Content uploaded by Kristopher Michael Goodrich
Author content
All content in this area was uploaded by Kristopher Michael Goodrich on Jun 16, 2015
Content may be subject to copyright.
This article was downloaded by: [University of New Mexico], [Kristopher M. Goodrich]
On: 18 August 2014, At: 11:33
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Homosexuality
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/wjhm20
An Examination of the Heterosexism
Scale
Kristopher M. Goodrich PhDa, James P. Selig PhDb & Gene Crofts BAc
a Counselor Education Program, College of Education, University of
New Mexico, Albuquerque, New Mexico, USA
b Educational Psychology Program, University of New Mexico,
Albuquerque, New Mexico, USA
c Counselor Education Program, University of New Mexico,
Albuquerque, New Mexico, USA
Accepted author version posted online: 02 Jun 2014.Published
online: 28 Jul 2014.
To cite this article: Kristopher M. Goodrich PhD, James P. Selig PhD & Gene Crofts BA (2014)
An Examination of the Heterosexism Scale, Journal of Homosexuality, 61:10, 1378-1392, DOI:
10.1080/00918369.2014.928168
To link to this article: http://dx.doi.org/10.1080/00918369.2014.928168
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Journal of Homosexuality,61:1378–1392,2014
Copyright © Taylor & Francis Group, LLC
ISSN: 0091-8369 print/1540-3602 online
DOI: 10.1080/00918369.2014.928168
An Examination of the Heterosexism Scale
KRISTOPHER M. GOODRICH, PhD
Counselor Education Program, College of Education, University of New Mexico, Albuquerque,
New Mexico, USA
JAMES P. SELIG, PhD
Educational Psychology Program, University of New Mexico, Albuquerque, New Mexico, USA
GENE CROFTS, BA
Counselor Education Program, University of New Mexico, Albuquerque, New Mexico, USA
Researchers explored the factor structure of Park’s Heterosexism
Scale (2001) with heterosexual parents of lesbian, gay, or bisex-
ual sons or daughters. Results suggested a two-factor solution, but
results varied from Park’s two-factor model. Additionally, rela-
tions between the Heterosexism Scale subscales and measures of
cognitive flexibility, religious commitment, and general family
functioning were examined. Implications for clinical practice and
future research are discussed.
KEYWORDS lesbian, gay, bisexual, LGB, heterosexism,
exploratory factor analysis, confirmatory factor analysis, family
Heterosexism is a construct that was originally identified by Herek (1990)and
is currently defined as a “systemic process of privilege toward heterosexuality
related to homosexuality based on the notion that heterosexuality is normal
and ideal” (Dermer, Smith, & Barto, 2010,p.327).Toalargeextent,ithas
replaced the earlier construct of homophobia, the irrational fear of homosex-
uals. Since the 1970s, there has been a desire to scientifically measure these
constructs and attitudes toward lesbians and gay males in general. The best-
known or most widely used scales include the Attitudes Toward Lesbians
and Gay Men (ATLG; Herek, 1994), the Homophobia Scale (Daly, 1990), the
Address correspondence to Kristopher M. Goodrich, Counselor Education Program,
College of Education, University of New Mexico, 129 Simpson Hall, MSC05-3040,
Albuquerque, NM 87131, USA. E-mail: kgoodric@unm.edu
1378
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1379
Homosexism Scale (Hansen, 1982), and the Heterosexual Attitudes Toward
Homosexuality Scale (Larsen, Reed, & Hoffman, 1980).
Across disciplines that study historically and culturally marginalized
groups, there has been a recognition over the past quarter century that
expressions of prejudice have become more covert and even unconscious
(Ellis, Kitzinger, & Wilkinson, 2002; Peel, 2001; Shelton & Delgado-Romero,
2011;Sueetal.,2007). In response, new instruments have been designed that
take into account the influence of social desirability or political correctness,
especially if items are too overtly or obviously worded (e.g., the Modern
Racism Scale [McConahay, 1986] and the Modern Sexism Scale [Swim, Aikin,
Hall, & Hunter, 1995]). While the early scales for measuring attitudes toward
lesbians and gay males contributed a great deal of knowledge about the atti-
tudes of the day, they, too, are now seen as addressing the topic too directly
and, as such, failing to measure what may be hidden attitudes (Morrison &
Morrison, 2002; Worthington, Dillon, & Becker-Schutte, 2005).
Besides this concern with a lack of subtlety, most of the older scales
have been criticized as lacking in psychometric evidence (Boysen, Vogel,
& Madon, 2006; Boysen, Vogel, Madon, & Wester, 2006; Gallor, 2006;Park,
2001), as the evidence for the validity and reliability of the scores from many
of the scales was based on a single study with no further investigation.
An exception is Herek’s (1994) ATLG scale, which was constructed out of
extensive studies and is broadly considered to be psychometrically sound
(Park, 2001). The ATLG consists of 20 items, 10 of which address attitudes
toward the lesbian population and 10 items addressing attitudes toward the
gay male population. The use of the ATLG today, however, is questioned
for the previously discussed overtness of its items as well as the omission of
items addressing attitudes toward the bisexual population (Park, 2001).
The Heterosexism Scale (Park, 2001) was developed to address the
need to more covertly measure attitudes of heterosexuals toward lesbian,
gay, and bisexual (LGB) persons and the need to acknowledge the bisexual
population in the sexual orientation spectrum. As such, the Heterosexism
Scale does not measure or separate attitudes toward individuals within the
subpopulations of lesbian, gay, or bisexual but instead measures attitudes
toward the LGB population as a whole. Extant research has supported that
there is a general belief system prevalent within our culture that positions
heterosexuality as superior to LGB status and can be found within law
(Lambda Legal, 2011;Park,2001), religion (Barnard, 2009;Garcia,Gray-
Stanley, & Ramirez-Valles, 2008; Goodwill, 2000; Love, Bock, Jannarone,
& Richardson, 2005;Park,2001; Sherry, Adelman, Whilde, & Quick, 2010),
education (Park, 2001; Goodrich & Luke, 2009,2010), family (Gallor, 2006;
Goodrich, 2009; Goodrich & Gilbride, 2010), and mass media (Park, 2001).
A further goal of its creator was for the Heterosexism Scale to be
psychometrically sound (Park, 2001). Toward this end, the scale was devel-
oped in five studies, including one exploratory and two confirmatory
analyses, primarily with a college student sample of convenience. Park (2001)
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1380 K. M. Goodrich et al.
found that a 17-item inventory, with two underlying factors (Superiority
of Heterosexuality and Tolerance/Acceptance of Nonheterosexuality), best
represented the construct under study.
The scale was further utilized with undergraduate populations (Boysen,
Vogel, & Madon, 2006; Boysen, Vogel, Madon, & Wester, 2006) and also used
for samples of heterosexual parents of LGB children (Gallor, 2006; Goodrich
& Gilbride, 2010). As the original validity evidence was based on use of the
scale with college students, the purpose of the current study is to extend
the validity evidence to use with heterosexual parents of LGB individuals,
exploring whether the same factor structure exists for populations of hetero-
sexual parents with LGB sons or daughters. Examining this instrument with
a broader population than college students is important, because college
students may respond to instruments differently than the general popula-
tion (Kazdin, 2002). In addition, as the Heterosexism Scale has been used
by a growing number of scholars exploring the experiences of heterosexual
parents with LGB children (e.g., Gallor, 2006; Goodrich & Gilbride, 2010),
it makes intuitive sense to explore the psychometric properties with other
populations. Therefore, the research question of this study is whether Park’s
two-factor solution for the Heterosexism Scale fit the data from a sample of
primarily middle-aged heterosexual parents of LGB sons or daughters.
Additionally, the researchers explored how the subscales of the
Heterosexism Scale relate to measures of cognitive flexibility (Martin &
Rubin, 1995), religious commitment (Worthington et al., 2003), and gen-
eral family functioning (Epstein, Baldwin, & Bishop, 1983). The measure
of cognitive flexibility was chosen because the concept of cognitive flexibil-
ity is associated with persons being able to consider multiple viewpoints and
potentially change their biases; researchers hypothesized that heterosexism
would be negatively correlated with cognitive flexibility. The measure of
religious commitment was used because there is an assumption that persons
who are more religious might also be more heterosexist. Finally, the con-
cept of general family functioning was chosen because this study measured
how functional the home environment was in families with heterosexual par-
ents with LGB-identified sons or daughters. The researchers believed that a
measure of heterosexism would be negatively correlated with general fam-
ily functioning because the assumption was that a functional family would
accept their LGB children.
METHOD
Participants
The participants in this study were recruited as part of a project (Goodrich &
Gilbride, 2010) to understand family functioning after a son’s or daughter’s
disclosure as LGB. All study participants identified as heterosexual parents of
self-identified LGB sons or daughters. Parents of transgender or intersex sons
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1381
or daughters were not surveyed as part of the project, because the study was
focused on parents’ response to their son’s or daughter’s sexual, rather than
gender, identity.
Of the 441 parents in the final sample, 339 parents (77%) identified as
female and 84 parents (19%) identified as male; 18 participants (4%) refused
to disclose their gender. Participants ranged in age from 35 to 88 years of
age, with a mean of 60 and standard deviation of 9.47. A large majority
of participants identified as Caucasian/White (429, 97%), with three partici-
pants (0.6%) identifying as Latino/Hispanic, one participant (0.2%) as Asian
American, one (0.2%) as biracial, four (0.8%) as multiethnic, and three (0.6%)
as “other.” A more comprehensive description of participant demographics
can be found in Goodrich and Gilbride (2010).
Sampling
The first author used purposive sampling, and participants were recruited
from several sources. The first sets of parents were recruited from support
organizations aimed at parents and friends of LGB persons (e.g., Parents,
Family, and Friends of Lesbians and Gays [PFLAG], 2011) and LGBT com-
munity and youth centers (Consortium of Higher Education LGBT Resource
Professionals, 2011; Funders for Lesbian and Gay Issues, 2011; theGLBT.com,
2011). The researchers specifically identified multicultural LGBT organiza-
tions (Funders for Lesbian and Gay Issues; theGLBT.com) because previous
scholarship had shown an underidentification of parents of color within LGB
research (Hom, 1994; Merighi & Grimes, 2000).
Next, parents from the nonaffliated/disidentified population, includ-
ing Exodus International (2011) and Focus on the Family (2011), were
recruited. These groups were contacted to increase the variability of scores
on the study’s instruments, because the first author was concerned it could
be restricted due to the self-report nature of the study. Finally, members
of religious/spiritual organizations were sampled as an additional way to
increase sample variability. This served to increase the likelihood that the
results would generalize not only to social support groups such as PFLAG but
also to parents not involved in specific LGB-related organizations. Religious
groups were identified from their national Web sites, and religious organiza-
tions with LGB-specific groups were found from the Sister Friends Together
(2011) Web site. Finally, snowball sampling was employed, in which partici-
pants were asked to forward the survey to other parents who might meet the
criteria for participation, including the second parent of the identified child.
Participant Screening
A total of 687 respondents received an invitation to participate and clicked to
the first page of the Web survey. Two of those respondents were immediately
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1382 K. M. Goodrich et al.
removed because it was apparent by their answers (e.g., participant’s age,
answers about “parenting relationship”) that they were not heterosexual par-
ents of LGB children, but instead the self-identified LGB child. As this was
a Web-based survey that snowballed to a variety of groups and organiza-
tions, the original size of the solicited sample is unknown, complicating
the calculation of a survey response rate. Of the remaining 685 respon-
dents, 450 parents completed the full survey for a completion rate of 65.7%.
After screening the data for normality and outliers, 10 responses were dis-
carded due to missing significant amounts of data across the instruments
used (Goodrich & Gilbride, 2010). The present analyses are based on the
data of the 441 subjects retained in the final data set.
Procedure
The groups described above were contacted via postal mail, phone calls, or
e-mail to inform them of the study and were asked to forward to each of
their members an informational letter and/or e-mail about the project. In the
correspondence, participants were directed to the online survey, using the
Surveymonkey.com platform. The survey was constructed using instruments
that reflect the operationalized definitions of the latent constructs from a pilot
study (Goodrich, 2009). As instrument reliability and validity has been a con-
cern in LGB research (Savin-Williams & Dube, 1998), special attention was
given to utilizing measures with reported internal consistency coefficients
≥0.7.
Instruments
HETEROSEXISM SCALE
The Heterosexism Scale consists of 17 items, each with a 6-point Likert-
type scale (i.e., strongly disagree, disagree, slightly disagree, slightly agree,
agree, and strongly agree). Higher scores indicate greater heterosexist atti-
tude and lower scores indicate lower levels of intolerance toward LGB
individuals (Park, 2001). The scale includes two subscales: the Superiority
of Heterosexuality and the Tolerance/Acceptance of Nonheterosexuality.
Examples of questions used on this scale include “Heterosexual couples
make better candidates for parents than do same-sex couples for adoption”
and “I believe the lives of lesbian, gay, and bisexual individuals could not
be as fulfilling as those of heterosexual individuals.” Consistent with Park
(2001), 10 of these items were reverse scored to maintain that meaning of
the subscales.
Park (2001) conducted five studies on this instrument to evaluate its
reliability and validity. Other researchers have shown the instrument to have
internal consistency of 0.92 in a study of undergraduate college students
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1383
(Boysen, Vogel, & Madon, 2006); in a separate study, Boysen, Vogel, Madon,
and Wester (2006) found the instrument had an internal consistency of
0.96 with samples of undergraduate college students and 0.91 with sam-
ples of therapist trainees. The validity of the Heterosexism Scale has been
supported by its relationship to attitudes toward ethnic minorities, biases
against women, authoritarian attitudes, and higher levels of homophobia, as
well as its lack of relation to social desirability (Park, 2001). Within this
study, the researchers found a Cronbach’s alpha reliability coefficient of
0.85 and 0.89 for the superiority of heterosexuality and tolerance/acceptance
of non-heterosexuality subscales, respectively.
Cognitive Flexibility
The Cognitive Flexibility Scale (CFS; Martin & Rubin, 1995) is a scale with
12 six-point Likert-type items constructed to measure cognitive flexibility,
defined by Martin and Rubin (1995) as a person’s (a) awareness that in any
given situation there are options and alternatives available, (b) willingness to
be flexible and adapt to the situation, and (c) self-efficacy in being flexible.
In two separate studies, scores on the CFS were positively related to com-
munication flexibility and negatively related to rigidity, as well as positively
correlated with interaction involvement and self-monitoring while negatively
correlated with participants’ unwillingness to communicate. No gender differ-
ences were observed. A 1-week test-retest correlation of 0.83 was found for
the two administrations of the instrument. Within this study, the Cronbach’s
alpha reliability coefficient of this instrument was 0.84.
RELIGIOUS COMMITMENT
The Religious Commitment Inventory (RCI-10; Worthington et al., 2003) con-
sists of 10 items with a 5-point Likert-type scale. Six of the items measure
intrapersonal religious commitment, or how one’s beliefs underpin his or
her approach to life, and four items measure interpersonal religious commit-
ment, or how much one relates his or her religious beliefs to other persons.
Worthington et al. (2003) found the RCI-10 to be a parsimonious measure
with reliability coefficients ranging from 0.88 to 0.98 for the entire scale,
0.92 to 0.94 for the intrapersonal subscale, and 0.83 for the interpersonal
subscale. Test-retest reliability results for administration three weeks apart
were reported at 0.87 for the entire scale, 0.86 for the intrapersonal subscale,
and 0.83 for the interpersonal subscale. At five months, the test-retest reli-
ability was 0.84 (Worthington et al., 2003). The RCI-10 was positively and
significantly correlated with single-item measures of religiosity, frequency
of attendance at religious activities, and self-rated spiritual intensity. Self-
identified religious participants taking this instrument scored significantly
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1384 K. M. Goodrich et al.
higher than participants identified as nonreligious, and participants who
endorsed salvation as one of the top five values on the Rokeach’s Value
Survey (Rokeach, 1967) had significantly higher scores on the RCI-10 than
those who did not (Worthington et al., 2003). The reliability coefficient for the
current study was 0.96; only Cronbach’s alpha for the total score is reported
because that is the score most commonly used in research with this scale.
GENERAL FUNCTIONING SUBSCALE OF THE FAMILY ASSESSMENT DEVICE
The General Functioning Subscale (GFS) of the Family Assessment Device
(FAD; Epstein et al., 1983) is a 12-item scale based on 4-point Likert-type
items pulled from the larger McMaster FAD. Numerous studies (e.g., Byles,
Byrne, Boyle, & Offord, 1988; Herring et al., 2006; Sawyer et al., 2001)have
shown that the GFS is a parsimonious measure that captures the general
functioning of families when compared to the full FAD and other scales.
The GFS has a reported alpha coefficient of 0.92 (Epstein et al., 1983), with
follow-up studies showing a test-retest reliability of 0.71 (Miller, Epstein,
Bishop, & Keitner, 1985). Within the present study, the researchers found
the reliability coefficient of this instrument to be 0.91. Descriptive statistics
for all measures are reported in Table 1.
RESULTS
A confirmatory factor analysis was conducted for the 17 items of the
Heterosexism Scale using Mplus 5.0 (Muthen & Muthen, 1998–2009). Given
the Likert scale of the items, the items were analyzed as ordered categorical
variables and the robust weighted least squares estimator was utilized. The
two-factor solution was tested with Superiority of Heterosexuality as the first
factor and Tolerance/Acceptance of Nonheterosexuality as the second. All
the factor loadings were statistically significant in this model; however, the
TABLE 1 Descriptive Statistics for Measures
Measure n M SD
HS: Superiority of Heterosexuality 441 12.66 6.45
HS: Tolerance/Acceptance of Non-heterosexuality 441 12.57 7.32
HS: Heterosexism 441 18.65 9.75
RCI: Religious Commitment 441 27.18 12.92
FAD: General Functioning 438 41.19 5.93
CFS: Cognitive Flexibility 441 59.74 6.92
Note. Original scoring reported here for the Heterosexism Scale and subscales; HS =Heterosexism Scale;
RCI =Religious Commitment Inventory; FAD =Family Assessment Device; CFS =Cognitive Flexibility
Scale.
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1385
model fit was not acceptable (χ2(35∗)=213.162; CFI =.88; RMSEA =.11).
[Note: ∗Mplus uses a special formula for degrees of freedom when using the
weighted least squares estimator.] Given the poor fit and the fact that the
measure was being used with a population for which it was not originally
normed, exploratory factor analysis was used to determine whether the fac-
tor structure for the instrument was similar to the factor structure proposed
by Park (2001).
Next, SPSS version 19 was used to conduct an exploratory factor anal-
ysis using Principal Axis Factoring. An oblique Promax rotation was used
because the factors of the scale were expected to be correlated. To identify
the appropriate number of factors, a parallel analysis (Horn, 1965)wascon-
ducted along with an inspection of the scree plot and including a careful
examination of the meaning of the factors from the exploratory factor anal-
ysis. A parallel analysis compares the eigenvalues from the analysis of the
observed items to the eigenvalues from an analysis of randomly generated
uncorrelated items. The eigenvalues from the parallel analysis of the uncor-
related items are used as a comparison showing expected values when no
factor structure is present. The logic of the parallel analysis is that the ana-
lyst should retain only those factors for the observed data with eigenvalues
larger than those for the corresponding factors in the randomly generated
data. When using the scree plot to identify the number of factors to extract,
the goal is to choose the number of factors that fall before the last large
decrease in eigenvalues.
Figure 1 shows the eigenvalues from the factor analysis and the parallel
analysis. Based on the scree plot and the results from the parallel analy-
sis, a two-factor solution was indicated. Table 2 lists the factor loadings for
0
2
4
6
8
10
01234567891011121314151617
Eigenvalues
Number of Factors
FIGURE 1 Eigenvalues from factor analysis and parallel analysis.
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1386 K. M. Goodrich et al.
TABLE 2 Factor Loadings for the Two-Factor Solution
Item Factor 1 Factor 2
HS1 −0.11 0.88
HS2 −0.03 0.77
HS3 −0.04 0.61
HS4 0.01 0.63
HS5 0.42 0.19
HS6 0.25 0.42
HS7 0.82 0.05
HS8 0.46 0.35
HS9 0.67 0.13
HS10 0.84 0.02
HS11 0.88 0.00
HS12 0.96 −0.15
HS13 0.78 −0.14
HS14 0.40 −0.06
HS15 0.83 0.00
HS16 0.53 0.15
HS17 0.57 0.19
Note. Bold factor loadings denote retained items.
the two-factor solution. Factor loadings of 0.3 or above were considered
evidence that an item loaded on a particular factor. To avoid the poten-
tial confusion of having items loading on both factors, items were retained
only if the loading was 0.3 or greater and the difference between the factor
loadings on the two factors was at least 0.15.
Based on these criteria, 16 of the 17 items were retained. Specifically,
item 8 was not retained because the difference between the loadings for the
first and second factors was not greater than 0.15. Of the 16 retained items, all
but two loaded on a factor consistent with Park’s original two factors. Items
5, 7, and 9–15 loaded on the first factor, corresponding to Park’s Tolerance
factor, and items 1–4 and 6 loaded on the second factor, corresponding to
Park’s Superiority factor. Contrary to Park’s solution, items 16 and 17 loaded
on the factor corresponding to Tolerance instead of the factor corresponding
to Superiority. The two factors showed a correlation of 0.65.
To examine how the two factors from the exploratory factor analysis
related to other measures, correlations were examined between the two
subscales of the Heterosexism Scale and three other measures: religious
commitment as measured by the RCI-10, general family functioning as mea-
sured by the GFS of the FAD, and cognitive flexibility as measured by the
CFS. Subscale scores were computed by taking the sum of the retained
items for the respective factors. Results were expected to show that more
tolerant attitudes toward LGB persons would be negatively correlated with
Religious Commitment and positively correlated with Family Functioning and
Cognitive Flexibility. Superiority (r=0.09, p=0.05), but not Tolerance (r
=0.06, p=0.19), was significantly related to religious commitment. Both
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1387
Superiority (r=−0.26, p<0.001) and Tolerance (r=−0.16, p=0.001) were
negatively related to general family functioning. Superiority was negatively
and significantly related to cognitive flexibility (r=−0.14, p=0.002), but
Tolerance was not (r=−0.06, p=0.21).
DISCUSSION
The purpose of this study was to find whether Park’s two-factor solution for
the Heterosexism Scale (2001) fit the data from a sample of primarily middle-
aged heterosexual parents of LGB sons or daughters. Additionally, the
researchers explored the relations between the subscales of the Heterosexism
Scale and measures of cognitive flexibility (Martin & Rubin, 1995), reli-
gious commitment (Worthington et al., 2003), and general family functioning
(Epstein et al., 1983). In the original confirmatory factor analysis using Park’s
(2001) subscales, the two-factor solution for the Heterosexism Scale pro-
vided an unacceptable fit with this sample. Instead, a two-factor solution
with one item omitted and the transfer of two items from the Superiority
to the Tolerance subscales was supported in the exploratory factor analysis.
As such, the overall functioning of the Heterosexism Scale was supported,
but the researchers recommend pilot testing the Heterosexism Scale with a
sample prior to using it with populations for whom the instrument has not
yet been validated.
The authors found that higher scores on both Heterosexism subscales
were associated with general family functioning. Additionally, higher scores
on the Superiority subscale corresponded to lower scores of cognitive flex-
ibility and higher scores on religious commitment. Not all the anticipated
relationships were found: the Tolerance subscale was not found to be signif-
icantly related to religious commitment or cognitive flexibility. This provided
mixed support for the researchers’ hypothesis that heterosexism would be
associated with religious commitment or cognitive flexibility. It does appear
that when heterosexuals have a superior attitude when interacting with
LGB persons, they tend to be more religious and less cognitively flex-
ibile, as hypothesized by the researchers. When more tolerant (although
not accepting) attitudes are found with heterosexuals, the associations with
religion and cognitive flexibility become less clear. As such, teasing out
the differences between tolerance and superiority would be a good next
step for future research. Although the findings of this study support the
potential for using the Heterosexism Scale with populations of individuals
other than college students, unacceptable fit with Park’s (2001)’s original
two-factor solution suggests other factor solutions may be more accept-
able with different populations. Additionally, the limited association with
other scales also raises questions about the scale’s applicability across all
contexts.
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1388 K. M. Goodrich et al.
Limitations
The first limitation of the study concerns the sample: the respondents were
predominately White, well-educated, middle-aged, middle to upper-middle
class, biological mothers of LGB sons or daughters who lived in the sub-
urbs and identified as liberal (for more information about the sample, see
Goodrich & Gilbride, 2010). Additionally, not all parents may have been
independent, because the sample could possibly have included two parents
of the same LGB son or daughter. As such, the parents found in this sam-
ple may not adequately represent all heterosexual parents of LGB sons or
daughters, such as families of color, of lower socio-economic status, with
less education, or politically identified as moderate or conservative. Differing
assumptions and values within distinct communities, such as the “down low”
concept found within the African American community (Boykin, 2004), might
make disclosure to one’s family by LGB children or openness by parents to
discuss their sons’ or daughters’ LGB identity more challenging. As such,
few studies have been able to explore the “outness” or disclosure of per-
sons in diverse racial/ethnic families (Hom, 1994; Merighi & Grimes, 2000).
One should use caution, therefore, before attempting to generalize to other
groups in the larger population.
The measures used to explore the relations between the Heterosexism
Scale and other constructs were related but indirect measures associated with
the construct. Although some results were found to be statistically signifi-
cant, the coefficients were not large. Finally, the fit indices did not support
the original two-factor solution for the Heterosexism Scale, suggesting a dif-
ferent two-factor solution with these data. Future researchers should test
this instrument with other populations to see if other factor solutions better
represent the data with those samples.
Clinical Implications
The results from this study support the use of the Heterosexism Scale with
heterosexual parents of LGB-identified sons or daughters. Professional coun-
selors could use this inventory when families come to counseling following a
disclosure by an LGB family member. Additionally, when strife occurs within
a family system, counselors might use the inventory as a means to under-
stand any unconscious biases or prejudices those heterosexual parents might
hold toward LGB individuals. The inventory could be used to start a dialogue
in the counseling room between family members to better understand where
each member of the family system is in relation to the son’s or daughter’s
identified sexual orientation.
Counselor educators and supervisors could use the Heterosexism Scale
when addressing multicultural and general counseling subject matter for stu-
dents and supervisees to identify, acknowledge, and understand unconscious
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1389
biases or prejudices they might hold toward the LGB population. The
inventory could be used as pre- and post-testing for students to track growth
or development over the course of a semester, as well as to measure how
prepared they might be for future counseling of LGB persons in practice
settings.
Implications for Future Research
As noted above, the findings of the study demonstrated that a two-factor
solution was supported for a sample of heterosexual parents of LGB-
identified sons or daughters, although it was a different two-factor solution
then Park (2001) originally proposed. As such, future researchers should
replicate this study with other samples of heterosexual parents of LGB sons
or daughters, as well as additional populations, to better understand the
Heterosexism Scale and whether a two-factor solution is the most appropri-
ate with those samples. Specifically, researchers might explore the use of
the Heterosexism Scale with other groups of heterosexual persons, includ-
ing heterosexual parents of color, fathers, and groups not typically found
within the LGB specific literature. Additionally, as an interesting differenti-
ation on associated scales were found for participants in this study, future
research could also explore how scores from the Heterosexism Scale relate
to those from other measures of the construct of heterosexism, including the
differences between persons who score high on the Superiority versus the
Tolerance scales.
Summary
Overall, it appears that the Heterosexism Scale is an appropriate instrument
to use with heterosexual parents of LGB sons or daughters. The results of
this study indicate that a two-factor solution did generalize from a sample of
college students to a sample of middle-aged parents. Additionally, the instru-
ment’s relationship with some related constructs provides some support for
the utility of the scale. This adds to the ability of clinicians and researchers to
utilize this subtler instrument while addressing issues related to heterosexism
with a broader sample of the population.
REFERENCES
Barnard, A. (2009). Lesbians’ constructions of depression. Health Care for Women
International,30, 373–389. doi:10.1080/07399330902785141
Boykin, K. (2004). Beyond the down low: Sex, lies, and denial in Black America.
New York, NY: Carroll & Graf.
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1390 K. M. Goodrich et al.
Boysen, G. A., Vogel, D. L., & Madon, S. (2006). A public versus private administra-
tion of the implicit association test. European Journal of Social Psychology,36,
845–856. doi:10.1002/ejsp.318
Boysen, G. A., Vogel, D. L., Madon, S., & Wester, S. R. (2006). Mental health
stereotypes about gay men. Sex Roles,54,69–82.doi:10.1007/s11199-006-8870-0
Byles, J., Byrne, C., Boyle, M. H., & Offord, D. R. (1988). Ontario child health study:
Reliability and validity of the general functioning subscale of the McMaster
Family Assessment Device. Family Process,27 ,97–104.doi:10.1111/j.1545-
5300.1988.00097.x
Consortium of Higher Education LGBT Resource Professionals (2011). Directory.
Retrieved from http://www.lgbtcampus.org/directory/
Daly, J. (1990). Measuring attitudes toward lesbians and gay men: Development
and initial psychometric evaluation of an instrument (Unpublished doctoral
dissertation). Southern Illinois University at Carbondale.
Dermer, S. B., Smith, S. D., & Barto, K. K. (2010). Identifying and correctly label-
ing sexual prejudice, discrimination, and oppression. Journal of Counseling &
Development,88, 325–331.
Ellis, S. J., Kitzinger, C., & Wilkinson, S. (2002). Attitudes towards lesbians and gay
men and support for lesbian and gay human rights among psychology students.
Journal of Homosexuality,44,121–138.doi:10.1300/J082v44n01_07
Epstein, N. B., Baldwin, L. M., & Bishop, D. S. (1983). The McMaster Family
Assessment Device. Journal of Marital and Family Therapy,9, 171–180.
doi:10.1111/j.1752-0606.1983.tb01497.x
Exodus International. (2011). Exodus ministry listings. Retrieved from http://
exodusinternational.org/find-help/exodus-ministry-listings/
Focus on the Family. (2011). Contact us. Retrieved from http://family.custhelp.com/
cgi-bin/family.cfg/php/enduser/stdadp.php?p faqid=14190
Funders for Lesbian and Gay Issues. (2011). List all funders. Retrieved from http://
www.lgbtfunders.org/seekers/search.cfm
Gallor, S. M. (2006). Heterosexual parents’ gender role attitudes, religious orientation,
heterosexist beliefs, support group experiences, and relationship functioning with
their lesbian and gay children (Unpublished doctoral dissertation). University of
Maryland, College Park.
Garcia, D., Gray-Stanley, J., & Ramirez-Valles, J. (2008). “The priest obviously doesn’t
know that I’m gay”: The religious and spiritual journeys of Latino gay men.
Journal of Homosexuality,55, 411–436. doi:10.1080/00918360802345149
Goodrich, K. M. (2009). Mom and dad come out: The process of identifying as
a heterosexual parent with a lesbian, gay, or bisexual child, Journal of LGBT
Issues in Counseling,3,37–61.doi:10.1080/15538600902754478
Goodrich, K. M., & Gilbride, D. D. (2010). The refinement and validation of a model
of family functioning after child’s disclosure as lesbian, gay, or bisexual. Journal
of LGBT Issues in Counseling,4(2), 92–121. doi:10.1080/15538695.2010.483575
Goodrich, K. M., & Luke, M. (2009). LGBTQ responsive school counseling. Journal
of LGBT Issues in Counseling,3, 113–127. doi:10.1080/15538600903005284
Goodrich, K. M., & Luke, M. (2010). Experiences of school counselors-in-training
in group work with LGBTQ adolescents. Journal for Specialists in Group Work,
35,143–159.doi:10.1080/01933921003705966
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
Heterosexism Scale 1391
Goodwill, K. A. (2000). Religion and the spiritual needs of gay Mormon men. Journal
of Gay and Lesbian Social Services,11(4), 23–37. doi:10.1300/J041v11n04_02
Hansen, G. L. (1982). Measuring prejudice against homosexuality (homosexism)
among college students: A new scale. Journal of Social Psychology,117 ,
233–236. doi:10.1080/00224545.1982.9713432
Herek, G. M. (1990). The context of anti-gay violence: Notes on cultural and
psychological heterosexism. Journal of Interpersonal Violence,5, 316–333.
doi:10.1177/088626090005003006
Herek, G. M. (1994). Assessing heterosexuals’ attitudes toward lesbians and gay men:
AreviewofempiricalresearchwiththeATLGscale.InB.Greene&G.M.Herek
(Eds.), Lesbian and gay psychology: Theory, research, and clinical application
(pp. 206–228). Thousand Oaks, CA: Sage.
Herring, S., Gray, K., Taffe, J., Tonge, B., Sweeney, D., Einfeld, S. (2006). Behaviour
and emotional problems in toddlers with pervasive developmental disor-
ders and developmental delay: Associations with parental mental health and
family functioning. Journal of Intellectual Disability Research,50, 874–882.
doi:10.1111/j.1365-2788.2006.00904.x
Hom, A. Y. (1994). Stories from the homefront: Perspectives of Asian American
parents with lesbian daughters and gay sons. Amerasia Journal,20(1), 19–32.
Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis.
Psychometrika,30, 179–185. doi:10.1007/BF02289447
Kazdin, A. E. (2002). Research design in clinical psychology (4th ed.). Upper Saddle
River, NJ: Pearson.
Lambda Legal (2011). Issues. Retrieved from http://www.lambdalegal.org/our-work/
issues/
Larsen, K. S., Reed, M., & Hoffman S. (1980). Attitudes of heterosexuals toward
homosexuality: A Likert-type scale and construct validity. Journal of Sex
Research,16,245–257.doi:10.1080/00224498009551081
Love, P., Bock, M., Jannarone, A., & Richardson, P. (2005). Identity interaction:
Exploring the spiritual experiences of lesbian and gay college students. Journal
of College Student Development,46 ,193–209.doi:10.1353/csd.2005.0019
Martin, M. M., & Rubin, R. B. (1995). A new measure of cognitive flexibility.
Psychological Reports,76, 623–626.
McConahay, J. B. (1986). Modern racism, ambivalence, and the Modern Racism Scale.
In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism
(pp. 91–125) Orlando, FL: Academic Press.
Merighi, J. R., & Grimes, M. D. (2000). Coming out to families in a multicultural con-
text. Families in Society: The Journal of Contemporary Human Services,81(1),
32–41.
Miller, I. W., Epstein, N. B., Bishop, D. S., & Keitner, G. I. (1985). The McMaster
Family Assessment Device: Reliability and validity. Journal of Marital and
Family Therapy,11,345–356. doi:10.1111/j.1752-0606.1985.tb00028.x
Morrison, M. A., & Morrison, T. G. (2002). Development and validation of a scale
measuring modern prejudice toward gay men and lesbian women. Journal of
Homosexuality,43(2), 15–37. doi:10.1300/J082v43n02_02
Muthen, L. K., & Muthen, B. O. (1998–2009). Mplus user’s guide (5th ed.). Los
Angeles, CA: Muthen & Muthen.
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
1392 K. M. Goodrich et al.
Park, J. (2001). Development of the heterosexism scale (Unpublished doctoral
dissertation). The Pennsylvania State University, University Park.
Peel, E. (2001). Mundane heterosexism: Understanding incidents of the every-
day. Women’s Studies International Forum,24,541–554.doi:10.1016/S0277-
5395(01)00194-7
Parents, Families, and Friends of Lesbians and Gays. (2011). Find a chapter.
Retrieved from http://community.pflag.org/Page.aspx?pid=803
Rokeach, M. (1967). Value survey. Sunnyvale, CA: Halgren Tests.
Savin-Williams, R. C., & Dube, E. M. (1998). Parental reactions to their child’s
disclosure of a gay/lesbian identity. Family Relations,47(1), 7–13.
Sawyer, M. G., Spurrier, N., Whaites, L., Kenney, D., Martin, A. J., & Baghurst,
P. (2001). The relationship between asthma severity, family functioning and
health related quality of life of children with asthma. Quality of Life Research,
9,1005–1115.doi:10.1023/A:1016655511879
Shelton, K., & Delgado-Romero, E. A. (2011). Sexual orientation microaggressions:
The experience of lesbian, gay, bisexual, and queer clients in psychotherapy.
Journal of Counseling Psychology,58, 210–221. doi:10.1037/a0022251
Sherry, A., Adelman, A., Whilde, M. R., & Quick, D. (2010). Competing selves:
Negotiating the intersection of spiritual and sexual identities. Professional
Psychology: Research and Practice,41,112–119.doi:10.1037/a0017471
Sister Friends Together. (2011). Find a church.Retrievedfromhttp://www.
sisterfriends-together.org/find-a-church/
Sue, D. W., Capodilupo, C. M., Torino, G. C., Bucceri, J. M., Holder, A. M. B.,
Nadal, K. L., & Esquilin, M. (2007). Racial microaggressions in everyday
life: Implications for clinical practice. American Psychologist,62(4), 271–286.
doi:10.1037/0003-066X.62.4.271
Swim, J. K., Aikin, K. J., Hall, W. S., & Hunter, B. A. (1995). Sexism and racism: Old-
fashioned and modern prejudices. Journal of Personality and Social Psychology,
68,199–214.doi:10.1037/0022-3514.68.2.199
theGLBT.com. (2011). GLBT communities of color.Retrievedfromhttp://www.
TheGLBT.com/S_ShowSupport.cfm
Worthington, R. L., Dillon, F. R., & Becker-Schutte, A. M. (2005). Development,
reliability, and validity of the lesbian, gay, and bisexual knowledge and atti-
tudes scale for heterosexuals (LGB-KASH). Journal of Counseling Psychology,
52,104–118.doi:10.1037/0022-0167.52.1.104
Worthington, E. L., Jr., Wade, N. G., Hight, T. L., Ripley, J. S., McCullough, M. E.,
Berry, J. W., Schmidt, M. M., Berry, J. T., Bursley, K. H., & O’Connor, L. (2003).
The religious commitment inventory—10: Development, refinement, and val-
idation of a brief scale for research and counseling. Journal of Counseling
Psychology,50,84–96.doi:10.1037/0022-0167.50.1.84
Downloaded by [University of New Mexico], [Kristopher M. Goodrich] at 11:33 18 August 2014
