Numbers and Needs: Identifying Demographics and Service Needs of Ethnoracial Communities

Masters Research Paper

Masters of Education
Department of Measurement, Evaluation
and Computer Applications (MECA)

Ontario Institute for
Studies in Education (O.I.S.E.)
University of Toronto
Professor Stacy Churchill
Academic Supervisor

Introduction

In spite of a constantly growing need for better data on the demographics and service needs of ethnoracial communities, including immigrants and refugees, there is a dramatic lack of information which is both useful and easily available. Why has it been so hard to get data that is appropriate for planning purposes, and what progress has been made in developing information sources that are relevant and accessible? These questions, which for several years have been a motivation for my professional activities and a source of stimulation in my academic studies, are addressed in the following paper.

Historical Background

Concerns, debates and practical experience on ethnoracial data collection have evolved considerably over the past twenty years. The first steps in making this issue one of public concern were government reports dealing with equity issues, such as employment equity (Abella, 1984), elimination of racism in Canadian society (Multiculturalism and Citizenship Canada, 1989), and equitable treatment of foreign-trained professionals (Task Force on Access to Trades and Professions, 1989).

These public concerns about equity issues have been motivated in large part by recent and dramatic changes in the demographics of the Canadian population. The ethnic composition of Canada has become more diverse as the source countries of immigration in the past two decades have shifted away from Europe and the United States to Asia, the Caribbean, South America, Africa and the Middle East (Kelly, 1995:7). Based on the 1991 Census, nearly 10 per cent of the Canadian population were visible minorities (Kelly, 1995:28); about 70 per cent of these visible minorities were immigrants (Kelly, 1995:38-41). Projections to the year 2001 suggest that visible minorities will make up approximately 18 per cent of the population of Canada, 40 per cent of Vancouver's population, and 50 per cent of Toronto's; about half of these visible minorities will reside in Ontario (Samuel, 1992:4-5).

As researchers and social planning professionals began to address these policy issues and the impact of the demographic changes, they reported a serious lack of accessible and useful information to support their work. This problem was highlighted in a number of ground-breaking studies including a review of the mental health needs of refugees in Canada (Canadian Task Force, 1988), a study of access to family services for minority populations in Toronto (Medeiros, 1991), and an examination of mental health services for Toronto's culturally-diverse population (Mental Health Program Services, 1992).

Some researchers have tried to respond to these information gaps by providing statistical evidence of discrimination (Billingsley and Muszynki, 1985; Frances and Ginzberg, 1985; Raskin, 1993). More recently, we have seen the development of grassroots research initiatives to document discrimination among particular ethnoracial communities, such as Ho-Lau's (1992a, 1992b) study which combined results of a survey of Chinese immigrants with relevant data from Statistics Canada.

Despite government policies to promote equity and some research initiatives to document discrimination, however, spokespersons for ethnoracial communities and professionals involved in social planning continued to report that their efforts to promote equitable access to services were seriously hampered by a lack of appropriate data (Davidson-Palmer, 1990; Sabloff, 1991; Herrera et. al., 1992; Access and Equity Centre, 1994). The concerns expressed most frequently related to the inadequacy or inaccuracy of official statistics on ethnoracial communities, especially recent immigrant and refugee arrivals, and the need for information on the ethnoracial demographics and cultural characteristics of clients in health and social services.

Without such information, a number of areas of essential planning for health and social provision are seriously weakened. These include responding to emerging and often acute needs among recently-arrived immigrant and refugee communities, identifying and removing linguistic and cultural barriers to access to services by ethnoracial communities, and incorporating equity principles into the governance mechanisms and human resource policies of public service institutions as well as private corporations.

Public Workgroups and Consultations

Growing interest in the subject of ethnoracial data collection has led in the past few years to the establishment of different working groups and public consultations, including several in which this author has been directly involved. In 1992 community representatives and public sector planners from the Metropolitan Toronto area came together in the Working Group on Ethnoracial Data Collection, chaired by the author, to promote both critical discussion and improved practices. The group was particularly concerned with making better use of client service data in the same sense expressed by the Corporation of Metropolitan Toronto:

All institutions need to implement... some system for reviewing the delivery of programs and services as a means of ensuring the efficient and effective utilization of limited resources. Within this context, the most reliable and efficient means of monitoring the effectiveness of the Metropolitan Council's Policy on Ethno-Racial Access to Metropolitan Services is to carry out a regular analysis of the service clients according to ethno-racial origin (Access and Equity Centre, 1994:5).

The first activity of the Working Group was a Pilot Project at selected health care sites to test a set of data elements identifying geographic origins, immigration status, preferred language of service delivery, religion, ethnicity and race based on voluntary self-identification by clients. The results of the project were judged to be very useful and included a high rate of voluntary response on the controversial issue of self-identification by racial origin (Working Group, 1993). The group also organized a well-attended Symposium which discussed research methodology and ethical issues as well as practical concerns (Working Group, 1994), and published a "Better Practices" guide to methods of ethnoracial data collection (Working Group, 1995).

In 1994 the federally-organized public consultations on Canadian immigration policy included a working group, in which the author participated as the sole community representative, which discussed the kinds of data needed to monitor and evaluate immigrant settlement in Canada (Citizenship and Immigration Canada, 1994:78). The most visible outcome of these discussions is the establishment of new research centres on immigration issues in Canada's largest urban centres with a collaborative perspective between academics and community representatives.

Also in the past few years, Statistics Canada reported that their consultations on the 1996 Census received more feedback on the subject of "ethno-cultural characteristics and Aboriginal status" than on any other topic (Statistics Canada, 1994:23). In response to submissions including one to which this author contributed (Murdie and Novac, 1993), the 1996 Census will for the first time treat questions of ethnicity, race and Aboriginal status as separate variables (Statistics Canada, 1995:11-15).

Examples of Research Initiatives

Most research concerned with the health and social service needs of ethnic, immigrant and refugee communities has to find some way of coming to terms with the current inadequacies of demographic and statistical information on ethnoracial communities. Two studies with informative approaches, both recent, dealt respectively with the settlement of Hispanic-speaking immigrants in Metro Toronto, and with the needs of refugee children in the Canadian school system.

The needs assessment of the Metro Toronto Hispanic community was conducted for the Centre for Spanish-Speaking Peoples (CSSP), and focused on the space and facility needs of the Centre, based on an assessment of community need (Sinclair-Jones, 1993). Along with a review of the history, role and structure of the CSSP, the study included an examination of immigration statistics, an analysis of client service statistics, an extensive literature review, in-depth interviews and focus groups with more than fifty people closely involved with the Centre; and confidential interviews with seventy-nine users of the Centre (the User Survey).

This research study found the official statistics on the demographics of Metro Toronto's Hispanic-speaking community to be quite inadequate:

It is difficult to estimate the size of the Spanish-speaking community in Metro Toronto with certainty.... The 1986 Census identified about 25,000 Hispanics in Metro Toronto, but most observers agree this drastically underestimated the size of the community.... It has been reported that the 1991 Census indicates an increase of 87% in the size of the Spanish-speaking community in Metro Toronto, using the 1986 Census as a base figure.... However... this figure is no more reliable than the 1986 base figure, especially because the definitions regarding to use of "mother tongue" were changed from one census to another (Sinclair-Jones, 1993:32).

In order to develop a more reasonable estimate of the current and future size of the community, the research took into account the following: total migration of Spanish-speaking persons to Canada based on immigration records; the percentage of these persons indicating Ontario as their intended destination; an estimate of the portion of the latter who likely came to Toronto; projections of the future rate of immigration to Canada and Ontario; rough estimates of the net effect of internal migration of Spanish-speaking persons to and from Toronto after arrival in Canada, and of the recent rate of population increase among Spanish-speaking persons in Metropolitan Toronto; consideration of the significant numbers of Spanish-speaking residents of Metropolitan Toronto who are without legal status and do not show up in census figures or other official statistics; and data on geographic origins of students enroled in the Metropolitan Toronto Public and Separate schools (Sinclair-Jones, 1993:35-38).

On the basis of this analysis, the study concluded that:

[T]he Spanish-speaking population in Metropolitan Toronto is between 100,000 and 120,000 people and will grow by some 50,000 people over the next 10 years (Sinclair-Jones, 1993:iv)

Further information on the demographics and needs of this immigrant and refugee community was provided by an extensive review of prior needs assessments and research reports. These demonstrated a number of acute health and social service needs for the Hispanic community, particularly in the area of employment (Sinclair-Jones, 1993:iv). Significant from the point of view of research methods is that most of the studies reviewed were non-academic projects, with narrow circulation, which would likely have been ignored by investigators who were unfamiliar with the community and its institutions.

The needs assessment conducted for the CSSP also compared the Centre's client statistics with Canadian immigration statistics to conclude that services were being provided to people from all of the countries of origin generally considered to be "Spanish-speaking", and that there was a close correlation between the numbers of people from each country of origin receiving CSSP services and their representation within the overall Hispanic community (Sinclair-Jones, 1993:iv,44-50). This bipolar method of analysis is common among researchers who are trying to make use of official census and immigration statistics which may be useful for broader aggregates while inadequate on the demographics of particular ethnoracial communities. Ho-Lau's study (1992a, 1992b), for example, compared unemployment rates among a sample of Chinese-born immigrants in Metro Toronto with official figures for the general population; while Choi et. al. (1989) made similar comparisons based on their own survey research concerning the health needs of four ethnoracial communities in relation to factors such as average size of families.

A second study, also recent and equally informative, examined the factors behind the school performance of refugee students in Ontario and across Canada (Churchill and Churchill, 1994). Along with an extensive review of documentary materials from both Canadian and international sources, this research project made use of numerous discussions and interviews with staff and administrators in the school system, community leaders, immigrant settlement service providers, and refugee children and their families.

As is common in research concerning the service needs of ethnoracial communities, the authors of this study reported that statistics such as census figures and Employment and Immigration Canada (EIC) data were in many ways inadequate or misleading for the purposes of their enquiry. Among the particular problems they noted were the lack of information on the internal migration of immigrants and refugees after arrival in Canada; the failure of EIC data to distinguish between young people arriving as Convention Refugees or Designated Classes and those arriving from refugee situations as Family Class immigrants; and the lack of information on undocumented or "illegal" immigrants and refugees (Churchill and Churchill, 1994:viii,1-2).

The fifth chapter of this study, "Factors Affecting School Performance", highlights the research methods used by the authors. This section deals with the characteristics of refugee children and their families including the refugee determination process, personal and family histories, mobility and resilience, age, and illiteracy. It also examines the culture of the school system and the impact of broader societal issues such as racism and the clashes of cultures. The analysis is derived largely from texts, including other research studies and policy documents, and from interviews and discussions with refugee children and their families and with educators and service providers (Churchill and Churchill, 1994:67-94).

This research report still makes extensive use of statistics, including studies done by the City of Toronto Board of Education (Every Secondary School Student Survey) and EIC and Statistics Canada census data. But rather than using quantitative information as the foundation of their investigation, the authors present the available statistics selectively and critically, as an additional and limited source of knowledge shedding further light on the situation of refugee school children. The basic method of investigation employed is qualitative (Churchill and Churchill, 1994:viii), in keeping with the subject and the nature of the research.

Ongoing Issues in Ethnoracial Data Collection

A recent document from Statistics Canada (Kelly, 1995) gives the impression that the main problem with the use of census data on ethnoracial communities in Canada lies in the identification of visible minorities. Relying heavily on Boxhill (1984, 1990, 1991), Kelly presents a coherent and useful summary of the ways that counts of visible minorities have been derived from other variables, principally ethnicity, and of the difficulties this presents.

If we follow the logic presented by Kelly, it would seem that the separation of the "ethnicity" and "race" variables in the upcoming 1996 Census will go a long way towards resolving the problems with ethnoracial demographics based on census data. And while I support these changes (having argued for them!), it also seems to me that the more fundamental problem with census data is a persistent and largely unavoidable undercounting of the most recently-arrived and socially-vulnerable ethnoracial communities.

On the most obvious level we can see that since the census is conducted only once every five years, and since it takes another few years to compile and publish the data, it will not provide us with good information on people who are new arrivals to our country. But we also need to consider the degree to which the very characteristics of ethnoracial communities may lead to their undercounting in the census.

The census is an essentially voluntary act of self-reporting, and there are a host of reasons why immigrant and refugee communities are not likely to report their true numbers. Lack of knowledge of Canada's two official languages is one such reason, as is also a lack of understanding of the purpose of the Census and its role in planning social services. The efforts by Statistics Canada to increase their community outreach and to provide information on the census in a variety of languages are laudable and indeed essential, but it is not reasonable to expect these initiatives to fully overcome the kinds of barriers to participation in the census that exist for recently-arrived communities. There is also the particular case of refugees fleeing violent political persecution in their previous homelands, and therefore deeply suspicious of voluntarily providing personal information to governments. Furthermore, many immigrants and refugees live in situations that are economically precarious and may therefore have reasons to not disclose the truth about such details as the number of persons residing in their dwelling-place. Finally, there are significant numbers of undocumented or "illegal" immigrants and refugees who have every reason to not report themselves in a census.

A common approach to these shortcomings in the information supplied by Statistics Canada is to combine census data with annual immigrant landing data from Employment and Immigration Canada (EIC). This was the role performed, for example, by the Ethnocultural Data Base of the Ontario Ministry of Citizenship, which also made these integrated statistics available by municipality. This approach is an improvement in the direction of more up-to-date data, but serious problems remain. The landing data does not take into account migration from Canada by immigrants after their arrival to this country, or internal migration by immigrants within Canada after arrival at their intended initial destination. The latter factor is important for many immigrant and refugee communities, and particularly significant in planning services for a region like Metropolitan Toronto which tends to attract immigrants from other parts of the country.

Furthermore, the combination of census and immigrant landing data provides no information on the net change in population (births versus deaths) for immigrant communities during the periods in between a census. Depending on the age and family characteristics of a particular immigrant or refugee community, this may be an important factor in assessing numbers.

Ultimately, though, the more fundamental problem associated with combining census and immigrant landing data lies with the validity of the census data itself. If, as I have argued, the census figures tend to undercount ethnoracial communities, then the combination of these numbers with immigrant landing data will still provide figures that are too low, although to a lesser degree.

When we examine the use and misuse of official statistics on ethnoracial communities, however, another question arises. Why is so much energy expended on compiling, manipulating and interpreting official numbers on ethnoracial communities, when the original motivation for the enquiry is very often a concern with identifying the needs of these communities? To provide even a partial answer to this question, we need to consider the role of quantitative and qualitative research methods in assessing the "numbers" and the "needs" of ethnoracial communities.

One of the things I found intriguing when I began to study the subject of ethnoracial data collection was the way that quantitative research methods were often used to minimize or deny what I perceived from my studies and my experience as the basic reality of discrimination against immigrants, refugees, and visible minorities. Examples of such studies that I tried to refute (Richmond, 1993) include Beaujot et. al. (1988) and Health and Welfare Canada (1989), which used census data to portray immigrants as slowly but steadily rising to income parity with their Canadian-born counterparts; and deSilva (1992) who analyzed census statistics to conclude that wage differentials between Whites and people of colour in Canada could not be attributed to racial discrimination.

Other researchers of course have used census data and other statistical sources to arrive at rather different conclusions. Boyd (1992) stresses that the social and economic status of particular groups of immigrants or ethno-racial communities is the product of a complex interplay of factors including gender, race and ethnicity, as well as length of time in Canada. She provides an important warning that attempts to over-simplify these patterns may well obscure the reality of discrimination based on gender, race and immigration status. Stasiulis (1990) makes the same point, relying on theoretical arguments as well as statistics. And Lautard and Guppy (1990), in an examination of occupational differentials among Canadian ethnic groups including visible minorities, place (White, European) Portuguese women at the bottom of an index of occupational status.

We could conclude therefore that it is dangerous to make sweeping generalizations about the economic and social status of ethnoracial communities on the sole basis of statistical aggregates. This is precisely the point made by (Anthony) Richmond (1993) in documenting and analyzing rapid changes in the educational experiences of Caribbean immigrants in Metropolitan Toronto that occured within a relatively short period of time.

We might also wonder whether there is too much attention to quantitative analysis of the socio-economic conditions of ethnoracial communities to the detriment of qualitative methods. There are at least three powerful arguments in favour of making greater use of qualitative methods such as in-depth interviews, focus groups, and literature reviews of community needs assessments. The first is that such methods can be more efficient in terms of human and financial resources than complex statistical studies. The second is that qualitative methods can be used very effectively to address the issues of "needs" rather than "numbers" within ethnoracial communities. The third argument is that qualitative approaches can be employed with communities for which reliable quantitative data is simply not available. Why, then, is so much emphasis placed on quantitative rather than qualitative methods? I addressed this question recently through in-depth interviews with three people whose daily work involves questions of ethnoracial data collection.

The persons I interviewed included one who works on questions of multicultural and ethnoracial equity within municipal government (T), a director of a community organization providing services to Africans (M), and another in municipal government who is responsible for translation and interpretation services (R) (Richmond, 1995). One of my concerns in these interviews was to understand to what degree the notion of "acceptable" data used as indicators of social needs of ethnoracial communities is linked to disputes over the relative merits of "quantitative" versus "qualitative" research methods.

All three persons interviewed wanted improvements in both quantitative and qualitative research in relation to ethnoracial demographics. However, the interviews revealed that the use of qualitative methods was a much more controversial issue.

One reason for favouring quantitative methods revealed by these interviews was the insistence by funders and politicians on this kind of information. This prejudice is confirmed even by researchers who favour qualitative methods, like Silverman who notes that it a common-sense axiom that "hard numbers" are the most convincing indicators of funding needs and program results (1993:163,173).

But the interviews also revealed a pervasive bias towards quantitative data. The language used by all three persons suggests that there is a special and powerful status attached to quantitative or statistical data, especially government statistical data such as census information. We can see this in the kinds of word pairings and associations employed. In the context of quantitative information and government statistics, the word "hard" describes "data" as does "exact", and "official" describes "statistics". In reference to qualitative data on needs provided by community representatives, the terms used are "estimates", "informal data", and "unofficial data".

For all the persons interviewed, the tension between "numbers" and "needs" is a source of controversy and practical difficulties in their work and is directly related to the relative merits of quantitative and qualitative approaches -- a subject identified by T as a "constant source of discussion". M revealed at the very end of his interview that he actually considered qualitative methods to be superior in assessing community needs, and that the constant insistence on quantitative indicators gave him his biggest headaches!

When the debate over the merits of these two approaches is pursued, however, the quality of the research is improved. R said that he has developed non-quantitative indicators of needs which have informed a number of funding decisions, and that this approach would not have been accepted had not other colleagues vigorously debated the merits of qualitative methods.

There are at least two types of theoretical issues that deserve proper attention in relation to ethnoracial data collection. The first is the existence of continuing controversies over the definitions of "ethnicity" and "race", and the social and political uses of such definitions. Good examples of these debates can be seen in the collection of essays edited by Li (1990), as well as in Ng (1981, 1991). This issue is also raised in a recent British study which used quantitative and qualitative methods to examine both ethnoracial identifiers and socioeconomic factors in relation to the self-assessed health status of individuals in an urban environment (Fenton et. al., 1995). The authors of this latter study said that, as ethnic and racial identification becomes more common in health research, some researchers are worried about a naive use of ethnic variables. They noted as well that "Measures of ethnic origin show no consistency and the concepts themselves are poorly understood" (Fenton et. al., 1995:56).

The Metro Toronto Working Group on Ethnoracial Data Collection addressed the controversies over ethnic and racial classifications as follows:

The members of the Working Group began work with an awareness that the collection of ethno-racial data is a controversial issue. The opinions of the ethno-racial communities on this topic are contradictory, with some spokespersons insisting on the need for improved data as a tool for more equitable access to services and employment opportunities, and other leaders emphasizing the dangers of the data being misused. Furthermore, the very notions of "race" and "ethnicity" are problematic in the context of data definitions and techniques of data collection. The concept of "race" is essentially a social construction based on historical power differentials between different communities of peoples, without any scientific basis; while the meaning of "ethnicity" is hotly disputed among academics and researchers.

Nevertheless, the members of the Working Group were also united in their conviction, based on practical experiences, that it was becoming impossible to plan for improved services without better data on the ethno-racial communities being served.... [W]hile the concept of "race" may be a social-constructed artifact, the problems of racism are very real. Systemic racial discrimination is a major and to some extent a growing problem in our society, and one with serious negative consequences for equitable service provision.

Given these realities, the members of the Working Group believe that the question of improved methods of ethno-racial data collection cannot be ignored. Rather it must be addressed on both a methodological and a practical level, with due attention to questions of cultural sensitivity and principles of ethics, confidentiality, and accountability. (Klimko et. al., 1994:120)

The second and directly related theoretical issue is the danger of misusing ethnic and racial identifiers by separating them from other important demographic variables. Again, Fenton et. al. address this issue directly.

We shall argue... that to treat 'ethnicity' as a variable in health research without due regard for material factors can be highly misleading. Differences in social experience which are expressed by the terms 'race' or 'ethnicity' are not reducible to social class; but the sociological construction of 'ethnic groups' outside the context of social class and material circumstances may appear to give a causal force to ethnicity which it does not deserve. (1995:55)

This point of view coincides completely with that of the Metro Toronto Working Group on Ethnoracial Data Collection, as expressed in the "Better Practices" guide:

Data on race and ethnicity should not be used as surrogates for socio-economic status, income or other predictors and should not be misinterpreted. (Working Group, 1995:32)

The debates on methodological and theoretical issues must continue, but there is also sufficient experience in the field of ethnoracial data collection to suggest some practical measures which can improve the quality of research and reduce the risk of serious errors.

We need an approach which treats ethnic and racial identifiers as partial elements within a broader set of data points, and which accepts the necessity of analyzing the whole set of variables to understand what information can be derived from the ethnoracial identifiers. This is the theoretical approach taken by people like Boyd (1992), and it is a sound one. The recent changes in data definitions for the 1996 Census should facilitate such an approach, within the previously-discussed limits in using official statistics. In applied research this was the approach taken by Fenton et. al. (1995:61), and the Metro Toronto Pilot Project on ethnoracial data collection (Working Group, 1993). These two projects combined factual identifiers such as country of origin, immigration status and religious affiliation with self-identification by race and ethnicity to produce useful data. The Pilot Project also showed that the collection of such data in relation to client service statistics has the added advantages of producing "numbers" which are not available from official statistics, and of linking ethnoracial identifiers to "needs" associated with providing health and social services.

We also need to give due attention to qualitative methods. The point here is not to pit qualitative methods against quantitative, but to recognize that qualitative methods are legitimate when used appropriately, and particularly useful for identifying the "needs" of ethnoracial communities that might be less important in terms of "numbers". There is some acceptance of this viewpoint, as with the statement from the Access and Equity Centre of the Corporation of Metropolitan Toronto that both quantitative and qualitative methods are needed to identify barriers to access, and to evaluate the effectiveness of programs (1994:1). But there is also resistance to this approach both within the research community and among government funders. In this respect it is worth noting that the the success of the two research initiatives described earlier in this paper was due in part to an approach which combined both quantitative and qualitative indicators.

Ethnoracial data collection also must be grounded in a knowledge of the needs and concerns of the communities addressed by the research. This issue was discussed by Dr. Michael Yeo, a Professor of Philosophy at the University of Western Ontario, in the keynote address at the 1994 Symposium on ethnoracial data collection organized by the Metro Toronto Working Group. In his presentation Dr. Yeo proposed a new research paradigm based on community consultation. The necessity for such a paradigm shift, he argued, is rooted in both ethical and highly practical considerations. Various social groups including ethnoracial communities are suspicious of traditional research methods which have failed to improve their social or economic conditions, and have also become more conscious of their citizenship and privacy rights. Without genuine community consultation, Dr. Yeo argued, it is increasingly impossible for researchers to gather the kind of data they need (Working Group, 1994:5-27).

The validity of Dr. Yeo's viewpoint is confirmed by the two research initiatives described earlier in this text. In both cases the appropriate use of quantitative and qualitative indicators, and the sensitive interpretation of the findings, was dependent on the researchers' knowledge of the conditions and concerns within the communities under investigation.

Conclusion

Over the past two decades, public concern over equity issues and for identifying the health and social service needs of ethnoracial communities has led to slow but steady progress in dealing with ethnoracial data collection. There is a growing understanding that official statistics, such as census information and immigrant landing data, can not and should not be used as the sole source of reference for the needs and numbers of ethnoracial communities. Other means are being used successfully to fill the information gap including smaller scale local surveys, linking ethnoracial identifiers to client service records, and qualitative research methods.

A number of theoretical and methodological issues require further attention. One is the meaning attached to identification of individuals or groups by "ethnicity" and "race". A second and related issue is the place of ethnic and racial identifiers within broader sets of demographic variables. A third and criticallly important subject is the appropriate use of qualitative methods within ethnoracial data collection.

Further progress on practical as well as theoretical and methodological issues depends on two fundamental conditions. One is a recognition that the research process must be rooted in a knowledge of the needs and dynamics of the communities under investigation. The second is a vision that seeks to develop ethnic and racial identifiers as essential, but inherently partial, elements within a broader set of economic and social indicators of service needs and barriers to access and equity.

References

Abella, Judge Rosalie Silberman, 1984, Equality in Employment: A Royal Commission Report (Report of the Commission on Equality in Employment), Ottawa: Supply and Services.

Access and Equity Centre, 1994, Why Collect Ethno-Racial and Aboriginal Data? A Guide for Managers, Toronto: Municipality of Metropolitan Toronto, Chief Administrator's Office.

Beaujot, R., K.G. Basavarajappa and R.B.P. Verma, 1988, Income of Immigrants in Canada A Census Data Analysis, Ottawa: Ministry of Supply and Services. Published by Statistics Canada as part of the "Current Demographic Analysis" series (J. Dumas, ed.).

Billingsley, Brenda and Leon Muszynki, 1985, No Discrimination Here? Toronto Employers and the Multi-Racial Work Force, Toronto: Urban Alliance on Race Relations and The Social Planning Council of Metropolitan Toronto.

Boxhill, Wally, 1984, Limitations to the Use of Ethnic Origin Data to Quantify Visible Minorities in Canada, Ottawa: Statistics Canada.

Boxhill, Wally, 1990, Making the Tough Choices in Using Census Data to Count Visible Minorities in Canada, Ottawa: Statistics Canada.

Boxhill, Wally, 1991, Approaches to the Collection of Data on Visible Minorities in Canada: A Review and Commentary, Ottawa: Statistics Canada.

Boyd, Monica, 1992, "Gender, Visible Minority and Immigrant Earnings Inequality: Reassessing an Employment Equity Premise", in Deconstructing a Nation: Immigration, Multiculturalism & Racism in '90s Canada (Vic Satzewich, ed.), Halifax: Fernwood Publishing.

Canadian Task Force, 1988, After the Door has been Opened: Mental Health Issues Affecting Immigrants and Refugees in Canada, Ottawa: Canadian Task Force on Mental Health Issues Affecting Immigrants and Refugees and Health and Welfare Canada.

Choi, Ferreira, Perez and Nguyen, 1989, Multicultural Health Survey Report, Four communities in Metropolitan Toronto, Toronto: Access Alliance for Multicultural Health.

Churchill, Isabel Kaprielian and Stacy Churchill, 1994, The Pulse of the World: Refugees in Our Schools, Toronto: OISE Press.

Citizenship and Immigration Canada, 1994, Immigration Consultations Report, Ottawa: Supply and Services Canada.

Davidson-Palmer, J. & Associates, 1990, Racial Data Collection, report prepared for Race Relations Directorate, Toronto: Ontario Ministry of Citizenship.

deSilva, Arnold, 1992, Earnings of Immigrants: A Comparative Analysis, Ottawa: Economic Council of Canada.

Ellis, Diana, Gayla Reid and Jan Barnsley, 1990, Keeping On Track: An Evaluation Guide for Community Groups, Vancouver: The Women's Research Centre.

Ellis, Diana and Jan Barnsley, 1992, Research for Change: Participatory Action Research for Community Groups, Vancouver: The Women's Research Centre.

Fenton, Steve, Anthony O. Hughes and Christine E. Hine, 1995, "Self-assessed health, economic status and ethnic origin", new community, 21(1):55-68.

Henry, Frances and Effie Ginzberg, 1985, Who Gets the Work: A Test of Racial Discrimination in Employment, Toronto: Urban Alliance on Race Relations and The Social Planning Council of Metropolitan Toronto.

Ho-Lau, Anita, 1992a, The Employment Situation and Social Participation of Chinese Immigrants in Metropolitan Toronto, Toronto: Chinese Information & Community Services.

Ho-Lau, Anita, 1992b, A Profile of Chinese Immigrants in Metropolitan Toronto, Toronto: Chinese Information & Community Services.

Health and Welfare Canada, 1989, Charting Canada's Future: A Report of the Demographic Review, Ottawa: Ministry of Supply and Services.

Herrera, M., I. Rootman, and P. Salamo (1992), Data Collection and Research on Multicultural Health: Ontario Consultations, Toronto: Ontario Multicultural Health Coalition, Centre for Health Promotion University of Toronto, City of Toronto Department of Public Health.

Kelly, Karen, 1995, Collecting Census Data on Canada's Visible Minority Population: A Historical Perspective, Ottawa: Statistics Canada.

Klimko, Natalia, Ted Richmond and Collette Rutherford, 1994, "Ethno-Racial Data Collection: Results of a Pilot Project", Health and Canadian Society, Volume 2, Number 1.

Lautard, Hugh and Neil Guppy, 1990, "The Vertical Mosaic Revisited: Occupational Differentials Among Canadian Ethnic Groups", in Peter S. Li (ed.), Race and Ethnic Relations in Canada, Toronto: Oxford University Press.

Li, Peter S. (ed.), 1990, Race and Ethnic Relations in Canada, Toronto: Oxford University Press.

Medeiros, John, 1991, Family Services for All: Study of Family Services for Ethnocultural and Racial Communities in Metropolitan Toronto, Toronto: Multicultural Coalition for Access to Family Services.

Mental Health Program Services, 1992, Improving Mental Health Supports for Diverse Ethno/Racial Communities in Metro Toronto: A Community Planning Project, Toronto: Metropolitan Toronto District Health Council.

Multiculturalism and Citizenship Canada, 1989, Eliminating Racial Discrimination in Canada, 1989, Ottawa: Supply and Services.

Murdie, Robert A. and Sylvia Novac, with Tom Baker, Ted Richmond, Laine Ruus and John Tibert, 1993, unpublished submission to Statistics Canada Census Content Determination Project.

Ng, Roxana, 1981, "Constituting Ethnic Phenomenon: An Account from the Perspective of Immigrant Women", Canadian Ethnic Studies, Vol. 13, No. 1:97-108.

Ng, Roxana, 1991, "Sexism, Racism and Canadian Nationalism", in Jesse Vorst et. al. (eds.), Race, Class and Gender: Bonds and Barriers, Toronto: Society for Socialist Studies and Garamond Press.

Richmond, Anthony H., 1993, "Education and qualifications of Caribbean migrants in Metropolitan Toronto", new community 19(2):263-280.

Richmond, Ted, 1993, "Income Equality of Canadian Immigrants: Myth or Reality?", unpublished paper written for studies at Ontario Institute for Studies in Education.

Richmond, Ted, 1995, "Needs versus Numbers: The Definition of 'Data' in Identifying Needs of Ethnoracial Communities", unpublished paper written for studies at Ontario Institute for Studies in Education.

Raskin, Carl, 1993, "De facto discrimination, immigrant workers and ethnic minorities: A Canadian overview", International Migration for Employment Working Paper, World Employment Programme, Geneva: International Labour Office.

Sabloff, A. et. al., 1991, Data Collection for Health Care Research, Policy Formulation, and Service Delivery in a Multicultural Society, Toronto: Ontario Ministry of Health.

Samuel, T. John, 1992, Visible Minorities in Canada: A Projection, report for the Race Relations Advisory Council on Advertising, Toronto: Canadian Advertising Foundation.

Silverman, David, 1993, Interpreting Qualitative Data: Methods for Analysing Talk, Text and Interaction, Thousand Oaks, California: Sage Publications.

Sinclair-Jones, Howy, 1993, Feasibility Study on Future Development of the Centre for Spanish-Speaking Peoples (Final Report), Toronto: Sinclair-Jones & Associates.

Stasiulis, Daiva K., 1990, "Theorizing Connections: Gender, Race, Ethnicity and Class", in Peter S. Li (ed.), Race and Ethnic Relations in Canada, Toronto: Oxford University Press.

Statistics Canada, 1994, Report No. 2 1996 Census Consultation Report, Ottawa: Supply and Services Canada.

Statistics Canada, 1995, Census 1996: Questions and reasons why questions are asked, Ottawa: Supply and Services Canada.

Task Force on Access to Professions and Trades in Ontario, 1989, ACCESS! Task Force on Access to Professions and Trades in Ontario, Peter A. Cumming Chair, Enid L.D. Lee and Dimitrios G. Oreopoulos Commissioners, Toronto: Ontario Ministry of Citizenship.

Working Group on Ethno/Racial Data Collection, 1993, Ethno/Racial Data Collection Project, Toronto: Metropolitan Toronto District Health Council.

Working Group on Ethno/Racial Data Collection, 1994, Planning for Diversity: Collection of Data on Culture, Language, Ethnicity and Race, Proceedings of A Symposium Held on May 11, 1994, Toronto: Metropolitan Toronto District Health Council.