<%BANNER%>

The Influence of Ego Identity on Heterosexual Individuals' Attitudes toward Lesbian and Gay Individuals

Permanent Link: http://ufdc.ufl.edu/UFE0018020/00001

Material Information

Title: The Influence of Ego Identity on Heterosexual Individuals' Attitudes toward Lesbian and Gay Individuals
Physical Description: 1 online resource (105 p.)
Language: english
Creator: Potoczniak, Daniel Jos
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: attitudes, authoritarianism, ego, gay, identity, lesbian
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of this study was to examine factors related to heterosexual individuals' attitudes of tolerance versus condemnation toward lesbian and gay (LG) individuals. Factors examined include ego identity commitment, ego identity exploration, authoritarianism, social dominance orientation, gender role attitudes, eroticism, and the number/intimacy of contact experiences with LG people. Using Herek's theory of attitude function, it is posited that attitudes, including those toward LG individuals, may serve an ego defensive function. Theoretically, according to ego identity development theory, the greater degree to which an individual has explored life options and has committed to those which are a satisfactory match, the greater degree of maturity of one's ego/ego-defensive functioning. Along these lines, having a more mature and adaptive ego identity might enable an individual to be more tolerant toward others who are different from her or him, especially LG individuals. Based in data collected from an Internet-based sample, results indicate that higher levels of authoritarianism and social dominance orientation are related to lower levels of ego identity exploration; also, higher levels of authoritarianism are related to high degrees of commitment. Additionally, greater numbers and greater intimacy of LG contact experiences were associated with higher levels of ego identity exploration. In turn, higher levels of ego identity exploration were related to more tolerant attitudes toward LG people. When considering individuals who demonstrated higher levels of authoritarian traits, ego identity exploration statistically moderated these individuals? relatively more condemning attitudes toward LG people. Previous findings in the literature on attitudes toward LG people, many times based in a college-aged sample, were also replicated in the current study, which used an older, non-college- student sample. Such replicated findings include heterosexual males having more negative attitudes toward LG individuals than do heterosexual women; however, this gender difference was eliminated when controlling for the influence of gender-role-based attitudes. Also, statistically significant differences in attitudes toward LG people were found among different races/ethnicities, especially among African American- and Hispanic-identified participants.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Jos Potoczniak.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Neimeyer, Greg J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0018020:00001

Permanent Link: http://ufdc.ufl.edu/UFE0018020/00001

Material Information

Title: The Influence of Ego Identity on Heterosexual Individuals' Attitudes toward Lesbian and Gay Individuals
Physical Description: 1 online resource (105 p.)
Language: english
Creator: Potoczniak, Daniel Jos
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: attitudes, authoritarianism, ego, gay, identity, lesbian
Psychology -- Dissertations, Academic -- UF
Genre: Counseling Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of this study was to examine factors related to heterosexual individuals' attitudes of tolerance versus condemnation toward lesbian and gay (LG) individuals. Factors examined include ego identity commitment, ego identity exploration, authoritarianism, social dominance orientation, gender role attitudes, eroticism, and the number/intimacy of contact experiences with LG people. Using Herek's theory of attitude function, it is posited that attitudes, including those toward LG individuals, may serve an ego defensive function. Theoretically, according to ego identity development theory, the greater degree to which an individual has explored life options and has committed to those which are a satisfactory match, the greater degree of maturity of one's ego/ego-defensive functioning. Along these lines, having a more mature and adaptive ego identity might enable an individual to be more tolerant toward others who are different from her or him, especially LG individuals. Based in data collected from an Internet-based sample, results indicate that higher levels of authoritarianism and social dominance orientation are related to lower levels of ego identity exploration; also, higher levels of authoritarianism are related to high degrees of commitment. Additionally, greater numbers and greater intimacy of LG contact experiences were associated with higher levels of ego identity exploration. In turn, higher levels of ego identity exploration were related to more tolerant attitudes toward LG people. When considering individuals who demonstrated higher levels of authoritarian traits, ego identity exploration statistically moderated these individuals? relatively more condemning attitudes toward LG people. Previous findings in the literature on attitudes toward LG people, many times based in a college-aged sample, were also replicated in the current study, which used an older, non-college- student sample. Such replicated findings include heterosexual males having more negative attitudes toward LG individuals than do heterosexual women; however, this gender difference was eliminated when controlling for the influence of gender-role-based attitudes. Also, statistically significant differences in attitudes toward LG people were found among different races/ethnicities, especially among African American- and Hispanic-identified participants.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Daniel Jos Potoczniak.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Neimeyer, Greg J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0018020:00001


This item has the following downloads:


Full Text





THE INFLUENCE OF EGO IDENTITY ON HETEROSEXUAL INDIVIDUALS' ATTITUDES
TOWARD LESBIAN AND GAY INDIVIDUALS























By

DANIEL JOSEPH POTOCZNIAK


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2007
































Copyright 2007 Daniel Joseph Potoczniak
































To my parents, for being so good at being parents;
To Rand, for being my everything, always;
To my brother, for being an example and a guide;
To my sister, for reminding me that it's good to be just who you are.

Thank you so much for helping me to get to where I am.









ACKNOWLEDGMENTS

I thank my supervisory committee chair, Greg Neimeyer, Ph.D., for his insight and support in

navigating this project. I am absolutely certain that it would not have gone so smoothly with anyone

else. I also thank M. Jeff Farrar, Ph.D., Mary Fukuyama, Ph.D., and Peter Sherrard, Ed.D., for all of

their help and input in making this study as interesting as it has been. I sincerely appreciate all of the

above individuals agreeing to work with me in the first place, and appreciate their helpful guidance.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. ...............4.....


LI ST OF T ABLE S ............ ..... .__ ...............7...


LI ST OF FIGURE S .............. ...............8.....


AB S TRAC T ......_ ................. ............_........9


CHAPTER


1 INTRODUCTION ................. ...............11.......... ......


2 LITERATURE REVIEW ................. ...............17................


The Motivational and Functional Approach to Understanding Attitudes .............. ................19
Experiential Attitudes ................. ...............20........._.....
Defensive Attitudes .............. ...............21....

Symbolic Attitudes ................. .. ........_. ........ ..._... ........ ........2
Research on Attitudes toward Lesbian and Gay (LG) Individuals..........._.._. ........._.._. ...25
Previous Contact with LG Individuals ................... ...... ........ ..........2
Gender-Based Differences in Attitudes toward LG Individuals .................. ...............29

Symbolic attitudes toward gender and gender roles............... ...............29.
African American-based gender differences in LG attitudes............... ................3
Religion and Religious Variables ............._. ...._.. ...._... ...........3
Personality-Based Factors .................. ...... ... ..........3
Defensive Attitudes with a Possible Basis in Gender .............. ...............39....

Purpose of the Study ............._. ...._... ...............42...

3 M ETHOD .............. ...............44....


Participants .............. ...............44....
Design and Procedure ............._. ...._... ...............46....
Instrum ents ............... .. ... .. ... .. ................. ...............4
Attitudes Toward Lesbians and Gay Men Scale, Short Form (ATLG-S) .......................47
Right-Wing Authoritarianism Scale, Short Version (RWA-S) ................... ...............48
Social Dominance Orientation Scale, Abbreviated (SDO-A) .............. ....................4
Attitudes toward Women (ATW) Scale .............. ...............49....
Ego Identity Process Questionnaire (EIPQ) ................ ...............50........... ...
Social Desirability .............. ...............51....
Eroti ci sm ................ ...............51........... ....

Analytic Strategies............... ...............5

4 RE SULT S .............. ...............54....











Order of Presentation Effects and Preliminary Analyses ................. ................. ....__54
Analyses............... ...............5
Hypothesis Tests .............__. ...... ._ ...............56....
Additional Research Questions .............. ...............59....

5 DIS CU SSION .............__....... ._ ...............71.


Hypotheses............... ...............7
Research Questions...................... ...............7
Limitations and Suggestions for Further Research ...........__ ........ ._ ......... ........7
Implications for Psychotherapy and Counseling .............__....... ........ ..........7

APPENDIX


A DEMOGRAPHIC QUESTIONNAIRE............... .............8

B THE ATTITUDES TOWARD LESBIANS AND GAY MEN SCALE .............. ..............84


C THE RIGHT-WING AUTHORITARIANISM SCALE, SHORT FORM ................... ..........868888

D THE SOCIAL DOMINANCE ORIENTATION SCALE, ABBREVIATED VERSION.....88

E THE ATTITUDES TOWARD WOMEN SCALE .....__.....___ ..............._........8


F EGO IDENTITY PROCES S QUESTIONNAIRE .....__.....___ ..............._........9

G MEASUREMENT OF EROTIC VALUE OF LESBIAN AND GAY SEX ........................93


H MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE, SHORT FORM ..................94

I EXAMPLE ADVERTISEMENT FOR PARTICIPANT RECRUITMENT ........................95

J INFORMED CONSENT ................. ...............96................

LIST OF REFERENCE S ................. ...............98................

BIOGRAPHICAL SKETCH ................ ............. ............ 105...










LIST OF TABLES


Table page

4-1 Means, Standard Deviations, and Scale/Subscale Correlations ................. ................ ...63

4-2 Summary of Multiple Regression Analy sis for the Prediction of EIPQ Commitment
and Exploration Scores by RWA and SDO .............. ...............64....

4-3 Summary of Multiple Regression Analysis for the Prediction of EIPQ Exploration
Scores by Number of Known LG Individuals and the level of Closeness in those
Rel ati on shi p s ................ ...............65...............

4-4 Summary of Multiple Regression Analysis for the Prediction of ATLG-S Scores by
EIPQ Exploration scores............... ...............66.

4-5 Moderating Effect of Ego Identity Exploration on the Relation between
Authoritarianism and Attitudes toward Lesbians and Gay Men ................. ................ ..67

4-6 Direct Effects Regression Analyses of Predictor Variables on Criterion Variables
(Gender, ATW Scores, and ATLG-S Scores)............... ...............68

4-7 Regression Analyses of Attitudes Toward Women Mediating the Effects of Gender
on ATLG-S scores .............. ...............69....









LIST OF FIGURES


Figure page

4-1 Plot of significant Authoritarianism X Ego Identity Exploration interaction, with
plotted values for scores on the Attitudes toward Lesbian and Gays (ATLG-S) ........._....70









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy.

THE INFLUENCE OF EGO IDENTITY ON HETEROSEXUAL INDIVIDUALS'
ATTITUDES TOWARD LESBIAN AND GAY INDIVIDUALS

By

Daniel Joseph Potoczniak

August 2007

Chair: Greg Neimeyer
Major: Counseling Psychology

The purpose of this study was to examine factors related to heterosexual individuals'

attitudes of tolerance versus condemnation toward lesbian and gay (LG) individuals. Factors

examined include ego identity commitment, ego identity exploration, authoritarianism, social

dominance orientation, gender role attitudes, eroticism, and the number/intimacy of contact

experiences with LG people. Using Herek' s theory of attitude function, it is posited that

attitudes, including those toward LG individuals, may serve an ego defensive function.

Theoretically, according to ego identity development theory, the greater degree to which an

individual has explored life options and has committed to those which are a satisfactory match,

the greater degree of maturity of one' s ego/ego-defensive functioning. Along these lines, having

a more mature and adaptive ego identity might enable an individual to be more tolerant toward

others who are different from her or him, especially LG individuals.

Based in data collected from an Internet-based sample, results indicate that higher levels

of authoritarianism and social dominance orientation are related to lower levels of ego identity

exploration; also, higher levels of authoritarianism are related to high degrees of commitment.

Additionally, greater numbers and greater intimacy of LG contact experiences were associated









with higher levels of ego identity exploration. In turn, higher levels of ego identity exploration

were related to more tolerant attitudes toward LG people. When considering individuals who

demonstrated higher levels of authoritarian traits, ego identity exploration statistically moderated

these individuals' relatively more condemning attitudes toward LG people.

Previous findings in the literature on attitudes toward LG people, many times based in a

college-aged sample, were also replicated in the current study, which used an older, non-college-

student sample. Such replicated findings include heterosexual males having more negative

attitudes toward LG individuals than do heterosexual women; however, this gender difference

was eliminated when controlling for the influence of gender-role-based attitudes. Also,

statistically significant differences in attitudes toward LG people were found among different

races/ethnicities, especially among African American- and Hispanic-identified participants.









CHAPTER 1
INTTRODUCTION

In 1998, the murders of Matthew Shepard and Billy Jack Gaither shocked thousands of

Americans (Herek, 2000). Shepard was a 21-year-old college student in Wyoming, and Gaither

was a 39-year-old factory worker in Alabama. The two had apparently little in common, save

the fact that both of them were targeted for attack because they were gay men. If the two

murders had not been grounded in anti-gay sentiment, they might not have been so noteworthy as

to garner the tremendous amount of attention they did in the mass media. However, due to the

hate-based nature of the crimes, they drew attention to the danger and violence faced by lesbian

and gay (LG) individuals on a regular basis. In early 2004, the National Coalition of Anti-

Violence Program's report on violent acts based in prejudice stated that more than 2000 violent

incidents over the previous year occurred because of the victim' s sexual orientation. Since a

great number of these types of crimes may never be reported to the police due to the stigmatized

status of the victim, there may be many more such crimes that are never reported (Herek, Gillis,

& Cogan, 1999).

Stigmatization of LG individuals is not a new phenomenon (Chauncey, 1994), existing

well into the previous century, and likely longer before that. With the rise of the civil rights

movement in the 1960s, however, LG individuals' condemnation was increasingly scrutinized.

Eventually, the American Psychiatric Association eventually declared that homosexuality was

not a mental illness (American Psychiatric Association, 1980), and the question of what led some

individuals to maintain negative perceptions of LG people began to receive closer attention.

Perhaps due to this attention and to LG individuals' increasing presence in visible society,

negative attitudes toward lesbians and gay men (hereafter referred to as "negative LG attitudes"

or "sexual prejudice") have decreased over the past 30 years (Loftus, 2001): whereas at least









two-thirds of respondents to the General Social Survey (GSS) considered same-sex sexual

behavior "always wrong" in the 1970s and 1980s, that figure declined significantly in the 1990s;

by 1996, only 56% of GSS respondents regarded such behavior as always wrong (Yang, 1997).

In addition to sexual behavior between two women or two men, much of the public also

holds negative attitudes toward LG individuals themselves (Herek, 2000). In a study conducted

using a random national sample, more than half of the heterosexual participants expressed

disgust for LG individuals (Herek, 1994). Respondents to the American National Election

Studies typically rate lesbians and gay men among the lowest of all social groups on a "feeling

thermometer," in which on a scale of 1 to 101, individuals are asked to rate how they feel about a

particular group, issue, or other factor (Yang, 1997). To understand this sexual prejudice (which

has sometimes been labeled "homophobia" or "heterosexism"), a growing number of studies

have explored various factors that appear to contribute to negative LG attitudes.

Findings from public opinion surveys have revealed higher levels of sexual prejudice

among individuals who are older, less educated, living in the U.S. South or Midwest, and living

in rural areas (Herek, 1994). Also, heterosexual men generally have more negative attitudes

toward LG individuals, especially gay men, than do heterosexual women (Kite & Whitley,

1996). One possible explanation for gender-based differences in LG attitudes rests in a gender

role attitudes/beliefs analysis. This point of view states that heterosexuals' evaluations of gay

men and lesbians are grounded in a much larger belief system concerning men and women and

their appropriate roles in society (Kite, 1994). Gender roles for men, a class of symbolic beliefs

that may be grounded in family-inspired values or in religious or political ideology, appear

somewhat more narrowly defined and inflexible than are those for women (Bem, 1993). In this

light, a violation of male gender roles is usually considered more inappropriate than violation of









female roles (Herek, 1988), and an individual's violation of those roles results in a challenge

toward concepts of gender appropriateness and a particular individual's acceptability. An

alternate explanation of heterosexual men's apparently greater sexual prejudice lies in the fact

that these men may attribute greater erotic value to sex between two women than to sex between

two men, thus rendering a more positive opinion of lesbians (Louderback & Whitley, 1997). In

contrast, heterosexual women appear to attribute low erotic value to both forms of sexual

behavior (Louderback & Whitley, 1997).

Apart from gender, religious variables are theoretically and empirically supported in

predicting attitudes toward LG individuals (Schwartz & Lindley, 2005). Variables such as

Allport and Ross's (1957) classification of intrinsic and extrinsic religious orientations, and

fundamentalism have been explored. Within the context of religion, a fundamentalist

understanding of religion (i.e., a literal interpretation of religious scripture) is one of the core

variables that predict more negative LG attitudes. Interestingly, however, when one includes a

consideration of authoritarianism, which is a personality variable extensively studied by

Altemeyer (1981, 1996), the significance of fundamentalism is largely eliminated (Laythe,

Finkel, & Kirkpatrick, 2001).

Altemeyer' s (1981, 1996) theoretical conceptualization of authoritarianism is

characterized by high degrees of submission to authorities, aggressiveness toward outgroups, and

adherence to conventions perceived to be endorsed by society and its authorities, suggesting that

people with high degrees of authoritarianism would be likely to exhibit anti-LG attitudes. More

specifically, these attitudes would be likely since LG persons are explicitly condemned by many

religious and political leaders, making LG persons an acceptable target for prejudice and

hostility. Additionally, lesbians and gay men represent a socially stigmatized out-group relative









to heterosexual women and men, and are perceived as advocates for political change in social

conventions. A number of studies have found significant positive links between authoritarianism

and anti-LG attitudes (e.g., Altemeyer, 1981, 2001; Ekehammar, Akrami, Gylje, & Zakrisson,

2004; Laythe, Finkel, & Kirkpatrick, 2001) indicating that persons with high levels of

authoritarianism express more anti-LG attitudes than do persons with low levels of

authoritari ani sm.

In addition to authoritarianism, social dominance orientation (SDO) also has been linked

consistently with negative attitudes toward lesbians and gay men. Pratto, Sidanius, Stallworth,

and Malle (1994) defined SDO as the "extent to which one desires that one' s in-group dominate

and be superior to out-groups" (p 724). SDO taps a general attitudinal orientation toward

intergroup relations that reflects a preference for hierarchical rather than egalitarian relations.

More specifically, the desire to maintain the superior position of their in-groups relative to

outgroups motivates people high in SDO to accept hierarchy-legitimizing myths that denigrate

members of out-groups and enforce the status quo of their in-groups' power position. Thus,

persons high in SDO tend to have negative LG attitudes because LG persons have lower social

status than do heterosexuals (Whitley & Lee, 2000). The empirical literature supports the

expected link between SDO and anti-LG attitudes. Across a number of studies, significant

correlations have been found between SDO and anti-LG attitudes (e.g., Altemeyer, 1998;

Whitley, 1999; Whitley & Lee 2000) such that higher SDO scores are related to higher levels of

anti-LG attitudes.

Another variable that seems to predict positive attitudes toward LG individuals fairly

consistently is prior contact with a lesbian or gay man (Herek & Capitanio, 1996). Interpersonal

contact appears strongly and significantly associated with favorable attitudes toward gay men,










and quantitatively more contact experiences are associated with more favorable attitudes.

Intimate contact is more likely than superficial contact to be associated with favorable attitudes,

and direct disclosure of another' s LG orientation was more likely than second-hand disclosure to

be associated with positive attitudes toward LG people (Herek & Capitanio, 1996).

Unfortunately, in exploring the above variables, almost none of the studies link their

hypotheses or findings to a theoretical framework. Rather, one study appears to build on

another' s variable-based findings, without providing a way for the findings to be expanded or

understood. This is especially surprising when one considers that in 1984, Herek provided a

theoretical foundation on which future research could be built. Within this framework, attitudes

toward LG people enable individuals to meet a psychological need or serve a psychological

function, usually driven by motivation to strengthen one' s personal identity (Kite, 1994). Herek

(1984) suggested three functions as most relevant to understanding attitudes toward LG

individuals. They are not meant to provide exhaustive explanations of all attitudes, but rather

provide the basis for a theory of understanding the sometimes disparate studies in the area of LG

attitudes. First, attitudes may be experiential, organizing one's social reality on the basis of

previous experiences and interactions with LG individuals. Second, attitudes may be defensive,

or ego defensive, helping a person contain an inner conflict that threatens a stable sense of self by

projecting that conflict onto LG individuals. Last, attitudes may be symbolic, embodying

abstract ideals or values that are linked to one's identity, social network, or reference groups

(Herek, 1984).

Although the above variables might be classified as symbolic (e.g., gender roles,

authoritarianism) or experiential (e.g., previous contact with an LG individual), no recent

published research has explored ego defensive options, which would presumably occur due to a









less strongly developed ego identity (Marcia, 1966, 1988, 1993). The present study aims to

address this limitation in the literature, by expanding current research on attitudes toward

lesbians and gay men through exploration of ego identity as a theoretically relevant predictor of

these attitudes. Additionally, as most extant studies employ samples composed of undergraduate

college students, the current study aims to incorporate an non-college student, adult population

through the use of Internet-based sampling. In doing so, it is anticipated that a greater degree of

diversity and external validity might be represented in the findings.

This paper will be structured in the following way: first, the psychosocial background of

attitudes toward lesbians and gay men will be examined; second, Herek's (1984) theory, which

frames LG attitudes as serving a psychological function will be reviewed; third, empirical studies

that demonstrate the relationship of various variables to LG attitudes will be explored; finally,

the subsequent sections will focus on the description of the current study, the methodology that

was used to test the study's hypotheses, the study's results and a discussion thereon.









CHAPTER 2
LITERATURE REVIEW

In a letter to an American mother, who had written to Sigmund Freud over her concern

about her son's homosexuality, Freud wrote back consolingly by saying that

[Homosexuality] is assuredly no advantage but it is nothing to be ashamed of, no vice, no
degradation, it cannot be classified as an illness; we consider it to be a variation of the
sexual function produced by a certain arrest of sexual development. Many highly
respectable individuals of ancient and modern times have been homosexuals, several of
the greatest men among them (Plato, Michaelangelo, Leonardo da Vinci, etc.). It is a
great injustice to persecute homosexuality as a crime, and cruelty, too (Freud, 1951, p
787).

Despite Freud' s seeming acceptance of homosexuality as a mere variation in sexual

function, certainly no "illness" or "crime," the psychiatric community labeled lesbian and gay

(LG) people as disordered for the maj ority of the 20th century. During the latter half of the

1960's, however, the North American gay rights movement began to mature, and LG individuals

began to protest their marginalized status. Among other political goals, one of the primary

efforts of the movement was the removal of homosexuality from the DSM-II. Beginning with

the work of Evelyn Hooker (1957), mental health professionals began to raise the question

whether mental disorder was truly a direct result of homosexuality. Hooker found that there

were, in fact, normal, healthfully functioning gay men, whose Rorschach test results could not be

differentiated from those of a "normal" heterosexual population. Rather, perhaps the increased

rate of mental disorders was a result of the burden of social stigma experienced by sexual

minorities (Hooker, 1957).

With no small degree of controversy from both within and outside its ranks, the

American Psychiatric Association voted to remove the diagnostic category of Homosexuality

from the DSM-III in 1973 (American Psychiatric Association, 1980). Even though the stigma of

being a sexual minority has eroded significantly since the 1970's, (Loftus, 2001), there still









remains a significant and pervasive prejudice regarding LG individuals (Herek, 2000). Whether

the medical community initially fueled society's view of homosexuality as deviant and perverse,

or vice versa, LG people continue to be socially stigmatized for medical, religious, and other

reasons (Boswell, 1980). Apart from the obvious hardships this places on lesbians and gay men,

such as social disregard for same-sex intimacy, legal barriers which deny same-sex couples

financial and legal rights enjoyed by heterosexual couples, and disenfranchisement from

religious and community organizations, stigma and prejudice toward LG people often pose

deleterious effects on these individuals' lives, including their mental health (Meyer, 1995, 2003;

Williamson, 2000).

LG people experience violence and discrimination due to their sexual orientation. In

November of 2004, the Federal Bureau of Investigation (FBI) released its annual report on hate

crimes across the nation. The report, which documented hate crimes across the country in 2003,

contained information on 1,239 incidents in which the perpetrators' motivating factor was

victims' actual or perceived sexual orientation. Such incidents represented 17% of the total

number of hate crimes in the FBI's report (NCAVP, 2004). The FBI's statistics on anti-lesbian

and gay incidents fell short of the number of incidents tracked in the National Coalition of Anti-

Violence Programs' 2003 report on hate violence, which recorded 2,051 incidents in only eleven

regions across the nation more than half of which involved criminal offenses. It is possible that

the FBI's apparent "blind eye" toward anti-LG attitudes and behavior reflects a more general

prejudice or stigma toward sexual minorities in the greater United States' society. Indeed, when

one regards the psychological literature on prejudice toward sexual minorities, it appears that

such prejudice and lack of regard for LG individuals continues to be a presence in the U. S.'s

general culture (Herek, 2000, 2002).









The Motivational and Functional Approach to Understanding Attitudes
toward Lesbians and Gay Men

Different theorists have suggested varying functions most often served by attitudes and

opinions. For example, opinions and attitudes may be a basis of security in the face of a

changing world for one individual, yet for another a goad to revolutionary activity. Attitudes are

part of one' s attempt to master the world and life in that world; more simply, they are a way in

which one adjusts to the social environment (Smith, Bruner, & White, 1956). Other historically

notable attitude functions include (1) an instrumental, adjustive, or utilitarian function, which

serve a person in terms of concrete rewards or punishments; (2) an ego-defensive function, in

which a person protects her or himself from acknowledging basic truths about one' s self or the

external world; (3) a value-expressive function, in which an individual derives satisfaction from

expressing attitudes appropriate to her or his personal values and concept of self; and (4) a

knowledge function, which is based upon an individual's need to give structure, organization,

and meaning to the world (Katz, 1960).

In attempting to understand and organize the variables that gird an individual's prejudice

or tolerance toward LG individuals, it is helpful to bear in mind the premise that different people

can express similar attitudes toward any number of social obj ects (LG individuals, racial ethnic

minorities, individuals with perceived social status) due to different attitudinal functions (Herek,

1984). For example, two heterosexual individuals may view LG individuals negatively: one

individual may be deeply religious and hold Christian fundamentalist views, and her or his

negative attitudes toward LG individuals might be accordingly motivated by a value-expressive

function; the other individual may have developed an ontology that organizes individuals into

traditional and "perverse" gender roles, and this person condemns LG individuals as a result of

the knowledge function. Within this functional approach, attitudes toward LG people enable









individuals to meet a psychological need, usually driven by motivation to strengthen one's

personal identity (Kite, 1994).

With specific regard to attitudes toward LG people, Herek (1984) suggested three

functions as most relevant to understanding attitudes toward LG individuals. They are not meant

to be an exhaustive taxonomy, but they do provide a parsimonious framework for understanding

the sometimes disparate studies in the area of LG attitudes. First, attitudes may be experiential,

organizing one' s social reality on the basis of previous experiences and interactions with LG

individuals. Second, attitudes may be (ego) defensive, helping a person contain an inner conflict

that threatens a stable sense of ego or self by proj ecting that conflict onto LG individuals. Last,

attitudes may be symbolic, embodying abstract ideals or values that are linked to one's identity,

social network, or reference groups (Herek, 1984). Each of Herek' s (1984) three attitude

functions will now be explored in turn.

Experiential Attitudes

Experiential attitudes arise when one has had prior experience with LG individuals, and

has emotions and cognitions associated with those experiences (Herek, 1984). For example, a

person who has had positive interpersonal contact with an LG person is more likely to hold

favorable attitudes toward other LG individuals on the basis of that contact. Similarly, negative

interpersonal contact will likely inspire negative attitudes toward LG individuals. Experiential

attitudes do not arise simply from contact per se, but importantly, the contact must provide

adequate basis for the formation of attitudes (Allport, 1954). Permitting further understanding of

experiential attitudes and contact between "in" and "out" groups is the contact hypothesis

(Allport, 1954). This hypothesis focuses on the change of an individual's prejudiced attitude in a









social situation, and proposes several conditions necessary for the contact to result in reduced

negative attitudes and greater tolerance:

1. support by authority figures and voluntary participation by the participants in the
experience;

2. equal perceived status of participants in the encounter;

3. cooperation and interdependence among participants across groups;

4. individualized contact with the potential for further friendship across groups;

5. behavior that that disconfirms stereotypes that the groups hold of each other (Wittig &
Grant-Thompson, 1998).

According to the theory behind "experiential attitudes" in Herek's theory, heterosexuals

who know or have met LG individuals may be better able than others to recognize negative

stereotypes as inaccurate, and would be also be more likely to hold tolerant attitudes. Similarly,

interpersonal contact with an LG person that has not met any one of the above requirements is

less likely to result in a tolerant attitude. As only a minority of individuals in the United States

have attitudes based on first-hand experience, it is likely that many people's attitudes are based

either in negative stereotype (Herek, 1984) or function according to a different motivational

principle.

Defensive Attitudes

Defensive attitudes, or ego-defensive attitudes, are those in which a person protects

herself or himself from acknowledging basic truths about the self or the external world. In one

sense, ego defense mechanisms are adaptive in that they temporarily remove the harshness of

internal conflict, and save the individual from psychodynamic disaster. In another sense,

however, such mechanisms may not always be adaptive in that they potentially handicap the









individual in her or his social adjustment, and prevent the acquiring of maximum satisfactions

from the external world (Katz, 1960).

According to Marcia' s theory of ego identity (Marcia, 1966, 1988, 1993), if an individual

has not sufficiently explored various personal domains (gender roles, sexual attraction) and

committed oneself to particular aspects of those (and other) domains, the result is a less stable

ego identity. A less stable identity translates into less optimal defenses, leaving an individual

more vulnerable to incoming threats. Theoretically, greater anxiety, depression, and other

maladies could be the result (Marcia, 1993).

The theory of defensive attitudes toward LG individuals would suggest that heterosexual

men and women who are genuinely secure in their own gender identity and sexual orientation,

having explored various options in those domains and having committed to personally

appropriate options, would feel less threatened by homosexuality than would those who are less

secure (Herek, 1984). Without such a stable identity, feelings of personal threat may more easily

bring about strong negative attitudes toward LG people. This idea appears to have some basis in

empirical truth. In one study (Adams, Wright, & Lohr, 1996), the role of homosexual arousal in

exclusively heterosexual men who admitted negative affect toward homosexual individuals was

investigated. Participants consisted of a group of homophobic men (n = 35) and a group of

nonhomophobic men (n = 29), assigned to groups on the basis of their scores on a measure of

homophobia (or, negative attitudes toward gay men). The men were exposed to sexually explicit

erotic stimuli consisting of heterosexual, male homosexual, and lesbian videotapes, and changes

in penile circumference were monitored. Both groups exhibited increases in penile circumference

to the heterosexual and female homosexual videos, but only the homophobic men showed an

increase in penile erection to male homosexual stimuli. The authors of the study concluded that










homophobia is likely associated with homosexual arousal of which the homophobic individual is

either unaware or denies. In support of this finding and of the link between negative LG

attitudes and ego development, other work focusing on college students has found a small but

significant correlation between higher levels of ego identity development and more positive

attitudes toward homosexuality (Weis & Dain, 1979).

Attitudes thus may serve a defensive function when an individual perceives some analogy

between an LG persona and her or his own unconscious conflicts (Herek, 1984). Subsequently,

that inner conflict is proj ected onto the external source of the conflict (i.e., the lesbian or gay

man), thereby reducing the experience of internal anxiety. This strategy permits an individual to

externalize conflict regarding sexual orientation, attraction, or gender role by rej ecting the

lesbians and gay men who exemplify the forbidden, unconscious desire; in doing so, the person

is never forced to recognize the urges as her or his own.

Symbolic Attitudes

Whereas defensive attitudes have the function of preventing the revelation of one' s true

nature to oneself or to others, symbolic attitudes have the function of giving positive expression

to one's central values and the type of person she or he would like to be (and, perhaps the type of

person by which she or he would like to be accepted). For example, an individual might

consider her or himself to be politically liberal and open-minded, and as such, may express

values that are dominant in liberal political circles. The function is not to gain public reward, but

to maintain a stable sense of self in relation to valued others, and to enj oy intrinsically the

expression of that self (Katz, 1960).

Whether symbolic attitudes are favorable or not favorable, prejudiced or tolerant, they are

derived from past and present socialization experiences (Katz, 1960; McConahy & Hough,









1976). They express attitudes and values that are important to one's sense of self, thereby

helping individuals to establish their identity and affirm their notion of the sort of person they

perceive and desire themselves to be. Simultaneously, these attitudes mediate the person's

relation to other important individuals and reference groups, forming part of a continuing social

exchange: one defines a sense of self and through reinforcement of one' s attitudes and values by

others in the relevant group, and also establishes the interpersonal relations that support that self.

Sexual attitudes tend to be consistent with a larger ideology typically found in socially

significant reference groups (religions, political parties, geographic moral climate, etc.) (Herek,

1984). With so many and varied opinions regarding the appropriateness of a lesbian or gay

orientation in society, depending on one's religion (Schulte & Battle, 2004), political preference

(Herek, 2000), geographic area (Herek, 1994), and family attitudes (Negy & Eisenman),

symbolic attitudes likely play a role in many individuals attitudes regarding LG people.

Symbolic attitudes toward minority groups may be seen as the expression by a dominant

group (i.e., White individuals) that racial or ethnic minorities are violating important societal or

religious values and demanding illegitimate changes in the status quo (McConahay and Hough,

1976). Along these lines, one may consider this third category of attitudes to serve a similar

function for heterosexuals regarding lesbians and gay men. Symbolic attitudes toward LG

individuals express the feeling that important personal or society values are being violated, and

that unreasonable demands are being made to change the status quo (Herek, 1984).

In recent years, relatively scant research has been completed on attitudes toward LG

people, focusing on disparate personal variables that are correlated with tolerant or prejudicial

attitudes (such variables are discussed in the following section). The attitudinal framework

outlined above is not intended to classify all the various variables that come to play in an









individual's forming an attitude toward LG individuals. Rather, the framework begins to

illustrate that individual personal variables (gender, religion, political affiliation, etc.) can serve

any number of attitudinal functions, and can play any number of roles in an individual's

understanding of her or his self.

Research on Attitudes toward Lesbian and Gay (LG) Individuals

Fairly consistent findings from national surveys reveal that higher levels of negative LG

attitudes exist among those who are older, less educated, living in the U.S. South or Midwest,

and living in rural areas (Herek, 1994). Various other personal characteristics and traits, such as

gender (Herek, 2002; Kite & Whitley, 1998), gender role beliefs and attitudes (Sakalli, 2002;

Sirin, McCreary, & Mahalik, 2004), previous contact with a lesbian or gay man (Anderssen,

2002; Liang & Alimo, 2005), authoritarianism (Altemeyer, 1996, 2001), religious orientation and

other religious variables (Herek & Capitanio, 1996; Plugge-Foust & Strickland, 2000), and

attributions of the etiology of sexual orientation (Landen & Innala, 2002; Oldham & Kasser,

1999), have been explored empirically. There is little research regarding attitudinal differences

among racial and ethnic minority groups, with the exception of scant empirical research

regarding African Americans' attitudes toward LG people (Negy & Eisenman, 2005; Schulte &

Battle, 2004).

Previous Contact with LG Individuals

The variable that seems to predict positive attitudes toward LG individuals most

consistently is one's prior contact with a lesbian or gay man. Using racial issues as an example,

if a White individual encounters an African American individual, she or he is likely to encode

information about that Black person in terms of the person' s minority status. In other words, the

apparent and visible "stigma" of being dark-skinned acts as a filter through which otherwise









unremarkable statements or actions are interpreted as indicative or stereotypical of the Black

community (Biernat, 2003). The minority person is thus evaluated solely in terms of her or his

minority status (Biernat, 2003). In contrast, for someone with a successfully concealed stigma,

such as a gay or lesbian person, there is no filter through which actions or statements would be

interpreted. For all intents and purposes, the LG person's behavior may be indistinguishable

from a heterosexual person's, and thus appears "normal" and acceptable. When LG individuals

disclose their stigmatized status to a heterosexual person after positive regard has already been

formed, the heterosexual person may respond to the information by individuating and

personalizing the minority group's members, resulting in reduced prejudice and increased

tolerance. This factor would appear to help form the "experiential attitudes" component of

Herek' s (1984) theory of attitude functions.

In a national telephone survey (Herek & Capitanio, 1996), a sample of adults in the

United States indicated their attitudes toward gay men and lesbians (n = 538; 45.9% male and

54.1% female; 81% White, 10.4% Black, 5% Hispanic, 2.8% Asian, and less than 1% not using

any of those labels; mean age 43.8 years; median income = 30,000 and 40,000; median level of

education was "some college"; 35.3% self-labeled as "Democrat", 31.6% "Republican", and

24.5% "Independent). Most respondents expressed negative attitudes toward gay men, with a

maj ority believing that "Sex between two men is just plain wrong" and that "I think male

homosexuals are disgusting" (69.8% and 54.1%, respectively). Interpersonal contact was

strongly and significantly associated with favorable attitudes toward gay men, and more contact

experiences with more individuals was associated with more favorable attitudes. Intimate

contact was more likely than superficial contact to be associated with favorable attitudes, and

direct disclosure of another' s LG orientation was more likely than second-hand disclosure to be









associated with positive attitudes toward LG people. The number of relationships with an LG

person accounted for a small but significant amount of variance in attitudes toward lesbians

(6.3%) and gay men (6.2%). The level of contact intimacy with an LG person (immediate family

member, distant relative, close friend, or acquaintance) also accounted for 9.5% of the variance

in attitudes toward lesbians, and 9.3% of the variance in attitudes toward gay men.

Liang and Alimo (2005) conducted a longitudinal study of heterosexual college students'

contact experiences with lesbian, gay, and bisexual individuals and the resulting attitudes toward

those individuals. Using 401 White heterosexual students (70% women and 30% men), the

effects of gender, pre-college attitudes and LGB contact, and college LGB contact were

explored. The authors believed that, based on previous literature, gender would influence the

attitudes and contact that the participants would have in LGB relationships. Additionally, based

on previous literature and on Allport' s (1957) contact hypothesis, the authors' expected that

interpersonal contact with LGB individuals prior to and while in college would have a positive

affect on prejudicial attitudes of the study participants. The effects of gender were relatively

small (p = .078 to .200), with women reporting somewhat more positive attitudes than men, and

having somewhat more contact with LGB individuals prior to and while in college (P = .192, p <

.01, and p = .106, p < .05, respectively). Attitudes toward LGB individuals prior to college were

predicted by pre-college contact (P = .426, p < .001); attitudes toward LGB individuals while in

college were predicted by pre-college attitudes (P = .190, p < .001) and LGB contact prior to

college (P = .320, p < .001); and, in-college attitudes were most strongly predicted by pre-college

attitudes and in-college LGB contact (P = .563, p < .001 and P = .166, p < .001, respectively). In

regarding these statistics, it is important to recognize that pre-college contact with LGB

individuals most strongly influenced later attitudes through both direct and indirect paths (i.e.,









through forming pre-college attitudes which influenced the formation of later attitudes), and that

LGB contact while in college had a direct influence. Among college students, similar patterns of

contact-based influence appears to hold in countries other than the U.S. as well (i.e., Norway,

Turkey) (Anderssen, 2002; Sakalli, 2002).

The analyses in most LG-attitude studies are univariate and correlational in nature, and

despite Herek' s (1984) theory of motivation and functional attitudes, it is uncommon to find

studies that make use of mediational or moderation-based analyses. One interesting exception

(Mohipp and Morry, 2004) investigated the relationship of symbolic beliefs and prior contact on

heterosexual attitudes toward LG people. The study demonstrated that positive symbolic beliefs

concerning LG individuals were associated with positive attitudes toward both lesbians (P = -.52,

p < .001) and gay men (P = -.31, p < .05); and, that when separated from other predictors, prior

contact was associated positive attitudes toward both lesbians and gay men (P = .19, p < .05). In

line with Herek' s theoretical ideas, the authors expected to find that symbolic beliefs would

statistically mediate the relation between prior contact and attitudes (Baron & Kenny, 1986).

However, no such relationship was found.

Clontatl I \ ithl LG individuals in the Afr~ican American community. It is commonly

assumed that African American communities, and especially the men of those communities, have

significantly less tolerance of homosexuality than do White communities (Battle & Lemelle,

2002). To a certain extent, it appears that a marginally higher degree of intolerance may actually

exist in the African American community despite having had contact with an LG individual. For

example, Herek & Capitanio (1995) conducted a national telephone survey, and found that

gender differences in LG attitudes among African American individuals were very similar to

those of White individuals. More tolerant attitudes were predicted by being highly educated,









unmarried, politically liberal, registered to vote, nonreligious, and included being Black in their

concept of gay men. Having contact with an LG individual also predicted more favorable

attitudes, but when other variables were controlled, having such contact lost any significance.

This may be due to 25% more of the African American sample than previous White samples

reporting that their contact with an LG person consisted of knowing a distant LG friend,

acquaintance, or distant LG relative. This is congruent with Herek' s (1984) theory regarding

experiential attitudes, as well as Allport's (1957) contact hypothesis, which states that less

intimate contact is less likely to lead to a reduction in prejudice.

Gender-Based Differences in Attitudes toward LG Individuals

Heterosexual men appear to hold more negative attitudes toward lesbians and gay men

than do heterosexual women (Kite & Whitley, 1996; Kite & Whitley, 1998). This balance is not

yet fully understood, due in part to empirical work that is not solidly grounded in theory, and in

part due to a fairly narrowly constructed empirical base that is formed largely of college

students' attitudes to the exclusion of other populations (Kite & Whitley, 1998). In order to

understand the basis for possible gender-based differences in attitudes toward LG people, an

exploration of possible reasons based in the theory of attitude function (Herek, 1984) will be

beneficial. The most relevant aspects of that theory for gender-based attitude differences lie in

symbolic attitudes about gender (i.e., gender roles) and ego defensive attitudes.

Symbolic attitudes toward gender and gender roles

One possible explanation for gender-based differences in LG attitudes rests in a gender

role attitudes/beliefs analysis. This point of view states that heterosexuals' evaluations of gay

men and lesbians are grounded in a much larger belief system concerning men and women, and

their appropriate roles in society (Kite, 1994). First, gender-associated beliefs appear to be









linked to one another; in other words, if one has stereotypically masculine traits, then it is

assumed that stereotypically masculine behaviors and characteristics are part and parcel of that

person' s gender role. If any one aspect of that gender role is violated, such as a gay man having

a stereotypically masculine physical appearance, but more stereotypically feminine behaviors,

prejudice against that individual will likely be the result (Kite & Whitley, 1998). Gender roles

for men, a class of symbolic beliefs that may be grounded in family-inspired values or in

religious or political ideology, appear somewhat more narrowly defined and inflexible than are

those for women (Bem, 1993). A violation of male gender roles is usually considered more

inappropriate than violation of female roles (Herek, 1988). An individual's violation of those

roles results in a challenge toward concepts of gender appropriateness, and thus toward the

particular individual's acceptability.

In a meta-analysis of gender-based differences in LG attitudes (Kite & Whitley, 1996,

1998), gender role attitudes had a mean correlation of r = .44 with scores on measures of

attitudes toward lesbians and gay men. Interestingly, when sex differences in gender role

attitudes were controlled, the average partial correlation between gender role attitudes and

attitudes toward lesbians and gay men fell to r = .02. These results indicate that sex differences

in LG attitudes are related to differences in gender role attitudes.

Clarifying further gender differences in symbolic beliefs, Herek (2002) conducted a

national telephone opinion survey and reported a number of factors on which gender-based

differences were found. Part of the sample was composed of previously polled respondents from

an survey that was completed two years prior (n = 666; 57% female, 43% male; 81% non-

hispanic White; mean age = 47, with an age range of 20-91; median education level = "some

college"; and, median income = $40,000 $50,000). The newly recruited respondents had a very









similar demographic composition with the exception of age, which was an average of two years

younger than the previously polled respondents. This reflects the period of time elapsed between

surveys. The findings were split into three major areas: (1) women generally hold more

favorable and less condemning attitudes toward LG people they were more supportive of

employment protection and adoption rights, were more willing to recognize same-sex

relationships as valid, were less likely to hold stereotypical beliefs, and displayed less negative

affective reactions; (2) both heterosexual men's and women's attitudes tended to be more hostile

toward gay men than lesbians, with gay men more likely than lesbians to be regarded as mentally

ill and to be perceived as child molesters, and with adoption rights more strongly supported for

lesbians than for gay men; (3) although heterosexual individuals expressed more negative

attitudes toward LG people of their same sex, this pattern was strongest among men (i.e.,

heterosexual men rated LG individuals more negatively than women, and rated gay men more

negative of all).

An additional interesting finding from the study (Herek, 2002) that replicated the findings

of the first telephone survey mentioned above (Herek & Capitanio, 1999), was a significant

effect on heterosexual men's attitudes determined by the order in which the survey questions

were posed. When questions concerning lesbians were asked first in a series, thus not casting

them in the context of attitudes toward gay men, heterosexual men were less hostile toward

lesbians than gay men in some cases, their attitudes toward lesbians were at least as favorable

as those of heterosexual women. This suggests that heterosexual men' s attitudes toward lesbians

are cognitively organized in a way that differs from attitudes toward gay men. In contrast,

heterosexual women's attitudes toward LG people are relatively unaffected by contextual

variables (question order in the survey). This may be due to findings in previous studies that










suggest heterosexual men's tendency to sexualize interactions with the opposite sex more often

than do heterosexual women (DeLamater, 1987). Presenting a potentially sexualized image of

lesbians before gay men may "prime" the respondents to exhibit more positive attitudes.

A study by Louderback and Whitley (1997) clarified the above suggestions regarding the

role of (a) gender role attitudes and (b) heterosexual men's perception of sex between lesbians as

erotic on attitudes toward lesbians and gay men. In a sample of 58 male and 109 female college

students (age range = 17 25 years, mean age = 18.7), the sex-based differences in perceptions

of LG people mentioned above were replicated. Additionally, men attributed more erotic value

to sex between two women than to sex between two men, whereas women attributed low erotic

value to both forms of sexual behavior. The authors posited a mediated model, in which sex

differences in attitudes toward lesbians and gay men would be mediated by the perceived erotic

value of homosexuality and sex-role attitudes. This hypothesis was supported. Whereas the

perceived erotic value of lesbian/gay sex previously accounted for 1 1.9% of the variance in

heterosexuals' LG attitudes, any differences between men's attitudes toward lesbians versus gay

men was eliminated when perceived eroticism was controlled. Men's LG attitudes, though, were

still significantly more negative than those of women, with this sex of participant by sex of

attitude target accounting for 8.3% of the variance in LG attitudes. When sex role attitudes

(which accounted for 17.3% of variance in LG attitudes) was also controlled, the remaining

gender-based differences in attitudes toward lesbians and gay men was eliminated. No main

effect remained for the sex of the participant or the sex of attitude target, and neither were there

any interaction effects of these two factors (Louderback & Whitley, 1997).









African American-based gender differences in LG attitudes

As previously mentioned, attitudes among African American and White individuals are

generally similar. For example, similar to large national samples, African American men appear

to hold less positive attitudes toward LG individuals than do African American females. The

etiology of this sex-based difference appears, though, not to lie in the gender/sex role attitudes

that predict White individuals' LG attitude differences (Lemelle & Battle, 2004). In data

provided by the National Black Politics Study in 1993, representing approximately 6.5 million

African American households, variables measured included age, church attendance, education,

household income, urban residency, and attitudes toward gay men (age = 40.97 years, SD =

15.86; Education = 13.07 years, SD = 2.98; Household income = 4.74, SD 2.42 [from a possible

9 categories that ranged from "under $10,000" to "over $75,000]; Urban Residency = .55, SD =

.50; and for religious attendance, 87% attended Church at least once a month). Whereas age,

income, education level, and urban residency were all significantly related to LG attitudes for

women in the study (p < .05), the only variable that was significant for men's LG attitudes was

the frequency of church attendance (p < .05). For these men, more frequent church attendance

was related to less favorable LG attitudes. Although significant, the effect size of church

attendance, even when combined with other the other predictors in one regression, was relatively

small (Adjusted R2 = .005 for men, and .033 for women).

Religion and Religious Variables

The popular media often discusses the controversy in religious circles regarding

individuals of a lesbian or gay orientation. For example, according to a Catholic encyclical, the

Church regards homosexual behavior to be "not a normal condition, the acts being against nature

[and] objectively wrong" (p 272, Broderick & Broderick, 1990). In this light, a number of









studies have regarded the importance of religion and religious values in attitudes toward LG

individuals. Allport & Ross (1967) defined two different religious orientations: extrinsic, a self-

serving, instrumental approach conforming to social conventions; and intrinsic, in which religion

provides a meaning-endowed framework in terms of which all life is understood. Allport &

Ross' s (1967) suggested that extrinsic individuals would likely be more prejudiced, and that

intrinsic individuals would be more tolerant toward non-dominant groups. Although previous

research has found intrinsically religious individuals to hold fewer racist attitudes and fewer

racial prejudices than do extrinsically religious individuals (Donahue, 1985), Herek (1987)

surprisingly found that intrinsically religious college students were more prejudiced against LG

individuals than their extrinsic counterparts. He suggested that an intrinsic orientation does not

foster an unequivocal acceptance of others, but rather encourages tolerance toward those that are

accepted by traditional religious teachings (Herek, 1987).

Further research on religious orientation revealed that the motivating factors behind

negative LG attitudes are more complex. In a sample of 257 college students at a conservative

Christian school in California (110 females and 66 males, all White; age range = 18 24, mean

age = 18.5; years of college experience range = 0 3, mean experience = 1.5), findings similar to

Herek' s (1987) regarding Intrinsic-Extrinsic religious orientation were initially found, with

Intrinsic individuals endorsing more negative LG attitudes. However, when the influence of the

variable Fundamentalism was controlled, High Intrinsics were more accepting of LG individuals

than Low Intrinsics. As such, Fundamentalism appears to be the motivating force behind some

religious individuals' anti-LG sentiment. Later research supports this suggestion, stating that

among other factors, Fundamentalism accounts for a significant portion of the variance in

predicting individuals' negative LG attitudes (P = .44, p < .001) (Schwartz & Lindley, 2005).










Afr~ican American religious service attendanced~~~~~ddddd~~~~ & LG attitudes. As mentioned in the

previous section on gender differences, Lemelle and Battle (2004) found that attendance at

religious services was the single distinguishing factor between African American men's and

women's LG attitudes. Another study affirmed that religious service attendance may indeed play

a unique role in African American individuals' perception of LG people (Schulte & Battle,

2004), although religious service attendance was not examined by participant gender. When

predicting attitudes toward lesbians, ethnic differences were present in the absence of religious

attendance; however, when religious attendance was entered into the model, ethnic differences

disappeared entirely (see also Negy & Eisenman, 2005). Somewhat similarly, regarding

attitudes toward gay men, although there were no racial differences in attitudes toward this

group, religious attitudes were always significant. The authors concluded that LG attitudes do

not necessarily vary as a function of ethnicity, but rather may be a function of religious

attendance in African American churches (Schulte & Battle, 2004). Race and religious

attendance alone explained variance in LG attitudes ranging from approximately 5 to 10%; when

other factors of sexual orientation, marital status, geographic region, and religion were entered in

the model, the variance explained in LG attitudes was approximately 20 to 23%.

Personality-Based Factors

Altemeyer (1981) defined the overarching concept of authoritarianism as comprised of

three attitudinal clusters. The first component, authoritarian submission, is defined as a "high

degree of submission to authorities who are perceived to be established and legitimate in the

society in which one lives" (p. 148). The second component is authoritarian aggression and

reflects "general aggression, directed at various persons, which is perceived to be sanctioned by

established authorities" (p. 148). Aggressiveness in this context is the predisposition to do










physical, psychological, economic, or social harm to others. Persons with high levels of

authoritarianism tend to be aggressive or support aggressiveness toward targets that they believe

are socially stigmatized. For example, persons high in authoritarianism might support police

violence against suspects in order to obtain confessions. The final component of Altemeyer' s

authoritarianism is conventionalism defined as a "high degree of adherence to the social

conventions which are perceived to be endorsed by society and its established authorities" (p.

148). Altemeyer (1981) described "established and legitimate authorities" to include individuals

in society who are usually considered to be legal or moral authorities such as parents for younger

persons, religious officials, heads of government, police officers, judges, and military officials.

Altemeyer' s (1981, 1996) theoretical conceptualization of authoritarianism is

characterized by high degrees of submission to authorities, aggressiveness toward outgroups, and

adherence to conventions perceived to be endorsed by society and its authorities, suggesting that

people with high degrees of authoritarianism would be likely to exhibit negative LG attitudes.

More specifically, this is likely because of LG persons (1) being explicitly condemned by some

religious and political leaders, making LG persons an acceptable target for prejudice and

hostility, (2) representing a socially stigmatized out-group relative to heterosexual women and

men, and (3) being perceived as advocates for political change and challenging social

conventions of traditional gender roles. Indeed, a number of studies have found significant

positive links between authoritarianism and anti-LG attitudes (e.g., Altemeyer, 1981, 2001;

Ekehammar, Akrami, Gylje, & Zakrisson, 2004; Laythe, Finkel, & Kirkpatrick, 2001) indicating

that persons with high levels of authoritarianism express more anti-LG attitudes than do persons

with low levels of authoritarianism.









An interesting study examined negative LG attitudes as predicted by authoritarianism and

religious fundamentalism (Laythe, Finkel, & Kirkpatrick, 2001). Their sample consisted of

college students from a Midwestern university (n = 140; 91 females and 47 males; age range =

18 to 48, mean age = 21.2, SD 6.94; 35% of the sample reported regular, once-weekly church

attendance, and 9.2% labeled themselves as atheist or agnostic; on a seven-point scale of

religiosity [1 = not religious, 7 = strongly religious], the mean was 3.9, SD = 1.7).

Authoritarianism (P = .37, p < .01) and fundamentalism (P = .21, p < .05) emerged as significant

predictors, each accounting for variance in LG prejudice above and beyond the prediction

afforded by the other. Together, the two predictors accounted for 28% of the variance (adjusted

R2). To explore further the relationship between fundamentalism and authoritarianism, the

authors completed a secondary post hoc analysis of data provided by previous Canadian studies

(Altemeyer & Hunsberger, 1992; Wylie & Forest, 1992). Demographic information for these

samples was not provided, beyond the fact that one study's participants were parents of college

students (247 women and 244 men), and a random sample of Canadian residents in Manitoba (43

women and 32 men). In both data sets, authoritarianism was a significant positive predictor of

negative LG attitudes (P = .69 and P = .67 for the two samples, both at p < .01). However,

fundamentalism was not a significant predictor in either data set apart from authoritarianism.

The authors point out that even though their own data found fundamentalism to predict negative

LG attitudes significantly, its actual effect in doing so was relatively small (P = .21). As such,

fundamentalism appears to predict very little if any variance beyond authoritarianism's relatively

strong predictive power (Laythe, Finkel, & Kirkpatrick, 2001).

In addition to authoritarianism, social dominance orientation (SDO) also has been linked

consistently with negative attitudes toward lesbians and gay men. Pratto, Sidanius, Stallworth,









and Malle (1994) defined SDO as the "extent to which one desires that one' s in-group dominate

and be superior to out-groups" (p 724). SDO taps a general attitudinal orientation toward

intergroup relations that reflects a preference for hierarchical rather than egalitarian relations.

More specifically, the desire to maintain the superior position of their in-groups relative to

outgroups motivates people high in SDO to accept hierarchy-legitimizing myths that denigrate

members of out-groups and enforce the status quo of their in-groups' power position. Thus,

persons high in SDO tend to have negative LG attitudes because LG persons have lower social

status than do heterosexuals (Whitley & Lee, 2000). The empirical literature supports the

expected link between SDO and anti-LG attitudes. Across a number of studies, significant

correlations have been found between SDO and anti-LG attitudes (e.g., Altemeyer, 1998;

Whitley, 1999; Whitley & Lee 2000) such that higher SDO scores are related to higher levels of

anti-LG attitudes. For example, in one study using a sample of 266 college students (122 men,

131 women; age range = 18 24, mean age = 19.2; 91% White, 4% African American, 2%

Hispanic, 2% "other ethnicity/race," and 1% Asian), SDO contributed significantly along with

authoritarianism, gender role beliefs, and gender in predicting attitudes toward LG individuals

(Whitley & Aegisdottir, 2000). When entered jointly in a regression, authoritarianism and SDO

accounted for 3 8.6% of the variance in attitudes toward lesbians, and 39.5% of the variance in

attitudes toward gay men (p < .001).

Although authoritarianism and SDO both include an element of prejudice toward

outgroups, Altemeyer (1988) argued that the two constructs are distinct in that SDO "does not

have the same psychological roots that previous studies have unearthed in right-wing

authoritarians" (p. 61). More specifically, a person high in authoritarianism is very accepting of

traditional values and institutional authorities and is likely to follow the instructions of those in










positions of power and influence. By contrast, an individual high in SDO is not motivated by a

sense of duty or morality (Heaven & Bucci, 2001). Thus, although authoritarianism and SDO

both can explain significant, unique and prejudicial variance in negative LG attitudes, they do so

in a complementary but not redundant fashion (Altemeyer, 1998). Indeed, empirical evidence

indicates that authoritarianism and SDO are only minimally correlated with each other, if at all,

(Altemeyer, 1998; Pratto et al., 1994) and the relationship of SDO to prejudice is not accounted

for by a joint relationship with authoritarianism (Altemeyer, 1988; Whitley & Lee, 2000).

Defensive Attitudes with a Possible Basis in Gender

Psychodynamically inspired defensive attitudes are associated with perceptions of one' s

own prized dissimilarity from gay men and lesbians, and with the previously mentioned

projection of internal conflict. Empirically speaking, heterosexual males consistently hold more

negative attitudes toward LG individuals than do heterosexual females; and, heterosexual males'

attitudes toward gay men are more negative than toward lesbians (Herek, 1988). Heterosexual

females also tend to regard LG people negatively, but their attitudes do not differ significantly

according to the gender of the target (Herek, 1988). Dynamically speaking, if an obj ect in the

external world recalls undesired traits one fears or dislikes in oneself, and if men' s gender roles

are more rigid than are those for women, the proj section of a proportionately greater negative

attitude from men toward other men makes theoretical sense (Kite & Whitley, 1998).

Regarding the concept of a well developed or strong ego, it is useful to recall Erickson's

(1968) stages of lifespan development. In one of the stages, Erikson posited that an individual

must endure an identity crisis" or struggle within her or himself to develop a consistent view of

the world, and how one fits into that world. In this way, there is a consistent picture of oneself as

belonging in the world, and knowledge of how the world is structured for one' s self. Marcia










(1964, 1966, 1988, 1993) elaborated on this concept, suggesting that a well developed or

"achieved" ego structure must have completed phases of development that he called

"exploration" and "commitment," which is necessary across many different domains (e.g.,

occupation, religion, sex roles). In terms of exploration, an individual with a well developed ego

will have explored options for herself/himself in various domains (i.e., for the domain of gender

roles: "Do I believe that women should remain at home to raise children? Do I believe that men

could be adequate 'stay-at-home' fathers?"). For commitment, an individual may have explored

various options and come to accept one in particular (i.e., "I really believe that both men and

women should be in the work place"); or, the individual may not have explored options, but still

committed to a particular one (i.e., "My father worked, and my mother stayed at home it

worked for them, and it should probably work for everyone").

Studies occasionally suggest that a lack of self-exploration or self-awareness may

contribute to attitudes toward LG individuals. For example, heterosexual male college students

in one study completed measures that identified that importance ascribed to traditional masculine

attributes, the perceived self-discrepancy between one's own attributes versus masculine

attributes, and homphobia (n = 85; 32% Protestant, 28% Catholic, 13% Jewish, 13% "Other

religions", 14% "No religion; 87% were White, 8% were African American, Asian American, or

Hispanic, and 5% identified as having "other" ethnicities) (Theodore & Basow, 2000).

Interestingly, participants who valued highly masculine attributes yet perceived themselves as

not having those attributes, and who did not value feminine attributes, had the highest scores on

the global measure of homophobia. In predicting homophobia, an interaction effect of

Masculine Attribute Importance x Self-discrepancy Along Masculine Traits significantly

predicted homophobia (P = .67, p < .001). Also, an interaction effect of Feminine Attribute









Importance x Self-discrepancy Along Masculine Traits significantly predicted homophobia (P = -

.51, p < .01). When viewed through a lens of ego identity development, these Eindings could

suggest that there may be some heterosexual college students who have "committed" to the

importance of stereotypically masculine attributes, but have not yet explored how their own

differing personal attributes could or could not also be of value. If one does not explore one' s

own personal characteristics and traits, but instead commits to an ideal without exploring its

personal significance, it is likely that rigid ego boundaries (i.e., high scores on homophobia),

may result to defend these beliefs.

No recently published studies have explicitly regarded the role of ego development in

attitudes toward lesbian and gay individuals. This is surprising, considering Herek's (2000)

research directive which stated that "relatively little research has been devoted to understanding

the dynamic cognitive processes associated with antigay attitudes" (p 21). In a study published

over 25 years ago that utilized a sample of 200 homosexual and heterosexual White, Protestant

college students, there was a significant correlation between higher levels of ego development

and more positive attitudes toward homosexuality (Weis & Dain, 1979). Also, in another more

recent sample of 440 undergraduates found in an unpublished dissertation, there existed a

significant but complex relationship between ego identity statuses and attitudes toward LG

people (Tureau, 2004). Specifically, among individual students who were either high or low on

the trait of absolutism, attitudes toward LG people were more similar in those who had an

"achieved" status. Due to the unavailability of that study's complete Eindings, further

information concerning contemporary correlations between ego identity and LG attitudes is

unknown. Regardless, it appears that there may be a relationship between more advanced or

"achieved" ego identity statuses, and positive attitudes toward LG individuals.









Purpose of the Study

According to Herek' s (1984) theory, attitudes toward LG individuals can fall into three

categories: symbolic, experiential, and ego defensive. Previous LG attitude research has

explored symbolic concepts including those found in authoritarian and socially dominant

personality styles/beliefs, religious beliefs, and also gender roles for men and women. In terms

of experiential beliefs, the concept of previous contact with an LG person has also been

explored. Each of these variables, in addition to other demographic information previously

mentioned (education level, urban/rural residence, gender, age), has been shown to be related

conceptually or empirically to attitudes toward lesbians and gay men. The role of ego

development, which may fuel Herek's (1984) defensive attitude function, has yet to be explored

in any published research.

This study served to explore the role of ego identity development in the context of

individual demographic information (including gender), authoritarianism, social dominance

orientation, and gender roles. It also served to replicate and extend findings from previous

studies by using an adult, non-college student population in collecting data. It was expected that

ego identity would be correlated with aspects of the contact experience with an LG person, and

would play a moderating role in the relationship between such contact and the resulting attitudes.

It was also expected that authoritarianism and social dominance would predict levels of ego

identity, which would in turn predict attitudes toward LG individuals. This study examined the

following hypotheses:

(1) Higher levels of Authoritarianism and Social Dominance Orientation will predict
higher levels of Ego Identity Commitment and lower levels of Ego Identity
Exploration.

(2) A higher number of LG contact experiences and greater intimacy in the contact
experience with an LG person will predict higher levels of Ego Identity Exploration.











(3) Higher levels of Ego Identity Exploration will predict more positive attitudes toward
LG individuals, and lower levels will predict more negative attitudes toward LG
individuals.

(3a) Ego Identity Exploration will moderate the relationship between
Authoritarianism and LG attitudes; it will also moderate the relationship between
SDO and LG attitudes.


Additional Research Questions:

(1) Previous findings regarding gender and LG attitudes in college students will be
replicated, but using a non-college student population.

(la) Males will have more negative attitudes toward LG individuals than will women;
this difference will be eliminated when controlling for gender role values.

(lb) Less condemning attitudes toward lesbians (as measured by the Attitudes toward
Lesbians subscale of ATLG-S) and more condemning attitudes toward gay men
(as measured by the Attitudes toward Gay Men subscale of ATLG-S) will be
associated with a male gender; however, when erotic values of LG sexuality are
controlled, they will account for gender-based difference in attitudes toward
lesbians and gay men.

(2) Authoritarianism, Social Dominance Orientation, and Church Attendance will
significantly predict negative attitudes toward LG individuals.

(2a) When controlling for the above variables, race will not be a significant predictor
of LG attitudes.









CHAPTER 3
METHOD

Participants

Considering that the maj ority of the literature regarding attitudes toward LG individuals

is based on data gathered from college students, it is unknown how the results from those data

might generalize to a larger, non-college student sample. As such, there is a need for studies of

sexual prejudice within different subsets of the population, including differently aged people

(Herek, 2000). Participants in this study were adult participants recruited not from a college

student population, but by means of advertisements on various e-mail listserys and Internet

discussion groups. A variety of sampling sources were chosen based on individuals' possible

interest in LG issues (e.g., The American Psychological Association's Divison 44 listserv);

individuals' likelihood of endorsing interest in social/political issues that are relevant to

ethnic/racial minorities, thus possibly providing a racially/ethnically diverse sample (e.g.,

Google's "Black Focus" discussion group, and newsgroups such as

alt. culture. african.american.issues); and, individuals' likelihood of having a defined attitude

toward other issues explored in this survey, such as gender-issues and nationalism-based issues

(e.g., Yahoo!'s Republican and Democrat discussion groups). An advertisement was posted on

each of these groups with the relevant Internet-link to the survey, so that individuals could easily

access the informed consent page and survey questionnaires. Demographic information was

collected, including age, sex, race/ethnicity, socio-economic status (education level and income

level), urban/suburban/rural residence, and geographic region of residence (e.g., Northeast,

South, Midwest), number and intimacy level of LG contact experiences, and frequency of

religious service attendance. The participant' s sexual orientation was assessed to in order to

locate data only from self-identified heterosexual individuals. This was done by asking










respondents to identify as one of the following: bisexual, gay, heterosexual, or lesbian. Further,

to ensure consistency among respondents' choice of sexual orientation, individuals were further

asked to respond to a scale that rated their sexual attraction to others, ranging from "1 =

exclusively the opposite sex" to "7 = exclusively the same sex." It was expected that

heterosexual respondents would endorse responses that indicate more attraction to the opposite

sex than the same sex.

To determine the necessary sample size for this study, a multiple regression using all

possible predictors to predict LG attitudes was considered, as this exploratory analysis will be

the most comprehensive analysis in the study (in that it would conceivably employ all predictors

simultaneously). Tabachnick and Fidell's (2001) formula for sample size with multiple

regressions was employed: n 104 + m, where "m" equals the number of independent variables.

With measured predictor variables (authoritarianism, social dominance orientation, gender role

attitudes, ego identity commitment and exploration, erotic value of same-sex intimacy, and social

desirability) and demographic information mentioned above all being possible independent

variables, "m" in the equation is equal to 17 (104 + 16 = 121). As such, the total number of

participants required for this study was 121. This is relatively conservative, as all independent

variables will be used simultaneously only for the purposes of non-hypothesized exploratory

analysis.

At the end of data collection, which lasted approximately 2 months, the final number of

participants was 235. After deleting 54 participants who identified as lesbian, gay, or bisexual,

the total number of participants was 181; one additional female participant' s responses were

deleted, as she identified as "heterosexual" but endorsed being attracted to "usually the same sex,

and sometimes the opposite sex." After searching for respondents who did not respond correctly









to two questions designed to indicate random responding (i.e., "Please choose 'disagree' from

the items below"), 19 response sets were further deleted.

The final sample was composed of 162 individuals (114 females, 48 males; mean age =

32.6, SD = 12.0; 1 participant identified as Native American [.6%], 28 identified as Asian or

Asian American [17.3%], 17 identified as African or African American [10.5%], 18 identified as

Hispanic [11.1%], no participants identified as Hawaiian/Pacific Islander [0%], and 20 identified

as Other [12.3%]. (The percentages listed above add to more than 100% due to participants

being allowed to check more than one category.) 71 participants endorsed living in an urban

setting, 80 endorsed a suburban setting, and 9 endorsed a rural setting (2 participants did not

respond to this question). The mean income level was 4.4 (SD = 2.2), with level 4 being

between $25,000 and $34,999, and level 5 being between $35,000 and $49,000. The mean

education level was 7 (SD = 2), indicating that some graduate study, but no graduate degree was

the average level of education; also, when considering standard deviations, the maj ority of

participants indicated having earned at least an associate's degree or higher.

Design and Procedure

The weblink to the survey was posted on various listserys and Internet discussion groups

so that individuals who were interested in participating in the study could easily access it.

Particular effort was exerted to have the weblink circulated to listserys or Internet groups that

could promote the ethnic and racial diversity of the sample. Participants were also encouraged to

circulate the weblink to others who may have been interested in participating in the research.

Prior to completing the survey, participants read about the purpose and the methods of the study

and electronically endorsed a consent form. Participants were informed that if they chose to










participate they would have the freedom to withdraw from the study at any time without any

penalty .

Instruments

Attitudes Toward Lesbians and Gay Men Scale, Short Form (ATLG-S)

The ATLG-S was used to measure attitudes toward LG individuals. Herek' s (1988) scale

is one of the more commonly used measures of anti-LG attitudes and assesses the affective

component of anti-LG attitudes referred to as "condemnation-tolerance." The ATLG-S uses a

5-point Likert scale from 1 = disagree strongly to 5 = agree strongly. The scale has 10 items:

5 items assess attitudes toward lesbians (ATL-S), and 5 items assess attitudes toward gay men

(ATG-S). Sample items include "Homosexual behavior between two men is just plain wrong"

and "Female homosexuality is a sin." With this scale, separate scores for attitudes toward lesbian

persons and attitudes toward gay persons can be computed by averaging or adding ATL-S or

ATG-S items, respectively. Item ratings were added to yield an overall ATLG-S score, with

higher scores indicating more negative attitudes. Construct validity for full ATLG scores has

been demonstrated through consistently high correlations with variables associated conceptually

with anti-LG attitudes such as dogmatism, conservative political ideology, and lack of personal

contact with LG persons (Herek, 1988, 1994; Whitley & Lee, 2000). Each short version of the

scale correlates highly with its longer counterpart (ATG with ATG-S, r = .96; ATL with ATL-S,

r = .95; and ATLG with ATLG-S, r = .97) (Herek, 1994). ATLG-S scores have yielded

satisfactory levels of internal consistency with community-sampled scores on the combined scale

(ATLG-S), ATL-S subscale, and ATG-S subscale, having Cronbach's alphas of .92, .85, and .87,

respectively (Herek, 1994). The Cronbach's alphas for the scale in the current study were .94,

.85, and .93, respectively.










Right-Wing Authoritarianism Scale, Short Version (RWA-S)

The RWA-S, developed by Zakrisson (2005), assesses Altemeyer' s concept of

authoritarianism (1981, 1996). It is a 15-item, 7-point Likert-type scale with item ratings ranging

from 1 = strongly disagree to 7 = strongly agree as recommended by Altemeyer (1996), and item

content was derived from the original RWA scale (Altemeyer, 1981, 1996). Higher scores

indicate higher authoritarian attitudes and beliefs. Sample items include, "The 'old-fashioned

ways' and 'old-fashioned values' still show the best way to live," and, "Our country needs a

powerful leader in order to destroy the radical and immoral currents prevailing in today's

society." With regard to validity, Zakrisson (2005) demonstrated that RWA-S scores are

correlated moderately and positively with scores on measures of racism and sexism, as does the

original RWA scale (Altemeyer, 1981, 1996), and is not significantly correlated with social

dominance. The RWA-S has been shown to have adequate reliability, yielding Cronbach's

alphas of approximately .80 (Zakrisson, 2005). The Cronbach's alpha for the scale in the current

study was .89.

Social Dominance Orientation Scale, Abbreviated (SDO-A)

The SDO-A, developed by Pratto et al. (1994), was used to measure social dominance

orientation, or respondents' preference for inequality among social groups. The SDO is an 8-

item, 7 point Likert-type scale and participants rate their responses on a range from 1 = very

negative to 7 = very positive. A scale score was computed by adding item ratings, with higher

scores indicating greater levels of SDO. Sample items include, "Some groups of people are

simply inferior to other groups," and, "It' s OK if some groups have more of a chance in life than

others." With regard to validity, SDO-A scores have been shown to correlate positively with

cultural elitism, ethnic prejudice, sexism, political-economic conservatism, and a desire to end









affirmative action, thus demonstrating a high degree of construct validity (Pratto et al., 1994).

Additionally, SDO-A scores have yielded high reliability estimates with an average Cronbach's

alpha of .86 across multiple samples (Pratto et al., 1994). The Cronbach's alpha for the scale in

the current study was .92.

Attitudes toward Women (ATW) Scale

The ATW (Spence & Hahn, 1997) was used to assess traditional gender role attitudes.

The ATW was created to assess people's beliefs about the responsibilities, privileges, and

behaviors in a variety of spheres that have traditionally been divided along gender lines, but

could, in principle, be shared equally by men and women. The scale's authors state that the title

of the scale is something of a misnomer, in that the scale taps beliefs about appropriate

responsibilities and rights for women versus appropriate responsibilities for men (Spence &

Hahn, 1997). As responsibilities for both genders are included in the scale, and it has been used

previously to determine gender role attitudes as they are related to LG attitudes (Louderback &

Whitley, 1997), it remains an appropriate measure for use in this study.

The ATW is a 15-item scale, with items scored on a response scale whose extremes are

labeled agree strongly and disagree strongly. Whereas in previous years Spence and Hahn

recommended using a 4-point scale, they currently recommend used a 5-point response scale as

this has been shown to produce a wider range of scores and less skewness (Hahn, 1993).

Approximately half of the items presents an egalitarian point of view and the remainder presents

a traditional point of view. The egalitarian items are reverse-scored. The item scores, which

range from 1 to 5, are summed to obtain a total scale score, with higher scores indicating more

egalitarian attitudes. Sample items include, "Under modern economic conditions, with women

active outside the home, men should share in household tasks such as washing dishes and doing









laundry," and, "There are many jobs in which men should be given preference over women in

being hired or promoted." With regard to validity, Spence and Hahn (1997) conducted a

longitudinal study, comparing scores of college students on the ATW since the early 1970's.

Findings revealed that as more women have entered the labor force in greater numbers, and as

legislation barring gender discrimination has come to pass, attitudes have shifted over time to

become more egalitarian as measured by the ATW. Additionally, ATW scores appear

unifactorial across genders, and have yielded high reliability estimates, with Spence and Hahn

(1997) reporting Cronbach's alphas in the mid-80's or higher. The Cronbach's alpha for the

scale in the current study was .91.

Ego Identity Process Questionnaire (EIPQ)

The EIPQ (Balistreri, Busch-Rossnagel, & Geisinger, 1995) is a 32-item scale that was

used to obtain continuous measures of participants' levels of exploration and commitment. Items

are rated on a 6-point Likert scale on subscales of exploration and commitment, which will be

separately added and averaged, with higher scores suggesting higher degrees of exploration or

commitment. Use of this scale is appropriate given that the measure provides greater

interpretability (on continuums of exploration and commitment) than do other measures which

generally use complex scoring techniques to assign one overall status to a participant. Sample

items include "My ideas about men's and women's roles have never changed as I became older,"

and, "I have definite views regarding the ways in which men and women should behave." With

regard to validity, Balistreri et al. (1995) report convergent validity with other measures of ego

identity, as well as a satisfactory item-loading on relevant factors. Cronbach's alpha values for

the exploration and commitment scales reported by the authors were .76 and .75, respectively;

one-week test-retest reliability was .90 for commitment, and .76 for exploration (Balistreri,









Busch-Rossnagel, & Geisinger, 1995). The Cronbach's alpha for the scale in the current study

was .87 for exploration, and .81 for commitment. Scores were transformed into z-scores, so that

higher (positive) scores indicated an above-average score, while lower (negative) scores

indicated a below-average score.

Social Desirability

A short form of the Marlowe-Crowne Social Desirability Scale (Reynolds, 1982) was

used to assess the tendency to provide socially desirable responses and thus to detect a common

source of response bias. Items, which are rated using a "true-false" format, were constructed so

that the socially desirable response would be unlikely to be true for most people. Sample items

include, "No matter who I'm talking to, I'm always a good listener," and "It is sometimes hard

for me to go on with my work if I am not encouraged." High scores on this scale reflect a

tendency to offer socially desirable responses. The original form of this measure has been shown

to be unrelated to measures of psychopathology and related to measures of underreporting

symptoms (Crowne & Marlowe, 1960). The short form used in this study consisted of the 13

items that had the highest loading in an exploratory factor analysis of the original 33-item

measure. The short form has been found to be highly correlated with the original scale, and the

internal consistency reliability was estimated to by .76. The Cronbach's alpha for the scale in the

current study was .79.

Eroticism

Eight items used by Louderback and Whitley (1997) were employed. Four items referred

to gay male sexuality, and four parallel items referred to lesbian sexuality: "I find the idea of a

man (woman) making love to another man (woman) erotic," "I find the idea of a man (woman)

making love to another man (woman) repulsive" (reverse scored), "I think that I would be









sexually aroused by watching two men (women) make love," and "I have viewed pornographic

materials involving male homosexual (lesbian) acts." The first item was originally used by

Nyberg and Alston (1977, as cited by Louderback & Whitley, 1997); the other items were

composed as variations on that theme. All statements will be rated on nine-point scales; the first

three items will be rated on scales ranging from strongly disagree (1) to strongly agree (9), and

the last item on scales ranging from never (1) to frequently (9). The perceived erotic value of

lesbianism scale had a reported reliability of .88, and the perceived erotic value of male

homosexuality scale had a reported reliability of .65. To maintain consistency with the only

previous study that examined eroticism with regard to gender and LG attitudes (Louderback &

Whitley, 1997), these items were interspersed with items from the ATW scale. The Cronbach's

alpha for the scale in the current study was .79 for the scale of gay erotic value, and .86 for the

scale of lesbian erotic value.

Analytic Strategies

To ascertain associations suggested in hypotheses 1-3 (excluding 3a) and research

question 2/2a, multiple regression analyses were employed, using the relevant EIPQ score or

ATLG score as the criterion variable, while entering the relevant predictor variables in one step.

Separate analyses were conducted for each individual hypothesis. For example, in hypothesis

two, Ego Identity Exploration was specified as the criterion variable, while LG contact

experiences and the degree of those experiences' intimacy was specified as predictor variables.

To evaluate the moderation-based relationships in hypothesis 3a, statistical strategies

frequently outlined in the literature was employed, and hierarchical multiple regressions were

used (Frazier, Tix, & Baron, 2004). Before proceeding, the predictor and the moderator variables

were standardized, thus reducing problems associated with multicollinearity among predictor and









moderator variables and the interaction terms created from them. To test the moderation effects,

variables were entered in the regression equation in a stepwise manner: the variable to be

controlled for (Social Desirability) was entered in the first step, followed by the standardized

predictor and moderator variables in the second step, followed by the product (interaction) terms

in the third step (Aiken & West, 1991; Holmbeck, 1997). Finally, all remaining possible

interactions with the controlled variable, Social Desirability, were checked in the final step to

ascertain no unexpected significant interactions. The significance of the moderation interaction

effect was then tested.

To evaluate research questions #1a-b, a dummy variable was assigned to all participants,

with "O" indicating a male gender, and "1" indicating a female gender, and regressions were

employed to analyze the data. Following the guidelines for mediational analyses that have been

detailed in the literature (Frazier, Tix, & Baron, 2004), the variables for question la were entered

stepwise in a hierarchical regression.









CHAPTER 4
RESULTS

Order of Presentation Effects and Preliminary Analyses

Prior to the main component of the study being posted on the Internet, a pilot study was

run to determine if there were order-based effects of the above measures. If the order of the

measures had affected the responses of participants on the Ego Identity Process Questionnaire

(as the central measure in this study), that questionnaire would have been placed first in the

survey to prevent influence from other measures, and the remaining measures would have been

placed following it in a random order.

The pilot test resulted in two pools of data (n = 22 for Group 1, and n = 21 for Group 2).

To analyze the results for the two groups on scores of Ego Identity exploration and commitment,

independent samples t tests were employed to determine if any significant difference between the

two groups' means existed. For the individuals who completed the ego identity measure at the

beginning of the survey, exploration (M = 69.4, SD = 1 1.5) was not significantly different from

individuals who completed the measure at the end of the survey (M = 67.9, SD = 12.2), t(41) =

.40, p = .69 (two tailed). Similarly, for the individuals who completed the ego identity measure

at the beginning of the survey, commitment (M = 66.2, SD = 9.0) was not significantly different

from individuals who completed the measure at the end of the survey (M = 69.3, SD = 7. 1), t(41)

=-1.3, p = .20 (two tailed). Due to the lack of significant difference on ego identity scores in the

above pilot test, the ego identity measure was placed at the beginning of the survey for all future

participants. Data from the pilot study were incorporated into the main sample's data, and were

included in the above sample description.









All measures were assessed for internal consistency. Cronbach's alphas were determined

to be adequate for the current study, and were consistent with other studies using the same

measures; values ranged from .79 to .94, and are summarized with measure intercorrelations in

Table 4-1. None of the correlations met the generally accepted standard of being greater than .80

to signify that two variables are significantly collinear (Lewis-Beck, 1980), and thus analyses by

multiple regression remained plausible. Data were further checked for multivariate normality

through assessment of skewness and kurtosis. All variables' values for skewness and kurtosis

fell between -2 and 2, with the exception of SDO skewnesss = 1.8; kurtosis = 2.9) ATW

skewnesss = -2.5; kurtosis = 7.4), and Number of LG People Known skewnesss = -2.76, kurtosis

= 7.27). A square-root transformation was applied to the values for SDO, due to its values being

positively skewed. This resulted in acceptable levels of skewness (1.3) and kurtosis (1.0). Due

to the values for ATW and Number of Known LG People being negatively skewed, the values

were reflected (each score substracted from the highest score, producing a "mirror" of the

original data), and the square root of each value was taken. For ATW, this resulted in acceptable

levels of skewness (1.0) and kurtosis (1.5). It is important to note that inverting the ATW values

reversed the meaning of the scores, with lower values now indicating more tolerant attitudes

toward women. For the Number of LG People Known, square root transformations were not

effective at reducing the skewness and kurtosis, and nor were log-based transformations. This is

likely due to the participants' frequent endorsement of relatively high levels of known-LG

individuals. However, an inversion transformation (in which the minimum scale value was

increased to 1 to avoid division by zero, and each value of the variable was divided into 1)

resulted in marginally acceptable levels of skewness (-1.9) and kurtosis (2.1), while

approximately preserving the original pattern of value distributions of the original data.









Analyses


Hypothesis Tests

Hypothesis 1 predicted that participants who had higher scores on authoritarianism

(RWA) and social dominance orientation (SDO) would predict higher levels of commitment and

lower levels of exploration. To examine the influence of the predictor variables (RWA and

SDO) on the two factors of ego identity, those variables were entered in a single block in two

separate regressions, with commitment as the criterion variable in the first regression, and with

exploration as the criterion variable in the second regression. Social Desirability was also

entered in the equation as a predictor, to control for its effects.

The regression with commitment as the dependent variable was significant, F(3, 158) =

19.73, p < .001, explaining a moderate amount of the variance (R2 = .27, adj R2 = .26) (Cohen,

1988). The R2 and the adjusted R2 in this regression were almost identical, suggesting that the

adjustment in these cases was very small because a fairly large sample size was utilized. Tests of

standardized partial regression coefficients revealed that higher levels of RWA were

significantly associated with higher levels of commitment (P = .44; p < .001), but that SOD was

not significantly associated with commitment (P = .03; p = .68). Social Desirability was also

significantly related to commitment scores (P = .15; p <.05). The regression with exploration as

the dependent variable was similarly significant, F(2, 159) = 36.60, p < .001, explaining a

relatively large amount of the variance (R2 = .41, adj R2 = .40) (Cohen, 1988). Tests of

standardized partial regression coefficients revealed that higher levels of RWA were

significantly associated with lower levels of exploration (P = -.54; p < .001), and that higher

levels of SOD were also significantly associated with lower levels of exploration (P = -. 15;










p < .05). Social desirability was also significantly related to exploration scores (P = -.19;

p < .001). Hypothesis 1 was partially supported (Table 4-2).

Hypothesis 2 predicted that higher number of LG contact experiences and greater

closeness/intimacy in the contact experience with LG people would predict higher levels of

exploration. To examine the influence of the predictor variables (Number of Known LG People

and Closeness) on exploration, those variables were entered in a single block. Social Desirability

was also entered in the equation, to control for its effects on exploration. The overall regression

was significant, F(3, 158) = 21.14, p < .001, explaining a moderate amount of the variance

(R2 = .29, adj R2 = .27) (Cohen, 1988). Tests of standardized partial regression coefficients

revealed that knowing greater numbers of LG individuals was significantly associated with

higher levels of exploration (P = .29; p < .001), as was the degree of closeness in the

relationships with those individuals (P = .18; p < .05) (Table 4-3). Social Desirability was also

significantly related to exploration (P = -.31; p < .001).

Hypothesis 3 predicted that higher levels of Ego Identity Exploration will predict more

positive attitudes toward LG individuals, and lower levels will predict more negative attitudes

toward LG individuals. To examine the influence of the predictor variable (exploration) on the

criterion variable (Attitudes toward LG People), both variables were entered in a regression. The

overall regression was significant, F(2, 159) = 36.66, p < .001, explaining a moderate-to-large

amount of the variance (R2 = .32, adj R2 = .31) (Cohen, 1988). Tests of the standardized partial

regression coefficient revealed that a higher level of ego identity exploration was significantly

associated with more tolerant attitudes (lower scores on the ATLG-S) toward LG people

(p = -.54; p < .001) (Table 4-4). Social Desirability was not significantly related to ATLG-S

scores (p = .05; p = .50).










Hypothesis 3a predicted that Ego Identity Exploration would moderate the relationship

between authoritarianism and LG attitudes; it was also predicted to moderate the relationship

between SDO and LG attitudes. However, when a regression was carried out using ATLG-S

scores as the criterion variable, and RWA and SDO were entered as predictor variables to gauge

their main effects, SDO was not a significant predictor of Attitudes toward LG individuals

(p = .03; p = .64). Due to this finding, no further analyses for this hypothesis were executed

using SDO as a predictor variable.

To test the hypothesis regarding the moderation of authoritarianism by exploration,

Frazier, Tix, and Baron's (2004) recommendation to use hierarchical multiple regression

procedures was followed. The variables were standardized to reduce multicollinearity between

the interaction term and the main effects when testing for moderator effects. In the first step of

the regression, social desirability was entered, in order to control for its effects throughout the

remainder of the steps. In the second step, the main effects (authoritarianism and exploration)

were entered. In the third step, the interaction (authoritarianism x exploration) was entered; at

this step, a significant change in R2 for the interaction term would indicate a significant

moderator effect. For the Einal step, all possible interactions were added to ascertain any further

effects (social desirability x authoritarianism, social desirability x exploration, social desirability

x authoritarianism x exploration). As shown in Table 4-5, there was a significant moderator

effect, indicating that exploration moderated the link of authoritarianism to attitudes toward LG

individuals. As indicated by the adjusted R2 for the regression equation, the main interaction

effects accounted for 57% of the variance in ATLG-S scores. The interaction term of

authoritarianism and exploration accounted for an additional 3% of the variance in ATLG-S

scores.









To explore patterns underlying this significant interaction effect, an Excel spreadsheet

program (at http://www.j eremydawson. co.uk/2-way standardised.xls) was used to plot the

interaction effects of authoritarianism and exploration, and to provide exact values according to

the moderator effects. In Figure 4-1, it may be seen that for individuals with low levels of

authoritarianism, there is little difference on ATLG-S scores between individuals with high

versus low levels of ego identity exploration. However, for individuals who endorsed high levels

of authoritarianism, negative or condemning attitudes toward LG individuals were moderated to

a significant degree by higher levels of ego identity exploration. In other words, ego identity

exploration appears to allow greater tolerance versus condemnation in attitudes toward LG

people, moderating the effects of even a high level of authoritarianism.

Additional Research Questions

As previously noted, the vast maj ority of research on LG attitudes has been conducted

using samples of college students. In addition to the above hypotheses, it was questioned

whether a sample that was older than a typical college-student-aged population would have

similar attitudes toward LG individuals. The first set of research questions (la and lb) were

based in the relationship between gender and LG attitudes.

Question la posited that males in the current sample would have more negative attitudes

toward LG individuals than will women; but, that this difference would be eliminated when

controlling for gender role values (measured by Attitudes toward Women ATW). In other

words, the relationship between gender and ATLG would be fully mediated by ATW scores. A

frequently used multi-step process for testing mediation was used(Baron & Kenny, 1986;

Frazier, Tix, & Barron, 2004; Holmbeck, 1997). In the first step, the predictor (e.g., social

support) should be significantly associated with the outcome (ego identity). In the second step,









the predictor should be significantly associated with the proposed mediator (self-concealment).

In the final step, in an equation in which the predictor and mediator are j ointly entered to predict

outcome, the mediator should be significantly associated with the outcome. Correspondingly, the

previously significant effects of the predictor should be non-significant or significantly reduced

once the role of the mediator is controlled. In this last step, it is also important to test the

significance of the mediated effect (Baron & Kenny, 1986; Frazier et al., 2004; Kenny, Kashy, &

Bolger, 1998).

In the first analysis step, recorded gender (0 = males, 1 = females) was entered in a

regression equation predicting scores on the Attitudes toward Lesbian and Gay Men, Short Form

(ATLG-S). Gender accounted for significant variability in ATLG-S scores (AR2 = .04, P = -.20,

p < .05). Thus, gender explained significant variance associated with ATLG-S scores and the

direction of this effect indicated that a male gender was significantly associated with less tolerant

attitudes toward LG individuals.

In the second step of testing mediation, the regression analysis indicated that gender also

accounted for significant variation in scores on the Attitudes toward Women (ATW) scale

(AR2 = .07, P = -.26, p < .01). The direction of the effect suggested that a male gender was

associated with less egalitarian views toward men's and women's gender roles. In the third step

of testing the mediated effect, both gender and ATW scores were entered in the regression

predicting ATLG-S scores. These results are summarized in Tables 6 and 7. Results revealed that

the mediator (ATW scores) was significantly associated with the outcome, while gender, initially

significantly associated with ATLG-S scores, ceased being a significant predictor. The results

indicated that ATW scores, indicative of gender role attitudes, mediated the relationship between

gender and attitudes toward lesbians and gay men. The mediated effect was evaluated for









statistical significance (Baron & Kenny, 1986; Kenny et al., 1998). The mediated effect was

statistically significant, Z >1.94, p < .05.

Research Question lb posited that less condemning attitudes toward lesbians and more

condemning attitudes toward gay men would be associated with a male gender; however, when

erotic values of LG sexuality are controlled, they would account for gender-based difference in

attitudes toward lesbians and gay men. Similar procedures to that used in research question la

were employed. In the first analysis step, recorded gender (0 = males, 1 = females) was entered

in a regression as a criterion variable, with Attitudes toward Gay Men (ATG) and Attitudes

toward Lesbians (ATL) scores as predictors. Neither predictor was significantly associated with

gender, and so analyses were not continued. In this light, the gender of the sexual minority

person in question does not appear to play a significant role in a heterosexual person's

developing attitudes toward lesbians or gay men. Rather, the fact that an individual is known to

be a sexual minority per se, regardless of gender, appear to be the important factor in attitude

development, as demonstrated by Research Question la.

Research Question 2 posited that Authoritarianism, Social Dominance Orientation, and

Church Attendance would significantly predict negative attitudes toward LG individuals (while

controlling for Social Desirability). To examine the influence of the predictor variables (RWA,

SDO, Frequency of Religious Service Attendence, and Social Desirability) on exploration, those

variables were entered in a single block in a multiple regression, with ATLG-S as the criterion

variable.

The overall regression was significant, F(4, 157) = 63.89, p < .001, explaining a large

amount of the variance (R2 = .62, adj R2 = .61) (Cohen, 1988). Tests of standardized partial

regression coefficients revealed that higher levels of RWA were significantly associated with









ATLG-S (P = .73; p < .001), but that SOD (P = .04; p = .50), Frequency of Religious Service

Attendance (P = .07; p = .20), and Social Desirability (P = -.01; p = .87) were not significantly

associated with ATLG-S. Maintaining these variables in the equation, all possible dummy-coded

race/ethnicity variables were entered in the equation (Native American, Asian/Asian American,

African American, Hispanic/Latino, Caucasian, Biracial, and Other). Again, the overall

regression was significant, F(11, 150) = 26.44, p < .001, explaining a large amount of the

variance (R2 = .66, adj R2 = .64) (Cohen, 1988). Tests of standardized partial regression

coefficients revealed that higher levels of RWA were significantly associated with higher levels

of commitment, but to a lesser degree than when not considering race (P = .66; p < .001). The

only racial category that was significant in predicting ATLG-S scores was an African American

identification (P = .22; p < .01). A Hispanic/Latino identification was marginally significant

(p = .15; p = .05). Thus, even when controlling for other relevant variables, including religion,

some ethnic/racial categories remained significant in predicting ATLG-S scores.











Table 4-1. Means, Standard Deviations, and Scale/Subscale Correlations
Min. Max. M SD 1 2 3 4 5 6 7 8
1. EIPQ Exploration -2.5 2.8 0.0 1 .87
2. EIPQ Commitment -3.5 1.7 0.0 1 -.57** .81
3. ATLG-S 9.0 45 16.8 9.5 -49** -.56** .94
4. RWA 16.0 99.0 43.5 18.0 .50** -.60** .78** .89
5. GEV 4.0 18.0 9.4 3.9 -.34** -.50** -.60** -.59** .79
6. LEV 4.0 20.0 12.2 4.5 -.19* .27** -.37** -.41** .48** .86
7. Social Desirability 0.0 13.0 4.5 3.2 .29** -.37** .25** .32** -.31** -.24** .79
8. SDO 2.83 7.48 3.9 1.2 .29** -.43** .43** .53** -.28** -. 15 .20** .92
9. ATW 1.0 7.8 3.1 1.4 .37** -.54** .64** .68** -.47** -.17* .18* .61** .91
EIPQ = Ego Identity Process Questionnaire; ATLG-S = Attitudes Toward Lesbians and Gay Men, Short Form; RWA: Right-Wing
Authoritarianism; GEV/LEV = Gay/Lesbian Erotic Value Questionnaire; SDO = Social Dominance Orientation; ATW: Attitudes
Toward Women. Cronbach's coeffieients alphas appear in bold italics on the diagonal.
Note: N= 162. p<.05. **p <.01.











Table 4-2. Summary of Multiple Regression Analysis for the Prediction of EIPQ Commitment and Exploration Scores by RWA
and SDO
Commitment Exploration
Predictor B SEb P t B SEb P t
RWA .03 .01 .44 5.36** -.03 .00 -.45 -6.11**
SDO .02 .07 .03 .32 -.13 .06 -.15 -2.10*
Social Desirability .05 .02 .15 2.06* -.06 .02 -.20 -3.10**
Note: Overall regression results for Commitment, F(3, 158) 19.73, p < .001, R2 .27; for Exploration, F(2, 159) = 36.60, p < .001,
R2 .41. RWA Right-Wing Authoritarianism, SDO Social Dominance Orientation; p < .05. ** p < .001.





Table 4-3. Summary of Multiple Regression Analysis for the Prediction of EIPQ Exploration
Scores by Number of Known LG Individuals and the level of Closeness in those
Relationships
Predictor B SEb P t
Number of LG people known 1.31 .33 .29 4.00**
Closeness .22 .09 .18 2.50*
Social Desirability -. 10 .02 -.31 -4.57**
Note: Overall regression results, F(3, 158)= 21.14, p < .001, R2 .29. p < .05. ** "p < .001


.









Table 4-4. Summary of Multiple Regression Analysis for the Prediction of ATLG-S Scores by
EIPQ Exploration scores


<001.


Predictor B SEb P t
Exploration -5.14 .67 -.54 -7.66*
Social Desirability .14 .21 .05 .69
Note: Overall regression results, F(2, 159)= 36.66, p < .001, R2 .32. p









Table 4-5. Moderating Effect of Ego Identity Exploration on the Relation between Authoritarianism and Attitudes toward Lesbians
and Gay Men
Step and Variable B p t Total R2 Adj. R2 R2 inc. F inc. df
Step 1
Social Desirability 2.38 .25 3.28* .06 .06 .06 10.73 1, 160
Step 2
Authoritari ani sm 6.66 .70 11.54** .63 .62 .57 120.42 2, 158
Exploration -1.43 -.15 -2.43*
Step 3
Authoritarianism X Exploration -1.43 -.20 -3.82** .66 .65 .03 14.65 1, 157
Step 4
Social Desirability X Authoritarianism .38 .04 .57 .67 .65 .01 .73 3, 154
Social Desirability X Exploration .87 .10 1.42
Social Desirability X Authoritarianism X -.09 -.02 -.25
Exploration
Note: Overall regression results, F(2, 159) = 36.66, p < .001, R2 = .32. *" p < .05, **' p < .001.










Table 4-6. Direct Effects Regression Analyses of Predictor Variables on Criterion Variables
(Gender, ATW Scores, and ATLG-S Scores)


Predictor

Gender
(0 = male, 1 = female)


Criterion

ATLG-S

ATW


B

-4.18

-.81


SE B


1.60

.24


Note: Nmale = 48, Neemale = 114, p < .05










Table 4-7. Regression Analyses of Attitudes Toward Women Mediating the Effects of Gender on
ATLG-S scores
Criterion Predictor B SEE BP

ATLG-S Gender -.79 1.30 -.04
(0 = male, 1 = female)
(Mediator)ATW .420 .42 .63*

* p <.01





































Low Authoritarianism High Authoritarianism
Low Ego Identity Exploration 9.24 24.44
High Ego Identity Exploration 10.26 19.70


Figure 4-1. Plot of significant Authoritarianism X Ego Identity Exploration interaction, with
plotted values for scores on the Attitudes toward Lesbian and Gays (ATLG-S)

30


25




I Low Ego Identity
S15 -I ...* Exploration
I I ~-- -'- -- High Ego Identity
10t -I ** Exploration


Low Authoritarianism


High Authoritarianism









CHAPTER 5
DISCUSSION

Hypotheses

Hypothesis 1, which predicted that participants who had higher scores on

authoritarianism (RWA) and social dominance orientation (SDO) would have higher levels of

ego identity commitment and lower levels of ego identity exploration, was partially supported.

RWA predicted higher levels of commitment, and lower levels of exploration; SDO predicted

lower levels of exploration, but did not significantly predict levels of commitment. This begins

to lend support to the concept that attitudes may help or hinder individual development, or vice

versa; more specifically, foreclosure of options (high commitment, low exploration) appears to

be related to authoritarian and/or socially dominant values. However, it is important to

remember that because this research is correlational in nature, it is uncertain whether more

authoritarian or socially dominant attitudes precede high commitment/low exploration of ego

identity development, or vice versa. It seems possible that one could embrace a conservative,

authoritarian, socially dominant ideology first, and when later confronted with others issues

germane to an authoritarian identity, adopt a viewpoint consistent with the larger body of

authoritarian views. However, it is also reasonable to conceive of an individual first developing

some beliefs and values consistent with an authoritarian viewpoint, and later Einding congruency

between her/his beliefs and the larger authoritarian group; from that point, the individual could

conceivably embrace other beliefs from the authoritarian group out of a feeling of identification

or solidarity. Lacking an experimental or longitudinal design, this study cannot begin to

postulate whether the "chicken or the egg" came first, but it is clear that identity development

and authoritarian/social dominance-based attitudes are related. To carry the finding one step

further and to place it more squarely in the language of Herek' s (1984) theory, ego-based









attitudinal functions (including the defensive functions) and symbolic attitudinal functions are

related to one another and may influence each other.

Hypothesis 2, which postulated that greater numbers of LG contact experiences and

greater intimacy in those contact experiences would predict higher levels of ego identity

exploration, was fully supported. Thus, not only do LG contact experiences and greater intimacy

therein have a relation to more tolerant attitudes toward LG individuals (Herek & Capitanio,

1996; Liang & Alimo, 2005), but also are related to greater levels of ego identity exploration.

Again, since this research is correlational, one cannot reasonable suggest that LG-based

acquaintances promote higher levels of exploration, or vice versa. However, it does seem

reasonable to imagine that the ego identity exploration, as inspired through interaction with an

LG person, could contribute to higher levels of exploration in certain domains of ego identity

(i.e., gender roles, relationships). Alternately, if ego identities that are high in exploration levels

indicate a certain type of personality (Schwartz, 2001; van Hoof, 1999), (e.g., an "open-minded"

or "open-to-experience" type of personality), then that personality might be more amenable to

increased exposure to other individuals' beliefs, values, and ways of living. As with the previous

hypothesis, it is difficult to imagine an experimental or longitudinal study that might detangle

causation from effect in this scenario.

At the core of this study, hypotheses 3 and 3a explored the role of ego identity

development in relation to attitudes toward LG individuals. Namely, the theoretical concept to

be tested stated that greater authoritarian/socially dominant attitudes would be related to

condemning LG attitudes, but that a more developed ego (namely, with greater levels of ego

identity exploration) would temper such condemning beliefs. To begin to test this concept,

Hypothesis 3 predicted that higher levels of ego identity exploration would predict more positive









attitudes toward LG individuals, and that lower levels of ego identity exploration would predict

more negative attitudes toward LG individuals. This hypothesis was supported. Confirming

Herek' s (1984) previous untested theory that one' s ego, namely the defenses embodied therein,

could affect attitudes toward LG individuals, this study found that a more developed ego identity

was related to greater tolerance in attitudes toward LG individuals. This is interesting because

although many domains of identity constitute the concept of ego identity (e.g., career, politics,

religion, sex roles), the overall construct was still significantly related to LG-based attitudes.

Schwartz (2001) and van Hoof (1999) both suggested that ego identity may be an indicator of

overall personality, perhaps having some relation to the "Big Five" construct of "Opennness to

Experience" in this scenario (McCrae & Costa, 1987). Greater flexibility in one's ego identity

appears to allow greater flexibility and adaptability in encountering differences throughout the

lifespan specifically in this scenario, greater flexibility and tolerance of others who are

different from one's self.

Tying all previous hypotheses together, hypothesis 3a suggested that ego identity

exploration would moderate the relationship between authoritarianism and LG attitudes, and that

it would also moderate the relationship between social dominance orientation and LG attitudes.

However, since SDO was not related to attitudes toward LG individuals, it was not included in

the analyses for this hypothesis. Apart from this deletion, the hypothesis was supported.

In order to understand this outcome, it will be important to clarify the nature of statistical

moderation. According to Frazier, Tix, and Baron (2004), questions involving moderators

address "when" or "for whom" a variable most strongly predicts or causes an outcome variable.

More specifically, a moderator is a variable that alters the direction or strength of the relation

between a predictor and an outcome (Baron & Kenny, 1986; Holmbeck, 1997; James & Brett,










1984). Thus, a moderator effect is basically an interaction whereby the effect of one variable

depends on the level of another. In other words, researchers are interested in moderation when

the relation between predictor (here, authoritarianism) and outcome (attitudes toward LG

individuals) variables are stronger for some people (i.e., those with lower levels of exploration)

than for others (i.e., those with higher levels of exploration).

When considering individuals with low levels of authoritarianism, there is little

difference in ATLG-S scores between individuals with high or low levels of ego identity

exploration. However, when considering individuals with higher levels of authoritarianism,

condemning attitudes toward LG individuals were moderated to a significant degree by higher

levels of ego identity exploration. This finding suggests that the degree to which one has

explored various domains of her or his life (perhaps demonstrating a degree of intrapersonal

adaptivity/flexibility) appears to allow greater tolerance in attitudes toward LG people, even as

levels of authoritarian values and attitudes become higher.

In essence, a better developed ego identity, specifically one consisting of higher levels of

ego identity exploration, implies less of a need to defensively proj ect negative reactions about a

gay or lesbian sexual orientation in the vehicle of condemning attitudes toward such individuals.

This would likely be due to a person being more secure through having already explored

identity-based options for her/himself. Herek's (1984) theory that attitudes may serve a variety

of functions (in this case, symbolic and defensive), and that the functions may interact with and

effect one another in determining a person's Einal attitude structure, is supported.

Research Questions

Apart from the above hypotheses, one of the aims of the current study was to explore

whether previous findings in the literature regarding attitudes toward LG individuals, based










largely in samples of college students, could be replicated in an adult-aged sample. As was

demonstrated by this study's results, only some of the results from previous studies generalized

to the present sample.

Research question la suggested that, as in previous studies based in college students

(Kite & Whitley, 1996, 1998), males would have more negative attitudes toward LG individuals

than would females, but that this difference would be eliminated (fully mediated) when

controlling for gender role values as measured by ATW. The current Eindings demonstrate that

this statement is true for adults as it is for college-aged individuals. As suggested by Kite

(1994), non-LG individuals' evaluations of gay men and lesbians are at least partly grounded in a

larger belief system concerning gender roles. If one has physical or behavioral traits that might

traditionally be labeled "masculine," then according to traditional gender role beliefs and values,

stereotypically masculine behaviors and characteristics should be a natural extension of that

individual's gender role. As previously reported, the literature on gender roles suggests that

male gender roles are more narrowly defined and inflexible than are those for women (Bem,

1993), and violating male gender roles is more obviously offensive than violation of female roles

(Herek, 1988). This study's finding clarifies that for adults as well as college student-aged

individuals, one's gender does not affect attitudes toward LG individuals, despite men typically

endorsing more condemning attitudes. Rather, gender role attitudes and beliefs, which may be

more inflexibly espoused by males than by females, are the deciding factor in formation of

attitudes toward LG people.

Carrying the above idea forward, and in line with previous findings in the literature

(DeLamater, 1987; Herek, 2002; Herek & Capitanio, 1999; Louderback & Whitley, 1997),

research question lb suggested that a male gender in the current sample would be related to less









condemning attitudes toward lesbians and more condemning attitudes toward gay men. Further,

and still in line with previous findings (Louderback & Whitley, 1997), the question postulated

that the above attitudinal differences on the part of heterosexual males in the sample would be

eliminated when erotic values of LG sexuality are controlled. However, when the above

question was tested using the current data, a male gender was not significantly related at all to

attitudes toward gay men or lesbians (considered separately). This suggests that attitudes toward

LG people function as a single construct in the current sample, and the gender of the LG person

is not significant in the formation of tolerant/intolerant attitudes. As demonstrated in research

question la, it is an overall violation of gender roles, rather than a gender (or an obj ect of

attitudes, as suggested by this question) that is important in the current sample when regarding

attitudes toward LG people.

Research Question 2 tested the suggestion that Authoritarianism, Social Dominance

Orientation, and Religious Service Attendance would significantly predict negative attitudes

toward LG individuals. Research question 2a posited that when controlling for the above

variables, race would not be a significant predictor of attitudes toward LG attitudes.

Interestingly, only authoritarianism among the variables listed above was significant in

predicting attitudes toward LG individuals. This is similar to previous findings in the literature

which suggest that authoritarianism is a motivating force behind some religious-based LG

prejudice (i.e., such as that dealing with fundamentalist religious belief, as in Laythe, Finkel, &

Kirkpatrick, 2001), and with prejudice again LG people in general (Altemeyer, 1981, 1998,

2001; Ekehammar, Akrami, Gylje, & Zakrisson, 2004; Laythe, Finkel, & Kirkpatrick,

2001;Whitley, 1999; Whitley & Lee 2000); however, it is contrary to other findings which

specifically found religious service attendance to predict an individual's level of tolerance









toward LG people (Schulte & Battle, 2004). This may be due to the latter study's focus on

African American individuals, whereas the current study included participants of various races

and ethnicities.

Lemelle and Battle (2004) reported that if one controls for religious service attendance,

ethnicity-based differences in tolerance levels toward LG people are eliminated. Analyses for

research question 2a, though, revealed that even when controlling for the influence of religious

service attendance, an African American racial identification remained a significant predictor of

attitudes toward LG people. While not technically significant, identifying oneself as Hispanic

was also nearly significant in this regard (p = .05). In attempting to understand this difference

and postulating that it may reside in religious differences among ethnicities/races, it is important

to note that this study did not regard all possible religious-based variables that have been used in

previous studies, such as fundamentalism or intrinsic versus extrinsic religiosity (Herek, 1987).

As such, the motivating religious factors behind racial differences in attitudes toward LG

individuals is unknown. It may be that different races'/ethnicities' understandings of appropriate

gender roles (i.e., some Hispanic cultures' "machismo" and "Marianisma") are a possible

alternative key to understanding the above differences.

Limitations and Suggestions for Further Research

A main limitation of the current study is relatively small cell sizes among some

racial/ethnic identifications (i.e., Hawaiian/Pacific Islander), thus preventing complete

understanding of the role of race/ethnicity in attitudes toward LG people. Although previous

studies suggest that when controlling for religious factors, there are no difference among some

races/ethnicities, the current study did not replicate those findings. It may be useful to obtain

samples composed entirely of a particular racial/ethnic group, and to employ more variables that










may be theoretically relevant to that group's cultural beliefs and values. The difficulty of this

undertaking may be formidable, though, as the effort required to accumulate even moderate

diversity in the current sample was significant. Despite such difficulties, gaining further

understanding of various races and ethnicities attitudes toward LG people is important,

especially as LG individuals continue to openly disclose their sexual/affectional orientations and

lobby for greater civil rights. A future study could replicate the current study using a larger

sample, or a sample with greater diversity, possibly allowing for such an understanding.

Additionally, noting that the current sample produced significant skewness and kurtosis on some

of the variables under study, greater effort could be put forth to gather a more "well rounded"

sample with more normally distributed scores.

Another important limitation is that this is s correlational study. The causal direction of

the statistically significant findings may only be postulated, and even then, only weakly so. For

example, in the test of Hypothesis 1, authoritarianism predicted higher levels of commitment,

and lower levels of exploration. It is empirically unjustified to suggest that a conservative or

authoritarian attitude is detrimental to aspects of psychological development. Equally unjustified

is the suggestion that all individuals with healthy psychological development will develop and

open-minded and tolerant stance toward LG individuals. Although the current study

demonstrates a clear relationship between the two factors, causation or a directional relationship

between them cannot be drawn from the current data. An experimental design, or perhaps a

longitudinal study that tracks individuals through a college career, might be useful in clarifying

this path.

Another conceivable limitation is the use of an Internet-based sample to the exclusion of

a more random sampling method. It is possible that some unexpected findings in the current










study are due to sampling bias generated through a limited arena for data collection. It is also

possible that a greater degree of racial/ethnic diversity could have been obtained through more

direct recruitment from community-based organizations. However, the empirical grounding for

such possibilities is not altogether clear. A significant study which compared Internet versus

paper-and-pencil sampling methods found that both means of data collection resulted in

approximately equivalent ratios of SES groupings, racial groupings, and response patterns

(Gosling, Vazire, Srivastava, & John, 2005). Nonetheless, the results of a future study that

replicates the current study, but using more varied methods of data collection, would be

interesting.

Other future work might explore the unexpected Eindings from the current study. For

example, although previous work indicated that heterosexual men would hold more condemning

attitudes toward gay men than toward lesbians, this study's results did not replicate that Einding.

It seems possible that greater exposure of the general population to gay men and lesbians over

the past decade may have eliminated differences in attitudes toward them. However, proof that

the current findings are not anomalous would be necessary before that hypothesis, or other

similar hypotheses, could be put forward.

Implications for Psychotherapy and Counseling

As prejudice against LG people sometimes leads to limited civil rights and violence

against that population, the present study has important implications for psychotherapy,

especially that geared toward ameliorating social injustices and decreasing the intolerance of

individuals who are intolerant of LG people. At the center of this study's findings is that ego

identity development, specifically the exploratory aspect of such development, is related to

greater tolerance toward LG individuals. Even more interestingly, such ego identity exploration










tempers or moderates the prejudicial aspects of an authoritarian personality. Thus, if an

individual expresses negative views toward LG individuals, psychotherapists may promote social

justice for the LG community not necessarily by confronting those views directly, but rather by

promoting greater exploration of life issues in general. This exploration appears to promote

greater tolerance toward LG people in an indirect way.

Additionally, psychoeducation for those in the LG community might focus on promoting

tolerance not only by bringing direct conflict against authoritarian political or religious groups,

but by allowing LG people to be known more personally to these groups. This study

demonstrates that such interpersonal knowledge promotes tolerance in and of itself. If it is true

that such exposure promotes ego identity exploration, then this tolerance will be somewhat stable

even in the face of the most rigorous authoritarian stance. In this way, the LG community may

be its own agent of social justice. Both this idea, as well as the concept of engendering ego

identity exploration in non-LG individuals, could be useful in planning outreach proj ects in

community centers or other mental health centers.

In summary, the fact that ego identity, most specifically the exploration aspect of ego

identity, is related to tolerance of LG people is a new finding in the literature on both attitudes

toward LG individuals and on ego identity. While the correlational nature of this research limits

one's ability to drawn causal implications from the study, this study does demonstrate a clear

relation among various socially based attitudes and one's own ego identity development. Most

interestingly, a more mature, well explored sense of self is related to one' s ability to tolerate LG

people, even in the context of an authoritarian-type personality or set of attitudes. As this type of

personality is typically more condemning of societal difference in general, and definitely toward

LG people, the fact that a more mature ego can promote tolerance in even such an intrapsychic









context is encouraging. In addition to providing empirical support for Herek' s (1984) theory of

LG prejudice, this study provides empirical grounding for psychotherapeutic and outreach-based

interventions aimed at reducing intolerance, and promoting social justice for LG individuals and

their community.









APPENDIX A
DEMOGRAPHIC QUENTIONNAIRE

(1) What is your current age?

(2) What is your race/ethnicity?
American Indian/Alaska Native
Asian/Asian American
Black/African American
Hispanic/Latino
Native Hawaiian or Other Pacific Islander
White/European American
Biracial/Multiracial
Other (Please specify):

(3) Where do you currently live?
United States (East Coast)
United States (West Coast)
United States (Northern Midwest)
United States (Southern Midwest)
United States (Southern)
United States (Alaska)
United States (Hawaii)
Canada
Mexico
Central America
South American
Europe
Asia
Africa
Australia
Other (Please Specify):

(4) In what type of town or city do you primarily live?
Urban
Suburban
Rural (population of less than 2500)

(5) What is your current yearly personal gross income?
Under $4,999
Between $5,000 and $14,999
Between $15,000 and $24,999
Between $25,000 and $34,999
Between $35,000 and $49,999
Between $50,000 and $74,999
Between $75,000 and $99,999









Between $100,000 and $149,999
More than $150,000

(6) What is your highest level of education?
Some High School
High School Graduate/Equivalent
Vocational Training and/or Certificate
Some College/University (no degree)
Associate's Degree(s)
Bachelor' sDegree(s)
Some Graduate Study (no graduate degree)
Master' sDegree(s)
Doctorate and/or Postgraduate Degree(s)
Other (please specify):

(7) What is your gender?
male
female
other (please specify)










APPENDIX B
THE ATTITUDES TOWARD LESBIANS AND GAY MEN SCALE

Below is a list of statements different people have made about men who are gay (homosexual),
or women who are lesbian (homosexual). Please mark a number from 1 ("strongly disagree") to
5 ("strongly agree") to describe your degree of agreement with each item.

1 = Strongly disagree
2 = Disagree somewhat
3 = Can't decide
4 = Agree somewhat
5 = Strongly Agree

(1) Lesbians just can't fit into our society.

(2) State laws regulating private, consenting lesbian behavior should be loosened.*

(3) Female homosexuality is a sin.

(4) Female homosexuality in itself is no problem, but what society makes of it can be a
problem.*

(5) Lesbians are sick.

(6) I think male homosexuals are disgusting.

(7) Male homosexuality is a perversion.

(8) Just as in other species, male homosexuality is a natural expression of sexuality in human
men.*

(9) Homosexual behavior between two men is just plain wrong.

(10) Male homosexuality is merely a different kind of lifestyle that should not be condemned.*

(11) How many times have you personally encountered a person (i.e., friend, acquaintance,
relative, distant relative) who you knew was a lesbian or gay man?




4 or 5
6 to 9
10 or more










(12) If you have known a lesbian or gay man, how would you rate the relationship in terms of
personal closeness?
I've never known a lesbian or gay man
very distant, not well known
somewhat distant, not very well known
somewhat close, somewhat well known
very close, very well known

(13) How frequently do you attend a religious service (i.e., church, temple, mosque, or other)
never
once or twice a year
once or twice a month
once a week
more than once a week

(14) On the following scale, with "O" indicating "the opposite sex, and "6" indicating "the same
sex," how would you describe the people to whom you are sexually attracted?
Always the opposite sex from me
Almost always the opposite sex, very occasionally the same sex
Sometimes the opposite sex, occasionally the same sex
About equally the opposite sex and the same sex
Sometimes the same sex, occasionally the opposite sex
Almost always the opposite sex, very occasionally the same sex
Always the same sex as me

(15) How would you describe your sexual orientation?
heterosexual
gay
lesbian
bisexual


















*These items are reverse scored.









APPENDIX C
THE RIGHT-WING AUTHORITARIANISM SCALE, SHORT FORM

Below is a list of questions concerning a variety of social issues. You will probably find that you
agree with some of the statements, and disagree with others, to varying extents. Please indicate
your reaction to each statement by checking the answers according to the following scale:

1 = Strongly disagree
2 = Moderately disagree
3 = Slightly disagree
4 = Neutral
5 = Slightly agree
6 = Moderately agree
7 = Strongly agree

(1) Our country needs a powerful leader, in order to destroy the radical and immoral currents
prevailing in society today.

(2) Our country needs free thinkers, who will have the courage to stand up against traditional
ways, even if this upsets many people.*

(3) The "old-fashioned ways" and "old-fashioned values" still show the best way to live.

(4) Our society would be better off if we showed tolerance and understanding for untraditional
values and opinions.*

(5) God's laws about abortion, pornography and marriage must be strictly followed before it is
too late; violations must be punished.

(6) The society needs to show openness towards people thinking differently, rather than a strong
leader; the world is not particularly evil or dangerous.*

(7) It would be best if newspapers were censored so that people would not be able to get hold of
destructive and disgusting material.

(8) Many good people challenge the state, criticize the church, and ignore "the normal way of
living."*

(9) Our forefathers ought to be honored more for the way they have built our society, and at the
same time we ought to put an end to those forces destroying it.

(10) People ought to pay less attention to the Bible and religion, and instead they ought to
develop their own moral standards.*

(11) There are many radical, immoral people trying to ruin things; the society ought to stop them.










(12) It is better to accept bad literature than to censor it.*

(13) Facts show that we have to be harder against crime and sexual immorality, in order to
uphold law and order.

(14) The situation in the society of today would be improved if troublemakers were treated with
reason and humanity.*

(15) If society so wants, it is the duty of every true citizen to help eliminate the evil that poisons
our country from within.








































*These items are reverse scored.









APPENDIX D
THE SOCIAL DOMINANCE ORIENTATION SCALE, ABBREVIATED VERSION

Which of the following obj ects or statements do you have a positive or negative feeling towards?
Beside each obj ect or statement, mark a number from '1' to '7' which represents the degree of
your positive or negative feeling.

1 = Very positive
2 = Positive
3 = Slightly positive
4 = Neither positive nor negative
5 = Slightly negative
6 = Negative
7 = Very negative

(1) Some groups of people are simply inferior to other groups.
(2) It' s OK if some groups have more of a chance in life than others.
(3) To get ahead in life, it is sometimes necessary to step on other groups.
(4) Inferior groups should stay in their place.
(5) Group equality should be our ideal.*
(6) We should do what we can to equalize conditions for different groups.*
(7) Increased social equality.*
(8) We would have fewer problems if we treated people more equally.*



















*These items are reverse scored.










APPENDIX E
THE ATTITUDES TOWARD WO1VEN SCALE

Please mark a number from 1 ("Disagree Strongly") to 4 ("Agree Strongly") to describe your
degree of agreement with each item. (Higher scores indicate more egalitarian attitudes)

1 = Agree strongly
2 = Agree somewhat
3 = Neutral
4 = Disagree somewhat
5 = Disagree strongly

(1) Swearing and obscenity are more repulsive in the speech of a woman than a man.

(2) Under modern economic conditions, with women active outside the home, men should share
in household tasks such as washing dishes and doing laundry.*

(3) It is insulting to women to have the "obey" clause still in the marriage service.*

(4) A woman should be as free as a man to propose marriage.*

(5) Women should worry less about their rights and more about becoming good wives and
mothers.

(6) Women earning as much as their dates should bear equally the expense when they go out
together.*"

(7) Women should assume their rightful place in business and all the professions along with
men.*

(8) A woman should not expect to go to exactly the same places or to have quite the same
freedom of action as a man.

(9) Sons in a family should be given more encouragement to go to college than daughters.

(10) It is ridiculous for a woman to run a locomotive and for a man to darn socks.

(11) In general, the father should have greater authority than the mother in the bringing up of
children.

(12) The intellectual leadership of a community should be largely in the hands of men.

(13) Economic and social freedom is worth far more to women than acceptance of the ideal of
femininity, which has been set up by men.*










(14) There are many jobs in which men should be given preference over women in being hired or
promoted.

(15) Women should be given equal opportunity with men for apprenticeship in the various
trades.*

















































*These items are reverse scored









APPENDIX F
EGO IDENTITY PROCESS QUESTIONNAIRE

Listed below are a number of statements describing different types of behavior. Please indicate
how you feel about each statement.

1 = Strongly disagree
2 = Disagree
3 = Slightly disagree
4 = Slightly agree
5 = Agree
6 = Strongly agree

(1) I have definitely decided on the occupation that I want to pursue.

(2) I don't expect to change my political principles and ideals.

(3) I have considered adopting different kinds of religious beliefs.

(4) There has never been a need to question my values.*

(5) I am very confident about what kinds of friends are best for me.

(6) My ideas about men's and women' s roles have never changed as I became older.*

(7) I will always vote for the same political party.

(8) I have firmly held views concerning my role in my family.

(9) I have engaged in several discussions concerning behaviors involved in dating relationships.

(10) I have considered different political views thoughtfully.

(1 1) I have never questioned my views concerning what kind of friend is best for me.*

(12) My values are likely to change in the future.*

(13) When I talk to people about religion, I make sure to voice my opinion.

(14) I am not sure about what type of dating relationship is best for me.*

(15) I have not felt the need to reflect upon the importance I place on my family.*

(16) Regarding religion, my beliefs are likely to change in the near future.*

(17) I have definite views regarding the ways in which men and women should behave.










(18) I have tried to learn about different occupational Hields to Eind the best one for me.

(19) I have undergone several experiences that made me change my views on men's and
women's roles.

(20) I have consistently re-examined many different values in order to Eind the ones which are
best for me.

(21) I think what I look for in a friend could change in the future.*

(22) I have questioned what kind of date is right for me.

(23) I am unlikely to alter my vocational goals.

(24) I have evaluated many ways in which I fit into my family structure.

(25) My ideas about men's and women's roles will never change.

(26) I have never questioned my political beliefs.*

(27) I have had many experiences that led me to review the qualities that I would like my friends
to have.

(28) I have discussed religious matters with a number of people who believe differently than I
do.

(29) I am not sure that the values I hold are right for me.*

(30) I have never questioned my occupational aspirations.*

(31) The extent to which I value my family is likely to change in the future.*

(32) My beliefs about dating are firmly held.












*These items are reverse scored.









APPENDIX G
MEASUREMENT OF EROTIC VALUE OF LESBIAN AND GAY SEX

Below is a list of statements different people have made about men who are gay (homosexual),
or women who are lesbian (homosexual). Please mark a number from 1 ("very strongly
disagree") to 9 ("very strongly agree") to describe your degree of agreement with each item.

1 = Very strongly disagree
2 = Strongly disagree
3 = Moderately disagree
4 = Slightly disagree
5 = Neutral
6 = Slightly agree
7 = Moderately agree
8 = Strongly agree
9 = Very strongly agree

(1) I Eind the idea of a man making love to another man erotic.

(2) I Eind the idea of a man making love to another man repulsive.*

(3) I think that I would be sexually aroused by watching two men make love.

(4) I have viewed pornographic materials involving male homosexual acts.

(5) I Eind the idea of a woman making love to another woman erotic.

(6) I Eind the idea of a woman making love to another woman repulsive.*

(7) I think that I would be sexually aroused by watching two women make love.

(8) I have viewed pornographic materials involving lesbian acts.













*These items are reverse scored.









APPENDIX H
MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE, SHORT FORM

Please indicate either "true" or "false" to describe your degree of agreement with each item.

(1) It is sometimes hard for me to go on with my work if I am not encouraged.
(2) I sometimes feel resentful when I don't get my way.
(3) On a few occasions, I have given up doing something because I thought too little of my
ability.

(4) There have been times when I felt like rebelling against people in authority even though I
knew they were right.

(5) No matter who I'm talking to, I'm always a good listener.
(6) There have been occasions when I took advantage of someone.
(7) I'm always willing to admit it when I make a mistake.
(8) I sometimes try to get even rather than forgive and forget.
(9) I am always courteous, even to people who are disagreeable.
(10) I have never been irked when people expressed ideas very different from my own.
(1 1) There have been times when I was quite j ealous of the good fortune of others.
(12) I am sometimes irritated by people who ask favors of me.
(13) I have never deliberately said something that hurt someone's feelings.









APPENDIX I
EXAMPLE ADVERTISEMENT FOR PARTICIPANT RECRUITMENT


I am a psychology researcher at the University of Florida, and am conducting a study on

how various people respond to important social issues. The input of African American

participants is often neglected by researchers, and the resulting research doesn't accurately reflect

how all people in this country feel or think.

If you are an African American person and would like to contribute to the study, it may

be accessed at http://survey .psych.ufl.edu/web social. The survey takes approximately 20 minutes

to complete, and the study has been approved by the University of Florida's Institutional Review

Board.

Once you click the consent link at the bottom of the web page above, the survey begins.

If you have questions or concerns, please feel free to contact me at the e-mail address above, or

using information found at the survey link above.

Thank you to anyone who is willing to participate. Please feel free to forward the link to

others who might be willing to participate as well.



Daniel Potoczniak, M. S.Ed.

Ph.D. Candidate, Counseling Psychology Program

University of Florida










APPENDIX J
INFORMED CONSENT

Informed Consent
Protocol Title: The Influence of Ego Identity on Heterosexual Individuals' Attitudes toward Lesbian
and Gay Individuals

Please read this consent document carefully before you decide to participate in this study.

Purpose of the research study:
The purpose of this study is to investigate how individuals regard different types of people, such as
women versus men, or lesbians versus gay men. An indicator of human development and of various
aspects of personality will be used to examine how various attitudes might be explained. As such,
this study involves research.

What you will be asked to do in this study:
You will be asked to complete a series of 102 questions concerning how you feel about a number of
topics, such as your ideas on politics, religion, values, family, friends, and significant others, to name
a few. You do not have to answer any question you do not wish to answer. You will then be asked
to provide some information about yourself, such as your gender, age, race, sexual orientation, and
how attracted you are to people of another sex/gender, or of the same sex/gender. You will not be
asked for your name or any other information that could identify you.

Time required:
Approximately 20 minutes.

Risks and Benefits:
This study has been approved by the University of Florida Institutional Review Board (IRB). There
are no anticipated risks associated with this study, other than the potential minimal to mild
discomfort that could potentially arise as a result of being asked personal questions regarding the
subject areas described above. The risk of such discomfort should be minimal, if it exists at all.
You may directly benefit through increased self-awareness and self-understanding as you
consider appropriate responses in completing the survey. Broader benefits of this study to others
include a more comprehensive social understanding of how some types of beliefs or values affect
attitudes toward lesbians and gay men. Additionally, through possible publication of the results, this
study may inform the fields of personal counseling and career counseling.

Compensation:
You will not be compensated in any form for your participation in this study.

Confidentiality:
You will not be asked to provide any identifying information about yourself, and no record will be
kept of your participation in any way.

Voluntary participation:
Your participation in this study is completely voluntary. There is no penalty for not participating.










Right to withdraw from the study:
You have the right to withdraw from the study at any time without consequence.

Technical Requirements:
You may complete this survey on any computer with Internet access using any later version Internet
browser with javascript and cookies enabled. Your browser must also be able to display 128-bit
encrypted web pages (which most Internet browsers support). If you experience technical difficulties
accessing the survey, and you have ensured that javascript and cookies are enabled within your
Internet browser, please send an e-mail to danpot~uHl.edu to report your technical difficulties, and
try accessing the survey once again the following day. You may want to also try accessing the survey
from a different computer.

Handling of data collected in this survey:
* Anonymity of data: You will not be personally identifiable by the data you submit via this Internet-
based survey so you are encouraged to answer all items honestly and completely. The data gathered
from this study may be presented in a Einal research paper in I!.- re-.t. and/or individual response
formats, however you will not be personally identifiable by the data presented within the Einal study.

* The data you submit via this survey is kept secure and private. The collected survey data will be
stored on highly secure and Birewalled servers, which can only be accessed by the research
Investigator via password-protected protocols.

* Survey data submissions cannot be traced back to the original research participant, and although
the IP address of the computers on which this survey is accessed for system administration and
record keeping purposes, no connection will be made between you and your computer's IP address.

Whom to contact if you have questions about the study:
For questions concerning the study or to request a copy of the results, contact Daniel Potoczniak,
M.S.Ed., Graduate Student, Department of Psychology, University of Florida, P.O. Box 112250,
Gainesville, Florida, 32611. Phone: (352) 392-0601 x536, E-mail: danpot~ufl.edu; or, Greg
Neimeyer, Ph.D., Professor, at the same address as above, phone: (352) 392-0601, x257, E-mail:
Neimeyer~uHl.edu.

For questions or concerns about your rig-hts as a research participant. contact the UFIRB office,
P.O. Box 112250, University of Florida, Gainesville, FL 32611-2250. Phone: (352) 392-0433.

Agreement:
By completing the following survey, you agree that you have read the above information; that you
voluntarily agree to participate in this study; and that you have received a copy of this description
either by paper or electronically. Please click on the button below if you agree.










LIST OF REFERENCES


Adams, H.E., Wright, L.W., & Lohr, B.A. Is homophobia associated with homosexual arousal?
Journal of Abnornzal Psychology, 105(3), 440-445.

Allport, G. (1954). The nature ofprejudice. New York: Addison-Wesley.

Allport, G.W., & Ross, J.M. (1967). Personal religious orientation and prejudice. Journal of
Personality and Social Psychology, 5, 432-443.

Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg, Canada: University of Manitoba
Press.

Altemeyer, B. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press.

Altemeyer, B. (2001). Changes in attitudes toward homosexuals. Journal ofHontosexuality,
42(2), 63-75.

Altemeyer, B., & Hunsberger, B. (1992). Authoritarianism, religious fundamentalism, quest, and
prejudice. International Journal for the Psychology ofReligion, 2, 113-133.

American Psychiatric Association. (1980). Diagnostic and Statistical Ma'~nual of2~ental
Disorders, 3rdEdition. Washington, DC: The American Psychiatric Association.

Anderssen, N. (2002). Does contact with lesbians and gays lead to friendlier attitudes? A two
year longitudinal study. Journal of Conanunity & Applied Social Psychology, 12, 124-
136.

Balistreri, E., Busch-Rossnagel, N.A., & Geisinger, K.F. (1995). Development and preliminary
validation of the Ego Identity Process Questionnaire. Journal ofAdolescence, 18, 179-
192.

Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1 173-1 182.

Battle, J., & Lemelle, A.J. (2002). Gender differences in African American attitudes toward gay
males. The Western Journal ofBlack Studies, 26(3), 134-139.

Bem, S.L. (1993). The lenses of gender: Transforming the debate on sexual inequality. New
Haven, CT: Yale University Press.

Biernat, M. (2003). Toward a broader view of social stereotyping. American Psychologist, 58,
1019-1027.










Boswell, J. (1980). Christianity, Social Tolerance, and Homosexuality. Chicago: The University
of Chicago Press.

Broderick, R.C., & Broderick, V. (1990). The Catholic Encyclopedia. Nashville, TN: Nelson
Reference.

Chauncey, G. (1994). Gay New York: Gender, Uhban Culture, and the Making of the Gay Male
World, 1890 -1940. New York: Basic Books.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:
Lawrence Earlbaum Associates.

Cohen, J. (1992). Statistical power analysis. Current Directions in Psychological Science, 1, 98-
101.

DeLamater, J. (1987). Gender differences in sexual scenarios. In K. Kelley (Ed.), Females, males
and sexuality: Theories and research (pp. 127-139). Albany: Status University of New
York Press.

Donahue, M.J. (1985). Intrinsic and extrinsic religiousness: Review and meta-analysis. Journal
ofPersonality and Social Psychology, 48, 400-419.

Ekehammar, B., Akrami, N., Gylj e, M., & Zakrisson, I. (2004). What matters most to prejudice:
Big five personality, social dominance orientation, or right-wing authoritarianism?
European Journal ofPersonality, 18, 463-482.

Erdfelder, E., Faul, F., & Buchner, A. (1996). GPOWER: A general power analysis program.
Behavioral Research M~ethods, hIstruntents & Computers, 28(1), 1-1 1.

Erikson, E.H. (1968). Identity: Youth and Crisis. New York: Norton.

Falkenberg, S.D., Hindman, C.D., & Masey, D. (1983, March). Measuring Attitudes Toward
Males in Society. Paper presented at the Southeastern Psychological Association annual
meeting, Atlanta, GA.

Frazier, P. A., Tix, A. P. & Barron, K. E. (2004). Testing Moderator and Mediator Effects in
Counseling Psychology Research. Journal of Counseling Psychology, 51, 115-134.

Freud, S. (1951). Historical notes: A letter from Freud. The American Journal ofPsychiatry,
107(10), 786-787.

Gosling, S.D., Vazire, S., Srivastava,S. John, O.P. (2005). Should we trust web-based studies? A
comparative analysis of six preconceptions about internet questionnaires. American
Psychologist, 59(2), 93-104.










Hahn, E.D. (1993). Sex-role Attitude Change and the Attitudes toward Women Scale.
Unpublished masters thesis. The University of Texas at Austin, Austin, TX.

Heaven, P. C. L., & Bucci, S. (2001). Right-wing authoritarianism, social dominance orientation
and personality: An analysis using the IPIP measure. European Journal ofPersonality,
15, 49-56.

Herek, G.M. (1984). Beyond "Homophobia": A social psychological perspective on attitudes
toward lesbians and gay men. Journal ofHomosexuality, 10(1-2), 1-21.

Herek, G.M. (1987). Religious orientation and prejudice: A comparison of racial and sexual
attitudes. Personality and Social Psychology Bulletin, 13(1), 34-44.

Herek, G.M. (1988). Heterosexuals' attitudes toward lesbians and gay men: Correlates and
gender differences. Journal of Sex Research, 25, 451-477.

Herek, G.M. (1994). Assessing attitudes toward lesbians and gay men: A review of empirical
research with the ATLG scale. In B. Greene & G.M. Herek (Eds.), Lesbian and gayggg~~~~~ggg
psychology (pp. 206-228). Thousand Oaks, CA: Sage.

Herek, G.M. (2000). The psychology of sexual prejudice. Current Directions in Psychological
Science, 9(1), 19-22.

Herek, G.M. (2000). Sexual prejudice and gender: Do heterosexuals' attitudes toward lesbians
and gay men differ? Journal ofSociallIssues, 56(2), 251-266.

Herek, G.M. (2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion
Quarterly, 66, 40-66.

Herek, G.M., & Capitanio, J. (1996). "Some of my best friends": Intergroup contact, concealable
stigma, and heterosexuals' attitudes toward gay men and lesbians. Personality and Social
Psychology Bulletin, 22(4), 412-424.

Herek, G.M., & Capitanio, J. (1999). Sex differences in how heterosexual think about lesbians
and gay men: Evidence from survey context effects. The Journal of Sex Research, 36(4),
348-360.

Herek, G.M., Gillis, J., & Cogan, J. (1999). Psychological sequelae of hate crime victimization
among lesbian, gay and bisexual adults. Journal of Consulting and Clinical Psychology,
67, 945-951.

Hooker, E. (1957). The adjustment of the male overt homosexual. Journal ofProjective
Techniques, 21, 18-3 1.

Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly,
24(2), 163-204.










Kite, M.E. (1994). When perceptions meet reality: Individual differences in reactions to lesbians
and gay men. In B.Greene & G.M. Herek, Lesbian and gay gg~~~~~ggggpsychology : Theory,
research, and clinical applications, (pp 25-53). Thousand Oaks, CA: Sage Publications.

Kite, M.E., & Whitley, B.E., Jr. (1996). Sex differences in attitudes toward homosexual persons,
behavior, and civil rights: A meta-analysis. Personality and Social Psychology Bulletin,
22, 336-353.

Kite, M.E., & Whitley, B.E., Jr. (1998). Do heterosexual women and men differ in their attitudes
toward homosexuality? A conceptual and methodological analysis. In Herek, G.M. (Ed.),
Stigma and sexual orientation: Thiderstanding prejudice against lesbians, gayggg~~~~~gggg men, and
bisexuals (pp 39-61). Thousand Oaks, CA: Sage Publications.

Landen, M., & Innala, S. (2002). The effect of a biological explanation on attitudes towards
homosexual persons: A Swedish national sample study. Nordic Journal ofPsychiatry,
56(3), 181-186.

Laythe, B., Finkel, D., & Kirkpatrick, L.A. (2001). Predicting prejudice from religious
fundamentalism and right-wing authoritarianism: A multiple-regression approach.
Journal for the Scientific Study ofReligion, 40(1), 1-10.

Lemelle, A.J., & Battle, J. (2004). Black masculinity matters in attitudes toward gay males.
Journal ofHontosexuality, 47(1), 39-51.

Lewis-Beck, M. (1980). Applied regression: An introduction. Beverly Hills, CA: sage
Publications

Liang, C.T.H., & Alimo, C. (2005). The impact of white heterosexual students' interactions on
attitudes toward lesbian, gay and bisexual people: A longitudinal study. Journal of
College Student Development, 46(3), 237-250.

Loftus, J. (2001). America's liberalization in attitudes toward homosexuality, 1973 to 1998.
American Sociological Review, 66(5), 762-782.

Louderback, L.A., & Whitley, B.E., Jr. (1997). Perceived erotic value of homosexuality and sex-
role attitudes as mediators of sex differences in heterosexual college students' attitudes
toward lesbians and gay men. The Journal of Sex Research, 34(2), 175-182.

Marcia, J.E. (1964). Determination and Construct Validity of Ego Identity Status. Unpublished
doctoral dissertation, Ohio State University.

Marcia, J.E. (1966). Development and validation of ego-identity status. Journal ofPersonality
and Social Psychology, 3(5), 551-558.










Marcia, J.E. (1988). Common processes underlying ego identity, cognitive/moral development,
and individuation. In D.K. Lapsley & F.C. Power (Eds.), SelfJ Ego, andldentity:
Integrative Approaches, pp 211-266. New York: Springer-Verlag.

Marcia, J.E. (1993). The ego identity status approach to ego identity. In Marcia, J.E., Waterman,
A.S., Matteson, D.R., Archer, S.L., & Orlofsky, J.L. (Eds.), Ego Identity: A H~andbook
for Psychosocial Research. New York: Springer-Verlag.

McConahay, J.B., & Hough, J.C., Jr. (1976). Symbolic racism. Journal of SociallIssues, 32(1),
23-45.

McCrae, R.R., & Costa, P.T. (1987) Validation of the five-factor model of personality across
instruments and observers. Journal ofPersonality and Social Psychology, 52, 81-90.

Meyer, I.H. (1995). Minority stress and mental health in gay men. Journal of Health and Social
Behavior, 36, 38-56.

Meyer, I.H. (2003). Prejudice, social stress and mental health in lesbian, gay, and bisexual
populations: Conceptual issues and research evidence. Psychological Bulletin, 129(5),
674-697.

National Coalition of Anti-Violence Programs (NCAVP). (2004, November). FBI releases
annual statistics on hate crimes: Annual report's deficiencies continue; eighteen-month
spike in anti-LGBT hate not acknowledged. Retrieved at
http://www.ncavp.org/media/MediaReleaseDeti sxp1067&d=1106 on 03/01/2006.

Negy, C., & Eisenman, R. (2005). A comparison of African American and White college
students' affective and attitudinal reactions to lesbian, gay, and bisexual individuals: An
exploratory study. The Journal of Sex Research, 42(4), 291-298.

Nyberg, K.L., & Alston, J.P. (1977). Homosexual labeling by university youths. Adolescence,
12, 541-546.

Oldham, J.D., & Kasser, T. (1999). Attitude change in response to information that male
homosexuality has a biological basis. Journal of Sex & Martial Therapy, 25(121-124).

Plugge-Foust, C., & Strickland, G. (2000). Homophobia, irrationality, and Christian ideology:
Does a relationship exist? Journal of Sex Education and therapy, 25(4), 240-244.

Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F., (1994). Social dominance orientation:
A personality variable predicting social and political attitudes. Journal ofPersonality and
Social Psychology, 67, 741-763.

Reynolds, W.M. (1982). Development of reliable and valid short forms of the Marlowe-Crowne
Social Desirability Scale. Journal of Clinical Psychology, 38(1), 119-125.










Sakalli, N. (2002). The relationship between sexism and attitudes toward homosexuality in a
sample of Turkish college students. Journal ofHontosexuality, 42(3), 53-64.
Schulte, L.J., & Battle, J. (2004). The relative importance of ethnicity and religion in predicting
attitudes towards gays and lesbians. Journal ofHontosexuality, 47(2), 127-142.

Schwartz, S.J. (2001). The evolution of Eriksonian and neo-Eriksonian identity theory and
research. Identity: An hIternational Journal of Theory and Research, 1(1), 7-58.

Schwartz, J.P., & Lindley, L.D. (2005). Religious fundamentalism and attachment: Prediction of
homophobia. The International Journal for the Psychology ofReligion, 15(2), 145-157.

Sirin, S.R., McCreary, D.R., & Mahalik, J.R. (2004). Differential reactions to men and women's
gender role transgressions: Perceptions of social status, sexual orientation, and value
dissimilarity. Journal of2~en 's Studies, 12(2), 119-132.

Spence, J.T., & Hahn, E.D. (1997). The attitudes toward women scale and attitude change in
college students. Psychology of Women Quarterly, 21, 17-34.

Tabachnick, B.G., and Fidell, L.S. (2001). Using Multivariate Statistics, Fourth Edition. Boston:
Allyn and Bacon.

Theodore, P.S., & Basow, S.A. (2000). Heterosexual masculinity and homophobia: A reaction to
self! Journal ofHontosexuality, 40(2), 3 1-48.

Tureau, Z.L. (2004). College student identity and attitudes toward gays and lesbians (Doctoral
dissertation, University of North Texas, 2004). Dissertation Abstracts hIternational,
64(9-B), 4657.

Van Hoof, A. (1999). The identity status field re-reviewed: An update of unresolved and
neglected issues with a view on some alternative approaches. DevelopnzentalReview, 19,
497-556.

Weis, C.B., Jr., & Dain, R.N. (1979). Ego development and sex attitudes in heterosexual and
homosexual men and women. Archives of Sexual Behavior, 8(4), 341-3 56.

Weiss, J.T. (2003). GL vs. BT: The archaeology of biphobia and transphobia within the U.S. gay
and lesbian community. Journal ofBisexuality, 3(3/4), 25-55.

Whitley, B.E., Jr. (1987). The relationship of sex-role orientation to heterosexuals' attitudes
toward homosexuals. Sex Roles, 17, 103-113.

Whitley, B.E., Jr. (1995). Sex-role orientation and attitudes toward homosexuality: A meta-
analy si s.










Whitley, B. E., Jr., & Aegisd6ttir, S. (2000). The gender belief system, authoritarianism, social
dominance orientation, and heterosexuals' attitudes toward lesbian and gay men. Sex
Roles, 42(11/2), 947-967.
Whitley, B. E., & Lee, S. E., (2000). The relationship of authoritarianism and related constructs
to attitudes towards homosexuality. Journal ofApplied Social Psychology, 30, 144-170.

Williamson, I. (2000). Internalized homophobia and health issues affecting lesbians and gay
men. Health Education Research, 15(1), 97-107.

Wittig, M.A., & Grant-Thompson, S. (1998). The utility of Allport's conditions of intergroup
contact for predicting perceptions of improved racial attitudes and beliefs. Journal of
Social Issues, 54(4), p795-8 12.

Wylie, L., & Forest, J. (1992). Religious fundamentalism, right wing authoritarianism, and
prejudice. Psychological Reports, 71, 1291-1298.

Yang, A. (1997). Trends: Attitudes toward homosexuality. Public Opinion Quarterly, 61, 477-
507.

Zakrisson, I. (2005). Construction of a short version of the Right-Wing Authoritarianism (RWA)
scale. Personality and'Individ'ual Differences, 39, 863-872.









BIOGRAPHICAL SKETCH

Daniel Joseph Potoczniak was born on April 24th, 1973 in Lakewood, NJ, to Joseph John

Potoczniak and Jane Marie Potoczniak. He grew up in Brick, NJ for the first 12 years of his life,

at which point his family moved to Brielle, NJ. During his high-school years, he attended

Christian Brothers Academy in Lincroft, NJ, where he dedicated his time to the pursuit of all

things French, editing the yearbook, and being a "band geek." After graduating from high

school, Daniel attended Georgetown University in Washington, D.C., where he maj ored in

English. Upon graduation, he worked in a number of different j obs, including being an assistant

to a pastry chef, copyediting children's books, selling eyeglasses, and being a supervisor for the

2000 Census. He also performed volunteer work, counseling others on the AIDS/HIV Hotline at

the Gay Men's Health Crisis.

Finding direction and motivation in counseling others, Daniel returned to graduate school

in 2001 at the University of Miami. There, he earned a Master of Science degree in Education

with a concentration in mental health counseling. In 2003, Daniel matriculated at the University

of Florida to begin his Ph.D. in counseling psychology, where he was offered the Marshall Criser

Presidential Fellowship. His counseling and research interests further developed there, and

included a focus on ego identity development across the lifespan, as well as issues that are

relevant to the gay, lesbian, bisexual, and transgender community.

On July 31st, 2007, Daniel completed his one-year pre-doctoral internship in counseling

psychology at the University of Pennsylvania' s Counseling and Psychological Services center.

After receiving his Ph.D., Daniel will seek employment that allows a balance among clinical

work, likely in a counseling center, and more assessment-based activities.





PAGE 1

1 THE INFLUENCE OF EGO IDENTITY ON HETEROSEXUAL INDIVIDUALS ATTITUDES TOWARD LESBIAN AND GAY INDIVIDUALS By DANIEL JOSEPH POTOCZNIAK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 Copyright 2007 Daniel Joseph Potoczniak

PAGE 3

3 To my parents, for being so good at being parents; To Rand, for being my everything, always; To my brother, for being an example and a guide; To my sister, for reminding me that its good to be just who you are. Thank you so much for helping me to get to where I am.

PAGE 4

4 ACKNOWLEDGMENTS I thank my supervisory committee chair, Greg Neimeyer, Ph.D., for his insight and support in navigating this project. I am absolutely certain that it would not have gone so smoothly with anyone else. I also thank M. Jeff Farrar, Ph.D., Mary Fukuyama, Ph.D., and Peter Sherrard, Ed.D., for all of their help and input in making this study as interesting as it has been. I sincerely appreciate all of the above individuals agreeing to work with me in the first place, and appreciate their helpful guidance.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ..............9 CHAPTER 1 INTRODUCTION..................................................................................................................1 1 2 LITERATURE REVIEW.......................................................................................................17 The Motivational and Functional Approach to Understanding Attitudes..............................19 Experiential Attitudes......................................................................................................20 Defensive Attitudes.........................................................................................................21 Symbolic Attitudes..........................................................................................................23 Research on Attitudes toward Lesbian and Gay (LG) Individuals.........................................25 Previous Contact with LG Individuals............................................................................25 Gender-Based Differences in Attitudes toward LG Individuals.....................................29 Symbolic attitudes toward gender and gender roles.................................................29 African American-based gender differences in LG attitudes...................................33 Religion and Religious Variables....................................................................................33 Personality-Based Factors...............................................................................................35 Defensive Attitudes with a Possible Basis in Gender.....................................................39 Purpose of the Study........................................................................................................... ....42 3 METHOD......................................................................................................................... ......44 Participants................................................................................................................... ..........44 Design and Procedure........................................................................................................... ..46 Instruments.................................................................................................................... .........47 Attitudes Toward Lesbians and Gay Men Scale, Short Form (ATLG-S).......................47 Right-Wing Authoritarianism S cale, Short Version (RWA-S).......................................48 Social Dominance Orientation Scale, Abbreviated (SDO-A).........................................48 Attitudes toward Wo men (ATW) Scale..........................................................................49 Ego Identity Process Questionnaire (EIPQ)....................................................................50 Social Desirability...........................................................................................................5 1 Eroticism...................................................................................................................... ....51 Analytic Strategies............................................................................................................ ......52 4 RESULTS........................................................................................................................ .......54

PAGE 6

6 Order of Presentation Effects and Preliminary Analyses.......................................................54 Analyses....................................................................................................................... ...........56 Hypothesis Tests..............................................................................................................5 6 Additional Research Questions.......................................................................................59 5 DISCUSSION.................................................................................................................... ....71 Hypotheses..................................................................................................................... .........71 Research Questions............................................................................................................. ....74 Limitations and Suggestions for Further Research.................................................................77 Implications for Psychotherapy and Counseling....................................................................79 APPENDIX A DEMOGRAPHIC QUESTIONNAIRE..................................................................................82 B THE ATTITUDES TOWARD LESB IANS AND GAY MEN SCALE................................84 C THE RIGHT-WING AUTHORITARIANISM SCALE, SHORT FORM.............................86 D THE SOCIAL DOMINANCE ORIENTATION SCALE, ABBREVIATED VERSION.....88 E THE ATTITUDES TOWA RD WOMEN SCALE.................................................................89 F EGO IDENTITY PROCESS QUESTIONNAIRE.................................................................91 G MEASUREMENT OF EROTIC VALU E OF LESBIAN AND GAY SEX..........................93 H MARLOWE-CROWNE SOCIAL DESIRA BILITY SCALE, SHORT FORM....................94 I EXAMPLE ADVERTISEMENT FOR PA RTICIPANT RECRUITMENT..........................95 J INFORMED CONSENT........................................................................................................96 LIST OF REFERENCES............................................................................................................. ..98 BIOGRAPHICAL SKETCH.......................................................................................................105

PAGE 7

7 LIST OF TABLES Table page 4-1 Means, Standard Deviations, and Scale/Subscale Correlations.........................................63 4-2 Summary of Multiple Regression Analysis for the Prediction of EIPQ Commitment and Exploration Scores by RWA and SDO..................................................................64 4-3 Summary of Multiple Regression Analysis for the Prediction of EIPQ Exploration Scores by Number of Known LG Individuals and the level of Closeness in those Relationships.................................................................................................................. ....65 4-4 Summary of Multiple Regression Analysis for the Prediction of ATLG-S Scores by EIPQ Exploration scores....................................................................................................66 4-5 Moderating Effect of Ego Identity Exploration on the Relation between Authoritarianism and Attitudes toward Lesbians and Gay Men........................................67 4-6 Direct Effects Regression Analyses of Predictor Variables on Criterion Variables (Gender, ATW Scores, and ATLG-S Scores)....................................................................68 4-7 Regression Analyses of Attitudes Toward Women Mediating the Effects of Gender on ATLG-S scores.............................................................................................................69

PAGE 8

8 LIST OF FIGURES Figure page 4-1 Plot of significant Authoritarianism X Ego Identity Exploration interaction, with plotted values for scores on the Attit udes toward Lesbian and Gays (ATLG-S)..............70

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. THE INFLUENCE OF EGO IDENTI TY ON HETEROSEXUAL INDIVIDUALS ATTITUDES TOWARD LESBIAN AND GAY INDIVIDUALS By Daniel Joseph Potoczniak August 2007 Chair: Greg Neimeyer Major: Counseling Psychology The purpose of this study was to examine f actors related to heterosexual individuals attitudes of tolerance versus condemnation towa rd lesbian and gay (LG) individuals. Factors examined include ego identity commitment, ego identity explor ation, authoritarianism, social dominance orientation, gender role attitudes, eroticism, and the number/intimacy of contact experiences with LG people. Using Hereks th eory of attitude func tion, it is posited that attitudes, including those toward LG individuals, may serve an ego defensive function. Theoretically, according to ego identity develo pment theory, the greater degree to which an individual has explored life options and has co mmitted to those which are a satisfactory match, the greater degree of maturity of ones ego/e go-defensive functioning. Along these lines, having a more mature and adaptive ego identity might enab le an individual to be more tolerant toward others who are different from her or him, especially LG individuals. Based in data collected from an Internet-based sample, results indica te that higher levels of authoritarianism and social dominance orientation are related to lower levels of ego identity exploration; also, higher levels of authoritarianism are rela ted to high degrees of commitment. Additionally, greater numbers and greater intimacy of LG contact experiences were associated

PAGE 10

10 with higher levels of ego identity exploration. In turn, higher leve ls of ego identity exploration were related to more tolerant attitudes toward LG people. When considering individuals who demonstrated higher levels of aut horitarian traits, ego identity e xploration statistically moderated these individuals relatively more c ondemning attitudes toward LG people. Previous findings in the liter ature on attitudes toward LG people, many times based in a college-aged sample, were also replicated in the current study, which used an older, non-collegestudent sample. Such replicated findings in clude heterosexual males having more negative attitudes toward LG individuals than do heterosexual women; how ever, this gender difference was eliminated when controlling for the infl uence of gender-role-based attitudes. Also, statistically significant differen ces in attitudes toward LG people were found among different races/ethnicities, especially among African Ameri canand Hispanic-identified participants.

PAGE 11

11 CHAPTER 1 INTRODUCTION In 1998, the murders of Matthew Shepard and Billy Jack Gaither shocked thousands of Americans (Herek, 2000). Shepard was a 21-year-o ld college student in Wyoming, and Gaither was a 39-year-old factory worker in Alabama. The two had apparently little in common, save the fact that both of them were targeted for attack because th ey were gay men. If the two murders had not been grounded in anti-gay sentiment, they might not have been so noteworthy as to garner the tremendous amount of attention they did in the mass media. However, due to the hate-based nature of the crimes, they drew at tention to the danger and violence faced by lesbian and gay (LG) individuals on a regular basis. In early 2004, the Nati onal Coalition of AntiViolence Programs report on violent acts based in prejudice stated that more than 2000 violent incidents over the previous year occurred because of the victim s sexual orientation. Since a great number of these types of crimes may never be reported to the police due to the stigmatized status of the victim, there may be many more such crimes that are never reported (Herek, Gillis, & Cogan, 1999). Stigmatization of LG individuals is not a new phenomenon (Chauncey, 1994), existing well into the previous century, a nd likely longer before that. W ith the rise of the civil rights movement in the 1960s, however, LG individuals condemnation was increasingly scrutinized. Eventually, the American Psychiatric Associatio n eventually declared that homosexuality was not a mental illness (American Psychiatric Associ ation, 1980), and the question of what led some individuals to maintain negative perceptions of LG people began to receive closer attention. Perhaps due to this attention and to LG indi viduals increasing presen ce in visible society, negative attitudes toward lesbians and gay men (hereafter referred to as negative LG attitudes or sexual prejudice) have decreased over the past 30 years (Loftus, 2001): whereas at least

PAGE 12

12 two-thirds of respondents to the General Soci al Survey (GSS) considered same-sex sexual behavior always wrong in the 1970s and 1980s, th at figure declined si gnificantly in the 1990s; by 1996, only 56% of GSS respondents regarded su ch behavior as always wrong (Yang, 1997). In addition to sexual behavior between two women or two men, much of the public also holds negative attitudes toward LG individuals themselves (Herek, 2000). In a study conducted using a random national sample, more than half of the heterosexual participants expressed disgust for LG individuals (Herek, 1994). Respondents to the American National Election Studies typically rate lesbians and gay men amon g the lowest of all so cial groups on a feeling thermometer, in which on a scale of 1 to 101, indivi duals are asked to rate how they feel about a particular group, issue, or other factor (Yang, 199 7). To understand this sexual prejudice (which has sometimes been labeled homophobia or het erosexism), a growing number of studies have explored various factors that appear to contribute to negative LG attitudes. Findings from public opinion surveys have reve aled higher levels of sexual prejudice among individuals who are older, le ss educated, living in the U.S. South or Midwest, and living in rural areas (Herek, 1994). Also, heterosexual men generally have more negative attitudes toward LG individuals, especially gay men, than do heterosexual women (Kite & Whitley, 1996). One possible explanation for gender-based di fferences in LG attitudes rests in a gender role attitudes/beliefs analysis. This point of vi ew states that heterose xuals evaluations of gay men and lesbians are grounded in a much larger belief system concerni ng men and women and their appropriate roles in societ y (Kite, 1994). Gender roles for me n, a class of symbolic beliefs that may be grounded in family-inspired values or in religious or pol itical ideology, appear somewhat more narrowly defined and inflexible than are those for women (Bem, 1993). In this light, a violation of male gender ro les is usually considered more inappropriate than violation of

PAGE 13

13 female roles (Herek, 1988), and an individuals violation of those roles results in a challenge toward concepts of gender appropriateness and a particular individual s acceptability. An alternate explanation of heterose xual mens apparently greater sexua l prejudice lies in the fact that these men may attribute greater erotic valu e to sex between two wome n than to sex between two men, thus rendering a more positive opini on of lesbians (Louderback & Whitley, 1997). In contrast, heterosexual women appear to attribut e low erotic value to both forms of sexual behavior (Louderback & Whitley, 1997). Apart from gender, religious variables are theoretically and empi rically supported in predicting attitudes toward LG individuals (S chwartz & Lindley, 2005). Variables such as Allport and Rosss (1957) classi fication of intrinsic and extrinsic religious orientations, and fundamentalism have been explored. Within the context of relig ion, a fundamentalist understanding of religion (i .e., a literal interp retation of religious script ure) is one of the core variables that predict more negative LG attitude s. Interestingly, however, when one includes a consideration of authoritarianism, which is a personality variable extensively studied by Altemeyer (1981, 1996), the signific ance of fundamentalism is la rgely eliminated (Laythe, Finkel, & Kirkpatrick, 2001). Altemeyers (1981, 1996) theoretical concep tualization of auth oritarianism is characterized by high degrees of submission to au thorities, aggressiveness toward outgroups, and adherence to conventions perceived to be endorse d by society and its author ities, suggesting that people with high degrees of author itarianism would be likely to exhibit anti-LG attitudes. More specifically, these attitudes woul d be likely since LG persons ar e explicitly condemned by many religious and political leaders, making LG pe rsons an acceptable target for prejudice and hostility. Additionally, lesbians and gay men repr esent a socially stigmatized out-group relative

PAGE 14

14 to heterosexual women and men, and are perceived as advocates for politic al change in social conventions. A number of studies have found si gnificant positive links between authoritarianism and anti-LG attitudes (e.g., A ltemeyer, 1981, 2001; Ekehammar, Akrami, Gylje, & Zakrisson, 2004; Laythe, Finkel, & Kirkpa trick, 2001) indicating that pe rsons with high levels of authoritarianism express more anti-LG attit udes than do persons with low levels of authoritarianism. In addition to authoritarianism, social domin ance orientation (SDO) also has been linked consistently with negative att itudes toward lesbians and gay me n. Pratto, Sidanius, Stallworth, and Malle (1994) defined SDO as the extent to which one desires that ones in-group dominate and be superior to out-groups (p 724). SDO ta ps a general attitudinal orientation toward intergroup relations that reflects a preference for hierarchical rath er than egalitarian relations. More specifically, the de sire to maintain the superior posi tion of their in-g roups relative to outgroups motivates people high in SDO to accept hierarchy-legitimizing myths that denigrate members of out-groups and enforce the status quo of their in-groups power position. Thus, persons high in SDO tend to have negative LG attitudes because LG persons have lower social status than do heterosexuals (Whitley & Lee, 2000). The empirical literature supports the expected link between SDO and anti-LG attitude s. Across a number of studies, significant correlations have been found between SDO a nd anti-LG attitudes (e.g., Altemeyer, 1998; Whitley, 1999; Whitley & Lee 2000) such that higher SDO scores are related to higher levels of anti-LG attitudes. Another variable that seems to predict posit ive attitudes toward LG individuals fairly consistently is prior contact w ith a lesbian or gay man (Herek & Capitanio, 1996). Interpersonal contact appears strongly and signi ficantly associated with favor able attitudes toward gay men,

PAGE 15

15 and quantitatively more contact experiences are associated with more favorable attitudes. Intimate contact is more likely than superficial cont act to be associated wi th favorable attitudes, and direct disclosure of anothers LG orientati on was more likely than second-hand disclosure to be associated with positive attitudes towa rd LG people (Herek & Capitanio, 1996). Unfortunately, in exploring the above variab les, almost none of the studies link their hypotheses or findings to a theoretical framew ork. Rather, one study appears to build on anothers variable-based findings without providing a way for th e findings to be expanded or understood. This is especially surprising when one considers th at in 1984, Herek provided a theoretical foundation on which future research c ould be built. Within this framework, attitudes toward LG people enable indivi duals to meet a psychological n eed or serve a psychological function, usually driven by motivation to strengthe n ones personal identity (Kite, 1994). Herek (1984) suggested three functions as most re levant to understanding attitudes toward LG individuals. They are not meant to provide exhaustive explanations of all attitude s, but rather provide the basis for a theory of understanding th e sometimes disparate studies in the area of LG attitudes. First, attitudes may be experiential organizing ones social reality on the basis of previous experiences and interactions with LG individuals. Second, attitudes may be defensive or ego defensive helping a person contain an inner conflict that threatens a stable sense of self by projecting that conflict onto LG individuals. Last, attitudes may be symbolic embodying abstract ideals or values that are linked to ones identity, so cial network, or reference groups (Herek, 1984). Although the above variables might be clas sified as symbolic (e.g., gender roles, authoritarianism) or experientia l (e.g., previous contact with an LG individual), no recent published research has explored ego defensive op tions, which would presumably occur due to a

PAGE 16

16 less strongly developed ego identity (Marci a, 1966, 1988, 1993). The present study aims to address this limitation in the literature, by e xpanding current research on attitudes toward lesbians and gay men through exploration of ego iden tity as a theore tically relevant predictor of these attitudes. Additionally, as most extant studies employ samples composed of undergraduate college students, the current study aims to incor porate an non-college st udent, adult population through the use of Internet -based sampling. In doing so, it is an ticipated that a greater degree of diversity and external validity might be represented in the findings. This paper will be structured in the followi ng way: first, the psychosocial background of attitudes toward lesbians and gay men will be examined; second, Hereks (1984) theory, which frames LG attitudes as serving a psychological f unction will be reviewed; third, empirical studies that demonstrate the relationship of various vari ables to LG attitudes will be explored; finally, the subsequent sections will focus on the desc ription of the current study, the methodology that was used to test the studys hypotheses, th e studys results and a discussion thereon.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW In a letter to an American mother, who ha d written to Sigmund Freud over her concern about her sons homosexuality, Freud wr ote back consolingly by saying that [Homosexuality] is assuredly no advantage but it is nothing to be ashamed of, no vice, no degradation, it cannot be classifi ed as an illness; we consider it to be a variation of the sexual function produced by a certain arrest of sexual development. Many highly respectable individuals of an cient and modern times have been homosexuals, several of the greatest men among them (Plato, Michaela ngelo, Leonardo da Vinci, etc.). It is a great injustice to persecute homosexuality as a crime, and cruelty, too (Freud, 1951, p 787). Despite Freuds seeming acceptance of homose xuality as a mere variation in sexual function, certainly no illness or crime, the psychiatric comm unity labeled lesbian and gay (LG) people as disordered for the majority of the 20th century. During the latter half of the 1960s, however, the North American gay rights m ovement began to mature, and LG individuals began to protest their marginalized status. Am ong other political goals, one of the primary efforts of the movement was the removal of hom osexuality from the DSM-II. Beginning with the work of Evelyn Hooker (1957), mental health professionals began to raise the question whether mental disorder was truly a direct re sult of homosexuality. Hooker found that there were, in fact, normal, healthfully functioning gay men, whose Rorschach test results could not be differentiated from those of a normal hetero sexual population. Rather, perhaps the increased rate of mental disorders was a result of the burden of soci al stigma experienced by sexual minorities (Hooker, 1957). With no small degree of controversy from both within and outside its ranks, the American Psychiatric Association voted to remo ve the diagnostic category of Homosexuality from the DSM-III in 1973 (American Psychiatri c Association, 1980). Even though the stigma of being a sexual minority has er oded significantly since the 1970 s, (Loftus, 2001), there still

PAGE 18

18 remains a significant and pervasiv e prejudice regarding LG indi viduals (Herek, 2000). Whether the medical community initially fueled societys view of homosexuality as deviant and perverse, or vice versa, LG people continue to be socially stigmatized for medical, religious, and other reasons (Boswell, 1980). Apart from the obvious ha rdships this places on lesbians and gay men, such as social disregard for same-sex intim acy, legal barriers which deny same-sex couples financial and legal rights en joyed by heterosexual couples, and disenfranchisement from religious and community organizations, stigma and prejudice toward LG people often pose deleterious effects on these individuals lives, including their mental health (Meyer, 1995, 2003; Williamson, 2000). LG people experience violence and discrimina tion due to their sexual orientation. In November of 2004, the Federal Bureau of Investig ation (FBI) released it s annual report on hate crimes across the nation. The re port, which documented hate cr imes across the country in 2003, contained information on 1,239 incidents in whic h the perpetrators motivating factor was victims actual or perceived sexual orientation. Such incidents represented 17% of the total number of hate crimes in the FBIs report (NC AVP, 2004). The FBIs statistics on anti-lesbian and gay incidents fell short of th e number of incidents tracked in the National Coalition of AntiViolence Programs 2003 report on hate violence, which recorded 2,051 incidents in only eleven regions across the nation more than half of which involved criminal offenses. It is possible that the FBIs apparent blind eye toward anti-LG at titudes and behavior reflects a more general prejudice or stigma toward sexual minorities in the greater United States society. Indeed, when one regards the psycholo gical literature on prejudice toward sexual minorities, it appears that such prejudice and lack of rega rd for LG individuals continues to be a presence in the U.S.s general culture (Herek, 2000, 2002).

PAGE 19

19 The Motivational and Functional A pproach to Understanding Attitudes toward Lesbians and Gay Men Different theorists have suggest ed varying functions most of ten served by attitudes and opinions. For example, opinions and attitudes ma y be a basis of security in the face of a changing world for one individual, yet for anothe r a goad to revolutionary activity. Attitudes are part of ones attempt to master the world and life in that world; more simply, they are a way in which one adjusts to the social environment (S mith, Bruner, & White, 1956). Other historically notable attitude functions include (1) an instrume ntal, adjustive, or utili tarian function, which serve a person in terms of concrete rewards or punishments; (2) an ego-defensive function, in which a person protects her or himself from ac knowledging basic truths about ones self or the external world; (3) a value-expr essive function, in which an indi vidual derives satisfaction from expressing attitudes appropriate to her or his personal values and con cept of self; and (4) a knowledge function, which is based upon an individua ls need to give structure, organization, and meaning to the world (Katz, 1960). In attempting to understand and organize the vari ables that gird an individuals prejudice or tolerance toward LG individuals it is helpful to bear in mind the premise that different people can express similar attitudes towa rd any number of social objects (LG individuals, racial ethnic minorities, individuals with perceived social status ) due to different attitudinal functions (Herek, 1984). For example, two heterosexual individua ls may view LG indivi duals negatively: one individual may be deeply religious and hold Ch ristian fundamentalist views, and her or his negative attitudes toward LG individuals might be accordingly motivated by a value-expressive function; the other individual ma y have developed an ontology th at organizes individuals into traditional and perverse gender roles, and this person condemns LG individuals as a result of the knowledge function. Within this functional approach, attitudes toward LG people enable

PAGE 20

20 individuals to meet a psychologi cal need, usually driven by mo tivation to strengthen ones personal identity (Kite, 1994). With specific regard to attitudes toward LG people, Herek (1984) suggested three functions as most relevant to understanding attitudes toward LG i ndividuals. They are not meant to be an exhaustive taxonomy, but they do provi de a parsimonious framework for understanding the sometimes disparate studies in the area of LG attitudes. First, attitudes may be experiential organizing ones social r eality on the basis of previous expe riences and interactions with LG individuals. Second, attitudes may be (ego) defensive helping a person cont ain an inner conflict that threatens a stable sense of ego or self by projecting that conf lict onto LG individuals. Last, attitudes may be symbolic embodying abstract ideals or values that are linked to ones identity, social network, or reference groups (Herek, 1984) Each of Hereks (1984) three attitude functions will now be explored in turn. Experiential Attitudes Experiential attitudes arise when one has had prior experience with LG individuals, and has emotions and cognitions associated with t hose experiences (Herek, 1984). For example, a person who has had positive interp ersonal contact with an LG person is more likely to hold favorable attitudes toward other LG individuals on the basis of th at contact. Similarly, negative interpersonal contact will likely inspire negative attitudes toward LG i ndividuals. Experiential attitudes do not arise simply fr om contact per se, but importan tly, the contact must provide adequate basis for the formation of attitudes (Allport, 1954). Permitting further understanding of experiential attitudes and cont act between in and out gro ups is the cont act hypothesis (Allport, 1954). This hypothesis fo cuses on the change of an individu als prejudiced attitude in a

PAGE 21

21 social situation, and proposes several conditions necessary for the contact to result in reduced negative attitudes and greater tolerance: 1. support by authority figures and voluntary participation by the participants in the experience; 2. equal perceived status of pa rticipants in the encounter; 3. cooperation and interdependence among participants across groups; 4. individualized contact with the poten tial for further friendship across groups; 5. behavior that that disconfirms stereotypes that the groups hold of each other (Wittig & Grant-Thompson, 1998). According to the theory behind experiential a ttitudes in Hereks theory, heterosexuals who know or have met LG individuals may be better able than others to recognize negative stereotypes as inaccurate, and would be also be more likely to hold tolerant attitudes. Similarly, interpersonal contact with an LG person that ha s not met any one of the above requirements is less likely to result in a tolerant attitude. As only a minority of individuals in the United States have attitudes based on first-hand experience, it is likely that many peoples attitudes are based either in negative stereotype (Herek, 1984) or function accordi ng to a different motivational principle. Defensive Attitudes Defensive attitudes, or ego-defensive attit udes, are those in which a person protects herself or himself from acknowledgi ng basic truths about the self or the external world. In one sense, ego defense mechanisms are adaptive in th at they temporarily remove the harshness of internal conflict, and save the individual fr om psychodynamic disaster. In another sense, however, such mechanisms may not always be ad aptive in that they potentially handicap the

PAGE 22

22 individual in her or his social adjustment, and prevent the ac quiring of maximum satisfactions from the external world (Katz, 1960). According to Marcias theory of ego iden tity (Marcia, 1966, 1988, 1993) if an individual has not sufficiently explored various personal do mains (gender roles, sexual attraction) and committed oneself to particular aspects of those (and other) domains, the result is a less stable ego identity. A less stable identity translates into less optimal defenses, leaving an individual more vulnerable to incoming threats. Theore tically, greater anxiet y, depression, and other maladies could be the result (Marcia, 1993). The theory of defensive attitudes toward LG individuals would suggest that heterosexual men and women who are genuinely secure in their own gender id entity and sexual orientation, having explored various options in those domains and having committed to personally appropriate options, would feel less threatened by homosexuality than would those who are less secure (Herek, 1984). Without such a stable identity, feelings of personal threat may more easily bring about strong negative attitude s toward LG people. This idea appears to have some basis in empirical truth. In one study (Ada ms, Wright, & Lohr, 1996), the ro le of homosexu al arousal in exclusively heterosexual men who admitted negativ e affect toward homosexual individuals was investigated. Participants cons isted of a group of homophobic men ( n = 35) and a group of nonhomophobic men ( n = 29), assigned to groups on the basi s of their scores on a measure of homophobia (or, negative attitudes toward gay men). The men were exposed to sexually explicit erotic stimuli consisting of hete rosexual, male homosexual, and lesbian videotapes, and changes in penile circumference were monitored. Both gr oups exhibited increases in penile circumference to the heterosexual and female homosexual vi deos, but only the homophobic men showed an increase in penile erection to male homosexual stimuli. The authors of the study concluded that

PAGE 23

23 homophobia is likely associated with homosexual arousal of which the homophobic individual is either unaware or denies. In support of this finding and of the link between negative LG attitudes and ego development, other work fo cusing on college students has found a small but significant correlation between higher levels of ego identity development and more positive attitudes toward homosexuali ty (Weis & Dain, 1979). Attitudes thus may serve a defensive function when an individual perceives some analogy between an LG persona and her or his own unc onscious conflicts (Herek, 1984). Subsequently, that inner conflict is projected onto the external source of the conflict (i .e., the lesbian or gay man), thereby reducing the experience of internal anxiety. This strategy pe rmits an individual to externalize conflict regarding sexual orientation, attraction, or gender role by rejecting the lesbians and gay men who exemplify the forbid den, unconscious desire; in doing so, the person is never forced to recognize the urges as her or his own. Symbolic Attitudes Whereas defensive attitudes have the function of preventing the revelation of ones true nature to oneself or to others, symbolic attitudes have the function of giving positive expression to ones central values and the type of person she or he would like to be (and, perhaps the type of person by which she or he would like to be acce pted). For example, an individual might consider her or himself to be politically li beral and open-minded, and as such, may express values that are dominant in liberal political circles. The function is not to gain public reward, but to maintain a stable sense of self in relation to valued others, and to enjoy intrinsically the expression of that self (Katz, 1960). Whether symbolic attitudes are favorable or not favorable, prej udiced or tolerant, they are derived from past and present socializat ion experiences (Katz, 1960; McConahy & Hough,

PAGE 24

24 1976). They express attitudes a nd values that are important to ones sense of self, thereby helping individuals to establish th eir identity and affirm their no tion of the sort of person they perceive and desire themselves to be. Simultaneously, these attitudes mediate the persons relation to other important individuals and refere nce groups, forming part of a continuing social exchange: one defines a sense of self and through reinforcement of ones attitudes and values by others in the relevant group, and al so establishes the inte rpersonal relations that support that self. Sexual attitudes tend to be cons istent with a larger ideol ogy typically found in socially significant reference groups (religions, political parties, geog raphic moral climate, etc.) (Herek, 1984). With so many and varied opinions rega rding the appropriateness of a lesbian or gay orientation in society, depending on ones religion (Schulte & Battle, 2004), political preference (Herek, 2000), geographic area (Herek, 1994), a nd family attitudes (Negy & Eisenman), symbolic attitudes likely play a role in many individuals attitudes regarding LG people. Symbolic attitudes toward minority groups ma y be seen as the expression by a dominant group (i.e., White individuals) that racial or ethnic minorities are violating important societal or religious values and demanding illegitimate ch anges in the status quo (McConahay and Hough, 1976). Along these lines, one may consider this third category of attitudes to serve a similar function for heterosexuals regarding lesbians and gay men. Symbolic attitudes toward LG individuals express the feeling th at important personal or society values are being violated, and that unreasonable demands are being made to change the status quo (Herek, 1984). In recent years, relatively scant research has been completed on attitudes toward LG people, focusing on disparate personal variables that are correlated with tolerant or prejudicial attitudes (such variables are discussed in the following section). Th e attitudinal framework outlined above is not intended to classify all th e various variables that come to play in an

PAGE 25

25 individuals forming an attit ude toward LG individuals. Ra ther, the framework begins to illustrate that individual personal variables (gende r, religion, political a ffiliation, etc.) can serve any number of attitudinal functions, and can play any number of roles in an individuals understanding of her or his self. Research on Attitudes toward Lesbian and Gay (LG) Individuals Fairly consistent findings from national surv eys reveal that higher levels of negative LG attitudes exist among those who are older, less e ducated, living in the U.S. South or Midwest, and living in rural areas (Herek, 1994). Various other personal char acteristics and traits, such as gender (Herek, 2002; Kite & Whitley, 1998), gender role beliefs and attitudes (Sakalli, 2002; Sirin, McCreary, & Mahalik, 2004), previous contact with a le sbian or gay man (Anderssen, 2002; Liang & Alimo, 2005), authoritarianism (A ltemeyer, 1996, 2001), religious orientation and other religious variables (Herek & Capita nio, 1996; Plugge-Foust & Strickland, 2000), and attributions of the etiology of sexual orientation (Landen & Innala, 2002; Oldham & Kasser, 1999), have been explored empirical ly. There is little research re garding attitudinal differences among racial and ethnic minority groups, with the exception of scan t empirical research regarding African Americans attitudes toward LG people (Negy & Eisenman, 2005; Schulte & Battle, 2004). Previous Contact with LG Individuals The variable that seems to predict positive attitudes toward LG individuals most consistently is ones prior contact with a lesbian or gay man. Using racial issues as an example, if a White individual enc ounters an African American individual she or he is likely to encode information about that Black person in terms of th e persons minority status. In other words, the apparent and visible stigma of being dark-ski nned acts as a filter through which otherwise

PAGE 26

26 unremarkable statements or actions are interprete d as indicative or stereotypical of the Black community (Biernat, 2003). The minority person is thus evaluated solely in terms of her or his minority status (Biernat, 2003). In contrast, for someone with a successfully concealed stigma, such as a gay or lesbian person, there is no filt er through which actions or statements would be interpreted. For all intents and purposes, the LG persons behavior ma y be indistinguishable from a heterosexual persons, and thus appears normal and acceptable. When LG individuals disclose their stigmatized status to a heterosexual person after positive regard has already been formed, the heterosexual person may respond to the information by individuating and personalizing the minority groups members, re sulting in reduced prejudice and increased tolerance. This factor would appear to help form the experiential at titudes component of Hereks (1984) theory of attitude functions. In a national telephone survey (Herek & Ca pitanio, 1996), a sample of adults in the United States indicated their attit udes toward gay men and lesbians ( n = 538; 45.9% male and 54.1% female; 81% White, 10.4% Black, 5% Hispan ic, 2.8% Asian, and less than 1% not using any of those labels; mean age 43.8 years; median income = 30,000 and 40,000; median level of education was some college; 35.3% self-label ed as Democrat, 31.6% Republican, and 24.5% Independent). Most respon dents expressed negative att itudes toward gay men, with a majority believing that Sex between two men is just plain wrong and that I think male homosexuals are disgusting (69.8% and 54.1%, respectively). Interpersonal contact was strongly and significantly associated with favorable attitude s toward gay men, and more contact experiences with more individuals was associated with more fa vorable attitudes. Intimate contact was more likely than superficial contact to be associated with favorable attitudes, and direct disclosure of anothers LG orientation was more likely than second-hand disclosure to be

PAGE 27

27 associated with positive attitude s toward LG people. The number of relationships with an LG person accounted for a small but significant amount of variance in attitudes toward lesbians (6.3%) and gay men (6.2%). The level of contac t intimacy with an LG person (immediate family member, distant relative, close fr iend, or acquaintance) also a ccounted for 9.5% of the variance in attitudes toward lesbians, and 9.3% of the variance in attitudes toward gay men. Liang and Alimo (2005) conducted a longitudina l study of heterosexua l college students contact experiences with lesbian, gay, and bisexual individuals a nd the resulting attitudes toward those individuals. Using 401 White heterose xual students (70% women and 30% men), the effects of gender, pre-college attitudes a nd LGB contact, and college LGB contact were explored. The authors believed that, based on pr evious literature, gend er would influence the attitudes and contact that the participants woul d have in LGB relationships. Additionally, based on previous literature and on All ports (1957) contact hypothesis, the authors expected that interpersonal contact with LGB individuals prio r to and while in college would have a positive affect on prejudicial attitudes of the study participants. The effects of gender were relatively small ( = .078 to .200), with women reporting somewh at more positive attitudes than men, and having somewhat more contact with LGB indi viduals prior to and while in college ( = .192, p < .01, and = .106, p < .05, respectively). Attitudes toward LGB individuals prior to college were predicted by precollege contact ( = .426, p < .001); attitudes toward LGB individuals while in college were predicted by pre-college attitudes ( = .190, p < .001) and LGB contact prior to college ( = .320, p < .001); and, in-college attitudes were most strongly predicted by pre-college attitudes and in-college LGB contact ( = .563, p < .001 and = .166, p < .001, respectively). In regarding these statistics, it is important to recognize that pre-college contact with LGB individuals most strongly influe nced later attitudes through both direct and indirect paths (i.e.,

PAGE 28

28 through forming pre-college attitude s which influenced the formation of later attitudes), and that LGB contact while in college had a direct influence. Among college students, similar patterns of contact-based influence appears to hold in coun tries other than the U.S. as well (i.e., Norway, Turkey) (Anderssen, 2002; Sakalli, 2002). The analyses in most LG-attitude studies ar e univariate and correlational in nature, and despite Hereks (1984) theory of motivation and functional attitu des, it is uncommon to find studies that make use of medi ational or moderation-based anal yses. One interesting exception (Mohipp and Morry, 2004) investigat ed the relationship of symbolic beliefs and prior contact on heterosexual attitudes toward LG people. The st udy demonstrated that positive symbolic beliefs concerning LG individuals were associated w ith positive attitudes toward both lesbians ( = -.52, p < .001) and gay men ( = -.31, p < .05); and, that when separate d from other predictors, prior contact was associated positive attitude s toward both lesbians and gay men ( = .19, p < .05). In line with Hereks theoretical ideas, the authors expected to find that symbolic beliefs would statistically mediate the rela tion between prior contact a nd attitudes (Baron & Kenny, 1986). However, no such relationship was found. Contact with LG individuals in the African American community. It is commonly assumed that African American communities, and especially the men of those communities, have significantly less tolerance of homosexuality than do White communities (Battle & Lemelle, 2002). To a certain extent, it app ears that a marginally higher de gree of intolerance may actually exist in the African American community despite having had contact with an LG individual. For example, Herek & Capitanio (1995) conducted a national telephone su rvey, and found that gender differences in LG attitudes among African American individuals were very similar to those of White individuals. Mo re tolerant attitudes were pr edicted by being highly educated,

PAGE 29

29 unmarried, politically liberal, regi stered to vote, nonreligious, a nd included being Black in their concept of gay men. Having contact with an LG individual also predicted more favorable attitudes, but when other variables were controll ed, having such contact lo st any significance. This may be due to 25% more of the African American sample than previous White samples reporting that their contact w ith an LG person consisted of knowing a distant LG friend, acquaintance, or distant LG srelat ive. This is congruent with Hereks (1984) theory regarding experiential attitudes, as well as Allports (1957) c ontact hypothesis, which states that less intimate contact is less likely to lead to a reduction in prejudice. Gender-Based Differences in Atti tudes toward LG Individuals Heterosexual men appear to hold more nega tive attitudes toward lesbians and gay men than do heterosexual women (Kite & Whitley, 1996; Kite & Whitley, 1998). This balance is not yet fully understood, due in part to empirical work that is not solidly grounded in theory, and in part due to a fairly narrowly c onstructed empirical base that is formed largely of college students attitudes to the exclusion of othe r populations (Kite & Whitley, 1998). In order to understand the basis for possible gender-based diffe rences in attitudes toward LG people, an exploration of possible reasons based in the theory of attitu de function (Herek, 1984) will be beneficial. The most relevant aspects of that th eory for gender-based attitude differences lie in symbolic attitudes about gender (i.e., ge nder roles) and ego defensive attitudes. Symbolic attitudes toward gender and gender roles One possible explanation for gender-based diffe rences in LG attitudes rests in a gender role attitudes/beliefs analysis. This point of vi ew states that heterose xuals evaluations of gay men and lesbians are grounded in a much larger belief system concerning men and women, and their appropriate roles in societ y (Kite, 1994). First, gender-asso ciated beliefs appear to be

PAGE 30

30 linked to one another; in other words, if one has stereotypically masculine traits, then it is assumed that stereotypically masculine behaviors a nd characteristics are part and parcel of that persons gender role. If any one aspect of that gender role is violated, such as a gay man having a stereotypically masculine physical appearance, but more stereotypically feminine behaviors, prejudice against that individual will likely be the result (Kite & Whitley, 1998). Gender roles for men, a class of symbolic beliefs that may be grounded in family-inspired values or in religious or political ideology, a ppear somewhat more narrowly de fined and inflexible than are those for women (Bem, 1993). A violation of ma le gender roles is usually considered more inappropriate than violation of female roles (H erek, 1988). An individual s violation of those roles results in a challenge toward concepts of gender appropriateness, and thus toward the particular individuals acceptability. In a meta-analysis of gender-based diffe rences in LG attitudes (Kite & Whitley, 1996, 1998), gender role attitudes had a mean correlation of r = .44 with scores on measures of attitudes toward lesbians and gay men. Intere stingly, when sex differences in gender role attitudes were controlled, the average partia l correlation between gende r role attitudes and attitudes toward lesbians and gay men fell to r = .02. These results indica te that sex differences in LG attitudes are related to differe nces in gender role attitudes. Clarifying further gender differences in symbolic beliefs, Herek (2002) conducted a national telephone opinion survey and reported a number of factors on which gender-based differences were found. Part of the sample was composed of previously polled respondents from an survey that was comp leted two years prior ( n = 666; 57% female 43% male; 81% nonhispanic White; mean age = 47, with an age ra nge of 20-91; median education level = some college; and, median income = $40,000 $50,000). The newly recr uited respondents had a very

PAGE 31

31 similar demographic composition w ith the exception of age, which was an average of two years younger than the previously polled respondents. Th is reflects the period of time elapsed between surveys. The findings were sp lit into three major areas: (1) women generally hold more favorable and less condemning attitudes toward LG people they were more supportive of employment protection and a doption rights, were more willing to recognize same-sex relationships as valid, were less likely to hol d stereotypical beliefs, a nd displayed less negative affective reactions; (2) both heterosexual mens a nd womens attitudes tended to be more hostile toward gay men than lesbians, with gay men more li kely than lesbians to be regarded as mentally ill and to be perceived as child molesters, and with adoption rights more strongly supported for lesbians than for gay men; (3) although hete rosexual individuals expressed more negative attitudes toward LG people of their same sex, this pattern was strongest among men (i.e., heterosexual men rated LG individuals more ne gatively than women, and rated gay men more negative of all). An additional interesting finding from the study (Herek, 2002) that replicated the findings of the first telephone survey mentioned above (Herek & Capitanio, 1999), was a significant effect on heterosexual mens attitudes determined by the order in which the survey questions were posed. When questions concerning lesbians were asked first in a series, thus not casting them in the context of attitudes toward gay men, heterosexual men were less hostile toward lesbians than gay men in some cases, their attit udes toward lesbians were at least as favorable as those of heterosexual women. This suggests th at heterosexual mens attitudes toward lesbians are cognitively organized in a way that differs from attitudes toward gay men. In contrast, heterosexual womens attitudes toward LG people are relatively unaffected by contextual variables (question order in the survey). This may be due to findings in previous studies that

PAGE 32

32 suggest heterosexual mens tendency to sexualize in teractions with the op posite sex more often than do heterosexual women (DeLamater, 1987). Pr esenting a potentially sexualized image of lesbians before gay men may prime the respondents to exhibit more positive attitudes. A study by Louderback and Whitley (1997) clarif ied the above suggestions regarding the role of (a) gender role attitudes and (b) heterosexual mens perception of sex between lesbians as erotic on attitudes toward lesbians and gay men. In a sample of 58 male and 109 female college students (age range = 17 25 years, mean age = 18.7), the sex-based differences in perceptions of LG people mentioned above were replicated. Additionally, men attribut ed more erotic value to sex between two women than to sex between two men, whereas women attributed low erotic value to both forms of sexual behavior. The au thors posited a mediated model, in which sex differences in attitudes toward lesbians and ga y men would be mediated by the perceived erotic value of homosexuality and sex-role attitudes. This hypothesis was supported. Whereas the perceived erotic value of lesbian/gay sex prev iously accounted for 11.9% of the variance in heterosexuals LG attitudes, any differences between mens attitudes toward lesbians versus gay men was eliminated when percei ved eroticism was controlled. Mens LG attitudes, though, were still significantly more negative than those of women, with this sex of participant by sex of attitude target acc ounting for 8.3% of the variance in LG attitudes. When sex role attitudes (which accounted for 17.3% of variance in LG attitudes) was also controlled, the remaining gender-based differences in attitudes toward le sbians and gay men was eliminated. No main effect remained for the sex of the participant or the sex of attitude target and neither were there any interaction effects of these tw o factors (Louderback & Whitley, 1997).

PAGE 33

33 African American-based gender differences in LG attitudes As previously mentioned, attitudes among Af rican American and White individuals are generally similar. For example, similar to larg e national samples, Africa n American men appear to hold less positive attitudes toward LG individu als than do African American females. The etiology of this sex-based differe nce appears, though, not to lie in the gender/sex role attitudes that predict White individuals LG attitude di fferences (Lemelle & Battle, 2004). In data provided by the National Black Politics Study in 1993, representing approximately 6.5 million African American households, variables measured included age, church attendance, education, household income, urban residency, and attitude s toward gay men (age = 40.97 years, SD = 15.86; Education = 13.07 years, SD = 2.98; Hous ehold income = 4.74, SD 2.42 [from a possible 9 categories that ranged from under $10,000 to over $75,000]; Urban Residency = .55, SD = .50; and for religious attendance, 87% attended C hurch at least once a month). Whereas age, income, education level, and urban residency were all significantly related to LG attitudes for women in the study ( p < .05), the only variable that was significant for mens LG attitudes was the frequency of church attendance ( p < .05). For these men, more frequent church attendance was related to less favorable LG attitudes. Although significant, the effect size of church attendance, even when combined with other the other predictors in one regression, was relatively small (Adjusted R2 = .005 for men, and .033 for women). Religion and Religious Variables The popular media often discusses the cont roversy in religious circles regarding individuals of a lesbian or gay orientation. For example, accordi ng to a Catholic encyclical, the Church regards homosexual behavior to be "not a normal condition, the acts being against nature [and] objectively wrong" (p 272, Broderick & Brode rick, 1990). In this light, a number of

PAGE 34

34 studies have regarded the importa nce of religion and religious va lues in attitudes toward LG individuals. Allport & Ro ss (1967) defined two different religious orientations: extrinsic a selfserving, instrumental approach conf orming to social conventions; and intrinsic in which religion provides a meaning-endowed framework in terms of which all life is understood. Allport & Rosss (1967) suggested that ex trinsic individuals would likely be more prejudiced, and that intrinsic individuals would be more tolerant toward non-dominant groups. Although previous research has found intrinsically re ligious individuals to hold fe wer racist attitudes and fewer racial prejudices than do ex trinsically religious individua ls (Donahue, 1985), Herek (1987) surprisingly found that intr insically religious college students were more prejudiced against LG individuals than their extrinsic counterparts. He suggested that an intr insic orientation does not foster an unequivocal acceptance of others, but rather encourages tolerance toward those that are accepted by traditional religious teachings (Herek, 1987). Further research on religious orientation re vealed that the motivating factors behind negative LG attitudes are more complex. In a sa mple of 257 college stude nts at a conservative Christian school in California (110 females a nd 66 males, all White; age range = 18 24, mean age = 18.5; years of college expe rience range = 0 3, mean experi ence = 1.5), findings similar to Hereks (1987) regarding Intrin sic-Extrinsic religious orientat ion were initially found, with Intrinsic individuals endorsing more negative LG at titudes. However, when the influence of the variable Fundamentalism was controlled, High Intr insics were more accepting of LG individuals than Low Intrinsics. As such, Fundamentalism appears to be the motivating force behind some religious individuals anti-LG sent iment. Later research supports this suggestion, stating that among other factors, Fundamentalism accounts for a significant portion of the variance in predicting individuals ne gative LG attitudes ( = .44, p < .001) (Schwartz & Lindley, 2005).

PAGE 35

35 African American religious serv ice attendance & LG attitudes. As mentioned in the previous section on gender differences, Leme lle and Battle (2004) found that attendance at religious services was the single distinguishi ng factor between African American mens and womens LG attitudes. Another study affirmed that religious se rvice attendance may indeed play a unique role in African Ameri can individuals perception of LG people (Schulte & Battle, 2004), although religious service attendance was not examined by participant gender. When predicting attitudes toward lesbians, ethnic differen ces were present in the absence of religious attendance; however, when religi ous attendance was entered into the model, ethnic differences disappeared entirely (see also Negy & Eise nman, 2005). Somewhat similarly, regarding attitudes toward gay men, although there were no racial differen ces in attitudes toward this group, religious attitudes were al ways significant. The authors concluded that LG attitudes do not necessarily vary as a func tion of ethnicity, but rather ma y be a function of religious attendance in African American churches (S chulte & Battle, 2004). Race and religious attendance alone explained variance in LG attitudes ranging from approximately 5 to 10%; when other factors of sexual orientation, marital status geographic region, and religion were entered in the model, the variance explained in LG attitudes was approximately 20 to 23%. Personality-Based Factors Altemeyer (1981) defined the overarching concep t of authoritarianism as comprised of three attitudinal clusters. The first component, authoritarian su bmission, is defined as a high degree of submission to authoriti es who are perceived to be es tablished and legitimate in the society in which one lives (p. 148). The sec ond component is authoritarian aggression and reflects general aggression, direct ed at various persons, which is perceived to be sanctioned by established authorities (p. 148). Aggressiveness in this contex t is the predisposition to do

PAGE 36

36 physical, psychological, economic, or social harm to others. Persons with high levels of authoritarianism tend to be aggressive or support aggressiveness toward targ ets that they believe are socially stigmatized. For example, persons high in authoritarianism might support police violence against suspects in or der to obtain confessions. The fi nal component of Altemeyers authoritarianism is conventionalism defined as a high degree of adherence to the social conventions which are perceived to be endorsed by society and its established authorities (p. 148). Altemeyer (1981) described e stablished and legitimate author ities to include individuals in society who are usually considered to be lega l or moral authorities su ch as parents for younger persons, religious officials, heads of government police officers, judges, and military officials. Altemeyers (1981, 1996) theoretical concep tualization of auth oritarianism is characterized by high degrees of submission to au thorities, aggressiveness toward outgroups, and adherence to conventions perceived to be endorse d by society and its author ities, suggesting that people with high degrees of authoritarianism woul d be likely to exhibit negative LG attitudes. More specifically, this is likel y because of LG persons (1) bei ng explicitly condemned by some religious and political leaders, making LG pe rsons an acceptable target for prejudice and hostility, (2) representing a socially stigmatized out-group relative to heterosexual women and men, and (3) being perceived as advocates fo r political change and challenging social conventions of traditional gende r roles. Indeed, a number of studies have found significant positive links between authoritarianism a nd anti-LG attitudes (e.g., Altemeyer, 1981, 2001; Ekehammar, Akrami, Gylje, & Zakrisson, 2004; La ythe, Finkel, & Kirkpa trick, 2001) indicating that persons with high levels of authoritarianism express more an ti-LG attitudes than do persons with low levels of authoritarianism.

PAGE 37

37 An interesting study examined negative LG att itudes as predicted by authoritarianism and religious fundamentalism (Laythe, Finkel, & Ki rkpatrick, 2001). Their sample consisted of college students from a Midwestern university ( n = 140; 91 females and 47 males; age range = 18 to 48, mean age = 21.2, SD 6.94; 35% of the sample reported regular, once-weekly church attendance, and 9.2% labeled them selves as atheist or agnosti c; on a seven-point scale of religiosity [1 = not religious, 7 = strongl y religious], the mean was 3.9, SD = 1.7). Authoritarianism ( = .37, p < .01) and fundamentalism ( = .21, p < .05) emerged as significant predictors, each accounting for variance in LG prejudice a bove and beyond the prediction afforded by the other. Together, the two predic tors accounted for 28% of the variance (adjusted R2). To explore further the relationship betw een fundamentalism and authoritarianism, the authors completed a secondary post hoc analysis of data provided by previous Canadian studies (Altemeyer & Hunsberger, 1992; Wylie & Forest 1992). Demographic information for these samples was not provided, beyond the fact that one studys participants were parents of college students (247 women and 244 men), and a random sample of Canadian residents in Manitoba (43 women and 32 men). In both data sets, authorit arianism was a significant positive predictor of negative LG attitudes ( = .69 and = .67 for the two samples, both at p < .01). However, fundamentalism was not a significant predictor in either data set apart from authoritarianism. The authors point out that even though their own data found fundamentalism to predict negative LG attitudes significantly, its actual effect in doing so was relatively small ( = .21). As such, fundamentalism appears to predict very little if any variance beyond authoritarianisms relatively strong predictive power (Laythe, Finkel, & Kirkpatrick, 2001). In addition to authoritarianism, social domin ance orientation (SDO) also has been linked consistently with negative att itudes toward lesbians and gay me n. Pratto, Sidanius, Stallworth,

PAGE 38

38 and Malle (1994) defined SDO as the extent to which one desires that ones in-group dominate and be superior to out-groups (p 724). SDO ta ps a general attitudinal orientation toward intergroup relations that reflects a preference for hierarchical rath er than egalitarian relations. More specifically, the de sire to maintain the superior posi tion of their in-g roups relative to outgroups motivates people high in SDO to accept hierarchy-legitimizing myths that denigrate members of out-groups and enforce the status quo of their in-groups power position. Thus, persons high in SDO tend to have negative LG attitudes because LG persons have lower social status than do heterosexuals (Whitley & Lee, 2000). The empirical literature supports the expected link between SDO and anti-LG attitude s. Across a number of studies, significant correlations have been found between SDO a nd anti-LG attitudes (e.g., Altemeyer, 1998; Whitley, 1999; Whitley & Lee 2000) such that higher SDO scores are related to higher levels of anti-LG attitudes. For example, in one study using a sample of 266 college students (122 men, 131 women; age range = 18 24, mean age = 19. 2; 91% White, 4% African American, 2% Hispanic, 2% other ethnicity/race, and 1% As ian), SDO contributed significantly along with authoritarianism, gender role be liefs, and gender in predicting at titudes toward LG individuals (Whitley & Aegisdottir, 2000). When entered join tly in a regression, authoritarianism and SDO accounted for 38.6% of the variance in attitudes to ward lesbians, and 39.5% of the variance in attitudes toward gay men ( p < .001). Although authoritarianism and SDO both in clude an element of prejudice toward outgroups, Altemeyer (1988) argued that the two cons tructs are distinct in that SDO does not have the same psychological r oots that previous studies ha ve unearthed in right-wing authoritarians (p. 61). More spec ifically, a person high in author itarianism is very accepting of traditional values and institutional authorities and is likely to follow the instructions of those in

PAGE 39

39 positions of power and influence. By contrast, an individual high in SDO is not motivated by a sense of duty or morality (Heaven & Bucci, 20 01). Thus, although authoritarianism and SDO both can explain significant, unique and prejudicial variance in negative LG attitudes, they do so in a complementary but not redundant fashion (Altemeyer, 1998). Indeed, empirical evidence indicates that authoritarianism and SDO are only minimally correlate d with each other, if at all, (Altemeyer, 1998; Pratto et al., 1994) and the rela tionship of SDO to prejudice is not accounted for by a joint relationship with authoritaria nism (Altemeyer, 1988; Whitley & Lee, 2000). Defensive Attitudes with a Possible Basis in Gender Psychodynamically inspired defensive attitudes are associated with perceptions of ones own prized dissimilarity from gay men and le sbians, and with the previously mentioned projection of internal conflict. Empirically speaking, heterosexual males consistently hold more negative attitudes toward LG individuals than do heterosexual females; and, heterosexual males attitudes toward gay men are more negative than toward lesbians (Herek, 1988). Heterosexual females also tend to regard LG people negative ly, but their attitudes do not differ significantly according to the gender of the ta rget (Herek, 1988). Dynamically speaking, if an object in the external world recalls undesired tr aits one fears or dislikes in oneself, and if mens gender roles are more rigid than are those for women, the projection of a proportionately greater negative attitude from men toward other men makes theoretical sense (Kite & Whitley, 1998). Regarding the concept of a well developed or strong ego, it is useful to recall Ericksons (1968) stages of lifespan development. In one of the stages, Erikson posited that an individual must endure an identity crisis or struggle within her or himself to develop a consistent view of the world, and how one fits into that world. In this way, there is a consistent picture of oneself as belonging in the world, and knowledge of how the world is structur ed for ones self. Marcia

PAGE 40

40 (1964, 1966, 1988, 1993) elaborated on this concept, suggesting that a well developed or achieved ego structure must have completed phases of development that he called exploration and commitment, which is necessary across many different domains (e.g., occupation, religion, sex roles). In terms of expl oration, an individual with a well developed ego will have explored options for herself/himself in various domains (i.e., for the domain of gender roles: Do I believe that women s hould remain at home to raise children? Do I believe that men could be adequate stay-at-home fathers?). Fo r commitment, an individual may have explored various options and come to accept one in particul ar (i.e., I really believe that both men and women should be in the work pl ace); or, the individual may not ha ve explored options, but still committed to a particular one (i.e., My father worked, and my mother stayed at home it worked for them, and it should probably work for everyone). Studies occasionally suggest that a lack of self-exploration or self-awareness may contribute to attitudes toward LG individuals. For example, he terosexual male college students in one study completed measures that identified that importance as cribed to traditional masculine attributes, the perceived self-discrepancy be tween ones own attributes versus masculine attributes, and homphobia ( n = 85; 32% Protestant, 28% Cat holic, 13% Jewish, 13% Other religions, 14% No religion; 87% were White, 8% were African American, Asian American, or Hispanic, and 5% identified as having o ther ethnicities) (Theodore & Basow, 2000). Interestingly, participants who valued highly ma sculine attributes yet pe rceived themselves as not having those attributes, and w ho did not value feminine attri butes, had the highest scores on the global measure of homophobia. In predic ting homophobia, an interaction effect of Masculine Attribute Importance x Self-discrep ancy Along Masculine Traits significantly predicted homophobia ( = .67, p < .001). Also, an interaction effect of Feminine Attribute

PAGE 41

41 Importance x Self-discrepancy Along Masculine Traits significantly predicted homophobia ( = .51, p < .01). When viewed through a lens of eg o identity development, these findings could suggest that there may be some heterosexual college students who have committed to the importance of stereotypically masculine attribut es, but have not yet explored how their own differing personal attributes could or could not also be of value. If one does not explore ones own personal characteristics and tr aits, but instead commits to an ideal without exploring its personal significance, it is lik ely that rigid ego boundaries (i.e., high scores on homophobia), may result to defend these beliefs. No recently published studies have explicitly regarded the role of ego development in attitudes toward lesbian and ga y individuals. This is surpri sing, considering Hereks (2000) research directive which stated that relatively little research has been devoted to understanding the dynamic cognitive processes associated with an tigay attitudes (p 21). In a study published over 25 years ago that utilized a sample of 200 homosexual and heterosexual White, Protestant college students, there was a significant correl ation between higher levels of ego development and more positive attitudes toward homosexuality (Weis & Dain, 1979). Also, in another more recent sample of 440 undergraduates found in an unpublished dissertation, there existed a significant but complex relationship between ego identity statuses and attitudes toward LG people (Tureau, 2004). Specifically among individual students who we re either high or low on the trait of absolutism, attitudes toward LG people were more similar in those who had an achieved status. Due to the unavailability of that studys complete findings, further information concerning contemporary correlations between ego identity and LG attitudes is unknown. Regardless, it appears that there may be a relationship between more advanced or achieved ego identity statuses, and pos itive attitudes toward LG individuals.

PAGE 42

42 Purpose of the Study According to Hereks (1984) theory, attitudes toward LG individuals can fall into three categories: symbolic, experiential, and ego defe nsive. Previous LG attitude research has explored symbolic concepts including those found in authorit arian and socially dominant personality styles/beliefs, religi ous beliefs, and also gender roles for men and women. In terms of experiential beliefs, the concept of previ ous contact with an LG person has also been explored. Each of these variab les, in addition to other dem ographic information previously mentioned (education level, urban/rural residence, gender, age), has been shown to be related conceptually or empirically to attitudes towa rd lesbians and gay men. The role of ego development, which may fuel Hereks (1984) defens ive attitude function, has yet to be explored in any published research. This study served to explore the role of e go identity development in the context of individual demographic information (including gender), authoritarianism, social dominance orientation, and gender roles. It also served to replicate a nd extend findings from previous studies by using an adult, non-coll ege student population in collecti ng data. It was expected that ego identity would be correlated with aspects of the contact experience with an LG person, and would play a moderating role in th e relationship between such contac t and the resulti ng attitudes. It was also expected that au thoritarianism and social dominan ce would predict levels of ego identity, which would in turn pr edict attitudes toward LG individuals. This study examined the following hypotheses: (1) Higher levels of Authoritarianism and Social Dominance Orientation will predict higher levels of Ego Identity Commitment and lower levels of Ego Identity Exploration. (2) A higher number of LG contact experi ences and greater intimacy in the contact experience with an LG person wi ll predict higher levels of Ego Identity Exploration.

PAGE 43

43 (3) Higher levels of Ego Ident ity Exploration will predict mo re positive attitudes toward LG individuals, and lower levels will pr edict more negative attitudes toward LG individuals. (3a) Ego Identity Exploration will moderate the relationship between Authoritarianism and LG atti tudes; it will also modera te the relationship between SDO and LG attitudes. Additional Research Questions: (1) Previous findings regarding gender and LG attitudes in college students will be replicated, but using a non-college student population. (1a) Males will have more negative attitude s toward LG individuals than will women; this difference will be eliminated when controlling for gender role values. (1b) Less condemning attitudes toward lesbians (as measured by the Attitudes toward Lesbians subscale of ATLG-S) and more condemning attitudes toward gay men (as measured by the Attitudes toward Gay Men subscale of ATLG-S) will be associated with a male gender; however, when erotic values of LG sexuality are controlled, they will account for gender-b ased difference in attitudes toward lesbians and gay men. (2) Authoritarianism, Social Dominance Orientation, and Church Attendance will significantly predict negative att itudes toward LG individuals. (2a) When controlling for the above variable s, race will not be a significant predictor of LG attitudes.

PAGE 44

44 CHAPTER 3 METHOD Participants Considering that the majority of the literature regarding at titudes toward LG individuals is based on data gathered from college students, it is unknown how the results from those data might generalize to a larger, non-co llege student sample. As such, there is a need for studies of sexual prejudice within different subsets of the population, including differently aged people (Herek, 2000). Participants in this study were adult participants recruited not from a college student population, but by means of advertisemen ts on various e-mail listservs and Internet discussion groups. A variety of sampling source s were chosen based on individuals possible interest in LG issues (e.g., The American Ps ychological Association s Divison 44 listserv); individuals likelihood of endorsing interest in social/politi cal issues that are relevant to ethnic/racial minorities, thus possibly providing a racially/e thnically diverse sample (e.g., Googles Black Focus discussi on group, and newsgroups such as alt.culture.african.american.issues); and, indivi duals likelihood of having a defined attitude toward other issues explored in this survey, su ch as gender-issues and nationalism-based issues (e.g., Yahoo!s Republican and Democrat discussi on groups). An advertisement was posted on each of these groups with the relevant Internet-link to the survey, so that individuals could easily access the informed consent page and survey questionnaires. Demographic information was collected, including ag e, sex, race/ethnicity, socio-economic status (education level and income level), urban/suburban/rural re sidence, and geographic region of residence (e.g., Northeast, South, Midwest), number and intimacy level of LG contact experiences and frequency of religious service attendance. The participants sexual orientation was assessed to in order to locate data only from self-identified hetero sexual individuals. This was done by asking

PAGE 45

45 respondents to identify as one of the following: bise xual, gay, heterosexual, or lesbian. Further, to ensure consistency among respondents choice of sexual orientation, indi viduals were further asked to respond to a scale that rated their sexual at traction to others, ranging from 1 = exclusively the opposite sex to = exclusively the same se x. It was expected that heterosexual respondents would endorse responses that indicate more attraction to the opposite sex than the same sex. To determine the necessary sample size for this study, a multiple regression using all possible predictors to predict LG attitudes was co nsidered, as this exploratory analysis will be the most comprehensive analysis in the study (in that it would conceivably employ all predictors simultaneously). Tabachnick and Fidells (2001) formula for sample size with multiple regressions was employed: n > 104 + m, where m equals the num ber of independent variables. With measured predicto r variables (authoritarian ism, social dominance orientation, gender role attitudes, ego identity commitmen t and exploration, erotic value of same-sex intimacy, and social desirability) and demographic information me ntioned above all bei ng possible independent variables, m in the equation is equal to 17 (104 + 16 = 121). As such, the total number of participants required for this st udy was 121. This is relatively c onservative, as all independent variables will be used simultaneously only fo r the purposes of non-hypothesized exploratory analysis. At the end of data collection, which lasted approximately 2 months, the final number of participants was 235. After dele ting 54 participants who identifie d as lesbian, gay, or bisexual, the total number of participants was 181; one additional female participants responses were deleted, as she identified as heterosexual but endorsed being attracted to usually the same sex, and sometimes the opposite sex. After searchi ng for respondents who did not respond correctly

PAGE 46

46 to two questions designed to indicate random re sponding (i.e., Please choose disagree from the items below), 19 response sets were further deleted. The final sample was composed of 162 indi viduals (114 females, 48 males; mean age = 32.6, SD = 12.0; 1 participant iden tified as Native American [.6%], 28 identified as Asian or Asian American [17.3%], 17 identified as African or African American [10.5%], 18 identified as Hispanic [11.1%], no participants identified as Ha waiian/Pacific Islander [0%], and 20 identified as Other [12.3%]. (The percentages listed above add to more than 100% due to participants being allowed to check more than one category. ) 71 participants endorsed living in an urban setting, 80 endorsed a suburban setting, and 9 endor sed a rural setting (2 participants did not respond to this question). The mean income level was 4.4 (SD = 2.2), with level 4 being between $25,000 and $34,999, and level 5 bein g between $35,000 and $49,000. The mean education level was 7 (SD = 2), indicating that some graduate study, but no graduate degree was the average level of education; also, when cons idering standard deviations, the majority of participants indicated havi ng earned at least an associates degree or higher. Design and Procedure The weblink to the survey was posted on various listservs and Intern et discussion groups so that individuals who were in terested in participating in the study could easily access it. Particular effort was exerted to have the weblink circulated to lis tservs or Internet groups that could promote the ethnic and racial diversity of the sample. Participants were also encouraged to circulate the weblink to others w ho may have been interested in participating in the research. Prior to completing the survey, participants read about the purpose and the methods of the study and electronically endorsed a cons ent form. Participants were info rmed that if they chose to

PAGE 47

47 participate they would have the freedom to w ithdraw from the study at any time without any penalty. Instruments Attitudes Toward Lesbians and Gay Men Scale, Short Form (ATLG-S) The ATLG-S was used to measure attitudes to ward LG individuals. Hereks (1988) scale is one of the more commonly used measures of anti-LG attitudes and assesses the affective component of anti-LG attitudes referred to as condemnation-to lerance. The ATLG-S uses a 5-point Likert scale from 1 = disagree strong ly to 5 = agree strongl y. The scale has 10 items: 5 items assess attitudes toward lesbians (ATL-S) and 5 items assess attitudes toward gay men (ATG-S). Sample items include Homosexual be havior between two men is just plain wrong and Female homosexuality is a sin. With this scale, separate scores for attitudes toward lesbian persons and attitudes toward ga y persons can be computed by averaging or adding ATL-S or ATG-S items, respectively. Item ratings were ad ded to yield an overall ATLG-S score, with higher scores indicating more ne gative attitudes. Construct va lidity for full ATLG scores has been demonstrated through consis tently high correlations with va riables associated conceptually with anti-LG attitudes such as dogmatism, conser vative political ideology, and lack of personal contact with LG persons (Her ek, 1988, 1994; Whitley & Lee, 2000) Each short version of the scale correlates highly with its l onger counterpart (ATG with ATG-S, r = .96; ATL with ATL-S, r = .95; and ATLG with ATLG-S, r = .97) (Herek, 1994). ATLG-S scores have yielded satisfactory levels of internal consistency with community-sampled scores on the combined scale (ATLG-S), ATL-S subscale, and ATG-S subscal e, having Cronbachs alphas of .92, .85, and .87, respectively (Herek, 1994). The Cronbachs alph as for the scale in th e current study were .94, .85, and .93, respectively.

PAGE 48

48 Right-Wing Authoritarianism Scale, Short Version (RWA-S) The RWA-S, developed by Zakrisson ( 2005), assesses Altemeyers concept of authoritarianism (1981, 1996). It is a 15-item, 7-point Likert-type scale with item ratings ranging from 1 = strongly disagree to 7 = strongly agre e as recommended by Altemeyer (1996), and item content was derived from the original RW A scale (Altemeyer, 1981, 1996). Higher scores indicate higher authoritarian at titudes and beliefs. Sample ite ms include, The old-fashioned ways and old-fashioned values still show the best way to live, and, Our country needs a powerful leader in order to destroy the radi cal and immoral currents prevailing in todays society. With regard to validity, Zakrisson (2005) demonstrated that RWA-S scores are correlated moderately and positively with scores on measures of racism and sexism, as does the original RWA scale (Altemeyer, 1981, 1996), and is not significantly corr elated with social dominance. The RWA-S has been shown to ha ve adequate reliability, yielding Cronbachs alphas of approximately .80 (Zakrisson, 2005). The Cronbachs alpha for the scale in the current study was .89. Social Dominance Orientation Scale, Abbreviated (SDO-A) The SDO-A, developed by Pratto et al. (1994), was used to measure social dominance orientation, or respondents pref erence for inequality among soci al groups. The SDO is an 8item, 7 point Likert-type scale a nd participants rate their resp onses on a range from 1 = very negative to 7 = very positive. A scale score was computed by adding item ratings, with higher scores indicating greater levels of SDO. Sample items include, Some groups of people are simply inferior to other groups, and, Its OK if some groups have more of a chance in life than others. With regard to validity, SDO-A scores have been shown to correlate positively with cultural elitism, ethnic prejudice, sexism, politic al-economic conservatism, and a desire to end

PAGE 49

49 affirmative action, thus demonstr ating a high degree of construct validity (Pratto et al., 1994). Additionally, SDO-A scores have yielded high relia bility estimates with an average Cronbachs alpha of .86 across multiple samples (Pratto et al ., 1994). The Cronbachs alpha for the scale in the current study was .92. Attitudes toward Women (ATW) Scale The ATW (Spence & Hahn, 1997) was used to a ssess traditional gender role attitudes. The ATW was created to assess pe oples beliefs about the res ponsibilities, privileges, and behaviors in a variety of sphere s that have traditionally been divided along gender lines, but could, in principle, be shared equally by men and women. The scales author s state that the title of the scale is something of a misnomer, in that the scale taps beliefs about appropriate responsibilities and rights for women versus ap propriate responsibilit ies for men (Spence & Hahn, 1997). As responsibilities fo r both genders are included in th e scale, and it has been used previously to determine gender ro le attitudes as they are relate d to LG attitudes (Louderback & Whitley, 1997), it remains an appropriate measure for use in this study. The ATW is a 15-item scale, with items sc ored on a response scale whose extremes are labeled agree strongly and disagree strongly Whereas in previous years Spence and Hahn recommended using a 4-point scale, they currentl y recommend used a 5-point response scale as this has been shown to produce a wider range of scores and less skewness (Hahn, 1993). Approximately half of the items presents an egal itarian point of view a nd the remainder presents a traditional point of view. Th e egalitarian items are reverse-scored. The item scores, which range from 1 to 5, are summed to obtain a total scale score, with higher scores indicating more egalitarian attitudes. Sample items include, Under modern economic conditions, with women active outside the home, men should share in hou sehold tasks such as wa shing dishes and doing

PAGE 50

50 laundry, and, There are many jobs in which men should be given preference over women in being hired or promoted. With regard to validity, Spence and Hahn (1997) conducted a longitudinal study, comparing scores of college students on the ATW since the early 1970s. Findings revealed that as more women have ente red the labor force in greater numbers, and as legislation barring gender discrimination has come to pass, attitudes have shifted over time to become more egalitarian as measured by th e ATW. Additionally, ATW scores appear unifactorial across genders, and have yielded high reliability es timates, with Spence and Hahn (1997) reporting Cronbachs alphas in the mid80s or higher. The Cronbachs alpha for the scale in the current study was .91. Ego Identity Process Questionnaire (EIPQ ) The EIPQ (Balistreri, Busch-Rossnagel, & Ge isinger, 1995) is a 32-item scale that was used to obtain continuous measures of participants levels of e xploration and commitment. Items are rated on a 6-point Likert scale on subscales of exploration and commitment, which will be separately added and averaged, with higher scores suggesting hi gher degrees of exploration or commitment. Use of this scale is appropria te given that the measure provides greater interpretability (on continuums of exploration and commitment ) than do other measures which generally use complex scoring techniques to assi gn one overall status to a participant. Sample items include My ideas about mens and womens roles have never changed as I became older, and, I have definite views regarding the ways in which men and women should behave. With regard to validity, Balistreri et al. (1995) report convergent valid ity with other measures of ego identity, as well as a satisfactory item-loading on re levant factors. Cronbachs alpha values for the exploration and commitment scales report ed by the authors were .76 and .75, respectively; one-week test-retest reli ability was .90 for commitment, a nd .76 for exploration (Balistreri,

PAGE 51

51 Busch-Rossnagel, & Geisinger, 1995). The Cronb achs alpha for the scale in the current study was .87 for exploration, and .81 for commitment. Scor es were transformed into z-scores, so that higher (positive) scores indicated an above-average score, wh ile lower (negative) scores indicated a below-average score. Social Desirability A short form of the Marlowe-Crowne Social Desirability Scale (Reynolds, 1982) was used to assess the tendency to provide socially desirable responses and th us to detect a common source of response bias. Items, which are rated using a truefalse format, were constructed so that the socially desirable respons e would be unlikely to be true for most people. Sample items include, No matter who Im talking to, Im alwa ys a good listener, and It is sometimes hard for me to go on with my work if I am not enc ouraged. High scores on this scale reflect a tendency to offer socially desirabl e responses. The original form of this measure has been shown to be unrelated to measures of psychopathol ogy and related to meas ures of underreporting symptoms (Crowne & Marlowe, 1960). The short form used in this study consisted of the 13 items that had the highest loading in an explor atory factor analysis of the original 33-item measure. The short form has been found to be hi ghly correlated with the original scale, and the internal consistency reliability was estimated to by .76. The Cronbachs alpha for the scale in the current study was .79. Eroticism Eight items used by Louderback and Whitley (1 997) were employed. Four items referred to gay male sexuality, and four parallel items re ferred to lesbian sexuality : I find the idea of a man (woman) making love to another man (woman ) erotic, I find the idea of a man (woman) making love to another man (wom an) repulsive (reverse scored), I think that I would be

PAGE 52

52 sexually aroused by watching two men (women) make love, and I have viewed pornographic materials involving male homosexual (lesbian) act s. The first item was originally used by Nyberg and Alston (1977, as cited by Loude rback & Whitley, 1997); the other items were composed as variations on that th eme. All statements will be rate d on nine-point scales; the first three items will be rated on scales ranging from strongly disagree (1) to strongly agree (9), and the last item on scales ranging from never (1) to frequently (9). The perceived erotic value of lesbianism scale had a reported reliability of .88, and the perceived erotic value of male homosexuality scale had a reporte d reliability of .65. To mainta in consistency with the only previous study that examined eroticism with re gard to gender and LG attitudes (Louderback & Whitley, 1997), these items were interspersed with items from the ATW scale. The Cronbachs alpha for the scale in the current study was .79 for the scale of ga y erotic value, and .86 for the scale of lesbian erotic value. Analytic Strategies To ascertain associations suggested in hypotheses 1-3 (excluding 3a) and research question 2/2a, multiple regression analyses were employed, using the relevant EIPQ score or ATLG score as the criterion variable, while enteri ng the relevant predictor variables in one step. Separate analyses were conducted for each indi vidual hypothesis. For example, in hypothesis two, Ego Identity Exploration was specified as the criterion variab le, while LG contact experiences and the degree of those experiences intimacy was specified as predictor variables. To evaluate the moderation-based relationshi ps in hypothesis 3a, st atistical strategies frequently outlined in the literature was employed, and hierarchical multiple regressions were used (Frazier, Tix, & Baron, 2004). Before proceed ing, the predictor and th e moderator variables were standardized, thus reducing problems associ ated with multicollinea rity among predictor and

PAGE 53

53 moderator variables and the interaction terms crea ted from them. To test the moderation effects, variables were entered in the regression equation in a stepwise manner: the variable to be controlled for (Social Desirability) was entered in the first step, followed by the standardized predictor and moderator variables in the second step, followe d by the product (interaction) terms in the third step (Aiken & West, 1991; Holm beck, 1997). Finally, all remaining possible interactions with the controlled variable, Social Desirability, were checked in the final step to ascertain no unexpected significant interactions. The significance of the moderation interaction effect was then tested. To evaluate research questions #1a-b, a dummy variable was assigned to all participants, with indicating a male gender, and indi cating a female gender, and regressions were employed to analyze the data. Following the guide lines for mediational analyses that have been detailed in the literature (Frazier, Tix, & Bar on, 2004), the variables for question 1a were entered stepwise in a hierarchical regression.

PAGE 54

54 CHAPTER 4 RESULTS Order of Presentation Eff ects and Preliminary Analyses Prior to the main component of the study be ing posted on the Inte rnet, a pilot study was run to determine if there were order-based effect s of the above measures. If the order of the measures had affected the responses of partic ipants on the Ego Ident ity Process Questionnaire (as the central measure in this study), that que stionnaire would have been placed first in the survey to prevent influence from other measures and the remaining measures would have been placed following it in a random order. The pilot test resulted in two pools of data ( n = 22 for Group 1, and n = 21 for Group 2). To analyze the results for the two groups on scores of Ego Identity exploration and commitment, independent samples t tests were employed to determine if any significant difference between the two groups means existed. For th e individuals who completed the ego identity measure at the beginning of the survey, exploration (M = 69.4, SD = 11.5) was not significantly different from individuals who completed the measure at th e end of the survey (M = 67.9, SD = 12.2), t (41) = .40, p = .69 (two tailed). Similarly, for the indivi duals who completed th e ego identity measure at the beginning of the survey, commitment (M = 66.2, SD = 9.0) was not significantly different from individuals who completed the measure at the end of the survey (M = 69.3, SD = 7.1), t (41) = -1.3, p = .20 (two tailed). Due to the lack of signi ficant difference on ego identity scores in the above pilot test, the ego identity measure was place d at the beginning of the survey for all future participants. Data from the pilot study were inco rporated into the main samples data, and were included in the above sample description.

PAGE 55

55 All measures were assessed for internal c onsistency. Cronbachs alphas were determined to be adequate for the current study, and were consistent with other studies using the same measures; values ranged from .79 to .94, and are summarized with measure intercorrelations in Table 4-1. None of the correlations met the gene rally accepted standard of being greater than .80 to signify that two variables are significantly collinear (Lewis-Beck, 1980), and thus analyses by multiple regression remained plausible. Data were further checked for multivariate normality through assessment of skewness and kurtosis. Al l variables values fo r skewness and kurtosis fell between -2 and 2, with the exception of SDO (skewness = 1.8; kurtosis = 2.9) ATW (skewness = -2.5; kurtosis = 7.4), and Number of LG People Known (skewness = -2.76, kurtosis = 7.27). A square-root transformation was applied to the values for SDO, due to its values being positively skewed. This resulted in acceptable le vels of skewness (1.3) and kurtosis (1.0). Due to the values for ATW and Numb er of Known LG People being ne gatively skewed, the values were reflected (each score substracted from th e highest score, producing a mirror of the original data), and the square root of each valu e was taken. For ATW, this resulted in acceptable levels of skewness (1.0) and kurtosi s (1.5). It is importa nt to note that inverting the ATW values reversed the meaning of the scores, with lower values now indicating more tolerant attitudes toward women. For the Number of LG People Known, square root transformations were not effective at reducing the skewness and kurtosis, and nor were log-ba sed transformations. This is likely due to the participants frequent endorsement of rela tively high levels of known-LG individuals. However, an i nversion transformation (in whic h the minimum scale value was increased to 1 to avoid division by zero, and each value of the variable was divided into 1) resulted in marginally acceptable levels of skewness (-1.9) and kurtosis (2.1), while approximately preserving the original pattern of value distributions of the original data.

PAGE 56

56 Analyses Hypothesis Tests Hypothesis 1 predicted that participants who had higher scores on authoritarianism (RWA) and social dominance orientation (SDO) w ould predict higher levels of commitment and lower levels of exploration. To examine the influence of the predic tor variables (RWA and SDO) on the two factors of ego identity, those va riables were entered in a single block in two separate regressions, with commitment as the cr iterion variable in the first regression, and with exploration as the criterion vari able in the second regression. Social Desirability was also entered in the equation as a predic tor, to control for its effects. The regression with commitment as the dependent variable was significant, F (3, 158) = 19.73, p < .001, explaining a moderate amount of the variance ( R2 = .27, adj R2 = .26) (Cohen, 1988). The R2 and the adjusted R2 in this regression were almost identical, suggesting that the adjustment in these cases was very small because a fairly large sample size was utilized. Tests of standardized partial re gression coefficients revealed th at higher levels of RWA were significantly associated with hi gher levels of commitment ( = .44; p < .001), but that SOD was not significantly associated with commitment ( = .03; p = .68). Social Desirability was also significantly related to commitment scores ( = .15; p <.05). The regr ession with exploration as the dependent variable wa s similarly significant, F (2, 159) = 36.60, p < .001, explaining a relatively large amount of the variance ( R2 = .41, adj R2 = .40) (Cohen, 1988). Tests of standardized partial re gression coefficients revealed th at higher levels of RWA were significantly associated with lo wer levels of exploration ( = -.54; p < .001), and that higher levels of SOD were also si gnificantly associated with lo wer levels of exploration ( = -.15;

PAGE 57

57 p < .05). Social desirability was also signi ficantly related to exploration scores ( = -.19; p < .001). Hypothesis 1 was part ially supported (Table 4-2). Hypothesis 2 predicted that higher number of LG contact experiences and greater closeness/intimacy in the contact experience w ith LG people would predict higher levels of exploration. To examine the influence of the predictor variables (Num ber of Known LG People and Closeness) on exploration, those variables were entered in a single block. Social Desirability was also entered in the equation, to control for its effects on exploration. The overall regression was significant, F (3, 158) = 21.14, p < .001, explaining a m oderate amount of the variance ( R2 = .29, adj R2 = .27) (Cohen, 1988). Tests of standardi zed partial regression coefficients revealed that knowing greater numbers of LG individuals was significan tly associated with higher levels of exploration ( = .29; p < .001), as was the degree of closeness in the relationships with those individuals ( = .18; p < .05) (Table 4-3). Social Desirability was also significantly related to exploration ( = -.31; p < .001). Hypothesis 3 predicted that hi gher levels of Ego Identity Exploration will predict more positive attitudes toward LG individuals, and lowe r levels will predict more negative attitudes toward LG individuals. To examine the influence of the predictor variab le (exploration) on the criterion variable (Attitudes towa rd LG People), both variables were entered in a regression. The overall regression was significant, F (2, 159) = 36.66, p < .001, explai ning a moderate-to-large amount of the variance ( R2 = .32, adj R2 = .31) (Cohen, 1988). Tests of the standardized partial regression coefficient revealed th at a higher level of ego ident ity exploration was significantly associated with more tolerant attitudes (l ower scores on the ATLG-S) toward LG people ( = -.54; p < .001) (Table 4-4). Social Desi rability was not significantly related to ATLG-S scores ( = .05; p = .50).

PAGE 58

58 Hypothesis 3a predicted that Ego Identity Exploration would mode rate the relationship between authoritarianism and LG attitudes; it wa s also predicted to moderate the relationship between SDO and LG attitudes. However, when a regression was carried out using ATLG-S scores as the criterion variable and RWA and SDO were entered as predictor variables to gauge their main effects, SDO was not a significant predictor of Attitudes toward LG individuals ( = .03; p = .64). Due to this finding, no furthe r analyses for this hypothesis were executed using SDO as a predictor variable. To test the hypothesis rega rding the moderation of aut horitarianism by exploration, Frazier, Tix, and Barons (2004) recommendation to use hierarchical multiple regression procedures was followed. The variables were st andardized to reduce multicollinearity between the interaction term and the main effects when tes ting for moderator effects. In the first step of the regression, social desirability was entered, in order to control for it s effects throughout the remainder of the steps. In the second step, th e main effects (authoritarianism and exploration) were entered. In the third step, the interaction (authoritarianism x exploration) was entered; at this step, a significant change in R2 for the interaction term would indicate a significant moderator effect. For the final st ep, all possible interactions were added to ascertain any further effects (social desirability x aut horitarianism, social desirability x exploration, social desirability x authoritarianism x exploration) As shown in Table 4-5, ther e was a significant moderator effect, indicating that exploration moderated the link of authoritarianism to attitudes toward LG individuals. As indicated by the adjusted R2 for the regression equation, the main interaction effects accounted for 57% of the variance in ATLG-S scores. The interaction term of authoritarianism and exploration accounted for an additional 3% of the variance in ATLG-S scores.

PAGE 59

59 To explore patterns underlying this significant interaction effect, an Excel spreadsheet program (at http://www.jeremydawson.co.uk/2-way standardised.xls ) was used to plot the interaction effects of au thoritarianism and exploration, and to provide exact values according to the moderator effects. In Figure 4-1, it may be seen that for individuals with low levels of authoritarianism, there is little difference on ATLG-S scores between individuals with high versus low levels of ego identity exploration. However, for individuals who endorsed high levels of authoritarianism, negative or condemning attit udes toward LG individuals were moderated to a significant degree by higher levels of ego identity exploration. In other words, ego identity exploration appears to allow gr eater tolerance versus condemna tion in attitudes toward LG people, moderating the effects of even a high level of authoritarianism. Additional Research Questions As previously noted, the vast majority of research on LG attitudes has been conducted using samples of college students. In addi tion to the above hypotheses, it was questioned whether a sample that was older than a typi cal college-student-age d population would have similar attitudes toward LG indi viduals. The first set of rese arch questions (1a and 1b) were based in the relationship between gender and LG attitudes. Question 1a posited that males in the current sample would have more negative attitudes toward LG individuals than will women; but, th at this difference would be eliminated when controlling for gender role values (measured by Attitudes toward Women ATW). In other words, the relationship between gender and ATLG would be fully mediated by ATW scores. A frequently used multi-step process for testing mediation was used(Baron & Kenny, 1986; Frazier, Tix, & Barron, 2004; Holmbeck, 1997). In the first step, the predictor (e.g., social support) should be significantly associated with the outcome (ego identity). In the second step,

PAGE 60

60 the predictor should be significantly associated with the proposed mediator (self-concealment). In the final step, in an equation in which the pred ictor and mediator are jointly entered to predict outcome, the mediator should be significantly associated with the outcome. Correspondingly, the previously significant effects of the predictor should be non-signi ficant or significantly reduced once the role of the mediator is controlled. In th is last step, it is also important to test the significance of the mediated effect (Baron & Kenny, 1986; Frazier et al., 2004; Kenny, Kashy, & Bolger, 1998). In the first analysis step, recoded gender (0 = males, 1 = females) was entered in a regression equation predicting scor es on the Attitudes toward Le sbian and Gay Men, Short Form (ATLG-S). Gender accounted for significan t variability in ATLG-S scores ( R2 = .04, = -.20, p < .05). Thus, gender explained significant vari ance associated with ATLG-S scores and the direction of this effect indicated that a male gender was signif icantly associated with less tolerant attitudes toward LG individuals. In the second step of testing mediation, the regression analysis indi cated that gender also accounted for significant variation in scores on the Attitudes toward Women (ATW) scale ( R2 = .07, = -.26, p < .01). The direction of the e ffect suggested that a male gender was associated with less egalitarian views toward men s and womens gender roles. In the third step of testing the mediated effect, both gender a nd ATW scores were entered in the regression predicting ATLG-S scores. These results are summari zed in Tables 6 and 7. Results revealed that the mediator (ATW scores) was significantly asso ciated with the outcome, while gender, initially significantly associated with ATLG-S scores, ce ased being a significant predictor. The results indicated that ATW scores, indicative of gender ro le attitudes, mediated the relationship between gender and attitudes toward lesbians and gay men. The mediated effect was evaluated for

PAGE 61

61 statistical significance (Baron & Kenny, 1986; Kenny et al., 1998). The mediated effect was statistically significant, Z >1.94, p < .05. Research Question 1b posited that less condemn ing attitudes toward lesbians and more condemning attitudes toward gay men would be a ssociated with a male gender; however, when erotic values of LG sexuality are controlled, they would account for gender-based difference in attitudes toward lesbians and gay men. Similar pr ocedures to that used in research question 1a were employed. In the first analysis step, rec oded gender (0 = males, 1 = females) was entered in a regression as a criterion variable, with Attitudes toward Gay Men (ATG) and Attitudes toward Lesbians (ATL) scores as predictors. Ne ither predictor was significantly associated with gender, and so analyses were not continued. In this light, the gender of the sexual minority person in question does not appear to play a significant role in a heterosexual persons developing attitudes toward lesbia ns or gay men. Rather, the fact that an individual is known to be a sexual minority per se, regard less of gender, appear to be th e important factor in attitude development, as demonstrated by Research Question 1a. Research Question 2 posited that Authoritar ianism, Social Dominance Orientation, and Church Attendance would significan tly predict negative attitudes to ward LG individuals (while controlling for Social Desirability). To examine the influence of the predictor variables (RWA, SDO, Frequency of Religious Service Attendence, and Social Desirabili ty) on exploration, those variables were entered in a singl e block in a multiple regression, with ATLG-S as the criterion variable. The overall regression was significant, F (4, 157) = 63.89, p < .001, explaining a large amount of the variance ( R2 = .62, adj R2 = .61) (Cohen, 1988). Tests of standardized partial regression coefficients revealed that higher leve ls of RWA were significantly associated with

PAGE 62

62 ATLG-S ( = .73; p < .001), but that SOD ( = .04; p = .50), Frequenc y of Religious Service Attendance ( = .07; p = .20), and Social Desirability ( = -.01; p = .87) were not significantly associated with ATLG-S. Maintaining these va riables in the equation, all possible dummy-coded race/ethnicity variables were entered in the equation (Native American, Asian/Asian American, African American, Hispanic/La tino, Caucasian, Biracial, and Other). Again, the overall regression was significant, F (11, 150) = 26.44, p < .001, explaini ng a large amount of the variance ( R2 = .66, adj R2 = .64) (Cohen, 1988). Tests of st andardized partial regression coefficients revealed that highe r levels of RWA were significantly associated with higher levels of commitment, but to a lesser degree than when not considering race ( = .66; p < .001). The only racial category that was significant in predicting ATLG-S scores was an African American identification ( = .22; p < .01). A Hispanic/Latino iden tification was marginally significant ( = .15; p = .05). Thus, even when controlling fo r other relevant variab les, including religion, some ethnic/racial categories remained significant in predicting ATLG-S scores.

PAGE 63

63 Table 4-1. Means, Standard Deviati ons, and Scale/Subscale Correlations Min. Max.M SD 1 2 3 4 5 6 7 8 1. EIPQ Exploration -2.5 2.8 0.0 1 .87 2. EIPQ Commitment -3.5 1.7 0.0 1 -.57** .81 3. ATLG-S 9.0 45 16.8 9.5 -49** -.56** .94 4. RWA 16.0 99.0 43.5 18.0 .50** -.60** .78** .89 5. GEV 4.0 18.0 9.4 3.9 -.34** -.50** -.60** -.59** .79 6. LEV 4.0 20.0 12.2 4.5 -.19* .27** -.37** -.41** .48** .86 7. Social Desirability 0.0 13.0 4.5 3.2 .29** -.37** .25** .32** -.31**-.24** .79 8. SDO 2.83 7.48 3.9 1.2 .29** -.43** .43** .53** -.28**-.15 .20** .92 9. ATW 1.0 7.8 3.1 1.4 .37** -.54** .64** .68** -.47**-.17* .18* .61** .91 EIPQ = Ego Identity Process Questionnaire; ATLG-S = Attitude s Toward Lesbians and Gay Men, Short Form; RWA: Right-Wing Authoritarianism; GEV/LEV = Gay/Lesbian Er otic Value Questionnaire; SDO = Social Dominance Orientation; ATW: Attitudes Toward Women. Cronbachs coefficients alphas appear in bold italics on the diagonal. Note: N = 162. p < .05. ** p < .01. 63

PAGE 64

64 Table 4-2. Summary of Multiple Regression Analysis for the Predicti on of EIPQ Commitment and Exploration Scores by RWA and SDO Commitment Exploration Predictor B SEb t B SEb t RWA .03 .01 .44 5.36** -.03 .00 -.45 -6.11** SDO .02 .07 .03 .32 -.13 .06 -.15 -2.10* Social Desirability .05 .02 .15 2.06* -.06 .02 -.20 -3.10** Note: Overall regression results for Commitment, F (3, 158) = 19.73, p < .001, R2 = .27; for Exploration, F (2, 159) = 36.60, p < .001, R2 = .41. RWA = Right-Wing Authoritarianism, SDO = Social Dominance Orientation; p < .05. ** p < .001. 64

PAGE 65

65 Table 4-3. Summary of Multiple Regression Analys is for the Prediction of EIPQ Exploration Scores by Number of Known LG Individuals and the level of Closeness in those Relationships Predictor B SEb t Number of LG people known 1.31.33 .29 4.00** Closeness .22 .09 .18 2.50* Social Desirability -.10 .02 -.31 -4.57** Note: Overall regression results, F (3, 158) = 21.14, p < .001, R2 = .29. p < .05. ** p < .001.

PAGE 66

66 Table 4-4. Summary of Multiple Regression Analysis for the Prediction of ATLG-S Scores by EIPQ Exploration scores Predictor B SEb t Exploration -5.14 .67 -.54 -7.66* Social Desirability .14 .21 .05 .69 Note: Overall regression results, F (2, 159) = 36.66, p < .001, R2 = .32. p < .001.

PAGE 67

67 Table 4-5. Moderating Effect of Ego Identity Exploration on the Relation between Aut horitarianism and Attitudes toward Lesbians and Gay Men Step and Variable B t Total R2 Adj. R2 R2 inc. F inc. df Step 1 Social Desirability 2.38 .25 3.28* .06 .06 .06 10.73 1, 160 Step 2 Authoritarianism Exploration 6.66 -1.43 .70 -.15 11.54** -2.43* .63 .62 .57 120.42 2, 158 Step 3 Authoritarianism X Exploration -1.43 -.20 -3.82** .66 .65 .03 14.65 1, 157 Step 4 Social Desirability X Authoritarianism Social Desirability X Exploration Social Desirability X Authoritarianism X Exploration .38 .87 -.09 .04 .10 -.02 .57 1.42 -.25 .67 .65 .01 .73 3, 154 Note: Overall regression results, F(2, 159) = 36.66, p < .001, R2 = .32. p < .05, ** p < .001. 67

PAGE 68

68 Table 4-6. Direct Effects Regre ssion Analyses of Predictor Va riables on Criterion Variables (Gender, ATW Scores, and ATLG-S Scores) Note: Nmale = 48, Nfemale = 114, p < .05 Predictor Criterion B SE B Gender (0 = male, 1 = female) ATLG-S -4.18 1.60 -.20* ATW -.81 .24 -.26*

PAGE 69

69 Table 4-7. Regression Analyses of Attitudes To ward Women Mediating the Effects of Gender on ATLG-S scores Criterion Predictor B SE B ATLG-S Gender (0 = male, 1 = female) -.79 1.30 -.04 (Mediator)ATW .420 .42 .63* p < .01

PAGE 70

70 Figure 4-1. Plot of significant Authoritarianism X Ego Identity Exploration interaction, with plotted values for scores on the Attit udes toward Lesbian and Gays (ATLG-S) 0 5 10 15 20 25 30 Low AuthoritarianismHigh AuthoritarianismAttitudes toward Lesbians and Gay s Low Ego Identity Exploration High Ego Identity Exploration Low Authoritarianism High Authoritarianism Low Ego Identity Exploration 9.24 24.44 High Ego Identity Exploration 10.26 19.70

PAGE 71

71 CHAPTER 5 DISCUSSION Hypotheses Hypothesis 1, which predicted that part icipants who had higher scores on authoritarianism (RWA) and social dominance orientation (SDO) would have higher levels of ego identity commitment and lower levels of e go identity exploration, was partially supported. RWA predicted higher levels of commitment, and lower levels of exploration; SDO predicted lower levels of exploration, but did not significantly predict levels of commitment. This begins to lend support to the concept that attitudes may help or hinder individual development, or vice versa; more specifically, foreclos ure of options (high commitment low exploration) appears to be related to authoritarian a nd/or socially dominant values. However, it is important to remember that because this research is correl ational in nature, it is uncertain whether more authoritarian or socially dominant attitudes precede high commitment/low exploration of ego identity development, or vice versa. It seem s possible that one could embrace a conservative, authoritarian, socially do minant ideology first, and when late r confronted with others issues germane to an authoritarian id entity, adopt a viewpoi nt consistent with the larger body of authoritarian views. However, it is also reasonable to conceive of an individual first developing some beliefs and values consistent with an au thoritarian viewpoint, a nd later finding congruency between her/his beliefs and the la rger authoritarian group; from th at point, the individual could conceivably embrace other beliefs from the authorita rian group out of a feeling of identification or solidarity. Lacking an experimental or longitudinal design, this study cannot begin to postulate whether the chicken or the egg came first, but it is cl ear that identity development and authoritarian/social domina nce-based attitudes are related. To carry the finding one step further and to place it more squarely in the language of Hereks (1984) theory, ego-based

PAGE 72

72 attitudinal functions (including the defensive f unctions) and symbolic attitudinal functions are related to one another and may influence each other. Hypothesis 2, which postulated that greater numbers of LG cont act experiences and greater intimacy in those contact experiences would predict higher levels of ego identity exploration, was fully supported. Thus, not only do LG contact experiences and greater intimacy therein have a relation to more tolerant attitudes toward LG individuals (Herek & Capitanio, 1996; Liang & Alimo, 2005), but also are related to greater levels of ego identity exploration. Again, since this research is correlational, one cannot reasonable suggest that LG-based acquaintances promote higher levels of explorat ion, or vice versa. However, it does seem reasonable to imagine that the ego identity explor ation, as inspired through interaction with an LG person, could contribute to higher levels of exploration in certain do mains of ego identity (i.e., gender roles, relationships). Alternately, if ego identities that are high in exploration levels indicate a certain type of personality (Schwa rtz, 2001; van Hoof, 1999), (e.g., an open-minded or open-to-experience type of personality), then that personality might be more amenable to increased exposure to other individu als beliefs, values, and ways of living. As with the previous hypothesis, it is difficult to im agine an experimental or longitudinal study that might detangle causation from effect in this scenario. At the core of this study, hypotheses 3 a nd 3a explored the role of ego identity development in relation to attitudes toward LG i ndividuals. Namely, the theoretical concept to be tested stated that greater authoritarian/s ocially dominant attitudes would be related to condemning LG attitudes, but that a more develo ped ego (namely, with greater levels of ego identity exploration) would temper such condemni ng beliefs. To begin to test this concept, Hypothesis 3 predicted that higher levels of ego identity explorat ion would predict more positive

PAGE 73

73 attitudes toward LG individuals, and that lower levels of ego id entity exploration would predict more negative attitudes toward LG individua ls. This hypothesis was supported. Confirming Hereks (1984) previous untested theory that ones ego, namely the defenses embodied therein, could affect attitudes toward LG individuals, this study found that a more developed ego identity was related to greater tolerance in attitudes towa rd LG individuals. This is interesting because although many domains of identity constitute the c oncept of ego identity (e.g., career, politics, religion, sex roles), the overall construct was stil l significantly related to LG-based attitudes. Schwartz (2001) and van Hoof (199 9) both suggested that ego iden tity may be an indicator of overall personality, perhaps having some relation to the Big Five c onstruct of Opennness to Experience in this scenario (McCrae & Costa, 1 987). Greater flexibility in ones ego identity appears to allow greater flexibili ty and adaptability in encount ering differences throughout the lifespan specifically in this scenario, greater flexibility a nd tolerance of others who are different from ones self. Tying all previous hypotheses together, hypot hesis 3a suggested that ego identity exploration would moderate the relationship betw een authoritarianism and LG attitudes, and that it would also moderate the relationship between so cial dominance orientation and LG attitudes. However, since SDO was not related to attitudes toward LG individuals, it was not included in the analyses for this hypothesis. Apart from this deletion, the hypothesis was supported. In order to understand this outco me, it will be important to clarify the nature of statistical moderation. According to Frazier, Tix, and Baron (2004), questions involving moderators address when or for whom a variable most strongly predicts or causes an outcome variable. More specifically, a moderator is a variable that alters the direction or strength of the relation between a predictor and an outcome (Baron & Kenny, 1986; Holmbeck, 1997; James & Brett,

PAGE 74

74 1984). Thus, a moderator effect is basically an in teraction whereby the e ffect of one variable depends on the level of another. In other words, researchers are interested in moderation when the relation between predictor (here, authorita rianism) and outcome (attitudes toward LG individuals) variables ar e stronger for some people (i.e., those with lower levels of exploration) than for others (i.e., those with higher levels of exploration). When considering individuals with low leve ls of authoritarianism, there is little difference in ATLG-S scores between individu als with high or low levels of ego identity exploration. However, when c onsidering individuals with highe r levels of authoritarianism, condemning attitudes toward LG individuals were moderated to a significant degree by higher levels of ego identity exploration. This fi nding suggests that the de gree to which one has explored various domains of her or his life (p erhaps demonstrating a degree of intrapersonal adaptivity/flexibility) appears to allow greater tole rance in attitudes toward LG people, even as levels of authoritarian values and attitudes become higher. In essence, a better developed ego identity, sp ecifically one consisting of higher levels of ego identity exploration, implies less of a need to defensively pr oject negative reactions about a gay or lesbian sexual orientation in the vehicle of condemning att itudes toward such individuals. This would likely be due to a person being more secure through having already explored identity-based options for her/himself. Hereks (1984) theory that attitudes may serve a variety of functions (in this case, symbolic and defensiv e), and that the functions may interact with and effect one another in determining a persons final attitude struct ure, is supported. Research Questions Apart from the above hypotheses, one of th e aims of the current study was to explore whether previous findings in the literature regarding attitudes toward LG individuals, based

PAGE 75

75 largely in samples of college students, could be replicated in an adult-aged sample. As was demonstrated by this studys results, only some of the results from previous studies generalized to the present sample. Research question 1a suggested that, as in previous studies based in college students (Kite & Whitley, 1996, 1998), males would have more negative attitudes toward LG individuals than would females, but that this difference would be eliminated (fully mediated) when controlling for gender role values as measured by ATW. Th e current findings demonstrate that this statement is true for adults as it is fo r college-aged individuals. As suggested by Kite (1994), non-LG individuals evaluati ons of gay men and lesbians are at least partly grounded in a larger belief system concerning gender roles. If one has physical or behavi oral traits that might traditionally be labeled masculine, then according to traditional gender role beliefs and values, stereotypically masculine behavior s and characteristics should be a natural extension of that individuals gender role. As previously reported, th e literature on gender roles suggests that male gender roles are more narrowly defined an d inflexible than are those for women (Bem, 1993), and violating male gender roles is more obvious ly offensive than violation of female roles (Herek, 1988). This studys fi nding clarifies that for adults as well as college student-aged individuals, ones gender does not affect attitudes toward LG indi viduals, despit e men typically endorsing more condemning attitudes. Rather, gender role attitudes and beliefs, which may be more inflexibly espoused by males than by female s, are the deciding factor in formation of attitudes toward LG people. Carrying the above idea forward, and in line w ith previous findings in the literature (DeLamater, 1987; Herek, 2002; Herek & Ca pitanio, 1999; Louderback & Whitley, 1997), research question 1b suggested that a male gender in the current sample would be related to less

PAGE 76

76 condemning attitudes toward lesbians and more condemning attitudes toward gay men. Further, and still in line with previous findings (L ouderback & Whitley, 1997), the question postulated that the above attitudinal differences on the part of heterosexual males in the sample would be eliminated when erotic values of LG sexuali ty are controlled. However, when the above question was tested using the current data, a male gender was not significan tly related at all to attitudes toward gay men or lesbia ns (considered separately). This suggests that attitudes toward LG people function as a single cons truct in the current sample, a nd the gender of the LG person is not significant in the formation of tolerant/intole rant attitudes. As demonstrated in research question 1a, it is an overall viol ation of gender roles, rather than a gender (or an object of attitudes, as suggested by this question) that is important in th e current sample when regarding attitudes toward LG people. Research Question 2 tested the suggestion that Authoritarianism Social Dominance Orientation, and Religious Service Attendance w ould significantly predict negative attitudes toward LG individuals. Research question 2a posited that when cont rolling for the above variables, race would not be a significant predictor of attitudes toward LG attitudes. Interestingly, only authoritarianism among the variables listed above was significant in predicting attitudes toward LG individuals. This is similar to previous findings in the literature which suggest that authoritarianism is a mo tivating force behind some religious-based LG prejudice (i.e., such as that dea ling with fundamentalist religious belief, as in Laythe, Finkel, & Kirkpatrick, 2001), and with prejudice again LG people in general (Altemeyer, 1981, 1998, 2001; Ekehammar, Akrami, Gylje, & Zakri sson, 2004; Laythe, Finke l, & Kirkpatrick, 2001;Whitley, 1999; Whitley & Lee 2000); however, it is contrary to other findings which specifically found religious servi ce attendance to predict an indi viduals level of tolerance

PAGE 77

77 toward LG people (Schulte & Battle, 2004). Th is may be due to the latter studys focus on African American individuals, whereas the current study included participants of various races and ethnicities. Lemelle and Battle (2004) reported that if one controls for religious service attendance, ethnicity-based differences in tolerance levels toward LG people are eliminated. Analyses for research question 2a, though, revealed that even when controlling for the influence of religious service attendance, an African American racial id entification remained a significant predictor of attitudes toward LG people. While not technica lly significant, identifying oneself as Hispanic was also nearly significant in this regard (p = .05). In attempting to understand this difference and postulating that it may reside in religious differences among ethnicitie s/races, it is important to note that this study did not rega rd all possible religious -based variables that have been used in previous studies, such as fundamentalism or intr insic versus extrinsic re ligiosity (Herek, 1987). As such, the motivating religious factors behind racial differen ces in attitudes toward LG individuals is unknown. It may be that different races/ethnicitie s understandings of appropriate gender roles (i.e., some Hispanic cultures machismo and Marianisma) are a possible alternative key to understa nding the above differences. Limitations and Suggestions for Further Research A main limitation of the current study is relatively small cell sizes among some racial/ethnic identifications (i.e., Hawaiian/P acific Islander), thus preventing complete understanding of the role of race/ethnicity in attitudes toward LG people. Although previous studies suggest that when controlling for religio us factors, there are no difference among some races/ethnicities, the current study did not replicat e those findings. It may be useful to obtain samples composed entirely of a particular racial/ethnic group, a nd to employ more variables that

PAGE 78

78 may be theoretically relevant to that groups cultural beliefs and values. The difficulty of this undertaking may be formidable, though, as the ef fort required to accumulate even moderate diversity in the current sample was significant. Despite such difficulties, gaining further understanding of various races and ethnicities attitudes toward LG people is important, especially as LG individuals continue to openly disclose their sexual/affectional orientations and lobby for greater civil rights. A future study co uld replicate the current study using a larger sample, or a sample with greater diversity, possibly allowing for such an understanding. Additionally, noting that the current sample pr oduced significant skewness and kurtosis on some of the variables under stu dy, greater effort could be put fort h to gather a more well rounded sample with more normally distributed scores. Another important limitation is that this is s correlational study. Th e causal direction of the statistically significant findi ngs may only be postulated, and ev en then, only weakly so. For example, in the test of Hypot hesis 1, authoritarianism predicte d higher levels of commitment, and lower levels of exploration. It is empirically unjustified to suggest that a conservative or authoritarian attitude is detrimental to aspects of psychological development. Equally unjustified is the suggestion that all indi viduals with healthy psychologica l development will develop and open-minded and tolerant stance toward LG individuals. Although the current study demonstrates a clear relationship between the two factors, causation or a directional relationship between them cannot be drawn from the current data. An experimental design, or perhaps a longitudinal study that tracks indivi duals through a college career, might be useful in clarifying this path. Another conceivable limitation is the use of an Internet-based sample to the exclusion of a more random sampling method. It is possible th at some unexpected findings in the current

PAGE 79

79 study are due to sampling bias gene rated through a limited arena for data collection. It is also possible that a greater degree of racial/ethnic diversity could ha ve been obtained through more direct recruitment from community-based organi zations. However, the empirical grounding for such possibilities is not altogether clear. A significant study which compared Internet versus paper-and-pencil sampling methods found that bo th means of data collection resulted in approximately equivalent ratios of SES groupi ngs, racial groupings, and response patterns (Gosling, Vazire, Srivastava, & John, 2005). Noneth eless, the results of a future study that replicates the current study, but using more varied methods of data collection, would be interesting. Other future work might explore the unexp ected findings from the current study. For example, although previous work indicated that heterosexual men would hold more condemning attitudes toward gay men than towa rd lesbians, this studys results did not replicate that finding. It seems possible that greater exposure of the general population to gay men and lesbians over the past decade may have eliminated differences in attitudes toward them. However, proof that the current findings are not anomalous would be necessary before that hypothesis, or other similar hypotheses, could be put forward. Implications for Psychotherapy and Counseling As prejudice against LG people sometimes l eads to limited civil rights and violence against that population, the pr esent study has important imp lications for psychotherapy, especially that geared toward ameliorating social injustices and decreasing the intolerance of individuals who are intolerant of LG people. At the center of this studys findings is that ego identity development, specifically the explorator y aspect of such development, is related to greater tolerance toward LG indivi duals. Even more interestingl y, such ego identity exploration

PAGE 80

80 tempers or moderates the prejudicial aspects of an authoritarian pers onality. Thus, if an individual expresses negative views toward LG individuals, psychotherapi sts may promote social justice for the LG community not necessarily by confronting thos e views directly, but rather by promoting greater exploration of life issues in general. This exploration appears to promote greater tolerance toward LG people in an indirect way. Additionally, psychoeducation fo r those in the LG community might focus on promoting tolerance not only by bringing dir ect conflict against authoritaria n political or religious groups, but by allowing LG people to be known more personally to these groups. This study demonstrates that such interpersonal knowledge prom otes tolerance in and of itself. If it is true that such exposure promotes ego identity explorati on, then this tolerance will be somewhat stable even in the face of the most ri gorous authoritarian stance. In this way, the LG community may be its own agent of social justice. Both this idea, as well as the concept of engendering ego identity exploration in non-LG individuals, coul d be useful in planning outreach projects in community centers or othe r mental health centers. In summary, the fact that ego identity, most specifically the explor ation aspect of ego identity, is related to tolerance of LG people is a new finding in the liter ature on both attitudes toward LG individuals and on ego identity. While the correlational nature of this research limits ones ability to drawn causal implications fr om the study, this study does demonstrate a clear relation among various socially based attitudes a nd ones own ego identity development. Most interestingly, a more mature, well ex plored sense of self is related to ones ability to tolerate LG people, even in the context of an authoritarian-type personality or set of attitudes. As this type of personality is typically more conde mning of societal difference in general, and definitely toward LG people, the fact that a more mature ego can promote tolerance in even such an intrapsychic

PAGE 81

81 context is encouraging. In addi tion to providing empiri cal support for Hereks (1984) theory of LG prejudice, this study provides empirical grou nding for psychotherapeutic and outreach-based interventions aimed at reducing in tolerance, and promoting social justice for LG individuals and their community.

PAGE 82

82 APPENDIX A DEMOGRAPHIC QUENTIONNAIRE (1) What is your current age?______________ (2) What is your race/ethnicity? ______American Indian/Alaska Native ______Asian/Asian American ______Black/African American ______Hispanic/Latino ______Native Hawaiian or Other Pacific Islander ______White/European American ______Biracial/Multiracial ______Other (Please specify):_________________________________ (3) Where do you currently live? ______United States (East Coast) ______United States (West Coast) ______United States (Northern Midwest) ______United States (Southern Midwest) ______United States (Southern) ______United States (Alaska) ______United States (Hawaii) ______Canada ______Mexico ______Central America ______South American ______Europe ______Asia ______Africa ______Australia ______Other (Please Specify):_____________________________ (4) In what type of town or city do you primarily live? ______Urban ______Suburban ______Rural (population of less than 2500) (5) What is your current y early personal gross income? ______Under $4,999 ______Between $5,000 and $14,999 ______Between $15,000 and $24,999 ______Between $25,000 and $34,999 ______Between $35,000 and $49,999 ______Between $50,000 and $74,999 ______Between $75,000 and $99,999

PAGE 83

83 ______Between $100,000 and $149,999 ______More than $150,000 (6) What is your highest level of education? ______Some High School ______High School Graduate/Equivalent ______Vocational Training and/or Certificate ______Some College/Unive rsity (no degree) ______Associates Degree(s) ______Bachelors Degree(s) ______Some Graduate Study (no graduate degree) ______Masters Degree(s) ______Doctorate and/or Po stgraduate Degree(s) ______Other (please specify):_____________________________________ (7) What is your gender? ______male ______female ______other (please specify)_________________________

PAGE 84

84 APPENDIX B THE ATTITUDES TOWARD LESB IANS AND GAY MEN SCALE Below is a list of statements different people have made about men who are gay (homosexual), or women who are lesbian (homosex ual). Please mark a number from 1 (strongly disagree) to 5 (strongly agree) to describe your degree of agreement with each item. 1 = Strongly disagree 2 = Disagree somewhat 3 = Cant decide 4 = Agree somewhat 5 = Strongly Agree (1) Lesbians just cant fit into our society. (2) State laws regulating private, consen ting lesbian behavior should be loosened.* (3) Female homosexuality is a sin. (4) Female homosexuality in itself is no probl em, but what society makes of it can be a problem.* (5) Lesbians are sick. (6) I think male homosexuals are disgusting. (7) Male homosexuality is a perversion. (8) Just as in other species, male homosexuality is a natural expression of sexuality in human men.* (9) Homosexual behavior between two men is just plain wrong. (10) Male homosexuality is merely a di fferent kind of lifestyle that should not be condemned.* (11) How many times have you personally enco untered a person (i.e., friend, acquaintance, relative, distant relative) who you knew was a lesbian or gay man? ______0 ______ 1 ______ 2 ______ 3 ______ 4 or 5 ______ 6 to 9 ______ 10 or more

PAGE 85

85 (12) If you have known a lesbian or gay man, ho w would you rate the relationship in terms of personal closeness? ______ Ive never known a lesbian or gay man ______ very distant, not well known ______ somewhat distant, not very well known ______ somewhat close, somewhat well known ______ very close, very well known (13) How frequently do you attend a religious service (i.e., churc h, temple, mosque, or other) ______ never ______ once or twice a year ______ once or twice a month ______ once a week ______ more than once a week (14) On the following scale, with indicati ng the opposite sex, and indicating the same sex, how would you describe the people to whom you are sexually attracted? _____Always the opposite sex from me _____Almost always the opposite sex, ve ry occasionally the same sex _____Sometimes the opposite sex, occasionally the same sex _____About equally the opposite sex and the same sex _____Sometimes the same sex, occasionally the opposite sex _____Almost always the opposite sex, ve ry occasionally the same sex _____Always the same sex as me (15) How would you describe your sexual orientation? ______ heterosexual ______ gay ______ lesbian ______ bisexual *These items are reverse scored.

PAGE 86

86 APPENDIX C THE RIGHT-WING AUTHORITARI ANISM SCALE, SHORT FORM Below is a list of questions concerning a variety of social issues. You will probably find that you agree with some of the statements, and disagree with others, to varying ex tents. Please indicate your reaction to each statement by checking th e answers according to the following scale: 1 = Strongly disagree 2 = Moderately disagree 3 = Slightly disagree 4 = Neutral 5 = Slightly agree 6 = Moderately agree 7 = Strongly agree (1) Our country needs a powerful leader, in orde r to destroy the radical and immoral currents prevailing in society today. (2) Our country needs free thinkers, who will have the courage to stand up against traditional ways, even if this upsets many people.* (3) The old-fashioned ways a nd old-fashioned values still show the best way to live. (4) Our society would be better off if we s howed tolerance and understanding for untraditional values and opinions.* (5) Gods laws about abortion, po rnography and marriage must be strictly followed before it is too late; violations must be punished. (6) The society needs to show openness towards pe ople thinking differently, rather than a strong leader; the world is not part icularly evil or dangerous.* (7) It would be best if newspape rs were censored so that people would not be able to get hold of destructive and disgusting material. (8) Many good people challenge the state, criticize the church, and ignore the normal way of living.* (9) Our forefathers ought to be hono red more for the way they have built our society, and at the same time we ought to put an e nd to those forces destroying it. (10) People ought to pay less attention to the Bible and religion, and instead they ought to develop their own moral standards.* (11) There are many radical, immoral people trying to ruin things; the so ciety ought to stop them.

PAGE 87

87 (12) It is better to accept bad literature than to censor it.* (13) Facts show that we have to be harder against crime and sexual immorality, in order to uphold law and order. (14) The situation in the society of today woul d be improved if troublemakers were treated with reason and humanity.* (15) If society so wants, it is th e duty of every true citizen to he lp eliminate the evil that poisons our country from within. *These items are reverse scored.

PAGE 88

88 APPENDIX D THE SOCIAL DOMINANCE ORIENTATI ON SCALE, ABBREVIATED VERSION Which of the following objects or statements do you have a positive or negative feeling towards? Beside each object or statement, mark a number from to which represents the degree of your positive or negative feeling. 1 = Very positive 2 = Positive 3 = Slightly positive 4 = Neither positive nor negative 5 = Slightly negative 6 = Negative 7 = Very negative (1) Some groups of people are simply inferior to other groups. (2) Its OK if some groups have more of a chance in life than others. (3) To get ahead in life, it is sometimes necessary to step on other groups. (4) Inferior groups should stay in their place. (5) Group equality should be our ideal.* (6) We should do what we can to eq ualize conditions for different groups.* (7) Increased social equality.* (8) We would have fewer problems if we treated people more equally.* *These items are reverse scored.

PAGE 89

89 APPENDIX E THE ATTITUDES TOWA RD WOMEN SCALE Please mark a number from 1 (Disagree Strongly) to 4 (Agree Strongly) to describe your degree of agreement with each item. (Higher sc ores indicate more egalitarian attitutdes) 1 = Agree strongly 2 = Agree somewhat 3 = Neutral 4 = Disagree somewhat 5 = Disagree strongly (1) Swearing and obscenity are more repulsive in the speech of a woman than a man. (2) Under modern economic conditions, with women active outside the home, men should share in household tasks such as wash ing dishes and doing laundry.* (3) It is insulting to women to have the obey clause still in the marriage service.* (4) A woman should be as free as a man to propose marriage.* (5) Women should worry less about their righ ts and more about becoming good wives and mothers. (6) Women earning as much as their dates should bear equally the expense when they go out together.* (7) Women should assume their rightful place in business and all the professions along with men.* (8) A woman should not expect to go to exactly the same places or to have quite the same freedom of action as a man. (9) Sons in a family should be given more en couragement to go to college than daughters. (10) It is ridiculous for a woman to run a locomotive and for a man to darn socks. (11) In general, the father should have greater authority than the mother in the bringing up of children. (12) The intellectual leadersh ip of a community should be largely in the hands of men. (13) Economic and social freedom is worth far more to women than acceptance of the ideal of femininity, which has been set up by men.*

PAGE 90

90 (14) There are many jobs in which men should be given preference over women in being hired or promoted. (15) Women should be given equal opportunity with men for apprenticeship in the various trades.* *These items are reverse scored

PAGE 91

91 APPENDIX F EGO IDENTITY PROCESS QUESTIONNAIRE Listed below are a number of stat ements describing different types of behavior. Please indicate how you feel about each statement. 1 = Strongly disagree 2 = Disagree 3 = Slightly disagree 4 = Slightly agree 5 = Agree 6 = Strongly agree (1) I have definitely decided on the occupation that I want to pursue. (2) I dont expect to change my political principles and ideals. (3) I have considered adopting diffe rent kinds of religious beliefs. (4) There has never been a n eed to question my values.* (5) I am very confident about what kinds of friends are best for me. (6) My ideas about mens and womens roles have never changed as I became older.* (7) I will always vote for the same political party. (8) I have firmly held views c oncerning my role in my family. (9) I have engaged in several discussions con cerning behaviors involved in dating relationships. (10) I have considered differe nt political views thoughtfully. (11) I have never questioned my views concer ning what kind of friend is best for me.* (12) My values are likely to change in the future.* (13) When I talk to people about re ligion, I make sure to voice my opinion. (14) I am not sure about what type of dating relationship is best for me.* (15) I have not felt the need to refl ect upon the importance I place on my family.* (16) Regarding religion, my beliefs are likely to cha nge in the near future.* (17) I have definite views regarding the ways in which men and women should behave.

PAGE 92

92 (18) I have tried to learn a bout different occupational fields to find the best one for me. (19) I have undergone several experiences that made me ch ange my views on mens and womens roles. (20) I have consistently re-examined many differ ent values in order to find the ones which are best for me. (21) I think what I look for in a fr iend could change in the future.* (22) I have questioned what ki nd of date is right for me. (23) I am unlikely to alter my vocational goals. (24) I have evaluated many ways in wh ich I fit into my family structure. (25) My ideas about mens and womens roles will never change. (26) I have never questioned my political beliefs.* (27) I have had many experiences that led me to review the qualities that I would like my friends to have. (28) I have discussed religious matters with a number of people who believe differently than I do. (29) I am not sure that the va lues I hold are right for me.* (30) I have never questioned my occupational aspirations.* (31) The extent to which I value my family is likely to change in the future.* (32) My beliefs about dating are firmly held. *These items are reverse scored.

PAGE 93

93 APPENDIX G MEASUREMENT OF EROTIC VALU E OF LESBIAN AND GAY SEX Below is a list of statements different people have made about men who are gay (homosexual), or women who are lesbian (homosexual). Pleas e mark a number from 1 (very strongly disagree) to 9 (very strongly agree) to describe your degree of agreement with each item. 1 = Very strongly disagree 2 = Strongly disagree 3 = Moderately disagree 4 = Slightly disagree 5 = Neutral 6 = Slightly agree 7 = Moderately agree 8 = Strongly agree 9 = Very strongly agree (1) I find the idea of a man making love to another man erotic. (2) I find the idea of a man making love to another man repulsive.* (3) I think that I would be sexually aroused by watching two men make love. (4) I have viewed pornographic materi als involving male homosexual acts. (5) I find the idea of a woman maki ng love to another woman erotic. (6) I find the idea of a woman making love to another woman repulsive.* (7) I think that I would be sexually ar oused by watching two women make love. (8) I have viewed pornographic ma terials involving lesbian acts. *These items are reverse scored.

PAGE 94

94 APPENDIX H MARLOWE-CROWNE SOCIAL DESIRA BILITY SCALE, SHORT FORM Please indicate either true or false to desc ribe your degree of agr eement with each item. (1) It is sometimes hard for me to go on with my work if I am not encouraged. (2) I sometimes feel resentful when I dont get my way. (3) On a few occasions, I have given up doing so mething because I thought too little of my ability. (4) There have been times when I felt like re belling against people in authority even though I knew they were right. (5) No matter who Im talking to, Im always a good listener. (6) There have been occasions wh en I took advantage of someone. (7) Im always willing to admit it when I make a mistake. (8) I sometimes try to get even rather than forgive and forget. (9) I am always courteous, even to people who are disagreeable. (10) I have never been irked when people ex pressed ideas very different from my own. (11) There have been times when I was qui te jealous of the good fortune of others. (12) I am sometimes irritated by people who ask favors of me. (13) I have never deliberately said so mething that hurt so meones feelings.

PAGE 95

95 APPENDIX I EXAMPLE ADVERTISEMENT FOR PA RTICIPANT RECRUITMENT I am a psychology researcher at the Universi ty of Florida, and am conducting a study on how various people respond to im portant social issues. The input of African American participants is often neglected by researchers, and the resulting research doesn't accurately reflect how all people in this country feel or think. If you are an African American person and would like to contribute to the study, it may be accessed at http://survey.psych.uf l.edu/websocial. The survey takes approximately 20 minutes to complete, and the study has been approved by th e University of Florida's Institutional Review Board. Once you click the consent link at the bottom of the web page above, the survey begins. If you have questions or concerns please feel free to contact me at the e-mail address above, or using information found at the survey link above. Thank you to anyone who is willing to participat e. Please feel free to forward the link to others who might be willing to participate as well. Daniel Potoczniak, M.S.Ed. Ph.D. Candidate, Counseling Psychology Program University of Florida

PAGE 96

96 APPENDIX J INFORMED CONSENT Informed Consent Protocol Title: The Influence of Ego Identity on Heterosexual Individuals Attitudes toward Lesbian and Gay Individuals Please read this consent document carefully be fore you decide to participate in this study. Purpose of the research study: The purpose of this study is to investigate how individuals regard different types of people, such as women versus men, or lesbians versus gay men. An indicator of human development and of various aspects of personality will be used to examine how various attitudes might be explained. As such, this study involves research. What you will be asked to do in this study: You will be asked to complete a series of 102 quest ions concerning how you feel about a number of topics, such as your ideas on politics, religion, values family, friends, and significant others, to name a few. You do not have to answer any question yo u do not wish to answer. You will then be asked to provide some information about yourself, such as your gender, age, race, sexual orientation, and how attracted you are to people of another sex/gend er, or of the same sex/gender. You will not be asked for your name or any other information that could identify you. Time required: Approximately 20 minutes. Risks and Benefits: This study has been approved by the Universi ty of Florida Institutional Review Board (IRB). There are no anticipated risks associated with this study, other than the potential minimal to mild discomfort that could potentially arise as a resul t of being asked personal questions regarding the subject areas described above. The risk of such discom fort should be minimal, if it exists at all. You may directly benefit through increased self-awareness and self-understanding as you consider appropriate responses in completing the surv ey. Broader benefits of this study to others include a more comprehensive social understanding of how some types of beliefs or values affect attitudes toward lesbians and gay men. Additionally through possible publication of the results, this study may inform the fields of personal counseling and career counseling. Compensation: You will not be compensated in any form for your participation in this study. Confidentiality: You will not be asked to provide any identifying information about yourself, and no record will be kept of your participation in any way. Voluntary participation: Your participation in this study is completely volu ntary. There is no penalty for not participating.

PAGE 97

97 Right to withdraw from the study: You have the right to withdraw from the study at any time without consequence. Technical Requirements: You may complete this survey on any computer with Internet access using any later version Internet browser with javascript and cookies enabled. Your browser must also be able to display 128-bit encrypted web pages (which most Internet browsers support). If you experience technical difficulties accessing the survey, and you have ensured that javascript and cookies are enabled within your Internet browser, please send an e-mail to danpot@ufl.edu to report your technical difficulties, and try accessing the survey once again the following day. You may want to also try accessing the survey from a different computer. Handling of data collected in this survey: Anonymity of data: You will not be personally identi fiable by the data you submit via this Internetbased survey so you are encouraged to answer all items honestly and completely. The data gathered from this study may be presented in a final rese arch paper in aggregate and/or individual response formats, however you will not be personally identifiab le by the data presented within the final study. The data you submit via this survey is kept secure and private. The collected survey data will be stored on highly secure and firewalled servers, which can only be accessed by the research Investigator via password-protected protocols. Survey data submissions cannot be traced back to the original research participant, and although the IP address of the computers on which this survey is accessed for system administration and record keeping purposes, no connection will be made between you and your computer's IP address. Whom to contact if you have questions about the study: For questions concerning the study or to request a copy of the results contact Daniel Potoczniak, M.S.Ed., Graduate Student, Department of Psychol ogy, University of Florida, P.O. Box 112250, Gainesville, Florida, 32611. P hone: (352) 392-0601 x536, E-mail: danpot@ufl.edu ; or, Greg Neimeyer, Ph.D., Professor, at the same addre ss as above, phone: (352) 392-0601, x257, E-mail: Neimeyer@ufl.edu For questions or concerns about your rights as a research participant, contact the UFIRB office, P.O. Box 112250, University of Florida, Gainesv ille, FL 32611-2250. Phone: (352) 392-0433. Agreement: By completing the following survey, you agree that you have read the above information; that you voluntarily agree to participate in this study; and that you have received a copy of this description either by paper or electronically. Please click on the button below if you agree.

PAGE 98

98 LIST OF REFERENCES Adams, H.E., Wright, L.W., & Lohr, B.A. Is homophobia associated with homosexual arousal? Journal of Abnormal Psychology, 105 (3), 440-445. Allport, G. (1954). The nature of prejudice New York: Addison-Wesley. Allport, G.W., & Ross, J.M. (1967). Pers onal religious orient ation and prejudice. Journal of Personality and Social Psychology, 5 432-443. Altemeyer, B. (1981). Right-wing aut horitarianism. Winnipeg, Canada: University of Manitoba Press. Altemeyer, B. (1996). The authoritarian specter Cambridge, MA: Harvard University Press. Altemeyer, B. (2001). Changes in attitudes toward homosexuals. Journal of Homosexuality, 42 (2), 63-75. Altemeyer, B., & Hunsberger, B. (1992). Author itarianism, religious fundamentalism, quest, and prejudice. International Journal for th e Psychology of Religion, 2 113-133. American Psychiatric Association. (1980). Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition Washington, DC: The Ameri can Psychiatric Association. Anderssen, N. (2002). Does contact with lesbians and gays lead to friendlier attitudes? A two year longitudinal study. Journal of Community & Applied Social Psychology, 12 124136. Balistreri, E., Busch-Rossnagel, N.A., & Geisin ger, K.F. (1995). Development and preliminary validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18 179192. Baron, R.M., & Kenny, D.A. (1986). The moderator-m ediator variable dis tinction in social psychological research: Con ceptual, strategic, and st atistical considerations. Journal of Personality and Social Psychology, 51 1173-1182. Battle, J., & Lemelle, A.J. (2002). Gender differences in African American attitudes toward gay males. The Western Journal of Black Studies, 26 (3), 134-139. Bem, S.L. (1993). The lenses of gender: Transformi ng the debate on sexual inequality. New Haven, CT: Yale University Press. Biernat, M. (2003). Toward a br oader view of social stereo typing. American Psychologist, 58, 1019-1027.

PAGE 99

99 Boswell, J. (1980). Christianity, Social Tolerance, and Homosexuality. Chicago: The University of Chicago Press. Broderick, R.C., & Broderick, V. (1990). The Catholic Encyclopedia Nashville, TN: Nelson Reference. Chauncey, G. (1994). Gay New York: Gender, Urban Culture, and the Making of the Gay Male World, 1890 1940. New York: Basic Books. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Cohen, J. (1992). Statistical power analysis. Current Directions in Psychological Science 1, 98101. DeLamater, J. (1987). Gender differences in sexual scenarios. In K. Kelley (Ed.), Females, males and sexuality: Theories and research (pp. 127139). Albany: Status University of New York Press. Donahue, M.J. (1985). Intrinsic and extrinsic religiousness: Review and meta-analysis. Journal of Personality and Social Psychology, 48 400-419. Ekehammar, B., Akrami, N., Gylje, M., & Zakrisson, I. (2004). What matters most to prejudice: Big five personality, social dominance orie ntation, or right-wing authoritarianism? European Journal of Personality 18 463-482. Erdfelder, E., Faul, F., & Buchner, A. (1996) GPOWER: A general power analysis program. Behavioral Research Methods, Instruments & Computers, 28 (1), 1-11. Erikson, E.H. (1968). Identity: Youth and Crisis. New York: Norton. Falkenberg, S.D., Hindman, C.D., & Masey, D. (1983, March). Measuring Attitudes Toward Males in Society. Paper presented at the S outheastern Psychologica l Association annual meeting, Atlanta, GA. Frazier, P. A., Tix, A. P. & Barron, K. E. (2004) Testing Moderator and Mediator Effects in Counseling Psychology Research. Journal of Counseling Psychology, 51, 115-134. Freud, S. (1951). Historical notes: A letter from Freud. The American Journal of Psychiatry, 107 (10), 786-787. Gosling, S.D., Vazire, S., Srivastava,S. John, O. P. (2005). Should we trust web-based studies? A comparative analysis of six preconcep tions about internet questionnaires. American Psychologist, 59 (2), 93-104.

PAGE 100

100 Hahn, E.D. (1993). Sex-role Attitude Change and the Attitudes toward Women Scale. Unpublished masters thesis. The Universi ty of Texas at Austin, Austin, TX. Heaven, P. C. L., & Bucci, S. (2001). Right-win g authoritarianism, social dominance orientation and personality: An analysis using the IPIP measure. European Journal of Personality, 15, 49-56. Herek, G.M. (1984). Beyond Homophobia: A soci al psychological perspective on attitudes toward lesbians and gay men. Journal of Homosexuality, 10 (1-2), 1-21. Herek, G.M. (1987). Religious orientation and prejudice: A comparison of racial and sexual attitudes. Personality and Social Psychology Bulletin, 13 (1), 34-44. Herek, G.M. (1988). Heterosexuals attitudes to ward lesbians and gay men: Correlates and gender differences. Journal of Sex Research, 25 451-477. Herek, G.M. (1994). Assessing attitudes toward le sbians and gay men: A review of empirical research with the ATLG scale. In B. Greene & G.M. Herek (Eds.), Lesbian and gay psychology (pp. 206-228). Thousand Oaks, CA: Sage. Herek, G.M. (2000). The psyc hology of sexual prejudice. Current Directions in Psychological Science, 9 (1), 19-22. Herek, G.M. (2000). Sexual prejudi ce and gender: Do heterosexua ls attitudes toward lesbians and gay men differ? Journal of Social Issues 56 (2), 251-266. Herek, G.M. (2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66 40-66. Herek, G.M., & Capitanio, J. (1996). Some of my be st friends: Intergroup contact, concealable stigma, and heterosexuals attitudes toward gay men and lesbians. Personality and Social Psychology Bulletin 22 (4), 412-424. Herek, G.M., & Capitanio, J. (1999). Sex differen ces in how heterosexual think about lesbians and gay men: Evidence from survey context effects. The Journal of Sex Research, 36 (4), 348-360. Herek, G.M., Gillis, J., & Cogan, J. (1999). Psychol ogical sequelae of hate crime victimization among lesbian, gay and bisexual adults. Journal of Consulting and Clinical Psychology, 67 945-951. Hooker, E. (1957). The adjustment of the male overt homosexual. Journal of Projective Techniques, 21, 18-31. Katz, D. (1960). The functional appr oach to the study of attitudes. Public Opinion Quarterly, 24 (2), 163-204.

PAGE 101

101 Kite, M.E. (1994). When perceptions meet reality: Individual differences in reactions to lesbians and gay men. In B.Greene & G.M. Herek, Lesbian and gay psychology: Theory, research, and clin ical applications (pp 25-53). Thousand Oaks, CA: Sage Publications. Kite, M.E., & Whitley, B.E., Jr. (1996). Sex differe nces in attitudes toward homosexual persons, behavior, and civil righ ts: A meta-analysis. Personality and Social Psychology Bulletin, 22 336-353. Kite, M.E., & Whitley, B.E., Jr. (1998). Do hetero sexual women and men differ in their attitudes toward homosexuality? A conceptual and me thodological analysis. In Herek, G.M. (Ed.), Stigma and sexual orientation: Understandi ng prejudice against lesbians, gay men, and bisexuals (pp 39-61). Thousand Oaks, CA: Sage Publications. Landen, M., & Innala, S. (2002). The effect of a biological explanation on attitudes towards homosexual persons: A Swedish national sample study. Nordic Journal of Psychiatry, 56 (3), 181-186. Laythe, B., Finkel, D., & Kirkpatrick, L.A. (2001). Predicting prejudice from religious fundamentalism and right-wing authoritar ianism: A multiple-regression approach. Journal for the Scientif ic Study of Religion, 40 (1), 1-10. Lemelle, A.J., & Battle, J. (2004). Black mascu linity matters in attitudes toward gay males. Journal of Homosexuality, 47 (1), 39-51. Lewis-Beck, M. (1980). Applied regression: An introduction. Beverly Hills, CA: sage Publications Liang, C.T.H., & Alimo, C. (2005). The impact of white heterosexual students interactions on attitudes toward lesbian, gay and bisexual people: A longitudinal study. Journal of College Student Development, 46 (3), 237-250. Loftus, J. (2001). Americas liberalization in attitudes toward homosexuality, 1973 to 1998. American Sociological Review, 66 (5), 762-782. Louderback, L.A., & Whitley, B.E., Jr. (1997). Per ceived erotic value of homosexuality and sexrole attitudes as mediators of sex differences in heterosexual college students attitudes toward lesbians and gay men. The Journal of Sex Research, 34 (2), 175-182. Marcia, J.E. (1964). Determination and Construct Valid ity of Ego Identity Status. Unpublished doctoral dissertation, Ohio State University. Marcia, J.E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3 (5), 551-558.

PAGE 102

102 Marcia, J.E. (1988). Common pro cesses underlying ego identity, cognitive/moral development, and individuation. In D.K. Lapsley & F.C. Power (Eds.), Self, Ego, and Identity: Integrative Approaches, pp 211-266. New York: Springer-Verlag. Marcia, J.E. (1993). The ego ident ity status approach to ego iden tity. In Marcia, J.E., Waterman, A.S., Matteson, D.R., Archer, S.L., & Orlofsky, J.L. (Eds.), Ego Identity: A Handbook for Psychosocial Research New York: Springer-Verlag. McConahay, J.B., & Hough, J.C., Jr. (1976). Symbolic racism. Journal of Social Issues, 32 (1) 23-45. McCrae, R.R., & Costa, P.T. (1987) Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology 52, 81-90. Meyer, I.H. (1995). Minority stre ss and mental health in gay men. Journal of Health and Social Behavior, 36 38-56. Meyer, I.H. (2003). Prejudice, social stress a nd mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin, 129 (5), 674-697. National Coalition of Anti-Violence Programs (NCAVP). (2004, November). FBI releases annual statistics on hate crimes: Annual repo rts deficiencies continue; eighteen-month spike in anti-LGBT hate not acknowledged. Retrieved at http://www.ncavp.org/media/MediaReleas eDetail.aspx?p=1067& d=1106 on 03/01/2006. Negy, C., & Eisenman, R. (2005). A comparison of African American and White college students affective and attitudinal reactions to lesbian, gay, and bisexual individuals: An exploratory study. The Journal of Sex Research, 42 (4), 291-298. Nyberg, K.L., & Alston, J.P. (1977). Ho mosexual labeling by university youths. Adolescence, 12 541-546. Oldham, J.D., & Kasser, T. (1999) Attitude change in response to information that male homosexuality has a biological basis. Journal of Sex & Martial Therapy, 25 (121-124). Plugge-Foust, C., & Strickland, G. (2000). Homophobi a, irrationality, and Christian ideology: Does a relationship exist? Journal of Sex Education and Therapy, 25 (4), 240-244. Pratto, F., Sidanius, J., Stallwor th, L. M., & Malle, B. F., (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741-763. Reynolds, W.M. (1982). Development of reliable and valid short forms of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology, 38 (1), 119-125.

PAGE 103

103 Sakalli, N. (2002). The relationship between se xism and attitudes toward homosexuality in a sample of Turkish college students. Journal of Homosexuality, 42 (3), 53-64. Schulte, L.J., & Battle, J. (2004). The relative importance of ethni city and religion in predicting attitudes towards ga ys and lesbians. Journal of Homosexuality, 47 (2), 127-142. Schwartz, S.J. (2001). The evolution of Eriksoni an and neo-Eriksonian identity theory and research. Identity: An Interna tional Journal of Theory and Research, 1 (1), 7-58. Schwartz, J.P., & Lindley, L.D. (2005). Religious fundamentalism and attachment: Prediction of homophobia. The International Journal for the Psychology of Religion, 15 (2), 145-157. Sirin, S.R., McCreary, D.R., & Mahalik, J.R. (2004) Differential reactions to men and womens gender role transgressions: Pe rceptions of social status, sexual orientation, and value dissimilarity. Journal of Mens Studies, 12 (2), 119-132. Spence, J.T., & Hahn, E.D. (1997). The attitudes toward women scale and attitude change in college students. Psychology of Women Quarterly, 21 17-34. Tabachnick, B.G., and Fidell, L.S. (2001). Using Multivariate Statistics, Fourth Edition. Boston: Allyn and Bacon. Theodore, P.S., & Basow, S.A. (2000). Hetero sexual masculinity and homophobia: A reaction to self? Journal of Homosexuality, 40 (2), 31-48. Tureau, Z.L. (2004). College student identity and attitudes toward gays and lesbians (Doctoral dissertation, University of North Texas, 2004). Dissertation Abstracts International, 64 (9-B), 4657. Van Hoof, A. (1999). The identity status fi eld re-reviewed: An update of unresolved and neglected issues with a view on some alternative approaches. Developmental Review, 19 497-556. Weis, C.B., Jr., & Dain, R.N. (1979). Ego develo pment and sex attitudes in heterosexual and homosexual men and women. Archives of Sexual Behavior, 8 (4), 341-356. Weiss, J.T. (2003). GL vs. BT: The archaeology of biphobia and transphobia within the U.S. gay and lesbian community. Journal of Bisexuality, 3 (3/4), 25-55. Whitley, B.E., Jr. (1987). The relationship of se x-role orientation to heterosexuals attitudes toward homosexuals. Sex Roles, 17 103-113. Whitley, B.E., Jr. (1995). Sex-role orientation and attitudes toward homosexuality: A metaanalysis.

PAGE 104

104 Whitley, B. E., Jr., & Aegisdttir, S. (2000). The gender belief system, authoritarianism, social dominance orientation, and heterosexuals attitudes toward lesbian and gay men. Sex Roles, 42(11/12), 947-967. Whitley, B. E., & Lee, S. E., (2000). The relationsh ip of authoritarianism and related constructs to attitudes towards homosexuality. Journal of Applied Social Psychology, 30, 144-170. Williamson, I. (2000). Internalized homophobia and health issues affecting lesbians and gay men. Health Education Research, 15(1), 97-107. Wittig, M.A., & Grant-Thompson, S. (1998). The u tility of Allport's conditions of intergroup contact for predicting perceptions of im proved racial attitudes and beliefs. Journal of Social Issues 54 (4), p795-812. Wylie, L., & Forest, J. (1992). Religious funda mentalism, right wing authoritarianism, and prejudice. Psychological Reports, 71 1291-1298. Yang, A. (1997). Trends: Attitudes toward homosexuality. Public Opinion Quarterly, 61 477507. Zakrisson, I. (2005). Construction of a short version of the Right -Wing Authoritarianism (RWA) scale. Personality and Indivi dual Differences, 39 863-872.

PAGE 105

105 BIOGRAPHICAL SKETCH Daniel Joseph Potoczniak was born on April 24th, 1973 in Lakewood, NJ, to Joseph John Potoczniak and Jane Marie Potoczniak. He grew up in Brick, NJ for the first 12 years of his life, at which point his family moved to Brielle, NJ. During his high-schoo l years, he attended Christian Brothers Academy in Lincroft, NJ, wher e he dedicated his time to the pursuit of all things French, editing the y earbook, and being a band geek. After graduating from high school, Daniel attended Georgetown University in Washington, D.C., where he majored in English. Upon graduation, he worked in a number of different jobs, incl uding being an assistant to a pastry chef, copyediting childrens books, sel ling eyeglasses, and being a supervisor for the 2000 Census. He also performed volunteer work counseling others on the AIDS/HIV Hotline at the Gay Mens Health Crisis. Finding direction and motivation in counseling ot hers, Daniel returned to graduate school in 2001 at the University of Miami. There, he earned a Mast er of Science degree in Education with a concentration in mental health counseling. In 2003, Daniel matriculated at the University of Florida to begin his Ph.D. in counseling psyc hology, where he was offered the Marshall Criser Presidential Fellowship. His counseling and rese arch interests further developed there, and included a focus on ego identity development acr oss the lifespan, as well as issues that are relevant to the gay, lesbian, bise xual, and transgender community. On July 31st, 2007, Daniel completed his one-year pr e-doctoral internship in counseling psychology at the University of Pennsylvanias Counseling and Psychologi cal Services center. After receiving his Ph.D., Daniel will seek empl oyment that allows a balance among clinical work, likely in a counseling center, a nd more assessment-based activities.