THESIS

llllllilllllllllllllllllllllllllllllllllllllllllUlllUJlll

3 1293 01691 8959

' This is to certify that the

dissertation entitled

HOW TIGHTLY IS THE KNOT TIED?: SPONTANEOUS
USE OF RELATIONSHIP CATEGORIES FOR INTER-
RACIAL MARRIAGES

presented by

Donna Annemarie Lewandowski

has been accepted towards fulfillment
of the requirements for

Ph. D- degree in _E_S¥Ch0109¥_

 

-’b\/‘tfl ' 1"} 4/ ‘— (if/visz
L/ b

Major professor

y.

Date December 16, 1997

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY
University

 

Michigan State

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

MTE DUE 1 DATE DUE

DATE DUE

 

”Ec1oz7oziol 4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M wmms-p.“

 

I-Iow TIGHTLY IS THE KNOT TIED?: SPONTANEOUS USE
OF RELATIONSHIP CATEGORIES FOR
INTER-RACIAL VERSUS lNTRA-RACIAL MARRIAGES
By

Donna Annemarie Lewandowski

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment Of the requirements
for the degree Of

DOCTOR OF PHILOSOPHY

Department Of Psychology

1 997

ABSTRACT
How TIGHTLY IS THE KNOT TIED?: SPONTANEOUS USE
OF RELATIONSHIP CATEGORIES FOR INTER-RACIAL
VERSUS lNTRA-RACIAL MARRIAGES
By

Donna Annemarie Lewandowski

The incidence Of interracial marriage between Blacks and Whites has grown
from 65,000 in 1970 to 322,000 in 1995 (US. Census Bureau, 1970, 1995). The
current study sought to determine whether information is spontaneously
organized in memory around the marital relationship for inter-racial couples as it
is for intra-racial couples (Sedikides, Olsen, 8. Reis, 1993). It was predicted that
it would not be. Couples matched or mismatched on occupation were included
since it was predicted that dissimilarities other than race might interfere with a
couple being viewed as a social unit. One hundred and eighty participants were
randomly assigned to one Of the cells Of a 2 (attribute: race or occupation) by 2
(relationship: matched on attribute vs mismatched on attribute) between-subjects
design. Participants received information about 8 target individuals who
comprised 4 couples. Their task was to memorize the information presented
about each target. After timed exposure to each target, participants completed
recall and recognition tasks, and belief similarity measures. Partial support for
the hypothesis that mismatched couples would be less likely to be perceived as
a social unit than matched couples was found for within couple and between-

couple errors on recognition tasks. However, responses to the free-recall

measure did not show greater spontaneous use Of the marital relationship
category for matched compared to mismatched couples. Belief similarity
measures suggested that couples matched on race or occupation were
perceived as more similar on a variety of dimensions (e.g., values, morals,
attitudes) compared to mismatched couples as expected. Since the same
pattern emerged for couples mismatched on occupation as for those mismatched
on race, findings suggest that it is not race per se but rather a lack Of perceived
similarity that discourages the perception Of couples as a unit. Implications Of the
lack of perceived social unity Of a couple by outsiders for the success of inter-

racial marriage are discussed.

This work is dedicated to my dearest one, Anne T. Lewandowski

ACKNOWLEDGMENTS

The completion of my Ph. D. would not have been possible without the
considerable guidance and support Of my committee members. I am fortunate tO
have benefited from their many (combined) years of experience.

TO Norbert Kerr and Larry Messé whose willingness over the years to
discuss any aspect Of graduate training (or life in general) has been invaluable.

l have the greatest respect for both Of you and greatly appreciate all Of your time
and efforts. To Rick DeShon to whom I can directly attribute my interest and
enjoyment Of statistics particularly as it relates to teaching. You were an
outstanding role model. Thank you for being so generous with your time.

TO Linda Jackson, who has fulfilled a variety of roles over the last five
years. I suspect that few students are fortunate enough to have the kind of
advisor-advisee relationship that we have shared. It’s been an exciting and
interesting metamorphosis from new graduate student to new colleague. My
most heart felt appreciation and gratitude goes to you.

Finally, to my dearest grandmother. \NIthout your unconditional love and
encouragement, I would never have attempted this endeavor nor been able to

complete it. I miss you.

TABLE OF CONTENTS

LIST OF TABLES ....................................... vii
INTRODUCTION ........................................ 1
REVIEW OF THE LITERATURE ............................ 1

I. Marriage as a socially potent relationship ................... 1
II. Balance theory ........................................ 3
III. Utilizing relationships in cognitive organization ............... 6
IV. Utilizing marital relationships in cognitive organization ....... 10
IV. Utilizing physical characteristics in cognitive organization . . . . 12
V. Hypotheses ....................................... . 15
METHODS ............................................ 1 7
l. Pretest Of stimulus materials ............................ 17
ll. Participants ......................................... 18
Ill. Design ............................................. 18
IV. Materials ........................................... 18
V. Dependent measures ................................. 19
VI. Procedure .......................................... 21
RESULTS ............................................. 23
I. Preliminary analyses .................................. 23
ll. Tests Of hypotheses .................................. 24
Ill. Additional analyses .................................. 28
DISCUSSION .......................................... 29
l. Interpretation of results ................................ 29
ll. Future research ...................................... 34
III. Conclusions ......................................... 36
APPENDICES .......................................... 38
LIST OF REFERENCES .................................. 65

LIST OF TABLES

Table 1 - Means and standard deviations for all dependent measures

by condition ...................................... 61
Table 2 - Means and standard deviations for all dependent measures

by type .......................................... 62
Table 3 - Means and standard deviations for all dependent measures

by subject sex .................................... 62
Table 4 - Means and standard deviations for all error types ........ 63

Table 5 - Means and standard deviations for all dependent measures
by attribute ....................................... 64

lNIBQQLLCIIQN

The importance Of interpersonal relationships can be seen in the central
position they hold in many social psychological theories (Heider, 1958; Kelley,
1979; Kelley 8 Thibaut, 1978; McCall, McCall, Denzin, Suttles, & Kurth, 1970).
Kelley & Thibaut’s (1979) theory Of interdependence, Heider’s (1958) unit
relationships theory, and McCall et al’s (1970) social bonds conception are only
a few of the theories that focus on social relationships. Theory and research
have suggested that relationships often function as an organizing tool in social
perception (Bond & Brockett, 1987; Fiske, Haslam, & Fiske, 1991; F iske &
Haslam, 1996; Sedikides, Olson, & Reis, 1993).

M' EIIS'IBII' l'

The marital alliance is considered tO be one of the most potent and
primary adult relationships. Research has demonstrated that this particular type
of relationship can function as a means Of cognitively organizing information in
memory (Sedikides, Olsen, & Reis, 1993). The present research is an attempt to
determine whether inter-racial couples are viewed as a “social unit” in the same
way that intra-racial couples are. In other words, is information cognitively
organized spontaneously in memory around the marital relationship for those
involved in mixed-race marriages as it is for same-race marriages?

The incidence Of inter-racial marriage between Blacks and Whites has
more than tripled from 65,000 in 1970 to 322,000 in 1995 (US Census,
1970,1995). Despite this increase, a Gallop Poll conducted in 1994 showed that

only 48% Of Americans approve of inter-racial marriage and 37% Openly oppose

2

it. In a 1995 Gallop Poll, 1 in 5 Americans stated that they would favor a law
prohibiting such unions (Gallop, 1995). If the current trend persists, inter-racial
marriages will continue to increase in number despite some enduring biases
against such unions (Schoen 8 WOOIdredge, 1989). Thus, how people who
marry a person Of another race are perceived, our attitudes toward them, and
our willingness (or unwillingness) to accept inter-racial marriage among family
and friends are issues that an increasingly larger portion Of individuals may have
to consider in the future.

Among numerous types of personal affiliations, marriage is in many
cultures the strongest of all mature relationships. The larger social world tends
to treat couples as a unit by the social considerations it extends to them.
Married people are invited to events as couples rather than as individuals. Often
rights, privileges, courtesies, and opportunities are extended to mates in
recognition Of their close relationship (McCall 8 Simmons, 1970).

Marriage is considered a socially defined and regulated alliance in all
human societies (Argyle 8 Henderson, 1985). Generally speaking, there are
and always have been strong social norms governing the selection Of marital
partners (Kerckhoff, 1974). We usually choose partners who are Of the same
race, religion, social class, educational background, and economic status as
ourselves (Kerckhoff, 1974; Rubin, 1973; Udry, 1971). While some departures
have become more acceptable over time (e.g., Italians marrying Poles), there
are still very strong social norms that favor selecting same-race partners

(Klemer, 1970; Rosenblatt, Karis, 8 Powell, 1995; Staples, 1992).

3

Rules governing matrimony are supported by religious, legal, moral, and
cultural systems. Engagements and marriages generally occur with some type
of formal procedure (civil and/or religious) and are Often attended by friends and
family who confirm that some change has taken place (Argyle 8 Henderson,
1985). Engaging in these practices results in a concurrence between the couple
and society that a long term partnership has been formed (Arger 8 Henderson,
1985). Thus, awareness of the members as a unit reaches well beyond the
couple themselves.

Balancelbeenr

While the balance theories of Heider (1958) and Newcomb (1961)
suggest that mere association is enough to produce the perception Of a unit
relationship, more recent work contends that the frequency, intensity, diversity,
and duration of social interaction (Kelley et al, 1983) or perceptions of entitativity
or unity (Campbell, 1958; Hamilton 8 Sherman, 1996) may be responsible for
creating perceptions Of closeness.

Heider (1958) speaks Of unit formation and balanced states by suggesting
that separate entities comprise a unit when they are perceived as belonging
together in an especially close way. Conditions that can lead to unit formation
include proximity, similarity, interaction, and common fate. Gestalt psychologists
have demonstrated that unit formation is an important part Of cognitive
organization (Heider, 1958). Further, there is an awareness Of the connections
that exist between entities (e.g., marital partners), the evaluation Of them (the

“sentiment" according to Heider) and the interdependence among them. For

4

example. members Of a family, partners in a marriage or a person and her
behaviors are viewed as units. A bond forms when the unit (e.g., the husband
and wife) and the evaluation Of it exist together with no tension. That is, there is
no pressure to change either the cognitive organization (i.e., “undo” the unit) or
the sentiment; the result is a balanced state. In short, the relationship between
unit formation and sentiments tend towards the preferred balanced state (Heider,
1958). Balanced relationships that.demonstrate a “good match' (i.e., similarity
and reciprocity) between partners are highly desirable (Forgas and Dobosz,
1980) and people have an implicit grasp Of what constitutes a balanced or good
match (Forgas, 1993). This is particularly true regarding marital partners (Argyle
8 Henderson, 1985; Kerckhoff, 1974; Klemer, 1970; Rubin, 1973; Udry, 1971).
Recent work by Hamilton 8 Sherman (1996) attempted to expand upon
Campbell’s (1958) notion Of social units or entitativity. Calling upon Gestalt
precepts, Campbell (1958) defined entitativity as the degree to which factors
such as similarity, proximity, and common fate lead one to perceive discrete
elements as parts Of a whole. Depending upon the combination of these factors,
“groups” will vary in the extent to which they are perceived to possess unity.
Hamilton 8 Sherman (1996) elaborate this point, suggesting that these
properties are correlated, but not perfectly, with perceived unity. Further, two
groups may be viewed as equally entitative but not for the same reasons; one
may he high on similarity while the other is high on common fate. In fact, while
acknowledging that the factors mentioned by Campbell (1958) certainly may

increase perceived unity, Hamilton 8 Sherman (1996) consider common fate,

5

interdependence (members have frequent and strong impact on each other
involving diverse activities over time), and organization (how the group is
structured) to be more important cues. The extent to which social perceivers
expect a group (e.g., married couple) to show consistency and coherence should
be influenced by the degree to which they are perceived as a “social unit“
(demonstrate high entitativity). Hamilton 8 Sherman (1996) proposed that
groups that are very tightly knit and interdependent are expected to be similar in
many respects. Thus, it seems reasonable to assume that social perceivers
would presume marital partners to be similar to each other based on
interdependence defined by their relationship.

In a sense, all married couples are likely assumed to possess a certain
level Of entitativity defined by the marital relationship. The present research
sought to explore the role of similarity as one Of the precursors to perceived
entitativity over and above interdependence. Thus, in addition to couples who
are similar or dissimilar on race, couples similar or dissimilar on occupation were
included. Occupation was selected as an attribute because, like race, it Often
may have preconceived or stereotypic characteristics associated with it. For
example, we may hold the belief that accountants are conservative, organized,
bookish, and wear glasses while an artist is thought to be liberal, free-spirited
and unpredictable.

Work by other researchers has taken a related but slightly different view
from Campbell (1958) and Hamilton and Shaman (1996) and shown that

formation Of a relationship category is the result of a process that Srull 8 Wyer

6

(1989) refer to as linking. Because Of linking, characteristics that belong to one
partner are Often tied to characteristics of the other partner. Couple members
are usually viewed as interdependent (Kemis 8 Wheeler, 1981), which can have
implications for how they are perceived as individuals. For example, Linda is a
runner so it may be assumed her husband Michael is concerned with staying in
shape. People may form an Opinion Of one partner given the information they
have about the other. Alternatively, one’s partner's habits or preferences might
have implications for the other partner. Maybe Lucy doesn’t like tO eat breakfast
so Ricky gets up earlier than necessary for work so that he has time to stop at
the bagel shop. This perceived interdependence or linking serves to blend two
people into a social unit.

!!!'l" Bll' l' 'E 'I' Q 'I'

Since Heider (1958), research has addressed specific kinds of
relationships (e.g., acquaintances, friends, romantic dyads) and the various
taxonomies employed by perceivers with regard to relationships, and has
attempted to determine which classifications are superordinate (Bond 8
Brockett, 1987; Fiske, Haslam, 8 Fiske, 1991; Taylor 8 Cracker, 1981). The
research discussed next addresses these and other questions regarding the use
Of how relationships are used as categories in cognitive organization.

Fiske and Cox (1979) considered the effect Of target familiarity and
descriptive purpose on how others are described. In a content analysis Of freely
recalled items, relationship descriptors (social roles and descriptions Of the

interaction one has with another) emerged as one Of the major categories of

7

description employed by subjects. The manner in which we organize information
in memory affects the way in which it is later recalled such that the category
used tO organize or cluster information emerges in free recall (Ostrom, Pryor, 8
Simpson, 1981). Relationship descriptors occurred particularly at the beginning
Of the description suggesting their primacy as an organizing structure.

Taylor and Crocker (1981) differentiated between two types Of cognitive
structures that help us to understand people: social structures based on the
relationships among people and personality structures based on characteristics
Of individuals. Their perspective was tested and extended by Bond and Brockett
(1987), who demonstrated that social contexts are superordinate to personality
characteristics in memory. They tested their social context-personality index
theory on subjects’ memory for acquaintances. In unconstrained recall, subjects
first named acquaintances in social clusters and then subdivided these clusters
into personality clusters. For example, subjects first recalled the people with
whom they bowl (e.g., Fred, VVIlma, Barney, and Betty), and then subdivided
these people into persons who are extroverted (e.g., Fred and VVIIma) versus
introverted (e.g., Barney and Betty). This finding demonstrates the central role
that relationships hold in the organization Of information in memory.

Even more recently, Fiske and his colleagues (F iske, Haslam, 8 F iske,
1991; Fiske, 1993; F iske, 8 Haslam, 1996) have conducted a series Of
experiments testing the relational-models taxonomy. This model maintains that
people generate, represent, coordinate, and evaluate social relations by

differentiating among four relationship types. Communal sharing, authority

8

ranking, equity matching, and market pricing represent discrete and coherent
cognitive structures that underlie a wide variety Of social roles (F iske 8 Haslam,
1996). Of particular interest here are communal sharing relationships. They are
exemplified by close kinship ties or love relationships such as marriage. In this
type of relationship, partners are equivalent and undifferentiated, focusing on
commonalities and disregarding differences and individual Identities (F iske 8
Haslam, 1996).

Fiske, Haslam, 8 Fiske (1991) tested the hypothesis that if relational
schemas play an important role in social cognition, then people would tend to
confuse those with whom they share the same type Of relationship. If, in
contrast, people use attribute schemas to organize acquaintances, they would
tend to confuse people who share the same attributes. For example, one would
be more likely to confuse an ex-boyfriend with a current boyfriend (communal
sharing) or confuse two teachers (authority ranking) if perceivers are thinking in
terms Of relationships. This pattern would be in contrast to confusing two Blacks
or two extroverts, which demonstrates the use Of attributes as a cognitive
organizing mechanism. F iske et al (1991) found that across five studies (looking
at naturally occurring confusions), confusions were better predicted by the
relationship model than the personal attribute schemas.

The same results held true for a diary study that collected naturally
occurring intentional substitutions (Fiske, 8 Haslam, 1996). Substitutions occur
when a subject selects an acquaintance to share in an activity (e.g., tennis) and

then must chose an alternate when her first choice is unavailable or changes her

9

mind. Fiske 8 Haslam (1996) sought to determine which characteristics
(relational or personal) were held in common by substitutable persons. First,
subjects kept a diary of naturally occurring substitutes. Later, as part of an
ostensibly different experiment, subjects recorded the personal attributes of their
substitutes using rating scales based on the Big Five personality dimensions.
Consistent with the first series of studies, use of relational taxonomies prevailed
over use of personal attributes. Three of the five personality dimensions failed
to significantly predict substitutions. Further, the effects for the personality
attributes disappeared or were attenuated when subjects’ tendency to interact
with people who share common attributes, rather than a tendency to base
substitutions on the attributes was taken into consideration.

F iske 8 Haslam (1996) maintain that relational models are utilized in
memory of person information to a far greater extent than has been previously
recognized. They suggest that these relationship schemas are implicit and are
mediated by structural rules that people use but are unable to articulate.

Overall, F iske and Cox, (1979), Taylor and Crocker, (1981), and Bond
and Brockett, (1987) all demonstrate that the use of relationships in cognitive
organization is a strategy employed early in the perception process and is
observable in the way in which information is later recalled (Ostrom, Pryor, 8
Simpson, 1981).

It seems reasonable that our desire for balanced, equitable relationships
combined with our tendency to link people together as social units might vary as

a function of the type of relationship present. Given the prominent and central

lO

role that marriage plays in our society, it also seems reasonable to assume that
individuals might rely on this unit relationship to organize social information
about the marriage partner.

Until recently, most research on person perception has focused not only
on hypothetical targets but also on unrelated, isolated targets. Yet rarely do we
meet someone without Ieaming something about his or her social network. In
real life we often meet others within a social context in which they are
accompanied (either physically or via implication) by spouses, friends, or co-

workers.

 

Sedikides, Olsen, 8 Reis (1993), in a series of five experiments,
demonstrated that subjects relied significantly on relationship categories to
organize target information in spite of instructions to memorize and form an
impression of each target person. Of specific interest to the present research,
Sedikides et al (1993, experiment I) demonstrated the spontaneous use of
marital relationships in cognitive organization. They exposed subjects to
information items about eight targets. Half the subjects were told that the eight
targets comprised four married couples and the dyads were identified (married
couple condition). The remaining subjects were supplied with the same target
information items, but targets were presented as married to unspecified others
(yoked couple condition). A clustering index called the adjusted ratio of
clustering or ARC score was employed to analyze the number and order of items

freely recalled (Gerjouy 8 Spitz, 1966; Roenker, Thompson, 8 Brown, 1971;

11

Simpson, 1979). The ARC score is “assumed to reflect the degree to which
information was cognitively organized around couples.” (Sedikides, Olsen, 8
Reis, 1993; Ostrom, Pryor, 8 Simpson, 1981).

Results of Sedikides et al (1993) study indicated that subjects
spontaneously organized the stimulus information around married couples more
so than around yoked couples (target person married to unspecified others).
Thus, subjects were using the marital relationship to organize information about
the marriage partners. The social unit, as an entity, provided a framework for
encoding incoming information.

An additional index of categorization used in Sedikides et al (1993), is
referred to as confusions. Confusions were defined as incorrect matches of
information with target names. VVIthin-couple confusions occur when subjects
recall an information item about one partner that belonged to the other partner
(e.g., It was Lisa’s husband that likes jazz, not Lisa). Between-couple
confusions consist Of recalling an information item as belonging to one partner in
a couple when it belongs to a member of a different couple.

Subjects made more within-couple confusions for married than yoked
couples, thereby demonstrating the strength of the marital relationship category.
Results also supported the hypothesis that fewer between-couple confusions
would occur for married than for yoked couples. This pattern was interpreted as
being relevant to the relative impermeability of boundaries present in the marital
relationship.

Overall, Sedikides et al (1993) demonstrated that perceivers

12

spontaneously organize information around relationship categories. In their
experiment, they selected information items such that categories were non-
overlapping, so as not to suggest competing categories. In other words, they did
not provide the same kind of information about each target (e.g., profession,
geographical location, age) because such information might offer additional
ways to categorize the targets. This raises the question addressed in the
present research of how robust relationship categories are as organizing
structures in the face Of other available categories (e.g., race).

In terms of perceptions of couples as social units, Sedikides et al (1993)
focus on interdependence as a precursor to entitativity. Half of their targets were
married to each other while the other half were presented in pairs married to
unspecified others. It is reasonable to assume that the former targets would be
viewed as sharing greater interdependence than the latter targets. The present
study goes one step further by trying to investigate the role of perceived
similarity given that interdependence exists. Thus, all the target couples in the
present research , described in greater detail later, are married to each other but

vary in similarity.

 

F iske and Taylor’s (1984) view of the social perceiver as a “cognitive
miser’ suggests that people will employ simple (minimal effort) and efficient
(minimal time) methods to organize the complex social world. Thus, one
categorizes and develops stereotypes about the categories. Categorization is

believed to be an inevitable, immediate, and an initially perceptual process that

13

occurs upon first encountering a target (Bruner, 1957).

Physical features are an important part of the categorization process
(Asch, 1946; Deaux 8 Lewis, 1984; F iske 8 Cox, 1979; Fiske 8 Neuberg, 1990;
McArthur 8 Baron, 1983; Posner, Nissen, 8 Klein, 1976). Race and sex are
particularly accessible categories that are difficult to suppress, and are believed
to be automatically encoded (Park 8 Rothbart, 1982). F iske and Neuberg’s
(1990) model of impression formation indicates this type of information (i.e., sex,
ethnicity, age, social class) is among the most basic of cues for activating a
meaningful social category, and this type of category-based processing has
priority over attribute-based processing. Other research has demonstrated the
pervasive use of such basic categories as race and gender.

Taylor, Fiske, Etcoff, and Ruderman (1978) had subjects listen to an
audio-taped discussion. Concurrently, slides of those speaking were shovm to
subjects. Taylor et al (1978) varied the sex and race of the discussion
participants. Subjects were later asked to match the audio statements with the
speakers. Results indicated that subjects did use the physical Characteristics of
race and sex as a means of categorization: intracategory errors were
significantly more common than intercategory errors.

Hewstone, Hantzi, and Johnson (1991) supported the results of Taylor et
al (1978) utilizing the same paradigm. They hypothesized that it is difficult to
interfere with a strong, automatic category such as race. Their work illustrates
that although memory for information about others improves as the task

becomes more engaging (e.g., anticipated interaction compared to simple

l4

recall), even the most demanding task did not override the strong tendency to
organize information along racial lines. In other words, even though an
engaging task resulted in better overall accuracy in recall, the kinds of errors still
reflected more intra- than inter-race mistakes. Thus, race information might
compete with relationship information as the primary organizing category.

Recall that the greater the perceived entitativity, the more social
perceivers expect those who comprise the unit to be similar (Hamilton 8
Sherman, 1996). Also, the more similar couples are perceived to be on a large
number of dimensions (race, education, attractiveness, etc), the more likely they
are to be viewed as a good match or a balanced relationship (Campbell, 1958;
Forgas, 1993; Forgas and Dobosz, 1980; Kerckhoff, 1974; Rubin, 1973; Udry,
1971).

The current study attempted to discover if inter-racial couples are viewed
as a “social unit“ to the same extent as intra-racial couples (Sedikides, Olsen, 8
Reis, 1993). It was hypothesized that they would not be. The next question is
whether it is something about race per se that interferes with the development of
entitativity or if other perceived dissimilarities are responsible. It was predicted
that although all married couples may be assumed to possess some degree of
interdependence, they may vary in the level of perceived similarity, which in turn
may effect their degree of entitativity. Therefore, the design included couples
who were either matched or mismatched on race as well as couples who were
matched or mismatched on another potentially discriminating attribute,

occupafion.

15

The paradigm for the current study is based on that used by Sedikides et
al (1993). Given the strong social norms governing marital partner selection, the
saliency of physical characteristics, and the tendency to use race as a primary
classification, the question of how information about inter-racial couples is

processed and organized is an important one.

MW
1111993119311: Dissimilar couples (those mismatched on race or occupation) will
not be perceived to be as strong a social unit as similar couples (those matched
on race or occupation). Therefore, marital relationship will be used as the
organizing category more frequently for matched than mismatched couples.
Wale, Clustering scores will be higher for matched couples compared to
mismatched couples indicating greater use of relationship categories.
W Matched couples will be viewed as more blended or intermeshed
than mismatched couples. It is expected that more within-couple confusions will
occur for matched couples than mismatched couples, reflecting a more
pronounced use of relationship categories for matched than mismatched.
WILLIE; There will be fewer Wuple confusions for matched
couples than mismatched couples reflecting stronger relationship category use.
W Subjects will correctly pair more of the matched couples than
mismatched couples on both semantic and visual matching tasks. Specifically,
subjects will correctly identify more of the matched pairings than the mismatched

pairings when presented with all target names and also when presented with all

16

target photos or occupations (depending on condition) in two matching tasks.
W12; Subjects in the mismatched couple conditions will organize
information around something other than the marital relationship (e.g., race,
occupation, gender) compared to those in the matched couple condition.
Prediction 2a: The sequence in which target items are recalled should be
informative as to the primary category used. If race and/or sex has been used
as the primary category then cluster analysis should reveal more same-
race/same-sex people or information items clustered together regardless of
marital relationships in the inter-racial conditions than to the intra-racial
conditions.

Prediction 2b: The between and MIMI couple confusions made will be accurate
with regard to the race/sex of the target the item actually belonged to but
incorrect with regard to the particular couple or partner for the inter-racial
condition than for the intra-racial condition. This pattern would demonstrate that
categorization by race or sex takes precedence over marital relationship when
couples are mismatched on race.

W Partners in the matched conditions will be perceived as more
similar to each other on underlying dimensions (e.g., morals, values, goals) than
partners in the mismatched conditions.

Prediction 3a: Couples in the matched conditions will have higher scores on the

belief similarity measure than those in mismatched conditions.

E I I E I' I I . I ,
The goal of the pre-test was to select photographs that equated physical
attractiveness (4-4.5 on a 7 point scale), depicted targets of 25-31 years of age,
and looked like candid shots that could have been taken by the experimenter. In
addition, in order to be selected, target photographs had to be a head and
shoulder pose with the target looking directly at the camera and smiling.
Attempts were made to equate other Characteristics (i.e., no target wore glasses,
all were dressed casually, etc). Eight photographs (4 African-Americans (2
male/2 female) and 4 Caucasians (2 male/2 female)) were selected.
Whiz Two hundred photographs of African-American and Caucasian
males and females were collected from catalogs and magazines. Photographs
were photocopied (resulting in black and white pictures) and all were placed
against a white background. One hundred and twenty psychology
undergraduates each rated a subset of seventy-five photographs for
attractiveness (1-7 point Likert scale; higher scores indicated greater
attractiveness) and perceived age of the target.
W: A list of a range of occupations was generated and 15
undergraduates were asked to indicate the relative status of each occupation on
a 7-point Likert scale with higher numbers indicating higher status. The goal of
the pre-test was to select 2 occupations that were equal and moderate in
perceived prestige (4 on a 7-point scale) but differed in Characteristics

stereotypically associated with it. The occupations of accountant (conservative,

l7

18

restrained) and graphic artist (creative, free-spirited) were selected.
Remnants; One hundred and eighty Michigan State University students (86
males and 94 females) recruited from introductory psychology classes
participated in exchange for course credit or extra credit.

WM Participants were randomly assigned to one of the cells of
a 2 (attribute: race vs occupation) by 2 (relationship type: matched on attribute
vs mismatched on attribute) between-subjects design. Participants were
presented with eight targets who comprised four couples. Those in the matched
relationship conditions saw couples who shared the same ethnicity (two Black
couples and two White couples) or four couples where partners shared the same
occupation (two couples in which both were graphic artists and two couples in
which both were accountants). Participants in the mismatched relationship
conditions read information about four couples who were mismatched on
ethnicity (Black males with White females (2) and White males with Black
females (2)) or four couples mismatched on occupation (accountants and
graphic artists paired together). Only those in the race conditions saw
photographs of the targets. Presentation of couples within conditions was
counterbalanced with regard to race, occupation and gender of target (i.e., half
the time the females were presented first and half the time the males were
presented first). (Appendix L)

Materials; The experiment utilized two booklets: one containing the stimulus
materials and one containing the distractor task and the dependent measures.

W The stimulus booklet consisted of sixteen pages, one for

19

each target person with a blank sheet between each (Appendix A). The eight
targets comprised 4 couples. The target’s name was presented at the top center
of each page. In addition, 5 pieces of information were presented on the page.
The first item of information identified the target’s spouse (e.g., Heather is
married to Todd). In the race conditions, race information was conveyed by
including a photograph of the target. In the occupation conditions, occupation
information was presented with spouse information (e.g., Rachael, a graphic
artist, is married to Mark). Target names (within each gender) were selected to
be equivalent in length. The four remaining items, chosen to be simple, neutral,
and feasible (i.e., hobby is photography) were also presented. At the bottom of
the page was a rating scale asking subjects for their impression of the target. It
was anchored by 1 (not at all favorable) and 9 (extremely favorable).
Witter; Dependent measures were presented in the
following order. Blank pages were inserted, where necessary to prevent
subjects from viewing forthcoming measures.

WEEK; The first page of booklet two consisted of a word search puzzle
requiring subjects to identify as many Dr. Seuss characters as possible within
the given time (2.5 minutes). (Appendix B)

W Thirty-two blank quarter-pages were provided for the free recall
task. Participants were asked to recall as many items of information as they
could in any order that came to mind with the following guidelines: Only one item
was to be written per quarter-page piece of paper and they were not to refer

back to prior pages. Participants were instructed to start each sentence with the

20

target’s name followed by the information item (e.g., Jennifer loves pizza).
W Subjects were presented with a list of the eight target names
along with a list of the 32 pieces of information (numbered) for the matching
task. Participants were instructed to write the number of the information items
that corresponded to each target in the blanks provided under each target name.
(Appendix C).

Wk; Female target names were listed in one column and male target
names were listed on another. Participants were asked to match the couples by
drawing a line from each female name to her partner’s name. (Appendix D).
BMW This task was completed by participants in the race
conditions only. Female target photos were presented on the left side of the
page and male target photos were presented on the right side. Participants
were asked to draw a line from each female target to her spouse. The photo
matching task and semantic task were counterbalanced. (Appendix E).
W This task was completed by participants in the
occupation conditions only. Each target name was listed and participants were
asked to write their occupation in the blank provided. The occupation matching
task and semantic task were counterbalanced. (Appendix F).
W Targets were presented as couples on the next four
pages (one page for each couple) along with seven questions designed to
measure how similar partners were perceived to be on a variety of dimensions
(e.g., values, religious beliefs, life goals). Seven—point Likert scales were used

with higher numbers indicating greater similarity. All information about the target

21

(photo or occupation information and information items) was presented for this

measure. (Appendix G).

Wu; The final page requested demographic information
and included a probe for suspicion of experimental intent. (Appendix H).

. -' Six questions probing for

 

attitudes towards inter-racial dating and marriage were embedded in a thirty-six
question survey. The survey, ostensibly for an unrelated study, used 7-point
Likert scales. (Appendix I).

Emeduce; Sessions included from 3 ~10 respondents and were conducted by
White (male or female) experimenters (Appendix J). Experimenters welcomed
participants to the experiment, described its purpose, and obtained informed
consent. Booklets, one containing the stimuli materials and the other containing
the dependent measures, were distributed to each person with the instructions to
keep them closed until instructed to open them. Individuals randomly received
one of the four possible stimuli booklets (matched/mismatched on race or
occupation). Participants were told they would see each target for 30 seconds
and that their task was to memorize as much of the information presented as
possible. They were also told that at the bottom of each target page was a scale
ranging from 1 (not at all favorable) to 9 (very favorable) and that they were to
indicate their impression of the target after the 30 seconds were over. They were
asked not to turn any other pages until instructed to do so. At this time,
respondents were informed that consecutive pairs of targets were married to

each other. Exposure time to each target was controlled by a pre-recorded tape

22

played by the experimenter. All instructions were briefly recapped on the tape
prior to beginning the experiment. Participants were instructed, via the tape, to
begin reading, to stop, to fill out the impression scale and to turn the page to
await further instructions. This procedure was repeated 8 times; once for each
target.

After exposure to the 8 target descriptions, participants were told to set
aside booklet one and place booklet two in front of them. Participants were
informed that the first task was the last timed portion of the experiment. They
were given 2.5 minutes to work on a word search puzzle intended to be a
distractor task. Next, subjects were asked to continue working their way through
the remaining pages, taking care to read all instructions completely and to raise
their hands if they had any questions. They were told to take as much time as
they needed to complete the entire booklet and that they should not turn back to
any pages nor should they look ahead. The dependent measures, proceeding
from recall to recognition tasks, were presented in the order described above.
After all participants had finished, materials were collected by the experimenter.
At this time, the second, supposedly unrelated study was introduced and
consent forms were distributed, signed, and collected. Under the guise of a
survey entitled WEE. the attitude questions
were collected. Participants were then debriefed (Appendix K), thanked for their
participation, and dismissed. Experimenters were able, via markings on the
back of both booklets and the surveys, to match each participant’s materials and

clip them together after subjects had left.

E 1' . I I ,
W In order to ascertain that all targets were viewed as
equivalent, subject’s impressions of each target were analyzed via ANOVA
procedures. Impressions did not differ significantly as a function of condition or
order of presentation. Impressions of five of the targets did differ significantly as
a function of subject sex with females rating them an average of .05 points
higher than males. All targets sharing the same race and gender (e.g., the two
Black males) were viewed as equivalent.

W Adjusted ratio of clustering (ARC; see appendix M) scores were
computed using the computational formula developed by Roenker, Thompson, 8
Brown (1971) and Gerjouy 8 Spitz (1966).

Wig; Incorrect matches between target names and information items
were determined to be either within-couple confusions or between-couple
confusions. A within-couple confusion consisted of attributing an information
item to the target’s partner rather than the actual target. A between-couple
confusion consisted of attributing an information item to someone other than the
correct target or that target’s partner. Proportions of each type of error over all
errors were calculated (between- or within-couple confusions/all confusions). All
confusions were further coded such that comparisons between the actual target
and the target named incorrectly by participants fell into one of the following
categories: same race/occupation and same sex, same race/occupation and

different sex, different race/occupation and same sex, different race/occupation

23

24

and different sex

' The number of correct

 

responses to each Of these tasks were tabulated by adding up correct responses
across all targets.

BelieLSimflaMMeesuce; Responses to each of the seven questions were added
across all four target couples to yield seven total scores for each subject. A
principal components factor analysis of the seven belief similarity questions
resulted in one factor, with an eigenvalue of 4.4, labeled “belief similarity” which

accounted for 63.6% of the variance. Reliability analysis of the seven questions

resulted in an alpha of .89.

 

- :1 e a ._ u . ' A principal
component factor analysis performed on the six attitude questions resulted in
one factor with an eigenvalue of 2.8. Two items were dropped due to low factor
loadings resulting in a four item factor labeled “attitudes toward inter-racial
dating and marriage” which accounted for 70.8% of the variance. Reliability

analysis of the four questions resulted in an alpha of .84.

Analyses;

tlxnetbesesi;

Hypothesis I predicted that couples matched on race or occupation would be
perceived more as a social unit than those mismatched on race or occupation.
Planned contrasts used to test the prediction that those in the matched

conditions would have higher adjusted ratio clustering (ARC) scores on the free

25

recall task compared to those in the mismatched condition were not significant
(Ms = 0.51 vs 0.53 respectively, t(159) = 0.841, p>.05. The free recall task was
also analyzed via a 3-way ANOVA of ARC scores by type (matched or
mismatched), attribute (race or occupation), and subject sex There were no
significant interactions or main effects to suggest any difference between
couples matched on race (M = 0.53) compared to those mismatched on race (M
= 0.53) or those whose occupation was matched (M =0.49) compared to
unmatched (M = 0.55). (See Table 1).

WIthin-couple confusions and between—couple confusions were analyzed
by forming proportions so that the total number of errors would be reflected in
the computation. Thus, the proportion of within-couple errors was formed by
dividing the total number of within-couple errors by the total number of errors
(both within and between).1 Planned comparisons indicated a trend in the
predicted direction such that those in matched conditions made more within-
couple confusions (and fewer between-couple confusions) compared to those in
the mismatched conditions (Ms =0.93 vs 0.96, t(169)=1.82, p<.07).

Three-way ANOVA’s were conducted for within-couple confusions by type
(matched, mismatched), attribute (race, occupation) and subject sex There
were no significant interactions. However, there was a main effect for type (F(1,
165) = 3.65, p< .05) such that more within-couple confusions (and fewer

between-couple confusions) were made for matched couples compared to

 

‘ Of course the between-couple confusions could have been analyzed and the results
would be the same since they are inversely related.

26
mismatched couples (Ms = .06 vs .03), as predicted. (See Table 2).

Planned comparisons for totaled within-and between-couple confusions
(instead of proportions) were not significant when comparing mismatched to
matched couples. There was no difference between matched (M=13.3) and
mismatched (M=14.4) couples for between-couple confusions (t(175)=.562,
p>.05) or within-couple confusions (Ms=.83 and .68 respectively, t(175)=.719,
p>.05). However, means were in the predicted direction for both types of
confusions.

Planned contrasts were also conducted for the semantic matching task
(matching the husband’s name to the wife’s name). There was no significant
difference between those in the mismatched conditions compared to the
matched conditions (Ms=2.41 vs 2.48 respectively, t(172)=.295, 9 >05. A
three-way ANOVA on the semantic matching task (matching the husband’s name
to the wife’s name) with type, attribute, and subject sex as factors. There was
one main effect for subject sex (F(1,168)= 7.17, p < .05. Female participants
correctly matched more of the couples by name (M = 2.8) than male participants
(M = 2.0 with t(174) = 2.79, p <.05. (See Table 3).

There were no significant differences between participants in the two
race conditions (targets matched or mismatched on race) on the photo matching
task. Participants in the intra-racial couple condition did not identify more
couples correctly than those in the inter-racial couple condition although the
means were in the predicted direction (M = 3.1 for matched versus M = 2.9 for

unmatched couples).

27

There were no significant differences on the occupation matching task
between those in the matched condition (M = 5.9) compared to those in the
unmatched conditions (M =6.0).

Hypothesis II:

Respondents in the mismatched conditions did not utilize race (or
occupation in the occupation conditions) or sex over marital category clustering
according to their ARC scores. Presence of such clustering would have
manifested itself as negative ARC scores. According to Roenker, Thompson, 8
Brown (1971), ARC scores of zero indicate chance clustering while negative
ARC scores indicate that some criteria (e.g., race) other than the one of interest
(marital relationship) had been used.

Errors on the recognition task which asked subjects to match each
information item to the correct target were analyzed. All errors (both within-
couple and between-couple) were included. The means and standard deviations
of each type of error are presented in Table 4. The total number of errors made
by subjects did not differ significantly across conditions (Ms = 13.6, 15.7, 15.9,
and 14.7 for conditions 1-4 respectively). Means did not'differ significantly for
any Of the kinds of errors across conditions or types (matched and mismatched).

A Chi-square analysis was conducted on each type of error by condition
so that observed frequencies could be compared to expected frequencies.
Results were not significant for any of the error types between conditions
suggesting that target race (or occupation in the occupation condition) or sex

was not used significantly more than chance for mismatched couples compared

28

to matched couples. Thus, hypothesis II which predicted that participants in the
mismatched conditions would use something other than the marital relationship
to organize information compared to those in the matched conditions, was not
supponed.

Hypothesis III:

A planned comparison demonstrated that couples in the matched
conditions were perceived as significantly higher in similarity (M=4.5) compared
to couples in the mismatched conditions (M=4.2 with t(176)=2.79, p<.006.

A three-way ANOVA of the belief similarity scale by type, attribute, and

subject sex resulted in one significant main effect for type. As expected,
couples matched on occupation or race were viewed as more similar (F(1,171) =
8.30, p < .05) on underlying beliefs (e.g., similar values, attitudes, background,
goals, etc) than couples who were mismatched (M = 4.55 and 4.22 respectively).
MW

Two-way ANOVA’s were conducted (for those in the race conditions) on
each of the dependent variables with type and subject sex as the factors.
Participants’ attitude measure was entered as a covariate. Attitude towards
inter-racial dating/marriage was not a significant covariate for any of the
dependent measures. The mean of the attitude scale (ranging from 1-7 with 7

being most favorable) was 5.47 (sd = 1.38).

29
215m

The purpose of this research was to determine if inter-racial couples are
viewed as a social unit to the same extent as intra-racial couples. It was
expected that they would not be, prompting a second research question of
whether it is race per se or perceived dissimilarity that prevents entitativity from
developing for inter-racial couples. Participants were provided with either race
or occupation information (along with 5 other information items) for eight targets
believed to comprise four married couples. Respondents completed free-recall
and recognition tasks about the information presented. In addition, responses
to matching tasks (i.e., pairing husbands and wives by name, matching pictures
or occupations with correct targets) and perceived belief similarity among pairs
were collected. Questions about attitudes toward inter-racial dating and
marriage were embedded in a questionnaire and administered as part of a
supposedly unrelated study.

Partial support was found for the hypothesis that couples matched on
race or occupation would be perceived as a social unit to a greater extent than
couples mismatched on race or occupation. Confusions between partners
indicated the relationship category of marriage was used differentially for
mismatched couples compared to matched couples. Subjects tended to confuse
partners within a couple more when they were matched on race or occupation
than when they were mismatched on these attributes. According to Sedikides et
al (1993), within-couple confusions indicate a greater use of the relationship

category of marriage. The conclusions reached by Sedikides et al (1993) (and

30

these findings) are consistent with Campbell’s (1958) and Hamilton and
Sherman’s (1996) discussion Of entitativity. Groups high in perceived entitativity
are expected to demonstrate greater unity, consistency, and coherence
compared to those not sharing entitativity (Hamilton and Sherman, 1996).
VVIthin-couple confusions are evidence that members within a couple are more
than two individuals. They are parts of a whole with a fading of individual
boundaries (Kemis 8Wheeler, 1981; Srull 8 Wyer, 1989).

This interpretation garners some support from the belief similarity
measure which demonstrated that couples matched on race or occupation were
perceived as being more similar in their values, attitudes, goals, and ideas on
child rearing, as sharing similar religious beliefs, interests, and backgrounds
compared to couples mismatched on these attributes.

As expected, participants made more between-couple confusions for
targets that were mismatched on race or occupation compared to those who
were matched on either of these attributes. Moreover, errors that indicated
confusion between couples may be interpreted as demonstrating a lack of
entitativity. The boundaries of the social unit are permeable and do not
demonstrate a strong use Of the relationship category of marriage.

In contrast, no differences emerged on the free-recall task (measured by
the adjusted ratio clustering index, ARC) between those in the two race
conditions or the two occupation conditions. It was anticipated that those in the
matched conditions would have higher ARC scores than those in the

mismatched conditions thereby demonstrating a greater use of the marital

31

relationship as an organizing category. Since other measures did support the
hypothesis that relationship information is used differentially by those in the
matched conditions compared to those in the mismatched conditions, these data
warrant discussion.

It may be that while the ARC measure is sensitive enough to differentiate
between married and unmarried couples (such as those in Sedikides et al, 1993
experiment), it may not be sensitive enough to differentiate between kinds of
married couples. Thus, in terms of Hamilton and Sherman’s (1996) discussion
of entitativity, the ARC measure may be sensitive to the interdependence of
couples but not the similarity/dissimilarity of couples.

Another interpretation is that the information imparted to subjects that
“consecutive pairs of targets were married to each other" was too salient or was
at least more salient than race or occupation thereby prompting participants to
believe they should view targets as couples in spite of the instructions to
memorize information about each person.

Thus, there appears to be some evidence that inter-racial couples are not
viewed as a social unit to the same extent as intra-racial couples. However, this
same pattern emerged for targets matched and mismatched on occupation
suggesting that it is not race per se (i.e., not the pure physicality of color) but
something about similarities/ dissimilarities (such as mismatched occupations)
that discourages the perception of entitativity.

The data suggest that it may not be race per se that accounts for society’s

continuing disapproval of inter-racial marriage (Rosenblatt, Karis, 8 Powell,

32

1995; US Census Bureau, 1995). Rather, it may be that dissimilarity on certain
dimensions (e.g., race, occupation) implies other underlying differences. Our
preconceived notions about Black males, conservative accountants or free-
thinking artists may underlie our reluctance to see balance and unity in a
relationship of mismatched partners (Argyle 8 Henderson, 1985; Heider, 1958;
Forgas, 1993; Forgas 8 Dobosz, 1980).

The lack of significant differences (as tested by ANOVA procedures) on
the tasks requiring respondents to match couples by name and photographs or
occupations (the semantic, photographic, and occupational matching tasks) is
puzzling. However, recall that the target’s name was presented with either a
photograph or occupation and spousal information before any other information
items were read. Recall also that the first task participants were asked to
perform was to indicate their impression of the target although they were told
that their ultimate goal was to memorize as much information as possible?Z
These conflicting instruction sets may be responsible for the lack of results.
Recent research has shown that different processing goals will have different
effects on how social information is processed (Hamilton and Sherman, 1996),
particularly when perceivers hold different expectations about the perceived
entitativity of the targets (McConnell, Sherman, 8 Hamilton, 1997). In short, it is
unclear if certain information (such as items presented earliest in a series) might

be better recalled than other information.

 

2Concern regarding the conflicting instructions was voiced prior to the study being conducted.
However, since part Of the goal was to replicate the results of Sedikides et al (1993), we decided tO follow
their procedure as closely as possible.

33

Alternatively, spousal information and (depending on condition) race or
occupation of the target, was information they had about every target they
viewed. None of the other information consisted of overlapping categories. In
other words, subjects did not receive hobby or religious preference information
about all targets. The study was designed this way in order to control the
number of ways subjects could categorize targets. However, it may have also
had the unexpected effect of causing subjects to pay particular attention to race,
occupation and spouses name since these data were overlapping category
information. Items like race, occupation or spouse’s name may have offered a
heuristic for organizing the incoming information because it offered alternative
categories other than marital relationship.

Less puzzling is the fact that the measure of attitudes toward inter-racial
dating and marriage was not a significant covariate. The undergraduate
population, as a whole is rather liberal and demonstrates a restriction of range
on such measures. A majority of the participants expressed very positive
attitudes towards inter-racial marriage and dating. Further, it is difficult to
access the extent of the social desirability bias often present in measures such
as this.

The unpredicted finding that female respondents correctly matched more
Of the couples compared to men is consistent with gender stereotypes. Females
tend to be more communal compared to males and are more concerned about
interpersonal relationships (Kelley et al, 1983). Therefore they are more likely to

be attentive to relationship information.

34

Participants did not appear to differentiate between kinds of couples
(matched and mismatched) at the perceptual level as measured by the clustering
index; the ARC scores for each type were not significantly different. Participants
were however, making distinctions later during the recognition task where they
were asked to match information items to the correct target as measured by
between- and within-couple confusions. It’s possible that as time passes
subjects have an opportunity to ask themselves the question “How do I feel
about that?“ Time may afford the social perceiver an opportunity to consider the
couple in other ways (i.e., moral, social, emotional). Perhaps the decision to
combine two individuals into a social unit comes after the perceptual process
when we have time to think about the partners. Implicit in the perceived
entitativity (or lack thereof) is the rendering of some valenced value judgment.
Do I (as a social perceiver) see these two partners as a balanced relationship or
a good match? (Forgas, 1993; Forgas 8 Dobosz, 1980; Kerckhoff, 1974).

One way of studying the effect of value judgments on perceived
entitativity would be to utilize information differing in valance (i.e., Jennifer
graduated at the top of her college class vs flunked out her first year in college)
to see if more negative information items are attributed to those married inter-
racially compared to intra-racially. We might expect that if there is some
negative evaluation being attributed to mismatched couples, they might be
associated with significantly more negative characteristics. This might also shed
light of the question of whether it is the inter-racial couple that is the focus of

social disapproval, the stigmatized out-group member (i.e., the Black partner for

35

White participants) or the deviant in-group member.

In summary, partial support was found for hypothesis I that matched
couples are viewed as a social unit to a greater extent than mismatched couples.
Evidence does not suggest that race or occupation information is utilized more
than relationship information for mismatched couples compared to matched
couples (hypothesis ll). Finally, perceived similarity of underlying dimensions
such as morals, values, background, life goals, religiosity, and attitudes is
viewed as greater for matched couples than mismatched couples (hypothesis
Ill).

One drawback to the current research is the artificiality of the four couple
composition. In real life, we do not meet the same number of couples married
inter-racially and intra-racially; inter-racial marriage accounts for only 1-1 ‘A
percent of all marriagesvin the United States (US Census Bureau, 1995). Nor is
it common for many of the couples we come in contact with to have the same
occupation. Research is currently being conducted that will consider a mixture
of inter- and intra-racial couples that more accurately depict the real world. It
seems reasonable to expect that one inter-racial couple in a group of intra-racial
couples, due to their increased saliency, might appear to be even less of a
social unit than those in the current study. The results found here might be even
more pronounced.

Additionally, it might be interesting to manipulate the kind of information
provided about each target (e.g., stereotypic and counter-stereotypic

statements) along with both negatively and positively valenced statements (as

36

mentioned earlier) to see which are recalled better. If negative stereotypic
statements are recalled better, it might be evidence that perceivers are relying
on (predominantly negative) stereotypes of social groups that bias how they see
the couple. If they recall counter-stereotypic statements, particUlarly positive
ones, it may suggest a willingness to see the couple as more than members of
their social groups.

It has been suggested that the support of family and friends is an
important factor in perceived marital satisfaction (Argyle 8 Henderson, 1985).
Ultimately, the lack of perceived social unity of a couple by outsiders might have
a detrimental effect on the partners who would not be granted the rights,
opportunities, and privileges extended to those whose relationships are
recognized (McCall 8 Simmons, 1970). If couples such as those who marry
inter-racially are not viewed as a social unit, it is possible that their future family
might also be viewed by others as less bonded, less cohesive than other families
(Rosenblatt, Karis, 8 Powell, 1995). Long-term effects on the adults and
potential effects on the children of such unions need to be considered.

Research is also currently underway to determine if other kinds of
dissimilarities interfere with the formation of the couple as a social unit. Couples
matched and mismatched on age or physical attractiveness are being
considered. Each of these attributes are connected to well established
stereotypes (Bar-Tal 8 Saxe, 1976; Berscheid 8 Walster, 1974; Jackson, 1992;
Kernis 8 Wheeler, 1981) that could act to inhibit the development of entitativity

between partners.

37

In conclusion, it appears that the reluctance of society to accept inter-
racial marriage may be due, in part, to the perception of underlying differences
between partners on certain belief dimensions rather than just the pure
physicality of race. More likely, it is a combination of these (and perhaps other)
components. Given that some of the same results were found for couples
mismatched on occupation, it seems reasonable to assume that other
differences (such as physical attractiveness) might show the same pattern of
results.

This is not to suggest that prejudice, negative stereotypes, historical and
social influences do not play a role in the perception of inter-racial couples. It
appears from other research that they do (Lewandowski, 1996; Rosenblatt,
Karis, 8 Powell, 1995; Staples, 1992). Attempting to study the specific
dimensions that social perceivers believe inter-racial partners differ on might
prove useful in discovering the core reasons for disapproval. The affect
associated with different kinds of mismatched couples (inter-racial, homosexual,
May-December romances) may add some interesting considerations.

Many questions remain for the interested researcher. For example, if
disapproval is not merely a result of perceptual processes (additional research
would have to rule out this possibility more conclusively) what underlying
dimensions are important for a “social unit“ to be perceived as such? Do certain
dimensions (i.e., religiosity) outweigh differences in physical attractiveness for
example? And perhaps most importantly, what effect does this have on couples

(and their families) engaged in atypical or mismatched marriages?

APPENDICES

APPENDIX A

APPENDIX A

Jennifer is married to Pete

Favorite color is blue

Hobby is photography

Loves pizza

College major was business

(PHOTOGRAPH)

What is your impression of Jennifer?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremelyfavorable

38

39

APPENDIX A

EEIE

Pete is married to Jennifer

Likes people who are “humorous”

Was born in Canada

Favorite actress is Julia Roberts

Is an early riser

(PHOTOGRAPH)

What Is your impression of Pete?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremely favorable

40

APPENDIX A

DANIEL

Daniel is married to Kimberly

Listens to Jazz

ls interested in politics

Once worked as a waiter

Likes animals

(PHOTOGRAPH)

What is your impression of Daniel?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremely favorable

41

K_|M_EE_B_L1

Kimberly is married to Daniel

Vacations in California

Fascinated by modern art

Is an avid reader

Pet peeve is heavy traffic

(PHOTOGRAPH)

What is your impression of Kimberly?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremely favorable

42

Mark is married to Rachael

Flosses every day

Favorite writer is Steinbeck

ls active in his faith

Collects antiques

(PHOTOGRAPH)

What is your impression of Mark?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremelyfavorable

43

Rachael is married to Mark

Belongs to a health club

Has occasionally traveled outside of Michigan

Favorite TV. shoe is “ER”

Preferred drink is red wine

(PHOTOGRAPH)

What is your impression of Rachael?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremely favorable

44

Heather is married to Todd

Belongs to a ski club

Likes to dress casually

Is a human rights activist

lsindependent

(PHOTOGRAPH)

What is your impression of Heather?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremelyfavorable

45

Todd is married to Heather

Lives on Meadowbrook Drive

Dreams of owning a Jaguar

Favorite US president is Roosevelt

Enjoys watching tennis

(PHOTOGRAPH)

What is your impression of Todd?

Not at all favorable 1 2 3 4 5 6 7 8 9 Extremelyfavorable

APPENDIX B

APPENDIX B

Please try and locate as many of the Dr. Seuss characters as you can in the
puzzle below. Circle the words as you find them as demonstrated by the

example.

(Word search puzzle placed here)

46

APPENDIX C

APPENDIX C

Please match each of the items with the correct person. Enter the corresponding number in the

blanks under the correct name.

TODD RACHAEL
MARK PETE
DANIEL HEATHER

1. Lives on Meadowbrook Dn've
. Avid reader
. Likes people who are “humorous”

. Favorite color is blue

2

3

4

5. Was born in Canada

6. Traveled occasionally outside of Michigan
7. Enjoys watching tennis
8. Likes to dress casually
9. Once worked as a waiter
10. Belongs to a health club
11. Vacations in Califomia
12. Flosses every day

13. Loves pl.

14. Listens to Jazz

15. Favorite writer is Steinbeck

16. Collects antiques

KIMBERLY

JENNIFER

17. Dreams of owning a Jaguar

18. Fascinated by modern art

19. Belongs to a ski club

20. Preferred drink is red wine

21 . Is an early riser

22. College major was business
23. Is independent

24. Pet peeve is heavy traffic

25. Favorite actress is Julia Roberts
26. Hobby is photography

27. Favorite US. president is Roosevelt
28. Favorite TV show is “ER”

29. Is active in his ram)

30. Is a human rights activist

31. Is interested in politics

32. Likes animals

47

APPENDIX D

APPENDIX D
Please draw a line from the wife on the left to her husband on the right so that

the four couples you saw are matched up.

RACHAEL DANIEL
HEATHER TODD
KIMBERLY PETE
JENNIFER MARK

48

APPENDIX E

APPENDIX E
Please match the wife on the left with her husband on the right by indicating the

proper number in the space.

PHOTO OF WIFE 1 1) HUSBAND PHOTO

This women’s husband is number

PHOTO OF \NIFE 2 2) HUSBAND PHOTO

This women's husband is number

PHOTO OF \NIFE 3 3) HUSBAND PHOTO

This women’s husband is number

PHOTO OF \NIFE 4 4) HUSBAND PHOTO

This women‘s husband is number

49

APPENDIX F

APPENDIX F

Please indicate whether each Of the following people you read about indicated

their occupation to be an accountant or a graphic artist.

Jennifer is a(n)

 

Rachael is a(n)

 

Heather is a(n)

 

Kimberly is a(n)

 

Pete is a(n)

 

Mark is a(n)

 

Daniel is a(n)

 

Todd is a(n)

 

SO

APPENDIX G

(Photograph or occupation

information here)

Favorite color is blue.
Hobby is photography.
Loves Pin.

College major was business.

APPENDIX G

(Photograph or occupation

information here)

Likes people who are “humorous
Was born in Canada.
Favorite actress is Julia Roberts

Is an early riser.

 

Circle the answer you select on the scale located directly below each question.

1. How similar do you think

Not at all similar 1
2. How similar do you think
Not at all similar 1
3. How similar do you think
Not at all similar 1
4. How similar do you think
Not at all similar 1
5. How similar do you think
Not at all similar 1
6. How similar do you think
Not at all similar 1
7. How similar do you think

Not at all similar 1

8

are in their values and beliefs?

4 5 6 7 Very similar
are in their religious beliefs?

4 5 6 7 Very similar
are in their attitudes?

4 5 6 7 Very similar
are in their background?

4 5 6 7 Very similar
are in their child rearing ideas?

4 5 6 7 Very similar
are in their life goals?

4 5 6 7 Very similar
are in their interests and hobbies?

4 5 6 7 Very similar

51

APPENDIX H

APPENDIX H

WON
1. SEx: MALE FEMALE 2. AGE:
3. MARITAL STATUS: SINGLE MARRIED DIVORCED
4. CLASS: FRESHMAN SOPHOMORE
JUNIOR SENIOR

5. RACEIETHNICITY:

ASIAN AMERICAN ANGLO AMERICAN

NATIVE AMERICAN OTHER

AFRICAN AMERICAN HISPANIC

6. Describe in your own words what you understand the purpose of this experiment to be.

 

 

 

7. How much did you enjoy participating in this experiment?
Not at all 1 2 3 4 5 6 7 Very much

52

APPENDIX 1

APPENDIX I

The following questions were embedded in a mock survey of 36 questions

entitled WW

1. I think that dating outside your race is acceptable.

2. I think that marrying outside your race is acceptable.

3. I would never marry someone of a different race.

4. I would date someone of a different race.

5. I have at least one good friend that is of a different race than me.
6. l have dated outside my race.

53

APPENDIX J

APPENDIX J

W

“I’d like to welcome you to the Person Perception Study. Please listen while I

give you a short description of the study and what your contribution to it will be.”

“You are going to be provided with some information about eight people. You’ll
be given 25 seconds to memorize the information. Then you will be asked to
recall the information about each person. If you are willing to participate in this

task, please read and sign the consent form you have in front of you.“

Have subjects read and sign the consent form. Collect the consent forms and set

them aside.

“Now let me tell you in greater detail how this experiment works. You have two
booklets in front on you. The top one says “TARGET DESCRIPTIONS” on it and
the second one says “MEMORY TASKS” First of all, certain portions of this
study are timed so it is VERY IMPORTANT that you DO NOT turn any pages in

the booklets until you are asked to do so.”

“There are eight people described in this booklet. Each description includes a

photograph and 5 pieces of information. When the tape says “BEGIN“, you will

54

55

open your booklet and begin reading. You will have 25 seconds to memorize
that person’s information. At that point, the tape will tell you to “STOP“. There is
a 9 point rating scale ranging from (1) “Not at all favorable“ to (9) “Extremely
favorable“ at the bottom of the description page. Please indicate your impression
of the person, turn the page and wait until you are told to continue.”

Ask to see if there are any questions about the procedure before beginning.
Make sure to pause for a few seconds. Be observant. If they have a confused

look on their face, briefly repeat the procedure.

Please set aside the booklet marked “MEMORY TASKS“ and place the booklet
marked “TARGET DESCRIPTIONS” in front of you. Again, remember to begin

and stop only when I ask you to do so.“

Make sure that after the first description, they turn to the colored filler page and

wait for you to say begin.

W

After you have completed the “TARGET DESCRIPTION” booklet:

“Now please set aside your “TARGET DESCRIPTION” booklet and place your
MEMORY TASKS” booklet in front of you. Only the first task, a word search

puzzle will be timed. When the tape says begin, open your booklet and begin

56

woking on the word search puzzle on the first page. Do not turn any other

pagesf
Tape will time this for 2.5 minutes and then instruct the participants to stop.

The rest of the booklet can be worked on at your own pace. Follow the directions
for each task. If you do not understand a task, raise your hand and I will assist

you. Take as much time as you need for the remaining pages“.

Again, ask if there are any questions before beginning. Alter all the participants
are finished ask them to paper clip their two booklets together and collect them.
Read the debriefing information to them, sign and stamp their Credit cards
and dismiss the participants. Give each participant a copy of the debriefing form

to take with them.

 

APPENDIX K

 

APPENDIX K

W

At this time. we’d like to thank you for participating in this Person Perception
Study. By taking the time to participate, you have made a valuable contribution
to psychological research. I’d like to tell you a bit more about what we are

looking at with this study and what you have helped us to accomplish.

Everyday all of us are bombarded with huge amounts of new information. We
watch television, read newspapers and magazines, converse with many people
(new and known). Our senses take in so much information that we depend on

certain tools to help us simplify our world. Usually we try to do this in a simple

 

and efficient way.

One very helpful tool for accomplishing this is to categorize or classify
information into groups. In other words, let’s say I asked you to memorize this list
of words (apple, snail, zebra, banana, pear, dog). Then I asked you to recall the
words. Chances are you would cluster your responses by “fruit” and “animal”.

Categorizing information provides a framework to organize incoming data.

We do the same thing with people. We can use all kinds of different
classifications: age, gender. occupation, race, hobbies, relationships (family,

friends, co-workers).

This study is trying to see if the category of marriage (a particularly potent type
of relationship in many cultures) Is a prevalent or common way of organizing
information. There is research that suggests relationships are used quite often in
cognitive organization. If I asked you to name 25 people that you know, you

might very well list family members, school friends, teammates in clusters (all of

57

58

one then all of another). Some researchers believe you are more likely to do that
instead of using individuating traits (all those with a good sense of humor, all the

extroverts, etc).

Now, you can easily imagine being at a party and meeting a woman, talking to
her for a while and then being introduced to her husband. So after learning
information about each of them, do you encode the information about 2 separate
people or do you organize it around the couple? The social world extends may
rights, courtesies, and privileges to spouses. In fact, there is research to
demonstrate that social support is a significant contributor to marital happiness
and success. It may be that couples who are not treated as a social unit, miss

out on this very important dimension.

So we are also interested in whether or not this relationship category is used in
the face of other, very salient categories such as race. Vlfill inter-racial couples
be considered a social unit like intra-racial couples are? If not, it may be up to
future research to determine what the ramifications of that are for inter-racial

couples.

Hopefully, this study will assist us in learning how information about people is

organized in memory. Again, thank you for taking the time to participate. If you
have any other questions about this research or are interested in knowing the

results when they are available, please feel free to contact the person listed

below.

Dr. Linda A. Jackson 432 Baker Hall 353-8690

Donna A. Lewandowski 424 Baker Hall 432-2169

 

APPENDIX L

 

In”... ... ._...

Appendix L
Couuterbglanciug Procedures

 

3:995;
2 Black males: Mark 2 Black females: Rachael
Todd Heather
2 White males: Pete 2 White females: Jennifer
Dan lfimberly
C o ' '0
MB Mfl MJ MK Bolded and underlined = Black couples
IR 111 TJ TK Underlined only = White couples
PR PH 3.1 % Bolded only = White male and Black female
DR DH DJ DIS Italicized = Black male and White female

 

luteziacial couditions;
4 orders of each of the following pairings:

 

MJ TK PR DH B,WIB,WNV,BNV,B
PH DR MR TJ W,BlW,BIB,W/B,W
RP JM HD KT B,WlW,B/B,W/W,B
KM HP JT RD W,B/B,WIW,BIB,W
I 'a o ' 'o

4 orders of each of the following:

PJ DK MR TH W,WNV,WIB,BIB,B
MH TR PK DJ B,BIB,BNV,WNV,W
KD HT JP RM W,WIB,BNV,WIB,B
RM JP HT KD B,BNV,W/B,BNV,W

 

Counterbalanoing of occupations followed the same procedures such that ‘/2 had the same and V:
had different occupations, ‘/2 had the wife presented first and ‘/2 had the husband presented first,
and ‘A had graphic artist presented first and ‘/2 had accountant presented first.

59

 

APPENDIX M

Appendix M

The Roenker, Thompson, and Brown (1971) formula for calculating adjusted

ratio of clustering (ARC) scores is as follows:

ARC = B - E(R)
MaxR - E(R)
R = The number of times a category item follows an item in the same
category.
E(R) = Expected (chance) number of category repetitions and is
calculated as Eu} -1
N
maxR = Maximum possible number of category repetitions and is equal to N

(total number of items recalled) - k (number of categories

represented in the recall protocol.

In the following string, the bolded words are the ones Sedikides, Olsen and Reis
(1993) would have coded as “hits” (R in the above formula) if Jack and Jill were
a category:

Ex 1: JACK JACK JILL JILL JILL JACK JACK JACK

Ex 2: JACK JILL JILL PETE JACK JACK JILL

60

 

Table 1

Means and standard deviations for all dependent measures by condition.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent measure: Court I1 Cond ll Cond III Cond IV
Clustering index (ARC) .53 (.24) .53 (.23) .49 (.24) .55 (.20)
Between-couple confusions (total) 12.9 (8.0) 15.0 (8.0) 14.8 (6.5) 13.9 (6.3)
Within-couple confusions (total) .67 (.83) .68 (1.8) 1.0 (1.1) .71 (1.0)
Between-couple confusions (proport)2 .94 (.10) .97 (.07) .94 (.07) .95 (.06)
\Mthin-couple confusions (proport) .06 (.10) .03 (.07) .06 (.08) .04 (.06)
Semantic (name) matching task 2.3 (1.5) 2.5 (1.4) 2.5 (1.5) 2.3 (1.4)
Photographic matching task3 3.11 (1.2) 2.95 (1.2) na na
Occupation matching task" na na 6.0 (1.5) 5.7 (1.9)
Belief similarity measure‘ 4.51 b (.63) 4.14,, (.74) 4.52,, (.57) 4.30,I (.97)
Attitude measure5 5.4 (1.6) 5.5 (1.3) 5.2 (1.3) 5.6 (1 .3)
Total number of errors 13.6 (8.2) 15.7 (8.4) 15.9 (6.8) 14.7 (6.6)

 

 

Standard deviations are shown in parentheses. Means with different subscripts are different at the

p<.05 level.
‘ Conditions:

I - Couple matched on race

III - Couple matched on occupation

61

ll - Couple mismatched on race

IV - Couple mismatched on occupation
2 Numbers represent a proportion of type of confusion (between or within)ltotal of all confusions.
’ Higher numbers indicate greater matching.

‘ Higher numbers indicate greater perceived similarity.
5 Higher numbers indicate higher approval of inter-racial dating and marriage.

Table 2

Means and standard deviations for all dependent measures by type.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent measure: Matched‘ Mlsmatched
Clustering index (ARC) .51 (.23) .53 (.20)
Between-couple confusions (total) 13.9 (6.9) 14.3“ (7.18)
Within-couple confusions (total) .83 (.95) .67 (1.5)
Between-couple confusions (proportY .93 (.09) .96“ (.07)
Within-couple confusions (proport) .06 (.09) .03“ (.07)
Semantic (name) matching task 2.41 (1.4) 2.48 (1.4)
Photographic matching task 3.1 (1.3) 2.9 (1.2)
Occupation matching task 6.0 (1.6) 5.9 (1.6)
Belief similarity measure 4.5 (.58) 4.2“ (.86)
Attitude measure 5.3 (1.5) 5.8 (1.3)

 

' Matched = matched on race or occupation; conditions I and Ill.

Msmatched = mismatched on race or occupation; conditions II and IV.

Table 3

Means and standard deviations of all dependent measures by subject sex

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent measure: Male Female
Clustering index (ARC) .52(.23) .53 (.20)
Between-couple confusions (total) 14.2 (7.0) 14.0 (7.1)
Within-couple confusions (total) .76 (1 .4) .75 .98)
Between-couple confusions (proport)2 .94 (.09) .94(.07)
WIthin-couple confusions (proport) .05 (.09) .05 .07)
Semantic (name) matching task 2.0(1.4) 2.8“ (1 .3)
Photographic matching task 2.7(1.2) 3.1(1.2)
Occupation matching task 5.6(1.5) 6.1 (1 .7)
Belief similarity measure 4.3 (.78) 4.5 (.70)
Attitude measure 5.4 (1 .3) 5.5 (1.5)

 

* = significantly different at the p< .05 level

 

 

 

Table 4

Means and standard deviations of error types.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type of error: Infra-racial: Inter-racial:
Different race 8 sex 1.2 (1 .5) 1.2 (2.3)
Different race/same sex 6.7 (4.8) 7.6 (4.7)
Same racel different sex 1.2 (1 .3) 1.5 (2.0)
Same race 8 sex 4.3 (3.3) 5.2 (3.7)
Type of error: Same Different
occupation: occupation:
Different occupation 8 sex 1.5 (1 .3) 1.5 (1 .3)
Different occupation/same sex 8.6 (4.4) 7.2 (3.6)
Same occupation/different sex 1.7 (1 .7) 1.6 (2.1)
Same occupation 8 sex 3.8 (2.5) 4.2 (3.2)
Type of error: Matched: Mismatched:
Different attribute 8 sex 1.3 (1.9) 1.3 (1 .4)
Different attribute/same sex 7.6 (4.7) 7.3 (4.3)
Same attribute] different sex 4.1 (3.0) 4.7 (3.5)
Same attribute 8 sex 1.5 (2.1) 1.6 (1 .5)

 

 

 

“ = means are different at the p < .05 level

63

 

 

 

 

Table 5

Means and standard deviations all dependent measures by attribute.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent measure: Race Occupation
Clustering index (ARC) .53 (.22) .52 (.22)
Between-couple confusions (total) 14.4 (7.8) 13.9 (6.12)
WIthin;couple confusions (total) .71 (1.5) .79 (.97)
Between-couple confusions (proport)2 .94 (.09) .94 (.07)
Within-couple confusions (proport) .05 (.09) .05 (.07)
Semantic (name) matching task 2.41(1.4) 2.48 (1 .4)
Photographic matching task na na
Occupation matching task na na

Belief similarity measure 4.3 (.72) 4.4 (.76)
Attitude measure 5.4 (1 .3) 5.4 (1 .5)

 

' Race = matched or mismatched on race; conditions I and II.
Occupation = matched or mismatched on occupation; conditions "I and IV.

64

 

 

 

 

LIST OF REFERENCES

BIBLIOGRAPHY
Argyle, M., 8 Henderson, M. (1985). NW
London: Heinemann.
Asch, S. (1946). Forming impressions of personality. Mums]

and Sgcial Psychglggy, 41, 258-290.
Bar-Tal, D., 8 Saxe, L. (1976). Perceptions of similarity and dissimilarity

attractive couples and individuals. WW1,
Q 772-781. Ir“
Berscheid, E., 8 Walster, E. (1974). Physical attractiveness. In L. '
Berkowitz (Ed-I. W New York:
Academic Press.
Bond, C. F., 8 Brockett, D. R. (1987). A social context-personality index

theory of memory for acquaintances. WWW =4.“
WC). 1110-1121.

Bruner, J. S. (1957). On perceptual readiness. WM
123-152.

Bruner, J. S. (1958). Social psychology and perception. In E. E. Maccoby,
T. M., Newcomb, 8 E. L. Hartley (Eds.), WM (85-94).
New York: Holt, Rinehart 8 VIfinston.

Campbell, D. T. (1958). Common fate, similarity, and other indices of the

status of aggregates of persons as social entities. W0), 14-
20.

 

Deaux, K., 8 Lewis, L. (1984). The structure of gender stereotypes:
Interrelationships among components and gender label. Jgurnal of Personally
and Sgcial Psychglogy, 46, 991-1004.

Fiske, A. P. (1991). Structures of sgcial Irf' e; Jue four elementary 1th§ gt
humanlelatigus, New York: Free Press.

Fiske. A. P- (1991). WWW
h_ur_r)au_r§_lati_gns, New York: Free Press.

65

66

Fiske, A P. (1993). Social schemata for remembering people:
Relationships and person attributes that affect clustering in free recall of
acquaintances. Journal of Quautitatiye Authrouolggy, 5, 305-324.

F iske, A P., Haslam, N., 8 F iske, S. T. (1991). Confusing one person with
another: What errors reveal about the elementary forms of social relations.

WWOL 656-674.
Fiske. A. P.. 8 Haslam. N. (1996)- mm

W Manuscript submitted for publication.
Fiske, S. T., 8 Cox M. G. (1979). Person concepts: The effect of target

familiarity and descriptive purpose on the process of describing others. Jgurual
gLEersgnaIM 136-161.

Fiske, S. T., 8 Neuberg, S. L. (1990). A continuum of impression
formation, from category-based to individuating processes: Influences of
information and motivation on attention
and interpretation. WWW 1-74.

Fiske, S. T. 8 Taylor, S. E. (1984). SgciaLngrIjtigu, New York: Random
House.

Forgas, J. P. (1993). On making sense of odd couples: Mood effects on
the perception of mismatched relationships. er 0 ' la 3 o 0
Bulletin, 19(1), 59-70.

Forgas, J. P., 8 Dobosz, B. (1980). Dimensions of romantic involvement:
Towards a taxonomy of heterosexual relationships. Will
43, 290-300.

Gerjouy, I. R., 8 Spitz, H. H. (1966). Associative clustering in free recall:
Intellectual and developmental variables. MW
10, 918-927.

Hamilton, D. L., 8 Sherman, S. J. (1996). Perceiving persons and groups.

Esychglogigal Review, 103, 336-355.
Heider, F. (1958). flag psychglggy of interpersonal relations, New York:

 

 

67

\NIIey.

Hewstone, M., Hantzi, A., 8 Johnston, L. (1991). Social categorization
and person memory: The pervasiveness Of race as an organizing principle.
WW 517-528

Jackson, L. A. (1992). ' '
W New York. State University of New York.

Kelley. H H (1979) W F
em Hillsdale, NJ: Erlbaum.

Kelley, H. H., 8 Thibaut, H. H. (1978). W

W New York: VVIley.
Kelley, H. H., Bersheid, E., Christensen, A., Harvey, J. H., Huston, T. L.,

Levinger, G., MCCIintock, E., Peplau, L. A, 8 Peterson, D. R. (1983). Clasa
Balatianahjpa, New York: W. H. Freeman 8 company.

Kerckhoff, A C. (1974). The social context of interpersonal attraction. In
T. L. Huston (Ed.), WW New York: Academic

press.

 

 

Kernis, M. H., 8 Wheeler, L. (1981). Beautiful friends and ugly strangers:

Radiation and contrast effects in perceptions of same-sex pairs. Earsonality and
Social anahology Bullatlr), 7, 617-620.

Klemer, R. H. (1970). W New York:
Harper 8 Row.

Lewandowski, D. A 8 Jackson, L. A. Perceptions of inter-racial couples:
Prejudice at the dyadic level. (Under review).

McArthur, L. 2., 8 Baron, R. (1983). Towards an ecological perspective of
social perception. W 215-238.

McCall, G. J., 8 Simmons, J. L. (1978). Identities and iutaractious: An
exammaflcdcthumauassosiaflensjmemfiavfie, (reV- ed-). New York: Free
Press.

McCall, G. J., McCall, M. M., Denzin, N. K., Suttles, G. D., 8 Kurth, S. B.

68

(1970). Social Relationships. Chicago, IL: Aldine Publishing Company.
McConnell, A R., Sherman, S. J., 8 Hamilton, D. L. (1997). Target
entitativity: Implications for information processing about individual and group

targets. WW9). 750-762.
Meiners, M. L., 8 Sheposh, J. P. (1977). Beauty or brains: Which image

for your mate? WWW 262-265.

Newcomb, T. M. (1961). The acquaintance process. New York: Holt, I:
Rinehart 8 \Mnston.

Rosenblatt, P. C., Karis, T. A, 8 Powell, R. D. (1995).Mu|1iranial_@_ungs_,
Sage Publications: Thousand Oaks, CA.

Rubin, Z. (1973). Liking and laying. New York: Holt, Rinehart, 8 Vtfinston.

Ostrom, T. M., Pryor, J. B., 8 Simpson, 0. D. (1981). The organization of
social information. In E. T. Higgins, C. P. Herman, 8 M. P. Zanna (Eds), Saga]
WWW 3-38.

Park, 8., 8 Rothbart, M. (1982). Perception of out-group homogeneity and

levels of social categorization: Memory of the subordinate attributes of in-group

 

and out-group members. rso I' ' I
1051-1068.

Posner, M. l., Nissen, M. J., 8 Klein, R. M. (1976). \fIsuaI dominance: An
information processing account of its origins and significance. Esyannlngiral
Revlayy, 83, 157-171.

Roenker, D. L., Thompson, C. P., 8 Brown, S. C. (1971). Comparison of
measures for the estimation of Clustering in free recall. EsyghalngiaaLBullatln,
16(1), 45-48.

Schoen, R., 8 WOOIdredge, J. (1989). Marriage choices in North Carolina
and Virginia, 1969-71 and 1979-81. JaumaliMaujagaangjnaEamflyfl, 465-
481.

Sedikides, C., Olsen, N., 8 Reis, H. T. (1993). Relationships as natural

categories. Journal or Personality and Social Psychology, 64(1), 71-82.

69

Sigall, H., 8 Landy, D. (1973). Radiating beauty: The effects of having a

physically attractive partner on person perception. JaumaLorEaLsnnalinLang
W 218-24.

Simpson, D. D. (1979). Four empirical investigations of organization of
person information in memory. Unpublished doctoral dissertation (Study 2), Ohio
State University.

Simpson, D. D. (1979). Four empirical investigations of organization of 1"
person information in memory. Unpublished doctoral dissertation (Study 3), Ohio I
State University.

Simpson, D. D. (1979). Four empirical investigations of organization of
person information in memory. Unpublished doctoral dissertation (Study 4), Ohio
State University. 5-

Srull, T. K., 8 Wyer, R. S. (1989). Person memory and judgment.
W0). 58-83-

Staples, R. (1992). Intermarriage. In E. W. Borgatta 8 M. L. Borgatta
(Eds.), Engrclpjadjafiguiulnmz, 968-974. New York: Macmillan.

Taylor, S. E., 8 Crocker, J. (1981). Schematic bases of social information
processing. In E. T. Higgins, C. P. Herman, 8 M. P. Zanna (Eds), mial
WW 89-134. Hillsdale, NJ: Erlbaum.

Taylor, S. E., Fiske, S. T., Etcoff, N. L., 8 Ruderman, A J. (1978).

Categorical bases of person memory and stereotyping. Journal a Personality
WWB-Nfi

Udry, J. R. (1971). Wales: (2nd ed.). Philadelphia,

PA: Lippincott.

U. S. Bureau of the Census (1995). Statiatical anstract of the Unites!
Stataa (110th Ed.) Washington D. C.: U. S. Government Printing Office.