 

THEdis

SITY LIBRAR RIES

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIII

31293001565

III

 

 

 

 

 

LIBRARY

Michigan State
University

 

w

 

This is to certify that the

thesis entitled

DEVELOPMENT OF THE PUERTO RICAN VERSION OF THE
INTERPERSONAL ADJECTIVE SCALES REVISED:
ESCALA DE ADJETIVOS INTERPERSONALES

presented by.

Marieva Puig

has been accepted towards fulﬁllment
of the requirements for

M.A. degree inEsxnhalogy.

 

MM

Major professor

Date 3/26/97

0-7 639 MS U is an Aﬁimatt‘vc Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove We checkout from your record.
TO AVOID FINES return on or before one due.

  

DATE DUE DATE DUE DATE DUE

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DEVELOPMENT OF THE PUERTO RICAN VERSION OF THE INTERPERSONAL
ADJECTIVE SCALES - REVISED: ESCALA DE ADJETIVOS INTERPERSONALES

By

Marieva Puig

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1997

ABSTRACT

DEVELOPMENT OF THE PUERTO RICAN VERSION OF THE INTERPERSONAL
ADJECTIVE SCALES-REVISED: ESCALA DE ADJETIVOS INTERPERSONALES

By

Marieva Puig

This study examined the structure of a Puerto Rican version of the Interpersonal
Adjective Scales -Revised (IAS-R) (Wiggins, Trapnell & Phillips, 1988). The “Escala de
Adjetivos Interpersonales” (EAI) was examined from a survey of 210 Puerto Rican
college undergraduates. The ﬁrst wave of analyses consisted of conﬁrmatory factor
analysis of the EAI scales. Indices of ﬁt indicated that the two factor structure was not
the best ﬁt. Further, principal component analyses suggested that a three factor solution
more adequately ﬁt the data. A second wave of exploratory analyses consisted on
performing ﬁrst and second order principal factor analysis on the 64 items of the EAI. In
this analysis a two factor structure was deemed appropriate. These ﬁndings suggest
similarities and raised the question of potential differences between the IAS-R and the
EAI. The possibility of these ﬁndings being the result of measurement error suggest the

need to continue the development of the EA].

To all the people who have supported me throughout this journey.

ACKNOWLEDGMENTS

The completion of this project represents a big step toward achieving my personal
and career goals. I owe special thanks to those whom with I counted for support and
guidance. To Aaron Pincus, who ﬁrst introduced me to research in Clinical Psychology
and reminded me that with perseverance I can accomplish everything. To Michael
C. Lambert, my chairperson, who challenged me and inspired me to continue my
dedication to ethnic-minority issues. Thanks to Bob Caldwell for his unconditional
support and serenity. Thanks to Joel Aronoﬁ‘ and John Hurley, two diﬂ‘erent mentors who
have helped me grow and learn since I left Puerto Rico. Finally, A mi familia (to my

family) whom I love and cherish.

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................... vii
LIST OF FIGURES ..................................................................................................... viii
CHAPTER 1
INTRODUCTION .......................................................................................................... 1
Why Puerto Ricans ..................................................................................................... 1
CHAPTER 2
INTERPERSONAL THEORY ........................................................................................ 6
Circumplex Models of Interpersonal Behavior ............................................................. 7
The Circumplex and its Usefulness for Diagnosis and Understanding
Psychopathology ....................................................................................................... 10
Interpersonal Adjective Scales-Revised ..................................................................... 11
CHAPTER 3 '
WHY EXAMINE THE IAS-R STRUCTURE WITH A PUERTO RICAN SAMPLE....14
CHAPTER 4
METHODS .................................................................................................................... 16
Sampling and Data Collection Procedures ................................................................. 16
Subjects .................................................................................................................... l6
Measure .................................................................................................................... 16
CHAPTER 5
RESULTS ..................................................................................................................... 18
Data Analyses ........................................................................................................... l8
Psychometric Properties of the EAI ........................................................................... 19
Internal consistency ............................................................................................ 19
Principal component ........................................................................................... 20
Cultural Inﬂuences on the EAI’s Responses ............................................................... 23
CHAPTER 6
DISCUSSION ............................................................................................................... 25

CHAPTER 7
REFERENCES ............................................................................................................. 39

vi

Table

LIST OF TABLES

Set of Interpersonal Adjective Scales-Revised (IAS-R)
Set of Adjective Scales for the EAI

Psychometric Characteristics of the EAI
Intercorrelation Matrix of the EAI Scales

Factor Loadings of the EAI

Factor Loadings of the EAI Scales ﬁom Principal Component
Analysis

Factor Loadings of the Second Order Principal Component
Analysis of the EAI

”I?
CD

32

33

34

35

36

38

LIST OF FIGURES

Fi ure _ag_P e

l Circumplex Structure of the Interpersonal Adjective Scales-Revised 30

viii

CHAPTER 1
INTRODUCTION

The ﬁrst section of this thesis contains a brief historical background of Puerto
Ricans, followed by a review of Sullivan’s interpersonal theory, Wiggins’ Circumplex
model of interpersonal traits, and the IAS-R measurement. Finally, the methods and
analyses performed in order to examine the factor structure of the EAI are discussed.
Why Puerto Ricans?

The 1989 Census listed Puerto Ricans as the second largest Hispanic subgroup
(2.5 million) residing in the United States (Alvarez, 1992). In addition, approximately 4
million Puerto Ricans reside on the island itself (Lamberty & Garcia C011, 1994). Despite
thelargenumberof Puerto Ricansontheislandandthemainland,thereisadearthof
research on this population. Most research groups Puerto Ricans with other Hispanic
groups in the US. mainland. Consequently, potentially important idiographic diﬁ‘erences
among the groups in these samples are overlooked and ignored.

The historical and migration experiences of various Hispanic groups in the United
States differ greatly (Baca Zinn, 1989). For example, many Cubans migrated to the
United States as political refugees. Meanwhile, Mexicans living in California, New
Mexico, Utah, Nevada and Colorado became US. residents as a result of the Guadalupe-
Hidalgo Treaty, (1848) while some of those still in Mexico later became illegal immigrants
(Amott & Matthaei, 1991). In the case of Puerto Ricans, they largely began to migrate in

1948 when Puerto Rico became a Commonwealth of the United States. These three

2

The problems caused by the singular classiﬁcation of Hispanics are numerous.
These problems are compounded by the difference among Hispanics in socioeconomic
factors. These factors include educational achievements, family income, and participation
in the labor market (Baca Zinn, 1989; Romero, 1986). Generalizations made from studies
of Hispanics as a single racial or ethnic group do not allow for each group’s accurate
representation. This study addressed this problem of heterogeneity in grouping by
focusing speciﬁcally on Puerto Ricans. Thus, it acknowledges the unique characteristics
of this group, including history and ethnicity.

The etlmic background of Puerto Ricans is a mixture of three races: Taino
(indigenous people), Spanish (colonizers) and Aﬁ’ican slaves brought by the Spaniards.
These tln'ee groups contributed to form the Puerto Rican culture. The gregarious Emily
patterns, interdependence among community members, and strong kinship bonds are
characteristic of Puerto Rican culture, and are where the African and Taino inﬂuence are
most evident. For Puerto Ricans, maintaining strong Emily bonds is the greatest
responsibility in life (Locke, 1992). Kinship, dependency, and cooperation between
immediate Emily members and other relatives are highly valued. In addition, Puerto Rican
Emilies encourage intimate personal relationships with extended family members (Canino
et al., 1980). The typical extended Emily consists of cousins, grandparents, neighbors,
“padrinos” (godparents) and other friends who have become part of the Emily even
though they are not biologically related. This Emily structure and relationships among
members of extended Emily in Puerto Rico represent the inﬂuence of African values on

the formation of present-day Puerto Rican culture.

3
When the Spanish colonizers arrived in Puerto Rico in 1493, they had a major

impact on the establishment of Spanish as the main language and of Christianity as the
predominant religion. However, the inﬂuence of the Aﬁ'icans and Tainos on the religious
beliefs of Puerto Ricans is also evident in the emphasis on spiritualism. Spiritualism is
rooted in the belief that certain persons in this world can establish contact with spiritual
forces and can use this ability to restrain unfavorable actions of evil spirits or encourage
the Evorable actions of good spirits (Locke, 1992). This belief in the power to
communicate with spirits is evident in Puerto Rico today. The traditional Catholic saints
are imbued with the power to inﬂuence the Etes of their town inhabitants. Saints are seen
as religious patrons of diﬂ‘erent towns around the island.

Spanish Catholicism permeates the traditional sex roles that exist in Puerto Rico.
The concept of ‘marianismo’ suggests that women should maintain their virginity for the
honor of the Emily and as a sign of respect for motherhood. Conversely, men’s sexual
activity is not important as long as they are good providers for Emily members. These
gender roles have also inﬂuenced the childrearing practices used among Puerto Rican
parents.

In the Puerto Rican culture there is a distinct difference between prescribed
behaviors for males and females. These gender related behaviors are similar to those seen
in Aﬁ'ican culture. On the one hand, girls are expected to be quiet, submissive, and giving,
while boys are expected to be respectful but active, aggressive, and daring (Alvarez,
1992). When girls grow up they are expected to be the anchors of the Emily, but they

have less social ﬁeedom and less control over decision making in the home than men

4
(Alvarez, 1992). Men, on the other hand, tend to have an authoritarian role in the Emily

and have unlimited ﬁeedom to socialize outside the home. However, regardless of these
differences in childrearing practices for girls and boys, respect for elders and authority
ﬁgures is highly valued for all Puerto Ricans. This respect for elders and authority ﬁgures
reﬂects both the Spanish and Aﬁican inﬂuence on Puerto Rican values.

During the last decade, Puerto Rico’s culture has also been inﬂuenced by the
United States. The United States invaded the island during the Spanish-American War in
1898. In 1917, Puerto Ricans were granted United States citizenship. Currently, Puerto
Rico is a Commonwealth of the United States. This sociopolitical relationship between
Puerto Rico and the mainland has facilitated the movement of Puerto Ricans to and ﬁ'om
the mainland with the full rights of naturalized citizens (Locke, 1992), a situation that is
unique to Puerto Ricans. The United States’ inﬂuence in the Puerto Rican community is
mostly seen in two broad areas: a) the industrialization of the island, and b) the
implementation of English as a second language in the schools, the workplace, and the
media The industrialization as implemented by the United States has resulted in an
increase in the presence of women in the workplace. This increase subsequently
inﬂuenced the role of women in the society, which has become perhaps the strongest
indication of the inﬂuence of the United States’ values on Puerto Rican culture.

The cultural characteristics outlined above are important to consider when
developing measures of interpersonal traits, such as Wiggins (1979, 1989) Circumplex
model, for use with a sample ﬁ'om the island of Puerto Rico. Speciﬁcally, although the

literature of interpersoml traits and Circumplex models of interpersonal behaviors has

5
evolved over a period of ﬁfty years (Benjamin, 1987; Freedman et al., 1951; Kiesler, 1982;

Leary, 1957; Wiggins, 1979), to my knowledge, no research has investigated the
applicability of interpersonal theory to the Puerto Rican culture. There is also no literature
base that speciﬁcally addresses the circumplex model of interpersonal traits and its
applicability to the Puerto Rican population. The present study was designed to address
this deﬁcit. To adequately accomplish this, a discussion of interpersonal theory is

necessary.

CHAPTER 2
INTERPERSONAL THEORY

Harry S. Sullivan’s (1953) interpersonal theory is based on the notion that
behavior exhibited by isolated individuals is an inadequate model for the study of
personality. Sullivan deﬁnes personality as “the relatively enduring patterns of recurrent
interpersonal situations which characterize human life” ( p. 110-111). According to
Sullivan, personality is only manifested in interpersonal interactions (Carson, 1991).
Sullivan and other interpersonal theorists assert that personality emerges within the
context of interpersonal relations with real and/or Entasized personiﬁcations of signiﬁcant
others (Carson, 1969; Sullivan, 1953). Contemporary interpersonal theorists continue to
maintain this perspective of personality by studying persons in social situations (Benjamin,
1987; Pincus, 1993; Wiggins, 1982).

Central to the notion of personality is the concept of the “self -system,” which
Sullivan views as emerging out of early interactions with signiﬁcant others (Carson, 1969;
Horowitz & Vitkus, 1986; Kiesler, 1982). Carson (1991) summarized Sullivan’s notion of
the development of the self-system and its relation to the interpersonal traits that
diﬂ‘erentiate people as follows: “Early interactions lay down memory traces, oﬂen aﬂ‘ect-
laden, imprecise, inarticulate and schematic in quality, that are activated by current
interpersonal situations and determine to a greater or lesser extent both experience and
behavior in those situations” (p. 186). Thus, both the behavior and the perception of
others’ behaviors toward one appear strongly affected by our concept of who we are

(Pincus, 1993), in other words, by our developed self-system.

7

One of the self-system’s major functions is to maintain a sense of security and to
avoid anxiety through the elicitation of a limited class of interpersonal reactions from
others (Kiesler, 1982). More speciﬁcally, when individuals interact with others, the self-
system is maintained by selectively unattending to information that is not consistent with
one’s view of oneself (Carson, 1969; Kiesler, 1982). This selective ﬁltering serves as a
mechanism to avoid anxiety (Carson, 1969). Sullivan referred to the need to avoid anxiety
and the need to relieve tension as the needs for interpersonal security and satisEction,
respectively. Sullivan’s theoretical assumptions have primarily been researched by
personality researchers who focus on circular representations of interpersonal behaviors
(Benjamin, 1987; Leary, 1957; Kiesler, 1982; Wiggins et al., 1988). Thus, as discussed

below, Sullivan’s assumptions can be described within the context of interpersonal

circumplex models.
Circmlex Models of Intgper_sonal Behavior

The circumplex model in its original form was inﬂuenced by Symonds (1939) and
Sullivan’s (1953) interpersonal theory. The representation of interpersonal behaviors in a
circular model emerged ﬁom observations of people in group psychotherapy. The goal of
researchers who designed the model was to deﬁne “interpersonal mechanisms” of overt
behavior arising during group interactions. From their efforts, a list of interpersonal
mechanisms was developed and presented in a circular continuum (La Forge & Sucze,
1956; Leary, 1957).

The literature on diﬁ‘erent types of circumplex models of interpersonal behavior is

extensive. Models described in the literature include Benjamin’s Structural Analysis of

8
Social Behavior (1987), Kiesler’s 1982 Interpersonal Circle (1983) and Wiggins’ et al.,

Interpersonal Circumplex model (1988). Despite their differences, these models have
similar theoretical assumptions based on Sullivan’s interpersonal theory. Thus, the
literature often refers to this body of work as research on the ‘Interpersonal Circle’
(Carson, 1991; Kiesler, 1983; Wiggins, 1982; Wiggins & Broughton, 1985). For the
purpose of this study, however, Wiggins’ circumplex model of interpersonal traits will
serve as the primary focus.

This model is widely accepted. It is inﬂuential in its contribution to psychotherapy
interventions, personality research, and in the development of diagnostic criteria for
personality disorders in the DSM-IV (Carson, 1991; Carson, 1969; Horowitz & Vitkus,
1986; Pincus, 1993; Pincus & Wiggins, 1990).

Wiggins’ circumplex model of interpersonal traits conceptualizcs traits in two
diﬂ‘erent ways: (a) as descriptions of how a person behaves in particular situations, and (b)
as descriptors of the intrapsychic process that affect how a person perceives the
environment (F under, 1991). This circumplex model is based on Sullivan’s notion that
interpersonal behaviors represent a blend of two basic motivations: status and aﬂiliation
(Kiesler, 1986). In other words, when two people interact they negotiate how ﬁiendly or
hostile they will be with each other, and how ‘in control’ each of them will be while
interacting. The two main orthogonal Ectors of the circumplex model are represented by
the bipolar axes of dominance vs. submissiveness and hostility vs. aﬁiliation. All
interpersonal behaviors can be described as joint functions of these dimensions (Wiggins,

1982). For example, arrogant behavior is a blend of dominant and hostile patterns of

9
interpersonal traits. As seen in Figure 1, Wiggins’ circumplex model contains eight

interpersonal trait variables arranged around the two main axes. Interpersonal traits
speciﬁcally describe enduring patterns of interpersonal behaviors characteristic of an
individual; that is, patterns which are consistent cross-situationally (Pincus, 1993).
Interpersonal traits in the circumplex model are best described within the context
of circular causality (Brokaw & McLemore, 1991). Circular causality refers to two
people interacting and mutually inﬂuencing each others’ behaviors. In other words, the
behavior of each person is seen as both a response to, and a stimulus for, the others’
behavior (Kiesler, 1983, 1988). In the circumplex model of interpersonal traits, this
circular causality is referred to as complementarity. Speciﬁcally, complementarity is based
on the assumption that a person’s interpersonal behavior tends to initiate or invite from
others reactions that lead to repetition of the original patterns of behavior (Kiesler, 1986).
Complementarity is evident among the two main axes of the circumplex, the horizontal
and vertical axes (see ﬁgure 1). Dominant behaviors elicit submissive responses ﬁ'om
interactants and, vice versa, submissive behavior elicits dominant responses. With regards
to the horizontal axis, hostile behaviors elicit hostile responses and amliative behavior
elicits ﬁiendly responses (Carson, 1991). For example, if Person A is ﬁ‘equently
interacting in a sly and tricky way a complementary response would be distant and ,
dissocial. This notion of complementarity has been challenged by research that indicates a
high frequency of hostile-dominant behaviors eliciting hosite-dominant responses (Orford,
1986). However, Pincus (1993) noted that is important to consider that complementarity

is a dynamic principle; therefore, it is not simply based on observable behavior but it

10
includes intrapsychic processes that inﬂuence the individual’s perceptions of their

interpersonal world.

Research with Wiggins’ circumplex model of interpersonal traits indicates that the
interpersonal variables represent a speciﬁc pattern of correlations (McCrae & Costa,
1989). Adjacent variables are highly correlated and variables opposite to each other are
inversely correlated. Thus, the interpersonal variable ‘warm’ is inversely correlated to
‘cold-hearted’. As discussed below, this aspect of the circumplex is particularly beneﬁcial
for assessment purposes.

The Circmplex and its Usefulness for Dggn' osmg' and Understandm' g chhopathology

Wiggins’ interpersonal circumplex model can provide a taxonomy to guide
assessment and diagnosis. An example of how this is of value is in its use with the DSM-
IV Axis II personality disorders (Benjamin, 1995; Kiesler, 1992; McLemore & Benjamin,
1979; Mggins, 1982). Personality disorders are deﬁned in the DSM-IV as “enduring
patterns of inner experience and behavior that deviate markedly from the expectations of
the individual’s culture, are pervasive and inﬂexible, have an onset in adolescence or early
adulthood, are stable over time, and lead to distress or impairment” (American Psychiatric
Association, 1994, p. 629). Research on interpersonal traits has concluded that
interpersonal dysﬁmction is central to histrionic, narcissistic, dependent, avoidant,
schizoid, and antisocial personality disorders (Benjamin, 1995; Wiggins & Pincus, 1989).

The circumplex can also be used to represent the inﬂexibility or rigidity of
personality disordered individuals. The vector length from the center of the circumplex can

be used as an indication of the intensity of an interpersonal trait (Wiggins et al., 1989).

ll

Calculation of the vector length as an index of interpersonal dysﬁmction can be measured
using a psychometrically sound instrument. One instrument that measures interpersonal
traits is the Interpersonal Adjective Scales-Revised (IAS-R; Wiggins et al., 1988). This
interpersonal measure can be given as a self-report or other’s report on the accuracy of the
64 adjectives selected for use as descriptors of the person’s interpersonal styles targeted
for assessment. The IAS-R was constructed with interpersonal adjectives that can be
scored and plotted in the circumplex model.

Inte_rp_ersonal Adjective Scales-Revised

The original Interpersonal Adjective Scales (IAS) (Wiggins, 1979) were developed
from the framework of Sullivan’s (1953) interpersonal theory. Building on Leary’s (1957)
work, the IAS were constructed to form a two-dimensional circumplex with interpersonal
trait descriptors around the coordinates of dominance and aﬁiliation (Wiggins, 1979).

The interpersonal adjectives were selected as a subset of Goldberg’s (1977) 1710
trait descriptive adjectives. From this subset, only adjectives that fell within the domain of
interpersonal traits were chosen. Interpersonal adjectives were deﬁned as those traits
reﬂected in dyadic interactions that have a relatively clear cut social (status) and emotional
(afﬁliation) consequence for both participants (Wiggins, 1979). Careful item selection and
other forms of psychometric validation resulted in the development of the IAS. The
measure consisted of 128 adjectives, which formed eight 16-item scales. Wiggins,
Trapnell, and Phillips (1988) later revised these scales and developed a 64-item short

form version known as the Interpersonal Adjective Scales-Revised (IAS-R). This test was

12
standardized on a sample (n=1 161) of college undergraduates from North American

universities (Canada and the United States).

The IAS-R is a shorter version of the IAS and contains 64 adjectives that form
eight 8-item scales. Construct Validity of the IAS-R was achieved ﬁom second order
principal components analyses of the intercorrelation matrix of the scales. The scales were
derived from ﬁrst order principal component analysis. The eight scales were labeled as
follows: Assured-Dominant (PA), Arrogant-Calculating (BC), Cold-Hearted (DE), Aloof-
Introverted (F G), Unassured-Submissive (HI), Unassuming-Ingenuous (IK), Warm-
Agreeable (LM), and Gregarious-Extraverted (NO). The two-word labels are meant to
include terms that share the same semantic features. The two-letter abbreviations are
arranged counterclockwise in alphabetical order for organizational purposes only. Table 1
describes the adjectives that belong to each interpersonal scale.

The internal consistency indices, as calculated by the coeﬂicient alpha values of
each of the eight scales are as follows: assured-dominant, .84; arrogant-calculating, .85;
cold-hearted, .80; aloof-introverted, .85; unassured-submissive, .83; unassuming-
ingenuous, .75; warm-agreeable, .86; and gregarious-extraverted, .85.

The convergent and discriminant validity of the IAS-R has been documented (see
Wiggins & Broughton, I985). For example, there is a correlation of .80 for the
“assured-dominant” scale and Bem’s masculinity scale (1974). Correlation coefﬁcients of
-.77 emerged ﬁ'om comparisons of the “aloof- introverted” scale with the sociability scale

of Buss and Plomin (1975).

13
How the IAS-R responses relate to actual interpersonal behavior was provided by

Giﬁ‘ord and O’Connor (1987). Their results indicated that preferred interpersonal space
and several conversation belmviors, as measured by Bales’ (1970) Interpersonal Process
Analysis, can be mapped in the interpersonal circumplex model (Giﬂ‘ord & O’Connor,
1987). Nonverbal behaviors have also been demonstrated to be related to interpersonal
traits represented in the circumplex (Giﬂ‘ord, 1991 ). For example, Giﬂ‘ord’s (1991)
ﬁndings indicated that subjects who score high on the gregarious-extraverted scale turn
their head toward others more, and wrap their arms less than aloof subjects. These
studies suggest that the IAS-R has good external validity.

Overall, there is evidence that the IAS-R is an adequate measure of interpersonal
traits and its eight scales represent a clear circumplex model of interpersonal traits as
suggested by the theory. However, as mentioned earlier, the IAS-R was developed using
a sample of college students ﬁ'om North American universities. Therefore, the
applicability of the IAS-R with a culturally distinct population such as Puerto Ricans has

not been established.

CHAPTER 3
WHY EXAMINE THE IAS-R STRUCTURE WITH A PUERTO RICAN SAMPLE
The existence of the eight interpersonal traits with similar structure as the

IAS-R inventory has not yet been investigated with a Puerto Rican sample. Although
interpersonal traits could be considered a universal phenomenon, diﬂ’erent cultures vary
in their norms of appropriate interpersonal behavior. Thus making the above question an
important one because the Puerto Rican culture differs in theoretically important ways
(e.g. language, community oriented society, traditional gender roles) ﬁ'om the United
States’ and Canada’s culture where the IAS-R was developed.

The ﬁrst goal of this study was to determine whether Mggins' circumplex model
and the Ector structure of the IAS-R could be replicated in a Puerto Rican sample. To
accomplish this, the factor structure of the Puerto Rican version of the IAS-R, the Escala
de Adjetivos Interpersonales (EAI) was examined. The questions that addressed the
psychometric properties of the EAI were the following:

a) Does the EAI contain the eight scales of interpersonal traits that compose the

IAS-R?

b) Are the ﬁrst order Ector structure and the second order Ector structure of the

EAI scales similar to those suggested by Wiggins' circumplex model of

interpersonal traits?

Answers to these questions provided the irrfornmtion necessary to examine the
applicability of Wiggins' circumplex model of interpersonal traits, as measured by the
EAI, on a Puerto Rican sample. Thus, the IAS-R was considered replicated if the EAI

eight scales loaded in a similar pattern as the IAS-R and if the second order Ector

l4

15

analysis revealed two main orthogonal Ectors similar to the Dominance and Aﬁiliation
Ectors of the IAS-R

A secondary goal of this study was to explore the pattern of responses of the
Puerto Rican sample in each of the EAI scales. The previously discussed cultural values
and norms of Puerto Ricans contribute to determine the interpersonal traits considered
appropriate in that society. Thus, one could speculate that Puerto Rican responses on the
scales of interpersonal traits would be consistent with the norms of interpersonal
behavior. To explore the pattern of responses among Puerto Rican college students the
following set of questions were addressed:

1) Will Puerto Rican college students endorse 'gregarious- extraverted' and

'warm-agreeable' interpersonal traits more than 'aloof-introverted' and 'cold-

hearted' traits?

2) Will Puerto Rican college women score signiﬁcantly higher in the

'unassured/submissive' interpersonal trait scale, than Puerto Rican men?

3) Will Puerto Rican college men score signiﬁcantly higher in the

'assured/dominant' interpersonal trait scale than Puerto Rican women?

CHAPTER 4

METHODS
Sampling and data collection procedures

As part of a project at Pennsylvania State University, the sample was collected

from two major universities located in the northeastern region of the island of Puerto
Rico. These institutions were the University of Puerto Rico and the Interamerican
University at Rio Piedras campus. These universities are located in the metropolitan
areas of Puerto Rico and students ﬁ'om all the different towns of the island attend these
two main campuses. Thus, student population is representative of college students
throughout the Island.

With permission from professors and oral informed consent ﬁ'om the students,
data were collected from undergraduate psychology classes. The EAI was administered
during class time. No monetary or other rewards were offered for participating in the
study.

.Su_biec_t§

The sample consisted of 210 Puerto Rican college undergraduate students, ages
17 to 35 (M=24; SD = 5.1 ), attending the University of Puerto Rico and the
Interarnerican University in Rio Piedras. Oﬁhe 210 subjects, 76% (n=158) were women
and 24% (n=52) men. This gender ratio can be explained, in part, by the Ect that in
Puerto Rico, twice as many women as men attend major universities.
M1113.

The measure used in this study was the translated Puerto Rican version of the

IAS-R( i.e. Escala de Adjetivos Interpersonales or EAI; see Appendix A). I developed

16

17

the EAI via a series of back translations (Draguns, 1982), and designed it to reﬂect the
idiomatic expressions of Puerto Ricans. Bilingual experts in scale construction on the
US. mainland and the island of Puerto Rico were consulted to ensure correct translation.
Table 2 includes the Puerto Rican version of the eight items on each the eight
interpersonal adjective scales.

The EAI is patterned after the IAS-R (see Appendix B). Both contain 64
adjectives, on which respondents rate themselves according to the descriptive accuracy of
each adjective. The ratings are done on an eight-point Likert scale ranging ﬁ‘om
1 (extremely inaccurate) to 8 (extremely accurate). Like the IAS-R, the EAI contains a
glossary that provides deﬁnitions of each of the 64 adjectives. Respondents were able to

refer to this glossary if they had questions about the meaning of any of the adjectives.

CHAPTER 5
RESULTS
Data Analyses
Two sets of analyses were done on the EAI to assess its psychometric properties
and to determine whether its scales matched those of the IAS-R.

The ﬁrst wave of analyses grouped the EAI items according to the format of the
IAS-R scales. The ﬁrst step was to test the reliability of these scales. Thus, the internal
consistency of the EAI scales was tested by calculating the coefﬁcient alpha values
(Cronbach, 1951). Having done this, the second step was to perform conﬁrmatory Ector
analysis of the intercorrelation matrix of the eight EAI scales to examine its Ector
structure. Indices of ﬁt determined if the two factor structure (i.e. Dominance and
Aiﬁliation) was an appropriate model for the EAI scales. The indices of ﬁt utilized were
the goodness of ﬁt index (GFI), Bentler and Bonett's (1980) Non-normed Index (NFI) and
the root-mean-square residual estimate (RMR).

The second wave of analyses tested the factor structure of the EAI without
groupiing the items in the same format of the eight IAS-R scales. In other words, the
items were not forced in the scales suggested by the IAS-R format. First and second
order principal component analyses were performed on the 64 items of the EAI to
. determine whether eight scales would emerge from the Puerto Rican sample and if their
contents would match those of the IAS-R. The scree test and the 21 eigenvalue criteria
were used to determine the Ector groupings to be retained for rotation. These Ectors
were then rotated according to the orthogonal varimax criteria. Those items with Ector

loadings of .30 or above were deemed as belonging to that particular scale.

18

chhomtrﬂromnies of the EAI
Internal consistency. As part of the ﬁrst wave of analyses the alpha coemcient
values (Cronbach, 1951) of internal consistency were computed for each of the eight EAI
scales. Kaplan and Saccuzo (1982) criteria for adequate coeﬂicient alpha values of z .70
was used to determined if the EAI scales had appropriate internal consistency. As part of
the reliability analysis the item total correlations for each EAI scale were examined.
According to Nunnally (1978) item-total correlations above .30 are considered good.
Based on this criterion, four of the items of the EAI scales were dropped because of their
low item-totalcorrelation with other items in the scales. The interpersonal adjectives with
low item-total correlations were: 'self-conﬁdent' (conﬁado(a) en si mismo (a); r = .14)
from theAssured/Dominant scale, 'craﬂy' ( listo(a); r = .13) from the Arrogant/Calculating
scale,'iron-hearted' (fuerte de caracter; r = .17) from the Cold-hearted scale, and
'unaggressive' (gentil; r = .04) from the Unassured/Submissive scale. The
resulting alpha values for each of the EAI scales are illustrated in Table 3. As
demonstrated in Table 3, coeﬁicient alpha values (Cronbach, 1951) for each of the eight
EAI scales indicated that the Umssuming/Ingenuous scale was the only scale that die not
have adequate internal consistency (alpha= .47). Overall, this analysis suggested that
seven out of the EAI eight scales have adequate internal consistency.
Conﬁnmtog Factor Mysis. The second part of this ﬁrst wave of analyses was
to examine if the EAI scales form a two Ector structure similar to that of the IAS-R.
Conﬁrmatory Ector analysis of the intercorrelation matrix of the EAI scales was

performed (see Table 4). Conﬁrmatory Ector analysis using the indices of ﬁt

20

aforementioned indicated that the two Ector structure is not an adequate model for the
eight EAI scales (GFI= .75; NFI= .21; RMR= .33). Thus these indices suggested that the
factor structure of the EAI scales is not similar to that of the IAS-R scales.

The next step was to examine which Ector structure could better represent the
EAI scales. Principal components analysis using both the scree test and the eigenvalue
21 criteria suggested the retention of three Ectors. Thus, three Ectors were rotated
according to the orthogonal varirmx criteria. Scales with Ector loadings of .30 or above
were deemed as belonging to that Ector. Table 5 indicates the Ector loadings of the
scales on the three Ectors identiﬁed. The eigenvalues for the ﬁrst three Ectors were 2.29,
2.08, and 1.66 respectively. The ﬁrst Ector included the following four scales:
arrogant /calculating (BC), cold-hearted (DE ), aloof / introverted (FG), and assured/
dominant (PA). This ﬁrst Ector was labeled 'Dominance-Cold' because the scales
represent a combination of dominant and cold-hearted interpersonal traits. The second
Ector included two scales: gregarious/ extraverted (N O) and warm/ agreeable (LM). On
the basis of the content of the scales this second factor was labeled 'Aﬁiliative'. The
third Ector included two scales: unassured/ submissive (HI) and unassuming/ ingenuous
(JK) scale. This Ector was labeled 'Submissive' .

In summary, these analyses suggested that the EAI contains seven out of the
eight IAS-R scales of interpersonal traits. Furthermore, the second order factor structure
of the EAI scales do not replicate the two Ector structure of the IAS-R.

Princm’ al commnent analysis. The second wave of analyses consisted of ﬁrst
order and second order principal components performed on the 64 items of the EAI.

Using the scree test and the eigenvalue 3 1 criteria, the ﬁrst order principal components

21

suggested the retention of seven Ectors. These seven factors were then retained and
rotated according to the orthogonal varimax criteria. Items with Ector loadings of 2 .30
were deemed as belonging to that scale.

Table 6 indicates the content and Ector loadings of the seven scales that
emerged. To compare the content of these scales with the original IAS-R scales, those
items with loadings of .30 or above on the EAI scales were compared to those items that
loaded on the matching IAS-R scales. If 85% (see Verhulst et al., 1988) of the items
loading on the EAI scales were similar to the items loading on the IAS-R, then it was
determined that the scales were replicated. For all of the seven EAI scales, less than 85%
of the items loading EAI match the items on the IAS-R scales. This indicated that the
IAS-R eight scales were not replicated. Thus suggesting that the content of the EAI
scales that emerged differ in structure from the IAS-R scales.

The seven scales were labeled according to the semantic characteristics of the
adjectives in each of the scales. The labels for the scales were: assured (e.g. self
conﬁdent, assertive, enthusiastic), arrogant/calculating (e. g. sly, cruel), cold-hearted (e. g.
antisocial, dissocial, warmthless), introverted (e.g. timid, distant), dominant (e.g. ﬁrm,
persistent, forceful), warm/agreeable (e.g. accommodating, charitable), and
gregarious/extraverted (e. g. neighbourly, ﬁiendly). These results indicated, as was the
case in the ﬁrst wave of analyses, that the scale of Unassuming/Ingenuous did not
emerge in the EAI. Furthermore, in this analysis dominant-related adjectives fornwd a

different scale ﬁ'om the ‘assured' scale.

22
Next, I was interested in examining the second order structure of the EAI scales to

examine whether they would form a two Ector structure. Second order principal
component analysis of these seven EAI scales was performed using the scree test and the
_>_ 1 eigenvalue criteria to determine how many Ectors to retain. Results suggested the
retention of two Ectors. These two Ectors were rotated according to the orthogonal
varirnax criteria. Table 7 indicates the Ector loadings of the seven scales on the two
Ectors. The structure of the two Ectors was determined based on those scales with
loadings of z .30 on a single Ector. The Ectors were labeled according to the semantics
of the scales grouped in one Ector. The ﬁrst Ector included the following four scales:
arrogant/calculating, cold-hearted, introverted and dominant. This Ector was labeled
'Dominance-Cold.’ Three scales luid Ector loadings of z .30 on the second Ector.
These scales were: warm-agreeable, gregarious/extraverted, and assured. This second
Ector was labeled 'Aﬁiliation.’ These two Ectors that emerged in this second order
principal component analysis differed in content from the original two Ector structure of
the IAS-R. The scales themselves are different ﬁ'om the eight IAS-R scales.

Furthermore, in the IAS-R, the factor labeled 'Dominance’ contained two main
scales:assured/dominant and unassured/ submissive. The factor labeled 'Aﬁiliation'
contained the scales cold-hearted and warm/agreeable. This was not the case for the EAI
two Ector structure (see Table 7).

Overall, these results suggested that while a two Ector solution emerged ﬁ'om
the second wave of analyses, there are differences between the EAI and the IAS-R
structure. That is, the ﬁrst order Ectors did not match the scales of the IAS-R. In both

analyses, the scale 'Unassuming/Ingenuous' did not emerge; thus suggesting that the EAI

23

contains seven scales of interpersonal traits as opposed to the eight scales of the IAS-R.
Furthermore, the second order two Ector structure of the IAS-R was not replicated in the

EAI as indicated by the conﬁrmatory factor analysis and ﬁrst and second order principal

component analyses.

Cultural Inﬂuences on the EAI Remnses

A secondary goal of this study was to explore whether Puerto Ricans' responses
on each of the EAI scales reﬂect a speciﬁc pattern that parallels the cultural norms. This
analyses were done with the EAI scales grouped in the format of the IAS-R.

The ﬁrst exploratory question was whether Puerto Rican college students
endorse ‘gregarious- extroverted' and 'warrn-agreeable' interpersonal traits more than
'aloof-introverted' and 'cold-hearted' traits. Diﬂ‘erences in these scales' scores were tested
via Chi-Squares. For the Chi-Square analyses scale scores above 4 were considered high
and scale scores 3 4 were considered low. Results suggested that Puerto Rican college
students scored signiﬁcantly higher in the gregarious/extroverted scale as compared to the
aloof/introverted scale (p= .013). For the warm/agreeable scale as compared to the
cold/hearted scale no students scored high in the cold/ hearted scale and no students
scored low in the warm/agreeable scale.

The other two exploratory questions were: a) whether college women score
signiﬁcantly higher in the 'unassured/submissive' interpersonal trait's scale than Puerto
Rican men and b) whether Puerto Rican college men score signiﬁcantly higher in the
'assured/dominant' interpersonal trait's scale than Puerto Rican women. To determine

gender effects on the score in these scales T-tests were used with an alpha of .05. Results

24

indicated no signiﬁcant diﬁ'erences between Puerto Rican college men and women score

in the unassured/ submissive nor the assured/dominant scale.

CHAPTER 7
DISCUSSION
This study was designed to determine whether the Puerto Rican version of the

Interpersonal Adjective Scales Revised (IAS-R), the 'Escala de Adjetivos
Interpersonales' (EAI), replicated the Ector structure of the IAS-R.

Two waves of analyses were performed. The ﬁrst wave of analyses grouped the
EAI items according to the IAS- R format. These analyses indicated that seven out of the
eight EAI scales had adequate internal consistency. The 'unassuming-ingenuous' scale
had low internal consistency (coefﬁcient alpha= .47; Cohen, 1951). The items included
in this scale were: undemanding, uncalculating, uncunning, unsly, boastless,
umrgumentative, unwily, and uncratty. These negatively worded (e. g. unsly) items are
used to describe the opposite of a trait (e.g. arrogant/calculating). In this study, one word
adjectives were used in the EAI to maintain the same format of the IAS-R Thus, careful
precautions were taken when deciding the Spanish adjective with equivalent semantic
meaning to the English version. However, it is possible that the use of the one word
adjective to describe such interpersonal traits (e. g. undemanding) is linguistically
inappropriate. The 'unassuming-ingenuous' scale may have emerged had longer phrases
been used in place of the one word adjectives. Future studies must examine the reliability
' of this scale by using phrases instead of one word adjectives.

When these scales, arranged according to the IAS-R format, were submitted to
conﬁrmatory Ector analysis the results indicated that the two Ector structure of the

IAS-R was not replicated in the EAI. Instead, a three Ector structure was indicated to be

25

26
a better model, statistically for the EAI. These three Ectors, labeled on the basis of the

semantic meanings of the adjectives, were 'Dominance-Cold', 'Amliation' and
'Submissive' It must be underscored that this factor structure may not necessarily be an
indication of cultural differences but instead a result of measurement error. In order to
examine whether this could be an artifact of measurement error it would be necessary to
collect a US. sample of 210 college undergraduates and examine the factor structure of
the IAS-R. I

Keeping in mind cultural differences, a second wave of analyses was performed to
further explore the possible structure of the EAI. Instead of grouping the EAI items in
the similar format of the IAS-R scales, principal component analyses were performed on
the 64 items of the EAI. Similar to the ﬁrst analyses, the EAI emerged with seven scales
of interpersonal traits as opposed to the eight IAS-R scales (Table 6). The 'unassuming-
ingenuous' scale did not emerge in this analysis. Again, this could be due to the use of
one word adjective to describe negatively worded adjectives which are not common
descriptors of interpersonal traits among Puerto Ricans. When these seven scales were
compared to the seven IAS-R matching scales, results suggested that the content of these
EAI scales differed in structure from the IAS-R For example, the 'Assured’ scale of the
EAI contained adjectives with positive loadings such as self-conﬁdent, self-assured,
jovial, enthusiastic, cheerful and adjectives with negative loadings such as meek, bashful,
shy, unbold, andforceles. In contrast to the ’Assured-dominant' scale of the IAS-R
which contains items such as self-assured and self-conﬁdent together with adjectives that
imply domimnt qualities (e. g. domineering, forceful, persistent, ﬁrm), the EAI did not

include dominant adjectives in the 'Assured' scale. In the EAI, these dominant-related

27

adjectives formed their own scale. Perhaps, these results suggests that Puerto Ricans
perceive being self-assured as different from being dominant. Thus indicating that culture
may inﬂuence the structure of interpersonal traits is some ways.

When these seven scales were submitted to second order principal component
analysis, they form a two Ector structure. The structure of this second order factors was
different ﬁ'om the IAS-R. The two Ectors were labeled: 'Dominance-Cold' and
'Aﬁliation'. Similar to the previous wave of analyses, the Ector 'Dominance-Cold'
suggested a pattern different from the IAS-R's 'Dominant’ Ector. The 'Dominant' Ector
on the IAS-R contains positive Ector loadings from the 'assured-dominant' scale and
negative loadings from the 'unassured-submissive' scale. In contrast,the 'Dominance-
Cold' Ector of the EAI contains the following four scales: 'arrogant-calculating', 'cold-
hearted', 'introverted' and 'dominant‘. This particular pattern, which also emerged in the
ﬁrst wave of analyses, may suggest that for Puerto Ricans, arrogant and cold-hearted
interpersonal behaviors are considered traits associated with a dominant trait.

Casting a critical eye on these Ector ﬁndings, it may be argued that the Ector
structure of the EAI may be an artifact of: a) this study sample size, b) the ratio of females
to males, c) the restriction of the sample to college undergraduate students or
(1) measurement error.

These methodological issues could be considered limitations of this study. The
sample size of 210 is smaller than those of many investigations with the IAS-R. The larger
percentageof women in this study may limit the generalizability of these ﬁndings to Puerto

Rican women. Furthermore, the college sample may not be representative of the Puerto

28
Rican population in general. Even though on these analyses the Ector structure of the

EAI is not exactly similar to the IAS-R, there is the possibility that it is not all due to
cultural diﬁerences but to random error issues of measurement. In order to make ﬁnal
conclusions about these differences ﬁrrther studies must be performed with equal sample
size from the U.S.and Puerto Rico.

In addressing the secondary goal of this study, the responses of Puerto Ricans on
each of the EAI scales were examined to determine whether there was a pattern that
paralleled the cultural norms. Using the EAI scales grouped in the IAS-R original format,
analyses indicated that Puerto Rican college students scored signiﬁcantly higher in the
'gregarious/extraverted' and 'warm-agreeable’ scale as compared to the
'aloof/introverted' scale (p= .013). This pattern supports the speculation that cultural
values (e. g. strong Emily bonds and interdependency with community) inﬂuenced Puerto
Ricans to endorse warm, agreeable, and gregarious interpersonal traits.

The next exploratory questions addressed whether there were gender differences
among the responses of Puerto Rican college students. No signiﬁcant differences were
found between Puerto Rican college men and women with regards to the 'assured-
dominant' scale and the 'unassured-submissive' scale. This can be interpreted in one of
two ways: a) the restriction of the sample to a college sample may have included men
and women with more liberal attitudes (e. g. androgenous gender roles) than the general
Puerto Rican population or, b) the inﬂuence of the United States in the last century has
had an impact in the gender roles of Puerto Rican men and women such that Puerto Rican
women and men perceive themselves as equal with regards to being dominant and/or

submissive.

29

In summary, the current study was the ﬁrst attempt to use Wiggins' et a1. (1988)
IAS-R to assess the Ector structure of interpersonal traits for a Puerto Rican sample.
Research is needed to further evaluate the psychometric characteristics of the EAI. The
limitations of this study may be circumvented by using: a) a larger more representative
sample from the community with a balance ration of male/female participants, b)
possrhly the use of phrases instead of one word adjectives were it may be necessary or c)
collect another sample of Puerto Rican and US undergraduates and examine the
similarities and differences of the factor structure of the EAI and IAS-R further.

Despite its limitations, this study has provided a foundation for further research
withinterpersonal traits and other personality related issues among Puerto Ricans. In
addtition, the study adds a potentially useful cultural component to the growing body of

research on the measurement of interpersonal traits.

FIGURE

 

 

PA
.8
s1.
‘35, Aeeu mitten! ‘5.
as O NO
Gregerloue-
O Extreverted
t- " ‘
Celculetlno
Celd- Verm-
Heerted ADMIN.
”- I _._J 0‘
E «LO-I. .A .4 .e L“
Aloot- - °-‘ Uneeeumlng-
Introverted Ingenuoue
0
F6 .K
225' "maize-Submissive 3'5.
J.
Ht
270'

Figure 1 Circumplex Structure of the Interpersonal Adjective Scales

Revised

TABLES

TABLE 1

Set Of Interpersonal Adjective Scales -Revised (IAS-R)

Assam
3 Assertive

17 Dominant

21 Self-assured

32 Domineering

38 Foreeful

42 Persistent

46 Self-conﬁdent

50 Firm

maximum:
24 Cunning

30 Crafty

48 Boastful

54 Wily

55 Calculating

57 Sly

61 Cocky

64 Tricky

W

11 Coldhearted
12 Ruthless

19 Iron-hearted
22 Cruel

26 Uncharitable
35 Unsyrnpathetic
53 Hard-hearted
59 Warmthless

Alene-Intranet“!
l Introverted

13 Dissocial

18 Anti-social

23 Unsparkling

45 Unneighbourly

52 Unsociable

56 Uneheery

60 Distant

II l-S l . .
4 Unauthoritative

9 Shy

25 Meek

28 Unaggressive

36 Timid

37 Unbold

49 Bashful

63 Forcelees

W
2 Undemanding

5 Uncalculating

10 Uncunning

27 Unsly

31 Boastless

33 Unargumentative

39 Unwily
51 Uncrafty

Elam-Agrees”:
6 Aecomodating
7 Kind
8 Charitable

14 Tender-hearted

15 Soft-hearted

34 Tender

41 Gentle-hearted

62 Sympathetic

WM
16 Cheerful

20 Enthusiastic

29 Jovial

40 Extraverted

43 Perky

44 Friendly

47 Outgoing

58 Neighbourly

TABLE 2

Set of A djective Scales for the EAI

Segumtahllcminante
3 Assertive
17 Dominante

21 Seguro(a) de si mismo

32 Mandon
38 Foraoso(a)
42 Insistente

*46 Confiado(a) en si mismo

51 Terco(a)

Amateurs-simmer

24 Malicioso(a)
*30 Listo(a)

48 Presumido(a)
54 Misterioso(a)
55 Calculador (a)
57 Enganoso(a)
61 Pretencioso(a)
64 Tramposo(a)

Hostil
ll Insensible
12 Despiadado(a)
*9 Fuerte de Caracter
22 Cruel
26 No caritativo(a)
35 Incompasivo(a)
53 Duro de Corazon
59 No afectuoso(a)

D' | | I | I. I

1 Introvertido (a)
13 Solitario(a)

18 Antisocial

23 Deeanimado(a)

45 Antipatico(a)

52 No sociable

56 Serio(a)

60 Distante

WW
4 No autoritario

9 Retraido(a)

25 Sumiso(a)

*28 Gentil

36 Timido

37 No atrevido(a)

49 Avergonzoso(a)

63 No Fonoso(a)

**

2 Modesto

5 No calculador(a)
10 No malicioso(a)
27 Honcsto

31 Humilde

33 Pasivo(a)

39 Verdadero

51 Sano

6 Servicial

7 Amable

8 Caritativo(a)

14 Tierno(a) de corazon
15 Compasivo(a)

34 Tierno(a)

41 De buen corazon

62 Simpatico(a)

S . I I E I I. I
16 Alegre

20 Entusiasmado(a)
29 Jovial

40 Extrovertido(a)

43 Animado(a)

44 Amigable

47 Sociable

58 Bien Amistoso

Note. * items with low item total correlation.

** scale with low coefﬁcient alpha

Table 3

Pachometric Characteristics of the EAI

 

 

Scale §_=_ M Coefﬁcient Alpha
(PA) Assured-Dominant
(Seguro-Dominante) 4.67 1 .27 0.70
(BC) Arrogant-Calculating
(Arrogante-Calculador) 2.65 1.12 . 0.72
(DE) Cold-Hearted
(Hostil) 1.99 0.91 0.72
(F G) Aloof-Introverted
(Distante-Introvertido) 2.84 1 . 15 0.79
(HI) Unassured-Submissive
(Inseguro-Sumiso) 3.79 l .19 .072
(JK) Unassuming-Ingenuous
(Modesto-Ingenuo) 5.87 0.83 0.47
(LM) Warm-Agreeable
(Amable-Agradable) 6.84 0.79 0.81
(N O) Gregarious-Extraverted
(Sociable-Extrovertido) 6.68 0.87 0.85

 

 

Table 4

34

I_ntercorrel_ation Matrix of the EAI Scales

Scale

PA

BC

DE

FG

LM

NO

PA

.39

.27

.05

-.27

-.05

-.10

.33

BC

.57

.13

-.30

-.08

.01

DE PG
.65 --
.24 .46
-.18 -.05
-.22 -.31
-.14 -.50

HI JK
.28 --

.01 .33
-.36 .17

LM

NO

Table 5
Factor Loadings of the EAI

Scales

(BC) Arrogant-Calculating
(Arrogante-Calculador)

(DE) Cold-Hearted
(Hostil)

(FG) Aloof-Introverted
(Distante-Introvertido)

(PA) Assured-Dominant
(Seguro-Dominantc)

(N O) Gregarious-Extraverted
(Sociable-Extrovertido)

(LM) Warm-Agreeable
(Amable—Agradable)

(HI) Unassured-Submissive
(Inseguro-Sumiso)

(JK) Unassuming-Ingenuous
(Modesto-Ingenuo)

35

Factor 1

.84

.84

.64

-.08

.11

-.I6

-.23

-.56

Factor 2

-.O6

-.16

.56

.55

.83

-.27

-.43

.29

Factor 3

-.O7

.14

.06

.26

.07

.76

.60

.60

Table 6

Factor Loadings of the EAI Sceﬁes from Principal Commnent Anaﬂsis

36

 

 

Scale Name Items Factor Loadings
Assured 46 self-conﬁdent .59
21 self-assured .55
29 jovial .52
20 enthusiastic .51
16 cheerful .48
43 perky .48
3 assertive .35
25 meek -.35
49 bashﬁil -.38
9 shy -.4O
37 unbold -.42
23 unsparkling -.53
63 forceless -.35
Gregarious/Extroverted 47 outgoing .70
58 neighbomly .70
62 sympathetic .68
44 ﬁiendly .68
51 uncraﬁy .38
40 extroverted .37
34 tender .35
Arrogant/Calculating 57 sly .64
24 cunning .57
64 tricky .52
12 ruthless .50
54 wily .50
53 hard-hearted .48
55 calculating .45
27 unsly -.45
10 uncunning -.51
22 cruel .35

37

 

 

Table 6 Continued

Scale Name Items Factor Loadings

Dominant 17 dominant .75

32 domineering .70

19 iron-hearted .65

_ 50 ﬁrm .56

42 persistent .47

30 crafty .43

38 forceful .43

4 unauthoritative -.34

Warm/Agreeable 6 accommodating .72

8 charitable .69

7 kind .57

28 unaggressive .47

31 boastless .36

39 unwily .33

15 soil-hearted .35

41 gentle-hearted .37

Cold-Hearted l8 anti-social .61

52 unsociable .52

13 dissocial .48

35 unsympathetic .46

45 unneighbourly .40

59 warmthless .38

26 uncharitable .34

Introverted 36 timid .38

l introverted .52

60 distant .47

56 uncheery .46

11 cold-hearted .42

Table 7

38

Factor Loadings of the Second Order Principal Commnents Analysis of the EAI

 

 

Scale Name Dominance-Cold Afﬁliation
Arrogant-Calculating .81 -.09
(Arrogante-Calculador)

Cold-Hearted .80 -.19
(Hostil)

Introverted .63 -.08
(Introvertido)

Dominant .45 .08
(Dominante)

Warm-Agreeable -. 14 .71

(Amable-Agradable)

Gregarious-Extroverted -.13 .68
(Sociable-Extrovertido)

Assured .33 .63

(Scam)

REFERENCES

Ammott,T.L. & Matthaei, J.A. (1991). Race, gender, and work: A multicultural
econom_i_c history of women in the United Stat_es. South End Press: Boston, MA.

Alvarez, M. D. (1992). Puerto Rican children on the mainland: current
perspectives. In A.N. Ambert & M.D. Alvarez (Eds), £u_erto ﬂan children on the

mainland: interdisciplm’ Lerspectives (pp. 3-15). New York: Garland.

 

American Psychiatric Association (1994). Dggp' ostic and statistical manual of
mental disorders (4th ed.). Washington, DC: Author.

American Psychological Association. (1993). Guidelines for providers of
psychological services to ethnic, linguistic, and culturally diverse populations. American
Pachologist, 48, 45-48.

Amrnott, T. L. & Matthaei, J.A. (1991). Race gender and work: a multicultultural
economy: history of women in the United State_s. Boston, MA: South End Press

Baca Zinn, M. (1989). Family, race and poverty in the eighties. Sigps: Journal of
Women in Culture and Society, 14, 20-24.

Bales, R. F. (1970). Perso ' and int nal behavior. New York: Holt,
Rinehart & Winston.

Bem, S. L. (1974). The measurement of psychological androgyny. Journal of
Consultarg' and Clinical Pachology, 42, 155-162.

Benjamin, LS. (1974). Structural Analysis of social behavior. Pachological
Review, _8_l, 392-425.

Benjamin, L. S. (1987). Use of the SASB dimensional model to develop
treatment plans for personality disorders. 1: Narcissism. Journal of Personality Disorders.
1, 43-70.

Benjamin, L. S. (1995). Diagnosis and treatment of personality disorders.
Journal of Personality Disorders. 1. 43-70.

Bentler, P.M. & Bonett. (1980). Signiﬁcance tests and goodness of ﬁt in the
analysis of covariance structures. Pachological Bulletm' 88, 588-606.

Brokaw, D.W. & McLemore, CW. (1991). Interpersonal models of personality
and psychopathology. In D.G. Gilbert & J.J. Conolly (Eds), PersoM’ , social skills and
Lsychopathology: an individual differences approach (pp. 49- 83). New York: Plenum.

39

40
Buss, A. H. (1989). Personality as traits. American Psychologist. 44. 1378-1388.

Buss, A. H., & Plomin, R. (1975). A temxrament theory of mrsonality
development. New York: Wiley.

Canino, I.A., Early, B.F., & Rogler, L. H. ( 1980). The Puerto Rican Child in New
York City: Stress and Meraal Heal}; New York: Hispanic Research Center.

Carson, RC. (1969). Interaction Concepts of Persotm. Chicago: Aldine.

 

Carson, RC. (1991). The social interactional viewpoint. In M. Hersonen & A.
Kazdin & A. Bellack (Eds), The Clinical Psychology Handbook (pp 185- 199), New
York: Pergamon.

Comas-Diaz, L. (1982). Mental health needs of Puerto Rican women in the
United States. In R. E. Zambrana (Ed.), Work, family and health: latina women in
transition (pp.1-10). New York, Hispanic Research Center.

Cronbach, L]. (1951). Coeﬂicient alpha and the internal structure of tests.
chhometrika, E, 297-334.

Draguns, J. G. (1982). Methodology in cross cultural psychology. In Culture and
Psychopatholoyg Journal of Cros_s Cultural Psychology, 5, 9-47.

Freedman, M.B., Leary, T., Ossorio, A.G., & Coﬂ’ey, HS. (1951). The
interpersonal dimension of personality. Journal of Personalig,ﬂ, 143- 161.

Funder, DC. (1991). Global Traits: A neo-Allportian approach to personality.
Pychological Science, 2, 31-39.

Giﬁord, R (1991). Mapping nonverbal behavior on the interpersonal circle.
of Persoﬂrty and Social Psycholo gyo, 61, 279-288

Gifford, R., & O’Connor, B. (1987). The interpersonal circumplex as a behavior
map. Journal of Personality and Social Pachology, 52, 1019-1026.

 

Goldberg, L. R (1977, August). e and ' : Develo '
taxonomy of trait descriptive terms. Invited address to the Division of Evaluation and
Measurement at the meeting of the American Psychological Association,

San Francisco, CA

Horowitz, L.M., & Vitkus, J. (1986). The interpersonal basis of psychiatric
symptoms. Clinical chhology Review, 6, 443-469.

41

Kaplan, R. M. & Saccuzzo, D. R. (1993 I. Psychological Test'arg: Principles,
Applicatioa and Issues, (3rd ed). Paciﬁc Grove, CA: Brook Cole Publishing Co.

Kiesler, DJ. (1982). Interpersoal theory for personality and psychotherapy. In
J .C. Anchin and DJ. Kiesler (Eds), Handbook of interp_ersonal pgchotherapy (pp. 3-24).
Elmsford, NY: Pergamon Press.

Kiesler, DJ. (1983). The 1982 Interpersonal Circle: A taxonomy for
complementarity in human transactions. Pachological Review 90 185-214.

,_—9

Kiesler, DJ. (1986). Interpersonal methods of diagnosis and treatment. In R.
Michels and J .O. Cavenar (Eds), Pglchiatpy (pp. 53-75). Philadelphia: Lippincott.

Lamberty,G. & Garcia Coll, C. T. (Eds). (1994). Puerto Rican women and
children: issues in healtl_1, growth. and development. New York: Plenum Press.

Leary, T. (1957). Intarmrsonal diagposis of mnalig. New York: Ronald.

Locke, D. C. (1992). Increasrr_rg' multicultural understandmg’ : A commehensive
model. Newburry, CA: Sage

McCrae, RR, & Costa P.T., Jr. (1989). The structure of interpersonal traits:
Wiggins’ circumplex and the ﬁve-Ector model. Journal of Personaligy and Social
ngchology, 56, 586-595.

McLemore, C.W. & Benjamin, LS (1979). Whatever happened to interpersonal
diagnosis? A psychosocial alternative to DSM-III. American Pachologist, 3_4, 17-34.

Orford, J. (1986). The rules of interpersonal complementarity: Does hostility
beget hostility and dominance, submission? Paychological Review, 23, 365-377.

Pincus, A.L., & Wiggins, J.S. (1990). Interpersonal problems and conceptions of
personality disorders. Jomal of Persona_li_ty Disorders. 5, 342-3 52.

Pincus, AL. (1993). The interpersonal circumplex and the interpersonal theory:
Perspectives on personality and its pathology. In S. Strack & M. Lorr (Eds),
Differentiating normal and abnormal mrsogm . New York: Springer.

Romero, M. (1986). Twice protected? assessing the impact of aﬁrmative action
on Mexican American women. In W.A. Van Horne & T.V. Tonnesen (Eds), Ethnicity'
and women. University of Wisconsin, American Ethnic Studies Coordination Committee.

Sullivan, HS. (1953). The intpgpe_r;_sonal theog of paychiapy. New York:
Norton.

42

Wiggins, J .S. (1979). A psychological taxonomy of trait-descriptive terms: The
interpersonal domain. Journal of Personality and Social Paychology, 3_7, 395-412.

Wiggins J.S. (1982). Circumplex models of interpersonal behavior in clinical
psychology. In P.C. Kendall and J .N. Butcher (Eds), HandbooylLof resegch methods 11
clinical pgchology. (pp. 183-221). New York: Wiley.

Wiggins, J .S. & Broughton, R. (1985). The Interpersonal Circle: A structural
model for the integration of personality research In R. Hogan & W.H. Jones (Eds),

Pmtives in mﬂg’ : A research annual (V 01.1, pp.l-47). Greenwich,CN: JAI
Press Inc.

Wiggins, J.S., Trapnell, P.D., & Phillips, N. (1988). Psychometric and geometric
characteristics of the revised Interpersonal Adjective Scales (IAS-R). Multivariate
Behavioral Research, Q, 17-30.

“Wiggins, J.S., & Pincus, AL. (1989). Conceptions of personality disorders and

dimensions of personality. chhologjcal Assessment: A Journal pf Consultmg' and
Clinical Pachology, 1, 305-316.

”‘IIIIIIIIIIIIIIIIIIII