Objectification Theory and Deaf Cultural Identity Attitudes: Roles ...

Journal of Counseling Psychology
Copyright 2007 by the American Psychological Association
2007, Vol. 54, No. 2, 178 –188
0022-0167/07/$12.00
DOI: 10.1037/0022-0167.54.2.178
Objectification Theory and Deaf Cultural Identity Attitudes: Roles in Deaf
Women’s Eating Disorder Symptomatology
Bonnie Moradi and Adena Rottenstein
University of Florida
This study examined the generalizability of direct and mediated links posited in objectification theory
among internalization of sociocultural standards of beauty, body surveillance, body shame, and eating
disorder symptoms with a sample of Deaf women. The study also examined the role of marginal Deaf
cultural identity attitudes within this framework. Data from 177 Deaf women indicated positive relations
among internalization, body surveillance, body shame, and eating disorder symptomatology. Consistent
with tenets of objectification theory, body shame mediated the links of internalization and body
surveillance with eating disorder symptoms. In addition, marginal Deaf identity attitudes (but not hearing,
immersion, or bicultural attitudes) were linked uniquely with eating disorder constructs and had
significant indirect relations through internalization with body surveillance, body shame, and eating
disorder symptoms. Implications for practice and future research are discussed.
Keywords: Deaf culture, Deaf identity, objectification theory, eating disorders, body image
Eating disorders and related symptoms are critical concerns,
experience, and identity (e.g., Foster & Kinuthia, 2003; Van Cleve
particularly to women’s health. Although men are also diagnosed
& Crouch, 1997; Wilcox, 1989). Thus, as with other cultural
with eating disorders, women constitute 90% of those diagnosed
minority groups, the cultural experiences and identities of Deaf
with such disorders (American Psychiatric Association, 2000); this
persons may shape their emotional, cognitive, and behavioral
represents roughly 10% of women (Striegel-Moore & Smolak,
functioning (Chovaz, 1998). Indeed, scholars have called for re-
2001). Also, the ubiquity of subclinical body image and weight
search that can inform culturally responsive mental health practice
concerns among women has been recognized as normative discon-
with Deaf clients in general (e.g., Freeman, 1989; Schauben, 2004;
tent (Rodin, Silberstein, & Striegel-Moore, 1984). Thus, eating
Williams & Abeles, 2004) and eating disorder interventions with
disorder theory, research, prevention, and intervention are impor-
Deaf women in particular (Rendon, Hills, & Rappold, 1992). In
tant foci for counseling psychologists (Kashubeck-West & Mintz,
response to such calls, in this study we tested the generalizability
2001). A key area to which counseling psychologists can contrib-
of direct and mediated links posited in objectification theory,
ute is the understanding of issues of diversity within eating disor-
among internalization of sociocultural standards of beauty, body
der research and practice. In fact, there is a paucity of eating
surveillance, body shame, and eating disorder symptoms, with a
disorder research with cultural minority women, and scholars have
sample of Deaf women. Within this framework, the role of mar-
called for attention to the roles of cultural marginalization and
ginal Deaf cultural identity attitudes, a posited culture-specific
devaluation in minority women’s experiences of eating and body
stressor, was also examined. Given limited research in this area,
image problems (e.g., Harris & Kuba, 1997; Root, 1990; Smolak
the roles of other Deaf cultural identity attitudes were examined in
& Striegel-Moore, 2001; Striegel-Moore & Cachelin, 2001;
an exploratory manner as well.
Striegel-Moore, Tucker, & Hsu, 1990; Thompson, 1992).
The present study addresses this need by focusing on the expe-
riences of Deaf1 women, a cultural group that is nearly invisible in
Objectification Theory
the eating disorder literature. Deaf persons are considered mem-
Objectification theory (Fredrickson & Roberts, 1997) has gar-
bers of a unique culture with its own language, history, shared
nered much attention and support as applied to understanding
women’s eating disorder symptoms. This theory posits that gender
role socialization and omnipresent sexual objectification experi-
ences lead women to treat their own bodies as an outsider would
Bonnie Moradi and Adena Rottenstein, Department of Psychology,
evaluate an object (Bartky, 1988, 1990; McKinley, 1998; Noll &
University of Florida.
This article is based on data collected for Adena Rottenstein’s under-
Fredrickson, 1998; Spitzack, 1990). This internalized observer’s
graduate senior thesis, and we thank Jamie Funderburk and Jodi Grace for
perspective on one’s body is called self-objectification and is
their feedback on that project. We also thank our consultants, Terry Coye,
Irene Leigh, Deborah Maxwell-McCaw, Hank Reidelberger, and Mike
1
Tuccelli, for their invaluable suggestions about our survey packet and data
Uppercase D is used to refer to Deaf persons as a cultural group,
collection.
whereas lowercase d signifies a hearing impairment rather than recognition
Correspondence concerning this article should be addressed to Bonnie
of Deaf culture (Napier, 2002; Olkin, 2004). We use the term Deaf
Moradi, Department of Psychology, University of Florida, P.O. Box
throughout the article given our attention to Deaf culture and related
112250, Gainesville, FL 32611-2250. E-mail: moradib@ufl.edu
identity attitudes.
178

---

OBJECTIFICATION THEORY AND DEAF IDENTITY
179
manifested by habitual body surveillance (McKinley & Hyde,
reported strict dieting or fasting, and 50% reported a strong fear of
1996). Greater levels of self-objectification or body surveillance
weight gain. On the basis of these data, Hills et al. argued that body
promote anxiety and body shame, reduce experiences of peak
image and eating disorder symptoms are a serious concern for
motivational or flow states, and diminish awareness of internal
Deaf women and in need of further attention. Hills et al.’s study
bodily states (e.g., hunger, fullness, sexual arousal). These expe-
provided important descriptive data, but its findings have to be
riences, in turn, promote mental health problems such as depres-
interpreted with caution because a presentation on eating disorders
sion, sexual dysfunction, and eating disorders, each of which is
was given before data collection and descriptions of anorexia and
more prevalent among women than among men (Fredrickson &
bulimia were posted in the room during data collection. The extent
Roberts, 1997).
to which these procedures influenced participants’ responses is
Research on objectification theory points to internalization of
unknown.
sociocultural standards of beauty, body surveillance (a manifesta-
In the second known study of eating disorder symptoms among
tion of self-objectification), and body shame as important corre-
Deaf women, DeWalt (1998) posited that Deaf culture may pro-
lates of eating disorder symptoms. Specifically, internalization is
mote a healthier image of beauty and less preoccupation with
related to greater body surveillance or self-objectification, self-
thinness for women than does hearing culture. With 89 adolescent
objectification/body surveillance is related to greater body shame,
girls attending residential schools for the Deaf, DeWalt found that
and body shame is linked with greater eating disorder symptoms
affiliation with Deaf culture was correlated with lower scores on
(e.g., Calogero, Davis, & Thompson, 2005; McKinley & Hyde,
three of eight Eating Disorder Inventory subscales and concluded
1996; Moradi, Dirks, & Matteson, 2005; Morry & Staska, 2001;
that Deaf cultural affiliation reduced eating disorder symptoms for
Slater & Tiggemann, 2002; Tiggemann & Slater, 2001; Tylka &
Deaf girls. Such an interpretation has to be tempered, however,
Hill, 2004). Data also support mediational processes posited in
given that the majority of relations between Deaf cultural affilia-
objectification theory within this chain of relations. Noll and
tion and eating disorder subscales were not significant. Also, there
Fredrickson (1998) found that body shame partially mediated the
was limited reliability and validity evidence for Deaf cultural
relation of self-objectification with eating disorder symptoms; this
affiliation scores, which were obtained from a measure developed
pattern emerged with body mass index (BMI) controlled and when
for and used only in this study. Finally, Deaf cultural identity was
symptoms of bulimia or anorexia were examined. The mediating
assessed unidimensionally (i.e., low to high Deaf affiliation) and
role of body shame in the self-objectification– eating disorder
thus did not account for the potential role of marginal attitudes,
symptoms link has been replicated with young and adult women,
characterized by tension between Deaf and hearing cultures. At-
ballet dancers and nondancers, and women diagnosed with eating
tention to marginal Deaf identity attitudes is important given the
disorders (e.g., Calogero et al., 2005; Slater & Tiggemann, 2002;
posited role of identity conflict, or tension between minority and
Tiggemann & Lynch, 2001; Tiggemann & Slater, 2001). In addi-
majority cultural identity, in minority women’s eating problems
tion, Moradi et al. found that body shame mediated the link of
(e.g., Harris & Kuba, 1997) and in symptomatology of Deaf
internalization of cultural beauty standards with eating disorder
persons (Glickman, 1996).
symptoms as well.
Unfortunately, research on objectification theory, as well as the
Deaf Cultural Identity
broader work on eating disorders, has been conducted mostly with
young, middle-class, White, heterosexual, college women. Al-
Glickman (1996) presented a model of Deaf cultural identity
though some eating disorder studies have expanded the literature
that includes four different Deaf identity formulations and is
to include African American, Asian American, Hispanic Ameri-
informed by Deaf persons’ experiences and the larger literature on
can/Latina, Native American, and international women (Altabe,
cultural and racial identity development. First, culturally hearing
1998; Crago, Shisslak, & Estes, 1996; Le Grange, Stone, &
attitudes reflect idealization of hearing ways of being (e.g., atti-
Brownell, 1998; Osvold & Sodowsky, 1993; Shaw, Ramirez,
tudes, behavior, communication style) and a view of deafness as a
Trost, Randall, & Stice, 2004; Smith & Krejci, 1991; Snow &
medical pathology. Second, culturally marginal attitudes reflect
Harris, 1989), women who are members of cultural minority
struggle or confusion with identity and lack of belonging to hear-
groups in terms of physical ability, including Deaf women, have
ing or Deaf cultures. Such attitudes can be accompanied by feel-
not yet received much attention in the literature.
ings of isolation, bitterness, and psychological symptomatology.
Third, immersion attitudes reflect enthusiastic embrace and poten-
Deaf Women and Eating Disorder Symptomatology
tial idealization of Deaf identity and community, and deprecation
of hearing persons. Finally, bicultural attitudes reflect recognition
A few case studies examining anorexia nervosa with Deaf
of strengths and weakness of Deaf and hearing cultures and per-
young women highlight client-specific life histories, risks, and
sons, along with comfort in both the Deaf and hearing worlds. Of
challenges associated with treatment (e.g., Chapman, Valmana, &
importance, these four formulations do not represent different
Lacey, 1998; De Leo & Santonastaso, 1987; Touyz, O’Sullivan, &
points on a single continuum but are four attitudes that Deaf
Beumont, 1994). Beyond such case studies, we identified only two
persons can hold simultaneously to varying degrees.
empirical studies on eating disorder symptoms with Deaf women.
Within the framework of Deaf cultural identity, Glickman
Hills, Rappold, and Rendon (1991) examined body image and
(1996) identified the isolation, conflict, and stress associated with
eating in 100 Deaf students (58% women) at Gallaudet University.
marginal identity as a potential source of psychological problems.
Among women in the sample, 21% reported current binge eating
This perspective is consistent with posited links between con-
behaviors, 46% overestimated their body size (suggestive of body
flicted identity and eating problems for minority women. Accord-
image disturbance), 9% reported vomiting for weight control, 17%
ing to Harris and Kuba (1997), conflicted identity occurs when

---

180
MORADI AND ROTTENSTEIN
“learned ways of behaving and interacting come into conflict with
women. This study also explored relations between Deaf cultural
the messages from a community with a different ethnocultural
identity attitudes and eating disorder constructs. Thus, the follow-
view” (p. 342). This tension between minority and majority cul-
ing hypotheses were examined with a sample of Deaf women:
tural identity can be particularly stressful for minority women
because dominant cultural standards of women’s beauty typically
1.
As in prior research on objectification theory, positive
devalue minority women’s characteristics; thus, minority women’s
correlations were expected among internalization of so-
internalization of dominant beauty standards can translate identity
ciocultural standards of beauty, body surveillance, body
conflict into body shame (Greene, 1994; Harris & Kuba, 1997;
shame, and eating disorder symptoms.
Neal & Wilson, 1989). At its extremes, such internalization and
shame might manifest as unhealthy attempts to emulate dominant
2.
Given limited research on Deaf cultural identity and
cultural ideals of beauty (e.g., extreme dieting, purging), react
eating disorder constructs, relations between eating dis-
against those ideals (e.g., bingeing), or vacillate between the two
order variables and all four Deaf cultural identity atti-
extremes, ultimately resulting in eating problems (Harris & Kuba,
tudes outlined in Glickman’s (1996) model were ex-
1997; Thompson, 1992). Indeed, the stress associated with adapt-
plored, with the expectation that marginal attitudes would
ing to conflicting cultural values has been linked with eating-
be linked positively with eating disorder constructs
disorder-related symptoms for African American and Latina
(Glickman, 1996; Harris & Kuba, 1997).
women (Perez, Voelz, Pettit, & Joiner, 2002). Although the con-
tent of the cultural experiences of Deaf women may differ from
3.
It was expected that internalization of cultural beauty
that of racial/ethnic minority women, the feeling of conflict be-
standards would mediate the links of marginal attitudes
tween minority and majority identity might exist for both minority
with eating disorder constructs (Greene, 1994; Harris &
groups. Thus, the stress that characterizes marginal identity (e.g.,
Kuba, 1997; Neal & Wilson, 1989).
conflict between messages from hearing and Deaf cultures, feel-
ings of being torn between Deaf and hearing identities) might be
4.
Consistent with prior research on objectification theory,
linked with eating problems for Deaf women, and internalization
it was expected that body shame would mediate the links
of dominant cultural standards of beauty might translate marginal
of internalization and body surveillance with eating dis-
identity into eating problems.
order symptoms.
In addition to the posited role of identity conflict in eating
problems, scholars have argued that some U.S. minority subcul-
Prior research suggests that age and BMI covary with eating
tures may have more flexible views about ideal body size than
disorder constructs (e.g., Jacobi, Hayward, de Zwann, Kraemer, &
exists in the dominant culture and that adoption of dominant
Agras, 2004; Stice, 2002). Thus, BMI and age were explored as
cultural values and rejection of minority cultural values may
potential covariates in order to provide a more stringent test of the
promote eating disorder symptoms for minority women. This
hypotheses.
perspective has been posited for visible racial/ethnic minority
women (e.g., Crago & Shisslak, 2003; Rubin, Fitts, & Becker,
Method
2003) as well as for nonvisible minority women such as lesbian
and Deaf women (e.g., Beren, Hayden, Wilfley, & Striegel-Moore,
Participants
1997; DeWalt, 1998; Siever, 1994). Support for this perspective is
mixed, however, with some studies suggesting that adoption of
Participants were 177 Deaf women ranging in age from 17 to 73
dominant cultural values is linked with greater eating disorder
years (M
36.80, SD
12.05; Mdn
36.00). About 86% of the
symptoms and other studies finding that the link is significant for
sample identified as White/Caucasian, 4% as Hispanic/Latina, 2%
some groups (e.g., Hispanic/Latina) but not others (e.g., Asian
as Asian American/Pacific Islander, 1% as African American/
American) and for some constructs (e.g., eating disorder symp-
Black, 1% as American Indian/Native American, and 6% as mul-
toms) but not others (e.g., body dissatisfaction; see, e.g., Cachelin,
tiracial or other. About 62% identified as exclusively heterosexual,
Veisel, Barzegarnazari, & Striegel-Moore, 2000; Gowen, Hay-
14% as mostly heterosexual, 5% as bisexual, 8% as mostly homo-
ward, Killen, Robinson, & Taylor, 1999). Based on these mixed
sexual, and 5% as exclusively homosexual (about 7% did not
findings, it is unclear whether Deaf identity attitudes that reflect
adoption of dominant cultural values (i.e., hearing attitudes), re-
answer). In terms of social class, 20% of the sample identified as
jection of such values and adoption of Deaf cultural values (i.e.,
low income, 72% as middle income, and 7% as high income.
immersion attitudes), or integration of the two sets of values (i.e.,
Roughly 46% of participants were employed full-time, 22% were
bicultural attitudes) would be related to eating disorder constructs
employed part-time, and 32% were unemployed (about 1% did not
for Deaf women. Given the limited research in this area, these
answer). Also, 25% were undergraduate students, 16% were grad-
possibilities are worthy of exploration.
uate students, and 55% were not students (about 3% did not
answer). In terms of hearing loss, 63% of the sample reported
severe hearing loss of 90 –120 dB, 16% a loss of 70 – 80 dB, 9% a
Overview of the Present Study
loss of 40 – 60 dB, and 11% did not know their level of hearing loss
On the basis of the literature reviewed here and to advance
(1% did not answer). Also, 57% of participants reported that they
understanding of eating disorder symptoms among Deaf women,
were born deaf, 23% became deaf at a later age, 11% had pro-
the present study examined generalizability of key aspects of
gressive hearing loss, and 4% were unsure about the age at which
objectification theory to eating disorder symptoms among Deaf
they became deaf (6% did not answer).

---

OBJECTIFICATION THEORY AND DEAF IDENTITY
181
Procedure
complexity to accommodate the average reading ability for Deaf
persons (Leigh & Anthony-Tolbert, 2001; Olkin, 2004). Resultant
Prior research suggests high levels of Internet access among
changes are noted in the description of instruments. Once consult-
Deaf persons, and Internet surveys have been used to recruit Deaf
ants’ feedback was addressed, the survey was reviewed by two
research participants (e.g., Bowe, 2002; Olkin, 2004). Thus, to
Deaf women who volunteered to complete the survey and provide
facilitate access to Deaf participants, the survey for the present
additional feedback. The volunteers found the survey to be clear
study was posted on an Internet site hosted by our institution.
and understandable and recommended no further revisions. Fi-
Participants were recruited through online organizations that were
nally, Microsoft Word was used to calculate the Flesch–Kincaid
selected based on the following criteria: The group had to be
grade level score for each instrument; the reading level of each
public but not for commercial (e.g., dating service) or sexual
instrument approximated or was below a fourth-grade reading
purposes (e.g., hearing persons who role-play deafness for sexual
level, which is the average reading ability for Deaf persons (Leigh
gratification), have over 50 members, target a Deaf audience (e.g.,
& Anthony-Tolbert, 2001; Olkin, 2004).
sites for interpreters or for hearing parents of Deaf children were
To ensure that participants were actively choosing their re-
excluded), and target primarily members who resided in the United
sponses rather than responding randomly, five validity questions
States (given that examining the role of nationality was beyond the
that asked participants to mark a particular response (e.g., “Please
scope of the present study). To increase diversity in terms of Deaf
click the button for Sometimes”) appeared throughout the survey.
cultural identity, we included groups that appeared highly identi-
Participants who marked an inaccurate response to more than one
fied with Deaf culture (e.g., mentioned Deaf culture, used capital
validity item were eliminated from analyses. The order of instru-
D notation, referenced Deaf pride) as well as those that appeared
ments was counterbalanced, and following Dillman’s (1978) rec-
less identified with Deaf culture (e.g., used hearing or hard-of-
ommendation, the demographic questionnaire appeared at the end
hearing terminology, discussed deafness as an auditory condition,
of the survey.
promoted cochlear implants). An invitation to participate in a study
Deaf identity.
Fischer and McWhirter’s (2001) Revised Deaf
about Deaf identity and attitudes about eating was distributed
Identity Development Scale (DIDS) is a shortened version of the
through group electronic mailing lists, message boards, and online
original DIDS (Glickman & Carey, 1993) and was used to assess
newsletters. The invitation provided a Web link to the survey and
attitudes reflective of hearing, marginal, immersion, and bicultural
encouraged recipients to forward the invitation to other potentially
Deaf identities. The 48 revised DIDS items are rated on a 5-point
interested participants. Upon accessing the survey link, partici-
continuum (1
strongly disagree, 5
strongly agree), and
pants received the informed consent page, clicked the “submit”
appropriate items are averaged to yield subscale scores corre-
button to indicate that they had read the consent form and agreed
sponding to each of the four Deaf cultural identity attitudes. Thus,
to participate, and were then brought to the survey.
each participant has a score on each of the four DIDS subscales;
A total of 217 surveys were submitted and screened to eliminate
higher scores indicate greater levels of the corresponding attitudes.
(a) 13 instances of duplicate submission (i.e., clicked submit
Fischer and McWhirter reported Cronbach’s alphas of .81, .84, .87,
button twice) and potential random responding (i.e., more than one
and .78 for hearing, marginal, immersion, and bicultural items,
inaccurate validity item response, described later), (b) 18 ineligible
respectively. With regard to validity, Fischer and McWhirter found
participants (i.e., no reported hearing loss, young adolescents), and
that prelingually auditorily deaf (auditorily deaf at birth or before
(c) 9 surveys missing substantial amounts of data, resulting in a
the age of 2), postlingually auditorily deaf, and hard-of-hearing
final sample size of 177 Deaf women for the present analyses.
individuals differed in expected directions on Deaf identity atti-
tudes (e.g., hearing scores were higher for hard-of-hearing than for
Instruments
pre- or postlingually deaf individuals; immersion scores were
higher for prelingually deaf than for postlingually deaf or hard-of-
We selected instruments used extensively in prior research to
hearing individuals). Also, as expected, bicultural attitudes, which
assess the objectification theory constructs of interest and the
reflect integration of hearing and Deaf cultural values, and mar-
measure grounded in Glickman’s (1996) model to assess Deaf
ginal attitudes, which reflect feeling marginalized from both hear-
cultural identity attitudes. Following prior recommendations, we
ing and Deaf cultures, were correlated negatively. For the present
took a number of steps to reduce unintentional bias in and maxi-
sample, Cronbach’s alphas were as follows: .81 for hearing, .85 for
mize the applicability of our survey for use with Deaf participants
marginal, .79 for immersion, and .73 for bicultural items.
(Poortinga, Bijnen, & Hagenaars, 1994; Olkin, 2004; Quintana,
Internalization of sociocultural standards of beauty.
The In-
Troyano, & Taylor, 2001). First, four consultants with expertise in
ternalization subscale of Heinberg, Thompson, and Stormer’s
research with Deaf populations and in Deaf communication, cul-
(1995) Sociocultural Attitudes Towards Appearance Questionnaire
ture, and identity evaluated the appropriateness of instruments for
is an 8-item measure of the level of adoption of dominant cultural
use with Deaf women. The consultants were a hearing interpreter
standards of beauty. Items are rated on a 5-point continuum (1
for Deaf students, a Deaf American Sign Language lecturer, a
completely disagree, 5
completely agree). Consultants recom-
Gallaudet professor with research expertise on Deaf culture and
mended eliminating the double negative created by disagreeing
identity, and a Gallaudet language consultant who specializes in
with the item “I do not wish to look like the models in the
research with Deaf populations. Consultants reviewed all instru-
magazines”; thus, we changed the item to “I wish to look like the
ments and recommended modifications to formatting, sentence
models in the magazines.” Appropriate items are reverse scored,
structure, item wording, and instructions. The reasons for these
and item ratings are averaged, with higher scores indicating greater
changes were to eliminate double negatives and expressions that
levels of internalization. Cronbach’s alphas for internalization
might be misunderstood (e.g., “even”) and to reduce sentence
items have been in the .80s (e.g., Heinberg et al., 1995; Morry &

---

182
MORADI AND ROTTENSTEIN
Staska, 2001). In terms of validity, internalization is related pos-
when I am not exercising as much as I should” was changed to “I
itively to body dissatisfaction (Griffiths et al., 2000) and body
worry that something is wrong with me when I am not exercising
image preoccupation (Morry & Staska, 2001). Cronbach’s alpha
as much as I should” and coded in the appropriate direction. Again,
for internalization items in the present sample was .90.
these changes reduced sentence complexity and eliminated double
Body surveillance and body shame.
McKinley and Hyde’s
negatives and expressions that might have been misunderstood by
(1996) Objectified Body Consciousness Scale contains two 8-item
participants (e.g., “even”). Cronbach’s alphas for body shame
subscales that were used in the present study: Body Surveillance
items have been in the .70s and .80s (McKinley, 1999; McKinley
and Body Shame. Items are rated on a 7-point scale (1
strongly
& Hyde, 1996). In terms of validity, body shame scores were
disagree, 7
strongly agree), and a nonapplicable (NA) option is
correlated positively with body surveillance and negatively with
selected if the item does not apply to the participant. Consistent
body esteem (McKinley & Hyde, 1996). Cronbach’s alpha in the
with McKinley and Hyde’s recommendation, NA responses are
current sample was .85.
coded as missing, appropriate items are reverse coded, and non-
Eating disorder symptomatology.
The Eating Attitudes
missing item ratings are averaged to yield subscale scores. Higher
Test–26 (EAT-26; Garner, Olmsted, Bohr, & Garfinkel, 1982) is a
scores indicate greater levels of each construct.
26-item measure of eating disorder attitudes and behaviors, with
The Body Surveillance subscale measures the extent to which a
continuous scores reflecting the continuum of eating problems
participant views her body from an outside observer’s perspective
(Kashubeck-West, Mintz, & Saunders, 2001). Items are rated on a
(i.e., self-objectification). On the basis of the consultants’ recom-
6-point continuum (1
always, 6
never), appropriate items are
mendations, the items “I think it is more important that my clothes
reverse coded, and item ratings are averaged. Higher scores indi-
are comfortable than whether they look good on me” and “I am
cate more maladaptive eating behaviors and attitudes (Mintz &
more concerned with what my body can do than how it looks”
O’Halloran, 2000). Cronbach’s alphas have ranged from .79 to .94
were changed to, respectively, “It is more important that my
(Kashubeck-West et al., 2001). The EAT-26 is one of the most
clothes are comfortable than that they look good on me” and “I
widely used measures of disordered eating (Garner, 1997). In their
care more about what my body can do than about how it looks.”
review of eating disorder measures, Kashubeck-West et al. re-
These changes reduced sentence complexity and potentially con-
ported that EAT-26 scores relate to other measures of eating
fusing expressions. Cronbach’s alphas for body surveillance items
disorder symptomatology as expected and differentiate between
have been in the .70s and .80s (McKinley, 1999; McKinley &
clinical and nonclinical groups. In the present sample, Cronbach’s
Hyde, 1996). With regard to validity, consistent with the tenets of
alpha for EAT-26 items was .88.
objectification theory, women’s body surveillance scores were
BMI.
Participants’ reports of current height and weight were
higher than men’s scores (McKinley, 1998). Also, body surveil-
used to compute BMI using the following formula: [Weight in
lance scores correlated negatively with body esteem and positively
pounds/(height in inches)2]
704.5 (American Obesity Associa-
with body shame (McKinley, 1998). With the present sample,
tion, n.d.). Self-reported and measured height and weight are
Cronbach’s alpha for surveillance items was .67.
highly correlated, and their use is recommended as a practical and
The Body Shame subscale measures participants’ feelings of
valid method to assess BMI (Goodman & Strauss, 2003).
shame when their body does not conform to cultural standards. On
the basis of the consultants’ recommendations, the item “Even
Results
when I can’t control my weight, I think I’m an okay person” was
changed to “When I can’t control my weight, I still think I’m an
Descriptive Information and Preliminary Analyses
okay person”; the item “When I’m not exercising enough, I ques-
tion whether I am a good enough person” was changed to “I
The sample’s mean scores on eating disorder constructs were
question whether I am a good enough person when I don’t exercise
generally close to the midrange of possible scores on each instru-
enough”; and “I never worry that something is wrong with me
ment (see Table 1) and also were comparable to those obtained
Table 1
Summary Statistics and Partial Intercorrelations Between Variables of Interest With Body Mass Index and Age Controlled
Possible
Variable
1
2
3
4
5
6
7
8
range
M
SD
Eating disorder constructs
1. Eating disorder symptoms
—
1–6
2.40
0.62
.88
2. Internalization of beauty standards
.52***
—
1–5
2.75
0.98
.90
3. Body surveillance
.38***
.46***
—
1–7
3.70
0.99
.67
4. Body shame
.59***
.55***
.39***
—
1–7
3.80
1.39
.85
Deaf cultural identity attitudes
5. Hearing
.19*
.11
.02
.19*
—
1–5
1.80
0.57
.81
6. Marginal
.31***
.28***
.13
.27***
.64***
—
1–5
2.10
0.70
.85
7. Immersion
.05
.13
.03
.07
.22**
.09
—
1–5
2.43
0.63
.79
8. Bicultural
.13
.16*
.06
.18*
.39***
.49**
.06
—
1–5
4.03
0.49
.73
Note.
Higher scores reflect higher levels of each construct.
* p
.05.
** p
.01.
*** p
.001.

---

OBJECTIFICATION THEORY AND DEAF IDENTITY
183
with women in prior studies. More specifically, the current sam-
and partial mediating roles of internalization in the links of mar-
ple’s means and standard deviations for internalization (M
2.75,
ginal Deaf identity attitudes with body surveillance, body shame,
SD
0.98), body surveillance (M
3.70, SD
0.99), body
and eating disorder symptoms; and paths required to test full and
shame (M
3.80, SD
1.39), and eating disorder symptoms (M
partial mediating roles of body shame in the links of internalization
2.40, SD
0.62) were comparable to those reported by Griffiths
and body surveillance with eating disorder symptoms. As such, not
et al. (2000) for internalization (M
3.12, SD
0.83), McKinley
all possible paths were estimated, and only those required to test
and Hyde (1996) for body surveillance (M
4.22, SD
0.91) and
the stated hypotheses based on the objectification theory frame-
body shame (M
3.24, SD
1.04), and Mazzeo (1999) for eating
work were included. We used maximum-likelihood estimation
disorder symptoms (M
2.49, SD
0.67). According to criteria
with the covariance matrix of the variables of interest as input. The
used by the National Center for Health Statistics (2004), our
data did not deviate substantially from multivariate normality. Fit
sample’s average BMI fell in the overweight category (M
27.25,
index values indicated a good fit and were as follows: goodness of
SD
6.40), with approximately 2% of the sample categorized as
fit index (GFI)
.99, adjusted goodness of fit index (AGFI)
underweight (BMI
18.5), 40% as healthy weight (BMI of 18.5
.92, comparative fit index (CFI)
.98, normed fit index (NFI)
to 24.9), 33% as overweight (BMI of 25 to 30), and 25% as obese
.97, nonnormed fit index (also known as the Tucker–Lewis index;
(BMI
30). We examined whether BMI and age covaried with
TLI)
.93, and root-mean-square error of approximation (RM-
variables of interest and should be controlled in analyses. BMI was
SEA)
.07. The model accounted for 8% of variance in internal-
correlated with scores on the EAT-26 (r
.19, p
.05) and body
ization, 22% of variance in body surveillance, 39% of variance in
shame (r
.34, p
.001); age was correlated with scores on the
body shame, and 46% of variance in eating disorder symptoms.
EAT-26 (r
.20, p
.01). Thus, these links were controlled in
We followed procedures outlined by Baron and Kenny (1986)
subsequent analyses.
and Frazier, Tix, and Barron (2004) to test proposed mediations in
Hypotheses 3 and 4. Partial correlations and standardized path
Hypothesis 1
coefficients indicated that most preconditions for mediation were
satisfied. Specifically, with regard to Hypothesis 3, preconditions
As indicated in Table 1 and consistent with Hypothesis 1, partial
were not met in the case of the mediating role of internalization in
correlations, controlling for BMI and age, indicated significant and
the link of marginal identity attitudes with body surveillance, given
positive relations among internalization, body surveillance, body
that the partial correlation between marginal attitudes (predictor)
shame, and eating disorder symptomatology.
and body surveillance (criterion) was not significant. On the other
hand, partial correlations indicated that marginal attitudes (predic-
Hypothesis 2
tor) were related to body shame and eating disorder symptoms
(criteria) as well as to internalization (mediator); internalization
Partial correlations indicated that bicultural attitudes were re-
(mediator), in turn, was related to body shame, and eating disorder
lated to lower internalization and body shame; marginal attitudes
symptoms (criteria) when marginal attitudes were accounted for in
were related to higher internalization, body shame, and eating
the path model. Similarly, with regard to Hypothesis 4, partial
disorder symptoms; immersion attitudes were not related to any of
correlations indicated that internalization and body surveillance
the eating disorder constructs; and hearing attitudes were related to
(predictors) each were related to eating disorder symptoms (crite-
higher body shame and eating disorder symptoms (see Table 1).
rion) and body shame (mediator); body shame (mediator), in turn,
Partial correlations consider each of the four Deaf identity attitudes
was related to eating disorder symptoms (criterion) when internal-
in isolation from the others. To attend to the fact that the set of
ization and body surveillance were accounted for in the path
attitudes exist simultaneously in Deaf persons’ cultural identity,
model.
we conducted multiple regression analyses to identify whether any
We multiplied indirect standardized path coefficients to com-
Deaf identity attitudes were related uniquely to eating disorder
pute indirect effects (Cohen & Cohen, 1983) and used Sobel’s
constructs. In these analyses, internalization, body surveillance,
formula (see Baron & Kenny, 1986; Frazier et al., 2004) to test
body shame, and eating disorder symptoms were the criterion
whether indirect effects were significant, indicating significant
variable in each of four separate regression equations with BMI
mediation. Consistent with Hypothesis 3, through internalization,
and age entered as covariates in Step 1 and the set of Deaf identity
marginal attitudes had a significant indirect link of .12 (.29
.42;
attitudes entered as predictors in Step 2. The set of Deaf identity
z
3.36, p
.001) with body shame and .06 (.29
.22; z
2.43,
attitudes accounted for significant variance, beyond BMI and age,
p
.05) with eating disorder symptoms; thus, internalization
in internalization, body shame, and eating disorder symptoms, but
mediated the links of marginal attitudes with body shame and
not in body surveillance (see Table 2). Consistent with Hypothesis
eating disorder symptoms. There was an additional significant
2, marginal attitudes was the only Deaf identity variable account-
direct link of marginal attitudes with eating disorder symptoms but
ing for unique variance in these regression equations. Therefore,
not with body shame (see Figure 1). Finally, through internaliza-
marginal attitudes were included along with objectification theory
tion, there was a significant indirect relation between marginal
variables in the path model testing the proposed mediations.
attitudes and body surveillance of .14 (.29
.47; z
3.46, p
.001), but in this case, the significant indirect relation does not
Path Analysis of Mediations in Hypotheses 3 and 4
suggest significant mediation because the precondition of a link
between predictor (i.e., marginal attitudes) and criterion (i.e., body
We used Amos (Version 4.01; Arbuckle, 1999) to conduct a
surveillance) was not satisfied.
path analysis of a model that included the previously described
With regard to Hypothesis 4, through body shame, internaliza-
covariate paths involving age and BMI; paths required to test full
tion of sociocultural standards of beauty had an indirect link of .17

---

184
MORADI AND ROTTENSTEIN
Table 2
Regression Equations Predicting Eating-Disorder-Related Constructs With Deaf Cultural Identity
Attitudes
Step and predictor
B
t
Total R2
Adjusted R2
Inc. R2
Inc. F
df
Eating disorder symptoms
1
BMI
.01
.13
1.73
.06
.05
.06
5.38**
2, 174
Age
.01
.19
2.55*
2
Hearing
.02
.02
0.22
.15
.12
.10
4.80**
6, 170
Marginal
.27
.31
3.10**
Immersion
.08
.08
1.11
Bicultural
.03
.02
0.24
Internalization of beauty standards
1
BMI
.02
.10
1.29
.02
.01
.02
1.79
2, 174
Age
.01
.08
0.99
2
Hearing
.14
.08
0.85
.12
.09
.10
5.02**
6, 170
Marginal
.45
.32
3.22**
Immersion
.22
.14
1.92
Bicultural
.08
.04
0.50
Body surveillance
1
BMI
.01
.08
1.01
.03
.02
.03
2.63
2, 174
Age
.01
.11
1.45
2
Hearing
.33
.19
1.89
.06
.03
.04
1.59
6, 170
Marginal
.33
.23
2.23*
Immersion
.08
.05
0.63
Bicultural
.04
.02
0.25
Body shame
1
BMI
.07
.32
4.46***
.12
.11
.12
11.29***
2, 174
Age
.01
.05
0.72
2
Hearing
.13
.05
0.56
.19
.16
.08
4.07**
6, 170
Marginal
.40
.20
2.08*
Immersion
.23
.10
1.45
Bicultural
.18
.06
0.79
Note.
Inc.
Incremental. BMI
body mass index.
* p
.05.
** p
.01.
*** p
.001.
(.42
.40; z
4.15, p
.0001), and body surveillance had an
model that constrained to zero (i.e., eliminated) links of marginal
indirect link of .06 (.15
.40; z
2.12, p
.05), with eating
Deaf identity attitudes with internalization, body surveillance,
disorder symptoms. Thus, consistent with Hypothesis 4, body
body shame, and eating disorder symptoms. The fit index values
shame mediated the links of internalization and body surveillance
for the alternative model were as follows: GFI
.95, AGFI
.85,
with eating disorder symptoms. Internalization also had a signifi-
CFI
.91, NFI
.89, TLI
.79, RMSEA
.12; and the
cant direct relation with eating disorder symptoms, but there was
chi-square statistic for the nested model comparison was signifi-
no additional significant direct link between body surveillance and
cant, 2(4)
22.97, p
.0001, indicating a poorer fit of the model
eating disorder symptoms (see Figure 1).
that eliminated the role of Deaf identity attitudes compared with
the fit of the original model.
Alternative Path Model
Discussion
To explore the importance of including marginal Deaf identity
attitudes in the model, we conducted a nested model comparison,
The present results support the generalizability of key aspects of
comparing the model depicted in Figure 1 with an alternative
objectification theory to Deaf women and also point to the impor-

---

OBJECTIFICATION THEORY AND DEAF IDENTITY
185
Age
.20
Internalization
.29
.22
.42
.47
.13
.24
Marginal Deaf
.40
Eating disorder
identity attitudes
Body shame
symptomatology
.15
Body surveillance
.29
Body mass
index
Figure 1.
Path model examining links among variables of interest. Values reflect standardized coefficients.
Dashed lines indicate nonsignificant paths; all other depicted paths are significant at p
.05.
tance of considering marginal Deaf identity attitudes in under-
consistent with Glickman’s (1996) conceptualization of marginal
standing eating disorder and related concerns among Deaf women.
Deaf identity attitudes as a source of psychological distress and
Thus, these results can inform counseling psychologists’ under-
also fit with conceptualizations that tension between internalizing
standing of disordered eating among Deaf women.
minority and majority cultural identities can be a source of body
First, positive relations among internalization of sociocultural
image and eating problems for cultural minority women (Greene,
standards of beauty, body surveillance, body shame, and eating
1994; Harris & Kuba, 1997; Neal & Wilson, 1989; Thompson,
disorder symptoms in the present sample of Deaf women were
1992). The fact that the relations of marginal Deaf identity atti-
consistent with the propositions of objectification theory and prior
tudes with body shame and eating disorder symptoms were medi-
research findings. Similarly, results supported the previously ob-
ated by internalization is consistent with suggestions that internal-
served mediating role of body shame in the links of internalization
izing dominant values regarding beauty can be a critical
and body surveillance with eating disorder symptoms (e.g., Moradi
mechanism for translating conflicted identity into eating problems
et al., 2005; Noll & Fredrickson, 1998; Tiggemann & Slater,
for cultural minority women (Greene, 1994; Harris & Kuba, 1997;
2001). Thus, the pattern of relations among objectification theory
Neal & Wilson, 1989; Thompson, 1992).
variables paralleled relations found in prior research and can
It is possible that links of marginal identity attitudes with eating
inform practice with Deaf women. Specifically, support for the
disorder constructs overlap with links of such attitudes with gen-
mediating role of body shame suggests that reducing body shame
eral psychological distress. Such overlap, however, may be clini-
might be a fruitful target of intervention in group and individual
cally and conceptually meaningful rather than reflecting only a
therapy and in prevention programs with Deaf women. Also, the
nuisance or confound. Indeed, research with women indicates
fact that body shame is linked empirically with internalization and
substantial comorbidity in psychological distress and eating disor-
body surveillance is consistent with conceptualizations of body
der symptoms and overlap in their predictors as well (Graber &
shame as resulting from monitoring one’s own body and compar-
Brooks-Gunn, 1996; Stice, Burton, & Shaw, 2004; Telch & Stice,
ing it with an internalized unrealistic ideal. Thus, strategies for
1998). Thus, parallels in the links of marginal attitudes with eating
reducing body shame with Deaf women might include bringing to
disorder and other psychological symptoms would be consistent
light dominant cultural beauty standards and clients’ body surveil-
with findings that many well established non-culture-specific eat-
lance coupled with helping clients to generate healthy alternative
ing disorder risk factors (e.g., body dissatisfaction, dietary re-
perspectives on beauty and self-evaluation. Generating and inter-
straint) are also risk factors for psychological distress (Bearman,
nalizing such healthy alternatives may be challenging, and clients
Stice, & Chase, 2003; Stice & Bearman, 2001; Stice, Hayward,
may need continued encouragement and reinforcement to combat
Cameron, Killen, & Taylor, 2000).
the omnipresent promotion of unrealistic beauty standards to
On the other hand, the fact that hearing, immersion, and bicul-
women (Fredrickson & Roberts, 1997; Kilbourne & Jhally, 2000).
tural attitudes did not emerge as uniquely related to eating disorder
Another important pattern of findings in the present study in-
constructs adds to prior mixed support for links of minority wom-
volved the role of marginal Deaf identity attitudes. When the set of
en’s body image and eating problems with adopting either domi-
Deaf identity attitudes was considered together, marginal attitudes,
nant or minority cultural values (e.g., Cachelin et al., 2000; Gowen
but not hearing, immersion, or bicultural attitudes, were related
et al., 1999). The present results involving Deaf cultural identity
uniquely with eating disorder constructs. Furthermore, in the path
attitudes suggest that adoption of dominant (i.e., hearing) or mi-
model, marginal identity attitudes were related, directly or indi-
nority (i.e., immersion) cultural identity or general values, or
rectly, to greater internalization, body surveillance, body shame,
balanced integration of the two sets of values (i.e., bicultural), may
and eating disorder symptoms. The significant direct and indirect
not be linked uniquely to eating and body image problems for Deaf
links of marginal attitudes with eating disorder constructs are
women. Instead, the extent to which Deaf women experience

---

186
MORADI AND ROTTENSTEIN
conflict between majority and minority identity (i.e., marginal
literature and tenets of objectification theory, the Cronbach’s alpha
identity) and internalize dominant cultural values, particularly
for body surveillance items was slightly lower in the present
regarding women’s beauty, may be the correlates of body image
sample than in some prior samples. One possibility is that, for Deaf
and eating problems. This pattern of findings suggests that facil-
women, some forms of attention to one’s own body might reflect
itating an open discussion about feelings of marginalization from
the fact that the body is used to communicate in sign language.
Deaf and hearing cultures is particularly important when working
Thus, for Deaf women, some body surveillance might reflect
with Deaf women. Also, working with clients to identify and
communication needs rather than self-objectification. Our consult-
connect with sources of support in Deaf and hearing communities
ants did not raise such concerns about body surveillance items, but
might be useful to the extent that those sources help to reduce
further qualitative exploration of the meaning of body surveillance
feelings of marginalization and foster internalization of healthy
items for Deaf women can inform whether there is a need for
standards of beauty and self-evaluation.
future refinement of this instrument for use with Deaf women.
Finally, the cross-sectional nature of the present data precludes
Limitations and Directions for Future Research
causal interpretations. Thus, these results are consistent with, but
do not directly test, the direction of causality proposed in objec-
A number of limitations must be considered in interpreting the
tification theory. Experimental and longitudinal data can extend
present findings. First, very few studies of body image and eating
the present findings and test temporal and directional relations
disorder symptoms have been conducted with Deaf women, and
implicit in objectification theory. For example, the extent to which
prevalence data for disordered eating within this population are
interventions that reduce proposed precursors (e.g., marginal atti-
needed. On average, our sample’s scores on the eating-disorder-
tudes) also lead to reductions in the subsequent chain of variables
related constructs fell near the midpoint of each scale, suggesting
(e.g., body surveillance, internalization) can be used to evaluate
that the sample did not evidence high levels of disordered eating.
temporal and causal relations. Such research can also inform
However, our sample was not randomly drawn, and our focus was
prevention and intervention efforts. We hope that the present
on continuous rather than clinically significant symptomatology.
findings foster further attention to research and practice that ad-
Thus, again, prevalence data about disordered eating among Deaf
dress body image and eating problems among Deaf women.
women are needed. Similarly, replication of the present findings
with future samples is critical to evaluating the generalizability of
References
results to samples of Deaf women from different backgrounds in
Altabe, M. (1998). Ethnicity and body image: Quantitative and qualitative
terms of race/ethnicity, sexual orientation, socioeconomic class,
analysis. International Journal of Eating Disorders, 23, 153–159.
and other dimensions. Such research can advance understanding of
American Obesity Association. (n.d.). AOA Fact Sheets. Retrieved
the links of Deaf cultural identity attitudes with eating disorder
December 30, 2006, from http://www.obesity.org/subs/fastfacts/
constructs in the context of additional minority cultural identities
morbidobesity.shtml
and values. Also, the present sample ranged in age from young
American Psychiatric Association. (2000). Diagnostic and statistical man-
adulthood to older adulthood, and the analyses controlled for
ual of mental disorders (4th ed., text rev.). Washington, DC: Author.
Arbuckle, J. L. (1999). Amos (Version 4.01) [Computer software]. Chi-
covariation between age and eating disorder symptoms. Explora-
cago: SmallWaters.
tion of the present pattern of findings with younger women is
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
needed, however, given that adolescence and early adulthood may
distinction in social psychological research: Conceptual, strategic, and
be high-risk periods for eating disorder onset (Hoek & van
statistical considerations. Journal of Personality and Social Psychology,
Hoeken, 2003; Jacobi et al., 2004; Stice, Killen, Hayward, &
51, 1173–1182.
Taylor, 1998).
Bartky, S. L. (1988). Foucault, femininity, and the modernization of
In addition, use of different methodology can facilitate further
patriarchal power. In I. Diamond & L. Quinby (Eds.), Feminism and
evaluation of the replicability of the present results. Specifically,
Foucault: Reflections of resistance (pp. 61– 86). Boston: Northeastern
despite its recruitment benefits, the use of an Internet survey in the
University Press.
present study may have limited the representation of Deaf women
Bartky, S. L. (1990). Femininity and domination: Studies in the phenom-
enology of oppression. New York: Routledge.
who do not have access to the Internet. Recruiting participants
Bearman, S. K., Stice, E., & Chase, A. (2003). Evaluation of an interven-
from schools and institutions that serve Deaf populations are
tion targeting both depressive and bulimic pathology: A randomized
additional strategies but can also limit participation to those who
prevention trial. Behavior Therapy, 34, 277–293.
have access to such resources. Thus, no single recruitment strategy
Beren, S. E., Hayden, H. A., Wilfley, D. E., & Striegel-Moore, R. H.
is likely to be free of limitations, but accrual of data using different
(1997). Body dissatisfaction among lesbian college students: The con-
strategies can provide a clearer understanding of eating problems
flict of straddling mainstream and lesbian cultures. Psychology of
among Deaf women. Similarly, careful translation and back-
Women Quarterly, 21, 431– 445.
translation of instruments into sign language would facilitate data
Bowe, F. G. (2002). Deaf and hard of hearing Americans’ instant messag-
collection with Deaf women who have limited reading abilities. To
ing and e-mail use: A national survey. American Annals of the Deaf, 147,
this end, psychometric evaluation of instruments administered in
6 –10.
Cachelin, F. M., Veisel, C., Barzegarnazari, E., & Striegel-Moore, R. H.
American and other sign language is needed.
(2000). Disordered eating, acculturation, and treatment-seeking in a
Relatedly, the present data provide needed evidence of reliabil-
community sample of Hispanic, Asian, Black, and White women. Psy-
ity and validity for scores on measures of important eating disorder
chology of Women Quarterly, 24, 244 –253.
constructs administered in written English to Deaf women. An
Calogero, R. M., Davis, W. N., & Thompson, J. K. (2005). The role of
important point to highlight is that although the present results
self-objectification in the experience of women with eating disorders.
involving body surveillance scores were consistent with prior
Sex Roles, 52, 43–50.

---

OBJECTIFICATION THEORY AND DEAF IDENTITY
187
Chapman, V., Valmana, A., & Lacey, J. H. (1998). A case report of
attitudes to appearance and general dissatisfaction. European Eating
Usher’s syndrome and anorexia nervosa. International Journal of Eating
Disorders Review, 8, 394 – 402.
Disorders, 23, 223–226.
Harris, D. J., & Kuba, S. A. (1997). Ethnocultural identity and eating
Chovaz, C. (1998). Cultural aspects of deafness. In S. S. Kazarian & D. R.
disorders in women of color. Professional Psychology: Research and
Evans (Eds.), Cultural clinical psychology: Theory, research, and prac-
Practice, 28, 341–347.
tice (pp. 377– 400). London: Oxford University Press.
Heinberg, L. J., Thompson, J. K., & Stormer, S. (1995). Development and
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation
validation of the Sociocultural Attitudes Towards Appearance Question-
analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
naire. International Journal of Eating Disorders, 17, 81– 89.
Crago, M., & Shisslak, C. M. (2003). Ethnic differences in dieting, binge
Hills, C. G., Rappold, E. S., & Rendon, M. E. (1991). Binge eating and
eating, and purging behaviors among American females: A review.
body image in a sample of the deaf college population. Journal of the
Eating Disorders, 11, 289 –304.
American Deafness & Rehabilitation Association, 25, 20 –28.
Crago, M., Shisslak, C. M., & Estes, L. S. (1996). Eating disturbances
Hoek, H. W., & van Hoeken, D. (2003). Review of prevalence and
among American minority groups: A review. International Journal of
incidence of eating disorders. International Journal of Eating Disorders,
Eating Disorders, 19, 239 –248.
34, 383–396.
De Leo, D., & Santonastaso, P. (1987). Anorexia nervosa in a prelingually
Jacobi, C., Hayward, C., de Zwann, M., Kraemer, H. C., & Agras, W. S.
deaf young woman: A case report. International Journal of Eating
(2004). Coming to terms with risk factors for eating disorders: Appli-
Disorders, 6, 317–320.
cation of risk terminology and suggestions for a general taxonomy.
DeWalt, J. L. (1998). “I’m proud that I’m deaf”: Deaf culture as a
Psychological Bulletin, 130, 19 – 65.
preventive factor against eating-disordered attitudes and behaviors for
Kashubeck-West, S., & Mintz, L. B. (2001). Eating disorders in women:
adolescent females. Dissertation Abstracts International, 59(08),
Etiology, assessment, and treatment. The Counseling Psychologist, 29,
2866A.
627– 634.
Dillman, D. A. (1978). Mail and telephone surveys: The total design
Kashubeck-West, S., Mintz, L. B., & Saunders, K. J. (2001). Assessment
method. New York: Wiley.
of eating disorders in women. The Counseling Psychologist, 29, 662–
Fischer, L. C., & McWhirter, J. J. (2001). The Deaf Identity Development
694.
Scale: A revision and validation. Journal of Counseling Psychology,