Assessing Race, Ethnicity and Gender in Health

Visit the link below to download the full version of this book:
https://cheaptodownload.com/product/assessing-race-ethnicity-and-gender-in-healt
h-full-pdf-download/
Assessing Race, Ethnicity
and Gender in Health

Sana Loue
Case Western Reserve University
Cleveland, Ohio
Sana Loue
Case Western Reserve University
Cleveland, OH 44106–4945
USA

Cover illustration

C 2006 Superstock

Library of Congress Control Number: 2006920917

ISBN 10: 0-387-32461-5 e-ISBN 0-387-32462-3

ISBN 13: 978-0387-32461-6

Printed on acid-free paper.

C 2006 Springer Science+Business Media, Inc.

All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer Science+Business Media, Inc., 233 Spring Street, New York,
NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use
in connection with any form of information storage and retrieval, electronic adaptation, computer
software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they
are not identified as such, is not to be taken as an expression of opinion as to whether or not they are
subject to proprietary rights.

Printed in the United States of America. (TB/MVY)

9 8 7 6 5 4 3 2 1

springer.com
Acknowledgments

This book would not have been possible without the vision of my former editor,
Mariclaire Cloutier, and my current editor, Bill Tucker. Appreciation is due to
a number of individuals for their assistance with the research. Gary Edmunds,
Nancy Mendez, Ingrid Vargas, and Jenice Contreras deserve recognition for the
many hours that they spent tracking down references.
I owe special thanks to members of the communities with which I work, too
numerous to name individually, for their contributions to my education and to the
evolution of my understanding of identities.

v
List of Tables

1. Comparison of categories and definitions for race and ethnicity

developed by Federal Office of Management and Budget ............ 27

2. Comparison of three approaches to color classification in

Nicaragua ...................................................................... 28

3. Diagnostic criteria and symptoms for gender identity

disorders........................................................................ 68

4. Classification systems of bisexuality...................................... 74

5. Selected studies indicating prevalence of sexual behavior in

comparison with sexual identity ........................................... 76

6. Estimated AIDS cases by race/ethnicity.................................. 91

7. Summary table of selected measures of ethnic and racial

identity and identification ................................................... 121

8. Summary table of selected measures of immigration status .......... 124

9. Characteristics of selected measures of acculturation.................. 128

10. Summary table of selected tools to assess gender role................. 138

11. Summary table of suggested dimensions of assessment for

sexual orientation ............................................................. 146

vii
Contents

Prologue xi

PART I FOUNDATIONS

1. Constructing Categories: Context and Consequences .................. 3

2. Methodological Considerations.............................................. 11

PART II CONSTRUCTS: THEIR DEFINITION AND USE

3. Defining Race, Ethnicity and Related Constructs ........................ 25

4. Defining Sex, Gender, and Sexual Orientation Constructs ............. 54
5. Race, Ethnicity, and Sexual Orientation in Health ....................... 86

PART III ASSESSING GENDER, ETHNICITY, AND RELATED

CONSTRUCTS

6. Measures of Ethnicity, Ethnic Identification, Acculturation,

and Immigration Status........................................................ 119
7. Measures of Sex, Gender, Gender Role, and Sexual Orientation...... 136

Index 153

ix
Prologue

American studies are five times more likely than European trials to report in pub-
lications the race or ethnicity of the study participants (Sheikh, Netuveli, Kai, and
Panesar, 2004). In a review of full-length articles appearing in three major pedi-
atric journals, for instance, it was found that over one-half of the published reports
contained data on participants’ race and/or ethnicity (Walsh and Ross, 2003).
Nearly 80% of hospitals in the United States collect data on race and ethnicity
(Runy, 2004). Despite the relatively high prevalence of collecting and reporting
race/ethnicity information, however, significant debate continues regarding the
wisdom of this practice. The following objections have been voiced with regard
to the collection and use of data relating to race and ethnicity.
r The categories once used are inadequate and categories change too frequently
to make the collection of the data worthwhile. As an example, during the period
from 1990 to 2003, hospitals in Rhode Island utilized three different classification
systems for the collection and recording of race and ethnicity data (Buechner,
2004).
r Categories that we construct may not be valid due to intermarriage. The most
recent census data indicates that over 2% of the American population, or more
than 7 million persons, now acknowledge a multiracial identity (Ahmann, 2005).
Children were more likely to be recorded as having a multiracial identity than
adults, which suggests an increase in interracial coupling.
r Data collection based on self-reports is not valid because individuals change
their self-identification depending upon the context in which they are asked to
so designate (Kaplan and Bennett, 2003).
r Individuals grouped into the same category demonstrate significant genetic di-
versity, so that the construction of the categories is questionable (Erickson, 2003;
Schultz, 2003). There are no gene variants that are present in all individuals of
one population and that are absent in the individuals of another population group
(Bonham, Warshauer-Baker, and Collins, 2005). For instance, Wilson and col-
leagues (2001) found in a study of drug-metabolizing enzymes and genotyping
in eight populations around the world that genotypes clustered into four groups,

xi
xii Prologue

but the four groups did not correspond to the populations from which they had
been drawn.
r Categorizations are too broad to have any definitive medical meaning (Schultz,
2003) and may obscure heterogeneity within groups (Kaufman, 1999; Williams,
2001). Gatrad and Sheikh (2000) have cautioned health care providers to re-
frain from making assumptions about patients’ willingness or unwillingness to
undergo particular screening tests or elective procedures on the basis of their
ethnicity.
r Socially determined categories cannot be applied to biological science (Schultz,
2003).
r Classifying individuals on the basis of socially constructed categories of race
and ethnicity serves to reinforce racial and ethnic divisions that already exist
(Azuonye, 1996; Bogue and Edwards, 1971; Fullilove, 1998; Stolley, 1999).
r The categories developed for race and ethnicity are often used to compare mi-
nority groups to the majority population and these comparisons often focus on
the negative aspects of the health and lives of minority group members (López,
2003).
r A focus on ethnic or racial groups may lead researchers to believe or encourage
them to disregard relevant social or cultural processes that are shared across
group boundaries (Garro, 2001).

Additional concerns have been voiced with regard to the assessment of accul-
turation level and immigration status, two constructs that are relevant to race and
ethnicity.
r Categories of immigration status change too frequently to be useful over time. As
an example, attempts to assess immigrants’ utilization of publicly funded health
care for specific services must be cognizant of the fluctuations in participants’
eligibility for benefits as a result of changes in their immigration status and/or
changes in the relevant legislation (Loue, Cooper, and Lloyd, 2005).
r Self-reports of immigration status will be inaccurate, if they are provided, due
to fears of deportation.
r Questions relating to immigration status will result in poor recruitment and
retention due to fears of the consequences that might ensue following disclosure.
Similar objections have been raised to the collection of data relating to sexual
orientation and sexual identity:
r Sexual orientation and/or sexual identity are fluid over a lifetime and assessment
at one point in time may be inaccurate.
r How an individual chooses to identify him- or herself may not be reflective of
his or her true orientation because of concern about the consequences of self-
disclosure and, accordingly, collection of this information is not valuable.
r Assessment of sexual orientation tends to compare homosexuals and heterosex-
uals, with the resulting inference that homosexuality is somehow “less than” or
“worse than” or undesirable.
 Prologue xiii

Indeed, with respect to any or all of these variables, there is little consensus among
researchers as to how categories should be defined or who should be assigned to
them. And, in view of such vociferous displeasure with the collection and recording
of these variables in the context of health care and research, one must necessarily
question why it is done.
Hospitals have been found to collect these data in order to meet the requirements
of a law or regulation, to improve the quality of care, to ensure the availability of
interpreter services, to improve or maintain community relations, to assist in tar-
geting marketing efforts, and/or because it is perceived as beneficial (Runy, 2004).
Pediatric researchers have reported collecting and reporting race and ethnicity
data because it was required by the institutional review board of their institution,
the National Institutes of Health, and/or the peer-reviewed journal in which they
wished to publish; to conform to a tradition of reporting race and ethnicity data;
to better describe the study population; and/or because they believed that it was
relevant (Walsh and Ross, 2003).
Scholars have suggested other reasons underlying the necessity for the collection
of data relating to race and ethnicity (Mays, Ponce, Washington, and Cochran,
2003). These reasons are relevant, as well, to data regarding sexual orientation.
(1) To describe vital and health statistics. These data provide information that
can be utilized by public health programming and planning to develop programs
targeting specific health issues of concern in a manner that is appropriate to the
affected communities.
(2) To identify risk indicators for specific health outcomes. It has been argued
that ethnicity itself constitutes a risk factor for specific diseases, such as Tay-
Sachs, which occurs predominantly in individuals of Eastern European Jewish
descent (Greenidge, 2004). Alleles associated with sickle cell anemia are not evenly
distributed across racial/ethnic groups, but are found more frequently in African-
American populations (Collins, 2004). The risk factor profile for breast cancer
among African-American women has been found to differ from that of white
women, although the underlying mechanisms of these differences require further
investigation (Bernstein, Teal, Joslyn, and Wilson, 2003).
(3) To improve the delivery of health care services (Hasnaian-Wynia and Pierce,
2005). Nonwhite patients have been found to rate the quality of and their satisfac-
tion with their health care lower than do whites (Haviland, Morales, Reise, and
Hays, 2003). To some extent, this difference may be attributable to disparate treat-
ment by health care providers associated with differences in patient self-identified
or perceived race or ethnicity (Bach, Pham, Schrag, Tate, and Hargraves, 2004;
van Ryn and Burke, 2000).
(4) To identify markers of unmeasured biological differences. Researchers have
reported slower metabolism of some drugs in persons of Asian ethnicity, compared
to white and blacks (Meadows, 2003). It has been hypothesized that this difference
may be attributable, in part, to genetic factors that have not yet been identified.
(5) To identify proxy variables for unmeasured social factors. Race, ethnicity, sex,
and sexual orientation may serve as markers for other variables that we are unable
xiv Prologue

to identify due to limitations in our knowledge and/or the methodologies available

to us.
Accordingly, if we are to collect these data, we must confront and address nu-
merous challenges. These include issues of operationalization of these constructs
in a manner that is appropriate to the research question and the study and target
populations, measurement, and sampling. Ethical issues are also raised by our con-
struction of these categories that must be addressed if we are to remain respectful
of the communities with which we work.
This text addresses many of these issues. Part I focuses on the foundations under-
lying the development of these categories and brings to the fore important ethical
and methodological issues in their construction and their use. Part II provides a
review of the literature that offers definitions of these constructs and their use in
health research. Examples of research that has relied on categories of race, eth-
nicity, and/or sexual orientation are provided, with commentary that discusses the
appropriateness of their use and the conclusions that were drawn as a result. The
final portion of the text provides a summary of many measures currently available
to assess race, ethnicity, sexual orientation, and related constructs.

References
Ahmann, E. (2005).Tiger Woods is not the only “Cablinasian”: Multi-ethnicity and health
care. Pediatric Nursing 31(2): 125–129.
Azuonye, I.O. (1996). Guidelines will encourage the thinking that underpins racism in
medicine. British Medical Journal 313: 426.
Bach, P.B., Pham, H.H., Scrag, D., Tate, R.C., Hargraves, J.L. (2004). Primary care physi-
cians who treat blacks and whites. New England Journal of Medicine 351: 575–584.
Bernstein, L., Teal, C.R., Joslyn, S., Wilson, J. (2003). Ethnicity-related variation in breast
cancer risk factors. Cancer 97 (1 Suppl): 222–229.
Bogue, G., Edwards, G.F. (1971). How to get along without race in demographic analysis.
Social Biology 18: 387–396.
Bonham, V.L., Warshauer-Baker, E., Collins, F.S. (2005). Race and ethnicity in the genome
era: The complexity of the constructs. American Psychologist 60(1): 9–15.
Buechner, J.S. (2004). Hospitalizations by race and ethnicity, Rhode Island, 1990–2003.
Medicine and Health/Rhode Island 87(7): 220–221.
Collins, F.S. (2004). What we do and don’t know about ‘race’, ‘ethnicity’, genetics and
health at the dawn of the genome era. Nature Genetics Supplement 36(11): S13–S15.
Erickson, A.K. (2003). Ethnicity puts clinical trials to the test. Nature Medicine 9(8): 983.
Fullilove, M.T. (1998). Comment abandoning “race” as a variable in public health research:
An idea whose time has come. American Journal of Public Health 88: 1297–1298.
Garro, L.C. (2001). The remembered past in a culturally meaningful life: Remembering as
cultural, social, and cognitive processes. In C. Moore, H. Mathews (Eds.), The Psychology
of Cultural Experience (pp. 105–147). U.K.: Cambridge University Press.
Gatrad, A.R., Sheikh, A. (2000). Birth customs: Meaning and significance. In A. Sheikh,
A.R. Gatrad (Eds.), Caring for Muslim Patients. Oxford, UK: Radcliffe Medical Press
Ltd.
 Prologue xv

Greenidge, K.C. (2005). Race, politics, ethnicity, and science. American Journal of Oph-
thalmology 139: 704–706.
Hasnaian-Wynia, R., Pierce, D. (2005). HRET Disparities Toolkit: A Toolkit for Collect-
ing Race, Ethnicity, and Primary Language Information from Patients. The Health Re-
search and Educational Trust. Available at http://www.hretdisparities.org. Last accessed
November, 2005.
Haviland, M.G., Morales, L.S., Reise, S.P., Hays, R.D. (2003). Do health care ratings differ
by race or ethnicity? Joint Commission Journal on Quality and Safety 29(3): 134–145.
Kaplan, J.B., Bennett, T. (2003). Use of race and ethnicity in biomedical publication. Journal
of the American Medical Association 289(20): 2709–2716.
López, S.R. (2003). Reflections on the Surgeon General’s Report on Mental Health, Culture,
Race, and Ethnicity. Culture, Medicine, and Psychiatry 27: 419–434.
Loue S., Cooper, M., Lloyd L.S. (2005). Welfare and immigration reform and use of prenatal
care among women of Mexican ethnicity in San Diego, California. Journal of Immigrant
Health 7(1): 37–44.
Mays, V.M., Ponce, N.A., Washington, D.L., Cochran, S.D. (2003). Classification of race
and ethnicity: Implications for public health. Annual Review of Public Health 24: 83–110.
Meadows, M. (2003). FDA issues guidance on race and ethnicity data. FDA Consumer
May–June: 36.
Runy, L.A. (2004). Collecting race & ethnicity data. Hospitals & Health Networks 78(8):
30.
Sheikh, A., Netuveli, G., Kai, J., Panesar, S.S. (2004). Comparison of reporting of ethnicity
in US and European randomised controlled trials. British Medical Journal 329 (7457):
87–88.
Schultz, J. (2003). FDA guidelines on race and ethnicity: Obstacle or remedy? Journal of
the National Cancer Institute 95(6): 425–426.
Stolley, P.D. (1999). Race in epidemiology. International Journal of Health Services 29:
905–909.
Van Ryn, M., Burke, J. (2000). The effect of patient race and socio-economic status on
physicians’ perceptions of patients. Social Science and Medicine 50: 813–828.
Wash, C., Ross, L.F. (2003). Whether and why pediatric researchers report race and ethnicity.
Archives of Pediatric and Adolescent Medicine 157: 671–675.
Wilson, J.F., Weale, M.E., Smith, A.C., Gratrix, F., Fletcher, B., Thomas, M.G., Bradman,
N., Goldstein, D.B. (2001). Population genetic structure of variable drug response. Nature
and Genetics 29(3): 265–269.
 Part I
Foundations
1
Constructing Categories: Context
and Consequences

The Social and Political Context of Categories

Increasingly, researchers, clinicians, policy makers, and others concerned with
health and public health have focused their attention on disparities in access to
health care, the quality of health care, and risk for various diseases across minority
populations, which are often defined in terms of their race or ethnicity. These
categories, however,

are human mental constructs . . . they are intellectual boundaries we put on the world in order
to help us apprehend it and live in an orderly way . . . [N]ature doesn’t have categories; people
do (Stone, 1988: 307).

Categories have been distinguished from groups and classes. Jenkins (2003) has
argued that the formation of groups and classes is rooted in the processes of internal
definition, by which individuals signal to others both inside and outside of their
group their self-definition of identity. These internal definitions and designations
may be critical, depending upon the goals of a particular study. As an example, it
may be important that a researcher studying social networks in the LGBT (lesbian,
gay, bisexual, and transgender) communities understand how individuals self-
classify because such distinctions establish their own position within a group and
signal their group membership to others; these distinctions may, for instance,
denote the degree of masculinity or femininity claimed, such as in the distinction
made between butch and femme lesbians or bear and twinkie gay men. (For a
discussion of some of these intra-group classifications, see Gibson and Meem,
2002; Stoller, 1991; Wright, 1997).
In contrast, categories are externally defined, although in actuality, the processes
of defining the “us” and the “them” is an interactive one: the categorization of
“them” assists in the definition of “us”, while the definition of “us” is a function
of the history of relationships with others (Hagendoorn, 1993; Jenkins, 2003). We
form these categories based on perceived commonalities that we believe somehow
distinguish those objects or persons that belong to a specified category and those
that do not (Jay, 1981).

3
4 1. Constructing Categories: Context and Consequences

The process of categorizing may be effectuated in a number of ways (Jenkins,

2003). First, the external categorization of a group may coincide with one or
more elements of the group’s self-identification. Second, the categorization may be
conducted by a group that the original group believes has the power and authority
to so categorize them as a function of their superior knowledge, power, status,
etc. This process of categorization is evident in the identification by scientists and
researchers of those groups believed to be at increased risk of HIV infection during
the early years of the epidemic in the United States. (See chapter 2). Third, the
imposition of the categories may have been effectuated through the use of power
that is exercised through the use of physical force or threat. The effective delineation
in the United States between those who are black and those who are white on
the basis of the “one drop rule” was implemented through the brutal institution of
slavery in a legal context that permitted it. (See chapter 2). Finally, some groups may
resist external categorization but the process of doing so requires the internalization
of that categorization as the focus of denial. As an example, the rejection by
homosexuals of their characterization by the American Psychiatric Association as
different, and therefore mentally ill, by virtue of their sexual orientation requires
the internalization of that differentness (although not the characterization as ill).
Jenkins (2003: 69) has cautioned:
The effective categorization of a group of people by a more powerful ‘other’ is not therefore
‘just’ a matter of classification (if, indeed, there is such a thing). It is necessarily an inter-
vention in that group’s social world which will, to an extent and in ways that are a function
of the specifics of the situation, alter that world and the experience of living in it . . . . [A]
concern with external definition and categorization demands that we pay attention to power
and authority, and the manner in which different modes of domination are implicated in
the social construction of ethnic and other identities . . . . Unless we can construct an under-
standing of ethnicity that can address all of ethnicity’s facets and manifestations, from the
celebratory communality of belonging to the final awful moment of genocide, we will have
failed both ourselves and the people among whom we undertake our research.

Although scientists and researchers may not conceive of themselves as indi-

viduals having power and authority, it is clear that as professionals they have
significantly greater education, knowledge, status, and voice in relation to many
of the groups who are the focus of research and, depending on context, the larger
community as well. As such, they hold significant power and are often viewed
as speaking with the voice of authority. This can be seen, for instance, in the re-
lationship of the investigators to the research subjects in the Tuskegee syphilis
experiment (see chapter 3), the previous characterization of homosexuality as a
mental illness by the American Psychiatric Association (see discussion below),
and the current classification of transssexuality as a mental disorder by that same
association (see chapter 3). Freire (1970: 87) has addressed the significance of the
labels that we assign:
Within the word we find two dimensions, reflection and action, in such radical interaction
that if one is sacrificed—even in part—the other immediately suffers. There is no true word
that is not at the same time a praxis. Thus, to speak a true word is to transform the world.
 Categories and Consequences: Ethical Implications 5

An unauthentic word, one which is unable to transform reality, results when dichotomy
is imposed upon its constitutive elements . . . . Either dichotomy, by creating unauthentic
forms of existence, creates also unauthentic forms of thought, which reinforce the original
dichotomy.

We see, then, that how we construct our categories and labels frames our world,
and that frame further reinforces our use of those categories. And, although our
construction of categories varies and shifts over time, place, and purpose, we often
act as if the categories to which we assign certain attributes and the members to
whom we attribute these attributes are static and remain so. As an example, research
indicates that the designation of an individual’s race or ethnicity may vary over the
course of his or her life depending upon who is charged with the responsibility or
authority to make this designation: the individual’s parent at the time of his or her
birth, the individual him- or herself, the employer reporting on the composition of
the workforce, and the coroner at the time of death (Yanow, 2003). Additionally,
whether an individual self-identifies as a member of a particular group, whatever
its nature, may vary depending upon the social, historical, and political context in
which the designation is to be made.

Categories and Consequences: Ethical Implications

Not infrequently, this process of developing categories involves reference to a
specific group that then becomes the category to which all others are compared.
The construction of this referent category and the subsequent comparison of all
other groups against this referent category may raise significant ethical issues.
As one scholar stated, “If men define situations as real, they are real in their
consequences” (Stone, 2003: 32, quoting W.I. Thomas).

The referent category often takes on a special significance in that it is isolated from
all else [and] is A and pure. Not-A is necessarily impure, a random catchall, to which
nothing is external except A and the principle of order that separates it from Not-A (Jay,
1981: 45).

As a result, members of those categories that are not-A, that are not part of the
referent group and are perceived as different, also may be perceived as being
deviant. This is best understood through an examination of the “social audience”
approach to deviance.
The social audience approach defines who and what are deviant as a function
of the viewers:

[W]hether or not an act is deviant depends on how others who are socially significant in
power and influence define the act. One could commit any act, but it is not deviant in its
social consequences if no elements of society react to it (Bell, 1971: 11).
[A]cts and actors violating the norms of society will be termed “rule-breaking behavior”
and “rule breakers,” while the terms “deviant behavior and “deviant” will be reserved for
acts and actors labeled as deviant by a social audience (Cullen and Cullen, 1978: 8).
6 1. Constructing Categories: Context and Consequences

Kitsuse’s view of deviance is particularly helpful in understanding how the cate-

gorization of individuals may lead to a perception of deviance:
Forms of behavior per se do not differentiate deviants from nondeviants; it is the responses
of the conventional and conforming members of the society who identify and interpret
behavior as deviant which sociologically transforms persons into deviants (Kitsuse, 1962:
253).
[D]eviance is not a property inherent in certain forms of behavior; it is a property conferred
upon these forms by the audience which directly or indirectly witness them (Erikson, 1962:
308).

Ben-Yehuda (1990: 36) has asserted that “who interprets whose behavior, why,
where, and when is very crucial . . . .” He offers two examples in support of his
view. The first is that of Joan of Arc, who was executed in 1431 after having been
convicted as a heretical deviant; she was later canonized by Pope Benedict XV.
The second example is that of the Nobel Laureate chemist Louis Pauling. Pauling
lobbied the United Nations during the late 1950s to end nuclear weapons testing.
Despite the widespread support that he received from other scientists, his stance
led to his interrogation by the U.S. Senate and a prohibition against his attendance
at various international scientific meetings. In 1962, he was awarded the Nobel
Peace Prize in recognition of his efforts to promote peace (Ben-Yehuda, 1990).
Depending upon the construction of the categories to be used in research and
the portrayal of members of those groups, the construction of categories may
also result in the marginalization of their members. Tucker (1990: 7) has defined
marginalization as
the complex and disputatious process by means of which certain people and ideas are privi-
leged over others at any given time . . . [and] the process by which, through shifts in position,
any given group can be ignored, trivialized, rendered invisible and unheard, perceived as
inconsequential, de-authorized, “other,” or threatening, while others are valorized.

Marginalization and consequent devaluation may result when populations,

groups, communities, or individuals do not conform to this idealized referent
group. Marginalization may result from differences in gender (Amaro and Raj,
2000; Bauer, Rodriguez, Quiroga, and Flores-Ortiz, 2000), race and ethnicity
(Gutierez and Lewis, 1997), social class (Marshall and McKeon, 1996), disability
(Braithwaite, 1996), and/or sexuality (Corey, 1996; Yep and Pietri, 1999).
Consider, for example, intersexuality (discussed in greater depth in chapter 4).
Epstein (1990: 104–105, 116) has argued that the medicalization of intersexuality
altered
the social condition of hermaphrodites [when] the availability of surgical and pharmaco-
logical interventions could control or create a public sexual identity for these individuals.
In mandating binary sex differentiation for legal purposes, medical jurisprudence has, then,
imposed a clearcut distinction even though in biomedical terms such a distinction has been
known not to exist . . . [T]he results of total medicalization return us to the semiotics of tera-
tology: individuals with gender disorders are permitted to live, but the disorders themselves
are rendered invisible, are seen as social stigmata to be excised in the operating room.”
 Categories and Consequences: Ethical Implications 7

In this way, the adherence to a belief that humans must necessarily be biolog-
ically of one sex or another reinforces and is then further reinforced by our con-
struction of maleness and femaleness. We construct categories and then engage in
activities to eliminate what is perceived to fall outside of these categories. We point
to these categories that we have constructed as if they had naturally arisen and then
proceed to interpret reality in the context of these “naturally-arising” categories.
A further example of the consequences of categorization of a group by those
holding power and authority is provided by the previous psychiatric designation of
homosexuality per se as a mental disorder. The second edition of the Diagnostic
and Statistical Manual (1968), previously utilized by the American Psychiatric
Association as a guide to the diagnosis of mental disorders, referred to homosexu-
ality as a mental illness. The inclusion of homosexuality in this nosology removed
it from the litany of behaviors that had previously fallen within the jurisdiction
of the church as the guardian of morality and, to some extent, the legal system,
which had viewed homosexuality as criminal in nature and therefore deserving of
punishment (Bayer, 1981).
Accordingly, the psychiatric profession viewed itself as the protector “of de-
viants who had suffered at the hands of society and the more traditional forces of
social control” (Bayer, 1981:11). However, its increasing assumption of responsi-
bility for the control of behaviors previously viewed as immoral and/or criminal,
such as substance abuse and sexual deviance, was viewed by others as an attempt
to widen the scope of its authority and foster a therapeutic state (Kittrie, 1972).
Further, in the context of the civil rights movement of the 1960s and 1970s, ho-
mosexuals began to see themselves as an oppressed minority that had been and
continued to be oppressed by social institutions and ideological standards and
began to campaign actively and vociferously to have homosexuality deleted as a
class of mental illness (Bayer, 1981: 12–13). Bayer has explained the significance
of this action:

To dismiss the significance of the debate over whether homosexuality ought to be included
in the APA’s nosological classification . . . is to miss the enormous importance it carried for
American society, psychiatry, and the homosexual community. By investing the dispute
with great meaning, the participants had themselves transformed it from a verbal duel into
a crucial, albeit symbolic, conflict. The gay community understood quite well the social
consequences of being labeled and defined by others, no matter how benign the posture of
those making the classification. A central feature of its struggle for legitimation therefore
entailed a challenge to psychiatry’s authority and power to classify homosexuality as a
disorder.

Gay organizations in New York City explained to the American Psychiatric

Association in a memorandum:

We are told, from the time that we first recognize our homosexual feelings, that our love for
other human beings is sick, childish and subject to “cure.” We are told that we are emotional
cripples forever condemned to an emotional status below that of the “whole” people who
run the world. The result of this in many cases is to contribute to a self-image that often
lowers the sights we set for ourselves in life, and many of us asked ourselves, “How could
8 1. Constructing Categories: Context and Consequences

anybody love me?” or “How can I love somebody who must be just as sick as I am?”
(Quoted in Bayer, 1981: 119).

Members of the psychiatric profession, too, campaigned for the elimination

of homosexuality from the APA nosological classification. Thomas Szasz argued
against the characterization of variance from behavioral norms as illnesses. In-
deed, Szasz viewed the association of homosexuality with disease as a form of
coercive social control (Bayer, 1981), recognizing the destructive consequences
that resulted:
Psychiatric preoccupation with the disease concept of homosexuality—as with the disease
concept of all so-called mental illnesses . . . conceals the fact that homosexuals are a group
of medically stigmatized and socially persecuted individuals. The noise generated by their
persecution and their anguished cries of protest are drowned out by the rhetoric of therapy—
just as the rhetoric of salvation drowned out the noise generated by the persecution of witches
and their anguished cries of protest. It is a heartless hypocrisy to pretend that physicians,
psychiatrists or normal laymen for that matter really care about the welfare of the mentally
ill in general, or the homosexual in particular. If they did, they would stop torturing him
while claiming to help him (Szasz, 1970: 168).

As alluded to by Szasz, the characterization of individuals or groups as deviant

may lead not only to their marginalization, but to their stigmatization as well. It
has been said that stigmatization
is essentially a relational construct; a stigmatized person must be marked or labeled as
deviating from a social standard or norm, and the label must be socially constructed as neg-
atively valued. An attributional component is inherent in most formulations of the stigma
construct; the mark is regarded as the result or manifestation of a personal attribute, dis-
position, or trait. In other words, the stigmatized individual is perceived as guilty in some
way for having caused or maintained their “marked” condition, even when no evidence for
their culpability is readily apparent (Luchetta, 1999: 2).

Many times, these attributions and the associated stigmatization are premised
on stereotypes. Those who are stigmatized
are pejoratively regarded by the broader society and [are] devalued, shunned or otherwise
lessened in their life chances and in access to the humanizing benefit of free and unfettered
social intercourse (Alonzo and Reynolds, 1995: 304).

Various international guidelines recognize the risk of stigmatization and

marginalization that may result or be associated with participation in health-related
research. The International Guidelines for Biomedical Research Involving Human
Subjects (Council for International Organizations of Medical Sciences, 2002) notes
in the commentary to Guideline 8, which addresses the benefits and risks of study
participation, that
research in certain fields, such as epidemiology, genetics, or sociology, may present risks to
the interests of communities, societies, or racially or ethnically defined groups. Information
might be published that could stigmatize a group or expose its members to discrimination.
Such information, for example, could indicate, rightly or wrongly, that the group has a higher
than average prevalence of alcoholism, mental illness or sexually transmitted disease, or is
 Summary 9

particularly susceptible to certain genetic disorders. Plans to conduct such research should
be sensitive to such consideration, to the need to maintain confidentiality during and after
the study, and for the need to publish the resulting data in a manner that is respectful of the
interests of all concerned, or in certain circumstances not to publish them. The ethical review
committee should ensure that the interests of all concerned are given due consideration; often
it will be advisable to have individual consent supplemented by community consultation.

Guidelines 19 and 21 of the International Guidelines for the Ethical Review

of Epidemiological Studies (Council for International Organizations of Medical
Sciences, 1991) also caution investigators to be aware of this potential risk and
to protect research participants from such risk to the extent possible. Guideline
19 provides that “ethical review must always assess the risk of subjects or groups
suffering stigmatization, prejudice, loss of prestige or self-esteem, or economic
loss as a result of taking part in a study . . . ,” while Guideline 21 notes that

Epidemiological studies may inadvertently expose groups as well as individuals to harm,

such as economic loss, stigmatization, blame, or withdrawal of services. Investigators who
find sensitive information that may put a group at risk of adverse criticism or treatment
should be discreet in communicating and explaining their findings. When the location or
circumstances of a study are important to understanding the results, the investigators will
explain by what means they propose to protect the group from harm or disadvantage; such
means include provisions for confidentiality and the use of language that does not imply
moral criticism of subjects’ behaviour.

The duty of the researcher to be cognizant of and to minimize such risks to

the research participants arises from the ethical principles of beneficence and
nonmaleficence. Beneficence refers to the “ethical obligation to maximize possi-
ble benefits and to minimize possible harms and wrongs,” while the principle of
nonmaleficence counsels researchers to protect participants from avoidable harms
(Council for International Organizations of Medical Sciences, 1991).

Summary
It is critical that researchers consider the larger ethical, social, and political im-
plications of their categorization of populations and research participants. Efforts
must be made to reduce the likelihood that these categorizations, which may be
formulated with the best of intentions to improve health, will result instead in the
reinforcement of stereotypes and the marginalization and/or devaluation of specific
groups.

References
Alonzo, A.A., Reynolds, N.R. (1995). Stigma, HIV and AIDS: An exploration and elabo-
ration of a stigma trajectory. Social Science and Medicine, 41(3), 303–315.
American Psychiatric Association. (1968). Diagnostic and Statistical Manual, DSM-II.
Washington, D.C.: Author.
10 1. Constructing Categories: Context and Consequences

Bayer, R. (1981). Homosexuality and American Psychiatry: The Politics of Diagnosis. New
York: Basic Books, Inc.
Bell, R.B. (1971). Social Deviance. Homewood, Illinois: Dorsey.
Ben-Yehuda, N. (1990). The Politics and Morality of Deviance: Moral Panics, Drug Abuse,
Deviant Science, and Diverse Stigmatization. Albany, New York: State University of
New York Press.
Comstock, G.D. (1991). Violence against Lesbians and Gay Men. New York: Columbia
University Press.
Cullen, F.T., Cullen, J.B. (1978). Toward a Paradigm of Labeling Theory. Lincoln, Nebraska:
University of Nebraska.
Erikson, K.T. (1962). Notes on the sociology of deviance. Social Problems, 9, 307–314.
Epstein, J. (1990). Either/or—neither/both: Sexual ambiguity and the ideology of gender.
Genders 7: 99–142.
Freire, P. (1970). Pedagogy of the Oppressed. (Trans. M.B. Ramos). New York: Continuum.
Gibson, M., Meem, D.T. (Eds.). (2002). Femme/Butch: New Considerations of the Way We
Want to Go. Binghamton, New York: Harrington Park Press.
Healey, D. (2001). Homosexual Desire in Revolutionary Russia: The Regulation of Sexual
and Gender Dissent. Chicago, Illinois: University of Chicago Press.
Jenkins, R. (2003). Rethinking ethnicity: Identity, categorization, and power. In J. Stone,
R. Dennis (Eds.), Race and Ethnicity: Comparative and Theoretical Approaches (pp. 58–
71). Malden, Massachusetts: Blackwell Publishing.
Kitsuse, J.L. (1962). Societal reaction to deviant behavior: Problems of theory and method.
Social Problems 9: 247–256.
Kittrie, N. (1972). The Right to Be Different. Baltimore, Maryland: Johns Hopkins Univer-
sity Press.
Luchetta, Y. (1999). Relationships between homophobia, HIV/AIDS stigma, and HIV/AIDS
knowledge. In L. Pardie, T. Luchetta (Eds.), The Construction of Attitudes Towards
Lesbians and Gay Men (pp. 1–18). Binghamton, New York: Harrington Park Press.
Mason, A., Palmer, A. (1996). Queer Bashing: A National Survey of Hate Crimes against
Lesbians and Gay Men. London: Stonewall.
Stoller, R.J. (1991). Pain and Passion: A Psychoanalyst Explores the World of S&M. New
York: Plenum Press.
Stone, D.A. (1988). Policy Paradox and Political Reason. Boston, Massachusetts: Little,
Brown.
Stone, J. (2003). Max Weber on race, ethnicity, and nationalism. In J. Stone, R. Dennis (Eds.),
Race and Ethnicity: Comparative and Theoretical Approaches (pp. 28–42). Malden,
Massachusetts: Blackwell Publishing.
Szasz, T. (1970). The Manufacture of Madness. New York: Delta Books.
Tucker, M. (1990). Director’s forward. In R. Ferguson, M. Gever, T.T. Minh-ha, C. West
(Eds.), Out There: Marginalization and Contemporary Cultures (pp. 7–8). New York:
The New Museum of Contemporary Art/MIT Press.
Wright, L. (1997). The Bear Book: Readings in the History and Evolution of a Gay Male
Subculture. Binghamton, New York: Harrington Park Press.
Yanow, D. (2003). Constructing “Race” and “Ethnicity” in America: Category-Making in
Public Policy and Administration. Armonk, New York: M.E. Sharpe.
2
Methodological Considerations

The selection of instruments for use in a particular study or the construction of new
instruments to assess ethnicity, race, gender, sex, and related constructs requires
a consideration of various methodological issues, in addition to the context in
which these tools are to be developed and the ethical implications of the categories
once they have been developed. Issues to be considered prior to deciding which
of existing instruments to use or whether and how to construct a new instrument
include the focus of the research question, the format to be used to collect the data,
and how the population of interest is to be sampled. The selection of an instrument
for use, or the development of a new instrument, also requires attention to the
instrument’s validity, reliability, and the possibility of misclassification associated
with its use (McDowell and Newell, 1996). A basic understanding of these issues
is important in order to better evaluate the literature that exists with regard to the
constructs that are the focus of this text. This is not, however, a comprehensive
discussion of these issues, which can be the focus of entire books themselves,
and the reader is urged to consult the sources listed at the end of this chapter for
additional guidance.

Framing the Research Question and the Research

How a research question is framed and the design of the study that will be un-
dertaken for its investigation are critical issues to be resolved prior to identifying
the instruments to be used or whether and how to develop a new instrument for
the assessment of any of the constructs discussed in this text. Issues requiring
consideration include the following.
(1) The time period of interest. Because one’s self-identity may change over
time with respect to ethnicity, race, sexual orientation, and related constructs, it is
important to determine at what point in time these are to be assessed. For instance,
does the research question demand an understanding of how an individual currently
self-identifies? This might be relevant, for instance, in studies assessing current
patient satisfaction with health care. Or, does the study focus on the impact of
stigmatization on one’s health status over time? In this case, it may be important

11
12 2. Methodological Considerations

to assess the individual’s identity over time and/or how an individual is perceived
by others in terms of his or her race, ethnicity, sexual orientation, etc.
This is an important consideration even in instances in which the researcher has
decided to rely on pre-formulated categories for the classification of the research
participants. For instance, the manner in which the federal government has defined
various ethnicities and races has changed over time. (See chapter 3.) A study that
spans time periods that use different classification systems may find that the choices
provided to respondents at the initiation of the study period may no longer be in
use towards the end, and researchers may have to reconcile responses to the newer
categories.
(2) The focus of the research question. The concepts of race, ethnicity, sex, gen-
der, sexual orientation and related concepts are multidimensional. As an example,
depending upon the focus of the research, a determination of ethnicity may require
an assessment of the ethnicity of an individual’s parents and grandparents in addi-
tion to a consideration of the origin of the individual research participant. A sexual
history that focuses on the number and sex of one’s sexual partners may be suffi-
cient to answer a question focusing on the sex of one’s sexual partners, but it may
not be adequate to determine an individual’s sexual orientation, which is a function
of emotional attraction, physical attraction, sexual fantasies, self-definition, and
opportunity.
These characteristics are also subject to identification not only by the individ-
ual who may be a participant in the research, but also by the observer as well.
For instance, an individual may self-identify his or her race (suspending, for the
moment, a discussion as to whether race exists), but an individual’s race is also
subject to the perception of the observer who, on the basis of criteria that he or
she as somehow developed, will make a judgment regarding the individual’s race.
Consequently, it is important to consider in framing the research question whether
participants’ self-identity as to race, ethnicity, sexual orientation, gender, etc. is
important or whether it is the perception of specified observers that is critical. For
instance, how an individual identifies him- or herself with respect to race may not
be as important as how the individual is viewed by others, if the focus of the study
is an exploration of the effects of political marginalization.

Selecting the Sample

How the study sample is selected and the size of the sample are critical issues. A
biased sample may lead to erroneous conclusions and an inadequate sample will
not have sufficient statistical power to detect the hypothesized effect. This section
very briefly reviews issues related to sampling. The issue is a complex one, and
readers are referred to other texts for an in-depth exploration of the topic (Cochran,
1977; Kish, 1965; Levy and Lemeshow, 1980).
The sampling procedure is framed around the sampling unit. In many cases, this
will be the individual, but it can be a family or household, or area of a community.
The sample that will be constructed consists of the sampling units that have been
 Selecting the Sample 13

selected from among those that are eligible for inclusion in the study (Kelsey,
Thomson, and Evans, 1986). For example, if an investigator wished to know the
proportion of households of a particular ethnic group had health insurance, the
sample would consist of a portion of those households where a designated member
was of that ethnicity.
The sampling frame refers to the list of the population from which the sample
will be drawn. In some cases, this is unknown and unknowable. For instance, a
study focusing on the experience of homophobia by gay and lesbian individuals
would have a difficult, if not impossible, task to construct a sampling frame,
since it would not be possible to know of every individual who self-identified
as gay or lesbian, since many may not wish to acknowledge their orientation
publicly.
In instances in which it would be difficult to locate and recruit study partici-
pants due to the nature of the research, investigators often rely on a convenience
sample comprised of volunteers. This strategy, however, can introduce bias into
the selection process. Snowball sampling, in which already-recruited participants
identify other individuals as potential participants, permits investigators to more
easily locate and recruit “hidden” individuals, but may also introduce selection
bias because the individuals recruited through respondents are more likely to be
like the respondents.
Probability sampling is advisable when it is feasible, in order to reduce the
possibility of selection bias. There are four basic designs for probability sampling:
simple random sampling, systematic sampling, stratified sampling, and cluster
sampling.
Simple random sampling requires knowledge of the complete sampling frame
in advance (Kelsey, Thompson, and Evans, 1986). This strategy means that each
sampling unit in the population has an equal chance of being selected for partici-
pation. This method does not require advance knowledge of the population itself,
but may be very inefficient.
Systematic sampling refers to the selection of sampling unit, such as individuals
or households, at regularly spaced intervals within the sampling frame, such as
every third household. This method has several advantages in that it does not
require advance knowledge of the sampling frame, as it can be constructed as the
process progresses and is generally relatively simple to implement.
Stratified sampling requires the division of the population into strata and a
sample is selected from each such strata. This process is significantly more complex
than the other strategies, but offers increased precision and may facilitate the
inclusion of specific groups of persons.
Like stratified sampling, cluster sampling divides the sample into groups, such
as clusters of homes. A sample is then taken of these clusters for inclusion in
the study or, alternatively, a subsample of these sample clusters is utilized. As
an example, an investigator wishing to study the prevalence of violence in public
housing projects might divide all such projects into clusters by geographic area and
then take a sample of these clusters. The households within these clusters could
then be queried about the violence in the public housing projects.