<%BANNER%>

Associative Stigma and Factors of Association

Permanent Link: http://ufdc.ufl.edu/UFE0020864/00001

Material Information

Title: Associative Stigma and Factors of Association
Physical Description: 1 online resource (72 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: associative, hepatitis, stigma
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Individuals affiliated with primarily-stigmatized others may, due to this relationship, be affixed with associative stigma; this likelihood should increase with different factors of the relationship, each mediated by particular assumptions made by the perceiver. In the current study, participants read vignettes about two roommates, one of them primarily-stigmatized, with high/low manipulations of three affiliation factors of interest. Participants then reported their perceptions of both the associatively-stigmatized and primarily-stigmatized roommates. Contrary to predictions, one affiliation factor (emotional attachment) decreased associative stigma, while the other two factors had no effect on it. Effects of the affiliation factors on primary stigmatization were comparable. In all, these findings suggest that increased emotional attachment may alleviate the effects of both primary and secondary stigmatization.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Cottrell, Catherine.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0020864:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020864/00001

Material Information

Title: Associative Stigma and Factors of Association
Physical Description: 1 online resource (72 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: associative, hepatitis, stigma
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Individuals affiliated with primarily-stigmatized others may, due to this relationship, be affixed with associative stigma; this likelihood should increase with different factors of the relationship, each mediated by particular assumptions made by the perceiver. In the current study, participants read vignettes about two roommates, one of them primarily-stigmatized, with high/low manipulations of three affiliation factors of interest. Participants then reported their perceptions of both the associatively-stigmatized and primarily-stigmatized roommates. Contrary to predictions, one affiliation factor (emotional attachment) decreased associative stigma, while the other two factors had no effect on it. Effects of the affiliation factors on primary stigmatization were comparable. In all, these findings suggest that increased emotional attachment may alleviate the effects of both primary and secondary stigmatization.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Cottrell, Catherine.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0020864:00001


This item has the following downloads:


Full Text





ASSOCIATIVE STIGMA AND FACTORS OF ASSOCIATION


By

DAVID A. R. RICHARDS












A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2008































To Devin









ACKNOWLEDGMENTS

I wish to express my thanks to my advisor for guiding me through the thesis process with a

patient hand, as well as to the members of my committee for their insightful oversight of this

process. I wish to express my gratitude to the many research assistants whose diligent aid were

indispensable to the execution of this study. Finally, I am thankful to my parents for their support

and encouragement.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ............ ...... ._._ ...............3....


LIST OF TABLES ........._.._ ..... ._._ ...............6....


AB S TRAC T ......_ ................. ............_........7


CHAPTER


1 INTRODUCTION ................. ...............8.......... ......


Contact Frequency ................. ...............11.......... .....
Perceived Volition ................. ...............12.......... .....
Emotional Attachment ................. ...............13.................
Current Research and Hypotheses ................. ...............14................

2 M ETHOD .............. ...............16....


Participants .............. ...............16....
Procedure ................ ............... ... ...............16.....
Measures of Dependent Variables ................. ...............18................
M measures of Mediators .............. ...............18....
Measures of Individual Differences. ................ ....................... ................19

Manipulation Checks ................. ...............19.................
Demographic Items. ............. ...............20.....

3 RE SULT S ................. ...............21.......... .....


Preliminary Analyses: Manipulation Checks and Composite Creation ................. ...............21
Hypotheses la, 2a, 3a: Effects of Relational Factors on Associative Stigmatization ............23
Hypotheses lb, 2b, 3b: Mediation of Effects on Associative Stigmatization ........................24
Alternative Mediational Pathways............... .... ..... ..........2
Effects of Relational Factors on Primary Stigmatization ................. ......... ................27

4 DI SCUS SSION ................. ...............3.. 0......... ....


Effects of Relational Factors on Associative Stigmatization .............. ....................3
Mediation of Effects on Associative Stigmatization ................. .............. ......... .....33
Effects on Primary Stigmatization................ ............3
Implications for Research on Associative Stigma ................. ...............36...............
Conclusion and Future Directions .............. ...............39....











APPENDIX

A EXPERIMENTAL MATERIALS ................. ...............41........... ....


B STIGMATIZATION OF Y (AS SOCIATIVELY-STIGMATIZED ROOMMATE)............_.45

C STIGMATIZATION OF X (PRIMARILY-STIGMATIZED ROOMMATE) ......................47

D ASSUMPTION OF Y'S RELATIONAL VALUE .............. ...............49....

E ASSUMPTION OF Y's RELATIONAL ESTEEM............... ...............51.

F ASSUMPTION OF Y' S POSSESSION OF STIGMATIZING ATTRIBUTE......................54

G ASSUMPTION OF Y's RELATIONAL OBLIGATIONS ................. .........................57

H MOTIVATION TO CONTROL PREJUDICE .............. ...............61....

I MANIPULATION CHECK S ........._.___..... ._ __ ...............66....

J DEMOGRAPHIC S ........._.___..... ._ __ ...............67....

LIST OF REFERENCES ........._.___..... .__. ...............69....

BIOGRAPHICAL SKETCH .............. ...............72....










LIST OF TABLES

Table page

3-1 Associative stigmatization across all factors ........._..... .......... ........ .. .........2

3-2 Primary stigmatization across all factors............... ...............29









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

ASSOCIATIVE STIGMA AND FACTORS OF ASSOCIATION

By

David A. R. Richards

May 2008

Chair: Catherine A. Cottrell
Major: Psychology

Individuals affiliated with primarily-stigmatized others may, due to this relationship, be

affixed with associative stigma; this likelihood should increase with different factors of the

relationship, each mediated by particular assumptions made by the perceiver. In our study,

participants read vignettes about two roommates, one of them primarily-stigmatized, with

high/low manipulations of three affiliation factors of interest. Participants then reported their

perceptions of both the associatively-stigmatized and primarily-stigmatized roommates. Contrary

to predictions, one affiliation factor (emotional attachment) decreased associative stigma, while

the other two factors had no effect on it. Effects of the affiliation factors on primary

stigmatization were comparable. Results suggest that increased emotional attachment may

alleviate the effects of both primary and secondary stigmatization.









CHAPTER 1
INTTRODUCTION

Erving Goffman (1963) defined a stigma as "an attribute that is deeply discrediting."

According to Goffman, such an attribute can be a physical abnormality, a personality or

character flaw, or a "tribal stigma" conferred by one's religion, race, or nation, so long as it has

the effect of reducing the stigma-bearer' s value in the eyes of society when possession of the

stigmatizing attribute is brought to light, and usually results in the discriminative treatment of the

stigma-bearer. A wide variety of groups face such social devaluation and rej section on an

everyday basis, including those who are obese (Crocker, Cornwell, & Major, 1993), physically

disabled (Farina, Sherman, & Allen, 1968), mentally ill (Farina, 1982) or retarded (Shears &

Jensema, 1969), homosexual (Herek, 1984), African American (Brigham, 1974), blind (Scott,

1969), or diagnosed with HIV/AIDS (Weitz, 1990) or cancer (Bloom & Kessler, 1994).

Individuals affixed with a stigma face many difficulties as a result of their stigmatization,

not the least of which is their difficulty in forging or maintaining relationships with

nonstigmatized individuals who may prefer not to associate with people bearing such stigma.

Indeed, non-stigmatized individuals who associate with stigma-bearers may become stigmatized

themselves, that association becoming a deeply discrediting attribute of its own. This

phenomenon has been termed stigma by association (Neuberg, Smith, Hoffman, & Russell,

1994)--known also as courtesy stigma (Goffman, 1963) or vicarious stigma (Corrigan & Miller,

2004)--and may involve difficulties similar to those that afflict the primarily-stigmatized (i.e.,

individuals possessing a stigma). Such relationships can cause otherwise "normal" individuals to

suffer associative stigma and its resultant social devaluation and rej section.

Although little research has been conducted on associative stigma, that research which

does exist has examined the phenomenon in the context of a number of types of social









associations, including friendships, romantic partnerships, occupational associations, and familial

relationships. In particular, the stigmatizing effects of having a homosexual roommate

(Sigelman, Howell, Cornell, Cutright, & Dewey, 1991) or friend (Neuberg et al., 1994), a

physically disabled dating partner (Goldstein & Johnson, 1997), or a depressed, alcoholic, or

incarcerated father (Mehta & Farina, 1988) have been studied.

Each of these studies found evidence for associative stigma by experimentally

manipulating the stigma in question. In addition, Neuberg and colleagues found that, while a

non-stigmatized individual incurred associative stigma as a consequence of associating with a

primarily-stigmatized individual (a homosexual friend), the reverse did not take place--that is,

the homosexual friend was not destigmatized as a consequence of associating with a

heterosexual. The interpersonal effects of social connections between primarily-stigmatized and

associatively-stigmatized individuals would appear to be, based on this limited prior research, a

one-way street.

However, much less attention has been given to how different types of relationships (or,

more precisely, how distinguishing features of relationships) confer associative stigma. Sigelman

and colleagues (1991) provide an exception. In addition to manipulating the discrediting attribute

(sexual orientation) borne by the primarily-stigmatized individual, they also manipulated the

nature of this individual's relationship with his heterosexual roommate, who was described as

rooming with the homosexual either by choice or after being involuntarily assigned.

Sigelman et al. found that the heterosexual individual described as being involuntarily

assigned to his roommate was not stigmatized by his association. So the volitional nature of an

association can be considered one feature of a relationship affecting associative stigma.









As additional examples, an otherwise non-stigmatized individual could experience

associative stigma as a result of relationships with either a coworker or a friend with a

stigmatizing condition (e.g., an infectious, debilitating disease). These social connections,

however, could differ greatly in several fundamental respects. A workplace association may

involve a lot of time (e.g., 40 hours per week) spent together, but few feelings of emotional

closeness. On the other hand, a friendship may involve far less face-to-face contact (e.g., fewer

than five hours per week), but strong feelings of emotional closeness. Perhaps these possible

differences in social connections lead to differences in the nature of the associative stigma. In the

case of the relationship with the coworker, society may stigmatize the non-stigmatized individual

because of an assumption that, as a result of the plentiful time spent in the diseased coworker' s

company, there is a very high chance that the associatively-stigmatized individual has also

contracted the coworker' s disease. On the other hand, in the case of the relationship with the

friend, society may assume that the associatively-stigmatized individual's emotional closeness to

the diseased friend burdens him or her with responsibility for the friend's welfare. This burden in

turn could cause the associatively-stigmatized individual to require the aid of others, making him

or her a less desirable relationship partner in the eyes of others and conferring associative stigma.

As this example suggests, it is necessary to identify exactly which critical features (e.g.,

volition of association, emotional closeness, frequency of contact, genetic relatedness) of these

different types of relationships are responsible for the associative stigma. A coworker is different

from a friend in many respects. Which of these respects are responsible for the differing injurious

effects of a person's stigmatized relationship on their perceived social worth? Moreover, what

are the inferential processes through which each relationship feature could exert its effects?









Toward an examination of these novel questions, the present study sought to test a

framework for several dimensions of non-familial associations. The dimensions of interest were

the volition of the relationship (i.e., how much the relationship is voluntary on the part of the

associatively-stigmatized individual), frequency of contact (i.e., how much time the

associatively-stigmatized individual spends in the company of the stigmatized relationship

partner), and emotional attachment (i.e., how close the associatively-stigmatized individual feels

toward the stigmatized relationship partner).

Contact Frequency

Social associations can vary in their contact frequency, the proportion of time that the

associatively-stigmatized and stigmatized individuals spend in each other's company.

Presumably, any time shared in this way provides opportunities for the stigma, if of an ostensibly

contagious nature, to infect the non-stigmatized individual. The impact of such shared company

is determined both by its context, and by the nature of the stigma in question. For example, time

spent together in an Internet chat room is likely to be of little relevance when the stigmatizing

attribute is a contagious disease (e.g., HIV/AIDS), but can become very relevant when the

stigmatizing attribute is ideological, such as subscription to a deviant belief system (e.g.,

S atani sm).

Assuming that the stigmatizing mark is perceived to be of a relevantly contagious nature,

higher contact frequency should, on average, confer greater associative stigma on non-

stigmatized individuals, an effect mediated by an assumption of possession of the stigmatizing

attribute, in which perceivers assume that such associatively-stigmatized individuals have also

come to possess the stigmatizing attribute itself. Non-stigmatized individuals who spend a great

amount of time in the company of an individual bearing a contagious stigmatizing attribute (like

a contagious disease) may be perceived as running an increased risk of becoming primarily










stigmatized themselves. This inference may lead an observer to stigmatize individuals in such

relationships as if they themselves were in fact primarily stigmatized. As a result, this

as sumpti on m ay account for the effects of c contact frequency with n the sti gma-by -as soci ati on

process (Mehta & Farina, 1988; Sigelman et al., 1991; Cottrell, 2000).

Perceived Volition

Social associations can also vary in their perceived volition, that is, the extent to which

the associatively-stigmatized individual's association with a stigmatized person is viewed as

being voluntary on the associatively-stigmatized individual's part (Cottrell, 2000). This factor of

association bears parallels to one of the dimensions of stigma discussed by Jones et al. (1984):

origin, which refers to how an individual came to bear a discrediting attribute. Much as,

according to Jones et al., primarily-stigmatized individuals seen as responsible for the

stigmatizing marks they bear will be stigmatized to a greater degree than those seen merely as

innocent victims, so will associatively-stigmatized individuals seen as responsible for their

relationship with a primarily-stigmatized individual be stigmatized to a greater degree than those

seen as involuntarily thrust into their affiliation by forces beyond their control (as demonstrated

by Sigelman et al., 1991).

The stigmatizing influence of a relationship partner' s stigma is likely to increase as the

rel ati on shi p b between the two i s vi ewed as voliti onal on the p art of the a s soci atively- sti gmati ze d

individual, an effect that may be mediated by an assumption of low relational value of the

associatively-stigmatized individual. Presumably, when forming relationships with one another,

we prefer to enter into associations with people who will, due to some valuable trait or attribute,

most benefit us personally. A person with a stigmatizing attribute makes a poor choice of such a

relationship partner because such a discrediting attribute has the effect of decreasing that

individual's social worth (or else the trait would not be stigmatizing). An observer may assume









that associatively-stigmatized individuals in voluntary relationships with stigmatized others offer

little of value as relationship partners, else they would not have to "settle" for associating with

people who are stigmatized (Mehta & Farina, 1988; Cottrell, 2000).

Emotional Attachment

Associations can also vary in their emotional attachment, the degree to which the

associatively-stigmatized individual's sentiments toward the stigmatized individual are

characterized by feelings of love, affection, and concern. Of interest here is the valence of the

attachment--the extent to which the associatively-stigmatized individual's feelings toward the

stigmatized individual are characterized by positive feelings.

The greater the associatively-stigmatized individual's emotional attachment to the

primarily-stigmatized associate, the greater the degree of associative stigma perceivers will

confer on the associatively-stigmatized individual as a result, an effect possibly mediated by an

assumption of relational obligations. Presumably, the warmer this emotional attachment, the

more motivated the associatively-stigmatized individual will be to invest time and effort into the

well-being of the stigmatized other. As a result, a perceiver may leap to the conclusion that the

associatively-stigmatized individual's relationship with the primarily-stigmatized individual

places a burden on the associatively-stigmatized individual. Because someone so burdened may,

in turn, require aid from others, the perceiver may stigmatize the associatively-stigmatized

individual, so as to avoid burdens that the associatively-stigmatized individual might otherwise

impose on the perceiver. In this way, the assumption of relational obligations may mediate the

relationship between the associatively-stigmatized individual's emotional attachment to the

stigmatized individual and the consequent associative stigma (Cottrell, 2000).

The three relationship dimensions (contact frequency, volition, and emotional attachment)

described here do not account for all the factors of relationships that may differ in ways that









influence the stigma-by-association process. For example, relationships may also differ in the

length of acquaintance as well as the genetic relatedness of the relationship partners, as well as in

the degree that the associatively-stigmatized individual may be perceived as responsible for the

stigmatizing condition possessed by their primarily-stigmatized relationship partner.

Additionally, the effects of these relationship dimensions on the process of associative

stigmatization may be mediated by assumptions other than those described here. For example,

the effect of volition on associative stigma may also, besides an assumption of low relational

value, be mediated by an assumption of low moral standards (if the primary stigma is a deviant

voluntary behavior). An assumption of low moral standards may also, besides an assumption of

relational obligations, mediate the effect of emotional attachment on associative stigma.

However, for purposes of the present study, the nature of the stigma of interest (possession of an

incurable disease, one for which the stigma-bearer may not be responsible) constrained the

choice of mediators to be examined.

Study Hypotheses

Our study explored the following: (1) the extent to which these three dimensions of

relationships (contact frequency, volition, and emotional attachment) confer associative stigma;

and (2) whether these effects are mediated by the described mediators (assumption of possession

of the stigmatizing attribute, assumption of low relational value, and assumption of relational

obligations, respectively). More specifically, the study experimentally manipulated these three

relationship features within an ostensible roommate relationship between a primarily-stigmatized

individual and an associatively-stigmatized individual, and then assessed participants'

impressions of both individuals. Specific hypotheses were as follows:

*Hypothesis la. The level of contact frequency between a primarily-stigmatized

individual and an associatively-stigmatized individual would affect associative










stigma such that greater contact frequency would produce greater associative

stigma.

* Hypothesis lb. The effect of contact frequency on associative stigma would be

mediated by an assumption, on the part of the perceiver, of the associatively-

stigmatized individual's possession of the stigmatizing attribute.

* Hypothesis 2a. The level of an associatively-stigmatized individual's volition in

associating with a primarily-stigmatized individual would affect associative stigma

such that greater volition would produce greater associative stigma.

* Hypothesis 2b. The effect of volition on associative stigma would be mediated by

an as sumpti on, on the p art of the p erceiver, of the as soci atively -sti gmati zed

individual's low relational value.

* Hypothesis 3a. The level of an associatively-stigmatized individual's emotional

attachment to a primarily-stigmatized individual would affect associative stigma

such that greater emotional attachment would produce greater associative stigma.

* Hypothesis 3b. The effect of emotional attachment on associative stigma would be

mediated by an assumption, on the part of the perceiver, of relational obligations

required of anybody in an association with the associatively-stigmatized individual.









CHAPTER 2
IVETHOD

Participants

Participants were 173 students who were enrolled in an introductory psychology course

and participated for partial satisfaction of a course requirement. The average age of the

participants was 19. 13 years (SD = 1.44). Of these, 51.4% identified themselves as

Cauca~sian white, and 25.4% identified themselves as Afr~ican American. The remaining

participants identified as Asian American, Hispanic, Native American, or "other." Females

comprised 78% of the participants. One participant was identified as having participated in a

prior phase of this same study. This participant' s data were excluded from all analyses.

Procedure

After giving informed consent, participants were informed that they were participating in

a study on friendship pairs that focused on roommates sharing one room, two-person dormitory

units (ostensibly because this was the most convenient example of friendship pairs available for

study). The participants were informed that they would be assisting in this research by providing

ratings of different attributes of the roommates in one such pair.

The target roommates were of the same gender, with that gender matched to the gender of

the participant. The target roommates were not referred to by name but, rather, "to protect the

anonymity of the people participating in this study," were identified as Roommate X (the

primarily-stigmatized roommate) and Roommate Y (the associatively-stigmatized roommate).

The participants were then presented with a brief essay ostensibly written by one of the

roommates in the study: roommate Y. Participants were told that, before writing the essay, Y was

provided the following instructions: "What we would like you to do now is write a very short

essay (between 10 and 20 sentences in length) telling us a little about yourself, your roommate,










and your relationship with each other. This essay is very open-ended, and you can give us any

information that you personally think is interesting or relevant. Remember that your responses

will be kept completely confidential, so be as honest and detailed as you possibly can without

going over the 20 sentence limit."

In addition to filler information intended to increase the believability of the essay, the

essay indicated that roommate X has Hepatitis. Previous research suggests that contagious

diseases (e.g., Weitz, 1990) often create primary stigma for their carriers--a necessary

precondition to examine stigma-by-association. In particular, Hepatitis has been shown to be a

stigmatizing disease (Crandall, 1991; Munoz-Plaza, Strauss, Astone, Des Jarlais & Hagan,

2004).

After describing this primarily stigmatizing attribute, the essay presented information

about the roommates' relationship, including the experimental manipulations: three independent

variables, each with two levels, making for a 2 x 2 x 2 design. Those variables were the

following: volition of the social association (the essay indicated that Y either voluntarily chose to

room with X, or was involuntarily assigned to room with X), frequency of contact (the essay

indicated that, as a result of their respective class and work schedules, X and Y either spends

very much to almost all of their waking time in the room together, or very little to almost none of

their waking time in the room together), and emotional attachment (the essay indicated that X

and Y are either very good friends, or not very good friends). Appendix A presents the

instructions and essay that participants viewed.

Participants were prompted to read this essay a second time before proceeding to the

measures, which asked participants to provide ratings of various characteristics of the two

roommates and of their relationship with each other. This questionnaire contained conceptually-









related sets of items with Likert-type answer scales ranging from 1 to 7, each with two anchors,

and measuring the dependent variables and proposed mediators as next described.

Measures of dependent variables

One set of items measured participants' (associative) stigmatization of Y (Appendix B).

This set of seven items was identical to Crandall's (1991) modified version of Bogardus' Social

Distancing Scale (1923), which Crandall used as a measure of stigmatization. An additional set

of identical items measured participants' (primary) stigmatization of X (Appendix C).

Measures of mediators

Another set of seven items (Appendix D) measured Y's perceived relational value. Some

of these items measured participants' perceptions of Y' s success and ability at forming

friendships (e.g., "How many close friends do you think Y (the writer of the essay) has?").

Participants were also asked to put themselves "in the shoes of people who know Y well" and to

consider how they think "people who know Y would rate their relationship with him/her" on

scales measuring the value of a relationship with Y (e.g., Unimportant/Important,

Worthless/Worthwhile). In addition to these items, participants also responded to a 10-item

"relational esteem" scale (Appendix E), a modified version of the Rosenberg Self-Esteem Scale

(Rosenberg, 1989), reworded so that participants, instead of indicating their perceptions of their

own worth, success, ability, and self-esteem, indicated their perceptions of Y' s worth, success,

ability, and self-perceptions.

Another set of 10 items (Appendix F) measured participants' assumptions of Y' s

possession of the stigmatizing attribute. These items consisted of health-related trait ratings of Y,

in which participants rated the likelihood that Y experiences a number of ill effects (e.g., nausea

and/or vomiting, headaches) on the reasoning that a person who has contracted a degenerative

and debilitating disease can be expected to suffer unpleasant symptoms as a result.









A set of 13 items (Appendix G) measured participants' assumptions of Y' s relational

obligations. For these items, participants were asked to put themselves "in the shoes of people

who know Y well" and to consider what they think such a relationship with Y is like. The items

then measured their expectations of how much time, energy, and resources Y's friends must

dedicate to helping Y, and also asked participants to indicate their expectations of the

burdensomeness of such a relationship, as well as Y' s reliance on his/her friends. This set also

included trait ratings of the costs and benefits (e.g., Rewarding/Punishing,

Overwhelming/Manageable) of a relationship with Y.

Measures of individual differences

Study participants may be reluctant to express attitudes that may be construed as

prejudicial or intolerant. To help account for this tendency, I asked participants to respond to a

modified version of Dunton and Fazio' s (1997) motivation to control prejudice scale, a 17-item

scale which measures the degree to which participants prefer to avoid prejudiced thoughts or

behaviors, either because they violate personal standards or because they violate social norms

(Appendix H). Because the original scale focuses on the respondent' s efforts to control prejudice

toward African Americans in particular, I modified the relevant original items to focus on

respondent' s efforts to control prejudice toward "someone different from people like me"; this

minor modification should allow for the measurement of participants' motivation to control

prejudice in general.

Manipulation checks

Finally, the questionnaire contained manipulation checks (Appendix I) in which

participants were asked to indicate their perceptions of Y' s volition in his or her relationship with

the diseased roommate, the percent of time that Y spends in X' s company, and Y' s emotional

attachment to X. A short, open-ended essay item prompted participants to recall information










from Y' s essay, to assess whether participants noted the information about X's stigmatizing

condition.

Demographic items

After completing these measures, participants provided demographic data: gender, age,

ethnicity, and religion (Appendix J). Finally, participants were thoroughly debriefed. None

expressed dissatisfaction with their treatment, and a few expressed interest in the results of the

study .









CHAPTER 3
RESULTS

Preliminary Analyses: Manipulation Checks and Composite Creation

To determine whether the manipulations had the desired effects on participants'

perceptions of the relationship between the two target roommates, analyses of variance

(ANOVAs) with contact frequency, perceived volition, and emotional attachment as between-

subj ects variables were conducted on participants' responses to each of the three manipulation

checks: how much free time the roommates spend in each other's company (contact frequency),

how much their relationship is voluntary on the part of the associatively stigmatized roommate

(volition), and how much the associatively-stigmatized roommate likes the primarily-stigmatized

roommate (emotional attachment).

For the item measuring the perceived amount of time the roommates spent in each other' s

company, the ANOVA revealed the desired main effect of contact frequency, F (1, 164) =

3 18.3 5, p < .001, partial eta-squared = .66, such that participants in the high contact frequency

condition (M~= 5.3 5, SD = 1.61) perceived the roommates as spending more time in each other' s

company than did participants in the low contact frequency condition (M~= 1.95, SD = .66). In

addition, the ANOVA revealed an unintended main effect of attachment on contact frequency, F;

(1, 164) = 11.30, p < .01, partial eta-squared =.06, such that participants in the high attachment

condition (M~= 4.08, SD = 2. 10) perceived the essay writer as spending more time around his or

her roommate more than did participants in the low emotional attachment condition (M~= 3.46,

SD = 2. 1 1).

For the item measuring the perceived volition behind the associatively-stigmatized

roommate's relationship with the primarily-stigmatized roommate, the ANOVA revealed the

desired main effect of volition, F (1, 164) = 61.53, p < .001, partial eta-squared = .27, such that










participants in the high volition condition (M~= 4.76, SD = 1.67) perceived the relationship as

being more volitional on the associatively-stigmatized roommate's part than did participants in

the low volition condition (M~= 2.89, SD = 1.62). In addition, the ANOVA revealed an

unintended main effect of attachment on perceived volition, F (1, 164) = 30.04, p < .001, partial

eta-squared = .27, such that participants in the high attachment condition (M~= 4.47, SD = 1.70)

perceived the essay writer's relationship with his or her roommate as being more volitional than

did participants in the low emotional attachment condition (M~= 3.09, SD = 1.83).

For the item measuring the perceived emotional attachment of the associatively-

stigmatized roommate to the primarily-stigmatized roommate, the ANOVA revealed the desired

main effect of attachment, F (1, 164) = 232.36, p < .001, partial eta-squared = .59, such that

participants in the high emotional attachment condition (M~= 5.48, SD = .97) were perceived as

being more emotionally attached to the primarily-stigmatized roommate than did participants in

the low emotional attachment condition (M~= 3.01, SD = 1.12). None of these ANOVAs revealed

any significant interactions between the effects of these three factors on these measures.

Examination of participants' answers to the open-ended essay item prompting them to

describe the information provided them by the essay showed that 150 of the 173 participants

made mention, in their answers, of the primarily-stigmatized roommate's Hepatitis, illness,

sickness, disease, or health, reassuring me that the essay's information about that roommate's

stigmatizing condition did not go unnoticed.

After the reverse-scored items were reverse-coded, the internal reliability of the various

scales was assessed by means of Cronbach alpha coefficients. The scales were shown to be

internally consistent, as follows: a = .829 for stigmatization of the associatively-stigmatized

roommate (roommate Y), a = .810 for stigmatization of the primarily-stigmatized roommate










(roommate X), a = .896 for assumption of the associatively-stigmatized roommate's relational

value, a = .810 for assumption of the associatively-stigmatized roommate's relational esteem, a

.869 for assumption of the associatively stigmatized roommate's possession of the stigmatizing

attribute, a = .881 for assumption of the associatively-stigmatized roommate's relational

obligations, and a = .780 for motivation to control prejudice. Finding all alphas to be

satisfactory, I averaged the responses to create composites for each scale.

Hypotheses la, 2a, 3a: Effects of Relational Factors on Associative Stigmatization

I predicted that higher levels of the three proposed features of association (frequency of

contact, voliti on, and em oti onal attachm ent) would re sult i n the as soci atively -sti gmatiz ed

individual being affixed with higher levels of stigmatization. To test these hypotheses, I

conducted a between-subj ects analysis of variance (ANOVA) with contact frequency, perceived

volition, and emotional attachment as between-subj ects variables on stigmatization of the

associatively-stigmatized roommate.l Means and standard deviations for the associative stigma

measure across the three factors are shown in Table 3-1.

I predicted that contact frequency would exert a main effect on associative stigmatization

such that higher contact frequency would confer greater associative stigma. This prediction was

not supported by the main effect of contact frequency on associative stigma, which was

nonsignificant, F (1, 165) = .56, p = .46, partial eta-squared < .01.

I also predicted that volition would exert a main effect on associative stigmatization such

that higher volition would confer greater associative stigma. This prediction was not supported





SWhen an analysis of covariance (ANCOVA) was performed on these data with motivation to control prejudice
included as a covariate, the significance of its effects did not differ from those found by the ANOVA. Therefore, I
present just the straightforward ANOVA here.










by the main effect of volition on associative stigma, which was nonsignificant, F (1, 165) = .01,

p = .90, partial eta-squared < .01.

I also predicted that attachment would exert a main effect on associative stigma such that

higher attachment would confer greater associative stigma. Results revealed a significant main

effect of attachment on stigmatization of the associatively-stigmatized roommate, F (1, 165) =

13.59, p < .01, partial eta-squared = .08. However, contrary to my prediction, the direction of this

main effect was opposite that which I had predicted, such that the associatively-stigmatized

roommate was stigmatized less when attachment was high (M~= 3.36, SD = .93) than when

attachment was low (M~= 3.89, SD = .911). Tests for interactions between the three factors also

revealed nonsignficant effects.

Hypotheses lb, 2b, 3b: Mediation of Effects on Associative Stigmatization

I predicted that the effects of the three factors of association (frequency of contact,

volition, and emotional attachment) on associative stigma would be mediated by assumptions of,

respectively, possession of the stigmatizing attribute, low relational value, and relational

obligations. I tested these hypotheses by testing for mediation in the manner recommended by

Baron and Kenney (1986)-specifically, by first regressing the mediator on the factor of

association, then by regressing stigmatization of the associatively-stigmatized roommate on the

mediator, and finally by regressing stigmatization of the associatively-stigmatized roommate on

both the mediator and the factor of association. If the factor of association affected the mediator

in the first equation, the factor of association affected stigmatization in the second equation, and

the mediator affected stigmatization in the third equation, and if the effect of the factor of

association on stigmatization was less in the third equation than in the second, then mediation

would be established. Whenever I found mediation, I conducted a Sobel test (1982) to confirm

its significance.









The only factor of association to affect associative stigma was emotional attachment.

Although the direction of this effect was the reverse of what I had predicted, I nonetheless tested

whether this effect was mediated by the predicted assumption of relational obligations.

Regression of participants' perceptions of relational obligations on attachment revealed that

attachment was a significant predictor of perceptions of relational obligations, B = -.23, p < .01,

such that higher attachment was associated with lower assumptions of relational obligations.

That is to say, when the essay writer was depicted as being emotionally close to the disease-

bearing roommate, participants perceived the essay writer as being less burdensome a

relationship partner than when the essay writer was depicted as not being emotionally close to

the disease-bearing roommate. Finding that attachment predicted assumptions of relational

obligations, I then examined whether relational obligations predicted associative stigmatization.

Regression of associative stigmatization on relational obligations revealed that relational

obligations were a significant predictor of associative stigmatization, a = .54, p < .001, such that

higher relational obligations were associated with higher stigmatization. I then examined whether

relational obligations mediated the effect of attachment on associative stigma. When associative

stigma was predicted from both relational obligations and attachment, I found that attachment

was a lessened, though still significant, predictor of associative stigma, a = -. 16, p < .05, and

relation obligations partially mediated its prediction of associative stigma a= .50, p < .001. This

partial mediation of the effect by assumptions of relational obligations was significant, Sobel = -

2.81, p <.01.

Alternative Mediational Pathways

Although I made no other specific mediational hypotheses, I also tested whether this effect

of attachment on associative stigma could be mediated by either of the other two assumptions.










To test whether participants' assumption of the associatively stigmatized roommate's

possession of the stigmatizing attribute mediated the effect of attachment on associative stigma, I

regressed the participant' s assumptions of poor health on attachment, and found that attachment

was not a significant predictor of assumption of poor health, a = .003, p = .97, ruling out its

mediation of attachment' s effect on associative stigma.

I also examined whether the measure of relational value and the measure of relational

esteem, intended to be measures of the same construct, may have mediated attachment's effect on

associative stigmatization. Though these two measures were significantly correlated with each

other, r = .54, p < .001, I tested them separately for any mediational role each may have played

in this effect of attachment. To test whether participants' assumption of the associatively

stigmatized roommate's relational value mediated the effect of attachment on associative stigma,

I first regressed assumptions of relational value on attachment, finding that attachment

significantly predicted relational value, a = .29, p < .001, such that higher attachment was

associated with higher relational value. I then regressed associative stigma on relational value,

finding that relational value significantly predicted associative stigma, a = -.53, p < .001, such

that higher relational value was associated with lower associative stigma. Finally, when

associative stigma was predicted from both attachment and relational value, I found that

attachment ceased to be a significant predictor (although its significance level was still

marginal), a = -. 13, p = .054, while relational value fully mediated its prediction of associative

stigma, B = -.49, p < .001. This full mediation was significant (Sobel = -3.50, p < .01).

To determine whether the measure of relational esteem also mediated this effect of

attachment on associative stigma, I first regressed relational esteem on attachment, finding that










attachment was not a significant predictor of relational esteem, f = .13, p = .08, ruling out its role

as a mediator of attachment' s effect on associative stigmatization.

Effects of Relational Factors on Primary Stigmatization

In exploratory analyses, I also tested whether the factors of association affected primary

stigmatization. I conducted a between-subj ects ANOVA with contact frequency, perceived

volition, and emotional attachment as between-subj ects variables on stigmatization of the

primarily-stigmatized roommate (the roommate with Hepatitis).2 Means and standard deviations

for the primary stigmatization measure across the three factors are presented in Table 3-2.

This ANOVA revealed a significant main effect of attachment on primary stigmatization,

F ( 1, 164) = 5.92, p < .0 5, p arti al eta- square d = .04, such th at the pri mari ly -sti gm ati zed

roommate was stigmatized less when attachment was high (M~= 4.03, SD = .91) than when

attachment was low (M~= 4.39, SD = .95).

The ANOVA also revealed a significant volition by contact interaction, F (1, 164) = 5.42,

p < .05, partial eta-squared = .03. Simple effect tests conducted on this interaction showed that,

when contact was high (i.e., frequent), volition had a significant effect, F (1, 169) = 10.49, p <

.01, partial eta-squared = .06, such that primary stigmatization was higher when volition was low

(M~= 4.53, SD = .89) than when it was high (M~= 3.91, SD = .93). No other significant simple

effects were found in this interaction.










2 As with associative stigma, when an ANCOVA was performed on these data with motivation to control prejudice
included as a covariate, the significance of its effects did not differ from those found by the ANOVA. Therefore, I
again present just the straightforward ANOVA here.









Table 3-1. Associative stigmatization across all factors
Table 3-1. Associative stigmatization across all factors
Low attachment
Low volition


High attachment

3.60 (.79)
3.18 (.88)

3.34 (.94)
3.33 (1.06)


Low contact frequency
High contact frequency
High volition
Low contact frequency
High contact frequencV


3.77 (.98)
3.91 (.79)

4.00 (1.17)
3.86 (.76)


Note: Higher values indicate higher mean levels of stigmatization. Parenthetical values
indicate standard deviations.









Table 3-2. Primary stigmatization across all factors.
Table 3-2. Primary stigmatization across all factors.


Low Attachment

4.28 (.95)
4.53 (1.06)

4.62 (.87)
4.12 (.86)


High Attachment

4.06 (1.00)
4.53 (.67)

3.84 (.77)
3.77 (.96)


Low Volition
Low Contact Frequency
High Contact Frequency
High Volition
Low Contact Frequency
High Contact FrequencV


Note: Higher values indicate higher mean levels of stigmatization. Parenthetical values
indicate standard deviations.









CHAPTER 4
DISCUS SION

The present study sought to determine how certain factors of association (contact

frequency, volition, and emotional attachment) would influence the degree of associative stigma

conferred on a target in a relationship with a primarily-stigmatized other. I predicted that, when

increased, these factors would confer greater associative stigma on the associatively-stigmatized

target. In addition, I predicted that the greater associative stigma conferred by increased contact

frequency would be mediated by an assumption of the associatively-stigmatized target' s

possession of the stigmatizing attribute, that the greater associative stigma conferred by

increased volition would be mediated by an assumption of the associatively-stigmatized target' s

low relational value, and that the greater associative stigma conferred by emotional attachment

would be mediated by an assumption of the associatively-stigmatized target' s high relational

obligations.

Effects of Relational Factors on Associative Stigmatization

In all, results did not support the study hypotheses. Contrary to predictions, none of the

factors of association, when increased, conferred greater degrees of associative stigma.

Furthermore, emotional attachment had the opposite effect; increases in emotional attachment

resulted in the associatively-stigmatized roommate actually being stigmatized less.

The lack of any effect of contact frequency or volition on associative stigmatization may

be simply due to these affiliative factors lacking a role in this particular associative stigma

process. That is, contrary to predictions, the extent to which an association with a primarily-

stigmatized other results in associative stigmatization may simply not be affected by the contact

frequency or perceived volition of that association, at least in the case of the stigma in question,

Hepatitis. Different relational factors influencing associative stigma are likely to come into play









for different kinds of stigmatizing attributes. Where volition has been shown to increase

associative stigmatization when the stigmatizing attribute is homosexuality (Sigelman et

al.,1991), it did not have that effect in conjunction with the present study's use of Hepatitis. As

for contact frequency, the essay writer may not have appeared likely to catch Hepatitis simply as

a result of spending more time in the company of the primarily-stigmatized roommate. If a

similar study were to instead manipulate the quality rather than the quantity of contact--i.e.,

whether or not the two targets were in a sexual relationship-participants might be more likely to

assume that the associatively-stigmatized target would catch Hepatitis.

Alternatively, it may be that the manipulations of contact frequency and volition

manipulations were simply too weak to have an effect on associative stigmatization. For

example, if a relationship with someone with Hepatitis were to confer associative stigma due to

an assumption of possession of the disease due to increased contact frequency, even roommates

who are "always around each other in the room, almost all the time" (as in the present study's

high contact frequency condition) may not be spending ample enough time in each other's

company for the participant to presume that the essay writer may be at risk for infection. As for

volition, the choice to room with someone with Hepatitis may not have been a meaningful

enough choice for participants to impute lower relational value on the associatively-stigmatized

target. If a future study were to use a more meaningful life choice, volition may be more likely to

affect associative stigma.

As to why increased levels of attachment did affect associative stigmatization, but in the

direction opposite of the predicted effect, any post hoc interpretations of this effect must

necessarily be speculative, but this result still requires an attempt at explanation. The

unanticipated beneficial result of the attachment manipulation on associative stigma may have









been due to a feature of the manipulation itself. The emotional attachment of the target pair was

manipulated by having the associatively-stigmatized roommate report, in the high attachment

condition, that he or she considered the primarily-stigmatized roommate a friend because they

were very close to each other (as opposed to reporting, in the low attachment condition, that he

or she did not consider the primarily-stigmatized roommate a friend and that they were not very

close to each other). An unanticipated effect of the attachment manipulation may also have been,

besides creating a representation of a certain kind of relationship, to create a representation of a

certain kind of relationship partner. An essay writer who described himself or herself as being in

a close, friendly relationship, regardless of the other person in the relationship, may implicitly

represent himself or herself as being friendly. A friendlier-appearing person would also likely

appear to be a more approachable person, resulting in the decreased social distancing displayed

by participants.

This explanation does not address, however, why high attachment's beneficial effects for

the essay writer were not outweighed by the hypothesized increase in the stigmatizing effects of

Hepatitis. The most obvious explanation is that, even though other research (Crandall, 1991;

Munoz-Plaza at al., 2004) suggests that Hepatitis is stigmatizing, it may not have been

sufficiently stigmatizing a disease to confer associative stigma. A more stigmatizing attribute

(e.g., HIV/AIDS, homosexuality) might have been necessary to offset the positive benefits

derived from the targets' appearance of friendliness. Future research might benefit from

observing how the kind of stigmatizing attribute interacts with association factors to stigmatize

or destigmatize targets. Such value-laden stigmas as HIV/AIDS or homosexuality might also be

mediated by assumptions other than those examined by this study--specifieally, participants who

observe an associatively-stigmatized target voluntarily rooming with someone who possesses a










stigma widely viewed as immoral (compared to participants who observe a target involuntarily

assigned to room with such a stigmatized individual) may assume that the target is himself or

herself an immoral person, and may stigmatize the target as a result.

Mediation of Effects on Associative Stigmatization

Although this effect of attachment on associative stigmatization was in the direction

opposite of that I had predicted, it was nonetheless partially mediated by the predicted mediator:

the participant's assumptions concerning the associatively-stigmatized roommate's

burdensomeness (i.e., relational obligations) as a relationship partner. The effect was also fully

mediated by an unpredicted mediator--the participants' assumptions concerning the

associatively-stigmatized roommate's value as a relationship partner.

These results are challenging to interpret. An essay writer who described himself or herself

as being in a close, friendly relationship, regardless of the other person in the relationship, may

implicitly represent himself or herself as being able and willing to enter into such relationships.

In contrast, someone who self-represents as not being very close to, or friendly with, a

relationship partner in his or her life may implicitly appear less willing or able to participate in

such relationships. A willingness and an ability to participate in friendships implies a willingness

and ability to carry out the responsibilities incumbent upon friends, such as the responsibility to

come to a friend's aid when needed. Such relationship partners may potentially be seen as less

burdensome than relationship partners who are not close friends, in that they help reduce, rather

than increase, one's own life burdens. This may have led participants to see the friendlier

roommate as conferring fewer relational obligations on relationship partners, which may have in

turn led to less social distancing (i.e., less stigmatization). With regard to the other confirmed

mediator of attachment's effect on associative stigmatization, relational value, a friendlier, more










helpful relationship partner might also, naturally, be perceived as a more valuable relationship

partner, which may in turn likewise lead to less social distancing on the part of the participant.

An alternative, less tenuous explanation may simply be that the increased perception of

friendliness elicited by the higher attachment, and the resultant increased attractiveness, may

have favorably biased participants' perceptions of the associatively-stigmatized roommate' s

relational value and relational obligations in a manner akin to a halo effect. However, this

interpretation is inconsistent with the fact the lack of attachment' s effect on participants'

perceptions of the associatively-stigmatized roommates healthiness. If higher attachment led

participants to more favorably evaluate the target' s traits, one would expect that participants

would attribute better health to the targets in the high attachment condition. This, however, was

not the case.

Effects on Primary Stigmatization

Attachment had the same effect on primary stigmatization as it did on associative

stigmatization in that higher attachment conferred less stigma on the primarily-stigmatized

roommate. Attachment may have had this effect on primary stigmatization for the same reason it

had the same effect on associative-stigmatization: The high attachment condition, by

representing the primarily-stigmatized roommate as being in a close, friendly relationship, may

have made the primarily-stigmatized roommate appear to be a less burdensome and more

valuable relationship partner, leading participants to engage in less social distancing.

Interestingly, even though volition and contact frequency did not exert any main effects on

either primary or associative stigmatization, they did interact with each other in affecting primary

stigmatization. The essay writer's choice to room with the primarily stigmatized roommate did

have an effect on primary stigmatization, such that higher volition resulted in lower










stigmatization, but only when the two roommates spent a great deal of time in each other's

company. When contact was infrequent, there was no effect of volition.

Though this interpretation is highly speculative, it seems that, when the essay writer

voluntarily chose to live with the primarily-stigmatized roommate, participants may have

inferred from this decision that the primarily-stigmatized roommate must have had some

appealing qualities (or else why choose to live with him or her?) leading the participants to

socially distance themselves from the primarily-stigmatized roommate to a lesser degree than

they would have if the roommate had been assigned to them. However, this decision to room

with the primarily-stigmatized roommate would likely be much more consequential if the essay

writer spent a great deal of time in his or her roommate's company, because a poor choice would

lead to the unfortunate result of spending a great deal of time in the company of someone

annoying or unpleasant. If the essay writer chose to room with someone they would spend a lot

of time with, then participants may have inferred that the person they chose to room with must be

especially likeable, if the essay writer was so willing to put himself or herself into a situation

where he or she would be so frequently in that person's company.

On the other hand, in the case of participants in the low contact condition, the decision

either to room or not to room with the primarily-stigmatized roommate may be seen as a much

less consequence-laden choice, as the essay writer may not have expected to spend much time in

the primarily-stigmatized roommate's company. Participants may infer that the essay writer, for

that reason, did not take the primarily-stigmatized roommate's likeability into consideration

when making this decision, making this choice less informative about the primarily-stigmatized

roommate's desirability as a roommate or relationship partner. This may have led participants to










weigh the essay writer' s volition less when evaluating the primarily-stigmatized roommate's

approachability, reducing the effect of volition on consequent primary stigmatization.

Again, these highly speculative interpretations of the interacting effects of volition and

contact should be further tested before given any acceptance.

Implications for Research on Associative Stigma

In their study of associative stigma, Neuberg et al. (1994) expected that the stigma-by-

association process would yield one of four possible outcomes: (1) The stigmatized, "marked"

individual would become destigmatized through his or her association with a "normal" other (a

"lifting" effect), (2) the "normal" would become stigmatized through his or her association with

the marked individual, (3) a contrast effect would cause the normal being less stigmatized and

the marked individual being more stigmatized as a result of the association, or (4) the association

would have no effect on the level of stigmatization of either the marked or normal individuals.

To discover which outcome would result, participants viewed an interaction between two targets,

one of which was presented as either heterosexual or homosexual. As the present study was

interested in how characteristics of the relationship (and not characteristics of the members of

the relationship) affected associative stigma, such manipulations were outside the present study's

scope, and it is impossible to say which, if any, of those four proposed outcomes occurred here.

However, it is interesting to note that the present study's manipulation of one aspect of

the targets' rel ati onship-attachment-re suited in an outcome that di d not map onto any of the

four predicted by Neuberg et al.: Both targets were destigmatized. It is encouraging to see that a

"marked" individual may enj oy a "lifting" effect from associating with a "normal" without

necessarily exacting an irreparable toll on the social acceptability of either relationship

participant, and that any associative stigmatization of the normal, if it occurs, may be offset by a









feature of the very relationship that, theoretically, is causing the normal to be stigmatized.

Associative stigma may not be a zero-sum dilemma.

Like the present study, Sigelman et al. (1991) experimentally manipulated whether a

target living with a stigmatized (homosexual, in this case) roommate had voluntarily chosen that

living arrangement. In addition, Sigelman et al. also assessed participants' tolerance or

intolerance of homosexuals. They found that, when participants rated as being high in

intolerance for homosexuals were provided information about the target living with a

homosexual roommate, targets who were depicted as having voluntarily chosen to room with a

gay student were perceived as being more homosexual and as having more stereotypically

homosexual traits than targets who were assigned to live with a gay roommate. More relevant to

comparisons to the present study, intolerant participants also liked targets who voluntarily chose

to live with a gay roommate less than they liked targets who were assigned to live with a gay

roommate. Although homosexuality is very different from the stigmatizing attribute I employed

(Hepatitis), it is still worth contrasting the effects of volition in their study to its effects in my

own. While their experimental increase in volition resulted in greater stigmatization and greater

attribution of stigma-relevant traits to the associatively stigmatized roommate, my own

manipulation of that same variable did not have any effects on stigmatization of, or attribution of

stigma-relevant traits to, associatively-stigmatized targets. This discrepancy may be due to

differing valence and stereotypic content of the two stigmatizing attributes, but it is still worth

noting that my own findings may represent an exception to the effects of volition on associative

stigmatization.

I suspect that one of the greatest challenges stigmatized individuals are likely to face (or

likely to believe they face) as a result of their stigmatization is an increased difficulty in forming









deep, meaningful, mutually healthy and beneficial relationships with other human beings as a

result of others' fears of being stigmatized by their associations with them. Given the present

findings that a greater emotional connection may actually have a beneficial result for both the

associatively-stigmatized and primarily stigmatized relationship partners, and assuming for the

moment that my post hoc interpretations of these results are valid, the implications of these

findings may actually bear considerable relevance to the area of impression management--

specifically, beneficial impression management. Human beings are constantly managing the

impressions they make on other people in strategic, motivated ways, a behavior referred to as

self-presentation (Schlenker, 1980). Additionally, people also make strategic presentations of

individuals with whom they are closely and positively affiliated, working to present their friends

in ways that are most likely to benefit their friends (Schlenker & Britt, 1999). Beneficial

impression management is a highly under-researched area, and to my knowledge, no extant

research on beneficial impression management (Schlenker & Britt, 1999; Schlenker & Britt,

2001; Pontari & Schlenker, 2004; Schlenker, Lifka, & Wowra, 2004; Pontari & Schlenker, 2006)

experimentally examined how beneficial impression management might be successfully

achieved by means of strategic presentations of the presenter's relationship with the beneficiary

instead of (or in addition to) strategic presentations of the beneficiary's personal qualities and

virtues. The mutually advantageous results of such a strategy, seemingly implied by the present

study's results, would do well to be examined by further research aimed at corroborating its

benefits, and, if real, could be beneficially applied by associatively-stigmatized individuals who

find themselves in situations where they are either unable or unwilling to conceal their

association with a primarily-stigmatized relationship partner and who wish to make the best of a

bad situation that, in the final analysis, perhaps needn't be so bad.









Conclusion and Future Directions

This is not to say that associative stigma does not pose a considerable problem for many

people, primarily- and associatively-stigmatized alike, or that the way different relational factors

may exacerbate this problem is yet understood by research into its processes. If the lessons

learned from this study are to prove at all useful, it will most likely be in their ability to inform

future research on this subj ect. To that end, I believe future research would benefit by focusing

exclusively on two relational factors--attachment (given its effects here) and volition (given its

effects verified by Sigelman et al., 1991)--as extant research seems to indicate that these are the

relational factors that seem to influence associative stigmatization most reliably. Additionally,

future research should manipulate the presence of the stigmatizing attribute, as well as the nature

of the attribute itself (e.g., Hepatitis, HIV/AIDS, or homosexuality), to see how these factors

interact with the associative factors under examination. It may be that more stereotypically

immoral stigmas may actually confer greater associative stigma when attachment and volition

are higher. Finally, given the finding that the relational factors did influence primary stigma,

future studies intended to further scrutinize how relational factors affect stigmatization of both

participants in the relationship would benefit from a more symmetrical presentation of the

targets, as opposed to the present study's more lopsided presentation of one particular target as

the source of all the participant' s information about the relationship of the targets to each other.

In sum, though higher attachment had a beneficial, instead of a deleterious, effect on

associative stigmatization, contrary to what I had hypothesized, and though the implications of

this effect require further empirical confirmation before they can be accepted with any

confidence, this finding may still shed light on how different features of a relationship with a

primarily-stigmatized individual can influence people's perceptions and stigmatization of both

members of such a relationship. The counter-hypothetical (yet ironically heartening) finding that










higher reported attachment had a "lifting" effect on participants' stigmatization of the primarily-

stigmatized target, but at no cost (and, indeed, with some improvement) to the associatively-

stigmatized writer's own social approachability, may be good news to those who find themselves

in such relationships and are uncertain how to portray these relationships to others.









APPENDIX A
EXPERIMENTAL MATERIALS

[Experimental manipulations are bracketed. Gender pronouns were matched to the gender

of the participant.]

First of all, thank you for helping us with this study. Your assistance is extremely vital to

our efforts to learn more about people and their relationships with each other. The focus of this

study is on pairs of University of Florida students formerly residing in one-room, two-person

dormitory units on campus. We are studying these roommates because we are interested in

characteristics of two-person relationships, and roommate pairs present the most convenient

example of these kinds of relationships. Last year, we recruited a number of roommates from the

University of Florida student body, and asked them to provide information about their living

situations. This is where you come in. During your time in this session today, you will be acting

as a rater.

You will read a statement that has been provided by one of the roommates who

participated in our study. These participants were asked to write a short essay about themselves,

their roommates, and their relationships with their roommates. In order to protect the anonymity

of the people participating in this study, the roommates will not be identified by name. Instead,

they will be referred to as X and Y. You will be considering the essay written by roommate Y.

Because the information you will be providing to us will be essential to our research goals,

we wish to stress to you how important it is that you read the essay closely and that you provide

only your most honest responses in the questionnaire. Keep in mind that your identity is also

being protected. Neither your name, nor any other information that can identify you, will be

attached to your responses, so please answer the questions as honestly as you can.









Thank you again for your assistance in this study. When you're ready to proceed, please go

to the next page.










As you just read, the roommates participating in this study were asked to write a short

essay. Here are the exact instructions these participants were given:

"What we would like you to do now is write a very short essay (between 10 and 20

sentences in length) telling us a little about yourself, your roommate, and your relationship with

each other. This essay is very open-ended, and you can give us any information that you

personally think is interesting or relevant. Remember that your responses will be kept completely

confidential, so be as honest and detailed as you possibly can without going over the 20 sentence

limit."

One of the statements written by one of the participants is provided below. Please read it

carefully, and then, when you feel ready to proceed, go to the next page.



This is my first year at the University of Florida. I'm originally from Tampa, Florida, and

I'm nineteen years old. I'm maj oring in Anthropology and am so far enjoying my time here, but

I'm thinking about switching maj ors.

Stuff I do for fun: I like all kinds of music. When I watch TV, I usually like Comedy

Central. I love dogs but I had to leave mine at home since the dorms don't allow them. I've been

to a couple of football games and I usually go out one or two times a week, sometimes more.

My roommate is from Jacksonville. One interesting thing about her is that she's the only

person I know with Hepatitis. (It's contagious but I'm not sure how she caught it.) There's no cure

and she has some bad health problems. She spends a lot of time online and is studying

Linguistics. She likes reading and she says she doesn't know what she wants to do when she gets

out of college.










This is the first time we've shared a dorm. [We were assigned as room mates, I didn't

actually choose to live with her. / We weren't assigned as room mates, I actually chose to live

with her.] Because of the way our work and class schedules overlap with each other (we both

have part time j obs) [we're never around each other in the room, almost none of the time. / we're

always around each other in the room, almost all the time.] All in all, especially after rooming

with her this long, I would have to say that [I don't really consider her a friend, because we

aren't very close to each other. / I really consider her a friend, because we're very close to each

other.]

That' s all I can think of to say, I'm not sure what else to write, I hope this is what you

needed.









APPENDIX B
STIGMATIZATION OF Y (ASSOCIATIVELY-STIGMATIZED ROOMMATE)

The following questions inquire about Y (the writer of the essay).

Indicate the degree to which you agree or disagree with the following statements:

1. Y appears to be a likeable person.


Strongly disagree


Strongly agree


2. I would like Y to be a close personal friend.


Strongly disagree


Strongly agree


3. I wouldn't mind it at all for Y to move into my neighborhood.


Strongly disagree


Strongly agree


4. I would like Y to come and work at the same place I do.


Strongly disagree


Strongly agree









5. Y is a person who is similar to me.

Strongly disagree


1 2 3


Strongly agree


7






Strongly agree


7






Strongly agree


7


6. I would like to have Y marry into my family.

Strongly disagree


1 2 3 4


7. Y is the kind of person that I tend to avoid.

Strongly disagree









APPENDIX C
STIGMATIZATION OF X (PRIMARILY-STIGMATIZED ROOMMATE)

The following questions inquire about X (the roommate of the essay's writer).

Indicate the degree to which you agree or disagree with the following statements:

X appears to be a likeable person.


Strongly disagree


Strongly agree


I would like X to be a close personal friend.


Strongly disagree


Strongly agree


I wouldn't mind it at all for X to move into my neighborhood.


Strongly disagree


Strongly agree


I would like X to come and work at the same place I do.


Strongly disagree


Strongly agree









X is a person who is similar to me.

Strongly disagree


1 2 3


Strongly agree


7






Strongly agree


7






Strongly agree


7


I would like to have X marry into my family.

Strongly disagree


1 2 3


X is the kind of person that I tend to avoid.

Strongly disagree









APPENDIX D
ASSUMPTION OF Y' S RELATIONAL VALUE

The following questions inquire about Y (the writer of the essay).

1. How many close friends do you think Y (the writer of the essay) has?


Very few


Very many


2. To what extent would the average UF student want Y as a friend?


Not at all


1


Very much


3. To what extent do you think people who know Y value their relationships with him/her?

Not at all Very much


1 2 3 4 5 6 7


Put yourself in the shoes of people who know Y well. How do you think people who

know Y would rate their relationship with him/her on the five scales below?

4.


Unimportant


1


Important












Valuable Not valuable


1 2 3 4 5 6 7







Worthless Worthwhile


1 2 3 4 5 6 7







Good Bad


1 2 3 4 5 6 7









APPENDIX E
ASSUMPTION OF Y'S RELATIONAL ESTEEM

Indicate the degree to which you agree or disagree with the following statements:

1. Y is a person of worth, at least on an equal plane with others.


Strongly disagree


Strongly agree


2. Y has a number of good qualities.


Strongly disagree


Strongly agree


3. All in all, Y is a failure.


Strongly disagree


Strongly agree


4. Y is able to do things as well as most other people.


Strongly disagree


Strongly agree


5. Y does not have much to be proud of.










Strongly disagree


Strongly agree


7


6. Y takes a positive attitude toward himself/herself.


Strongly disagree


Strongly agree


7. On the whole, Y is satisfied with himself/herself.


Strongly disagree


Strongly agree


8. Y wishes he/she could have more respect for himself/herself.


Strongly disagree


Strongly agree


9. Y certainly feels useless at times.

Strongly disagree


Strongly agree


7


10. At times Y thinks he/she is no good at all.










Strongly agree


Strongly disagree









APPENDIX F
ASSUMPTION OF Y'S POSSESSION OF STIGMATIZING ATTRIBUTE

The following questions inquire about Y (the writer of the essay).

How likely do you think it is that Y (the writer of the essay) experiences each of the

following?

1. General feelings of poor health


Very unlikely


Very likely


2. Difficulty sleeping


Very unlikely


Very likely


3. Nausea and/or vomiting


Very unlikely


Very likely


4. Light headedness and/or fainting


Very unlikely


Very likely









5. Headaches


Very unlikely


Very likely


6. Susceptibility to infections


Very unlikely


Very likely


7. Weakness


Very unlikely


Very likely


8. Exhaustion from minor exertion


Very unlikely


Very likely


9. Dizziness


Very unlikely


Very likely









10. Unhealthy body weight (either overweight or underweight)

Very unlikely


Very likely









APPENDIX G
ASSUMPTION OF Y'S RELATIONAL OBLIGATIONS

The following questions inquire about Y (the writer of the essay).

Put yourself in the shoes of people who know Y (the writer of the essay) well. Consider

what you think their relationship with Y is like. For the following questions, indicate what you

think a relationship with Y is like.

1. How much time do you think Y' s friends must dedicate to helping Y deal with his/her

problems?

None at all Very much


1 2 3 4 5 6 7




2. How much energy do you think Y' s friends must dedicate to helping Y deal with his/her

problems?

None at all Very much


1 2 3 4 5 6 7




3. How much of their resources do you think Y' s friends must dedicate to helping Y deal with

his/her problems?

None at all Very much


1 2 3 4 5 6 7










4. How burdensome do you think Y's friends find their relationship with Y?


Not at all


1


Very much


5. How inconvenient do you think Y' s friends find their relationship with Y?


Very much


6. How demanding do you think Y's friends find their relationship with Y?


Not at all


1


Very much


7. How problematic do you think Y' s friends finds their relationship with Y?


Very much


8. How much assistance do you think Y requires from his/her friends?


None at all


1


Very much


Not at all


1


Not at all


1









Put yourself in the shoes of people who know Y well. Consider what you think their

relationship with Y is like. On each of the ten scales below, indicate what you think their

experience of the relationship with Y is like, from their point of view:



People who know Y well find their relationship with Y...


Refreshing


Draining


Rewarding


Punishing


Low maintenance


High maintenance


Overwhelming


Manageable










13.


Energizing


Exhausting


7









APPENDIX H
MOTIVATION TO CONTROL PREJUDICE



1. In today's society it is important that one not be perceived as prejudiced in any manner.


Strongly disagree


Strongly agree


2. I always express my thoughts and feelings, regardless of how controversial they might be.

Strongly disagree Strongly agree


3. I get angry with myself when I have a thought or feeling that might be considered prejudiced.


Strongly disagree


Strongly agree


4. If I were participating in a class discussion and a student who was different from people like

me expressed an opinion with which I disagreed, I would be hesitant to express my own

viewpoint.


Strongly disagree


Strongly agree









5. Going through life worrying about whether you might offend someone is just more trouble

than it's worth.


Strongly disagree Strongly agree


7







Strongly agree


7


6. It's important to me that other people not think I'm prejudiced.


Strongly disagree


7. I feel it's important to behave according to society's standards.


Strongly disagree


Strongly agree


8. I'm careful not to offend my friends, but I don't worry about offending people I don't know or

don't like.

Strongly disagree Strongly agree









9. I think that it is important to speak one's mind rather than to worry about offending someone.

Strongly disagree Strongly agree


10. It's never acceptable to express one's prejudices.

Strongly disagree


Strongly agree


11. I feel guilty when I have a negative thought or feeling about someone different from people

like me.


Strongly disagree Strongly agree


12. When speaking to someone different from people like me, it's important to me that he/she not

think I'm prejudiced.

Strongly disagree Strongly agree









13. It bothers me a great deal when I think I've offended someone, so I'm always careful to

consider other people's feelings.

Strongly disagree Strongly agree


7






Strongly agree


7






Strongly agree


7


14. If I have a prejudiced thought or feeling, I keep it to myself.

Strongly disagree


1 2 3 4 5


15. I would never tell jokes that might offend others.

Strongly disagree


1 2 3 4


16. I'm not afraid to tell others what I think, even when I know they disagree with me.

Strongly disagree Strongly agree


1 2 3 4 5 6 7










17. If someone who made me uncomfortable sat next to me on a bus, I would not hesitate to

move to another seat.


Strongly disagree Strongly agree


1 2 3 4 5 6 7









APPENDIX I
MANIPULATION CHECKS

In the short statement that you read, Y provided information about his/her self, his/her

roommate, and his/her relationship with the roommate. We'd like you to answer some questions

about that information. Even if the essay you read didn't have any information relevant to these

questions, please answer them to the best of your ability:



How much of Y' s total free time does he/she spend in X' s company?


Very little


Very much


How much would you say Y's relationship with X is voluntary?


Not at all voluntary


Very voluntary


How much would you say Y likes X?


Not at all


1


Very much


In the space below, please briefly describe the information that was provided in the essay

that you read:










APPENDIX J
DEMOGRAPHICS

Please answer the following questions. As mentioned before, your responses are

confidential. These demographic questions only serve to help us explore the sample helping us

with our study.



What is your gender? Male Female



How old are you? years



Which of the following ethnic groups) do you consider yourself a member of! You can

check multiple groups.

African American

Asian Americans

Hispanic

Native American

Caucasian/White

Other:










Which of the following religions best describes you?

Protestant (Methodist, Lutheran, Episcopalian, etc.)

Catholic

Fundamentalist/Evangelical Christian

Jewi sh

Muslim

Hindu

Buddhi st

Athei st

Other:

No religious affiliation









LIST OF REFERENCES


Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1 173-1 182.

Bloom, J. R., & Kessler, L. (1994). Emotional support following cancer: A test of the
stigma and social activity hypothesis. Journal of Health and Social Behavior, 35, 1 18-133.

Bogardus, E. S. (1923). Immigration and race attitudes. Boston: D.C. Heath.

Brigham, J. C. (1974). Views of black and white children concerning the distribution of
personality characteristics. Journal ofPersonality, 42, 144-158.

Corrigan, P. W., Miller, F. E. (2004). Shame, blame, and contamination: A review of the
impact of mental illness stigma on family members. Journal of2~entalHealth, 13, 537-548.

Crandall, C. S. (1991). Multiple Stigma and AIDS: Illness stigma and attitudes toward
homosexuals and IV drug users in AIDS-related stigmatization. Journal of Community &
Applied Social Psychology, 1, 165-172.

Crocker, J., Cornwell, B., & Major, B. (1993). The stigma of overweight: Affective
consequences of attributional ambiguity. Journal ofPersonality and Social Psychology, 64, 60-
70.

Dunton, B. C. & Fazio, R. H. (1997). An individual difference measure of motivation to
control prejudiced reactions. Personality & Social Psychology Bulletin, 23, 316-326.

Farina, A. (1982). The stigma of mental disorders. In A. G. Miller (Ed.), In the eye of the
beholder: Contemporary approaches to stereotyiping (pp. 305-363). New York: Praeger.

Farina, A., Sherman, M., & Allen, J. G. (1968). The role of physical abnormalities in
interpersonal perception and behavior. Journal ofAbnormal Psychology, 73, 590-593.

Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood
Cliffs, NJ: Prentice Hall.

Goldstein, S. B., & Johnson, V. A. (1997). Stigma by association: Perceptions of the
dating partners of college students with physical disabilities. Basic andApplied Social
Psychology, 19, 495-504.

Herek, G. M. (1984). Attitudes toward lesbians and gay men: A factor-analytic study.
Journal ofHomosexuality, 10, 39-51.

Jones, E. E., Farina, A., Hastorf, A. H., Markus, H., Miller, D. T., & Scott, R. A. (1984).
Social stigma: The psychology of marked relationships. New Y ork: Freeman.










Mehta, S. I., & Farina, A. (1988). Associative stigma: Perceptions of the difficulties of
college-aged children of stigmatized fathers. Journal of Social and Clinical Psychology, 7, 192-
202.

Munoz-Plaza, C. E., Strauss, S. M., Astone, J. M., Des Jarlais, D. C., Hagan, H. (2004).
Drug treatment programs as sites of opportunity for delivery of Hepatitis C prevention education:
Client and staff perspectives. Journal of Drug Issues, 34, 861-878.

Neuberg, S. L., Smith, D. M., Hoffman, J. C., & Russell, F. J. (1994). When we observe
stigmatized and "normal" individuals interacting: Stigma by association. Personality and Social
Psychology Bulletin, 20, 196-209.

Park, J. H., Faulkner, J., & Schaller, M. (2003). Evolved disease-avoidance processes and
contemporary anti-social behavior: Prejudicial attitudes and avoidance of people with
disabilities. Journal ofNonverbal Behavior, 27, 65-87.

Pontari, B. A., & Schlenker, B. R. (2001). Providing and withholding impression
management support for romantic partners: Gender of the audience matters. Journal of
Experimental Social Psychology, 40, 41-51.

Pontari, B. A., & Schlenker, B. R. (2006). Helping friends manage impressions: We like
helpful liars but respect nonhelpful truth tellers. Basic andApplied Social Psychology, 28, 177-
183.

Rosenberg, M. (1989). Society and the Adolescent Self-Image. Revised edition.
Middletown Cliffs, NJ: Wesleyan University Press.

Schlenker, B. R. (1980). Impression Management: The Self-concept, Social Identity, and
Interpersonal Relations. Monterey, CA: Brooks/Cole Publishing Company.

Schlenker, B. R., & Britt, T. W. (1999). Beneficial impression management: Strategically
controlling information to help friends. Journal ofPersonality and Social Psychology, 76, 559-
573.

Schlenker, B. R., & Britt, T. W. (2001). Strategically controlling information to help
friends: Effects of empathy and friendship strength on beneficial impression management.
Journal of Experimental Social Psychology, 3 7, 3 57-3 72.

Schlenker, B. R., Lifka, A., & Wowra, S. A. (2004). Helping new acquaintances make
the right impression: Balancing image concerns of others and self. Selfandldentity, 3, 191-206.

Scott, R. A. (1969). The making of blind men: A study of adult socialization. New York:
Russell Sage Foundation.

Shears, L. M., & Jensema, C. J. (1969). Social acceptability of anomalous persons.
Exceptional Children, 36, 91-96.











Sigelman, C. K., Howell, J. L., Cornell, D. P., Cutright, J. D., & Dewey, J. C. (1991).
Courtesy stigma: The social implications of associating with a gay person. Journal of Social
Psychology, 131, 45-56.

Sobel, M.E. (1982). Asymptotic confidence intervals for indirect effects in structural
equation models. In S. Leinhardt (ed.), Sociological M~ethodology 1982 (pp. 290-312).
Washington DC: American Sociological Association.

Weitz, R. (1990). Living with the stigma of AIDS. Qualitative Sociology, 13, 23-38.









BIOGRAPHICAL SKETCH

David Richards was born in Miami. After attending Jesuit High School of Tampa, he

earned a B.A. in English at the University of Florida, and a B.A. in psychology, with honors, at

the University of South Florida, Tampa, before returning to the University of Florida to pursue a

graduate education in social psychology.





PAGE 1

ASSOCIATIVE STIGMA AND FACTORS OF ASSOCIATION By DAVID A. R. RICHARDS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

To Devin 2

PAGE 3

ACKNOWLEDGMENTS I wish to express my thanks to my advisor fo r guiding me through the thesis process with a patient hand, as well as to the members of my committee for their insightful oversight of this process. I wish to express my gratitude to the many research assistants whose diligent aid were indispensable to the execution of this study. Finally, I am thankful to my parents for their support and encouragement. 3

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................3 LIST OF TABLES ...........................................................................................................................6 ABSTRACT .....................................................................................................................................7 CHAPTER 1 INTRODUCTION ................................................................................................................ ....8 Contact Frequency ..................................................................................................................11 Perceived Volition ..................................................................................................................12 Emotional Attachment ............................................................................................................13 Current Research and Hypotheses ..........................................................................................14 2 METHOD ...................................................................................................................... .........16 Participants .............................................................................................................................16 Procedure ................................................................................................................................16 Measures of Dependent Variables. ..................................................................................18 Measures of Mediators. ...................................................................................................18 Measures of Indivi dual Differences. ...............................................................................19 Manipulation Checks. ......................................................................................................19 Demographic Items. ........................................................................................................20 3 RESULTS ..................................................................................................................... ..........21 Preliminary Analyses: Manipulati on Checks and Composite Creation .................................21 Hypotheses 1a, 2a, 3a: Effects of Relati onal Factors on Associ ative Stigmatization ............23 Hypotheses 1b, 2b, 3b: Mediation of Ef fects on Associative Stigmatization ........................24 Alternative Mediational Pathways ..........................................................................................25 Effects of Relational Factor s on Primary Stigmatization .......................................................27 4 DISCUSSION .................................................................................................................. .......30 Effects of Relational Factors on Associative Stigmatization .................................................30 Mediation of Effects on A ssociative Stigmatization ..............................................................33 Effects on Primary Stigmatization ..........................................................................................34 Implications for Research on Associative Stigma ..................................................................36 Conclusion and Future Directions ..........................................................................................39 4

PAGE 5

5 APPENDIX A EXPERIMENTAL MATERIALS ..........................................................................................41 B STIGMATIZATION OF Y (ASSOCIATI VELY-STIGMATIZED ROOMMATE) .............45 C STIGMATIZATION OF X (PRIMARI LY-STIGMATIZED ROOMMATE) ......................47 D ASSUMPTION OF YS RELATIONAL VALUE ................................................................49 E ASSUMPTION OF Y's RELATIONAL ESTEEM ................................................................51 F ASSUMPTION OF YS POSSESSION OF STIGMATIZING ATTRIBUTE ......................54 G ASSUMPTION OF Ys R ELATIONAL OBLIGATIONS ....................................................57 H MOTIVATION TO CO NTROL PREJUDICE ......................................................................61 I MANIPULATION CHECKS .................................................................................................66 J DEMOGRAPHICS .................................................................................................................67 LIST OF REFERENCES ...............................................................................................................69 BIOGRAPHICAL SKETCH .........................................................................................................72

PAGE 6

LIST OF TABLES Table page 3-1 Associative stigmatization across all factors ...........................................................................28 3-2 Primary stigmatization across all factors. ................................................................................29 6

PAGE 7

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ASSOCIATIVE STIGMA AND FACTORS OF ASSOCIATION By David A. R. Richards May 2008 Chair: Catherine A. Cottrell Major: Psychology Individuals affiliated with primarily-stigmatized others may, due to this relationship, be affixed with associative stigma; this likelihood should increase with different factors of the relationship, each mediated by particular assu mptions made by the perceiver. In our study, participants read vignettes a bout two roommates, one of them primarily-stigmatized, with high/low manipulations of three a ffiliation factors of interest. Pa rticipants then reported their perceptions of both the a ssociatively-stigmatized and primaril y-stigmatized roommates. Contrary to predictions, one affiliation factor (emotional attachment) decreased associative stigma, while the other two factors had no effect on it. Effects of the affiliation factors on primary stigmatization were comparable. Results suggest that increased emotional attachment may alleviate the effects of both prim ary and secondary stigmatization. 7

PAGE 8

CHAPTER 1 INTRODUCTION Erving Goffman (1963) defined a stigma as an attribute that is deeply discrediting. According to Goffman, such an attribute can be a physical abnormality, a personality or character flaw, or a tribal stigma conferred by ones religion, race, or nation, so long as it has the effect of reducing the stigma-bearers value in the eyes of society when possession of the stigmatizing attribute is brought to light, and usually results in th e discriminative treatment of the stigma-bearer. A wide variety of groups face such social devaluation and rejection on an everyday basis, including those who are obese (Crocker, Cornwell, & Major, 1993), physically disabled (Farina, Sherman, & Allen, 1968), menta lly ill (Farina, 1982) or retarded (Shears & Jensema, 1969), homosexual (Herek, 1984), African American (Brigham, 1974), blind (Scott, 1969), or diagnosed with HIV/AIDS (Weitz, 1990) or cancer (Bl oom & Kessler, 1994). Individuals affixed with a stigma face many diff iculties as a result of their stigmatization, not the least of which is their difficulty in forging or maintaining relationships with nonstigmatized individuals who may prefer not to associate with people bearing such stigma. Indeed, non-stigmatized individual s who associate with stigma-b earers may become stigmatized themselves, that association becoming a deep ly discrediting attribute of its own. This phenomenon has been termed stigma by association (Neuberg, Smith, Hoffman, & Russell, 1994)known also as courtesy stigma (Goffman, 1963) or vicarious stigma (Corrigan & Miller, 2004)and may involve difficulties similar to those that afflict the primarily-stigmatized (i.e., individuals possessing a stigma). Su ch relationships can cause othe rwise normal individuals to suffer associative stigma and its result ant social devaluat ion and rejection. Although little research has been conducted on associative stigma, th at research which does exist has examined the phenomenon in the context of a number of types of social 8

PAGE 9

associations, including friendships, romantic partnerships, occupational associations, and familial relationships. In particular, the stigmatizi ng effects of having a homosexual roommate (Sigelman, Howell, Cornell, Cutright, & Dewey, 1991) or friend (Neuberg et al., 1994), a physically disabled dating partne r (Goldstein & Johnson, 1997), or a depressed, alcoholic, or incarcerated father (Mehta & Fa rina, 1988) have been studied Each of these studies f ound evidence for associative stigma by experimentally manipulating the stigma in question. In addi tion, Neuberg and colleagues found that, while a non-stigmatized individual incurre d associative stigma as a cons equence of associating with a primarily-stigmatized individual (a homosexual fr iend), the reverse did not take placethat is, the homosexual friend was not de stigmatized as a consequenc e of associating with a heterosexual. The interpersonal effects of social connections between primarily-stigmatized and associatively-stigmatized individu als would appear to be, based on this limited prior research, a one-way street. However, much less attention has been given to how different types of relationships (or, more precisely, how distinguishing features of relationships) confer associative stigma. Sigelman and colleagues (1991) provide an exception. In ad dition to manipulating the discrediting attribute (sexual orientation) borne by the primarily-stigma tized individual, they also manipulated the nature of this individuals re lationship with his heterosexual roommate, who was described as rooming with the homosexual either by choice or after being i nvoluntarily assigned. Sigelman et al. found that the heterosexual individual described as being involuntarily assigned to his roommate was not stigmatized by his association. So the volitional nature of an association can be considered one feature of a relationship affecting associative stigma. 9

PAGE 10

As additional examples, an otherwise non-stigmatized individual could experience associative stigma as a result of relationships with either a coworker or a friend with a stigmatizing condition (e.g., an infectious, debi litating disease). Thes e social connections, however, could differ greatly in several funda mental respects. A workplace association may involve a lot of time (e.g., 40 hours per week) spen t together, but few feelings of emotional closeness. On the other hand, a friendship may involve far less face-to-face contact (e.g., fewer than five hours per week), but strong feelings of emotional cl oseness. Perhaps these possible differences in social connections lead to differences in the nature of the associative stigma. In the case of the relationship with the coworker, soci ety may stigmatize the non-stigmatized individual because of an assumption that, as a result of the plentiful time spent in the diseased coworkers company, there is a very high ch ance that the associatively-s tigmatized individual has also contracted the coworkers disease. On the othe r hand, in the case of the relationship with the friend, society may assume that the associatively-stigmatized individuals emotional closeness to the diseased friend burdens him or her with responsibility for the fr iends welfare. This burden in turn could cause the associatively-stigmatized in dividual to require the aid of others, making him or her a less desirable relationship partner in the eyes of others and conferring associative stigma. As this example suggests, it is necessary to identify exactly which critical features (e.g., volition of association, emotional closeness, freq uency of contact, genetic relatedness) of these different types of relationships are responsible fo r the associative stigma. A coworker is different from a friend in many respects. Which of these re spects are responsible for the differing injurious effects of a persons stigmatized relationship on their perceived social worth? Moreover, what are the inferential processes through which each relationship feature could exert its effects? 10

PAGE 11

Toward an examination of these novel que stions, the present study sought to test a framework for several dimensions of non-familial a ssociations. The dimensions of interest were the volition of the relatio nship (i.e., how much the relationship is voluntary on the part of the associatively-stigmatized individual), frequency of contact (i.e., how much time the associatively-stigmatized individual spends in the company of the stigmatized relationship partner), and emotional attachment (i.e., how clos e the associatively-stigmatized individual feels toward the stigmatized relationship partner). Contact Frequency Social associations can vary in their contact frequency the proportion of time that the associatively-stigmatized and stigmatized i ndividuals spend in eac h others company. Presumably, any time shared in this way provides oppor tunities for the stigma, if of an ostensibly contagious nature, to infect the non-stigmatized individual. The impact of such shared company is determined both by its context, and by the natu re of the stigma in question. For example, time spent together in an Internet chat room is likel y to be of little relevance when the stigmatizing attribute is a contagious dis ease (e.g., HIV/AIDS), but can be come very relevant when the stigmatizing attribute is ideological, such as subscription to a deviant belief system (e.g., Satanism). Assuming that the stigmatizing mark is perceived to be of a relevantly contagious nature, higher contact frequency should, on average, confer greater associative stigma on nonstigmatized individuals, an effect mediated by an assumption of possessio n of the stigmatizing attribute in which perceivers assume that such asso ciatively-stigmatized individuals have also come to possess the stigmatizing attribute itself. Non-stigmatized individuals who spend a great amount of time in the company of an individual bearing a contagious stig matizing attribute (like a contagious disease) may be perceived as r unning an increased risk of becoming primarily 11

PAGE 12

stigmatized themselves. This inference may lead an observer to stigmatize individuals in such relationships as if they themselves were in fact primarily stigmatized. As a result, this assumption may account for the effects of contac t frequency within the stigma-by-association process (Mehta & Farina, 1988; Sigelman et al., 1991; Cottrell, 2000). Perceived Volition Social associations can also vary in their perceived volition that is, the extent to which the associatively-stigmatized individuals associ ation with a stigmatized person is viewed as being voluntary on the associativel y-stigmatized individuals part (C ottrell, 2000). This factor of association bears parallels to one of the dimensi ons of stigma discussed by Jones et al. (1984): origin which refers to how an individual came to bear a discrediting attribute. Much as, according to Jones et al., primarily-stigmati zed individuals seen as responsible for the stigmatizing marks they bear will be stigmatized to a greater degree than those seen merely as innocent victims, so will associ atively-stigmatized i ndividuals seen as re sponsible for their relationship with a primarily-stigm atized individual be stigmatized to a greater degree than those seen as involuntarily thrust into their affiliation by forces beyond their control (as demonstrated by Sigelman et al., 1991). The stigmatizing influence of a relationship partners stigma is likely to increase as the relationship between the two is vi ewed as volitional on the part of the associatively-stigmatized individual, an effect that may be mediated by an assumption of low relational value of the associatively-stigmatized individual. Presumably, when forming relationships with one another, we prefer to enter into associations with people who will, due to some valu able trait or attribute, most benefit us personally. A person with a stigma tizing attribute makes a poor choice of such a relationship partner because such a discrediting attribute has the effect of decreasing that individuals social worth (or el se the trait would not be stigma tizing). An observer may assume 12

PAGE 13

that associatively-stigmatized indi viduals in voluntary relationships with stigmatized others offer little of value as relationship partners, else they woul d not have to settle for associating with people who are stigmatized (Mehta & Farina, 1988; Cottrell, 2000). Emotional Attachment Associations can also vary in their emotional attachment, the degree to which the associatively-stigmatized indivi duals sentiments toward the stigmatized individual are characterized by feelings of love affection, and concern. Of intere st here is the valence of the attachmentthe extent to which the associativel y-stigmatized individuals feelings toward the stigmatized individual are char acterized by positive feelings. The greater the associatively-stigmatized individuals emotional attachment to the primarily-stigmatized associate, the greater th e degree of associative stigma perceivers will confer on the associatively-stigma tized individual as a result, an effect possibly mediated by an assumption of relational obligations Presumably, the warmer this emotional attachment, the more motivated the associatively-stigmatized indivi dual will be to invest time and effort into the well-being of the stigmatized other. As a result, a perceiver may leap to the conclusion that the associatively-stigmatized individuals relati onship with the primarily-stigmatized individual places a burden on the associativel y-stigmatized individual. Because someone so burdened may, in turn, require aid from othe rs, the perceiver may stigmatize the associatively-stigmatized individual, so as to avoid burdens that the associatively-stigmatized i ndividual might otherwise impose on the perceiver. In this way, the assu mption of relational obligations may mediate the relationship between the associatively-stigmatized individuals emotional attachment to the stigmatized individual and the consequent associative stigma (Cottrell, 2000). The three relationship dimensions (contact fr equency, volition, and emotional attachment) described here do not account for all the factors of relationships that may differ in ways that 13

PAGE 14

influence the stigma-by-association process. For example, relationships may also differ in the length of acquaintance as well as the genetic relate dness of the relationship partners, as well as in the degree that the associativelystigmatized individual may be perceived as responsible for the stigmatizing condition possessed by their prim arily-stigmatized relationship partner. Additionally, the effects of these relationship dimensions on the process of associative stigmatization may be mediated by assumptions ot her than those describe d here. For example, the effect of volition on associ ative stigma may also, besides an assumption of low relational value, be mediated by an assumption of low mora l standards (if the primary stigma is a deviant voluntary behavior). An assumption of low moral standards may also, besides an assumption of relational obligations, mediate the effect of emotional at tachment on associative stigma. However, for purposes of the present study, the nature of the stigma of inte rest (possession of an incurable disease, one for which the stigma-bearer may not be responsible) constrained the choice of mediators to be examined. Study Hypotheses Our study explored the following: (1) the ex tent to which these three dimensions of relationships (contact frequency, volition, and emo tional attachment) confer associative stigma; and (2) whether these effects are mediated by th e described mediators (assumption of possession of the stigmatizing attribute, assumption of lo w relational value, and assumption of relational obligations, respectively). More specifically, the study experimentally manipulated these three relationship features within an ostensible roommate relationship between a primarily-stigmatized individual and an associativel y-stigmatized individual, and then assessed participants impressions of both individuals. Sp ecific hypotheses were as follows: Hypothesis 1a. The level of contact frequenc y between a primarily-stigmatized individual and an associa tively-stigmatized individual would affect associative 14

PAGE 15

15 stigma such that greater contact freq uency would produce greater associative stigma. Hypothesis 1b. The effect of contact frequency on associative stigma would be mediated by an assumption, on the part of the perceiver, of the associativelystigmatized individuals possession of the stigmatizing attribute. Hypothesis 2a. The level of an associatively-stigmatized individuals volition in associating with a primarily-stigmatized i ndividual would affect associative stigma such that greater volition would produ ce greater associative stigma. Hypothesis 2b. The effect of volition on associa tive stigma would be mediated by an assumption, on the part of the percei ver, of the associatively-stigmatized individuals low relational value. Hypothesis 3a. The level of an associatively-stigmatized individuals emotional attachment to a primarily-stigmatized indi vidual would affect associative stigma such that greater emotional attachment would produce greater associative stigma. Hypothesis 3b. The effect of emotional attachment on associative stigma would be mediated by an assumption, on the part of the perceiver, of relational obligations required of anybody in an association with the associatively-stig matized individual.

PAGE 16

CHAPTER 2 METHOD Participants Participants were 173 students who were en rolled in an introductory psychology course and participated for partial sa tisfaction of a course requirement. The average age of the participants was 19.13 years ( SD = 1.44). Of these, 51.4% identified themselves as Caucasian/white and 25.4% identified themselves as African American The remaining participants identified as Asian American, Hispanic Native American or other. Females comprised 78% of the participants One participant was identified as having participated in a prior phase of this same study. This particip ants data were exclude d from all analyses. Procedure After giving informed consent, participants were informed that they were participating in a study on friendship pairs that focused on room mates sharing one room, two-person dormitory units (ostensibly because this wa s the most convenient example of friendship pairs available for study). The participants were inform ed that they would be assisti ng in this research by providing ratings of different attributes of the roommates in one such pair. The target roommates were of the same gender, with that gender matched to the gender of the participant. The target roommates were not referred to by name but, rather, to protect the anonymity of the people participating in this study, were identified as Roommate X (the primarily-stigmatized roommate) and Roommate Y (the associatively-stigmatized roommate). The participants were then presented with a brief essay ostensibly written by one of the roommates in the study: roommate Y. Participants were told that, before writing the essay, Y was provided the following instructions: What we w ould like you to do now is write a very short essay (between 10 and 20 sentences in length) telling us a little about yourself, your roommate, 16

PAGE 17

and your relationship with each other. This essay is very open-ended, and you can give us any information that you personally think is interest ing or relevant. Rememb er that your responses will be kept completely confidential, so be as honest and detailed as you possibly can without going over the 20 sentence limit. In addition to filler information intended to increase the believability of the essay, the essay indicated that roommate X has Hepatitis. Previous research suggests that contagious diseases (e.g., Weitz, 1990) often create prim ary stigma for their carriersa necessary precondition to examine stigma-by-association. In particular, Hepatitis has been shown to be a stigmatizing disease (Crandall, 1991; Munoz-Pla za, Strauss, Astone, Des Jarlais & Hagan, 2004). After describing this primarily stigmatizing attribute, the essay presented information about the roommates relationship, including the experimental ma nipulations: three independent variables, each with two levels, making for a 2 x 2 x 2 design. Those variables were the following: volition of the social a ssociation (the essay indicated that Y either voluntarily chose to room with X, or was involuntarily assigned to room with X), frequency of contact (the essay indicated that, as a result of their respective class and work schedules, X and Y either spends very much to almost all of their waking time in th e room together, or very little to almost none of their waking time in the room together), and em otional attachment (the essay indicated that X and Y are either very good friends, or not very good friends). Appendix A presents the instructions and essay th at participants viewed. Participants were prompted to read this essay a second time before proceeding to the measures, which asked participants to provide ratings of various characteristics of the two roommates and of their relationshi p with each other. This questi onnaire contained conceptually17

PAGE 18

related sets of items with Likert -type answer scales ranging from 1 to 7, each with two anchors, and measuring the dependent variables and proposed mediators as next described. Measures of dependent variables One set of items measured participants (a ssociative) stigmatization of Y (Appendix B). This set of seven items was identical to Crandall s (1991) modified version of Bogardus Social Distancing Scale (1923), which Crandall used as a measure of stigmatization. An additional set of identical items measured participants (primary) stigmatization of X (Appendix C). Measures of mediators Another set of seven items (Appendix D) meas ured Ys perceived relational value. Some of these items measured part icipants perceptions of Ys success and ability at forming friendships (e.g., How many close friends do you think Y (the writer of the essay) has?). Participants were also asked to put themselves in the shoes of people who know Y well and to consider how they think people who know Y w ould rate their relationship with him/her on scales measuring the value of a relati onship with Y (e.g., Unimportant/Important, Worthless/Worthwhile). In additi on to these items, participants also responded to a 10-item "relational esteem" scale (Appendix E), a modified version of the Rosenberg Self-Esteem Scale (Rosenberg, 1989), reworded so that participants, instead of indicat ing their percep tions of their own worth, success, ability, and self-esteem, indi cated their perceptions of Ys worth, success, ability, and self-perceptions. Another set of 10 items (Appendix F) meas ured participants assumptions of Ys possession of the stigmatizing attribute. These items consisted of health-related trait ratings of Y, in which participants rated the likelihood that Y experiences a number of ill effects (e.g., nausea and/or vomiting, headaches) on th e reasoning that a person who ha s contracted a degenerative and debilitating disease can be expected to suffer unpleasant symptoms as a result. 18

PAGE 19

A set of 13 items (Appendix G) measured pa rticipants assumptions of Ys relational obligations. For these items, participants were as ked to put themselves in the shoes of people who know Y well and to consider what they think such a relationship with Y is like. The items then measured their expectati ons of how much time, energy, a nd resources Ys friends must dedicate to helping Y, and also asked participants to indicate their expectations of the burdensomeness of such a relationship, as well as Ys reliance on his/her friends. This set also included trait ratings of the costs and benefits (e.g., Rewarding/Punishing, Overwhelming/Manageable) of a relationship with Y. Measures of individual differences Study participants may be reluctant to express attitudes that may be construed as prejudicial or intolerant. To he lp account for this tendency, I aske d participants to respond to a modified version of Dunton and Fazios (1997) motivation to cont rol prejudice scale, a 17-item scale which measures the degree to which partic ipants prefer to avoi d prejudiced thoughts or behaviors, either because they violate personal standards or becau se they violate social norms (Appendix H). Because the original scale focuses on the respondents efforts to control prejudice toward African Americans in particular, I modi fied the relevant original items to focus on respondents efforts to control pr ejudice toward someone different from people like me; this minor modification should allow for the measur ement of participants motivation to control prejudice in general. Manipulation checks Finally, the questionnaire contained ma nipulation checks (Appendix I) in which participants were asked to indica te their perceptions of Ys volition in his or her relationship with the diseased roommate, the percent of time that Y spends in Xs company, and Ys emotional attachment to X. A short, open-ended essay item prompted participants to recall information 19

PAGE 20

20 from Ys essay, to assess whet her participants noted the information about X's stigmatizing condition. Demographic items After completing these measures, participants provided demographic data: gender, age, ethnicity, and religion (Appendix J). Finally, participants were thoroughly debriefed. None expressed dissatisfaction with their treatment, and a few expressed interest in the results of the study.

PAGE 21

CHAPTER 3 RESULTS Preliminary Analyses: Manipulatio n Checks and Composite Creation To determine whether the manipulations had the desired effects on participants' perceptions of the rela tionship between the two target r oommates, analyses of variance (ANOVAs) with contact frequenc y, perceived volition, and emotional attachment as betweensubjects variables were conducted on participants' responses to each of the three manipulation checks: how much free time the roommates spe nd in each other's company (contact frequency), how much their relationship is voluntary on the pa rt of the associatively stigmatized roommate (volition), and how much the a ssociatively-stigmatized roommate likes the primarily-stigmatized roommate (emotional attachment). For the item measuring the perceived amount of time the roommates spent in each others company, the ANOVA revealed the desired main effect of contact frequency, F (1, 164) = 318.35, p < .001, partial eta-squared = .66, such that pa rticipants in the high contact frequency condition ( M = 5.35, SD = 1.61) perceived the roommates as spending more time in each others company than did participants in the low contact frequency condition ( M = 1.95, SD = .66). In addition, the ANOVA revealed an unintended main effect of attachment on contact frequency, F (1, 164) = 11.30, p < .01, partial eta-squared =.06, such that participants in the high attachment condition ( M = 4.08, SD = 2.10) perceived the essay writer as spending more time around his or her roommate more than did participants in the low emotional attachment condition ( M = 3.46, SD = 2.11). For the item measuring the perceived volit ion behind the associatively-stigmatized roommates relationship with the primarily-stigmatized roommate, the ANOVA revealed the desired main effect of volition, F (1, 164) = 61.53, p < .001, partial eta-squared = .27, such that 21

PAGE 22

participants in the high volition condition ( M = 4.76, SD = 1.67) perceived the relationship as being more volitional on the associatively-stigmati zed roommates part than did participants in the low volition condition (M = 2.89, SD = 1.62). In addition, the ANOVA revealed an unintended main effect of attachment on perceived volition, F (1, 164) = 30.04, p < .001, partial eta-squared = .27, such that participan ts in the high atta chment condition ( M = 4.47, SD = 1.70) perceived the essay writer's relationship with his or her roommate as being more volitional than did participants in the low em otional attachment condition ( M = 3.09, SD = 1.83). For the item measuring the perceived emo tional attachment of the associativelystigmatized roommate to the primarily-stigmatiz ed roommate, the ANOVA revealed the desired main effect of attachment, F (1, 164) = 232.36, p < .001, partial eta-squared = .59, such that participants in the high emo tional attachment condition ( M = 5.48, SD = .97) were perceived as being more emotionally attached to the primarily -stigmatized roommate than did participants in the low emotional attachment condition ( M = 3.01, SD = 1.12). None of these ANOVAs revealed any significant interactions between the eff ects of these three fact ors on these measures. Examination of participants answers to th e open-ended essay item prompting them to describe the information provided them by the essay showed that 150 of the 173 participants made mention, in their answers, of the prim arily-stigmatized roommates Hepatitis, illness, sickness, disease, or health, r eassuring me that the essays in formation about that roommates stigmatizing condition did not go unnoticed. After the reverse-scored items were reverse-coded, the internal reliability of the various scales was assessed by means of Cronbach alpha coefficients. The scales were shown to be internally consistent, as follows: = .829 for stigmatization of the associatively-stigmatized roommate (roommate Y), = .810 for stigmatization of the primarily-stigmatized roommate 22

PAGE 23

(roommate X), = .896 for assumption of the associativ ely-stigmatized roommate's relational value, = .810 for assumption of the associativelystigmatized roommate's relational esteem, = .869 for assumption of the associatively stigmatized roommate's possession of the stigmatizing attribute, = .881 for assumption of the associativ ely-stigmatized roommate's relational obligations, and = .780 for motivation to control pr ejudice. Finding all alphas to be satisfactory, I averaged the responses to create composites for each scale. Hypotheses 1a, 2a, 3a: Effects of Relational Factors on Associative Stigmatization I predicted that higher levels of the three pr oposed features of asso ciation (frequency of contact, volition, and emotional attachment) w ould result in the asso ciatively-stigmatized individual being affixed with higher levels of stigmatization. To te st these hypotheses, I conducted a between-subjects analysis of varian ce (ANOVA) with contac t frequency, perceived volition, and emotional attachment as between -subjects variables on stigmatization of the associatively-stigmatized roommate.1 Means and standard deviations for the associative stigma measure across the three factor s are shown in Table 3-1. I predicted that contact frequency would exer t a main effect on associative stigmatization such that higher contact frequency would confer greater associative stigma. This prediction was not supported by the main effect of contact frequency on associativ e stigma, which was nonsignificant, F (1, 165) = .56, p = .46, pa rtial eta-squared < .01. I also predicted that volition would exert a ma in effect on associative stigmatization such that higher volition would confer greater associ ative stigma. This pred iction was not supported 1 When an analysis of covariance (A NCOVA) was performed on these data with motivation to control prejudice included as a covariate, the significance of its effects did not differ from those found by the ANOVA. Therefore, I present just the straightforward ANOVA here. 23

PAGE 24

by the main effect of volition on associative stigma, which was nonsignificant, F (1, 165) = .01, p = .90, partial eta-squared < .01. I also predicted that attachment would exert a main effect on associative stigma such that higher attachment would confer greater associative stigma. Results revealed a significant main effect of attachment on stigmatization of the associatively-stigmatized roommate, F (1, 165) = 13.59, p < .01, partial eta-squared = .08. However, contra ry to my prediction, the direction of this main effect was opposite that which I had predic ted, such that the asso ciatively-stigmatized roommate was stigmatized less when attachment was high ( M = 3.36, SD = .93) than when attachment was low ( M = 3.89, SD = .911). Tests for interactions between the three factors also revealed nonsignficant effects. Hypotheses 1b, 2b, 3b: Mediation of E ffects on Associative Stigmatization I predicted that the effects of the three factors of associ ation (frequency of contact, volition, and emotional attachment) on associativ e stigma would be mediated by assumptions of, respectively, possession of the stigmatizing attribute, low relational value, and relational obligations. I tested these hypot heses by testing for mediation in the manner recommended by Baron and Kenney (1986)specifically, by first regressing the mediator on the factor of association, then by regressing s tigmatization of the associativel y-stigmatized roommate on the mediator, and finally by regressing stigmatization of the associatively-stigmatized roommate on both the mediator and the factor of association. If the factor of a ssociation affected the mediator in the first equation, the factor of association affected stigmatiz ation in the second equation, and the mediator affected stigmatizat ion in the third equation, and if the effect of the factor of association on stigmatization was less in the third equation than in the second, then mediation would be established. Whenever I found mediati on, I conducted a Sobel te st (1982) to confirm its significance. 24

PAGE 25

The only factor of association to affect a ssociative stigma was emotional attachment. Although the direction of this effect was the reve rse of what I had predicted, I nonetheless tested whether this effect was mediated by the pred icted assumption of relational obligations. Regression of participants' percep tions of relational obligations on attachment revealed that attachment was a significant predictor of perceptions of relational obligations, = -.23, p < .01, such that higher attachment was associated w ith lower assumptions of relational obligations. That is to say, when the essay writer was depicted as being emotionally close to the diseasebearing roommate, participants perceived th e essay writer as being less burdensome a relationship partner than when th e essay writer was depicted as not being emotionally close to the disease-bearing roommate. Finding that attachment predicted assumptions of relational obligations, I then examined wh ether relational obligations pred icted associative stigmatization. Regression of associative stig matization on relational obligati ons revealed that relational obligations were a significant pred ictor of associa tive stigmatization, = .54, p < .001, such that higher relational obligations were associated with higher stigmatization. I then examined whether relational obligations mediated th e effect of attachment on associative stigma. When associative stigma was predicted from both relational obliga tions and attachment, I found that attachment was a lessened, though still significant, predictor of associative stigma, = -.16, p < .05, and relation obligations partially mediated its prediction of associative stigma = .50, p < .001. This partial mediation of the effect by assumptions of relational obligations wa s significant, Sobel = 2.81, p < .01. Alternative Mediational Pathways Although I made no other specific mediational hypotheses, I also tested whether this effect of attachment on associative stigma could be me diated by either of the other two assumptions. 25

PAGE 26

To test whether participants' assumption of the associatively stigmatized roommate's possession of the stigmatizing attribute mediated th e effect of attachment on associative stigma, I regressed the participants assu mptions of poor health on attach ment, and found that attachment was not a significant predictor of assumption of poor health, = .003, p = .97, ruling out its mediation of attachments e ffect on associative stigma. I also examined whether the measure of relational value and the measure of relational esteem, intended to be measures of the same construct, may have mediated attachment's effect on associative stigmatization. Though these two meas ures were significantly correlated with each other, r = .54, p < .001, I tested them separately for any mediational role each may have played in this effect of attachment. To test whethe r participants' assumption of the associatively stigmatized roommate's relational value mediated the effect of attachment on associative stigma, I first regressed assumptions of relational value on attach ment, finding that attachment significantly predicte d relational value, = .29, p < .001, such that higher attachment was associated with higher relational value. I then regressed associative stig ma on relational value, finding that relational value significantly predicted associative stigma, = -.53, p < .001, such that higher relational value was associated w ith lower associative stigma. Finally, when associative stigma was predic ted from both attachment and relational value, I found that attachment ceased to be a significant pred ictor (although its significance level was still marginal), = -.13, p = .054, while relational value fully me diated its prediction of associative stigma, = -.49, p < .001. This full mediation was significant (Sobel = -3.50, p < .01). To determine whether the measure of relationa l esteem also mediated this effect of attachment on associative stigma, I first regres sed relational esteem on attachment, finding that 26

PAGE 27

attachment was not a significant predictor of relational esteem, = .13, p = .08, ruling out its role as a mediator of attachments e ffect on associative stigmatization. Effects of Relational Factors on Primary Stigmatization In exploratory analyses, I also tested whethe r the factors of associ ation affected primary stigmatization. I conducted a be tween-subjects ANOVA with contact frequency, perceived volition, and emotional attachment as between -subjects variables on stigmatization of the primarily-stigmatized roommate (t he roommate with Hepatitis).2 Means and standard deviations for the primary stigmatization measure across th e three factors are presented in Table 3-2. This ANOVA revealed a significan t main effect of attachment on primary stigmatization, F (1, 164) = 5.92, p < .05, partial eta-squared = .04, such that the primarily-stigmatized roommate was stigmatized less when attachment was high ( M = 4.03, SD = .91) than when attachment was low ( M = 4.39, SD = .95). The ANOVA also revealed a signifi cant volition by contact interaction, F (1, 164) = 5.42, p < .05, partial eta-squared = .03. Simple effect test s conducted on this inte raction showed that, when contact was high (i.e., frequent ), volition had a significant effect, F (1, 169) = 10.49, p < .01, partial eta-squared = .06, such that primary s tigmatization was higher when volition was low ( M = 4.53, SD = .89) than when it was high ( M = 3.91, SD = .93). No other significant simple effects were found in this interaction. 2 As with associative stigma, when an ANCOVA was perfor med on these data with motivation to control prejudice included as a covariate, the significance of its effects did not differ from those found by the ANOVA. Therefore, I again present just the straightforward ANOVA here. 27

PAGE 28

Table 3-1. Associative stigmatization across all factors Table 3-1. Associative stigmatization across all factors Low attachment High attachment Low volition Low contact frequency 3.77 (.98) 3.60 (.79) High contact frequency 3.91 (.79) 3.18 (.88) High volition Low contact frequency 4.00 (1.17) 3.34 (.94) High contact frequency 3.86 (.76) 3.33 (1.06) Note: Higher values indicate higher mean leve ls of stigmatization. Parenthetical values indicate standard deviations. 28

PAGE 29

29 Table 3-2. Primary stigmati zation across all factors. Table 3-2. Primary stigmati zation across all factors. Low Attachment High Attachment Low Volition Low Contact Frequency 4.28 (.95) 4.06 (1.00) High Contact Frequency 4.53 (1.06) 4.53 (.67) High Volition Low Contact Frequency 4.62 (.87) 3.84 (.77) High Contact Frequency 4.12 (.86) 3.77 (.96) Note: Higher values indicate higher mean leve ls of stigmatization. Parenthetical values indicate standard deviations.

PAGE 30

CHAPTER 4 DISCUSSION The present study sought to determine how certain factor s of association (contact frequency, volition, and emotional attachment) woul d influence the degree of associative stigma conferred on a target in a relationship with a primarily-stigma tized other. I predicted that, when increased, these factors would conf er greater associative stigma on the associatively-stigmatized target. In addition, I predicted th at the greater associative stigma conferred by increased contact frequency would be mediated by an assumption of the associatively-stigmatized targets possession of the stigmatizing attribute, that the greater associative stigma conferred by increased volition would be mediated by an assump tion of the associatively-stigmatized targets low relational value, and that the greater associ ative stigma conferred by emotional attachment would be mediated by an assumption of the asso ciatively-stigmatized targets high relational obligations. Effects of Relational Factors on Associative Stigmatization In all, results did not support the study hypotheses. Contrary to predictions, none of the factors of association, when increased, conferred greater de grees of associative stigma. Furthermore, emotional attachment had the oppos ite effect; increases in emotional attachment resulted in the associatively-stigmatized roommate actually being stigmatized less. The lack of any effect of contact frequenc y or volition on associat ive stigmatization may be simply due to these affiliative factors lacking a role in this particular associative stigma process. That is, contrary to predictions, the extent to which an association with a primarilystigmatized other results in associative stigmatiz ation may simply not be affected by the contact frequency or perceived volition of that associatio n, at least in the case of the stigma in question, Hepatitis. Different relational fact ors influencing associative stigma are likely to come into play 30

PAGE 31

for different kinds of stigmatizing attributes Where volition has been shown to increase associative stigmatization when the stigmatiz ing attribute is homosexuality (Sigelman et al.,1991), it did not have that effect in conjunction with the present studys use of Hepatitis. As for contact frequency, the essay writer may not have appeared likely to catch Hepatitis simply as a result of spending more time in the compa ny of the primarily-stigmatized roommate. If a similar study were to instead manipulate the quality rather than the quantity of contacti.e., whether or not the two targets were in a sexual re lationshipparticipants mi ght be more likely to assume that the associatively-stigma tized target would catch Hepatitis. Alternatively, it may be that the manipula tions of contact frequency and volition manipulations were simply too weak to have an effect on associat ive stigmatization. For example, if a relationship with someone with Hepa titis were to confer associative stigma due to an assumption of possession of the disease due to increased contact frequency, even roommates who are always around each other in the room, almost all the time (as in the present study's high contact frequency condition) may not be spending ample enough time in each other's company for the participant to presume that the essa y writer may be at risk for infection. As for volition, the choice to ro om with someone with Hepatitis may not have been a meaningful enough choice for participants to impute lower re lational value on the associatively-stigmatized target. If a future study were to use a more mean ingful life choice, voliti on may be more likely to affect associative stigma. As to why increased levels of attachment di d affect associative s tigmatization, but in the direction opposite of the predicted effect, any post hoc interpretations of this effect must necessarily be speculative, but this result st ill requires an attempt at explanation. The unanticipated beneficial result of the attachment manipulation on associative stigma may have 31

PAGE 32

been due to a feature of the manipulation itself. The emotional attachment of the target pair was manipulated by having the associatively-stigmati zed roommate report, in the high attachment condition, that he or she considered the primar ily-stigmatized roommate a friend because they were very close to each other (as opposed to re porting, in the low attach ment condition, that he or she did not consider the primarily-stigmatized roommate a friend and that they were not very close to each other). An unanticip ated effect of the attachment manipulation may also have been, besides creating a representation of a certain kind of relationship, to create a representation of a certain kind of relationship partner. An essay writer who described hi mself or herself as being in a close, friendly relationship, re gardless of the other person in th e relationship, may implicitly represent himself or herself as being friendly A friendlier-appearing person would also likely appear to be a more approachable person, resul ting in the decreased soci al distancing displayed by participants. This explanation does not address, however, why high attachment's beneficial effects for the essay writer were not outweighed by the hypothes ized increase in the s tigmatizing effects of Hepatitis. The most obvious explanation is th at, even though other research (Crandall, 1991; Munoz-Plaza at al., 2004) suggests that Hepatitis is stigmatizing, it may not have been sufficiently stigmatizing a disease to confer a ssociative stigma. A more stigmatizing attribute (e.g., HIV/AIDS, homosexuality) might have been necessary to offset the positive benefits derived from the targets appearance of frie ndliness. Future research might benefit from observing how the kind of stigmatizing attribute in teracts with association factors to stigmatize or destigmatize targets. Such value-laden stigma s as HIV/AIDS or homosexuality might also be mediated by assumptions other than those examined by this studyspecifically, participants who observe an associatively-stigmatized target vol untarily rooming with someone who possesses a 32

PAGE 33

stigma widely viewed as immoral (compared to participants who observe a target involuntarily assigned to room with such a stigmatized individu al) may assume that the target is himself or herself an immoral person, and may s tigmatize the target as a result. Mediation of Effects on Associative Stigmatization Although this effect of attachment on asso ciative stigmatization was in the direction opposite of that I had predicted, it was nonetheless partially mediated by the predicted mediator: the participant's assumptions concerning the associatively-stigmatized roommate's burdensomeness (i.e., relational oblig ations) as a relationship partne r. The effect was also fully mediated by an unpredicted mediatorthe participants' assumptions concerning the associatively-stigmatized roommate's value as a relationship partner. These results are challenging to interpret. An essay writer who descri bed himself or herself as being in a close, friendly relationship, regardless of the other person in the relationship, may implicitly represent himself or herself as being ab le and willing to enter into such relationships. In contrast, someone who self-represents as not being very close to, or friendly with, a relationship partner in his or her life may implicitly appear less willing or able to participate in such relationships. A willingness and an ability to participate in friendships implies a willingness and ability to carry out the res ponsibilities incumbent upon friends, su ch as the responsibility to come to a friend's aid when needed. Such relationship partners may potentially be seen as less burdensome than relationship partners who are not close friends, in that they help reduce, rather than increase, ones own life burdens. This may have led participants to see the friendlier roommate as conferring fewer re lational obligations on relationshi p partners, which may have in turn led to less social distancing (i.e., less stig matization). With regard to the other confirmed mediator of attachment's effect on associative st igmatization, relational value, a friendlier, more 33

PAGE 34

helpful relationship partner might also, naturally, be perceived as a more valuable relationship partner, which may in turn likewise lead to less social distancing on the part of the participant. An alternative, less tenuous explanation may simply be that the in creased perception of friendliness elicited by the higher attachment, and the resultant increased attractiveness, may have favorably biased particip ants perceptions of the associatively-stigmatized roommates relational value and relational obligations in a manner akin to a halo effect. However, this interpretation is inconsistent with the fact the lack of attachments effect on participants perceptions of the associatively-stigmatized r oommates healthiness. If higher attachment led participants to more favorably evaluate the targ ets traits, one would e xpect that participants would attribute better health to the targets in the high attach ment condition. This, however, was not the case. Effects on Primary Stigmatization Attachment had the same effect on primary stigmatization as it did on associative stigmatization in that higher attachment conferred less stigma on the primarily-stigmatized roommate. Attachment may have had this effect on primary stigmatization for the same reason it had the same effect on associative-stigma tization: The high attachment condition, by representing the primarily-stigmatized roommate as being in a close, fr iendly relationship, may have made the primarily-stigmatized roommate appear to be a less burdensome and more valuable relationship partner, leading participants to engage in less social distancing. Interestingly, even though volition and contac t frequency did not exert any main effects on either primary or associative stig matization, they did interact with each other in affecting primary stigmatization. The essay writer's choice to room with the primar ily stigmatized roommate did have an effect on primary stigmatization, such that higher voliti on resulted in lower 34

PAGE 35

stigmatization, but only when the two roommates spent a great deal of time in each other's company. When contact was infrequent, there was no effect of volition. Though this interpretation is highly speculative, it seems that, when the essay writer voluntarily chose to live with the primarily-s tigmatized roommate, participants may have inferred from this decision that the primarily-stigmatized roommate must have had some appealing qualities (or else why choose to live with him or her?) leading the participants to socially distance themselves from the primarily -stigmatized roommate to a lesser degree than they would have if the roommate had been assi gned to them. However, this decision to room with the primarily-stigmatized roommate would lik ely be much more consequential if the essay writer spent a great deal of tim e in his or her roommate's company, because a poor choice would lead to the unfortunate result of spending a gr eat deal of time in the company of someone annoying or unpleasant. If the essay writer chose to room with someone they would spend a lot of time with, then participants may have inferred that the person they chose to room with must be especially likeable, if the essay writer was so wi lling to put himself or he rself into a situation where he or she would be so fre quently in that person's company. On the other hand, in the case of participants in the low contact condition, the decision either to room or not to room with the primar ily-stigmatized roommate may be seen as a much less consequence-laden choice, as the essay writer may not have expected to spend much time in the primarily-stigmatized roommates company. Participants may infer that the essay writer, for that reason, did not take the primarily-stigmatiz ed roommates likeability into consideration when making this decision, making this choice le ss informative about the primarily-stigmatized roommates desirability as a room mate or relationship partner. This may have led participants to 35

PAGE 36

weigh the essay writers volition less when ev aluating the primarily-stigmatized roommates approachability, reducing the effect of voliti on on consequent primary stigmatization. Again, these highly speculative interpretations of the interacting effects of volition and contact should be further tested before given any acceptance. Implications for Research on Associative Stigma In their study of associative stigma, Neuberg et al. (1994) expected that the stigma-byassociation process would yield one of four possible outcomes: (1) The stigmatized, "marked" individual would become destigmatized through his or her association with a "normal" other (a "lifting" effect), (2) th e "normal" would become stigmatized through his or her association with the marked individual, (3) a cont rast effect would cause the no rmal being less stigmatized and the marked individual being more stigmatized as a result of the associatio n, or (4) the association would have no effect on the level of stigmatizati on of either the marked or normal individuals. To discover which outcome would re sult, participants viewed an in teraction between two targets, one of which was presented as either heterose xual or homosexual. As the present study was interested in how characteristics of the relationship (and not characteristics of the members of the relationship) affected associ ative stigma, such ma nipulations were outside the present study's scope, and it is impossible to say which, if any, of those four proposed outcomes occurred here. However, it is interesting to note that the present study's manipulati on of one aspect of the targets' relationshipattachmentresulted in an outcome that did not map onto any of the four predicted by Neuberg et al.: Both targets were destigmatized. It is en couraging to see that a marked individual may enjoy a lifting effect from associating with a normal without necessarily exacting an irrepa rable toll on the social accepta bility of either relationship participant, and that any associative stigmatizati on of the normal, if it occurs, may be offset by a 36

PAGE 37

feature of the very relationshi p that, theoretically, is causing the normal to be stigmatized. Associative stigma may not be a zero-sum dilemma. Like the present study, Sigelman et al. ( 1991) experimentally manipulated whether a target living with a stigmatized (homosexual, in this case) roommate had voluntarily chosen that living arrangement. In addition, Sigelman et al. also assessed partic ipants' tolerance or intolerance of homosexuals. They found that, when participants ra ted as being high in intolerance for homosexuals were provided in formation about the target living with a homosexual roommate, targets who were depicted as having voluntar ily chosen to room with a gay student were perceived as being more homosexual and as having more stereotypically homosexual traits than targets w ho were assigned to live with a gay roommate. More relevant to comparisons to the present study, intolerant partic ipants also liked targ ets who voluntarily chose to live with a gay roommate less than they liked targets who were assigned to live with a gay roommate. Although homosexuality is very different from the stigmatizing attribute I employed (Hepatitis), it is still worth contrasting the eff ects of volition in their study to its effects in my own. While their experimental increase in voliti on resulted in greater st igmatization and greater attribution of stigma-relevant traits to th e associatively stigmatized roommate, my own manipulation of that same variable did not have any effects on stig matization of, or attribution of stigma-relevant traits to, associatively-stigmatiz ed targets. This discrepancy may be due to differing valence and stereotypic co ntent of the two stigmatizing attr ibutes, but it is still worth noting that my own findings may represent an exception to the e ffects of voliti on on associative stigmatization. I suspect that one of the gr eatest challenges stigmatized indi viduals are likely to face (or likely to believe they face) as a result of their stigmatization is an increased difficulty in forming 37

PAGE 38

deep, meaningful, mutually healthy and beneficial relationships with other human beings as a result of others' fears of being stigmatized by their associations with them. Given the present findings that a greater emotional connection may actually have a benefici al result for both the associatively-stigmatized and primarily stigmatiz ed relationship partners, and assuming for the moment that my post hoc interpre tations of these results are va lid, the implications of these findings may actually bear considerable relevance to the area of impression management specifically, beneficial impressi on management. Human beings are constantly managing the impressions they make on other people in strate gic, motivated ways, a behavior referred to as self-presentation (Schlenker, 1980). Additionally, people also make strategic presentations of individuals with whom they are closely and positively affiliated, working to present their friends in ways that are most likely to benefit their friends (Schlenker & Br itt, 1999). Beneficial impression management is a highly under-resea rched area, and to my knowledge, no extant research on beneficial impressi on management (Schlenker & Brit t, 1999; Schlenker & Britt, 2001; Pontari & Schlenker, 2004; Schlenker, Lifk a, & Wowra, 2004; Pontari & Schlenker, 2006) experimentally examined how beneficial im pression management might be successfully achieved by means of strategic presentations of th e presenter's relationship with the beneficiary instead of (or in addition to) strategic presentations of the beneficiary's personal qualities and virtues. The mutually advantageous results of such a strategy, seemingly implied by the present study's results, would do well to be examined by further research aimed at corroborating its benefits, and, if real, could be beneficially applied by associatively-stigmatized individuals who find themselves in situations where they ar e either unable or unwilling to conceal their association with a primarily-stigmatized relationship partner and who wish to make the best of a bad situation that, in the final an alysis, perhaps needn't be so bad. 38

PAGE 39

Conclusion and Future Directions This is not to say that associative stigma does not pose a considerable problem for many people, primarilyand associatively-stigmatized a like, or that the way different relational factors may exacerbate this problem is yet understood by research into its processes. If the lessons learned from this study are to prove at all useful, it will most likely be in their ability to inform future research on this subject. To that end, I believe future research would benefit by focusing exclusively on two relationa l factorsattachment (given its e ffects here) and volition (given its effects verified by Sigelman et al., 1991)as extant research seems to indicate that these are the relational factors that seem to influence asso ciative stigmatization most reliably. Additionally, future research should manipulate the presence of the stigmatizing attribute, as well as the nature of the attribute itself (e.g., He patitis, HIV/AIDS, or homosexuali ty), to see how these factors interact with the associative f actors under examination. It may be that more stereotypically immoral stigmas may actually confer greater asso ciative stigma when attachment and volition are higher. Finally, given the fi nding that the relational factors did influence primary stigma, future studies intended to further scrutinize how relational factors affect stigmatization of both participants in the relationship would benefit from a more symmetrical presentation of the targets, as opposed to the present studys more lopsided presentati on of one particular target as the source of all the participant s information about the relationship of the targets to each other. In sum, though higher attachment had a bene ficial, instead of a deleterious, effect on associative stigmatizatio n, contrary to what I had hypothesize d, and though the implications of this effect require further empirical confirmation before they can be accepted with any confidence, this finding may still shed light on ho w different features of a relationship with a primarily-stigmatized individual can influence peoples perceptions and stigmatization of both members of such a relationship. Th e counter-hypothetical (yet iron ically heartening) finding that 39

PAGE 40

40 higher reported attachment had a lifting effect on participants stigmatization of the primarilystigmatized target, but at no cost (and, indee d, with some improvement) to the associativelystigmatized writers own social approachability, may be good news to those who find themselves in such relationships and are uncertain how to portray these rela tionships to others.

PAGE 41

APPENDIX A EXPERIMENTAL MATERIALS [Experimental manipulations are bracketed. Ge nder pronouns were matched to the gender of the participant.] First of all, thank you for helping us with this study. Your assistance is extremely vital to our efforts to learn more about people and their re lationships with each other. The focus of this study is on pairs of University of Florida students formerly residing in one-room, two-person dormitory units on campus. We are studying thes e roommates because we are interested in characteristics of two-person relationships, a nd roommate pairs presen t the most convenient example of these kinds of relationships. Last ye ar, we recruited a number of roommates from the University of Florida student body, and asked th em to provide information about their living situations. This is where you come in. During your time in this session today, you will be acting as a rater. You will read a statement that has been provided by one of the roommates who participated in our study. These participants were asked to write a short essay about themselves, their roommates, and their relati onships with their roommates. In order to protect the anonymity of the people participating in th is study, the roommates will not be identified by name. Instead, they will be referred to as X and Y. You will be considering the essay written by roommate Y. Because the information you will be providing to us will be essential to our research goals, we wish to stress to you how impor tant it is that you r ead the essay closely and that you provide only your most honest responses in the questionnaire. Keep in mi nd that your identity is also being protected. Neither your name, nor any ot her information that can identify you, will be attached to your responses, so please answer the questions as honestly as you can. 41

PAGE 42

Thank you again for your assistance in this study. When youre ready to proceed, please go to the next page. 42

PAGE 43

As you just read, the roommate s participating in this study were asked to write a short essay. Here are the exact instructions these participants were given: What we would like you to do now is writ e a very short essay (between 10 and 20 sentences in length) telling us a little about yourself, your roommate, and your relationship with each other. This essay is ve ry open-ended, and you can give us any information that you personally think is interesting or relevant. Rememb er that your responses will be kept completely confidential, so be as honest and detailed as you possibly can without going over the 20 sentence limit. One of the statements written by one of the pa rticipants is provided below. Please read it carefully, and then, when you feel ready to proceed, go to the next page. This is my first year at the University of Fl orida. Im originally from Tampa, Florida, and Im nineteen years old. Im majoring in Anthropo logy and am so far enjoying my time here, but Im thinking about switching majors. Stuff I do for fun: I like all kinds of music. When I watch TV, I usually like Comedy Central. I love dogs but I had to leave mine at home since the dorms dont allow them. Ive been to a couple of football games and I usually go out one or two times a week, sometimes more. My roommate is from Jacksonville. One intere sting thing about her is that she's the only person I know with Hepatitis. (It's contagious but I'm not sure how she caught it.) There's no cure and she has some bad health problems. She sp ends a lot of time online and is studying Linguistics. She likes reading and she says she do esn't know what she wants to do when she gets out of college. 43

PAGE 44

44 This is the first time weve shared a dorm. [We were assigned as room mates, I didnt actually choose to live with her. / We werent a ssigned as room mates, I actually chose to live with her.] Because of the way our work and cl ass schedules overlap with each other (we both have part time jobs) [were never around each othe r in the room, almost none of the time. / were always around each other in the room, almost all the time.] All in all, es pecially after rooming with her this long, I would have to say that [I dont really consider her a friend, because we arent very close to each other. / I really consider her a friend, be cause were very close to each other.] Thats all I can think of to say, Im not sure what else to write, I hope this is what you needed.

PAGE 45

APPENDIX B STIGMATIZATION OF Y (ASSOCIATI VELY-STIGMATIZED ROOMMATE) The following questions inquire about Y (the writer of the essay). Indicate the degree to whic h you agree or disagree with the following statements: 1. Y appears to be a likeable person. Strongly disagree Strongly agree 1 2 3 4 5 6 7 2. I would like Y to be a close personal friend. Strongly disagree Strongly agree 1 2 3 4 5 6 7 3. I wouldn't mind it at all for Y to move into my neighborhood. Strongly disagree Strongly agree 1 2 3 4 5 6 7 4. I would like Y to come and work at the same place I do. Strongly disagree Strongly agree 1 2 3 4 5 6 7 45

PAGE 46

46 5. Y is a person who is similar to me. Strongly disagree Strongly agree 1 2 3 4 5 6 7 6. I would like to have Y marry into my family. Strongly disagree Strongly agree 1 2 3 4 5 6 7 7. Y is the kind of person that I tend to avoid. Strongly disagree Strongly agree 1 2 3 4 5 6 7

PAGE 47

APPENDIX C STIGMATIZATION OF X (PRIMARI LY-STIGMATIZED ROOMMATE) The following questions inquire about X (t he roommate of th e essays writer). Indicate the degree to whic h you agree or disagree with the following statements: X appears to be a likeable person. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like X to be a close personal friend. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I wouldn't mind it at all for X to move into my neighborhood. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like X to come and work at the same place I do. Strongly disagree Strongly agree 1 2 3 4 5 6 7 47

PAGE 48

48 X is a person who is similar to me. Strongly disagree Strongly agree 1 2 3 4 5 6 7 I would like to have X marry into my family. Strongly disagree Strongly agree 1 2 3 4 5 6 7 X is the kind of person that I tend to avoid. Strongly disagree Strongly agree 1 2 3 4 5 6 7

PAGE 49

APPENDIX D ASSUMPTION OF YS RELATIONAL VALUE The following questions inquire ab out Y (the writer of the essay). 1. How many close friends do you think Y (the writer of the essay) has? Very few Very many 1 2 3 4 5 6 7 2. To what extent would the averag e UF student want Y as a friend? Not at all Very much 1 2 3 4 5 6 7 3. To what extent do you think people who know Y value their relationships with him/her? Not at all Very much 1 2 3 4 5 6 7 Put yourself in the shoes of people who know Y well. How do you think people who know Y would rate their relationship with him/her on the five scales below? 4. Unimportant Important 1 2 3 4 5 6 7 49

PAGE 50

50 5. Valuable Not valuable 1 2 3 4 5 6 7 6. Worthless Worthwhile 1 2 3 4 5 6 7 7. Good Bad 1 2 3 4 5 6 7

PAGE 51

APPENDIX E ASSUMPTION OF Y'S RELATIONAL ESTEEM Indicate the degree to whic h you agree or disagree with the following statements: 1. Y is a person of worth, at least on an equal plane with others. Strongly disagree Strongly agree 1 2 3 4 5 6 7 2. Y has a number of good qualities. Strongly disagree Strongly agree 1 2 3 4 5 6 7 3. All in all, Y is a failure. Strongly disagree Strongly agree 1 2 3 4 5 6 7 4. Y is able to do things as well as most other people. Strongly disagree Strongly agree 1 2 3 4 5 6 7 5. Y does not have much to be proud of. 51

PAGE 52

Strongly disagree Strongly agree 1 2 3 4 5 6 7 6. Y takes a positive attitude toward himself/herself. Strongly disagree Strongly agree 1 2 3 4 5 6 7 7. On the whole, Y is satisfied with himself/herself. Strongly disagree Strongly agree 1 2 3 4 5 6 7 8. Y wishes he/she could have mo re respect for himself/herself. Strongly disagree Strongly agree 1 2 3 4 5 6 7 9. Y certainly feels useless at times. Strongly disagree Strongly agree 1 2 3 4 5 6 7 10. At times Y thinks he/she is no good at all. 52

PAGE 53

53 Strongly disagree Strongly agree 1 2 3 4 5 6 7

PAGE 54

APPENDIX F ASSUMPTION OF YS POSSESSION OF STIGMATIZING ATTRIBUTE The following questions inquire ab out Y (the writer of the essay). How likely do you think it is that Y (the writer of the essay) experiences each of the following? 1. General feelings of poor health Very unlikely Very likely 1 2 3 4 5 6 7 2. Difficulty sleeping Very unlikely Very likely 1 2 3 4 5 6 7 3. Nausea and/or vomiting Very unlikely Very likely 1 2 3 4 5 6 7 4. Light headedness and/or fainting Very unlikely Very likely 1 2 3 4 5 6 7 54

PAGE 55

5. Headaches Very unlikely Very likely 1 2 3 4 5 6 7 6. Susceptibility to infections Very unlikely Very likely 1 2 3 4 5 6 7 7. Weakness Very unlikely Very likely 1 2 3 4 5 6 7 8. Exhaustion from minor exertion Very unlikely Very likely 1 2 3 4 5 6 7 9. Dizziness Very unlikely Very likely 1 2 3 4 5 6 7 55

PAGE 56

56 10. Unhealthy body weight (either overweight or underweight) Very unlikely Very likely 1 2 3 4 5 6 7

PAGE 57

APPENDIX G ASSUMPTION OF YS R ELATIONAL OBLIGATIONS The following questions inquire abou t Y (the writer of the essay). Put yourself in the shoes of people who know Y (the writer of the essay) well. Consider what you think their relationship with Y is like. For the following questions, indicate what you think a relationship with Y is like. 1. How much time do you think Ys friends must dedicate to helping Y deal with his/her problems? None at all Very much 1 2 3 4 5 6 7 2. How much energy do you think Ys friends must dedicate to helping Y deal with his/her problems? None at all Very much 1 2 3 4 5 6 7 3. How much of their resources do you think Ys fr iends must dedicate to helping Y deal with his/her problems? None at all Very much 1 2 3 4 5 6 7 57

PAGE 58

4. How burdensome do you think Ys friends find their relationship with Y? Not at all Very much 1 2 3 4 5 6 7 5. How inconvenient do you think Ys friends find their relationship with Y? Not at all Very much 1 2 3 4 5 6 7 6. How demanding do you think Ys friends find their relationship with Y? Not at all Very much 1 2 3 4 5 6 7 7. How problematic do you think Ys friends finds their relationship with Y? Not at all Very much 1 2 3 4 5 6 7 8. How much assistance do you think Y requires from his/her friends? None at all Very much 1 2 3 4 5 6 7 58

PAGE 59

Put yourself in the shoes of people who know Y well. Consider what you think their relationship with Y is like. On each of the te n scales below, indicate what you think their experience of the relationship with Y is like, from their point of view: People who know Y well find their relationship with Y. 9. Refreshing Draining 1 2 3 4 5 6 7 10. Rewarding Punishing 1 2 3 4 5 6 7 11. Low maintenance High maintenance 1 2 3 4 5 6 7 12. Overwhelming Manageable 1 2 3 4 5 6 7 59

PAGE 60

60 13. Energizing Exhausting 1 2 3 4 5 6 7

PAGE 61

APPENDIX H MOTIVATION TO CONTROL PREJUDICE 1. In today's society it is important that one not be perceived as prejud iced in any manner. Strongly disagree Strongly agree 1 2 3 4 5 6 7 2. I always express my thoughts and feelings, re gardless of how controversial they might be. Strongly disagree Strongly agree 1 2 3 4 5 6 7 3. I get angry with myself when I have a thought or feeling that might be considered prejudiced. Strongly disagree Strongly agree 1 2 3 4 5 6 7 4. If I were participating in a class discussion and a student who was different from people like me expressed an opinion with which I disagr eed, I would be hesitant to express my own viewpoint. Strongly disagree Strongly agree 1 2 3 4 5 6 7 61

PAGE 62

5. Going through life worrying about whether you might offend someone is just more trouble than it's worth. Strongly disagree Strongly agree 1 2 3 4 5 6 7 6. It's important to me that othe r people not think I'm prejudiced. Strongly disagree Strongly agree 1 2 3 4 5 6 7 7. I feel it's important to behave according to society's standards. Strongly disagree Strongly agree 1 2 3 4 5 6 7 8. I'm careful not to offend my friends, but I don 't worry about offending people I don't know or don't like. Strongly disagree Strongly agree 1 2 3 4 5 6 7 62

PAGE 63

9. I think that it is important to speak one's mi nd rather than to worry about offending someone. Strongly disagree Strongly agree 1 2 3 4 5 6 7 10. It's never acceptable to express one's prejudices. Strongly disagree Strongly agree 1 2 3 4 5 6 7 11. I feel guilty when I have a negative thought or feeling about someone different from people like me. Strongly disagree Strongly agree 1 2 3 4 5 6 7 12. When speaking to someone different from people like me, it's important to me that he/she not think I'm prejudiced. Strongly disagree Strongly agree 1 2 3 4 5 6 7 63

PAGE 64

13. It bothers me a great deal when I think I' ve offended someone, so I'm always careful to consider other people's feelings. Strongly disagree Strongly agree 1 2 3 4 5 6 7 14. If I have a prejudiced thought or feeling, I keep it to myself. Strongly disagree Strongly agree 1 2 3 4 5 6 7 15. I would never tell jokes that might offend others. Strongly disagree Strongly agree 1 2 3 4 5 6 7 16. I'm not afraid to tell others what I thi nk, even when I know they disagree with me. Strongly disagree Strongly agree 1 2 3 4 5 6 7 64

PAGE 65

65 17. If someone who made me uncomfortable sat next to me on a bus, I would not hesitate to move to another seat. Strongly disagree Strongly agree 1 2 3 4 5 6 7

PAGE 66

APPENDIX I MANIPULATION CHECKS In the short statement that you read, Y provi ded information about his/her self, his/her roommate, and his/her relationship with the room mate. Wed like you to answer some questions about that information. Even if the essay you read didnt have any information relevant to these questions, please answer them to the best of your ability: How much of Ys total free time doe s he/she spend in Xs company? Very little Very much 1 2 3 4 5 6 7 How much would you say Ys rela tionship with X is voluntary? Not at all voluntary Very voluntary 1 2 3 4 5 6 7 How much would you say Y likes X? Not at all Very much 1 2 3 4 5 6 7 In the space below, please brie fly describe the information th at was provided in the essay that you read: 66

PAGE 67

APPENDIX J DEMOGRAPHICS Please answer the following questions. As mentioned before, your responses are confidential. These demographic qu estions only serve to help us explore the sample helping us with our study. What is your gender? ______ Male ______ Female How old are you? ______ years Which of the following ethnic group(s) do you c onsider yourself a member of? You can check multiple groups. ______ African American ______ Asian Americans ______ Hispanic ______ Native American ______ Caucasian/White ______ Other: _________________________________________ 67

PAGE 68

68 Which of the following religi ons best describes you? ______ Protestant (Methodist, Lutheran, Episcopalian, etc.) ______ Catholic ______ Fundamentalist/Evangelical Christian ______ Jewish ______ Muslim ______ Hindu ______ Buddhist ______ Atheist ______ Other: _________________________________________ ______ No religious affiliation

PAGE 69

LIST OF REFERENCES Baron, R. M., & Kenny, D. A. (1986). The m oderator-mediator dist inction in social psychological research: Con ceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51 1173. Bloom, J. R., & Kessler, L. (1994). Emotional support following cancer: A test of the stigma and social activity hypothesis. Journal of Health and Social Behavior, 35, 118. Bogardus, E. S. (1923). Immigration and race attitudes. Boston: D.C. Heath. Brigham, J. C. (1974). Views of black and white children concerni ng the distribution of personality characteristics. Journal of Personality, 42, 144. Corrigan, P. W., Miller, F. E. (2004). Shame, blame, and contamination: A review of the impact of mental illness stigma on family members. Journal of Mental Health, 13, 537. Crandall, C. S. (1991). Multiple Stigma and AIDS: Illness stigma and attitudes toward homosexuals and IV drug users in AIDS-related stigmatization. Journal of Community & Applied Social Psychology, 1, 165. Crocker, J., Cornwell, B., & Major, B. (1993). The stigma of overweight: Affective consequences of attributional ambiguity. Journal of Personality and Social Psychology, 64, 60 70. Dunton, B. C. & Fazio, R. H. (1997). An indi vidual difference measure of motivation to ontrol prejudiced reactions. Personality & Social Psychology Bulletin, 23 316. Farina, A. (1982). The stigma of mental disorders. In A. G. Miller (Ed.), In the eye of the beholder: Contemporary approaches to stereotyping (pp. 305). New York: Praeger. Farina, A., Sherman, M., & Allen, J. G. ( 1968). The role of physical abnormalities in interpersonal perception and behavior. Journal of Abnormal Psychology, 73, 590. Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ: Prentice Hall. Goldstein, S. B., & Johnson, V. A. (1997). Stigma by association: Perceptions of the dating partners of college stude nts with physical disabilities. Basic and Applied Social Psychology, 19, 495. Herek, G. M. (1984). Attitudes toward lesbia ns and gay men: A factor-analytic study. Journal of Homosexuality, 10, 39. Jones, E. E., Farina, A., Hastorf, A. H., Mar kus, H., Miller, D. T., & Scott, R. A. (1984). Social stigma: The psychology of marked relationships. New York: Freeman. 69

PAGE 70

Mehta, S. I., & Farina, A. (1988). Associativ e stigma: Perceptions of the difficulties of college-aged children of stigmatized fathers. Journal of Social and Clinical Psychology, 7, 192 202. Munoz-Plaza, C. E., Strauss, S. M., Astone, J. M., Des Jarlais, D. C., Hagan, H. (2004). Drug treatment programs as sites of opportunity for delivery of He patitis C prevention education: Client and staff perspectives. Journal of Drug Issues, 34, 861. Neuberg, S. L., Smith, D. M., Hoffman, J. C., & Russell, F. J. (1994). When we observe stigmatized and normal individuals interacting: Stigma by association. Personality and Social Psychology Bulletin, 20, 196. Park, J. H., Faulkner, J., & Schaller, M. ( 2003). Evolved disease-avoidance processes and contemporary anti-social behavior: Prejudicia l attitudes and avoida nce of people with disabilities. Journal of Nonverbal Behavior, 27, 65. Pontari, B. A., & Schlenker, B. R. (2001). Providing and withholding impression management support for romantic part ners: Gender of th e audience matters. Journal of Experimental Social Psychology, 40 41. Pontari, B. A., & Schlenker, B. R. (2006). Helping friends manage impressions: We like helpful liars but respect nonhelpful truth tellers. Basic and Applied Social Psychology, 28 177 183. Rosenberg, M. (1989). Society and the Adolescent Self-Image. Revised edition. Middletown Cliffs, NJ: Wesleyan University Press. Schlenker, B. R. (1980). Impression Management: The Selfconcept, Social Identity, and Interpersonal Relations. Monterey, CA: Brooks/Cole Publishing Company. Schlenker, B. R., & Britt, T. W. (1999). Be neficial impression management: Strategically controlling information to help friends. Journal of Personality and Social Psychology, 76 559 573. Schlenker, B. R., & Britt, T. W. (2001). St rategically controlling information to help friends: Effects of empathy and friendship st rength on beneficial impression management. Journal of Experiment al Social Psychology, 37 357. Schlenker, B. R., Lifka, A., & Wowra, S. A. (2004). Helping new acquaintances make the right impression: Balancing imag e concerns of others and self. Self and Identity, 3 191. Scott, R. A. (1969). The making of blind men: A st udy of adult socialization. New York: Russell Sage Foundation. Shears, L. M., & Jensema, C. J. (1969). So cial acceptability of anomalous persons. Exceptional Children, 36, 91. 70

PAGE 71

71 Sigelman, C. K., Howell, J. L., Cornell, D. P., Cutright, J. D., & Dewey, J. C. (1991). Courtesy stigma: The social implicati ons of associating with a gay person. Journal of Social Psychology, 131, 45. Sobel, M.E. (1982). Asymptotic confidence inte rvals for indirect e ffects in structural equation models. In S. Leinhardt (ed.), Sociological Methodology 1982 (pp. 290). Washington DC: American So ciological Association. Weitz, R. (1990). Living with the stigma of AIDS. Qualitative Sociology, 13, 23.

PAGE 72

BIOGRAPHICAL SKETCH David Richards was born in Miami. After attending Jesuit High School of Tampa, he earned a B.A. in English at the University of Florida, and a B.A. in psychology, with honors, at the University of South Florida, Tampa, before re turning to the University of Florida to pursue a graduate education in social psychology. 72