be expressed "covertly" or "overtly" and these differential expressions need to be studied. As the research is undertaken, it is important that all dimensions be measured to determine the nature and extent of each type/dimension of prejudice and discrimination in each institutional order (context). This research will include sub-populations that have been ignored in the past, e.g., ethnic groups, age groups, occupational groups, of all types.

The two concepts under consideration can be considered either "independent" or "dependent" variables. This means that one can focus on the antecedents of prejudice/discrimination, e.g., what causes people to be prejudiced or what causes people to engage in discriminatory behaviour? Alternatively, one can focus on the consequences of prejudice/discrimination, e.g., what impact does discrimination have on the behaviour, productivity, creativity, citizenship of an individual/organization or society?

While considerable concern and interest has focused on "what causes prejudice and/or discrimination," no systematic set of empirical research projects exists. A meta analysis of existing research results in the area is presented in Appendix A. Researchers focusing on this question want to know what are the "roots" of such attitudes and/or behaviours. They want to look at both the individual and structural factors that give rise to prejudicial attitudes and/or discrimination. At the individual level, one would want to look at various attributes of individuals that predispose them to have such attitudes and/or behaviour. One would also look at the socio-demographic factors, e.g., age, sex, social class, residence, education, that are linked to individuals exhibiting various degrees of prejudice/discrimination. At the structural level, the interest is in whether or not individuals that operate in institutions or specific social/organizational contexts have differential levels or prejudice and/or discrimination. The general question would be "How does the contextual environment influence the attitudes and behaviours of individuals?" It also will focus on how organizations promote prejudice/discrimination regardless of the attitude of the individuals in the organization. Unfortunately, almost all research carried out in the past has focused on the individual unit of analysis.

In this research setting, researchers would look at how institutions impact the attitude and behaviour of individuals. For example, in the area of education, how do schools affect the attitudes and behaviour of the individual? Are teachers important? Do peers have an impact? Do the curricula have a role to play? How does the ethnic composition of the school impact attitudes or behaviours? Do students in private ethnic/religious schools think and behave differently than those in public schools? Does the administrative structure in a school have an impact on student’s attitudes and behaviour about immigrants and/or visible minorities?

In addition, research focusing on the transmission of attitudes and behaviour needs to be undertaken. This research, focusing first on the "socialization" activities of families, would then go on to look at how institutions carry their message and impact the recipient, e.g., the role of media. Research of this nature would focus on the impact of each of the various institutional sectors of our society to determine if any one is more or less important in terms of impacting upon citizens. Does the political sector disproportionately influence people’s attitudes and/or behaviour toward ethnic groups or immigrants? How does the media influence how we think and behave?

The second aspect of prejudice and discrimination focuses on the issue of "consequences". What are the influences on individuals living in a society that engages in expressions of prejudice and discrimination? Considerable research will need to be undertaken to look at how both the social and economic capital (at both the individual and/or organizational level) are impacted by discriminatory behaviour taking place in a society (see Appendix A). The impact of such attitudes and behaviour will have to be researched at different levels of analysis. For example, what impact does a discriminatory act have on an individual’s productivity in the work place? Her level of learning in an educational institution? Commitment to Canadian values and goals? Do acts of discrimination impact upon the mental and physical health of the individual. What are the indirect impacts of such behaviour, such as on other family members? Other co-workers? Is there a cycle of events that act in a recursive manner? The impact of discrimination also would need to be assessed at a higher level of analysis such as organizational. Here the question might be how effective and/or efficient is an organization structured in such a way that it supports or perpetuates certain types of prejudice and/or discrimination. Finally, we need to investigate the impact of discrimination at the societal level, such as how is the quality of life in Canada impacted by such attitudes and behaviours? Is social cohesion of a community or society threatened by such behaviour?

A third type of research project that will need to be carried out will be "evaluation" research. This research will assess the impact of various social policies and/or programs that are seen to diminish or exacerbate prejudice/discrimination. Until we live in a society devoid of prejudice and discrimination, we also must develop coping techniques that can be drawn upon by targets of such behaviour. These coping techniques can be at the organizational or individual level. For example, what kinds of "anti-racist" programs could be put in various educational or work place settings? Again, evaluation research will have to be carried out in a variety of organizational contexts. What kinds of teacher training programs would be most beneficial in reducing the amount of prejudice/discrimination exhibited in the schools? These "evaluation" research projects must be undertaken to assess the impact of the programs/policy and monitor the extent of change of prejudice and/or discrimination.

There also is an increasing interest in looking at the intersection of the two factors (prejudice and discrimination). Put another way, is there an interaction effect between the two concepts that enhances or retards levels of prejudice or discrimination? This issue also addresses the question of linkage between attitudes and behaviour. Does having a negative attitude (prejudice) mean that discriminative behaviour will be exhibited? The answers to these questions will require research projects that exhibit a variety of research designs, measurement procedures, and data analytical techniques.

How do we proceed? The following sequential steps seem appropriate.

    1. What is the incidence/nature of prejudice and discrimination?
    2. What are the causes of prejudice/discrimination?
    3. How are prejudice and discrimination transmitted over time and among people?
    4. What are the consequences of prejudice/discrimination?
    5. How can we translate "pure" research results into applied policy?




Appendix A

Meta Analysis of Research in the Area of

Prejudice and Discrimination

Variables Related to Expressions of Prejudice and Discrimination



High depression


high (perceived)

High anxiety


high (perceived)

Low intergroup, interaction


high (perceived)

Affirmative action mentioned



Social class



High concentration with low interaction



Manual labour force



Sex (males)



Ethnicity (Caucasian/latino)



High self esteem



High economic competition






Less knowledge of outgroup member



High measures of psychological well being



high sense of powerlessness



Ethnicity (and ethnic identity)



Demographic variables (except sex)



Immigrants (vs) native born



Greater perceived threat (realistic/symbolic) of outgroup



High intergroup anxiety



High negative stereotypes



Low sense of mastery over life events


perceived high

High acculturation level


low perceived

Good language ability


low perceived




Age (older)



Strong ethnic identity


low perceived

High intergroup competence


low perceived

High education



Proximity of housing (of minorities)



National pride



Contact with minority in work setting



Post materialism vs (materialism)



varied by subtle vs blatant discrimination or prejudice



Discrimination related to:

    • lower sense of social (cultural) coherence - decreased employment
    • physical illness - lower salaries
    • psychological stress - low performance expectancies
    • low self esteem - low achievement motivation
    • low life satisfaction - heightened ingroup identification
    • anxiety - increase aggression
    • depression - increased sadness

Appendix B

At the heart of research is good science and science has clearly articulated rules and conduct of inquiry. The fundamental assumptions of science cannot be violated or the research becomes unreliable and invalid. This will mean that research projects that do not exhibit "good science" cannot be undertaken or funded. As well, "exploratory" research projects must be set aside in favour of theoretical "confirmatory" research projects. The key to any research is the "generalizability" of the research results and this also demands "good science" that requires samples drawn from the population to be reflective of the population. In studying any social issue, it must be clearly demonstrated empirically that specific factors are "causal" and thus provide an explanation for the behaviour. Thus, one of the key factors in establishing reliable and valid indicators and "causal" factors is "good science". We take as a given that all research to be undertaken must first pass the test of "good science".

There are well established criteria for establishing "causality" and this must remain the central focus of the proposed research. The kinds of research required will focus on "confirmatory" research and only in a secondary sense will exploratory and/or descriptive research be funded. The kinds of research employed to study prejudice and discrimination should not simply take the form of large-scale pan-Canadian surveys. Many of the issues to be researched will require different types of research designs, measurements, and analysis. Moreover, cross sectional formats will have to be expanded to include longitudinal types of research projects.

There also will need to be a translation from theoretical results to practical application. This means that either basic or applied research will need to be "re-cast" into policy and program strategies. Just as researchers are not clear how to recast their results into policy perspectives, it is not easy for policy makers to take research results from "pure research" and show its relevance to their work. An interim group will need to be established to "translate" between the two worlds and ensure that research results are both informative to policy makers as well as understandable. Related to this is the unpredictability of politics in the world. Thus, while we all aspire to engage in "evidence based results", there are times when pressures are placed on policymakers that are politically motivated. Hence there will be times when evidence based results are not implemented because of political reasons. These should be used as opportunities to investigate the impact of such activities.

Table 1. Research Methodology Considerations

Research Design

Table 1 reflects the diverse kinds of data and types of research that will need to be carried out over the next decade to better answer the questions posed by scholars, practitioners, and policy makers. Some of the data required will be based upon pan-Canadian samples while other data need not meet that criterion. Results obtained from local, random samples can be generalizable to a variety of settings, populations, and regions of the country. For example, there are no "a-priori" reasons to suspect that a random sample of workers from Manitoba will be differently impacted by discriminatory behaviour than workers from Nova Scotia or Quebec. The theoretical principles are the same in many settings and that will provide the key for explaining either the "roots" of the behaviour/attitude or the consequences.

The nature of the data collected will also vary. Some of the data will need to be obtained through primary sources (original research) while other research projects can be just as productive in using secondary data (data collected by another person for another purpose). For example, it is expected that other agencies will carry out research on issues of interest to Metropolis researchers as well as policymakers. One example might be that the Council of Ministers of Education undertake a nation-wide study on "educational equity" for which the results would be available to Metropolis researchers and policymakers. Some data will be collected through the use of surveys (individual, groups) through e-mail, mail, personal interviews, or telephone. Research data may emerge from experimental (or quasi experimental) conditions (variables are contrived and manipulated by the researcher) or in more naturalistic settings. Interviews, observation, participant observation, unobtrusive measures, e.g., content analysis, will all be used in order to confirm and triangulate the research results.

The focus of the research also will need to take into account a cross section of Canadians and their institutions. Thus, gender, age, social class, occupation, ethnic background, region, and a host of other factors will need to be included in the data collection procedures. Factories as well as banking organizations, schools and grocery stores, fabrication plants as well as law firms, religious institutions as well as the military need to be included in the samples subjected to analysis.

The research to be carried out must be interdisciplinary and ensure that different perspectives be integrated into the research design and data collection. A host of factors reflecting different disciplines will need to be incorporated into the data collection. If, for example, the research is focusing on "citizenship" or "participation in the political process", then researchers from a variety of disciplines will need to be involved in the research process. These interdisciplinary perspectives will need to be in place as the research is being developed. Research projects will need to focus on different "units of analysis", specifically focusing on organizational and structural levels.


          Home | About Us| Events | Partners | Publications | Media Centre | Policy Priorities | Ottawa Team