Iowa State University Capstones, Theses and

Retrospective Theses and Dissertations

Dissertations

1993

Three papers on homosexual sexual scripts

Brian Magruder
Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd

Part of the Family, Life Course, and Society Commons

Recommended Citation
Magruder, Brian, "Three papers on homosexual sexual scripts " (1993). Retrospective Theses and Dissertations. 10841.
https://lib.dr.iastate.edu/rtd/10841

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University
Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University
Digital Repository. For more information, please contact [email protected].
 U-M-I
MICROFILMED 1994
 INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the

copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by

sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced

xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.

University Microfilms International

A Bell & Howell Information Company
300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA
313/761-4700 800/521-0600
 Order Number 9414002

Three papers on homosexual sexual scripts

Magruder, Brian, Ph.D.

Iowa State University, 1993

UMI
300 N. ZeebRd.
Ann Arbor, MI 48106
 Three papers on homosexual

sexual scripts

by

Brian Magruder

A Dissertation Submitted to the

Graduate Faculty in Partial Fulfillment of the

Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Department; Sociology
Major; Sociology

Approved;

Signature was redacted for privacy.

In Charge of Major Work

Signature was redacted for privacy.

For the Major Department

Signature was redacted for privacy.

Fop/^e Graduate College

Iowa State University

Ames, Iowa

1993
 ii

TABLE OF CONTENTS

Page

GENERAL INTRODUCTION AND THEORY 1

Introduction 1

The Sexual Script Perspective 5

Organization of Dissertation 28

PAPER 1. ADOLESCENT DISCLOSURE OF HOMOSEXUALITY

TO THEIR FAMILIES 31

ABSTRACT 32

INTRODUCTION 33

Gay and lesbian youths 35

Disclosure and family dynamics 36

THEORETICAL ANALYSIS 40

METHODS 45

Sample 45

Measurement 46

Procedures 49

RESULTS 51

DISCUSSION 55

CONCLUSION 58

BIBLIOGRAPHY 60
 iii

PAPER 2. THE DEFINITION OF ATTRACTIVENESS AND

SENSUALITY: A SCHEMATIC COMPARISON
BETWEEN HOMOSEXUALS AND HETEROSEXUALS 67

ABSTRACT 68

INTRODUCTION 69

REVIEW OF LITERATURE 71

Objective versus subjective definitions of

attractiveness 71

Objective criteria of attractiveness 72

Gender differences 75

A SCHEMATIC ANALYSES 77

Schematics 78

Hypotheses 81

METHODS 83

Homosexual sample 83

Heterosexual sample 85

Measurement 85

Cautions and limitations 89

RESULTS 90

Definitions of attractiveness 90

Definitions of sensuality 92

Definitional differences between attractiveness

and sensuality 94

Group comparisons 95

DISCUSSION 99

BIBLIOGRAPHY 103
 iv

PAPER 3. THE VIOLENT EFFECTS OF DISCRIMINATION

ON GAY AND LESBIAN RELATIONSHIPS 118

ABSTRACT 119

INTRODUCTION 120

MODELING VIOLENCE IN GAY/LESBIAN RELATIONSHIPS 122

Cultural scenario scripts 123

Interrelational scripts 127

Intrapsychic scripts 128

Hypothesized models 130

METHODS 132

Measurement 132

Procedures 137

RESULTS 139

DISCUSSION 150

BIBLIOGRAPHY 154

GENERAL SUMMARY 163

Summary of Results 163

Limitations and Future Research 182

Conclusion 191

BIBLIOGRAPHY 195

APPENDIX 199
 1

GENERAL INTRODUCTION AND THEORY

Introduction

Homosexuality is either a feeling of sexual desires for

persons of one's own sex or sexual relations with a member of

one's own sex, or both (Thio, 1983). The past forty years

have witnessed several shifts in sociological research

concerning the subject of homosexuality. At the onset of the

late forties and up to the early seventies, research on sexual

orientation was a waxing field of study. As research began to

wane on approximating the parameters of the homosexual

population, inquiry into causal development began to dominate

the scientific investigation of homosexuality. This research

trend spanned the early sixties to the early eighties. The

late eighties brought a psychological shift in the sexual

orientation research, focusing primarily on the social

processes of homosexuals, including concepts such as

homosexual identity development. The nineties has undertaken

the scientific study of homosexual relationships, denoting

differences and similarities between heterosexual

relationships. This dissertation will continue with this

trend by exploring homosexual relationships as extrapolated

from a sexual script perspective.

Kinsey and others pioneered the first extensive study of

homosexuality in 1948. Their research established the extent

of participation in homosexual activities by white males

(Kinsey, Pomeroy, and Martin, 1948). Further research has

 2

attempted to approximate the parameters - boundaries - of the

homosexual population by replicating Kinsey et al.'s initial

study. Such researchers have concluded that between ten and

twelve percent of any given population to be of a same-sex

orientation (Athanasiou, Shaver, and Travis, 1970; Hayes and

Oziel, 1976; Hunt, 1974). Applying these percentages to the

current population of the United States would roughly be

equivalent to over thirty million homosexuals residing in the

United States.

With regard to causal development, the origin of

homosexuality and heterosexuality remain a mystery. However,

it seems apparent both biological and psychosocial

(sociological and psychological) factors are implicated in

homosexuality. The extensive work of Money (1980) on genetic

and hormonal factors influencing sexual drives compounded with

the recent research being conducted on the organic dimensions

of the brain provide a convincing argument for the biological

premise of a homosexual orientation. However, other

researchers indicate no clear genetic differences between

homosexuals and heterosexuals (Gagnon, 1977; Masters and

Johnson, 1979).

Sexual identities, perceptions of self as heterosexual,

homosexual, or bisexual in relation to sexual, romantic, and

broader cultural arrangements are constructed through various

forces. The homosexual identity is one of several identities

incorporated into an individual's concept of self. According

 3

to Cass (1983) depending on the context, the homosexual

identity may function as a self-identity, a perceived

identity, a presented identity, or all three^ The homosexual

identity is a self-identity when people see themselves as

homosexual in relation to romantic or sexual encounters and

engagements. It is a perceived identity in situations where

people think or know that others view them as homosexual. It

is a presented identity when people present or announce

themselves as homosexual in concrete social settings, i.e.,

occupational environments. Troiden (1984/1985) expands on

Cass's research by asserting homosexual identity is firmly

established when an individual has developed a positive gay

self-image.

During the past decade, several researchers have proposed

theoretical models, attempting to explain the formation and

continuance of homosexual identities (Cass, 1979, 1983; Lee,

1977; Ponse, 1978; Shafer, 1976; Troiden, 1977, 1979;

Weinberg, 1977, 1978). Although the various models suggest

different stages in the formation of homosexual identities,

striking similarities emerge as transitional growth periods

for homosexual identity development.

The social stigma and sanctioning surrounding

homosexuality in the United States affects both the formation

and expression of homosexuality. Therefore, homosexual

identities are described as developing over a protracted

period of time and involving a number of transitional stages.

 4

In short, homosexual identity formation entails increasing

acceptance of the label "homosexual" as applied to the self in

spite of possible resistance or sanctioning from society. The

homosexual individual, accepting this stigmatized identity,

initiates disclosure of their identity, and finally commits to

the identity through love relationships or membership in the

gay subculture (Cass, 1983; Coleman, 1982; Lee, 1977).

Although a substantial amount of empirical attention has

been given to the formation of homosexuality, its causes and

consequences; little sociological theorizing has been

submitted regarding scripted intimate relationships among

homosexuals. The role of social structure, culture, and

various systems have been neglected in the literature through

the sole analysis of the gay/lesbian relationship as an entity

within itself, apart from cultural context. Not only must

researchers carefully analyze the dynamics of relationships,

and the psyche of those involved; consequently, researchers

must also study the phenomenon under the context of the

structural and cultural arrangements which frame the

relationship. Therefore, a sexual script perspective, which

entails all three dimensions (the individual, the

relationship, and culture), demonstrates a strong utility for

a sociological inquiry into gay/lesbian relationships.

 5

The Sexual Script Perspective

In general, sexual scripts are understood as general

guidelines, procedural diagrams for interpreting sexually

relevant expressions and behaviors (Simon and Gagnon, 1984).

In short, sexual scripts are typologized by three distinct,

yet interconnected, levels of analysis. Cultural scenario

scripts dictate the normatively accepted or shared values

attached to sexuality. Interrelational scripts guide the

contexts of relational sexuality. Intrapsychic scripts define

the motivations, fantasies, and desires for individual

sexuality.

The sexual script perspective actually emerged from

schema theory. The term script, under a schematic definition,

refers to the dynamic knowledge surrounding particular events

and phenomenon; consequently, schema is defined as the static

knowledge unique to a given situation (Schwartz and Reisberg,

1991). In relation to sexual scripts, scripted and schematic

information is geared to the static and dynamic processes

relevant to sexual attitudes, behaviors, and life styles.

Sexual schema and scripts are means for cognitively ordering

normative or redundant and dynamic or interactive processes

associated with sexuality.

The tenets of schema theory are generally centered around

the concepts of active perception and interpretation. As a

microlevel theory, a schema reflects a generalized perception

based on redundant situations, phenomenon, and events (Schank,

 6

1982). Therefore, a schema consists of familiar

understandings about people, places, norms, and actions.

Schema theory does not necessarily explain why a certain

phenomenon may occur, as much as it focuses on the

interpretation of the phenomenon by individuals.

In short, schema theory is a learning-oriented theory.

Schematas are maintained and modified by repetition and other

various learning models (i.e., reinforcement, etc.). The

assumptions represented by schema theory suggest the level of

analysis is the individual. However, schema theory does not

necessarily ignore the social context of the individual.

Socially constructed norms and values are embedded within the

frameworks of schemata, thus establishing certain parameters

for acceptable social behavior.

Not unlike symbolic interactionism, schema theory

emphasizes decoding and encoding processes in the

interpretation of phenomenon. For example, Bartlett (1932)

focused on the effect of relevant schematic patterns,

hypothesizing that a phenomenon which is not relevant or

particularly meaningful to an active schema will not receive

detailed attention. And phenomenon, which is pertinent to an

individual's schema, will be encoded.

The decoding process entails the act of interpretation,

categorizing phenomenon into compartmentalized schematas

resulting from perceptual filtering (Anderson and Pichert,

1978). The filtering process entails decoding mechanisms in

 7

order to "filter out" meaningless information, and cognitiveiy

order relevant information about a given phenomenon. The

filtering process is subject to various schematic mechanisms

which may alter the perception and meaning of the phenomenon,

resulting in source confusion and schematic errors (Spiro,

1977).

The sexual script perspective borrows several key

elements and assumptions from schema theory to explain

sexuality. Again, the focus of a sexual script perspective is

on the formation and maintenance of those schemata which are

sexual in nature. Encoding, decoding, schematic errors, and

the general cognitive ordering of sexual phenomenon and

terminology are incorporated in the sexual script perspective.

However, the two theories diverge on the point of analysis.

Where schema theory tends to view the individual in isolation,

centering only on the cognitions of the individual; sexual

scripts tends to partition schemata into environmental

contexts and effects. In short, schema theory maintains the

individual is the centralized element of interpretation. The

sexual script perspective contends the individual is an active

component, but must be understood in terms of the social

context of the phenomenon.

The domain assumption of the sexual script perspective

maintains humans are primarily social in their learning of

sexuality (Gagnon, 1977). At the beginning of life, humans

are defined as being polymorphously perverse, in other words.

 8

a "blank slate" of sexuality. During childhood and

adolescence, men and women begin to learn and incorporate

general norms and values regarding sexuality via their

socialization agents. By early adulthood, men and women have

developed their sexual scripts concerning sexual behaviors,

attitudes, relationships, desires, and sanctions. Sexual

scripts define the affective and cognitive boundaries for

appropriate and inappropriate sexuality. Scripts designate

acceptable and unacceptable partners, behaviors, fantasies,

techniques, and emotions. Although, sexuality does maintain

an imperative biological element, the sexual script

perspective facilitates attention to important social forces

which guide and shape sexuality. In many cases, these scripts

are congruent with previous gender-role socialization (Levine,

1987). For example, males tend to view sexuality in

instrumental, active means; consequently, females perceive

sexuality in expressive, process-oriented methods (Laws and

Schwartz, 1977).

Cultural scenario scripts reflect the broader cultural

and structural arrangements which represent the official and

public means to legitimate or illegitimate sexual behaviors,

and life styles. Cultural scenario scripts can be

characterized by three distinct categories; procreative,

recreational, and relational (Gagnon and Simon, 1977).

Procreative scripts maintain sexuality is only legitimated for

reproductive purposes. Recreational scripts propose sexuality

 9

is for enjoyment and pleasure, regardless of the relational or

reproductive facets. Relational scripts dictate the context

for appropriate sexual relations. If a culture warrants a

relational script, members in the culture adopt the belief

patterns that sexuality is only permitted and legitimate in

committed, affective relationships.

Interrelational (or interpersonal) scripts serve to

mediate between intrapsychic and cultural scenario scripts.

Interrelational scripts facilitate the norms and values

surrounding sexuality, from the broader social forces, in

preexisting and future relationships with others. However,

interrelational scripts also expedite personal fantasies and

sexual motivations into the relationship. Interrelational

scripts further incorporate the fantasies and sexual

curiosities of the significant other. In short,

interrelational scripts bridge the gap between general

cultural forces and personal desires. The concept of sexual

identity, whether heterosexual or homosexual, emerges at this

particular level. Sexual identity, from an interrelational

script, consists of expectations and the social roles embedded

in sexuality with others, primarily significant others.

Interrelational scripts are constantly being modified

throughout the relationship, and with each new relationship

which follows. Modification of interrelational scripts may

occur when sexual partners or significant others present new,

different, or variant sexual expectations. When opposing

 10

interrelational scripts converge, individuals negotiate new

interrelational scripts for accommodating the disparity, under

the constraints of the prevailing cultural scenario scripts.

Intrapsychic scripts serve as the latent features of

individual sexuality. Intrapsychic scripts are the internal

desires, motivations, fantasies, those characteristics which

foster sexual arousal and excitement. In short, intrapsychic

scripts are responsible for stimulating and maintaining sexual

arousal. Intrapsychic scripts also incorporate those

characteristics which are personality-oriented, such as self-

esteem, body image, feelings of sexual adequacy, and the like.

Such personality features serve to facilitate sexual and

affective motivations.

Sexual scripts can be divided into two categories,

homosexual and heterosexual sexual scripts. Although both are

not mutually exclusive in content, homosexual scripts do

possess unique characteristics which deviate from the dominant

heterosexual script. One might suggest, all participants in a

society are socialized with heterosexual scripts; however,

only integrated homosexuals (maintaining a positive homosexual

V
self-image or identity) are privy to homosexual scripts.

Homosexual sexual scripts incorporate the cultural,

heterosexist norms governing acceptance of homosexuality; and,

in turn, homosexual scripts incorporate the homosexual

subcultural characteristics which reflect those norms and

values specific to the gay subculture. Homosexual scripts

 11

differ from heterosexual scripts in one facet due to the vast

stigma attached to homosexuality. In other words, homosexual

scripts must consequently deal with a divisive stigma attached

to their entire sexual life style. Heterosexual structural

and cultural influences impact on definitions of

attractiveness and sensuality, homosexual identity formation

and development, and relational dynamics as affected by

discriminatory stress, the themes of this dissertation.

As cultural scenario scripts influence the context or

domain in which homosexuality transpires, interrelational

scripts, for homosexuals, represent the dimensions of their

relationships which are disparate from heterosexual

relationships, perhaps due to stigmatizing or isolating

structural influences. Intrapsychic scripts reflect the

internalized dimensions of cultural and interrelational

scripts via motives, and desires relevant to homosexual

expressions, identity salience, and sexual symbolism. All

three levels of scripting are interdynamic processes,

constantly modified throughout the life course by interchanges

with various structural and relational influences.

Theory or perspective?

The sexual script perspective builds its theoretical

interpretation on many sociological theories, borrowing

particular concepts and propositions which place the context

of homosexuality in the eye of the culture or system (cultural

 12

scenario scripts). Questions of sexuality are cultural

constructions, specific to time and place: what sexuality is,

the purposes it serves, its manner of expression, and what it

means to be sexual. Lesbianism and male homosexuality are

similarly constructed and culture bound. Sexual behavior and

responsiveness span the spectrum from exclusive

heterosexuality to exclusive homosexuality; the positions that

people occupy on this spectrum result from social learning in

specific environments.

According to Niklas Luhmann (Turner, 1991), society is

composed of three dynamic "social systems": interaction

systems, organization systems, and societal systems. The

"interaction system" revolves around the presence of the

individual, and those in contact with the individual.

Organizational systems "coordinate the actions of individuals

with respect to specific conditions" (Turner, 1991: 96). The

societal system establishes the boundaries and guidelines for

interactions between individuals and culture. Therefore,

cultural scenario scripts view homosexual identity as grounded

in specific social contexts or interaction systems. Normative

structures prevailing in the societal system govern activity

in specific contexts that determines which roles and

identities may be expressed legitimately. And organizational

systems place limits on the identities and roles that may be

enacted appropriately.
 13

The sexual script perspective treats the homosexual

identity as a construct embedded within an interrelational

script and an intrapsychic script. Further, a homosexual

identity will only be mobilized or set forth in situations

which norms permit it to be presented-an important function

provided by the various subcultures and scenes that constitute

the homosexual subculture. A homosexual intrapsychic script,

and perhaps a homosexual interrelational script, is manifested

only in systems that honor homosexual conduct and allow him or

her to claim the identity or express the corresponding role,

although the experience of homosexual identity may be

triggered when homosexual elements are introduced in

heterosexual systems (Troiden, 1984/1985). The homosexual

identity is suppressed, and becomes latent, when an individual

leaves homosexual systems for more conventional environments.

In other words, the homosexual intrapsychic script is

confounded outside of a cultural scenario script tolerant of

homosexuality.

The most extreme statement of the sexual script

perspective is expressed by sociologist Edward Sagarin (1973,

1975, 1976, 1979). He points out that many people have

homosexual desires, but that relatively few individuals

translate these desires into behavior; fewer yet define

themselves as homosexual in the sense of an intrapsychic

script. He argues that people commit the fallacy of

reification when they define themselves as "being" what they

 14

are "doing." In other words, discontinuity between a

homosexual intrapsychic script and a heterosexual cultural

scenario script results in fragmentation or

compartmentalization in the interrelational script. In order

for a continuous homosexual identity to emerge, from a sexual

script perspective, all three scripts (cultural,

interrelational, and intrapsychic) must be congruous, despite

the cultural resistance provided by the dominant heterosexual

cultural scenario scripts. Therefore, it is imperative to

examine the role of cultural and structural forces which shape

the sexual scripting among homosexuals.

Again, the influence of culture or social systems plays

an integral component in the manifestation of the various

sexual scripts. For example, several patterns of formal and

informal lesbian relationships emerge in nonclass societies

where women have greater autonomy. Among the polygynous

Azande of Africa, for example, where each wife had her own

dwelling and her own plot of land, and controlled whatever

profits from her work she made through trade, some women

established lesbian relationships with their co-wives within

the formal, organizational structure of polygynous marriage.

Husbands could not forbid these arrangements, but the wives

kept them secret to avoid threatening their husbands

(Blackwood, 1985). Lesbian relationships are also reported to

exist between co-wives of the Nyakyusa, another polygynous

African group (Adam, 1985).

 15

In class societies, where women lack autonomy and power,

formal lesbian relationships, if they exist at all, occupy a

marginal status in relation to the dominant culture or the

societal system. In the Near East, for example, lesbian

behavior among women in harems and within the Muslim

institution of purdah was informal. It should be noted that

homosexual behavior is outlawed under Islamic law, and

adultery is punishable by divorce or death (Blackwood, 1985).

Therefore, concepts such as stratification guide the normative

context, ensuing the development of a cultural scenario script

governing homosexual relations.

The ethnographic evidence also suggests that age-

structured homosexuality among unmarried male youths is

extremely common in societies where "bachelorhood" is a

transitional status between childhood and adulthood.

Standards, associated with age appropriate sexual behavior, is

but one of the many aspects enveloped under cultural scenario

scripts. Among the Nyakyusa, for example, young males between

the ages of ten and or eleven and twenty-five separate from

their families to form peer groups that set up entirely new

villages for themselves. "It is generally accepted that these

youths engage in reciprocal homosexual relations" (Adam, 1985:

21).

Ritualized homosexual relations between older males and

younger males are even more common and provide a contrast to

the peer-oriented models described above. An older married

 16

male enters into a role-structured pedagogic/sexual

relationship with a young man. The older man socializes the

youth to male culture and gender functions. A role

differentiation consistent with this "one-way" socialization

emerges: The older man is the provider and the younger man is

the recipient, and this differentiation "structures anal and

oral intercourse" (Adam, 1985; 25). The social categories of

heterosexual, homosexual, and bisexual are meaningless and

irrelevant. The fundamental social division in these

societies is between the relatively "free, adult male citizen

and all others. Sexuality is merely one of many rights

exercised by the free male over women, youth, and slaves"

(Adam, 1985: 22). This example portrays the means in which

the cultural scenario scripts, of a given society, influences

the structure of not only gay and lesbian relationships, but

also heterosexual engagements.

For other societies, older bachelors enter into role-

structured sexual relationships with younger boys. "Gender

ideology plays the primary role in determining sexuality among

males" (Adam, 1985; 22). Homosexuality is associated with

gender differentiation (masculinization of the male) rather

than gender mixing (effiminization of the male), a view that

prevails in the contemporary United States. Sexuality is only

one of the many ways in which males set themselves apart from

women and perpetuate the male group.

 17

The anthropological case studies reveal structural

components in the evolution of homosexual sexual scripts.

Homosexuality is manifested in its behaviors concurrent with

interaction, organization, and societal systems. However,

Luhmann's neofunctionalism is not adequately adept in

describing or explaining the social conflict between the

homosexual role and institutions, for example, in the United

States. Luhmann's analysis also does not easily lend itself

toward explaining why individuals chose expressions of their

social-roles continuously over settings. For a further

explanation of the social-role approach, the essay will now

turn toward Randall Collins' exchange conflict theory, Peter

Blau's structural exchange theory, Claude Levi-Strauss's

structuralism, and Karl Marx's theory of exchange and conflict

to increase the depth of theoretical analysis regarding the

homosexual sexual script perspective.

The tenets of exchange theory are numerous, depending on

the theoretical variation in question. However, the basic

principles underlying all variations of exchange theory can be

reduced to several key axioms. One of the propositions

includes the rationality principle. According to Peter Blau

(Turner, 1991), rationality constitutes the ability to expect

and perceive a profit from one another through a particular

activity. And the more profitable an activity is, the more

likely the activity will be repeated. A second principle is

reciprocity. Reciprocity represents any obligation which

 18

arises when people have exchanged a reward or rewards with one

another. The justice principle asserts established exchange

relations and are likely to be governed by norms of equality

and fairness. The imbalance principle proposes the more

unequal the exchanges between social groups (individuals), the

more likely future exchanges will be unequal and unstable.

These general tenets comprise our understanding of structural

exchange theory from the perspective of both Claude Levi-

Strauss and Peter Blau.

Randall Collins and Karl Marx possess slightly different

theoretical perspectives, guiding their interpretation of

exchange relations. Both theorists view the exchange network

as embedded with conflict. In general, both Marx and Collins

contend exchange relations are based on the availability and

value of particular resources (Turner, 1991). Conflict and

organization arise when those who possess scarce and valued

resources control (suppress) those who do not have access to

such resources.

The rationality of homosexuality precludes the

heterosexual dogma of many structures and institutions in

heterosexist societies, such as the United States. To

understand the conflict and exchange principles of homosexual

identity acquisition, an understanding of the social

structures impairing homosexuality must be emphasized. The

analysis of innate costs and rewards attached to homosexuality

 19

by social structures broadens the sociological insight to

homosexual sexual scripts.

According to Barry Adam (1978), the socio-structural

relationships between heterosexual and homosexual groups are

framed by patterns of domination and subordination. Dominant

groups produce and maintain social order because they control

the bulk of the socially valued resources. According to

Turner's interpretation of Marx's theory of exchange and

conflict, "Those who controlled valued resources have power

over those who do not" (Turner, 1991: 299). Control over

resources enables these groups to create and control social

institutions, which are structured to maximize the life

chances of dominant groups by minimizing those of subordinate

groups. Inferiorization, then, refers to the creation of

social inequality.

The inferiorization of homosexuals is evident at all

levels of society and in all major institutions. Dominant

institutions inferiorize subordinate groups by constructing

the characteristics that allegedly set them apart (e.g.,

effeminacy, same-sex attraction) and that justify their

exclusion from the hierarchy of access. Peter Blau's vision

of exchange systems and macrostructure reiterates a similar

proposition: "the significance of 'shared values' increases,

for it is through such values that indirect exchanges ... are

mediated" (Turner, 1991: 340). "Shared values" could connote

heterosexism, the belief that all individuals are and should

 20

be heterosexual. If institutions are comprised of these

"shared values", then "as people and various forms of

collective organizations become dependent upon particular

networks of indirect exchanges for expected rewards, pressures

for formalizing exchange networks through explicit norms

increase" (Turner, 1991: 342). Institutions of social

control-police and the courts, for example-assume attitudes of

official blindness and neglect toward acts of aggression

leveled against inferiorized groups (e.g., hate crimes, or

discrimination) and "at the same time, the inferiorized

themselves are more frequently subject to arrest, police

harassment, and conviction by courts" (Adam, 1978: 28).

Further, "institutions exert a kind of external constraint on

individuals and various types of collective units, bending

exchange processes to fit their prescriptions and

proscriptions" (Turner, 1991: 342). Blau envisioned

institutions as possessing normative guidelines contingent

with society's "shared values." Institutions project

authority in order to establish social objectives based on

"shared values" (Turner, 1991). In the case of homosexuals,

the mode of sexual expression itself has been criminalized

under many state statutes.

Legal institutions condone inferiorization explicitly by

defending the constitutionality of discriminatory practices

(e.g., state sodomy statutes, physical harassment, etc.); they

routinize inferiorization implicitly by not outlawing

 21

discrimination in employment and place of residence on the

basis of sexual orientation. Economic institutions maintain

inferiorization through discriminatory hiring and barring

homosexuals from certain occupations (e.g., the clergy,

national security), trade unions (e.g., firefighters), and

voluntary associations (e.g., Big Brothers). Political

institutions also remain, for the most part, the monopoly of

noninferiorized people.

Institutions that transmit culture-schools, churches, the

publishing industry, and the mass media-define the personal

and social characteristics of inferiorized people: "The

systematic selection of attributes of inferiorized peoples for

public presentation constructs an image which rationalizes

inferiorized status for both privileged and inferiorized

groups. Inferiorized people discover their 'objective'

identity lives beyond their control; the image of self,

institutionalized by cultural agents, exists alien to their

own experience and self-expression." (Adam, 1978: 30-31).

Inferiorized people tend to be described (if acknowledged at

all) in disrespectable contexts: crime, physical disease,

immorality, mental illness. Homosexuals, for example, have

been variously portrayed as child molesters, mentally ill, and

sinful.

Medical institutions also contribute to inferiorization

by presenting the life patterns of inferiorized people as

unnatural (e.g., the presumption that heterosexuality is the

 22

biological norm), as reflecting disease (e.g., homosexuality

is caused by pathological fears of the opposite sex), as

invalid or unauthentic (e.g., homosexual relations are not

meant to be gratifying, they are simply sexually confused

situations). The maintenance of the social order, as

currently constituted, depends on the continuing

inferiorization of lesbians and gay men. The general practice

of inferiorization and the more specific pattern of stigma

shape the experience and the life chances of homosexuals. The

social-role of homosexuality from the institutional

perspective has been equally confused with the concept of

"being" and "doing." Institutions seek to influence the role

of homosexuals, yet subordinate the "being" (identity) of

homosexuality. In other words, institutions create obstacles

for homosexual identity formation, and the manifestation of

integrated homosexual sexual scripts.

Why then does the social-role of homosexuality pervade in

societies constructed by heterosexist institutions? The

decision to participate in homosexual roles is based on the

perceived rewards inherent in the organizational structure.

Sexual behaviors are one of the most valued experiences for

human beings as a species and for most people as individuals.

Its valuation varies from those who are indifferent, to those

that are relatively addicted to sexual gratification (Burgess

and Wallin, 1953; Cuber and Haroff, 1965). Sexual

gratification, feelings of intimacy, and innate desires

 23

(intrapsychic scripts) are individual responses to the

socially constructed rewards and costs attached to variant

sexualities. According to exchange theory, conceptualization

of homosexual identities are expressed at a cost/benefit

ratio. Those behaviors or identities which have the greatest

rewards or benefits will increase the salience or expression

of a particular behavior or identity. Those identities or

behaviors which entail the greatest costs will decrease the

salience of identity constructs or the probability of

retaining certain behaviors or beliefs.

Benefits and costs are unevenly skewed toward

heterosexual relations in the United States. Religion,

political ideologies, family structures tend to reinforce

those behaviors which are heterosexual in nature, and sanction

behaviors linked to homosexuality. Thus, development of

homosexual identities procure perceiving homosexual relations

as a benefit as opposed to a cost. Such a process may be

indefinite for some individuals. However, others are capable

of solidifying a homosexual identity at relatively young ages.

From an exchange perspective, establishing a homosexual

identity involves structural arrangements which enable the

homosexual to perceive a greater benefit than cost ratio in

expressing and developing homosexual identities.

Davenport (1965) has described the sexual scripts of the

Melanesians in the Southwest Pacific. Premarital intercourse

is strictly forbidden in this culture, both males and females

 24

are encouraged to masturbate. In addition, all unmarried

males engage in homosexual relations with the full knowledge

and acceptance of the community. This cross-cultural case

study validates how perceived rewards and costs provide

opportunities for salient homosexual identity development.

Given the rewards in the United States are not those of the

Melanesians, it is presumed the perceived costs outweigh the

perceived rewards associated with a presented homosexual

identity.

The realization of costs associated with homosexual

identities forces gays and lesbians to develop resources which

enhance the benefits of their unique identity. As Randall

Collins suggests "the chains of interactions develop a

structure, whose profile depends on the respective levels and

types of resources possessed by participants" (Turner, 1991:

248). Such resources may include friendship networks

supportive of homosexuality, or participation in homosexual-

oriented organizations. Given the validity and relativity of

perceptions, gay love relationships may also procure salience

in homosexual identity development.

From an exchange perspective, several voluntarily and

involuntarily imposed factors explain the stability of

homosexual sexual scripts. Plummer (1979) lists ease,

pleasure, and secondary gain as factors that individuals

impose voluntarily upon themselves. Involuntarily imposed

 25

factors include problems of access, lack of "in-group"

support, and public labeling as homosexual.

The ease of remaining committed to familiar patterns of

behavior and the difficulties posed by adopting new lines of

action encourage people to retain the various sexual scripts

associated with homosexuality. Randall Collins describes the

familiarity with the social role as "rituals provide payoffs,

per se, especially those critical resources revolving around

group membership" (Turner, 1991; 238). Once individuals

become stabilized in homosexual roles, they may come to view

personal costs involved in taking on bisexual or heterosexual

roles (e.g., time, energy, anxiety, and diminished sexual

arousal) as outweighing the benefits of occupying a more

conventional social status. Comfort and familiarity with

homosexual identities and roles may foster the idea that it is

"easier, more attractive, less costly to remain homosexual"

(Plummer, 1975: 50). Individuals may also chose to retain

their homosexual identities and roles because they are more

pleasurable. Lesbian and gay males have learned that the

"acts of falling, making, and being 'in love' with a member of

the same sex can be both pleasurable and satisfying" (Plummer,

1975: 151). Homosexual experience comes to be sought as an

end in itself.

Secondary gains may also lead individuals to retain

homosexual roles and identities. Clear advantages accrue to

homosexual lifestyles. Lesbian and gay male sexuality is

 26

nonprocreative. Worries about pregnancy are nonexistent.

Homosexuals are also more tolerant of others labeled

"deviant," perhaps as a consequence of their own experience

with stigma (Corbett, Troiden, and Dodder, 1974). Greater

social mobility and a higher standard of living (at least for

gay males) have also been cited as advantages (Plummer, 1975).

Similar advantages, however, are open increasingly to

heterosexual cohabitators and child-free, married couples.

Factors over which an individual has little or no control

may also maintain the stability of homosexual roles and

identities over time. One involuntarily imposed factor is the

problem of access. As lesbians and gay males become older and

more entrenched in homosexual experience, they become

increasingly alienated from heterosexual experience, finding

it: "difficult to make or maintain heterosexual contacts and

increasingly disturbing to contemplate the idea of

heterosexual activity. Earlier problems of access and

identity may re-emerge in reverse if [they] should contemplate

departure from the homosexual role: the secure world is nor

the 'deviant' world, and the problematic world becomes the

'straight' world" (Plummer, 1975: 152).

The lack of in-group social support is a second

involuntarily imposed factor. Plummer (1975) contends that

homosexuals who attempt to reenter the heterosexual world

receive little or no support from homosexual friends and

acquaintances. More often than not, those who try to "go

 27

straight"-that is, attempt to incorporate heterosexual sexual

scripts-are ridiculed, not taken seriously, or rejected. A

person's homosexual friends are likely to see his or her

desires as unrealistic, self-deceptive, faddish, or a form of

fence sitting. The individual is often accused of denying his

or her "true" sexual or romantic nature.

Public labeling also fosters an involuntary commitment to

the homosexual script. Public labeling and denouncement as

homosexual by official social-control agents, such as the

police or courts, although relatively infrequent, may lead to

role imprisonment; "To come before a court in a blaze of

public scandal is to be publicly ushered into a deviant role,

with very few chances of receiving official declarations of

exit" (Plummer, 1975; 152).

In summary, social, cultural, and historical forces,

rather than inherent traits or essences, shape or construct

the characteristics thought to correspond with homosexual

sexual scripts. The term homosexual describes a particular

script, an expected pattern of behavior that flows along lines

shaped by age, gender conceptions, economic arrangements,

kinships, and other concepts. It is a shorthand summary of

how a given culture at a given historical point expects

homosexually experienced people to be cognitively,

emotionally, and behavioral.

The balance of rewards and costs as extracted by

heterosexist institutions and fostered by the presences of

 28

available resources distinguishes homosexual sexual scripts as

being quite significant. The homosexual sexual script is a

dynamic, cultural concept linked to the social structure.

Claude Levi-Strauss asserted "all exchange relations involve

costs for individuals, but, in contrast with economic or

psychological explanations of exchange, such costs are

attributed to society-to those customs, rules, laws, and

values" (Turner, 1991: 293). The social structure, as

depicted in the anthropological evidence, determines the cost

and benefits of acquiring the specified script. The emergence

of "being" from the "doing" infers commitment to the

designated script, despite the legacy of heterosexism embodied

in social institutions.

Organization of Dissertation

Given the predominance of heterosexual scripts, and their

relative strength in contemporary society, this dissertation

will apply the sexual script perspective to three selected

dimensions of homosexuality. This dissertation follows a

three article format. The organization of the dissertation

consists of a general literature review of the issue of

homosexual sexual scripts, three articles which apply sexual

scripts to homosexual relational dynamics, and a general

summary chapter relating each article back to the sexual

script perspective. Further, the dissertation also contains

an appendix section which presents the factor analysis results

 29

of certain scales used within some of the articles. The

research conducted in articles two and three have been

approved by the Human Subjects Committee at Iowa State

University. The results of the first article is based on

secondary data anylsis.

Each article, in this dissertation, will primarily focus

on one of the three analytical levels of sexual scripts,

merged with other sociological theories (i.e., exchange,

conflict, schema, etc.) to explain the dynamics of homosexual

relationships. In the first article, sexual scripts will be

employed to inspect the disclosure process among gay and

lesbian adolescents to their parents. This article will

center on the role identification with the family affects the

extent of perceived homosexual-supportive resources,

expression of the adolescents homosexual identity, and the

disclosure of their identity. Through social exchange

analysis, the article will balance the rewards and costs

associated with homosexual sexual scripts, compared to the

heterosexual scripts implicitly presumed in family

arrangements.

The second article will contrast homosexual scripts with

heterosexual scripts, by investigating the definitions of

attractiveness and sensuality manifested in the unique

cultural scenario scripts for each group. By extrapolating on

schema theory, the study will delineate the difference between

normative and individual definitional constructs. In this

 30

particular case, normative components in the definition of

attractiveness or sensuality will correspond to cultural

scenario scripts; consequently, individual-oriented

definitions will validate schematic suppositions.

The last article will examine physically and sexually

aggressive lesbian and gay relationships. Given the

institutionalized conflict between heterosexual and homosexual

scripts, heterosexual harassment of homosexuals may

circuitously lead to physical or sexual aggression in gay and

lesbians relationships. When homosexual scripts obviously

confront heterosexual scripts (i.e. being "out" of the

closet), heterosexuals will be more likely to enforce the

heterosexual script through, perhaps, violent means. Such

aggressive encounters, endured by homosexuals, may foster

learned helplessness or truculent frustration within

homosexual relationships, leading gays and lesbians

susceptible to potentially violent relationships.

 31

PAPER 1. ADOLESCENT DISCLOSURE OF HOMOSEXUALITY

TO THEIR FAMILIES
 32

ABSTRACT

The focus of this paper is the disclosure of homosexual

identities among adolescents, when considering their

relationships with their families. Over 170 gay adolescents

responded to a survey, addressing the expression of their

homosexual identity, the extent the adolescents identified

with their immediate family, the extent of resources they

perceived which were supportive of homosexuality, and the

degree of homosexuality disclosure to their family. The

results demonstrated identification with the family detracted

from both the extent of perceived, homosexual-supportive

resources available to gay adolescents, and the salience of

the adolescents' homosexual identity. Further, perceived,

homosexual-supportive resources and the salience of

homosexual identity increased the probability the gay

adolescents would disclose their identity to their family.

Family identification was not directly associated with

disclosure of homosexual identity to the adolescents'

families. The overall model explained more of the common

variance of disclosing to family members for males, than for

lesbians. In short, families of gay adolescents posed

barriers to homosexual identity acquisition. An exchange

analysis would suggest families reflect costs to disclosing

homosexuality; whereas, homosexual-supportive resources and

identity salience provide rewards for disclosure of a

homosexual identity.
 33

INTRODUCTION

Contemporary research trends on homosexuality have

shifted from causal inquiries surrounding developmental and

life style issues. In particular, research on homosexual

identity formation has been a growing field of theoretical

analysis in recent years. Such theoretical formulations

implied sexual identities, perceptions of self as

heterosexual, homosexual, or bisexual in relation to sexual,

romantic, and broader cultural arrangements are constructed

through various forces.

The homosexual identity is one of several identities

incorporated into an individual's concept of self. According

to Cass (1984) and depending on the context, the homosexual

identity may function as a self-identity, a perceived

identity, a presented identity, or all three. A self-identity

is characterized as self realization of same-sex desires and

behaviors; however, this may not be known or revealed to

others. When individuals suspect another individual to be

gay, but this notion remains to be confirmed constitutes a

perceived identity." Finally, .a presented identity occurs when

a homosexual announces their life style or identity to others.

These general concepts readily translate into an indicator of

identity disclosure. The primary focus of this research is on

the concept of presented identity.

During the past decade, several researchers have proposed

theoretical models, attempting to explain the formation and

 34

continuance of homosexual identities (Cass, 1979, 1984; Lee,

1977; Ponse, 1978; Schafer, 1976; Troiden, 1979; Weinberg,

1979). Although the various models suggested different stages

in the formation of homosexual identities, disclosure of the

homosexual identity appears to be a significant hallmark in

settling homosexual identity development.

The objective of this study was to analyze a specific

model for the disclosure of homosexual identities among gay

and lesbian adolescents to their immediate families. This

research hypothesized that disclosure of identity to the

family was predicated by the salience of the adolescents'

homosexual identity and the extent of perceived resources

serving as social support networks for the adolescents. It

was also posited the interpersonal dynamics of the family,

such as the degree the adolescent identifies with the family,

will influence both the salience of the adolescents'

homosexual identity and the breadth of perceived resources.

The second objective of this study was the consideration

and analysis of sex differences associated with each postulate

of the proposed model. The model contends structural forces,

impeding the expression of homosexual identities, will vary

for each sex, given different gender expectations for male and

female adolescents.
 35

Gay and lesbian youths

Very little is empirically known about the life style of

gay and lesbian youths. Theoretical analysis of homosexual

identities suggested the period of adolescence is critical for

the development of a positive gay self-image (Troiden, 1989).

Further, Dank (1972) asserted "coming out" and disclosure may

be occurring at younger ages, and will continue to do so, in

the United States. This is due, in part, to an increased

emphasis and acceptance of sexuality in general, tolerance of

homosexual behavior, and availability of information

concerning homosexuality (Dank, 1972).

Evidence that homosexual identities and homosexual

behaviors (including disclosure) are occurring at younger ages

is reflected in the ramifications of adolescent homosexuality.

The recent impact of AIDS on the gay community may force

disclosure of homosexuality among sexually active gay

adolescents. There is disturbing evidence that the dramatic

rise of teen-age, male prostitution in the United States is

partially due to the intolerance of homosexuality by their

parents (Bales, 1985). Further, the negative consequences of

"coming out" during adolescence may also be mirrored in the

distinct suicide rates plaguing gay and lesbian youth

(Hollinger, 1978; Kournay, 1987).

Other significant problems associated with gay and

lesbian youth relate to identity formation and development.

According to Erikson (1963), identity formation is the most

 36

salient developmental task during adolescence. Pressure to

resolve a heterosexual identity from external social forces,

and the internal need to foster homo-erotic drives can be

overwhelming. Maylon (1982) noted such identity confusion can

induce low self-esteem, depression, denial, suppression, and

compartmentalization in homosexual adolescents. To increase

the intensity of the trauma, Maylon (1981) asserted gay and

lesbian adolescents usually do not have full access to gay and

lesbian resources or communities, given the adult-oriented

nature of the communities and resources, which may foster

alienation among adolescents (see also Martin, 1982).

Disclosure and family dynamics

Disclosure of sexual orientation to the family of origin

is, needless to say, a traumatic experience for both the

adolescent and nonhomosexual family members. To date, the

analysis of family reactions to disclosure of homosexual

identities has been dominated by parental responses. Weinberg

(1972) and Jones (1978) concluded parental reactions to

disclosure tended to be quite negative, and consisted of

primarily two characteristics; (1) negative conceptions of

the homosexual identity, and (2) distinct feelings of failure

and guilt associated with their parental role. The child is

often severed from familial ties due to misconceptions about

homosexuality, thus parents may react to the disclosure by

treating the child as a stranger or as a stigmatized.

 37

stereotyped entity (Collins and Zimmerman, 1983; Devine, 1984;

Fairchild and Hayward, 1979; Strommen, 1989). Collins and

Zimmerman (1983) and Devine (1984) asserted structural,

regulative, cohesive, and thematic issues particular to the

family guide the parental reaction to the child's disclosure.

One central theme apparent in numerous writings on the subject

is that a positive relationship with parents prior to

disclosure is a good indicator of a healthy resolution

(Borhek, 1983; Fairchild and Hayward, 1979; Silverstein, 1977;

Weinberg, 1972).

Familial factors affecting identity formation or

disclosure of identity have not been confirmed. For example,

birth order has been sufficiently researched, yet has

maintained contradictory findings, regarding the etiology of

homosexual identity formation. Grundlach (1977) concluded

birth order and family atmosphere influenced the disclosure

process among lesbians. However, Perkins (1978) ascertained

from her study of 212 lesbians birth order, and, specifically

being an only child, were not significantly related to a

lesbian identity. Siegelman (1973) asserted birth order and

family size were not significantly distinct between male and

female homosexuals or heterosexuals. In general, the most

accurate empirical generalization regarding the link between

homosexual disclosure of their identity to their family of

origin was the unpredictability associated with familial

response (Borhek, 1983).

 38

The role of siblings in the disclosure process has been

recently introduced as a role relationship potentially

affecting homosexual identity formation. However, there has

been little discussion on the particular effects of disclosure

to siblings. Jones (1978) reported siblings tend to respond

to disclosure in a similar fashion as parents. Siblings view

the disclosing sibling as a stranger, or as a stigmatized

entity; however, siblings do not suffer the feelings of guilt

and blame as do parents.

The most significant research, pertaining to the

disclosure process of adolescents to family members, has

integrated not only structural concepts (such as family

structure), but also psychological and interactional

correlates. Savin-Williams' study of 317 gay and lesbian

college students and adolescents demonstrated the significance

of parents and self-evaluation for the "coming-out" process

(1989). Satisfying personal relationships and younger parents

were found to be good predictors for lesbians to disclose

their identity to their parents; however, this was not the

case for males. Further, self-esteem of the youth was

considered as a mild predictor of disclosure. Although the

study was most noteworthy, given the skewed age distribution

of the sample, the findings may not be generalizable to those

adolescents under the age of eighteen, and still residing with

their parents.
 39

It is evident further research is needed to examine the

role of internal and structural components influencing the

disclosure process of gay and lesbian youth to their family

origin. It is also necessary to place the role of the family

in context with other structural components which may affect

the salience of an adolescent's homosexual identity.

 40

THEORETICAL ANALYSIS

The proposed model in this study focuses on the

correlates which determine the salience of homosexual identity

and identity disclosure, despite structural resistance and

adversity to such identity progression. The model was based

on several theoretical premises, all of which posit homosexual

identity is a negotiated identity, internally and externally

between the self and structural arrangements. It is believed

identity disclosure is partially dependent on environmental

influences and the extent of perceived resources which either

push or pull the individual toward/away from expressing a

salient identity, and disclosure of the identity to family

systems. The model asserts, in part, that disclosure is a

cognitive process; however, the articulation and expression of

the identity (or salience and disclosure of the identity) is

negotiated, involving perceived costs and rewards linked to

structural arrangements. The model is theoretically founded

by both social exchange and sexual script perspectives.

According to social exchange theory, conceptualization of

identities can be expressed by the psychic profit associated

with a cost-benefit ratio. Those behaviors or identities

which incur the greatest rewards or benefits will increase the

salience of a particular behavior or identity complex. Those

identities or behaviors which entail the greatest costs will

decrease the salience of identity constructs or the

probability of retaining certain behaviors.

 41

Benefits and costs are decidedly skewed toward

heterosexual relations in the United States. Religion,

political ideologies, family structures tend to reinforce

heterosexual behaviors and identities, and sanction

behaviors/identities linked to homosexuality. Thus,

development and disclosure of homosexual identities procure

perceiving homosexual relations as a benefit as opposed to a

cost. Such a process may be indefinite for some individuals.

However, others are capable of solidifying a homosexual

identity at relatively young ages.

The means of distributing the cost and benefits attached

to sexual orientations are mediated by sexual scripts. Sexual

scripts represent sexual articulations by culture,

interpersonal relationships, and cognitive desires. Scripts

are mental trajectories, a cognitive view of official beliefs

(cultural scenario scripts), relational meanings and

expectations (interpersonal scripts), and sexual fantasies

(intrapsychic scripts). Sexual scripts designate cognitive

and affective limits, indicating appropriate and inappropriate

sexual behaviors (Gagnon and Simon, 1973).

In general, cultural scenario scripts encourage

heterosexuality. The media, religious forces, and society,

overall, dictate a heterosexist norm prevailing through the

embodiment of scripts geared to opposite-sex relations.

Because cultural scenario scripts value heterosexual relations

over homosexual relations, internalizing these official views

 42

negatively impacts on homosexual identity formation.

Interpersonal scripts associated with family arrangements also

discourage homosexuality, as documented in the literature.

Given the prevalence of a heterosexual norm among cultural

scenario and interpersonal scripts, homosexual adolescents

must undergo a resocialization process. This resocialization

process involves the formation and expression of a salient

homosexual identity despite relational and structural

constraints.

Figure l illustrates the postulated model of this

research project. The model predicts that a basic structural

and relational arrangement, the family, will influence the

salience of homosexual identities among adolescents, and the

disclosure of their identities to their family. The family

serves as a primary socializing agent which potentially

reinforces the heterosexual norm, and sanctions homosexual

development. The model posits strong identification with the

family of origin will impede the extent of homosexual-

supportive resources perceived by lesbian and gay adolescents

(Arrow A). Given the heterosexual orientation of families, in

general, the structure of these families will limit the

perception of gay adolescents toward viable resources,

supportive of developing a homosexual identity. Further, the

parents of gay and lesbian adolescents in all probability will

not avail resources to their children, assuming their children

to be heterosexual. It is also hypothesized that family

 43

identification will negatively impact on the expression of a

homosexual identity (Arrow B). Although past research has

found a positive correlation between family relationships and

acceptance of gay youth, family identification may not be a

significant predictor of disclosure to the family. Though a

positive relationship between the adolescent and their parents

may be conducive to disclosure, given the dissenting sanctions

for homosexuality, adolescents may be reluctant to disclose

their identity to their families for fear of being sanctioned.

It is hypothesized perceived supportive resources will foster

the expression of a homosexual identity among gay adolescents

(Arrow C). Such resources as gay organizations, supportive

friendships and relationships, counselors, etc. should, in

theory, facilitate the process of identity acquisition versus

identity confusion. In short, the resources should pull the

adolescent toward a positive gay self-image, and push the

adolescent away from the fear of parental sanctioning.

Figure 1 about here

There is little past research to speculate on the

relationship between family identification and the decision

regarding disclosure of gay identities. Therefore, it is

hypothesized an exchange occurs between the "pulls" and

"pushes" of the social structure on disclosure of sexual

orientation. The model asserts gay adolescents weigh the

 44

perceived costs associated with disclosing a potentially

stigmatized identity to their families (Arrow D), with the

benefits related to perceived supportive resources empowering

the salience and disclosure of gay identities (Arrow E).

Finally, the model maintains homosexual identity expression

will positively influence the extent of disclosure among gay

adolescents to their parents (Arrow F). Because the final

stage of identity formation, in the literature, suggests

commitment to the gay identity, it is hypothesized disclosure

is a sufficient component of commitment, and will entail

disclosure to the adolescents' primary groups (i.e. the

family).
 45

METHODS

Sample

The data for this study were obtained from a

questionnaire administered to self-identified gay adolescents,

ranging in age from 14 to 18. The sample originated from a

clinical-based, support group designed to provide resources

for gay and lesbian youth dealing with personal issues

pertinent to their particular life style. The support group

initially involved 35 adolescents. Through snowball sampling,

the adolescents involved in the support group were asked to

administer the survey to their personal network of friends and

acquaintances. A pre-addressed, postage-free envelope was

included with the questionnaire. No identifying marks were

placed on either the questionnaire or the envelope to ensure

confidentiality. A cover letter was also attached to the

questionnaire indicating the purpose of the study and means

undertaken by the research staff to assure anonymity for those

who might feel uncomfortable with their responses or with

others knowing of their participation in the project. The

final sample consisted of 172 self-identified gays and

lesbians, with 85 females and 87 males. The majority of

respondents, 95%, were from a suburban, upper East coast

community. The families of the adolescents were primarily

middle-class, with fairly stable incomes and professional

 46

occupations (61%). Further, the family structures of the

families were substantially dual earner, intact families

(78%).

Measurement

In order to operationalize a questionnaire in terms

familiar with gay adolescent needs and comprehension,

consultation with the gay and lesbian support group assisted

in the development of wording for several items on the

questionnaire. It should be noted many of the responses on

the questionnaire rely on the perceptions of the adolescents.

Further, many of the indices consisted of a single indicator;

consequently, single indicator concepts are generally

considered problematic regarding validity and reliability.

Several researchers support perceptual measures as valid

indices of measurement (Acock and Bengston, 1980; Gecas and

Schawlbe, 1986; Kerckoff and Huff, 1974).

Identification with Family. The extent the gay youth

identified with their family was measured by the following

question: "Before you realized (or suspected) you were gay,

how well did you get along with your family?" Respondents

chose one of five responses, with 1="I have never related to

my family", 2="I did not get along with my family to well",

3="I could somewhat relate to my family", 4="I got along with

my family fairly well", and 5="l related to my family very

well."
 47

Perceived Resources (Homosexual Social Support Networks).

The extent of perceived resources, functioning as social

support networks, was assessed by asking the respondents to

approximately indicate (to the best of their immediate

knowledge) how many friends, support groups, family members,

organizations, and acquaintances they believed to be

supportive of homosexuals. The adolescents selected one of

the following five response categories: l="0-5", 2="6=10",

3="11-15", 4="16-20", and 5="21 or more."

Disclosure of Identity to Family. Disclosure of gay

identities was ascertained by the degree of knowledge the

adolescents' parents were cognizant of their child's sexual

orientation. The adolescents chose one of five responses to

complete the following statement: "My parents ..." The

response categories ranged from l="have no idea that I may be

gay.", 2="raay suspect me to be gay.", 3="have asked me if I

was gay.", 4="have been told that I am gay, but do not believe

it.", and 5="are fully aware that I am gay."

Expression of Homosexual Identity. The degree the

adolescents' homosexual identity had been enacted upon was

assessed by five questions. The first two questions

ascertained participation and importance of homosexual-

oriented organizations. Respondents were asked to endorse one

of five response categories to the question: "During the past

month, how many gay organizations, groups, or meetings have

you attended?" The responses ranged from 1="0", 2="1", 3="2",

 48

4="3" and 5="4 or more." It should be noted frequency of

attendance and defined importance of gay-oriented

organizations are not synonymous. Given extenuating

circumstances within such a stigmatized population,

participation in gay organizations may not always be

commensurate with desired level of attendance. Therefore,

importance of such organizations, as perceived by the

adolescents, may be indicative of participation if homosexual

founded organizations were readily available in the gay

community. Importance of gay-oriented organizations were

assessed by a 5-point Likert scale in response to the

question: "How important are gay organizations, groups, and

meetings to you?" The respondents were asked to select one of

the following categories: l="not important", 2="mildly

important", 3="fairly important", 4="very important", and

5="extremely important."

The third question on the expression of homosexual

identity scale ascertained the importance of homosexual

friends to the gay and lesbian adolescents. Respondents were

asked to select one of five possible responses to the

following question: "How important are gay friends to you?"

The response categories included: l="not important",

2="mildly important", 3="fairly important", 4="very

important", and 5="extremely important."

The last two questions on the expression of homosexual

identity scale assessed gay sexual relations among the

 49

adolescents. Frequency and Importance of gay sexual activity

served as the indicators for sexual identity.

The adolescents' responses on the five items were summed

to comprise the scale, objectively measuring the expression of

their homosexual identity formation and development. The

scale produced a possible range of scores from 5 to 25.

Higher scores on the scale were indicative of strong gay

identity expression and formation, lower scores on the scale

suggested suspicion, nonexpression of the identity, or

identity confusion. The Cronbach's alpha coefficient for the

expression of homosexual identity scale was .77.

Procedures
Statistical analysis of the data consisted of several

techniques to distinguish significant associations between the

variables used in this analysis. First, a Pearson correlation

matrix of zero-ordered correlations was computed for both male

and female adolescents. Second, path analysis was implemented

to test the hypothesized associations of the proposed model

regarding the influence of the family on gay and lesbian

adolescents. The utilization of path analysis was considered

as a statistical technique to evaluate the strength of the

bonds between key variables, while supplying information

toward causal constructs, given the theoretical disposition of

the model. The variables constituting the formation of the

path model included: identification with family, perceived

 50

homosexual-supportive resources, salience of homosexual

identity, and disclosure of identity to the adolescents'

family. To reiterate, the model posits disclosure of identity

to the adolescent's family is contingent upon the extent the

adolescent identifies with their family, as mediated by the

salience of their homosexual identity and the extent of

perceived homosexual-supportive resources available to them.

Third, LISREL VII (Joreskog and Sorbom, 1989) was used to

determine the significance of direct and indirect effects for

each of the preceding variables on disclosure.

 51

RESULTS

The zero-ordered correlation coefficients for all the

variables in the study (refer to Table 1), provided strong

evidence that structural factors do impede the extent

homosexual-supportive resources are perceived by adolescents,

expression of homosexual identities among gay adolescents, and

whether or not adolescents disclosed their identities to

immediate family members. For lesbians, identification with

family was significantly correlated with the dependent

variables in the study. For gay males, identification with

family was significantly associated with perceived homosexual-

supportive resources, salience of homosexual identity, and

disclosure of identity to their family.

Table 1 about here

The results of the path model (see Figure 2) indicated

identification with family was significantly negatively

correlated with the extent of perceived resources or social

support networks available for both gay and lesbian

adolescents (males; beta=-.472, p<.001; females: beta=-.385,

p<.001). The degree of identification the adolescent had with

their family negatively affected the amount of resources

perceived by the adolescent, assisting in the development of a

positive gay self-image. A strong identification with family

impeded or detracted from perceiving the availability of

 52

resources, congruent with homosexual-oriented support

networks. Identification with their family explained more of

the total common variance in perceived resources for gay

adolescents than for lesbian adolescents (males: R^=.223;

females: R^=.148).

Figure 2 about here

Identification with the family and the extent of

perceived resources available for both gay and lesbian

adolescents appeared to be strong predictors of the expression

of homosexual identities. The stronger the identification

with the adolescents' families, the less the homosexual

identity was expressed for both males and females (males:

beta=-.631, p<.001; females; beta=-.570, p<.001). In other

words, identification with the family and expression of

homosexual identities for gay and lesbian adolescents were

incongruent. The family inhibits the gay youth from

expressing homosexual relationships and associating in the gay

community. The extent of perceived homosexual-supportive

resources positively contributed to expression of homosexual

identities among gay and lesbian youth (males; beta=.431,

p<.00l; females: beta=.392, p<.001). The more resources,

perceived by both gay and lesbian adolescents which were

considered as supportive to their homosexual identity, the

more their identity as a homosexual was expressed.

 53

Identification with the family and perceived homosexual-

supportive resources explained approximately one-half of the

total common variance associated with a salient homosexual

identity for both gay and lesbian adolescents (males:

R^=.482; females; R^=.421).

Identification with family was not significantly

associated with the extent gay adolescents disclosed their

identities to their parents; however, both expression of their

identities and perceived resources did positively affect

disclosure. Expression of homosexual identity was the

strongest predictor of disclosure to parents for both gay and

lesbian adolescents (males: beta=.479, p<.001; females:

beta=.253, p<.05). The extent of perceived, supportive

resources was also positively related to identity disclosure

for both males and females (males: beta=.311, p<.01; females:

beta=.261, p<.05). Given, the significance of family

identification on perceived resources and salient identity

expression, family identification mediated the nature of

disclosure to parents. Therefore, the model suggests

perceived social support and expression of identities may be

more responsible for disclosure than the endurance of family

identification among gay adolescents. The model also implied

perceived resources may be more predictive to disclosure than

expression of gay identities among lesbians. And, salience of

identity appeared to maintain a stronger effect on disclosure

than perceived resources for the males in this study. The

 54

proposed model explained twice the total common variance among

gay males than for lesbians (males: R^=.540; females:

1^^.257).

Table 2 about here

A test of the significance for indirect effects provided

intriguing, if not compelling results. Although modest,

identification with family significantly, yet indirectly,

affected the expression of the adolescent's homosexual

identity (males: beta=-.210; females: beta=-.157).

Identification with family significantly and indirectly

decreased the probability that gay adolescents would disclose

his or her identity to their family (males: beta=-.449;

females: beta=-.236). The extent of perceived resources, by

the adolescent, affected the disclosure process indirectly

(males: beta=.213; females: beta=.098). This trend was

significant only for males. Therefore, family identification

did maintain an indirect effect on the disclosure process;

however, this process is mediated by the extent of perceived

resources and the expression of a homosexual identity among

adolescents.
 55

DISCUSSION

Identification with parents detracted from the extent of

perceived homosexual-supportive resources among gay

adolescents. Further, identification with the family and

perceived resources significantly predicted the expression, or

salience of gay identities. Finally, social resources and

salience of identity positively affected the disclosure of

identity among adolescents to their parents. Identification

with the family did not directly affect whether or not the gay

adolescent disclosed their identity to their family; however,

family identification is mediated by perceived resources and

the expression of homosexual identities.

The model explained more of the common variance for males

than for females. In general, the tested model maintained

twice the explanatory power (R^) for the gay males when

contrasted with the lesbian adolescents. Perhaps, the

expression of identity scale was more conducive toward

assessing male sexuality and identity than for females. The

slight gender difference could also be a result of community

differences. Theoretically, in the gay male community,

community networks and resources tend to be highly saturated

with sexually schematic themes. In the lesbian community, the

resources involve more intimate friendship-based networks.

Therefore, the measurement of perceived resources and identity

expression may be more valid and reliable indicators for gay

males than for lesbians. The findings of this study could

 56

also be the result of structural versus internal negotiations.

Perhaps, for males external, structural forces are more

influential than internal motivations; whereas, for lesbians,

the process may be more defined among internal forces (i.e.,

self-esteem, self-efficacy, etc.).

Disclosure of a gay identity among adolescents entails a

resocialization process, whereby adolescents must secure those

resources which facilitate homosexual scripts over the

crystallized heterosexual scripts indoctrinated by parents.

In short, disclosure is a means of securing enough social

supportive resources to counter the heterosexual norm

maintained through familial interactions. By exploring and

utilizing homosexual-oriented resources, adolescents procure

the means possible to establish a homosexual identity, and by

the same means, initiate a resocialization conducive to

homosexual scripts. After an indefinite period of time,

adolescents become familiar with the relatively new scripts

and, in turn, nourish the continuity of their identity as a

gay individual. As the formation of their new identity comes

to completion, no longer are the adolescents necessarily

dependent on the heterosexual scripts fostered by the family.

When the balance of support provided by the gay community

exceeds the possible rejection or sanctions will the

adolescent disclose their identity to their families.

Homosexual networks and scripts empower adolescents, and

 57

adults alike, with confidence and support to disclose their

identity to loved ones.

 58

CONCLUSION

The data supported a generalized exchange between

homosexual-supportive resources, salience of identity, and

family identification. Disclosure of sexual identity can be

conceptualized as an exchange system, where the costs of

disclosing a homosexual identity to parents was weighed

against resources supporting the disclosure process. The

impact of heterosexual scripts, via family identification,

reduced the amount of perceived resources available to gay and

lesbian adolescents. Further, identification with the

heterosexual scripts, abundant in the family, detracted from

the expression or salience of acquiring a homosexual identity.

However, family identification did not directly influence

whether gay and lesbian adolescents disclosed their identity

to the parents. Yet, the extent of supportive resources and

the expression of identity were strong predictors of the

disclosure process.

Sexual scripts mediate the exchange of heterosexual and

homosexual rewards and costs. Given the dominating influence

of heterosexual scripts, disclosure becomes a process of

acquiring resources supportive of homosexuality, and

expressing a salient identity based on the perception of such

resources. When the extent of perceived resources and the

internalization of the support, acquired through the

resources, provides more rewards than the associated costs of

risking family identification, then disclosure becomes viable

 59

for the adolescent. In short, adolescents are more likely to

disclose their identity to their parents, when the resources

are available to support the potentially negative consequences

associated with disclosure.

 60

BIBLIOGRAPHY

Acock, A., and Bengston, V. (1980). "The socialization and

attribution processes: Actual versus perceived
similarity among parents and youth." Journal of Marriage
and the Family^ 42, 501-515.

Bales, J. (1985). "Gay adolescents' pain compounded." APA

Monitor, 16(12), 21.

Borhek, M. (1983). Coming Out to Parents. New York, NY:

Pilgrim Press.

Cass, V. (1979). "Homosexual identity formation: A

theoretical model." Journal of Homosexuality, 4(3),
219-35.

Cass, V. (1984). "Homosexual identity formation: Testing a

theoretical model." Journal of Sex Research, 20(2), 143-
67.

Collins, L., and Zimmerman, N. (1983). "Homosexual and

bisexual issues." In J. Hansen, J. Woody, and R. Woody
(Eds.), Sexual Issues in Family Therapy, (pp. 82-100).
Rockville, MD: Aspen Publications.

Dank, B. (1972). "Why homosexuals marry heterosexual women."

Medical Aspects of Human Sexuality, 6, 14-23.

DeVine, J. (1984). "A systemic inspection of affectional

preference orientation and the family origin." Journal
of Social Work and Human Sexuality, 2, 9-17.

Erikson, E. (1963). childhood and Society. New York:

Norton.

Fairchild, B., and Hayward, N. (1979). Now That You Know:

What Every Parent Should Know About Homosexuality. New
York, NY: Harcourt, Brace, and Jovanovich.

Gagnon, J., and Simon, W. (1973). Sexual Conduct: The

Social Sources of Human Sexuality. Chicago, IL: Aldine.

Gecas, v., and Schwalbe, M. (1986). "Parental behavior and

adolescent self-esteem." Journal of Marriage and the
Family, 48, 37-46.

Grundlach, R. (1977). "Sibling size, sibsex, and

homosexuality among females." Transnational Mental
Health Newsletter, 19(1), 3-7.
 6l

Jones, C. (1978). Understanding Gay Relatives and Friends.

New York, NY; Seabury.

Joreskog, K. M. and Sorbom, D. (1989). LISREL 7: A Guide to

the Program and Applications. Chicago, IL; SPSS, Inc.

Kerkoff, A., and Huff, J. (1974). "Parental influence on

educational goals." Sociometry, 37, 307-327.

Kourany, R. (1987). "Suicide among gay adolescents."

Journal of Homosexuality, 13(4), 111-17.

Lee, J. (1977). "Going public: A study in the sociology of

homosexual liberation." Journal of Homosexuality, 3(1),
49-78.

Martin, D. (1982). "Learning to hide: The socialization of

the gay adolescent." In Feinstein, Looney, Schwartzberg,
& Sorosky (Eds.), Adolescent Psychiatry, Vol. 10, (pp.
52-65). Chicago, Illinopis: University of Chicago
Press.

Maylon, A. (1981). "The homosexual adolescent; Development

isuues and social bias." Child Welfare, 60, 321-30.

Maylon, A. (1982). "Biphasic aspects of homosexual identity

formation." Psychotherapy: Theory, Research, and
Practice, 19, 335-40.

Perkins, M. (1978). "On birth order among lesbians."

Psychological Reports, 43, 814.

Ponse, B. (1978). Identities in the Lesbian World: The

Social Construction of Self. Westport Connecticut:
Greenwood Press.

Savin-Williams, R. (1989). "Coming out to parents and self-

esteem among gay and lesbian youths." Journal of
Homosexuality, 18(1/2), 1-35.

Scahfer, S. (1976). "Sexual and social problems among

lesbians." Journal of Homosexuality, 12( 1),-5.0-69.

Siegelman, M. (1973). "Birth order and family size of

homosexual men and women." Journal of Consulting and
Clinical Psychology, 41, 164.

Silverstein, C. (1977). A Family Matter: A Parent's Guide

to Homosexuality. New York, NY: McGraw-Hill.
 62

Strommen, E. (1989). "'You're a what?': Family member

reactions to the disclosure of homosexuality." Journal
of Homosexuality, 18(1/2), 37-58.

Troiden, R. (1979). "Becoming homosexual: A model of gay

identity acquisition." Psychiatry, 142(4), 362-73.

Troiden, R. (1989). Gay and Lesbian Idneity: A Sociological

Analysis. Dix Hills: General Hall, Inc.

Weinberg, G. (1972). Society and The Healthy Homosexual.

New York, NY: St. Martin's Press.

Weinberg, T. (1979). "On 'doing' and 'being' gay; Sexual

behavior and homosexual male self-identity." Journal of
Homosexuality, 4, 143-56.
 63

Expression of
Homosexual Identity

/
Identification
with Family
Disclosure
of Identity

Perceived Homosexual-
Supportive Resources

Figure 1. Proposed model illustrating the effect of family

identification on disclosure of identity, mediated
by the extent of perceived supportive resources and
identity expression
 64

Table 1. Correlation matrix for the variables in the study

among both gay males and lesbians

1. 2. 3. 4.

1. Identification —.472 -.631 -.471

with Family

2. Extent of Supportive -.385 .643 .630

Resources

3. Expression of -.570 .566 X .693

Identity

4. Disclosure of -.340 .431 .445

Identity to Parents

Mean

(males) 3.14 2.33 14.43 2.58

(lesbians) 3.09 2.60 12.92 2.28

Standard Deviation

(males) .11 .13 .47 .15

(lesbians) .13 .14 .44 .14

Note: coefficients for lesbians appear below the diagonal.

•V
Table 2. Decomposition of effects for the proposed model by
gay males and lesbians

TOTAL DIRECT INDIRECT

Hales Females Males Females Males Females

Perceived Resources
Family Identification -.472 -.385 -.472 -.385 .000 .000

Expression of Identity
Family Identification -.631 -.570 -.421 -.413 -.210* -.157*
Perceived Resources .444 .407 .444 .407 .000 .000

Pisploswe gf Identity
Family Identification -.471 -.340 -.022 -.104 -.449* -.236*
Perceived Resources .525 .352 .307 .254 .213* .098
Expression of Identity .479 .242 .479 .242 .000 .000

* indirect effect significant at the .05 level.

 66

/r= .551 (.466)

Expression of
Homosexual Identity

- .421»(- .413*)

Identification \ .479» (.242*)

with Family .444» (.407»)
- .022(- .104)

\ Disclosure
.472*(- 385*)
of Identity
.312» (.254^)^,.'
\
«2=.539 (.253)
Perceived Homosexual-
Supportive Resources

/?2= .223 (.148)

Note: Standardized regression coefficients for lesbian

adolescents appear in parentheses.

* p value significant at the .05 level.

Figure 2. The impact of family identification on disclosure

of identity to the family among gay and lesbian
adolescents
 67

PAPER 2. THE DEFINITION OF ATTRACTIVENESS AND SENSUALITY:

A SCHEMATIC COMPARISON BETWEEN HOMOSEXUALS AND HETEROSEXUALS
 68

ABSTRACT

The objective of this study was to develop the

relationship between sexual scripts and schema toward the

definition of attractiveness and sensuality. By contrasting

the definitions of attractiveness and sensuality among gay

males and heterosexual females, and heterosexual males and

lesbians; gender and sexual orientation differences could be

identified, regarding conceptual definitions of

attractiveness. Content analysis was employed to probe the

differences as they relate to schematic and

cultural/subcultural influences. For the study, two data sets

were utilized. The first data set included over 300 gays and

lesbians. The second data set consisted of 400 Iowa state

University college students. In general, heterosexual females

and homosexual males shared similar definitions of

attractiveness; however, the two groups did not share similar

definitions of sensuality. Among lesbians and heterosexual

males did not share any significant similarities in

definitions of attractiveness or sensuality.

 69

INTRODUCTION

Several metaphors have been constructed to designate

general connotations attached to the definition of beauty.

For example, the popular remark "beauty is in the eye of the

beholder," suggests the determinants of beauty are personal,

individual-oriented. "The face of an angel," and "what is

beautiful is good" reflect beneficent personality standards in

the definition of attractiveness. Further, a statement such

as "clothes make a man," incurs subliminal concepts of social

and gender stratification which may play an integral part in

defining beauty. Does culture, the values and norms embedded

within the United States, influence definitions of

attractiveness? Is sensuality linked to the definition of

attractiveness, or is sensuality a markedly different concept

than beauty? And, is the definition of attractiveness and

sensuality composed of different traits based on sexual

orientation differences?

The definition of attractiveness and sensuality employed

in this research is exploratory. The definitions of

attractiveness and sensuality are, by nature, the subject'of

this article. Therefore to define both terms would be

premature. However, the notion of attractiveness in this

research projects connotes general aesthetics appeal. And

sensuality was preconceived as those characteristics which

incur sexual desire.

 70

The purpose of this study was to analyze the various

components in the definition of attractiveness and sensuality

among heterosexual females when compared to the responses of

homosexual males, and heterosexual males when compared to

lesbians. The objectives to be accomplished by such research

were focus on (1) normative versus individual-oriented means

of definition construction for sensuality and attractiveness;

and (2) the sexual schematic comparisons among heterosexual

females and homosexual males, and heterosexual males and

lesbians regarding specific traits implicit in defining what

is attractive and what is sexy regarding both males and

females.
 71

REVIEW OF LITERATURE

Numerous researchers have conducted studies demonstrating

the importance of being attractive in the United States. The

commonality emerging among these studies concur that

attractive individuals receive preferential treatment. The

effects of facial appearance influences the context of dating

(see Krebs and Aldinolfi, 1975); marriage (Nida and Williams,

1977); politics (Efran and Patterson, 1974); employment

(Waters, 1985); and criminality (Âgnew, 1984). The role of

attractiveness is an important element in social life. The

effects of attractiveness are uncontested by this study;

however, what remains more of a mystery are our perceptions

about the definition of attractiveness.

Objective versus subjective definitions of attractiveness

Patzer (1985) has criticized many research efforts

concerning the effect of attractiveness for inadequate

definitions of attractiveness employed in their research

designs. Many researchers have assumed a common, perhaps

universal, definition of attractiveness (Touhey, 1979;

Sprecher, 1989; Gladue and Delaney, 1990). Such studies have

defined what is attractive by varying means, usually

subjective assessment strategies made by a designated research

team. Consequently, these studies inferred the definition of

attractiveness is similar for all individuals. Such

assumptions are qualified and legitimated by theorizing that

 72

normative rules guide the definition of attractiveness. In

other words, attractiveness is not in the eye of the beholder,

it remains in the eye of the culture.

Research on subjectivity and intersubjective agreement

(objectivity), regarding the definition of attractiveness, has

only recently become a concern among attractiveness

investigators. Downs (1990) concluded "interjudge agreement

concerning target subjects' levels of attractiveness is

generally high, which suggests that attractiveness is both

scaleable and culturally-defined" (p. 458). Further, Udry

(1965) found interjudge agreement on female faces tended to be

quite reliable. However, Patzer (1985) suggested the

definition of attractiveness is not a quantitative element.

Patzer contended the definition of attractiveness is a

subjective concept, which eludes codification. Lucker and

colleagues (1981) asserted attractiveness is composed of

objective (empirically measured) criteria and a subjective

element. Therefore, definitions of attractiveness vary

empirically as an objective element; and, according to some

theorists, definitions also rely on a subjective, unmeasurable

quality unique to individuals.

Objective criteria of attractiveness

If attractiveness can be theoretically and empirically

defined at a scaleable level, what constitutes the cultural or

objective definition of attractiveness? In the past, many

 73

studies have focused solely on the face in their attempt to

define attractiveness. A study conducted by Alicke, Smith,

and Klotz (1986) argued the body of an individual may be as

equally important as their face in defining attractiveness.

Another study by Franzoi and Herzog (1987) determined women

and men concur buttocks, eyes, legs, and health are

significant features in defining male attractiveness. Women

were also more inclined to consider body scent and physical

stamina as important determinants in judging male

attractiveness.

Other studies have yielded similar findings involving the

male body. Symons (1987) concluded females were more

receptive of moderate-sized men. This finding was supported

by Horvath (1981), asserting moderately broad shoulders were

attractive. Studies conducted in the 1970's revealed similar

premises about the size and muscularity of males, regarding

the definition of attractiveness. Lavrakas (1975) and

Graziano, Brothen, and Bersheid (1978) affirmed a medium-sized

stature were more appealing in defining male attractiveness

among women.

Facial qualities as a determinant for attractiveness has

received greater scientific evaluation. Therefore, the

research on facial characteristics tend to be more advanced

and sophisticated than research conducted on the male body.

Analyses of the male face have proposed maturity, dominance,

social status, and sociability are distinguishing dimensions

 74

which increase attractiveness in males (Keating, 1985;

Cunningham, Barbee, and Pike, 1990). In other words, male

facial characteristics corresponding to such personality

qualities tend to connote higher levels of attractiveness.

According to Keating (1985), male faces possessing a square

jaw and thin lips were more likely to be defined as dominating

and attractive. Cunningham et al. (1990) reported large eyes,

prominent cheekbones, a large chin, high-status clothing, and

a big smile were significantly associated with male

attractiveness. Both of these studies suggest not only

physical features are conducive in defining what is attractive

in males, but also assert personality characteristics and

social class may be equally as important in defining

attractiveness.

According to Furnham, Hester, and Weir (1990), males

prefer women possessing large breasts and an hourglass figure

(see also Gitter, Lomranz, Saxe, and Bar-Tal, 1983). Franzoi

and Herzog (1987) concluded body scent, waist, thighs, body

build, buttocks, breasts, eyes, face, and weight were

designated as important features for males defining female

attractiveness. Keating (1985) discovered a strong gender

stereotype in males' ratings of female facial characteristics.

The males in her study delineated immature, and nondominant

facial features as being attractive. Kulik and Harackiewicz

(1979) asserted males found feminine females as being more

attractive than androgynous females; however, males drifted

 75

toward androgynous females for companionship rather than

romance. Feinman and Gill (1978) suggested females with

lighter coloration (complexion, hair, eyes) tended to be rated

more attractive among male respondents.

Gender differences

Research studies on attractiveness between men and women

have demonstrated convincing evidence for substantial gender

differences. According to Jackson and associates (1986),

women were more concerned and less satisfied with their

appearance than men. Rand and Hall (1983) concluded men and

women were not congruent in judgements of their own

attractiveness. The researchers argued females were more

susceptible to the cultural norms which based female

attractiveness as more relevant compared to males. The most

thorough analysis of gender differences and attractiveness

stems from the meta-analytic review conducted by Feingold

(1990). Feingold reviewed five research paradigms which

yielded similar findings: men place greater emphasis on

attractiveness than females.

Studies assessing gender differences in what males and

females define or judge to be attractive have illuminated

further theoretical considerations regarding attractiveness.

Keating (1985) discovered mature features were linked to

judging men as attractive, while female faces were considered

more attractive when the face exhibited immature and

 76

nondominant features. Feinman and Gill (1978) investigated

attractiveness preferences among college students, resolved a

greater desire for lighter coloration by males, and a stronger

preference for darker coloration by females. Kulik and

Harackiewicz (1979) researched gender identities as a

dimension of attractiveness. They found that females

preferred androgynous male partners, while males favored

feminine women for romantic involvements. The summation of

the research suggested interesting gender trends with how

males and females define attractiveness. Of particular

interest was the influence of personality (gender traits) with

definitions of attractiveness, which had been somewhat

neglected in past research. Further, past studies have

neglected the role of sexual orientation on delineating the

definition of attractiveness along gender lines.

 77

A SCHEMATIC ANALYSIS

Although there has been substantial research on

attractiveness, the methodologies employed by some studies may

be considered as problematic for several theoretical reasons.

First, few of the studies which link attractiveness to

preferential treatment have neglected actually defining the

concept of attractiveness. Rather, the investigators have

assumed a commonality between their personal assessment of

attractiveness and the subjects. Such studies have fostered a

fairly strong, perhaps unreasonable, assumption in their

studies: everyone has the same definition of attractiveness.

Second, many of the studies conducted on attractiveness have

neglected a personality dimension to the definition of

attractiveness. Though, some of the studies have focused on

physical attributes and performance, they have ignored general

personality constructs (i.e. warmth and sensitivity,

confidence, etc.).

Third, many of the researchers have made another quite

robust assumption. They have assumed definitions of

attractiveness are automatically congruent with what an

individual denotes as romantic or sensual. This assumption

has not been proven valid within any study conducted on

attractiveness and romantic intention. Further, the

definition of sensuality may be conceptually and categorically

different from definitions of attractiveness.

 78

Finally, past analyses of gender differences, regarding

attractiveness, have focused their comparisons on

predominantly heterosexual male and female populations. On a

conceptual and theoretical level, it could be argued that

definitions between males and females of opposite sex

individuals are incompatible. In other words, when males

define characteristics important for female attractiveness,

and are then compared to what females define attractive in

males, is like comparing the definitions of apples to oranges.

Males and females may share similar, analogous qualities in

their definition of attractiveness; however, considering the

subjects of their definitions are different, many unique

qualities, specific to males or females, may be expressed.

The Donovan, Hill and Jankowiak (1989) study concluded sex and

sexual orientation of perceivers or observers were irrelevant

to prescaling of female stimuli. They did find sexual

orientation and sex were significant mediating variables when

judging male stimuli.

Schematics

Gender schémas attribute definitions of attractiveness on

which individuals structure, and process their cognitions of

the social world into understandable components. In general,

schémas are much like sexual scripts (Simon and Gagnon, 1984),

which are perhaps more relevant to the understanding of

attractiveness and schematic orientations. Where schémas

 79

refer to static normative knowledge, scripts are defined as

dynamic interchanges between the social environment and the

individual based on normative interaction patterns. Sexual

scripts are understood as general guidelines, procedural

diagrams for interpreting sexually relevant expressions and

behaviors. Sexual scripts are typologized by three distinct,

yet interconnected, levels. Cultural scenario scripts dictate

the normatively accepted or shared values attached to

sexuality. Interrelational scripts guide the contexts of

relational sexuality. Intrapsychic scripts define the

motivations, fantasies, and desires for individual sexuality.

If we assume schematics theoretically parallel sexual scripts,

then schémas can be broken into three levels of abstraction;

that is, there are cultural schémas, relational schémas, and

intrapsychic schémas.

From this sexual schematic perspective, definitions of

attractiveness and sensuality vary along schematic levels of

abstraction. Therefore, definitions of attractiveness reflect

cultural standards of attractiveness, relational aspects of

attractiveness, and romantic or sexually desired

characteristics of attractiveness. An individual's definition

of attractiveness constitutes multiple components reflecting

societal, relational, and personal influences. If a common

culture dictates a definition of attractiveness, this may

account for relatively high levels of interrater reliability

in past studies. Definitions of what is attractive for males

 80

or females should be shared among all individuals regardless

of sexual orientation.

Definitions of attractiveness at the relational and

intrapsychic schematic levels should not be readily shared or

even similar among individuals. Because each individual

possesses varied past experiences and meanings associated with

sexuality, definitions of attractiveness should be varied.

Furthermore, it is posited definitions of attractiveness at

the relational and intrapsychic level will be more aptly

conceptualized as sensual. In other words, what is sexy is

more likely to be determined by an individual's past sexual

interactions and their respective meanings which they attach

to sexuality. Though it is believed definitions of

attractiveness and sensuality will be dissimilar, a

substantial link between the two concepts should be apparent

because of the reflexivity between the three schematic levels.

Theoretically, the definition of sensuality among raters could

account for the subjective error variations found in many of

the studies conducted on attractiveness.

Gender schematic differences, theoretically, should, be a

relevant concern for researchers. Once again, if we assume

schémas parallel sexual scripts, the gender differences

apparent in sexual scripting should be manifested within

definitions of both attractiveness and sensuality. If

traditional gender roles are present within definitions of

attractiveness and sensuality, then males would be more

 81

instrumental in terming their definitions of attractiveness

and sensuality, and females would be more inclined toward

expressive characteristics. If attractiveness is culturally

or normatively defined, and traditional gender roles are an

obvious component in American society, then definitions of

attractiveness should reveal traditional gender differences.

The same should be equally true for definitions of sensuality.

Because traditional gender roles are frequently exhibited

within the context of relational interactions, definitions of

sensuality should also indicate gender biases.

Hypotheses

To reiterate the objectives of this study, the primary

concern of this project is to provide further clarifications

on the definition of attractiveness. The second objective is

to analyze the differences in definitional constructs between

sensuality and attractiveness. The third objective is to

examine gender differences between males and females regarding

the definitions of attractiveness and sensuality, by

controlling for sexual orientation. It should be noted that

definitions of sensuality would be incompatible if given

different targeted stimuli. In other words, males defining

sensual characteristics of females, and females defining

sensual traits in men are incongruent stimuli.

It is expected homosexual males and heterosexual females,

and heterosexual males and lesbians will define attractiveness

 82

using the same general characteristics. To reiterate, is it

assumed attractiveness is guided by general norms within the

cultural scenario script which define male and females

attractiveness the same, regardless of sexual orientation.

However, it is predicted sensuality will differ between the

heterosexual and homosexual comparison groups because of

individual motives in the intrapsychic scripts. The

hypotheses stated for this project are delineated below.

(1.) Definitions of attractiveness will significantly

differ when compared to definitions of sensuality

within both comparison groups.

(2.) Definitions of attractiveness will be significantly

similar between heterosexual females and homosexual

males; and heterosexual males and lesbians.

(3.) Definitions of sensuality will be significantly

different between heterosexual females and

homosexual males; and heterosexual males and

lesbians.
 83

METHODS

The data for this analysis was theoretically based on

four distinct categories: heterosexual males and females, and

homosexual males and females. To test the propositions

asserted in this study, two samples were collected from these

designated categories. Both surveys were administered during

the spring and summer of 1992.

Homosexual sample

Contact with the homosexual subjects, participating in

this project, was made in three ways: by stratified sampling

through (1) gay/lesbian organizations, (2) gay/lesbian pride

events, and via snowball sampling tactics by (3) acquaintances

in the gay/lesbian community. Numerous organizations were

asked to participate in the project, those who chose to take

part in the study were sent surveys to administer to their

members. Organizers of gay pride events were contacted for

permission to distribute questionnaires for interested

respondents. Several respondents took additional surveys to

give to their gay/lesbian/bisexual friends apart from the gay

pride and the organizational activities. These sampling

procedures were used to gain access to a highly stigmatized

population. Despite the difficulties in reaching a

stigmatized sample, 816 surveys were distributed and 306

surveys were returned. The questionnaire response rate was

calculated at 37.7%. The relatively low response rate

 84

resulted due to several factors, including: snowball

sampling, apprehension on the confidentiality of the survey,

and an inability to follow-up on all the surveys. Of the

completed surveys, 169 of the respondents were identified as

gay/bisexual males, and 130 were lesbian/bisexual females.

The average age of the subjects, participating in the

study, was 32. The majority of respondents (79%) identified

themselves as exclusively gay. Only fifteen percent of the

subjects in the project distinguished their orientation

ranging from mostly homosexual to primarily heterosexual with

slight homosexual tendencies. Approximately twenty-two

percent of the respondents classified themselves as students;

forty-four percent as managers, professionals, or executives;

and twenty-three percent as skilled or unskilled workers. The

other eleven percent of subjects were scattered across

miscellaneous occupations, or retired. The majority were

highly educated, possessing or in the process of obtaining a

college degree. Seventy percent of the participants claimed

their annual income was below $30,000, with a mean categorical

response of "between $15,000 and $30,000." The sample

consisted of respondents from over fourteen different states,

from California to New York; however, most of the respondents

were from Iowa. Eighty percent of the sample resided in

metropolitan areas or suburbs. Approximately half of the

sample was committed to a relationship of some sort, while the

 85

other half of the respondents were not dating currently, or

were dating casually.

Heterosexual sample

Slightly over two-hundred heterosexual females and one-

hundred and forty-nine males responded to a survey

administered in a large sociology service course at a major

midwestern university. The mean age of the respondents was

approximately 20-21. The majority of the participants were

from a town or small city (41%). Over seventy-one percent of

the respondents were classified as either sophomores or

juniors. The students were from a wide range of majors. The

majority of the respondents lived off-campus (43% for females,

and 46% among males), while thirty percent resided in

dormitories. Approximately twenty-six percent of females and

twenty-three percent of males participated in

sorority/fraternity housing. The families of the participants

in this study tended to possess medium-level incomes. Over

half of the sample reported their family's income was over

forty-thousand dollars, and only three percent responded their

family's income was below fifteen-thousand dollars.

Measurement

Two general, open-ended questions ascertained general and

specific attributes guiding definitions of attractiveness and

sensuality among both heterosexual and homosexual respondents.

 86

The first question asked both samples to identify three to

five characteristics which makes an individual attractive.

The second question asked the respondent to identify three to

five characteristics which makes an individual sexy. Open-

ended questions were believed to be superior for this study as

opposed to predefined scale items for several reasons.

Predefined scale items would have limited the respondents to

specify the importance of several designated items in which

past researchers had proposed were qualities of

attractiveness. This method would have restricted the

definition of attractiveness to attributes imposed by

researchers. Secondly, the categories offered by researchers

in the past were usually measured by the degree of importance

for each particular item. The problem with such a format

transposes nominal categories into interval level data, which

may lose the subjective essence of the category through the

digitization of responses. "Importance" scales also produce

asymmetrical distributions, as evidenced by the substantial

means presented in studies using such a design (see Franzoi

and Herzog, 1987).

The subjects' responses to the two questions were coded

in a parallel fashion. A random selection of fifty surveys

(twenty-five from each sample) were previewed to denote

possible categories in the subjects' responses to definitions

of attractiveness and sensuality. A general coding scheme was

established using the parameters initially defined by the

 87

previewed surveys. Each new response by the participants, not

undertaken by the initial coding scheme, was added to the

scheme. By the end of the content analysis, the coding scheme

included over forty categories used by the respondents to

define attractiveness and sensuality.

The coding categories represent generalities, as opposed

to specific attributes. For example, the category mouth

included such aspects as lips and teeth. Further, the

category mouth does not specify specific attitudes reflected

on the mouth (i.e. thick lips, small mouth, etc.). One

category, in particular, should be cautiously interpreted:

muscularity. The concept muscularity employed muscle tone and

weight as general dimensions. Weight was not considered as a

distinct dimension because of the extended variability

associated with weight. The same coding scheme was applied to

the personality categories identified by the respondents.

In order to capture very general definitions, not

specific to body, face or personality characteristics, four

"catch-all" categories were formed. General body, face and

personality categories were developed to secure general

preferences. The categories were only used for respondents

who simply indicated a one-word statement, such as "their

personality", or "their body." The last general category

established was overall looks or appearance. This category

included statements like "good-looking", or "the way they

look." Such statements do not reflect specific attributes

 88

from a definitional stance; however, the statements, though

obtuse, do connote general attitudes about attractiveness and

sensuality.

After the content analysis was conducted on the data, the

categories were collapsed into three general concepts to be

tested between and within groups on their definitions of

attractiveness and sensuality. The coding scheme was broken

into three major components: face, body, and

personality/style. The categories were fairly simplistic to

divide, except for a few intangible qualities which were not

as obvious. Clothes and hygiene were conceptualized as

personality dimensions, as was the monetary dimension.

Because a smile represents a particular mood or personality

condition, it was operationalized as a personality dimension

as opposed to a facial characteristic. The category age was

dropped from the comparative analysis, due to the obtuse

nature of the concept, and its low frequency among the two

samples. Responses in the general looks/overall appearance

category were divided equally among the base categories of

body, face, and personality.

The responses were tabulated by summing the frequency of

responses, not the number of respondents specifying a

particular category. Given the respondents had an opportunity

to list several characteristics, it seemed reductionistic to

limit the unit of analysis to the respondent. Tests for

significant differences between sample groups was ruled out

 89

due to the nominal context of the data, and that the unit of

analysis was responses versus respondents. The frequency of

responses were crosstabulated within the two male-oriented

sample groups (heterosexual females and homosexual males) and

the female-oriented sample (heterosexual males and lesbians)

regarding the differentiation of the perspective groups

definition for sensuality and attractiveness. The responses

were then compared between the heterosexual and homosexual

samples regarding the definitions of sensuality and

attractiveness.

Cautions and limitations

The data for this analysis should be interpreted with

some degree of caution. First, the samples are not readily

comparable for a precise analysis between groups. The gay

males used in the analysis are substantially older than the

females, which may suggest an age or cohort differentiation as

opposed to a gender difference. Secondly, it is assumed the

gender identity of the two samples are normally distributed.

This remains a substantial assumption providing gender tests

were not administered to the participants. Third, the

analysis is simplistic. The categories reflect only

dimensions of attractiveness or sensuality, not the extent

(variability in the dimension) or the relative importance of

the dimensions.
 90

RESULTS

Definitions of attractiveness

Tables 1, 3, 5, and 7 reproduce the frequency of

responses specified by homosexual males and females, and

heterosexual females and males regarding the definition of

attractiveness. Based on the percentage of responses for the

variety of traits, several traits emerged as more viable than

others.

Homosexual males. If based on the criteria of

exceeding twenty percent of the total responses as being

highly relevant traits, than homosexual males designated eyes,

hair, height, muscularity, buttocks (approximately), chest,

and general facial features as being attractive. When

compared to heterosexual females, gay males exhibited a

stronger preference for hands, genitalia, chest, hygiene, the

general categories: face, body, and looks. Gay males also

responded more often to the subject of body hair contrasted

with the responses of heterosexual females. The category of

body hair for gay males exceeded female responses by nine

percent. For many of the gay males specifying body hair,

implied the more body hair the better, whereas for

heterosexual females they indicated the less body hair the

better. With regard to hygiene, only seven percent of the

responses made by gay males reported the way a man smelled was

significant to them; consequently, only two percent of

heterosexual female responses specified this particular

 91

category. The most startling preference indicated by gay

males, not issued by heterosexual females, was that of

genitalia. Seventeen percent of the gay males responses

favored large genitalia, as compared to zero responses by

heterosexual females.

Heterosexual females. Heterosexual females specified

eyes, hair, height, muscularity, buttocks, general facial

characteristics, and smiling as being prominent categories.

It appears heterosexual females place a greater emphasis on

eyes, mouth, and hair when compared to homosexual males.

Twice as many females indicated hair as being relevant in

their definitions of attractiveness than the gay males in this

study. Slightly more heterosexual females (by six percent of

responses) over males indicated eyes and mouth were of

interest in defining attractiveness for males. Heterosexual

female responses were also more favorable toward height,

smiling, and muscularity when compared with the responses of

gay males. The difference in responses describing muscularity

was over twenty percent. Both height and smiling

differentiated by approximately eleven percent between the

.... ^

males and females in this study.

Heterosexual males. Again, using the twenty percent

and over criteria as a means to specify particular traits as

analytically meaningful, then heterosexual males selected

eyes, hair, muscularity, legs, breasts, buttocks, and face (in

general) as being substantially significant. Twice as many

 92

males stipulated height, legs, and breasts when compared to

lesbians. Further, approximately nineteen percent of

heterosexual males designated overall body features when

contrasted with less than one percent of the lesbian sample.

Lesbians. Lesbians indicated eyes, hair, muscularity,

general looks, and smile as relevant features in defining

female attractiveness. When comparing lesbians to

heterosexual males in definitional content for female

attractiveness, thirteen percent more lesbians specified the

category of eyes over heterosexual males. Eight percent more

lesbians cited smile as being considered attractive over

heterosexual males. Twice as many lesbians indicated

mannerisms/movement, and affectionate/romantic characteristics

as being more attractive when compared to heterosexual males.

And fourteen percent more lesbians were more likely to

designate eyes as being attractive when contrasted with

heterosexual males.

Definitions of sensuality

Analyzing the percentage of responses between homosexual

males and heterosexual females revealed some intriguing

contrasts. Tables 2 and 4 reflect the number of responses,

and percentage of responses for each category among homosexual

males and heterosexual females. First, it appeared the

percentage of responses for the definition of sensuality were

more varied among the designated categories than the responses

 93

regarding the definition of attractiveness. Once again,

heterosexual females indicated, by their responses, a stronger

preference for muscularity (by eleven percent) than gay males.

The women in the study also exhibited higher responses for the

categories of smiling^ affectionate/romantic behaviors, and

warmth/sensitivity. The differences between sample groups

were not exceptionally large, but perhaps substantively

significant. On the other hand, homosexual males were more

likely to respond to the categories of hair, genitalia,

intelligence, and body hair when compared to females.

Tables 6 and 8 reflect the frequency of responses in the

definition of sensuality among heterosexual males and

lesbians. Again, more heterosexual males (fourteen percent)

indicated hair was sensuous than lesbians; however, more

lesbians responded eyes (11 percent) were sensuous over

heterosexual males. Twice as many males concluded muscularity

was a sensuous characteristic compared to lesbians.

Heterosexual males were also more likely to designate

buttocks, breasts, and clothes as sensuous features for

females when contrasted to lesbians. Thirteen percent more

lesbians than heterosexual males targeted confidence as

sensuous; whereas, twenty-seven percent more heterosexual

males over lesbians specified affectionate/romantic features

as sensuous for females.

 94

Definitional differences between attractiveness and sensuality

A major proposition posited in this study asserted what

is sensual and what is attractive are two unique, and distinct

concepts. This hypothesis was verified in the crosstabulation

analysis of the respondents' definitions of attractiveness

when compared to sensuality. Tables 9, 10, 11, and 12 present

the number of responses for each major classification (face,

body, and personality) concerning the definitions of

attractiveness and sensuality within each targeted sample.

Among homosexual males (Table 9), body characteristics were

more important than facial features for definitions of

attractiveness and sensuality. However, personality traits

were least relevant regarding definitions of attractiveness;

conversely a man's personality was preferred over facial and

body characteristics for gay males' definition of sensuality.

The Chi square product yielded by the test of significance was

overwhelming (chi square with 2 degrees of freedom = 224.163,

p < .000). A parallel trend was evident for heterosexual

females (see Table 10). A man's body tended to be more

important than his face when judging levels of both

attractiveness and sensuality in men. And personality

characteristics were very irrelevant for definitions of

attractiveness, but were exceptionally important in judging a

male's sensuality among heterosexual females. The chi square

product was considerably higher for heterosexual females [chi

square (2)=334.470; p < .001], indicating a significant

 95

difference between definitions of attractiveness and

sensuality.

Table 11 reflects the crosstabulation analysis between

the definition of attractiveness with the definition of

sensuality among lesbians. A considerable difference was

apparent in the contrasting definitions. With regard to

attractiveness, a woman's face and body were equally

important, more so than personality. However, personality

was, by far, the most explicit category cited by the lesbian

sample in defining sensuality (chi square with two degrees of

freedom = 130.947, p < .001). Table 12 reports the contrast

in definitional components between attractiveness and

sensuality among heterosexual males. Similarly, heterosexual

males reported the face and body were significant features

which defined female attractiveness. However, personality

characteristics appeared to be the most influencing factor in

the definition of sensuality, with body being second (chi

square with two degrees of freedom = 195.103, p < .001).

Group comparisons

The numerous categories of attractiveness were collapsed

into three major categories: face, body, and personality.

The categories were then contrasted between the four sample

groups using a crosstabulation of responses. A chi square

product was then calculated to estimate the significance of

differences associated in heterosexual females' and homosexual

 96

males' definitions of male attractiveness; and, between

heterosexual males and lesbians in definitions of female

attractiveness. Table 13 presents the percent of responses

made by gay males and heterosexual females. The chi square

indicated no significant differences between gay males and

straight females concerning definitions of attractiveness (chi

square with two degrees of freedom = 4.08, p < .200). It

appeared a man's body dominated the percentage of responses

for both target groups; a man's face was second in percentage

of responses. And a man's personality was the least indicated

among the participants in this study. Table 15 illustrates

the comparison of responses between lesbians and heterosexual

males. The chi square product indicated a significant

difference in the means lesbians and heterosexual males

specify qualities of female attractiveness among the

categories of face, body, and personality (chi square with two

degrees of freedom = 24.104, p < .001). It appeared a woman's

body is of more interest to heterosexual males than lesbians;

consequently, twice as many lesbians indicated personality

characteristics were important if defining female

attractiveness.

As with attractiveness, a chi square test was conducted

on the definition of sensuality between the four groups.

Again, the categories were condensed into three basic

classifications. Table 14 illustrates the differences in

definitions of sensuality between heterosexual females and

 97

homosexual males. The highest percentage of responses for

both males and females fell into the classification of

personality. It appeared personality was a convincing trait

in the definition of sensuality. A man's body was the second

most identified category among the respondents, and facial

characteristics maintained the lowest percentage of responses.

The response differential between heterosexual females and

homosexual males regarding the perspective groups' definition

of sensuality was significant (chi square = 18.71, p < .001).

The results indicated a significant difference in percentage

of responses for personality traits, and facial

characteristics among the two samples. A man's body yielded

similar response rates for straight females and gay males.

More gay males than straight females reported facial

characteristics, in general, were more significant in their

definitions of sensuality. And more females than males

demonstrated a stronger preference for personality traits.

Table 16 presents the crosstabulation of responses

between heterosexual males and lesbians regarding the

definition of sensuality. The overwhelming majority of

respondents denoted personality characteristics over face and

body features in the definition of female sensuality. More

lesbians indicated preference for personality characteristics

than heterosexual males. And more heterosexual males

delineated body features over lesbians. Slightly more

lesbians specified facial qualities when compared to

 98

heterosexual males. The results of the chi square test was

significant (chi square with two degrees of freedom = 18.97, p

< .001).
 99

DISCUSSION

The results of this study suggested that the construct

definitions of sensuality and attractiveness were

significantly different. The definition of sensuality was

markedly different from the definition of attractiveness for

both females and males. This finding critically challenged

past studies conducted on the effect of attractiveness on the

selection of romantic partners. Given the two unique

constructs, future researchers should be concerned with

topological definitions utilized in their projects. Further,

future investigators should employ personality dimensions in

their analysis of both attractiveness and particularly

sensuality. Since the definition of sensuality was

significantly composed of personality characteristics,

including a personality dimension to definitions of. sensuality

should heighten the results of future research efforts.

The schematic arrangements of cognitive structures

employed in defining attractiveness and sensuality varied

along gender lines. Although some characteristics

demonstrated similar importance, other categories yielded

differential importance between males and females. Eyes,

hair, height, muscularity, and general facial characteristics

were commonly judged as important components of attractiveness

for both males and females. Both lesbians and heterosexual

females were more likely to signify smiling as important in

their definitions of attractiveness. On an abstract level.

 100

smiling infers an expressive element dependent on mood, and

feelings. Gay males denoted genitalia and chest as being

relevant characteristics in assessing attractiveness.

Genitalia and chest as characteristics important for sexual

relations may be defined as instrumental. The same was true

for heterosexual males, defining both buttocks and breasts as

attractive. Therefore, both heterosexual and gay males were

more likely to indicate instrumental qualities as attractive,

and females were more likely to focus on expressive qualities.

Gender differences were more apparent regarding

definitions of sensuality. By analyzing the crosstabulation

of sensuality between heterosexual females and homosexual

males, it appears females were significantly more interested

in personality categories; whereas homosexual males favored

physical attributes. Once again, the two groups differed

along gender schematics: males endorsing instrumental

features of strength and masculinity, and females subscribing

to expressive traits like personality dimensions. In short,

gender schémas are relevant to the definitions of sensuality

and attractiveness and should be studied further.

The definition of attractiveness was found to be

significantly similar among heterosexual females and

homosexual males, but not so between lesbians and heterosexual

males. This result supports the hypothesis that

attractiveness is linked to a cultural schema which dictates a

normative structure to the definition of attractiveness for

 101

both gay males and heterosexual females. It could be inferred

from the data that lesbians, perhaps, maintain a unique

cultural schemata pertaining to the definition of female

attractiveness when compared to heterosexual males. The data

indicated lesbians admired personality qualities as attractive

over obvious physical attributes. Whereas, heterosexual males

paid more attention to the physical attributes of body and

face. The cultural scenario script of lesbians focused on

personality traits such as warmth and sensitivity, emphasizing

intimacy attributes over the physical. The cultural scenario

script for heterosexual males indicated straight males

centered on more implicitly sexual qualities, such as buttocks

and breasts. The contrast between the cultural scenario

scripts among heterosexual males and lesbians reflected the

stark subcultural differentiation between intimacy versus

sexuality, regarding the definition of attractiveness.

Definitions of sensuality were significantly different

between the four comparison groups. This finding infers

sensuality is conceptually linked to the cognitive structures

congruent with relational and intrapsychic schémas. Since

relational and intrapsychic schémas are individually variant,

this finding suggests definitions of sensuality are

individual-oriented. What is sensual to one individual is not

necessarily sensual for another. Considering the variance of

responses among all four comparison groups, it appears the

definition of sensuality is in the eye of the beholder.

 102

Though there is some overlap among the subjects' responses

between attractiveness and sensuality, sensuality appears to

be more of a personality quality as opposed to a complete

physical attribute.

In conclusion, it is hoped that future investigators

first clarify and substantiate their definitions of

attractiveness and sensuality in their research. Second,

further analyses is in order to substantiate tests of

significance on gender, and sexual orientation differences.

Third, an analyses on the importance of certain features over

others should be evaluated. The study has indicated gender,

and sexual orientation differences exist regarding the

definitions of attractiveness and sensuality. Perhaps, it

could be inferred from this study that attractiveness is in

the eye of the culture or cultures, and sensuality is in the

eye of the beholder.

 103

BIBLIOGRAPHY

Agnew, R. 1984. "Appearance and delinquency." criminology:

An Interdisciplinary Journal 22:421-440.

Alicke, M. D., R. H. Smith, and M. L. Klotz. 1986.

"Judgments of physical attractiveness: the role of faces
and bodies." Personality and Social Psychology Bulletin
12:381-389.

Cunnigham, M. R., A. P. Barbae, and C. L. Pike. 1990. "What

do women want? Facialmetric assessment of multiple
motives in the perception of male facial physical
attractiveness." Journal of Personality and Social
Psychology 59:61-72.

Donovan, J. M., E. Hill, and W. R. Jankowiak. 1989. "Gender,

sexual orientation, and truth-of-consensus in studies of
physical attractiveness." The Journal of Sex Research
26:264-271.

Downs, A. C. 1990. "Objective and subjective physical

attractiveness judgments among young adults." Perceptual
and Motor Skills 70:458.

Efran, M., and E. Patterson. 1974. "Voters vote beautiful;

the effects of physical attractiveness on a national
debate." Canadian Journal of Behavioral Science 6:352-
356.

Feingold, A. 1990. "Gender differences in effects of

physical attractiveness on romantic attraction: a
comparison across five research paradigms." Journal of
Personality and Social Psychology 59:981-993.

Feinman, S., and G. W. Gill. 1978. "Sex differences in

physical attractiveness preferences." The Journal of
Social Psychology 105:43-52.

Franzoi, S. L., and M. E. Herzog. 1987. "Judging physical

attractiveness: what body aspects do we use?"
Personality and Social Psychology Bulletin 13:19-33.

Gladue, B. A., and H. J. Delaney. 1990. "Gender differences

in perception of attractiveness of men and women in
bars." Personality and Social Psychology Bulletin
16:378-391.

Glaser, B. G., and A. L. Strauss. 1967. The Discovery of

Grounded Theory: Strategies for Qualitative Research.
New York, NY: Aldine De Gruyter.
 104

Graziano, W. G., T. Brothen, and E. Berscheld. 1978. "Height

and attraction: do men and women see eye-to-eye?"
Journal of Personality 46:128-145.

Harry, J. 1990. "A probability sampling of gay males."

Journal of Homosexuality 19:89-104.

Hovrath, T. 1981. "Physical attractiveness; the influence of

selected torso parameters." Archives of Sexual Behavior
10:21-24.

Jackson, L. A., L. A. Sullivan, and J. S. Hymes. 1986.

"Gender, gender role, and physical appearance." The
Journal of Psychology 121:51-56.

Keating, C. F. 1985. "Gender and the physiognomy of

dominance and attractiveness." Social Psychology
Quarterly 48:61-70.

Krebs, D., and A. Adinolfi. 1975. "Physical attractiveness,

social relations, and personality style." Journal of
Personality and Social Psychology 31:245-253.

Kulik, J. A., and J. Harackiewicz. 1979. "Opposite-sex

interpersonal attraction as a function of the sex-roles
of the perceiver and the perceived." Sex Roles 5:443-
452.

Lavrakas, p. L. 1975. "Female preferences for male

physique." Journal of Research in Personality 9:324-334.

Lucker, G. W., W. E. Beane, and K. Guire. 1981. "The

ideographic approach to physical attractiveness
research." The Journal of Psychology 107:57-67.

Nida, S., and J. Williams. 1977. "Sex stereo-typed traits,

physical attractiveness, and interpersonal attraction."
Psychological Reports 41:1311.

Patzer, G. 1985. The Physical Attractiveness Phenomena. New

York, NY: Plenum Press.

Rand, C. S., and J. A. Hall. 1983. "Sex differences in the

accuracy of self-perceived attractiveness." Social
Psychology Quarterly 46:359-363.

Simon, W., and J. H. Gagnon. 1984. "Sexual scripts."

Society 22:53-60.
 105

Sprecher, S. 1989. "The importance to males and females of

physical attractiveness, earning potential, and
expressiveness in initial attraction." Sex Roles 21:591-
607.

Symons, D. 1987. "An evolutionary approach: can Darwin's

view of life shed light on human sexuality?" Pp. 91-126
in Theories of Human Sexuality, edited by J. H. Geer and
W. T. O'Donohue. New York, NY: Plenum Press.

Touhey, J. C. 1979. "Sex-role stereotyping and individual

differences in liking for the physically attractive."
Social Psychology Quarterly 42:285-289.

Udry, J. 1965. "Structural correlates of feminine beauty

preferences in Britain and the U.S.: a comparison.
Sociology and Social Research 49:330-342.

Waters, J. 1985. "Cosmetics and the job market." Pp. 51-87

in The Psychology of Cosmetic Treatments^ edited J.
Graham and A. Kligman. New York, NY: Praeger Press.
 106

Table 1. Number of responses, and percent of homosexual males

responding to definitions of attractiveness for
males (n=158)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 48 30.38%
2. Hair 48 30.38%
3. Nose 2 1.27%
4. Chin 4 2.53%
5. Jaw 1 .63%
6. Cheeks (cheekbone) 0 0%
7. Mouth 5 3.16%
8. Skin Color (race) 5 3.16%
9. Skin Complexion (tan) 6 3.80%
10. Facial Hair 10 6.33%
11. Height 41 25.95%
12. Muscularity 85 53.80%
13. Shoulders 2 1.27%
14. Arms 10 6.33%
15. Legs 24 15.19%
16. Hands 11 6.96%
17. Feet 2 1.27%
18. Hips 1 .63%
19. Genitals 27 17.09%
20. Abdomen 5 3.16%
21. Buttocks 31 19.62%
22. Chest/Breasts 37 23.42%
23. Clothes 4 2.53%
24. Hygiene 12 7.60%
25. Sense of Humor 2 • 1.27%
26. Energetic 1 .63%
27. Sensitivity 2 1.27%
28. Confident 1 .63%
29. Af f ect ionate/RomantIc 1 .63%
30. Seductive 0 0%
31. Intelligence 4 2.53%
32. Monetary 1 .63%
, 33. Mannerisms/Movement 2 1.27%
34. Voice/Communication 2 1.27%
35. Nice Personality 0 0%
36. Looks/Overall Appearance 20 12.66%
37. Face (general) 86 54.43%
38. Body (general) 23 14.56%
39. Personality (general) 5 3.16%
40. Smile 18 11.39%
41. Body Hair 16 10.13%
42. Style 3 1.90%
43. Age 3 1.90%

Note: respondents were given an opportunity to specify more

than one item.
 107

Table 2. Number of responses, and percent of homosexual males

responding to definitions of sensuality for males
(n=154)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 42 27.27%
2. Hair 25 16.23%
3. Nose 1 .65%
4. Chin 1 .65%
5. Jaw 1 .65%
6. Cheeks (cheekbone) 1 .65%
7. Mouth 5 3.25%
8. Skin Color (race) 4 2.60%
9. Skin Complexion (tan) 3 1.30%
10. Facial Hair 6 3.90%
11. Height 8 5.19%
12. Muscularity 38 24.68%
13. Shoulders 0 0%
14. Arms 3 1.30%
15. Legs 8 5. 19%
16. Hands 4 2.60%
17. Feet 1 .65%
18. Hips 2 .65%
19. Genitals 23 14.94%
20. Abdomen 3 1.30%
21. Buttocks 14 9.09%
22. Chest/Breasts 8 5.19%
23. Clothes 22 14.29%
24. Hygiene 4 2.60%
25. Sense of Humor 23 14.94%
26. Energetic 14 9.09%
27. Sensitivity 14 9.09%
28. Confident 17 11.04%
29. Affectionate/Romantic 11 7.14%
30. Seductive 12 7.79%
31. Intelligence 33 21.43%
32. Monetary 1 .65%
33. Mannerisms/Movement 27 17.53%
34. Voice/Communication 22 14.27%
35. Nice Personality 11 7.14%
36. Looks/Overall Appearance 24 15.58%
37. Face (general) 23 14.94%
38. Body (general) 21 13.64%
39. Personality (general) 39 25.32%
40. Smile 15 9.74%
41. Body Hair 14 9.09%
42. Style 8 5.19%
43. Age 2 1.30%

Note: respondents were given an opportunity to specify more

than one item.
 108

Table 3. Number of responses, and percent of heterosexual

females responding to definitions of attractiveness
for males (n=188)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 120 63.82%

2. Hair 68 36.17%
3. Nose 2 1.06%
4. Chin 3 1.60%
5. Jaw 2 1.06%
6. Cheeks (cheekbone) 2 1.06%
7. Mouth 17 9.04%
8. Skin Color (race) 0 0%
9. Skin Complexion (tan) 15 7.98%
10. Facial Hair 2 1.06%
11. Height 71 37.77%
12. Muscularity 139 73.94%
13. Shoulders 12 6.38%
14. Arms 10 5.32%
15. Legs 17 , 9.04%
16. Hands 12 6.38%
17. Feet 0 0%
18. Hips 1 .53%
19. Genitals 0 0%
20. Abdomen 5 2.66%
21. Buttocks 43 22.87%
22. Chest/Breasts 22 11.70%
23. Clothes 5 2.66%
24. Hygiene 5 2.66%
25. Sense of Humor 3 1.60%
26. Energetic 0 0%
27. Sensitivity 1 .53%
28. Confident 3 1.60%
29. Affectionate/Romantic 0 0%
30. Seductive 0 0%
31. Intelligence 1 .53%
32. Monetary 0 0%
33. Manner isms/Movement 7 3.72%
34. Voice/Communication 0 0%
35. Nice Personality 2 1.06%
36. Looks/Overall Appearance 15 7.98%
37. Face (general) 42 22.34%
38. Body (general) 13 6.91%
39. Personality (general) 5 2.66%
40. Smile 52 27.66%
41. Body Hair 3 1.60%
42. Style 4 2.13%
43. Age 0 0%

Note: respondents were given an opportunity to specify more

than one item.
 109

Table 4. Number of responses, and percent of heterosexual

females responding to definitions of sensuality for
males (n=176)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1.Eyes 56 31.82%
2.Hair 14 7.95%
3.Nose 1 .57%
4. Chin 0 0%
5. Jaw 2 1.14%
6. Cheeks (cheekbone) 1 .57%
7. Mouth 3 1.70%
8. Skin Color (race) 0 0%
9. Skin Complexion (tan) 8 4.55%
10. Facial Hair 2 1.14%
11. Height 15 8.52%
12. Muscularity 63 35.80%
13. Shoulders 2 1.14%
14. Arms 2 1.14%
15. Legs 9 5.11%
16. Hands 2 1.14%
17. Feet 0 0%
18. Hips 0 0%
19. Genitals 0 0%
20. Abdomen 4 2.27%
21. Buttocks 18 10.23%
22. Chest/Breasts 5 2.84%
23. Clothes 26 14.77%
24. Hygiene 9 5.11%
25. Sense of Humor 34 19.32%
26. Energetic 24 13.64%
27. Sensitivity 40 22.73%
28. Confident 26 14.77%
29. Affectionate/Romantic 27 15.34%
30. Seductive 13 7.39%
31. Intelligence 14 7.95%
32. Monetary 1 .57%
33. Mannerisms/Movement 41 23.30%
34. Voice/Communication 22 12.50%
35. Nice Personality ' ' ...1^ 10.23%
36. Looks/Overall Appearance 28 15.91%
37. Face (general) 14 7.95%
38. Body (general) 16 9.09%
39. Personality (general) 36 20.45%
40. Smile 43 24.43%
41. Body Hair 4 2.27%
42. Style 8 4.55%
43. Age 0 0%

Note: respondents were given an opportunity to specify more

than one item.
 110

Table 5. Number of responses, and percent of heterosexual

males responding to definitions of physical
attractiveness for females (n=149)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 59 39.59%
2. Hair 71 47.65%
3. Nose 4 2.68%
4. Chin 0 0%
5. Jaw 0 0%
6. Cheeks (cheekbone) 1 .67%
7. Mouth • 6 4.03%
8. Skin Color (race) 1 .67%
9. Skin Complexion (tan) 10 6.71%
10. Facial Hair 0 0%
11. Height 23 15.44%
12. Muscularity 39 26.17%
13. Shoulders 1 .67%
14. Arms 0 0%
15. Legs 42 28.19%
16. Hands 3 2.01%
17. Feet 1 .67%
18. Hips 1 .67%
19. Genitals 0 0%
20. Abdomen 4 2.68%
21. Buttocks 48 32.21%
22. Chest/Breasts 49 32.89%
23. Clothes 8 5.37%
24. Hygiene 8 5.37%
25. Sense of Humor 1 .67%
26. Energetic 2 1.34%
27. Sensitivity 2 1.34%
28. Confident 1 .67%
29. Affectionate/Romantic 2 1.34%
30. Seductive 1 .67%
31. Intelligence 5 3.36%
32. Monetary 0 0%
33. Mannerisms/Movement 3 2.01%
34. Voice/Communication 0 0%
35. Nice Personality 3 2.01%
36. Looks/Overall Appearance 11 • 7.38%
37. Face (general) 58 38.93%
38. Body (general) 29 19.46%
39. Personality (general) 5 3.36%
40. Smile 18 12.08%
41. Body Hair 0 0%
42. Style 7 4.70%
43. Age 0 0%

Note; respondents were given an opportunity to specify more

than one item.
 Ill

Table 6. Number of responses, and percent of heterosexual

males responding to definitions of sensuality for
females (n=149)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 34 22.82%
2. Hair 41 27.52%
3. Nose 1 .67%
4. Chin 0 0%
5. Jaw 0 0%
6. Cheeks (cheekbone) 1 .67%
7. Mouth 4 2.68%
8. Skin Color (race) 0 0%
9. Skin complexion (tan) 8 5.37%
10. Facial Hair 0 0%
11. Height 3 2.01%
12. Muscularity 35 23.49%
13. Shoulders 0 0%
14. Arms 1 .67%
15. Legs 27 18.12%
16. Hands 2 1.34%
17. Feet 1 .67%
18. Hips 2 1.34%
19. Genitals 0 0%
20. Abdomen 7 4.70%
21. Buttocks 19 12.75%
22. Chest/Breasts 25 16.78%
23. Clothes 43 28.86%
24. Hygiene 6 4.03%
25. Sense of Humor 20 13.42%
26. Energetic 10 6.71%
27. Sensitivity 14 9.40%
28. Confident 11 7.38%
29. Affectionate/Romantic 51 34.23%
30. Seductive 20 13.42%
31. Intelligence 17 11.41%
32. Monetary 0 0%
33. Mannerisms/Movement 25 16.78%
34. Voice/Communication 12 8.05%
35. Nice Personality 12 8.05%
36. Looks/Overall Appearance 22 14.77%
37. Face (general) 11 7.38%
38. Body (general) 18 12.08%
39. Personality (general) 27 18.12%
40. Smile 18 12.08%
41. Body Hair 0 0%
42. Style 17 11.41%
43. Age 0 0%

Note: respondents were given an opportunity to specify more

than one item.
 112

Table 7. Number of responses, and percent of homosexual

females responding to definitions of physical
attractiveness for females (n=130)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 69 53.08%
2. Hair 48 36.92%
3. Nose 2 1.54%
4. Chin 1 .77%
5. Jaw 1 .77%
6. Cheeks (cheekbone) 2 1.54%
7. Mouth 12 9.23%
8. Skin Color (race) 5 3.85%
9. Skin Complexion (tan) 7 5.38%
10. Facial Hair 2 1.54%
11. Height 10 7.69%
12. Muscularity 46 35.38%
13. Shoulders 4 2.68%
14. Arms 6 4.62%
15. Legs 16 12.31%
16. Hands 10 7.69%
17. Feet 1 .77%
18. Hips 2 1.54%
19. Genitals 1 .77%
20. Abdomen 3 2.31%
21. Buttocks 8 6.15%
22. Chest/Breasts 22 16.92%
23. Clothes 7 5.38%
24. Hygiene 11 8.46%
25. Sense of Humor 1 .77%
26. Energetic 3 2.31%
27. Sensitivity 2 1.54%
28. Confident 10 7.69%
29. Aff ectionate/Romantic 0 0%
30. Seductive 0 0%
31. Intelligence 6 4.62%
32. Monetary 1 .77%
33. Manner isms/Movement 8 6.15%
34. Voice/Communication 2 1.54%
35. Nice Personality 2 1.54%
36. Looks/Overall Appearance 7 5.38%
37. Face (general) 30 23.08%
38. Body (general) 1 .77%
39. Personality (general) 7 5.38%
40. Smile 27 20.77%
41. Body Hair 3 2.31%
42. Style 6 4.62%
43. Age 1 .77%

Note; respondents were given an opportunity to specify more

than one item.
 113

Table 8. Number of responses, and percent of homosexual

females responding to definitions of sensuality for
females (n=130)

Items No. of Responses % of Respondents

Specifying Trait Specifying Trait

1. Eyes 45 34.62%
2. Hair 17 13.08%
3. Nose 1 .77%
4. Chin 2 1.54%
5. Jaw 2 1.54%
6. Cheeks (cheekbone) 1 .77%
7. Mouth 7 5.38%
8. Skin Color (race) 2 1.54%
9. Skin Complexion (tan) 4 3.08%
10. Facial Hair 1 .77%
11. Height 2 1.54%
12. Muscularity 12 9.23%
13. Shoulders 1 .77%
14. Arms 0 0%
15. Legs 5 3.85%
16. Hands 3 2.31%
17. Feet 2 1.54%
18. Hips 3 2.31%
19. Genitals 1 .77%
20. Abdomen 1 .77%
21. Buttocks 7 5.38%
22. Chest/Breasts 9 6.92%
23. Clothes 11 8.46%
24. Hygiene 5 3.85%
25. Sense of Humor 24 18.46%
26. Energetic 12 9.23%
27. Sensitivity 12 9.23%
28. Confident 30 23.08%
29. Affectionate/Romantic 9 6.92%
30. Seductive 13 10.00%
31. Intelligence 23 17.69%
32. Monetary 0 0%
33. Mannerisms/Movement 23 17.69%
34. Voice/Communication 19 14.62%
35. Nice Personality 14 10.77%
36. Looks/Overall Appearance 13 10.00%
37. Face (general) 5 3.85%
38. Body (general) 14 10.77%
39. Personality (general) 32 24.62%
40. Smile 15 11.54%
41. Body Hair 3 2. 31%
42. Style 16 12.31%
43. Age 1 .77%

Note; respondents were given an opportunity to specify more

than one item.
 114

Table 9. Comparison of general qualities in definitions of

attractiveness and sensuality among homosexual males

Face Body Personality Total

Attractiveness 214 339 60 613

Sensuality 137 149 272 558

Total 351 488 332 1171

Chi Square (2) = 224.163

p < .001

Table 10. Comparison of general qualities in definitions of

attractiveness and sensuality among heterosexual
females

Face Body Personality Total

Attractiveness 273 363 88 724

Sensuality 100 167 383 650

Total 373 530 471 1171

Chi Square (2) = 334.470

p < .001
 115

Table 11. Comparison of general qualities in definitions of

attractiveness and sensuality among homosexual
females

Face Body Personality Total

Attractiveness 180 164 93 437

Sensuality 88 76 256 420

Total 268 240 349 857

Chi Square (2) = 130.947

P < .001

Table 12. Comparison of general qualities in definitions of

attractiveness and sensuality among heterosexual
males

Face Body Personality Total

Attractiveness 204 265 58 522

Sensuality 98 158 263 519

Total 302 423 321 1041

Chi Square (2) = 195.103

p < .001
 116

Table 13. Comparison of general qualities in definitions of

attractiveness between heterosexual females and
homosexual males

Percentiles Total No.

Face Body Personality of Responses

Heterosexual 37.71% 50.14% 12.15% 724

Females

Homosexual 34.91% 55.30% 9.79% 613

Males

Total 36.42% 52.51% 11.07% 1337

Chi Square (2) = 4.08

p < .200

Table 14. Comparison of general qualities in definitions of

sensuality between heterosexual females and
homosexual males

Percentiles Total No.

Face Body Personality of Responses

Heterosexual 15.383 25.69% 58.92% 650

Females

Homosexual 24.553 26.70% 48.75% 558

Males

Total 19.62% 26.16% 54.22% 1208

Chi Square (2) = 18.71

p < .001
 117

Table 15. Comparison of general qualities in definitions of

attractiveness between heterosexual males and
homosexual females

Percentiles Total No.

Face Body Personality of Responses

Heterosexual 38.71% 50.28% 11.01% 527

Males

Homosexual 41.19% 37.53% 21.28% 437

Females

Total 39.84% 44.50% 15.66% 964

Chi Square (2) = 24.104

p < .001

Table 16. Comparison of general qualities in definitions of

sensuality between heterosexual males and
homosexual females

perçentiles Total No.

Face Body Personality of Responses

Heterosexual 18.88% 30.44% 50.684 519

Males

Homosexual 21.00% 18.004 61.00' 420

Females

Total 19.81% 24.92% 55.27% 939

Chi Square (2) = 18.97

p < .001
 118

PAPER 3. THE VIOLENT EFFECTS OF DISCRIMINATION

ON GAY AND LESBIAN RELATIONSHIPS
 119

ABSTRACT

The objectives of this study were, in general, to analyze

the relationship between physical discrimination and

aggression in gay and lesbian relationships. The proposed

models asserted physical and sexual discrimination was caused

by the extent of being "out of the closet". Further, it was

hypothesized discrimination would increase the extent of

physical victimization, the perpetration of -abuse, and sexual

victimization in gay and lesbian relationships. Finally,

abuse incurred in the relationship would lead to unstable

relationships. The models were moderated by self-esteem. The

sample consisted of 126 lesbians, and 167 gay males. The

results of the analyses indicated being "out" fostered

discrimination among lesbians with high and low self-esteem;

however, being "out" lead to discrimination for gay males with

low self-esteem only. Discrimination was significantly

associated with physical victimization, the use of violence,

and being a victim of sexual aggression among gay males

regardless of self-esteem. Discrimination was related to

physical victimization, the use of violence, and sexual

aggression only among lesbians with low self-esteem. The

" * •..

various forms of abuse were highly associated with relational

stability. The more abuse encountered in the relationship,

the less likely the relationship was maintained for both gay

males and lesbians, regardless of self-esteem.

 120

INTRODUCTION

Very little empirical research has been conductcd on

violent gay and lesbian relationships. The sparse research,

which has been available on homosexual victimology, is

dramatic in its presentation. However, duo to the qualitative

nature of the data, past research sacrifices generalization

and probability analysis to the phenomenon. Of the studies

which have been conducted on gay/lesbian abuse, the results

have primarily been descriptive. For example, Brand and Kidd

(1986) concluded lesbians reported slightly less abuse (both

physical and sexual) when compared to heterosexual females.

On the other hand, Kelly and Warshafsky (1987) found the vast

majority of a self-selected sample of gays and lesbians were

verbally abusive; and, over forty percent of the sample had

used physically aggressive tactics during the course of their

relationships. However, the researchers did not examine the

extent of perpetration or victimization regarding the

aggressive relationships. Kelly and Warshafsky (1987) did

conclude gay males were more likely to encounter violence in

their relationships than lesbians, but not significantly.

According to Duncan (1990), gays and lesbians were more likely

to experience forced sexual participation than heterosexual

college students, although, their sample consisted of less

than seventy gay/lesbian subjects. Finally, substance abuse

has been documented, among lesbian partners, as a correlate of

domestic violence. Schilit, Lie, and Montagne (1990)

 121

recounted two-thirds of their sample asserted alcohol or drugs

were implicated during an abusive relationship.

 122

MODELING VIOLENCE IN GAY/LESBIAN RELATIONSHIPS

Gay and lesbian relationships have posed unique problems

to family violence researchers. Generally, homosexual

relationships are structured differently than heterosexual

relationships, from the inherent structure of the relationship

to the prevailing stigma of homosexuality in the dominant

culture. Therefore, various considerations are necessary when

modeling the potential indicators and causes of violence in

gay and lesbian relationships. First, a model diagramming gay

and lesbian violence needs to account for the general cultural

context gay and lesbian relationships are consistently exposed

to. Second, a model of gay and lesbian violence has to

incorporate how the stigma associated with homosexuality

affects the nature of violence in their relationships, whether

physical, sexual, victimization, or perpetration. Third, the

model should investigate the affect of the violent episodes on

the stability of the relationships.

This study focuses on applying a sexual script

perspective to various violent episodes in gay and lesbian

relationships. The analysis in this study incorporates the

three interdynamic levels of sexual scripts. The three levels

of sexual scripts include cultural scenario scripts,

interrelational scripts, and intrapsychic scripts. In

general, sexual scripts are general cognitive guidelines

ordering sexual definitions, attitudes and behaviors.

Cultural scenario scripts are scripts directed by broader

 123

social forces, detailing appropriate and inappropriate sexual

conduct. Interrelational scripts are scripts which reflect

the interdynamic qualities of past and present relationships

regarding an individual's sexual identity. Intrapsychic

scripts describe sexual desires, fantasies, and motivations.

An examination of how all three levels of scripting influence

not only the effect of violence, but also the correlates of

abuse in gay and lesbian relationships. For example, by

investigating the relationship between cultural perspectives

toward homosexuality (cultural scenario scripts), with the

psychological stamina of gays and lesbians (intrapsychic

scripts), the effect of violence in gay/lesbian relationships

(interrelational) can be further evaluated. Therefore, it is

necessary to examine all three aspects of sexual scripts to

obtain a fully developed understanding of how violence is

initiated, mediated, and sustained within gay and lesbian

relationships.

Cultural scenario scripts

It is posited being "out of the closet", gays and

lesbians would incur more discrimination than those gays and

lesbians who do not readily disclose their identity.

Although, disclosure of homosexuality may be a precursor to

discrimination, there are different intervals of disclosure.

It could be argued those gays and lesbians who are flamboyant

with their sexual identity are more likely to incur

 124

discrimination. While those gays and lesbians who disclose

their sexual identity to only a few individuals are not as

likely to face the extent of discrimination as those who

conspicuously disclose their orientation. Further,

conspicuous homosexuals would incur more discrimination than

those who do not disclose their sexual identity. From a

cultural scenario perspective, it could be argued heterosexual

sexual scripts entail moderate social approval of violence

directed at gays and lesbians. According to Wolfgang and

Ferracuti (1967) subcultural approval toward violence is a

matter of differential identification and association with the

predominant values of a given culture. Therefore, the extreme

forms of physical and sexual harassment endured by gays and

lesbians could amplify frustration or extend current

definitions of violence within gay/lesbian relationships.

Hvpothesis 1. Being "out of the closet" will increase

the amount of discrimination incurred among gays and lesbians.

According to the National Gay and Lesbian Task Force over

ninety percent of gays and lesbians are victimized, harassed,

and discriminated against on the basis of sexual orientation

(Mohr, 1988). Given cultural scenario scripts are decisively

geared toward heterosexual-oriented life styles and behaviors,

discrimination toward gays/lesbians reflects the extent of

homophobia apparent in society. Discrimination based on

sexual orientation serves as a stress producing mechanism

within gay/lesbian relationships, amplifying frustration and

 125

aggression. Physical or sexual violence or harassment of

homosexuals because of their sexual orientation may be

conducive as an arousal stimulus toward perpetrating violence,

or serve to displace aggression in their own intimate

relationships.

Hypothesis 2. High levels of discrimination will

increase the perpetration of violence in gay and lesbian

relationships.

Discrimination may also serve as a means to broaden an

individual's definition of victimization. By employing a

symbolic interactionist perspective to explore discrimination

and relational violence, for the focus of such a theory would

examine "the different meanings of violence people hold ...

and the consequences of some meanings in situational settings"

(Bersani and Chen, 1988; p. 65). The violence, imposed on

gays and lesbians via discrimination, may actually broaden

their symbolic criteria of victimization; and, perhaps they

will redefine similar episodes, occurring in their

relationship, as being violent.

Another potential correlate of abuse in gay/lesbian

relationships is power. Bologna, Waterman, and Dawson (1987)

suggested a perceived lack of power was associated with both

victimization and perpetration in lesbian relationships. Both

physical and sexual discrimination undermines the perception

of social power for both gays and lesbians. Given

heterosexuals maintain the power in society, and enforce their

 126

power via discrimination, prejudices based on sexual

orientation eventually disrupt the relational quality of gay

and lesbian relationships. Further, the detrimental effects

of discrimination on gay and lesbian relationships cuts across

all power or socio-economic levels. For example, Kelly and

Warfshafsky (1987) reported no significant relationship

between status differentials and incidence of abuse in

gay/lesbian relationships.

Hypothesis 3. High levels of discrimination will

increase physical victimization in gay and lesbian

relationships.

Hypothesis 4. High levels of discrimination will

increase sexual victimization in gay and lesbian

relationships.

The relationship between discrimination and victimization

may be a product of several sources. One explanation could be

a sensitization effect, whereby gay males and lesbians who had

been victimized by heterosexuals may be more likely to define

themselves as victims in their own intimate relationships. A

second explanation could stem from an unmeasured intervening

variable, such as deviant coping mechanisms. Further,

discriminated homosexuals may suffer from low self-esteem and

seek out potentially violent relationships because they have

been conditioned to accept deviance as a lifestyle. A third

explanation could be assertiveness.

 127

Interrelational scripts

It is posited the more violent the relationship, the more

likely the relationship will be unstable. It is assumed

violence in a relationship will detract from the stability of

the relationship, as a whole. Stemming from this postulation,

is a further inquiry. If violence has no significant effect

on the stability of gay/lesbian relationships, then it might

imply the gay and lesbian subculture leans toward a high

approval of violence within their relationships, supporting

the subcultural violence thesis.

Because gays and lesbians tend to maintain "closed"

relationships, the balancing of autonomy and attachment is a

more prevalent concern for homosexual relationships (Peplau,

Cochran, Rook, and Padesky, 1978; Kreston and Bepko, 1980).

As a "closed" system, relational, discrimination faced by

lesbians and gays may foster higher degrees of attachment,

which may prove to be conflictual when problems of autonomy

surface (McCandlish, 1982; Lindenbaum, 1985). Therefore, gay

and lesbian partners tend to be more dependent on their

partners within their relationships. Renzetti (1988)

demonstrated dependency appears to be a strong correlate with

abuse in lesbian relationships. As a lesbian or, perhaps, a

gay male partner attempts to achieve autonomy, and if their

partner is highly dependent and aggressive, the relationship

may become abusive. The correlate between dependency and

 128

abuse may further be extenuated when discrimination is

encountered by gays and lesbians.

According to Strube (1988), the decision to leave an

abusive relationship is, in part, based on perceived

entrapment, learned helplessness, and a cost/benefit analyses.

Given there are few mediation agencies and support services

for victims of gay and lesbian abusive relationships, many gay

males and lesbians, who are sexually or physically abused,

must feel entrapped by their relationship and by society's

general disapproval of their sexual life style. Further, gays

and lesbians do not have the social resources available to

them to promote a perceptual benefit for leaving an abusive

relationship. The lack of available resources also fosters a

feeling of learned helplessness. Therefore, it would not be

surprising to find gays and lesbians are less likely to leave

abusive relationships. However, according to Celles (1976)

heterosexual women who had incurred severe or frequent abuse

were more likely to seek intervention, with separation and

divorce as the most prominent forms of intervention.

Hypothesis 5. Perpetration of violence, physical

victimization, and sexual victimization will increase the

instability of gay and lesbian relationships.

Intrapsychic scripts

It is hypothesized the self-esteem of the subjects will

moderate the extent discrimination will influence the level of

 129

abuse (whether physical, sexual, or battery) and the stability

of the relationship after the violent episode. Self-esteem

can be conceptualized as a variable at the intrapsychic level,

conducive to explain sexual and physical abuse, and

perpetration of violence in intimate relationships. Low self-

esteem, in the violence literature, has been substantiated as

a characteristic of batterers or aggressors (Gayford, 1975;

Goldstein and Rosenbaum, 1985). Low self-esteem has also been

correlated with being a victim of abuse (Walker, 1979;

Carlson, 1977). It is apparent self-esteem, as a personality

construct, is associated with violence, whether through

perpetration or victimization. Therefore, it is posited self-

esteem, as a moderating variable, will influence the

relationship between discrimination and perpetrating, or being

a victim of aggression. Lesbians and gay males with low self-

esteem will be more likely to adopt a victim perspective

whether through discriminatory acts or abusive relationships.

Further, lesbians and gay males with low self-esteem will be

more likely to abuse their partners, because they will not

possess the means to adequately deal with the stress

associated with discrimination.

According to Abramson, Seligman, and Teasdale (1978), one

means for assessing learned helplessness is self-esteem. In

their research (Abramson, Garber, and Seligman, 1980) proposed

self-blame for uncontrollable events are more likely to

manifest low self-esteem. Victims of discrimination, a

 130

relatively uncontrollable event from the perception of gays

and lesbians, are more likely be influenced by learned

helplessness. Therefore, lesbians and gay males who are

physically discriminated against are more likely to consider

themselves as victims in abusive relationships.

Hypothesis 6. Gays and lesbians with high self-esteem,

when compared to those with low self-esteem, will be less

influenced by the association between discrimination and

perpetration of violence, physical victimization, and sexual

victimization.

Hypothesized models

Figure 1 illustrates the three proposed models of this

study. The three models differ regarding the paths from

discrimination to the three forms of violence examined in this

study. Model A illustrates the model examining the effect of

discrimination of physical victimization. Model B diagrams

the model analyzing the effects of discrimination on

perpetration of violence. Model C indicates the effect of

discrimination on sexual aggression. To reiterate, it is

hypothesized being "out" will increase the probability for

gays and lesbians to be victims of physical/sexual

discrimination by heterosexuals (Arrows A). Gays and

lesbians, who are victims of violent forms of discrimination,

will be more apt to define aggressive episodes in their

relationships as violent (Arrow B) or sexually abusive (Arrow

 131

C). Further, discrimination may increase stress within gay

and lesbian relationships, which could amplify aggressive

tendencies; consequently, discrimination may increase the

probability gays and lesbians resort to using violence within

their relationships (Arrow D). And, those gays and lesbian

who incur violence within their relationships will be more

likely to maintain relatively unstable relationships (Arrows

E). Finally, it is posited gays and lesbians with high self-

esteem will not be as likely to be victimized or perpetrate

violence in, their relationships as gays and lesbians with low

self-esteem (Arrows F).

Figure 1 about here

 132

METHODS

Measurement

The key variables in this study include: being "out"

(disclosure), physical and sexual discrimination, physical

violence, sexual aggression, the effect of the violent episode

or sexual aggressive act on the relationship, and self-esteem.

Phvsical and sexual discrimination. Physical and

sexual discrimination were assessed by the summing of two

general dichotomous items. Both items attempted to ascertain

the extent of victimization by heterosexuals. The

introductory statement for the two items read; "The following

list are examples of things that could have happened to you.

For each item, circle either yes or no. Circle yes only if

this directly happened to you AND if it occurred because

others believed you are a gay/lesbian." Two items were

selected from a list of possible discriminatory outcomes. The

initial list included: physical, sexual, occupational,

housing, property-related, and familial. Both physical and

sexual discrimination were targeted because of the direct

violent nature of both forms of discrimination. Though the

other forms of discrimination faced by gays and lesbian, may

constitute relational and psychological stress, they are not

necessarily direct, physical confrontations.

Physical discrimination was measured as a dichotomous

category (either "yes" or "no") to the following question:

"Someone threatened to or did physically hurt me (verbal

 133

threats, hitting, slap, punch, kick, beat-up, strike with an

object, or use a weapon) because I am gay/lesbian." Sexual

discrimination was assessed in a similar fashion. Respondents

selected either the category of "yes" or "no" to the following

statement: "Someone threatened to or did sexually assault me

or attempt to assault me because I am gay/lesbian (i.e. verbal

threats or made an unsuccessful or successful attempts to

touch breasts or genitals, or force oral, anal, or vaginal

intercourse)." The scores for the respondents were added to

compile a physical/sexual discrimination scale, ranging in

possible responses from 0 to 2. The two items were correlated

at .372 for lesbians, and .268 for gay males. Both

correlations were significant at the .01 level.

Out. Being "out" served as an indicator of gay

identity disclosure and lifestyle activity. Respondents

selected one of five possible response categories to the

following question; "On a scale of 1-5, how would you rate

yourself as 'out of the closet'." The categories included

l="very out", 2="somewhat out", 3="out", 4="somewhat in", and

5="in the closet."

Self-esteem. Self-esteem was assessed by five, general

questions adapted from Rosenberg's Self-Esteem Scale

(Rosenberg, 1965). The five items consisted of "on the whole

I am satisfied with myself"; "at times I think I am no good at

all"; "I certainly feel useless at times"; "I feel that I am a

person of worth, at least on an equal plane with others"; and

 134

"I take a positive attitude towards myself." Respondents were

requested to select one of five possible response categories

to indicate the extent with which they strongly agreed,

agreed, were unsure, disagreed, or strongly disagreed with the

accorded statements. The items were then summed to create a

scale, with possible values of 5 to 25. A factor analysis

demonstrated all the items loaded fairly well on a single

factor. The Cronbach's alpha coefficient for males was .746,

and for females it was .836.

Physical tactic conflict scale. The physical conflict

scale was devised to isolate using violence and being a victim

of violence. The scale was a modified version of the Conflict

Tactic Scale (Straus, Celles, and Steinmetz, 1980). Each

conflict scale ascertained the extent of physical violence

incurred during a gay and lesbian relationship. With regards

to being a victim of violence, participants responded to the

question; "Please indicate the type of violence used toward

you in an intimate homosexual/lesbian relationships (circle

all that apply)." Further, respondents selected the various

forms of violence which applied to the statement: "Please

indicate the types of violence YOU used against your partner

in an intimate homosexual/lesbian relationship. The types of

violent episodes considered for each tactic scale included;

threats, pushing, slapping, punching, striking with an object,

and using a weapon. Each conflict tactic was assessed by a

dichotomous category with l="experienced", and 2="not

 135

experienced." None of the respondents indicated they had used

a weapon against their partner during the course of their

relationship history. Therefore, using a weapon was omitted

from the use of violence scale. The various modes of

violence, for each scale, were summed to create an expression

of physical confrontation in gay and lesbian relationships.

The scale results ranged in possible values from 0 to 6 for

the physical victimization scale, and 0 to 5 for the use of

violence scale. The alpha coefficient for the victimization

scale was .838 for gay males, and .742 for lesbians. The

alpha coefficient on the use of violence scale was .787 for

gay males, and .745 for lesbians. A factor analysis conducted

on both scales for both gays and lesbians indicated all items

loaded on a single factor, respective of perpetration and

victimization.

Sexual aggression. The sexual aggression scale

initially consisted of 12 items. The items consisted of a

vast range of possible methods used to gain sexual behaviors

from the subject. The indicators included such methods as:

intoxication, threats to terminate the relationship, threats

to disclose negative information, guilt, physical detainment,

use of false promises, use of lies, persistent physical

attempts, being held down, use of a weapon, threats of

physical force, and use of physical force (Waldner-Haugrud and

Magruder, 1991). Respondents were asked to indicate the most

intimate behavior that occurred with a person they were

 136

involved with in a lesbian/homosexual relationship despite

their wish not to participate in the sexual behavior. The

response categories for the items were ranked and coded as a

hierarchy of intense sexual involvement, where O="not

applicable", l="kissing, breast/buttocks fondling, genital

fondling", and 2="oral/anal/sexual intercourse." The method

of using a weapon was dismissed from the scale because none of

the respondents indicated this type of event had occurred. In

a factor analysis on the scale, all items loaded moderately on

a single factor, except for physical detainment among

lesbians. Due to the item's substantive strength, and to be

consistent across sexes, physical detainment remained in the

scale despite its poor factor loading. It was believed

physical detainment, although uncorrelated with the other

items in the scale among lesbians, contributed as a extensive

measure of sexual aggression. The eleven items were summed to

composite a scale, indicative of the extent of sexual

behaviors incurred by various means based on undesired

participation. The range of the scale consisted of possible

values, extending from 0 to 20. The alpha reliability

coefficients for the sexual aggression scale were .762 for

lesbians, and .746 for gay males.

Relational stability. The effect of the violent

episode on the relationship, whether inflicted upon

(physically or sexually) or caused by the respondent, was

assessed by three similar questions. Each relational

 137

stability question directly followed the victim of violence

scale, the use of violence scale, and the sexual aggression

scale. The effect of the violent or sexually aggressive

episode was assessed by the following question: "How did the

incident affect your relationship?" Respondents selected 1 of

five possible categories for each aggressive conflict scale.

The categories included; O="not applicable", l="no change",

2="relationship became worse, but still involved",

3="relationship improved" and 4="relationship ended." The

scale was recoded to be indicative of a relational stability

continuum. Therefore, the higher the score on the relational

stability index, the more the subject was involved in the

relationship after the violent episode; consequently, the

lower the score the more likely the subject was to abandon the

relationship.

Procedures

In accord with the exploratory nature of this study,

frequencies on all the key variables were tabulated to profile

the extent of violence occurring in gay and lesbian

relationships. Second, the self-esteem scale was dichotomized

-' » .

at the median to represent a four cell typology: high self-

esteem gay males, high self-esteem lesbians, low self-esteem

gay males, and low self-esteem lesbians. Third, t-tests were

conducted within and between typological groups to test for

significant differences for all the variables in the study.

 138

save self-esteem. Fourth, four sets of correlation matrices,

including all the variables in the study were computed among

gay males and lesbians by high and low self-esteem. Finally,

stacked structural equation models were employed to test for

significant differences along the various paths in the

hypothesized models, controlling for both sex and self-esteem.

 139

RESULTS

The frequency distributions on the variables in the study

yielded interesting findings on the extent of victimization

faced by both gays and lesbians. In general, 46 percent of

lesbians and 29 percent of males recounted they were

physically victimized during the course of their

relationships. The maximum values for both males and females

was 6, on a possible six point scale. The analysis on the

scale was continued on each individual item to further

document the extent of victimization. For example, 32

lesbians and 32 gay males reported being threatened by their

partner during the course of a past relationship. Further, 45

lesbians and 30 males asserted being pushed by their partner.

One quarter of the lesbians were slapped, and similarly,

fifteen percent of lesbians were punched; consequently,

approximately seventeen percent of gay males were slapped, and

fifteen percent reported being punched by their partners.

Only 9 lesbians and 11 males accounted being struck by an

object. Finally, only 3 lesbians and 4 males recounted being

assaulted with a weapon by their partner in a relationship.

With regard to the sample using violence during the

course of their past relationships, a parallel frequency

distribution was noted. As a whole, 37 percent of lesbians

and only 22 percent of gay males reported using some sort of

violence in their relationships. Both males and females

yielded a maximum value of five on the 6-point scale. Fifteen

 140

percent of lesbians, and eleven percent of gay males stated

they had threatened to use violence against their partner.

Twenty-seven percent of lesbians had pushed their partners;

consequently, slightly less lesbians (seventeen percent)

recounted slapping their partners. Eleven percent of the

males in the sample admitted to pushing their partners, and

twelve percent of the males had slapped their partners.

Twelve percent of lesbians, and nine percent of males

confessed to punching their partner at least once in their

relationships. Only four lesbians acknowledged striking their

partner with an object, while none stated they had used a

weapon against their partner. Finally, 5 males reported

striking their partner with an object, and none of the males

recounted using a weapon during his intimate relationship.

The sexual aggression items revealed a similar trend, as

evident in the conflict tactic items. In general, 46 percent

of lesbians and 56 percent of gay males stated they had been

sexually victimized during the course of their relationships.

For lesbians, the maximum scale value was 12, and for males it

was slightly higher at 16. An examination for each of the

various items detailed the extent of sexual victimization

among males and lesbians. Eighteen percent of lesbians and

nineteen percent of gay males reported being purposely

intoxicated by a partner to engage in unwanted sexual

participation. Four percent of lesbians and gay males stated

their partners had threatened to terminate their relationship

 141

unless they were to participate in unwanted kissing, petting,

oral, anal, or vaginal intercourse. Only one lesbian and two

gay males admitted to engaging in unwanted sexual activity

because their partner had threatened to disclose negative

information about them. Twenty-two percent of lesbians and

twenty-four percent of males reported being made to feel

guilty in order to participate in forced sexual relationships.

Approximately four percent of lesbians and eleven percent of

males stated they were physically detained in order to take

part in unwanted sexual activities. A relatively high percent

of lesbians (twenty-seven percent) and males (thirty-five

percent) recounted being victimized by persistent physical

attempts in order to be "turned on." Sixteen percent of

lesbians and twenty-five percent of males asserted being lied

to in order for a partner to obtain unwanted sexual activity.

Similarly, nine percent of lesbians and nine percent of males

acknowledged the use of false promises employed by past

partners to procure undesired sexual relations. Although not

included in the scale, five percent of lesbians and nine

percent of males asserted they were physically held down in

order to obtain unwanted sexual activity. None of the sample,

either lesbians or males, admitted to having a weapon used

against them in order to gain unwarranted sexual

participation. Only one lesbian and five males engaged in

unwanted sexual relations because their partners had

threatened physical force. Finally, three lesbians and seven

 142

gay males stated they had engaged in undesired sexual activity

because their partners had employed physical coercion.

Regarding the stability of the relationships after the

incidence of physical victimization, the majority of males (17

percent) who had endured physical victimization stated the

relationship had ended. Five percent of males reported the

relationship had worsened, but were still involved in the

relationship. Six percent recounted there was no change in

the relationship, two percent stated the relationship

improved. Twenty percent of lesbians reported their

relationship ended after the physical victimization. Seven

percent of lesbians stated there was no change in the

relationship. Fifteen percent asserted the relationship

worsened, but remained involved. And only four percent of

lesbians reported the relationship improved after the violent

incident.

Nine percent of males asserted there was no change in the

relationship after they had employed violent tactics. Five

percent stated the relationship worsened, however they

remained involved. Only one percent of males reported the

relationship had improved. And eight percent of the gay male

sample recounted the relationship had ended after the violent

episode. In contrast, thirteen percent of lesbians reported

their relationship worsened, yet remained involved, after they

had employed some act of violence. Twelve percent of lesbians

stated there was no change in their relationship after they

 143

had used a violent tactic. Four percent of lesbians asserted

the relationship had improved, and nine percent of lesbians

stated the relationship ended.

For sexually aggressive relationships, nine percent of

lesbians reported the relationship remained the same after the

sexually aggressive incident. Only two percent stated the

relationship had improved. The majority of lesbians (27

percent of the lesbian sample) asserted the relationship had

worsened. Only one percent of the gay males asserted the

sexually aggressive relationship had improved. Fifteen

percent stated there was no change in the relationship after

the aggressive episode. And, again, the majority of the gay

males (twenty-seven percent) recounted the relationship had

worsened.

The distribution on the other variables in the study

indicated eighty percent of lesbians considered themselves to

be "out", "somewhat out", or "very out" of the closet;

consequently, eighty-seven percent of the males reported being

"out", "somewhat out", or "very out" with regard to their

sexual identity. One third of lesbians and forty-seven

percent of males recounted being hurt or physically threatened

by discrimination. The average number of assaults reported by

those lesbians and males who were harassed were 7 and 13,

respectively. The incidence of physical discrimination ranged

from 1 to fifty for lesbians, and up to 98 for males.

Fourteen percent of lesbians and nineteen percent of males

 144

stated being sexually assaulted or threatened by

discriminatory harassment. Lesbians reported an average of

ten sexual assaults; however, gay males contended an average

of only 4 sexual assaults.

The self-esteem scale yielded fairly normative results

for both males and lesbians. The range of values for lesbians

extended from six to twenty-five, with a mean value of 18.8.

For males, the values on the self-esteem scale ranged from 10

to 25, with a mean of 19.3. Therefore, to create two groups

representing high and low self-esteem, the scale was

dichotomized at the mean/median. Scores of 18 and lower were

considered to be indicative of low self-esteem, and scores

consisting of nineteen and higher were believed to represent

individuals with high self-esteem.

Table 1 about here

Table 1 presents the means and the standard deviations

for the assorted variables in the study, by sex and self-

esteem. A test of significant differences, controlling for

sex and self-esteem, yielded no significant differences among

lesbians possessing high or low self-esteem. However, a few

significant differences were found between gay males with high

or low self-esteem. Low self-esteem males reported

experiencing significantly more instances of physical

victimization, sexual victimization, and using violence in

 145

their relationships than males with high self-esteem.

Further, low self-esteem males were more likely to recount

they stayed within the sexually aggressive relationship than

males with high self-esteem. And low self-esteem males stated

they maintained the relationships in which they used physical

violence. Among low self-esteem groups, males were

significantly more likely to report they remained in

relationships which they were physically victimized, and they

had employed violent tactics. Among high self-esteem groups,

males again stated they were significantly more likely to be

physically victimized and use violence than high self-esteem

lesbians.

Tables 2 and 3 about here

Tables 2 and 3 present the correlation matrices of the

variables used in the study for lesbians and gay males,

controlling for high and low self-esteem. Among lesbians with

low self-esteem when contrasted to lesbians with high self-

esteem (refer to Table 2), being "out" maintained a'higher

degree of correlation between discrimination, physical

victimization, and the degree of stability for relationships

where lesbians were physically victimized and employed violent

tactics. This finding was also characteristic of gay males

(refer to Table 3). Therefore being "out" or public displays

of identity were more readily associated with violence with

 146

homosexuals possessing low self-esteem. However, gay males

with low self-esteem maintained a higher degree of association

between being "out", sexual aggression, and the stability of

the relationship after a sexually aggressive incident than

lesbians with low self-esteem.

Incurring physical discrimination was moderately

associated with all the variables in the analysis for lesbians

with low self-esteem when compared to lesbians with high self-

esteem. Again, this trend was repeated among the gay males.

The correlation between discrimination and sexual aggression

relational stability was higher among gay males with low self-

esteem than lesbians with low self-esteem. Further, the

correlations between discrimination and the other variables in

the study were higher among gay males with high self-esteem

when contrasted with lesbians with high self-esteem.

Physical victimization was highly correlated with using

violence and sexual aggression for both high and low self-

esteem lesbians, and gay males. This finding supports a

reciprocal effect in violent tactics, although it was not an

strongly correlated with sexual aggression among lesbians.

Therefore, relational violence among gay males is more likely

to be manifested in sexual relationships when paralleled to " •

lesbians. Further, physical victimization was highly

correlated with the relational stability variables. In

particular, the correlation between physical victimization and

relational stability for sexually aggressive relationships was

 147

twice as high among lesbians with high self-esteem than

lesbians with low self-esteem. The association between using

violence and sexual aggression was twice as high for lesbians

with high self-esteem when compared to lesbians with low self-

esteem. Using violence maintained a comparatively higher

correlation with sexual aggression for gay males with low

self-esteem than lesbians with low self-esteem.

In short, the correlation matrices suggest violence, all

forms, incurred in a relationship was negatively, and highly

associated with relational stability. Therefore, gays and

lesbians are not likely to maintain the relationship after

being physically victimized, using violence, or enduring

sexual aggression. Further, the relatively high correlations

among the conflict tactic scales indicate violence leads to

violence. This effect was especially warranted between

physical victimization and the use of violence, for both gay

males and lesbians regardless of self-esteem. However, gay

males with low self-esteem were more likely to transcend their

victimization from physically aggressive relationships to

sexually aggressive relationships, more so than lesbians with

low self-esteem.

Figures 2 and 3 about here

Figures 2, and 3 illustrate the findings of the proposed

models for lesbians and gay males, controlling for high and
 148

low self-esteem. For lesbians (refer to Figure 2), being out

was positively and significantly related to discrimination for

lesbians with both high and low self-esteem. Naturally, this

finding was evident in all three models displaying the effects

of various violent tactics.

However, among gay males (refer to Figure 3), being "out"

was positively and significantly related to enduring

discrimination for low self-esteem males, but not

significantly for males with high self-esteem. Respondents,

describing themselves as being "out of the closet", were more

likely to endure both physical and sexual discrimination from

heterosexuals, except among gay males with high self-esteem.

This trend was evident across all three models.

For lesbians, discrimination was positively related to

physical victimization (Model A), using violence (Model B),

and sexual aggression (Model C); however, this finding was

only significant among lesbians with low self-esteem. Fixing

this path to be equal between lesbians with high and low self-

esteem, significantly changed the chi square of the stacked

model. This significant moderating effect was only true for

the effect of discrimination on using violence (Model B), and

being a victim of physical violence (Model A) [Model A; change

in chi square (1 degree of freedom) = 3.59, p < .1; Model B:

change in chi square (1 degree of freedom) = 4.55, p < .05].

This significant change in the chi square value indicated a

significant contrast between lesbians with low self-esteem and

 149

lesbians with high self-esteem. A significant change in chi

square was not noted regarding the moderating effect of self-

esteem on the relationship between discrimination and sexual

aggression (Model C).

For gay males, discrimination was positively related to

physical victimization (Model A), using violence (Model B),

and sexual aggression (Model C). The trend was significant

regardless of controlling for high and low self-esteem. There

was no significant change in chi square values when the paths

were set to be equal between high and low self-esteem among

the gay males in the study. Therefore, a significant

moderating effect of self-esteem on discrimination effecting

physical victimization (Model A), using violence (Model B), or

sexual aggression (Model C) for gay males was not confirmed.

All three of the conflict tactic scales (physical

victimization, using violence, and sexual aggression) were

negatively related to the extent of relational stability after

the violent episode. This finding was highly significant

among both gay males and lesbians, after controlling for self-

esteem. This finding indicated gay males and lesbians with

either ^igh or low self-esteem did not maintain their

relationship after the violent incidence. Further, the more

violent the incident, or the more extensive the violence, the

less likely the relationship continued.

 150

DISCUSSION

This study demonstrated the violent effects of physical

and sexual discrimination on the intimate relationships of

gays and lesbians. The more identified gays and lesbians were

as homosexuals, or being "out", the more likely they were to

suffer from violent discrimination and abuse by heterosexuals.

Further, the effects of the discrimination were manifested in

relatively violent forms within their relationships.

Discrimination was positively related to the extent of

physical victimization gays and lesbians reported.

Discrimination also lead to an elevated level of perpetrating

violent acts in the relationship by gays and lesbians. And

finally, discrimination was positively related to the extent

of sexual aggression incurred by gays and lesbians in their

relationships. These aggressive encounters undermined the

stability of the relationships. The respondents indicated the

more violent the relationship, the more likely the

relationship was terminated. The moderating effects of self-

esteem were only significant in defining the relationship of

discrimination with both physical victimization and the use of

violence among lesbians. In short, discrimination was not

significantly related to either physical victimization and the

use of violence among lesbians with high self-esteem; however,

the relationships were significantly associated among lesbians

with low self-esteem.

 151

The proposed models offered support to the sexual script

perspective, by implicating the violent effects of

discrimination on gay and lesbian relationships. Given the

cultural scenario scripts in American society are skewed

toward heterosexuality, tumultuous discrimination serves as a

means for reinforcing this dominant sexual perspective. The

increase in hate crimes, directed toward gays and lesbians, in

recent years have confirmed this proposition. It should also

be noted, physical discrimination is more probable among

individuals who identify with the gay culture. By being "out"

homosexuals place themselves at risk for being victimized by

the violent nature of heterosexual scripts. The effects of

physical discrimination on gay and lesbians relationships

tends to further violent outcomes. Unfortunately, the rise

and legitimation of prejudice towards homosexuals increases

the extent of violence in gay and lesbian relationships.

Although this study documented a relatively high amount

of violence, either physical or sexual, this study did not

find support of the subcultural approval of violence

hypothesis. The strong association between the various

dimensions of violence and relational stability indicated that

gays and lesbians who incurred some form of violence were

likely to terminate the relationship. This finding inferred

gays and lesbians do not, in general, maintain or accept

violent relationships. Theoretically, it was considered gays

and lesbians might be prone to a general acceptance of

 152

violence in their relationships due to the stigmatized nature

of their life style. This was not, however, substantiated.

Another theoretical consideration stemmed from the

autonomy/dependency correlate found in past literature. It

appeared gays and lesbians do not confront problems in leaving

violent relationships based on problems with overdependency,

lack of perceived resources, learned helplessness, and

entrapment. However, further research in the area of leaving

an abusive gay and lesbian relationship is needed. The

results of this study are only tentative findings based on

cross-sectional data. More extensive quantitative and

qualitative data are necessary to rule out the influence of

mediating correlates which impair or foster leaving an abusive

gay and lesbian relationship.

Very few significant gender differences were established

in this study. The only significant association, after

controlling for gender, was among the intercorrelations of the

various violent tactic scales. Gay males were more likely to

report a strong interrelationship between physical

victimization, sexual victimization, and using violence. In

contrast, physical victimization and the use of violence were

highly correlated among lesbians; however, sexual aggression

was not as strongly correlated with the former conflict

scales. Future analyses on sexual aggression among lesbians

should focus careful attention on physical detainment, which

was excluded from this study. This finding may be the result
 153

of an emphasis placed on sexuality among gay males. Gay males

may utilize a sexual script which accentuates the role of

sexuality, linking it with other aspects of a relationship, at

least more so than lesbians. In short, violence appears to be

highly reciprocal in gay and lesbian relationships, given the

strong correlations between using violence and being a victim

of violence.

In conclusion, the violence in gay and lesbian

relationships mirror heterosexual relationships. Abuse is

reciprocal, and individuals who are abused maintained unstable

relationships. However, gays and lesbians are confronted with

a more divisive conflict, discrimination. The more

homosexuals exhibit their identity as a gay or lesbian, the

more likely they are to incur the scripted wrath of the

dominant, heterosexual culture. And with this discrimination,

gays and lesbians are more likely to endure elevated forms of

violence in their own relationships.

 154

BIBLIOGRAPHY

Abramson, L. Y., Garber, J., and Seligman, M. E. P. (1980).

Learned helplessness in humans: An Attributional
analysis. In J. Garber and M. E. P. Seligman (Eds.),
Human Helplessness: Theory and Applications (pp. 3-34).
New York: Academic Press.

Abramson, L. Y., Seligman, M. E. P., and Teasdale, J. (1978).

Learned helplessness in humans: Critique and
reformulation. Journal of Abnormal Psychology, 87, 49-
74.

Bersani, Carl A., and Chen, Huey-Tsyh. (1988). Sociological

perspectives in family violence. In V. B. Van Hasselt,
R. L. Morrison, A. S. Bellack, M. Hersen (Eds.), Handbook
of Family Violence (pp. 58-87). New York: Plenum Press.

Bologna, M. J., Waterman, C. K., and Dawson, L. J. (1987).

Violence in gay male and lesbian relationships:
Implications for practitioners and policy makers. Paper
presented at the Third National Conference for Family
Violence Researchers, Durham, NH, July.

Brand, P. A., and Kidd, A. H. (1986). Frequency of physical

aggression in ,heterosexual and female homosexual dyads.
Psychological Reports, 59, 1307-1313.

Carlson, B. E. (1977). Battered women and their assailants..

Social Work, 22, 455-461.

Duncan, David F. (1990). Prevalence of sexual assault

victimization among heterosexual and gay/lesbian
university students. Psychological Reports, 66, 65-66.

Gayford, J. (1975). Wife battering: A preliminary survey of

100 cases. British Medical Journal, 1, 195-197.

Gelles, R. J. (1976). Abused wives: Why do they stay?

Journal of Marriage and the Family, 38, 659-668.

Goldenstein, D., and Rosenbaum, A. (1985). An evaluation of

the self-esteem of maritally violent men. Family
Relations, 4, 425-428.

Kelly, E. E., and Warshafsky, L. (1987). Partner abuse in

gay male and lesbian couples. Paper presented at the
Third National Conference for Family Violence
Researchers, Durham, NH, July.
 155

Krestan, J., and Bepko, C. S. (1980). The problem of fusion

in the lesbian relationship. Family Process, 19, 277-
289.

Liddenbaum, J. P. (1985). The shattering of an illusion:

The problem of competition in lesbian relationships.
Feminist Studies, 11, 85-103.

McCandlish, B. M. (1982). Therapeutic issues with lesbian

couples. Journal of Homosexuality, 7, 71-78.

Mohr, Richard D. (1988). Gay Justice: A Study of Ethics,

Society, and Law. New York: Columbia University Press.

Peplau, L. A., Cochran, S., Rook, K., and Padesky, C. (1978).

Loving women: Attachment and autonomy in lesbian
relationships. Journal of Social Issues, 34, 84-100.

Renzetti, C. (1988). Violence in lesbian relationships: A

preliminary analysis of causal factors. Journal of
Interpersonal Violence, 3, 381-399.

Renzetti, C. (1989). Building a second closet; Third party

responses to victims of lesbian partner abuse. Family
Relations, 38, 157-163.

Rosenberg, M. (1965). Society and the Adolescent Self-Image.

Princeton: Princeton University Press.

Schilit, Rebecca, Gwat-Yong Lie, and Marilyn Montagne.

(1990). Substance use as a correlate of violence in
intimate lesbian relationships. Journal of
Homosexuality, 19, 51-65.

Straus, Murray, Celles, Richard, and Steinmetz, Suzanne.

(1980). Behind Closed Doors: Violence in the American
Family. New York: Doubleday.

Strube, Michael J. (1988). The decision to leave an abusive

relationship: Empirical evidence and theoretical.issues.
Psychological Bulletin, 104, 236-250.

Waldner-Haugrud, Lisa, and Magruder, Brian. (1991). Male and

female sexual victimization in dating relationships:
justification for the use of gender-neutral scales.
Presented at the Annual Midwest Sociological Society
Meeting, Des Moines, Iowa.

Walker, L. E. (1979). The Battered Woman. New York: Harper

and Row.
 156

Wolfgang, M. E. and Ferracuti, F. (1967). The Subculture of

Violence.. Beverly Hills, CA: Sage.

Wolfgang, M. E., and Ferracuti, F. (1982). The Subculture of

Violence (2nd ed.). Beverly Hills, CA: Sage.
 157

Model A:

Being "out" Physical

victimization
A
B E

Physical/Sexual Relational
Discrimination Instability
Self-Esteem

Model B:

Being "out" Use of

violence

Physical/Sexual Relational
Discrimination Instability
Self-Esteem

Model C:

Being "out" Physical

victimization

Physical/sexual Relational
Discrimination Instability
Self-Esteem

Figure 1. Proposed models illustrating the relationship

between disclosure, discrimination, physical
victimization, using violence, sexual
victimization, and relational stability among gay
males and lesbians
 158

Table 1. Comparison of means among all the variables in the

study by sex and self-esteem

Variables Hales Females

Low High Low High

Self-Esteem Self-Esteem Self-Esteem Self-Esteem

Being "Out*
mean 3.912 4.133 3.725 4.000
standard deviation 1.129 1.068 1.176 1.043

PieqrininatiM
mean .735 .651 .532 .483
standard deviation .745 .688 .718 .682

Physical Victimization
mean 1.088 .470^ 1.242 .948^
standard deviation 1.646 1.162 1.724 1.161

Relational Stability
mean 1.080 .817 .985^ 1.417
standard deviation 1.549 1.553 1.398 1.759

Dse of Violence
mean .853 .229® .807 .690^
standard deviation 1.417 .816 1.329 • 1.111

Relational Stability
mean 1.029 .312® .794b .900
standard deviation 1.464 .932 1.241 1.349

Sexual Victimization
mean 3.147 1.759® 2.097 1.633
standard deviation 3.311 2.662 3.217 2.490

Relational Stability , ^
mean 1.409 .918® .969 1.172
standard deviation 1.379 1.276 1.321 1.403

® significant difference at the .05 level between high and low

self-esteem groups among males and females.

^ significant difference at the .05 level between males and

females, controlling for self-esteem.
 159

Table 2. Correlation matrix for all the variables in the

study among lesbians, controlling for high and low
self-esteem [N; (low self-esteem=54, high self-
esteem=72)]

1. 2. 3. 4. 5. 6. 7. 8.

1. Being "Out" X .370* .252 -.278 .154 -.216 .150 .049

2. Discrimination .247 X .464* -.368* .350* -.248 .417* -.176

3. Physical victimization .000 .099 X -.751* .708* -.638* .365* -.262

4. Relational Stability .086 .060 -.794* X -.620* .850* -.359* .357*

5. Using Violence -.061 .016 .694* -.649* X -.702* .100 -.189*

6. Relational Stability .093 .037 -.654* .724* -.742* X -.224 .374*

7. Sexual Aggression -.162 .126 .327* -.340* .295 -.230 X -.605*

8. Relational Stability .067 -.306 • -.432* .323* -.308 .341* -.734* X

Note: correlations for lesbians with low self-esteem appear

above the diagonal.

* significant at the .05 level.

 160

Table 3. Correlation matrix for all the variables in the

study among gay males, controlling for high and low
self-esteem [N: (low self-esteem=75, high self-
esteem=92)]

1. 2. 3. 4. 5. 6. 7. 3.

1. Being "Out" X .246 .134 -.115 .163 -.221 .200 -.205

2. Discrimination .114 X .484* -.341* .324* -.225 .490* -.410*

3. Physical Victimization .067 .330* X -.710* .772* -.527* .518* -.315

4. Relational stability .046 -.237 -.675* X -.629* .776* -.424* .334*

5. Using Violence -.021 .318 .747* -.472* X -.684* .541* -.314

6. Relational Stability .085 -.180 -.581* .550* -.714* X -.383 .286

7. Sexual Aggression .080 .333 .502* -.382* .284 -.315 X -.656*

8. Relational Stability -.095 -.128 -.237 .318 -.097 .239 -.704* X

Note: correlations for gay males with low self-esteem appear

above the diagonal.

* significant at the .05 level.

 161

.109 (.239)
Model A:

Being "out" Physical

fl^.oi3(,o6i) victimization
I
.n4(.J46«) I .()75*( .7in"l «• •1561 SIM'
I .3.(0 •(.4S9')

Physical/sexual Relational
Discrimination instability
.*''(6)= 1.70 p=.945
GFI=.991
«^-.lOl(.lOJ)
Model B:

Being "out" use of

/
Violence «^ = .5ioi.)fi«>
.II4(.246»)X, / ., -.714•( .684») /

Y ,,--'''^318«(.324«) ./

Physical/sexual Relational
Discrimination instability
.*2(6) = 3.977 p =. 681
GFI=.991
R^-.in(.240 )
Model C:
/>

Being "out" ff^-:.oi3(.o6i ) Sexual

I Victimization .4,6,430,

•\ ^ .333'(.490»)
• \

Physical/sexual
Discrimination Relational
X- IT,) = 5.4.^/, =. 4')» Instability
GFI=.984

Note: standardized regression coefficients for gay males with

low self-esteem appear in parentheses.

* significant at the .05 level.

Figure 2. The impact of discriminatory stress on physical

victimization, use of violence, and sexual
victimization in gay male relationships,
controlling for high and low self-esteem
 162

Model A:
Being "out" , Physical
R :-.06t(.i37) Victimization
7114'( ,711") 0)1) I tW I
.24T(..)70*)

Physical/sexual Relational
Discrimination Instability
(6) = 5.99 p = . 424
GFI=.970
/î^ =.004(.123)

Model B: 1/
Being "out" «^=.06i(.i37) yge of
\ j Violence

Physical/sexual Relational
Discrimination Instability
X- (6) - 2.36 p - . 883
GFI=.994
«^=.016(.174)
Model C:
Being "out" sexual
Victimization
/'
/ ~ , ,734 •( .605'I
y// -<I26(.4I7«) \
:
Physical/sexual Relational
Discrimination Instability
.v^ (()) = 12.270 p = . ()5()
GFI=.934

Note: standardized regression coefficients for lesbians with

low self-esteem appear in parentheses.

* significant at the .05 level

Figure 3. The impact of discriminatory stress on physical

victimization, use of violence, and sexual
victimization in lesbian relationships, controlling
for high and low self-esteem
 163

GENERAL SUMMARY

Summary of Results

Each of the three articles in this dissertation has

focused on the sexual script perspective as applied to

homosexuality. The common theme in the three articles was the

examination of how cultural scenario scripts impact on the

interrelational and intrapsychic scripts for both gay males

and lesbians. Paper 1 analyzed the impact of identification

with heterosexual scripts, via the family, on the disclosure

of homosexual identities among gay and lesbian adolescents.

Paper 2 examined the diversity between cultural and

intrapsychic scripts among homosexuals and heterosexuals,

regarding the definition of attractiveness. Paper 3

investigated the negative influence of stigmatization

(discrimination) toward homosexuality, apparent in the

dominant cultural scenario script, on gay and lesbian

relationships. Each of the papers examined the role cultural

scenario scripts assert on homosexuality, whether it was

manifested in disclosure, the definition of attractiveness, or

in abusive relationships.

Disclosure of identity to the family

The model, proposed in Paper 1, asserted disclosure of

homosexual identities among gay and lesbian adolescents was

indirectly influenced by heterosexual scripts, manifested

through family identification. However, this process was

 164

mediated by the extent of perceived resources which were

supportive of homosexuality, and the expression of homosexual

identities among the adolescents in the study. The model was

based on the social exchange and the sexual script

perspectives. It was posited family identification was one of

the primary socializing agents for heterosexual sexual

scripts. Such an influence could be conceived as a cost when

disclosing a gay identity. This cost was weighed against the

degree of perceived supportive resources and the exploration

of the adolescent's own homosexual identity. In short, the

family was seen as a "push" toward a heterosexual life style,

whereas expression of homosexual identities and the extent of

perceived resources represented a "pull" toward a homosexual

life style. The ratio between the heterosexual scripted

influence and the homosexual scripted influence would affect

whether or not the adolescents disclosed their identity to

their parents.

The results of the study indicated identification with

the family was negatively related to the extent of perceived

resources, which were considered by the adolescent respondents

as supportive of homosexuality. Family identification was

negatively related to the expression of homosexual identities

among both gay and lesbian adolescents. However, family

identification was not significantly related to disclosing the

identity to the parents of the respondents. The extent of

perceived homosexual-supportive resources was positively

 165

associated with the expression or salience of homosexual

identities. The extent of perceived resources was also

positively associated with disclosure of gay and lesbian

identities to parents. Finally, the expression of homosexual

identities, among the gay and lesbian adolescents, was

positively related to disclosure. Few significant gender

differences were found in the general model. The one

contrast, between gay males and lesbians, was the extent of

common variance explained by the significant relationships in

the model. Identification with family, the extent of.

perceived resources, and expression of homosexual identity

explained twice the common variance in disclosure for gay male

adolescents, when compared to lesbians.

The merging of the sexual script perspective with social

exchange theory explained the process by which homosexual

adolescents disclosed their identities to their parents. In

short, identification with the sexual scripts fostered by

parents and families was not significantly linked to whether

the adolescents disclosed their identities to their parents.

However, identification with heterosexual scripts, as

socialized via the family, did significantly detract from

perceiving homosexual-supportive resources and expressing the

adolescents' identities through organization participation,

sexual contact, and friendship networks. In turn, the more

the adolescents secured and perceived various resources,

supportive of homosexuality, the higher the benefit of

 166

disclosing their identity. Similarly, the more the

adolescents investigated homosexual scripts, by perceiving

extensive resources supporting homosexuality, the lower they

perceived the costs of disclosing their identities to their

parents. Social support resources facilitated the disclosure

process by socializing gay and lesbian adolescents with

appropriate homosexual sexual scripts.

By educating and resocializing gay and lesbian

adolescents with sexual scripts, favorable toward

homosexuality, provided the adolescents with the ability to

envision potential benefits of adopting their homosexual

identities. Whereas, identification with the family, dictated

the inherent costs of breaking the normative guidelines of

heterosexuality. Therefore, expression of homosexual

identities, for adolescents, may be conceptualized as the

theoretical weight associated with the costs of dissolving the

bonds of heterosexual scripts, advocated by the family, versus

the perceived benefits facilitated by homosexual-oriented

organizations and resources. The two competing forces are

best expressed by the regression coefficients between

identification with the family and the extent of perceived

supportive resources on expression of homosexual identities.

The "push/pull" mechanisms, affecting the expression of

homosexual identities among gay and lesbian adolescents,

reflects the innate struggle between heterosexual and

homosexual sexual scripts. Again, identification with the

 167

family "pushes" the adolescent away from expressing their

homosexual identities, via the sanctions toward homosexuality

as indoctrinated by heterosexual sexual scripts. While

homosexual-oriented resources "pull" the adolescent toward

expressing their homosexual identities by resocializing the

adolescent with homosexual sexual scripts.

Definitions of attractiveness and sensuality

The second article in this project focused on the

differences in definitions of attractiveness among homosexuals

and heterosexuals. The general assertion posited by the

study, contended the definition of attractiveness was based on

shared cultural scenario scripts. It was asserted males and

females maintain different cultural scenario scripts which

guide the definition of attractiveness for the appropriate

sex. For this study, it was assumed cultural scenario scripts

comprise the unwritten rules and guidelines in society which

dictate the normative standards associated with

attractiveness. The study posited heterosexual females and

gay males would share similar cultural scenario scripts

regarding the definition of attractiveness for males.

Similarly, lesbians and heterosexual males would share

corresponding cultural scenario scripts guiding the definition

of attractiveness for women.

Further, the study asserted several conceptual and

methodological considerations for studying the definition of

 168

attractiveness. First, researchers on attractiveness should

consider the scripted difference between sensuality and

attractiveness. The study posited the definition of

attractiveness is standardized, based on the dominant cultural

scenario script. Whereas, the composite definition of

sensuality is individual-based, guided by an individual's

intrapsychic scripts (fantasies, desires, motivations,

arousal). Second, the study contended the definition of

attractiveness should not be limited to only physical

qualities. It was posited the definition of both

attractiveness and sensuality would be partially constructed

with personality characteristics, as well as physical

attributes.

In general, the study found no significant differences

between the definitions of attractiveness among gay males and

heterosexual females. This finding indicates gay males and

heterosexual females share a similar definition of

attractiveness, and therefore similar cultural scenario

scripts regarding the definition of attractiveness for men.

However, significant differences were discovered when

contrasting the definitions of sensuality among homosexual

males and heterosexual females. This finding suggests that

though gay males and heterosexual females share similar

cultural scenario scripts, they significantly differ regarding

the definition of sensuality, or the composition of their

respective intrapsychic scripts.

 169

In the comparison of the definition of attractiveness

among heterosexual males and lesbians, a significant gender

difference was established. Also, significant gender

differences were found in definitions of sensuality among

heterosexual males and lesbians. Both findings indicate

lesbians and heterosexual males share different cultural

scenario scripts and intrapsychic scripts for both the

definitions of attractiveness and sensuality. Lesbians tended

to focus more on personality constructs regarding

attractiveness. Heterosexual males, on the other hand,

concentrated on more physical attributes, such as the face

and, in particular, the body.

It could be inferred from these results that heterosexual

males and lesbians possess different cultural scenario scripts

at the subcultural level. It appears the lesbian subculture

prizes personality qualities, such as warmth and affection,

over physical attributes, such as the face and body.

Consequently, the cultural scenario script guiding the

definition of attractiveness among heterosexual males targets

physical characteristics over personality attributes. Perhaps

this difference in scripts stems from a gender contrast

between lesbians and heterosexual males. Based on traditional

notions of gender scripts, women should view warm and

affectionate women as being attractive (Dew, 1985). Given

cultural scenario scripts define women who are sensitive and

warm as attractive, it is not surprising that lesbians would

 170

place preference on those women possessing such attributes.

However, this distinction in scripting appears to be gender

biased. Heterosexual males maintain a cultural script which

emphasizes physical characteristics above personality

attributes. Again, this is not necessarily surprising. A

further examination of the results indicate sensuality and

attractiveness scripts for heterosexual males tend to overlap.

There is much similarity in the definitions of attractiveness

and sensuality among heterosexual males. This finding

suggests heterosexual males do not separate the constructs of

attractiveness and sensuality. To this grouping, the two

constructs are synonymous.

The general conclusion of the study suggests heterosexual

males, heterosexual females, and homosexual males place a

strong emphasis on physical attributes in defining

attractiveness. However, lesbians appear to prioritize

personality characteristics over physical elements when

ordering attractive qualities. Further, heterosexual females,

homosexual males, and lesbians lean toward personality

qualities in defining sensual features. However, heterosexual

males tend toward physical attributes in defining sensuality.

The broader message of the scripted perspective, when given

these general trends, suggest that what is beautiful is not

necessarily in the eye of the beholder. In defining what

beautiful is, one must first differentiate between sensuality

and attractiveness. What is attractive appears to be in the

 171

eye of the culture, whereas what is sensual appears to be in

the eye of the individual.

The violent effects of discrimination

The three models, proposed in Paper 3, asserted gays and

lesbians, who considered themselves as being "out", were more

likely to incur both physical and sexual discrimination by

heterosexuals. The extent of discrimination incurred by gays

and lesbians would affect the degree of violence within their

relationships. The three models examined in the study,

focused on three forms of relational violence; physical

victimization, perpetration or use of violence, and sexual

aggression. It was predicted the more violent the

relationship, the more likely the relationship would be

unstable. Finally, it was postulated the effects of

discrimination on abuse would be moderated by the self-esteem

of the gays and lesbians in the study.

The findings of the study indicated gays and lesbians,

who described themselves as being "out", were more likely to

incur discrimination. However, this finding was not supported

for gay males with high self-esteem. Discrimination was

positively related to all three forms of abuse for gay males,

regardless of self-esteem. Discrimination was significantly

and positively associated with physical victimization, the use

of violence, and sexual aggression only for lesbians with low

self-esteem. In comparing stacked structural equation models

 172

between lesbians with high and low self-esteem, the data

yielded a significant change in the chi square products when

the path from discrimination to physical victimization and the

use of violence was set to be equal for the two groups. For

the model concerning physical victimization, the significance

level was less than .05; and, the significance level, for the

model concerning the use of violence, was less than .10. A

comparison between the stacked structural equation models for

lesbians with high and low self-esteem, regarding the path

from discrimination to sexual aggression, did not yield any

significant changes in the chi square product. Finally, for

both gay males and lesbians, regardless of self-esteem,

violence in the relationship made the relationship unstable.

Many of the gays and lesbians, who incurred violence in their

relationship (whether it was physical victimization,

perpetration of violence, or sexual aggression) eventually

terminated the relationship after the violent episode.

The theoretical basis for the three models in the study

was structured around a sexual script perspective. Given

cultural scenario scripts assert negative sanctions associated

with homosexual behaviors and life styles, the social

acceptance of sanctioning homosexuality is manifested and

perpetuated by cultural scenario scripts. Cultural scenario

scripts mandate homosexuality is an inappropriate sexual

behavior. Discrimination toward homosexuals appears to be the

end result of this general negativity toward homosexuality in

 173

heterosexual scripts. The abuse of homosexuals has reached an

extended level in American society, based on such cultural

scenario scripts. It has become socially accepted to

physically and sexually harass gays and lesbians, based solely

on their sexual orientation. In fact, it may be asserted such

discrimination is actually encouraged by heterosexual sexual

scripts, in defining the appropriate sexual behavioral

standards for members in society. The data Indicated the more

socially "out" homosexuals perceived themselves to be, the

more discriminatory episodes they encountered by

heterosexuals.

The effect of discrimination increases the probability

gays and lesbians would incur some form of violence within

their relationships, whether physical victimization,

perpetrating violent behaviors, or being a victim of sexual

aggression. The effect of the cultural scenarios scripts,

which foster discrimination toward homosexuals, appears to

facilitate violence in gay and lesbians relationships.

Therefore, the discrimination directed at gays and lesbians

becomes filtered into the interrelational scripts among gays

and lesbians. This process of violent "filtration" among gays

and lesbians may be a result of several factors, including:

stress amplification, frustration-aggression, redefining

victimization, learned helplessness, etc.. Both stress

amplification and frustration aggression hypotheses assert

gays and lesbians who incur discrimination tend channel or

 .174

displace the stress and frustration, associated with the

discriminatory acts, to the perpetration of violence in their

relationships. Or, the stress and frustration, associated

with the discrimination, fosters alienation and helplessness

among gay and lesbian victims. The stress of discrimination,

compiled with the lack of supportive resources designed for

gays and lesbians, could become manifested in further forms of

aggression under the context of relational violence. Cultural

scenario scripts, which facilitate discrimination toward

homosexuals, simultaneously defines gays and lesbians as

legitimate victims of cultural abuse. The effect of such

cultural scenario scripts ultimately impacts on the

interrelational scripts of gays and lesbians in defining

themselves as legitimate victims. Gays and lesbians who

utilize violence within their relationships, definitively

support violence toward gays and lesbians, and become

legitimated by the broader sexual scripts which foster such

violence based on sexual orientation. In turn, those gays and

lesbians who are victims of physical or sexual abuse, define

themselves as deserving victims if they have incorporated the

values embedded in cultural scenario scripts into their

interrelational scripts. Further, gays and lesbians who are

abused, may have learned they are helpless victims under the

broader sexual schema as appropriated by cultural scenario

scripts.
 175

It was, perhaps, encouraging to note the data indicated

that many of the abusive relationships were unstable

relationships. The frequency results suggested many of the

participants in abusive relationships terminated the

relationship after the violent incident. The cultural

scenario scripts, which define abused gays and lesbians as

legitimate victims, is a transient aspect among homosexual

interrelational scripts. Despite the lack of social services,

providing assistance to abused gays and lesbians, victimized

gays and lesbians still maintained enough social resources to

abandon abusive relationships. It is posited the

interrelational scripts among gays and lesbians, in general

cases, does not condone long-term, abusive relationships. It

could also be the same cultural scenario scripts which remit

discrimination toward gays and lesbians, also perceive gay and

lesbian relationships as illegitimate associations. And such

illegitimacy, attached to gay and lesbian relationships, may

predispose gay and lesbians to maintaining relatively unstable

relationships. Therefore, leaving an abusive gay and lesbian

relationship may not pose as many social sanctions, when

compared to abusive heterosexual relationships. The "closed"

nature attached to homosexual relationships may actually

reduce the social disapproval mechanisms, which are generally

associated with divorce and separation incurred in

heterosexual relationships.
 176

The only gender differences reported in this study tended

to be correlated with self-esteem. It was found among gay

males with high self-esteem, that being "out" was not

significantly related to incurring physical and sexual

discrimination. However, this finding was significant for

lesbians, regardless of self-esteem, and among gay males with

low self-esteem. The intrapsychic scripts, guiding both

lesbians and gay males, reflects the internal mechanisms

linked to a homosexual identity. Such internal mechanisms,

underlying intrapsychic scripts, includes concepts like self-

esteem. It is posited that gay males with low self-esteem do

not possess the identity salience to sufficiently cope with

being "out" and the occurrence of discrimination. Perhaps gay

males with low self-esteem are more likely to attribute

discrimination to the extent they flaunt or disclose their

homosexual identity. Whereas, gay males with high self-esteem

maintain confidence about their identity and intrapsychic

scripts. Therefore, gay males with high self-esteem do not

readily associate any discriminatory acts they may incur with

their homosexual life style.

The second gender difference found in this study focused

on the impact of discrimination on sexual aggression, physical

victimization, and the use of violence. Among gay males, and

lesbians with low self-esteem, the effect of discrimination on

relational violence was significant. However, among lesbians

with high self-esteem, the relationships between the three

 177

forms of violence and discrimination were found to be

nonsignificant. Lesbians with high self-esteem appear to be

able to diffuse the negative effects of discrimination on

relational violence. Given lesbians have interrelational

scripts which tend to be built on friendship networks and

intimacy, this may account for the apparent gender difference

concluded in this study. If lesbians with high self-esteem

built their relationships on the basis of intimacy and

friendship, perhaps their interrelational scripts are more

forgiving of discriminatory violence, and capable of

nonviolently coping with discrimination.

Integrating a sexual script perspective

The three studies presented in this dissertation have

investigated the effect of cultural and structural

arrangements as they have affected the perspective of

homosexual sexual scripts. The general attempt by each of the

projects has been to analyze the various means by which

structural or cultural arrangements have influenced the

development of homosexual sexual scripts. The general

postulate governed by the homosexual sexual script perspective

asserts homosexual relationships are sanctioned by the broader

cultural values and norms.

The sexual script perspective defines homosexuality in

terms of public labeling and the attendant roles and scripts

associated with this label. The criticism attached to the

 178

sexual script perspective contends the approach does not allow

for the development of sexual identities or the influence of

physiological conditions. Where the sexual script perspective

views homosexuality as loosely defined behavior patterns based

on interactions with others, it tends to neglect the power of

cultural arrangements which influence identity construction

and development. The first article in this dissertation

attempted to merge the two approaches in the homosexuality

literature: identity development and sexual scripts. The

model proposed in the first article asserted homosexual

identities are facilitated or detracted from developing into

an integrated identity (disclosing the homosexual identity)

based on the stigma associated with the label of homosexuality

(via family identification). The staunch theoretical position

asserted by Sagarin (1975), contended homosexuality is simply

comprised of a set of roles pertaining to a series of scripted

situational contexts, was not necessarily supported by the

article. Rather, the identity/script perspective, advocated

by Troiden (1979, 1984/1985), was sustained by the findings in

the study. In general, the results of the project indicated

structural forces, such as the family, affect the expression

and disclosure of homosexual identities among adolescents.

The heterosexual scripts, promoted by the family, detracted

from the salience of a homosexual identity.

The second criticism attached to the sexual script

perspective is directed at the ambiguity of scripted

 179

behaviors. Sexual scripts represent general guidelines,

interactions, values, and norms; however, the detail of the

scripting process tends to be quite innocuous. The aim of the

second article was to detail the characteristics of one

scripted dimension, attractiveness. Analyzing the open-ended

responses of the participants, revealed a normative script

which denoted general characteristics composing the definition

of attractiveness. Specific personality, facial, and bodily

characteristics were emphasized by the subjects. These

representations comprised a mosaic of cultural expressions

which delineate the cultural dimensions and norms defining

attractiveness.

The separation of attractiveness and sensuality, as two

distinct definitive constructs, further detailed the

specificity associated with sexual scripts. The findings in

the study indicated the composite qualities associated with

attractiveness were fairly similar among homosexual males and

heterosexual females. However, the two groups significantly

differed in the characteristics which define sensuality among

men. The definition of sensuality is perhaps centered around

intrapsychic scripts, suggesting what is sexy is guided by

individual fantasies and desires. Among both sample groups,

homosexuals and heterosexuals, responses varied significantly

between compatible groups. Homosexual males and heterosexual

females differed significantly in their responses with regard

to defining male sensuality. And heterosexual males and

 180

lesbian differed significantly when the characteristics of

body, face and personality were contrasted in the definition

of female sensuality. Therefore, the sexual scripts guiding

the definition of sensuality is either subculturally specific,

or individual-oriented.

It is the viewpoint of this dissertation that the sexual

scripts, defining sensuality are individual-oriented, rather

than subcultural specific. After reviewing the qualitative

responses of each sample group, it appeared individuals

maintained unique expressions of sensuality, regardless of

subcultural affiliation. For example, some homosexual males

asserted bald men were sexy, and some heterosexual males

stated women with small breasts were sensual. Although such

responses were not normative, they represented a faction of

responses which were unique. It is posited individuals

construct intrapsychic scripts which define sensuality on an

individual level. This construction could be due to

socialization patterns or symbolic attachment to various media

representations. Whatever the origin of the intrapsychic

script, the subjects' responses reflected individual qualities

which delineated certain responses from a "normative" picture

of attractiveness. In short, by examining the individual

responses to the open-ended questions, it appeared the

definition of attractiveness bared striking similarities

across responses; consequently, an inspection of responses to

sensuality yielded dissimilar characteristics.

 181

A third criticism of the sexual script perspective is the

narrow field in which the perspective is utilized. The

perspective tends to be offered to explain patterns in sexual

interactions, but is neglected when offering commentary about

other relational aspects over and above sexuality. The third

paper in this dissertation attempted to generalize the sexual

script perspective to other relational dynamics other than

sexuality. The third article focused on the effects of

discrimination on relational violence. Physical

discrimination, based on sexual orientation, was found to

increase the probability of being a victim of violence in gay

and lesbian relationships, and increase the likelihood of

abusing. Cultural scenario scripts define homosexuality as a

deviant status in the United States, which subjugates

homosexuals to discrimination. The stress associated with

physical harassment eventually becomes a part of the dynamics

of gay and lesbian relationships, and can be correlated with

being a victim or using violence. Further, those who

maintained some aspect of physical violence in their

relationships were more likely to have unstable relationships.

The sexual script perspective can assist other theories

in explaining not only sensuality, but interrelational

dynamics and exchanges. The sexual script perspective can

offer a certain understanding about the way culture, social

structure, and individuals interact with one another.

Scripts, as a theory can be used to explain the general

 182

guidelines and norms encompassing relational interactions, or

by extrapolating the perspective to other relational

phenomenon, above sexuality. Further, sexual scripts can be

utilized to approach specific issues in intimate

relationships, for example attractiveness. By understanding

the interdynamic levels of sexual scripting, the various

levels of analysis can be centered to explain the relationship

between the individual and culture, or social structure.

Limitations and Future Research

It is hoped each of the three studies have contributed to

the understanding of homosexual sexual scripts. By noting how

sexual scripts affect the development of homosexual

identities, norms, and relationships, each article has

furthered the research knowledge on homosexuality. However,

much remains to be investigated, especially concerning

homosexual relationships. Prior to discussing potential

avenues for future research, it is necessary to address the

criticisms and limitations of each of the three studies in

this dissertation.

Limitations

Family disclosure. In general, there are three major

limitations associated with the article on the disclosure of

identities to the family among gay and lesbian adolescents.

First, the representativeness of the sample should be

 183

evaluated. Given the data for the study emerged from a

snowball sampling technique, the sample does not necessarily

convey the extent of representativeness as associated with a

random sample. However, the limitations posed by the snowball

sample must be realistically addressed. Gay and lesbian

adolescents comprise a unique and highly stigmatized

population. To obtain a random sample of gay and lesbian

adolescents verges on the improbable.

The second limitation with the article stemmed from the

measurement of several key concepts. Single indicators were

used to measure the concepts of family identification, the

extent of perceived homosexual-supportive resources, and

disclosure of their identities to their family. The use of

single indicators posed a threat to both the reliability and

validity of the tested model. The utilization of a single

indicator for family identification may have "tapped" into

variant dimensions of the general construct, such as

satisfaction with family members, and family value

association. Further, the use of a single indicator for the

variable of family identification may be confounded with

potentially spurious or intervening variables, such as

parenting style, religious values, peer associations, and

sexual values. The same problem exists with the extent of

perceived resources measure. Again, the construct may be

drawing upon the dimensions or spurious correlates of

community size, the actual versus perceived resources, and the

 184

utilization of such supportive résources. The disclosure

variables may have targeted single dimensions of disclosure,

or the intervening correlates of disclosure. For example, the

disclosure variable is not directly ascertaining whether or

not the adolescents disclosed their sexual identity to their

parents, rather the variable attempted to estimate whether or

not the parents knew of the adolescent's sexual identity.

The third limitation of the disclosure study focuses on

the causal ordering assumed by the hypothesized model. The

model asserted family identification detracted from the extent

of perceived resources, and the expression of the adolescents'

homosexual identity. It was also posited the extent of

perceived resources and the expression of a homosexual

identity would facilitate the disclosure process among gay and

lesbian adolescents. Given the study was conducted with

cross-sectional data, there is no way to validate the extent

identification with the family was a product of the disclosure

process or directed the disclosure process. Although the

literature does advocate that parental attitudes and values

precipitate the acquisition of homosexual identities.

Definition of attractiveness. The major limitation

posed by the attractiveness study consists of comparable

samples. In short, the two major samples of heterosexuals and

homosexuals were not directly analogous. The homosexual

sample was older, maintained a higher social economic status,

and were more diverse in regional composition. This

 185

limitation may cause problems in asserting that unique sexual

scripts exist for homosexual and heterosexual cultures, in

particular for scripts dealing with sensuality. For example,

the difference between lesbians and heterosexual males in the

definition of attractiveness could be the result of older

lesbians.

A second limitation posed by the project was in the

assessment of attractiveness and sensuality. General

characteristics were utilized, versus specific qualities.

Although respondents indicated they found, for example,

muscularity was attractive, the respondents did not indicate

how important muscularity was when compared to other

characteristics. The elusive nature of the measures did not

ascertain the extent to which respondents favored various

characteristics over others.

The third limitation of the study is centered on the

interpretation of the results. The tests of significance in

the study should be interpreted very carefully. In order to

conduct the test of significance the categories were collapsed

into three major sections: face, body, and personality. The

tests offer limited information regarding the general priority

of general attributes over others, but failed to provide

specific information regarding particular characteristics.

The violent effects of discrimination. Although one of

the first studies to quantitatively examine gay and lesbian

violence, the study suffers from several limitations. First,

 186

the measures in the study could be improved upon. The

discrimination variable was measured by using two dichotomous

items, with no indication of time or place of occurrence. The

study cannot delineate the causal influence of discrimination

on violence due to the cross-sectional data. For example, it

could be possible those who used violence in their

relationships, or were victims of violence (physical or

sexual) established themselves as potential targets of

discrimination after the violent episode.

Further, homosexual identity was assessed by the degree

the subjects were "out" regarding their life style. The

measure did not address the extent the respondents were

satisfied or happy with their sexual identity. Although the

association between being committed to a homosexual identity

and discrimination may appear to be trivial. The relationship

assumes those who were committed to a homosexual identity were

more likely to be discriminated against, and thus incur some

form of violence in their relationships. It may be more

appropriate to assess not just the extent the respondents were

"out of the closet", but also the degree they were comfortable

with their sexual identity.

The second criticism of the .study is with the self-esteem

variable. By dichotomizing the variable at the median reduced

the variance associated with self-esteem. In other words,

those with high self-esteem and moderately high self-esteem

were grouped together. Using self-esteem as a dichotomous

 187

grouping was selected for the ease of modeling the mediating

effects of the variable. It is possible to expand on the

results of the study by treating self-esteem as a continuous

interaction variable, and utilizing the entire variance

associated with self-esteem.

Future research
Future research in the area of homosexual sexual scripts

is encouraged. Little is empirically known about homosexual

life styles and relationships. Any further attempt to clarify

the scripts in homosexual relationships would contribute to

the sociological literature on both sexual scripts and

homosexuality. Not only is general research encouraged on

homosexual sexual scripts, but each article has particular

areas of concern which should be addressed by further

investigations.

Family disclosure. Much research on homosexual

identity and disclosure have been conducted, yet little of the

research has been done on adolescents. It is hoped more

research on homosexual identity acquisition will be devoted

toward adolescent samples, instead of adult retrospective

studies. Future studies, utilizing gay and lesbian

adolescents, should focus on the processes and resources used

in the development of a homosexual identity. As the study

demonstrated, perceived resources and support networks

promoted a sense of security toward the disclosure process.

 188

Further, future research should incorporate stronger more

diverse measures of support and resource utilization to build

a stronger and more reliable index of social support. In

addition, it is hoped researchers will elaborate on the family

relationship/identification variable. Homosexuality

investigators should consider such variables as parent/child

communication patterns, and supportiveness.

Ideally, a longitudinal or panel design would be the most

extensive means to document the actual processes adolescents

go through in the development and disclosure of a homosexual

identity. A qualitative study could facilitate further

understanding of homosexual identity formation. In addition

to identifying the various resources used in the development

process, a qualitative study could address the particular

concerns and identity struggles among gay and lesbian

adolescents.

Second, it is hoped future research will focus on

contextual variables which would moderate the proposed

relationships in the model. Some variables which may be

considered would include: socio-economic status, peer group

affiliation, gender identity, the role of siblings, etc. The

model would also benefit by contrasting urban from rural

adolescents. Third, it is hoped the model could be

generalized from adolescents to adults who pursue a homosexual

identity or disclosure in adulthood. By targeting other

 189

factors which could affect the disclosure process, should shed

insight on the subject of homosexual identity formation.

Definition of attractiveness. Although quite a bit of

research has been conducted on physical attractiveness, it is

hoped future research will investigate the dimensions of

attractiveness beyond physical characteristics. One of the

contributions of the study, to the conceptual definition of

attractiveness, was the focus on personality features. Future

research on attractiveness should further explore the

personality dimension in relation to the physical dimension.

It is also hoped attractiveness researchers could

differentiate between concepts like sensuality and

attractiveness. The study found significant differences in

the definitions of attractiveness and sensuality.

Second, it is hoped research on the differences between

homosexuals and heterosexuals will be further addressed. For

example, it would be interesting to decompose various

subcultural influences in developing respective definitions of

attractiveness and sensuality. Or to test the interaction

effect of gender identity in delineating the definitions of

attractiveness and sensuality. Finally, subcultural

influences, such socio-economic status or friendship

affiliation, would be an intriguing correlate to definitions

of attractiveness or sensuality.

The violent effects of discrimination. As one of the

first articles to quantify physical and sexual violence in gay

 190

and lesbian relationships, it is hoped much research will stem

from this article. First, a qualitative analysis is

recommended on the subject. It is important to learn why and

how harassment leads to using violence in gay and lesbian

relationships, and being a victim of physical and sexual

aggression. A qualitative analysis may improve on the

theoretical development of the article in providing a

contextual framework in which the association between

discrimination and violence is exacerbated.

Second, another study is recommended to validate the

findings of this study. It is advisable that a follow-up

study consist of a longitudinal or panel design in which the

causal relationships between the various forms of violence and

discrimination could be further documented. Further, better

measures of the critical variables in the study should be

considered. For example, a follow-up study should expand on

the discrimination measure by distinguishing when and how the

act of harassment had occurred. It is also recommended other

forms of discrimination should be investigated, such as

employment and housing crises. Further research on the

effects of discrimination is encouraged. It would be

interesting to note the effects of discrimination on health or

relationship quality, for example.

Homosexuality researchers need to broaden their general

concept of homosexual identity. In the past, many researchers

have primarily focused on the psychological aspects of a

 191

homosexual identity. In general, it is hoped research trends

will move toward a structural perspective which can

incorporate structural influences with psychological factors.

It is also hoped that homosexual identity researchers will

build on the presented studies by elaborating on the

conceptual ideas and models.

Conclusion
The purpose of this dissertation was to apply the sexual

script perspective to various dimensions of homosexuality.

The second objective of the dissertation was to show the

conflict between heterosexual and homosexual sexual scripts.

In doing so, the dissertation attempted to map homosexual

sexual scripts, and heterosexual scripts dealing with

homosexual issues. Three studies were conducted to examine

various aspects of homosexual sexual scripts and the influence

of heterosexual scripts on homosexual themes. The first paper

addressed homosexual identities, and the problems faced by

adolescents in disclosing their identities to their parents.

The project found adolescents were caught in an exchange

struggle between the "pushes and pulls" of cultural and

subcultural influences. In other words, adolescents must

weigh the costs and benefits of forfeiting heterosexual sexual

scripts for the possible benefits of acquiring homosexual

sexual scripts.
 192

The second paper targeted the dimensions of interpersonal

dynamics among heterosexual and homosexuals. The definition

of attractiveness and sensuality were contrasted among

heterosexuals and homosexuals. The study concluded

homosexuals and heterosexuals overlap in their definition of

attractiveness, with the exception of the lesbian sample;

however, homosexuals and heterosexuals were dissimilar in

their definition of sensuality. Once engaging a homosexual

script, gays and lesbians (in particular) were more likely to

view definitions of sensuality apart from heterosexual

influences.

The third study focused on violence in gay and lesbian

relationships. The more gays and lesbians were committed to a

homosexual script (being "out", so to speak), the more likely

gays and lesbians were physically harassed and discriminated

against. Further, the more physical discrimination gays and

lesbians incurred, the more likely they were to be victims of

sexual aggression or violence in their relationships. And

discriminated gays and lesbians were more likely to use

violence within their relationships. The results in the study

suggested gays and lesbians, though articulating a unique

sexual script, were not beyond the sanctions and influences of

heterosexual scripts. In this case, the influence of

heterosexual scripts (via discrimination) appears to have

decisively negative consequences for gay and lesbian

relationships.
 193

In conclusion, homosexual sexual scripts cannot be

examined as an entity by themselves. Homosexual sexual

scripts operate under the influence of the dominant

heterosexual culture, and therefore, are subjugated to

heterosexual influences. In the first and third article, both

studies illustrated the influence heterosexual scripts have on

the development of homosexual identities, and the continuance

of homosexual relational scripts. In short, it appears

heterosexual scripts can detract from the development of

homosexual scripts and identities, and serve as potential

barriers or obstacles after an individual embraces a

homosexual identity and the scripts which accommodate such an

identity.

Homosexual sexual scripts, in some ways, are similar to

heterosexual scripts. The second article demonstrated

similarities in the definition of male attractiveness among

heterosexual females and homosexual males. However, in other

ways, homosexual sexual scripts are different from

heterosexual sexual scripts. For example, in the same

article, heterosexual males and lesbian possessed different

attitudes about what characteristics constitute female

attractiveness.

The sexual script perspective tends to be viewed as an

incomplete theory. The three studies in this dissertation

have attempted to fill in some of the gaps in the theory, to

present a broader application of the perspective. By merging

 194

the sexual script perspective with other existing theories and

hypotheses, facilitated a broader and perhaps more complete

understanding of intimate relationships.

 195

BIBLIOGRAPHY

Adam, Barry D.
1978. The Survival of Domination. New York: Elsevier.
1985. "Age, structure, and sexuality: reflections on
the anthropological evidence on homosexual
relations." Journal of Homosexuality^ 11(3/4); 19-
33.

Anderson, R. C. and J. Pichert

1978. "Recall of previously unrecallable information
following a shift in perspective." Journal of
Verbal Learning and Verbal Behavior, 17: 1-12.

Athanasiou, R., P. Shaver, and C. Travis

1970. "Sex." Psychology Today, July; 50.

Bartlett, F. C.
1932. Remembering: A Study in Experimental and Social
Psychology. Cambridge; Cambridge University Press.

Blackwood, Evelyn
1985. "Breaking the mirror; the construction of
lesbianism and the anthropological discourse on
homosexuality." Journal of Homosexuality, 11(3/4):
1-17.

Burgess, E.W. and P. Wallin

1953. Engagement and Marriage. Chicago; Lippincott.

Cass, Viviene c.
1979. "Homosexual identity formation: a theoretical
model." Journal of Homosexuality, 4(3); 219-35.
1983. "Homosexual identity: a concept in need of
definition." Journal of Homosexuality, 9(2/3): 105-
26.

Coleman, Eli
1982. "Developmental stages of the coming-out process:"
In Homosexuality: Social, Psychological, and
Biological Issues, edited by William Paul, James D.
Weinrich, John C. Gonsiorek, and-Mary E. Hotvedt,
149-58. Beverly Hills: Sage.

Corbett, Sherry, Richard Troiden, and Richard Dodder

1974. "Tolerance as a correlate of experience with
stigma: the case of the homosexual." Journal of
Homosexuality, 3(1): 3-13.
 196

Cuber, J.F. and P.B. Harroff

1965. The Significant Americans. New York: Appleton-
Century.

Davenport, W.
1965. "Sexual patterns and their regulation in a society
of the Southwest Pacific." In Sex and Behavior.
New York: Wiley.

Dew, Mary A.
1985. "The effects of attitudes on inferences of
homosexuality and perceived physical attractiveness
in women." Sex Roles, 12(1/2); 143-155.

Gagnon, John
1977. Human Sexualities. Glenview, IL: Scott,
Foresman.

Gagnon, John H. and William Simon

1973. Sexual Conduct: The Social Forces of Human
Sexuality. Chicago; Aldine.

Goode, Erich
1973. The Drug Phenomenon: Social Aspects of Drug
Taking. Indianapolis: Bobbs-Merrill.

Hayes, T. and F. Oziel

1976. "Capturing the invisible population." Journal of
Homosexuality, 10(1): 125-37.

Hunt, Morton
1974. Sexual Behavior in the 1970s. Chicago; Playboy
Press.

Kinsey, Alfred, Wardell Pomeroy, and Clyde Martin'

1948. Sexual Behavior in the Human Male. Philadelphia:
Saunders.

Laws, Judith L. and Pepper Schwartz

1977. Sexual Scripts: The Social Construction of Female
Sexuality. Hinsdale, IL: Dryden Press.

Lee, John
1977. "Going public: a study in the sociology of
homosexual liberation." Journal of Homosexuality,
3(1): 49-78.

Levine, Martin P.
1987. "Gay Macho: Ethnography of the Homosexual Clone."
Doctoral dissertation. New York University.
 197

Masters, W. and V. Johnson

1979. Homosexuality in Perspective. Boston; Little
Brown.

Money, J.
1980. Love and Love Sickness: The Science of Sex,
Gender Difference, and Pair-Bonding. Baltimore:
John Hopkins University Press.

Plummer, Kenneth
1975. Sexual Stigma: An Interactionist Account.
London: Routledge and Kegan Paul.

Ponse, Barbara
1978. Identities in the Lesbian World: The Social
Construction of Self. Westport, CT; Greenwood
Press.

Rainwater, Lee
1970. Behind Ghetto Walls: Black Families in a Federal
Slum. Chicago: Aldine.

Sagarin, Edward
1973. "The good guys, the bad guys, and the gay guys."
Contemporary Sociology, 2(1): 3-13.
1975. "The tyranny of 'isness." In Deviants and
Deviance: An Introduction to the Study of Disvalued
People and Behavior, 144-54. New York: Praeger-
Holt.
1976. "Thieves, homosexuals, and other deviants: the
high personal cost of wearing a label." Psychology
Today, 9(10): 25-31.
1979. "Deviance without deviants: the temporal quality
of patterned behavior." Deviant Behavior, 1(1): 1-
13.

Schank, R. C.
1982. Dynamic Memory: A Theory of Reminding and
Learning in Computers and People. New York:
Cambridge University Press.

Schwartz, Barry and Daniel Reisberg

1991. Learning and Memory. New York: W. W. Norton and
Company.

Shafer, Siegrid
1976. "Sexual and social problems among lesbians."
Journal of Sex Research, 12(1): 50-69.
 198

Shutz, Alfred
1962. Collected Papers. Vol. I: The Problem of Social
Reality. The Hague: Martinus Nijhoff.
1967. The Phenomenology of the Social World. Evanston,
IL: Northwestern University Press.

Simon, William , and Gagnon, John H.

1984. "Sexual scripts." Society, 22(1): 53-60.

Thio, Alex
1983. Deviant Behavior. Boston: Houghton Mifflin.

Troiden, Richard
1977. "Becoming homosexual: research on acquiring a gay
identity." Journal of Homosexuality, 1(1): 31-46.
1979. "Becoming homosexual: a model of gay identity
acquisition." Psychiatry, 42(4): 362-73.
1984/1985. "Self, self-concept, identity, and homosexual
identity: constructs in need of definition and
differentiation." Journal of Homosexuality,
10(3/4): 97-109.

Turner, Jonathan
1991. The Structure of Sociological Theory. Belmont,
CA: Wadsworth, Inc.

Weinberg, Thomas
1977. "Becoming homosexual: self-disclosure, self-
identity, and self-maintenance." Doctoral
dissertation. University of Connecticut.
1978. "On 'doing' and 'being' gay; sexual behavior and
homosexual male self-identity." Journal of
Homosexuality, 4(2): 143-56.
 199

APPENDIX
 200

Table 1. Factor analysis results for the expression of

homosexual identity scale for gay males and lesbians
as reported in Paper 1

Variables in the Scale Males Females

Frequency of attendance .740 .83 3

in gay organizations

Importance of gay .801 .817

organizations

Frequency of gay .678 .662

sexual contact

Importance of gay .742 .771

sexual contact

Importance of gay .756 .443

friendships

Standardized alpha .753 .798

Cronbach's alpha .745 .794

Table 2. Factor analysis results for the physical

victimization scale as reported in Paper 3

Variables in the Scale Males Females

Threatened .839 .649

Pushed .837 .795

Slapped .779 .681

Punched .826 .781

Struck with an Object .646 .616

Use of a Weapon .485 .436

Standardized alpha .833 .745

Cronbach's alpha .838 .742
 201

Table 3. Factor analysis results for the use of violence

scale as reported in Paper 3

Variables in the Scale Males Females

Threatened .780 .708

Pushed .843 .780

Slapped .778 .717

Punched .836 .751

Struck with an Object .594 .587

Use of a Weapon X X

Note: "X" indicates zero cases occurred for the particular

variables.

Standardized alpha .825 .753

Cronbach's alpha .825 .745
 202

Table 4. Factor analysis results for the sexual aggression

scale as reported in Paper 3

Variables in the Scale Males Females

Intoxication .512 .446

Threat to Terminate .412 .665

Relationship

Threatened to Disclose .570 1 Case

Negative Information

Guilt .660 .694

Physical Detainment .552 .191

Partner Tried to Turn On .607 .702

Use of Lies .626 .764

Use of False Promises .431 .779

Physically Held Down .572 .709

Use of a Weapon X X

Threatened to Use .647 1 Case

Physical Force

Use of Physical Force .457 .655

Note: "X" indicates zero cases occurred for the particular

variables.

Standardized alpha .763 .747

Cronbach's alpha .746 .762
 203

Table 5. Factor analysis results for the self-esteem scale as

reported in Paper 3

Variables in the Scale Males Females

On the whole I am .694 .811

satisfied with myself.

At times I think I am .710 .749

no good at all.

I certainly feel .617 .690

useless at times.

I feel that I am a person .793 .823

of worth, at least on an
equal plane with others.

I take a positive .800 .871

attitude towards myself.

Standardized alpha .773 .849

Cronbach's alpha .746 .836