..
3.: mar!
2.. v.

    

 

 

Ech

 

Er

 

s

 

E

 
 

 

 
 

 

 
 
 
 
 

 

 
  
   

 

 

b... :1

51,...

       

 

4%..
J J. awmwmwfl.
Em? x. m...

  

       
 

I A... .n
.45....»

    

 

$4.. €31.41:
: Lawfixhi
75:93! v.33}...
1:53:- 2'1; if.”

:1: h. EL»... 3....»

   

 

~

      

.511.
it. l.....fl.uu-.«Véi..hnwl
1:412... .
.1551: 2.... 167%... .-
51:59.»..ruilxllnfn
«lift... :83

a 1.

 

     

   

 

  

 

 
 

 

)»y(w...y(1.‘l) 1.: \ A... .

  

 

’31.?

 

 

 

 

 

  

. '- lI \ .N/
)....I.v.1.¢’.vl../!

...l. 3st,?

     

lllllllll lllll llllllllllllllll -

3 1293 10181 2497 £11594 i1! 7
Michigan State g.

University r5

l l

BjThis is to certify that the

4 i ‘ thesis entitled

The Effects of Intuition,
Feeling, and Similarity
on Predictive Empathy

presented by
Martha L. Aldenbrand

has been accepted toWards fulfillment
of the requirements for

Ph.D. mag“... in Psychology

9" lfj/gk 4/ ‘5:sz7—\

Major professor

DateLL". '~ y’ 4 E

0-7 639

 

 

 

 

 

W278".
,139 [1149 -

 

0X A350

 

 

 

.. ._.____.—_-—l

ABSTRACT
THE EFFECTS OF
INTUITION, FEELING, AND SIMILARITY
0N PREDICTIVE EMPATHY
By
Martha Lynn Aldenbrand

This investigation concerned relationships between empathic
ability and preferences in the use of Intuition and Feeling as modes
of understanding others. A secondary feature was the exploration
of how similarity of Intuition and Feeling preferences between subject
and target persons related to empathy. Empathy was defined in the
"person perception” tradition as ”the ability to accurately predict
the feelings, thoughts, and behaviors another person would ascribe to
himself.” The Myers-Briggs Type Indicator was used to assess Intui-
tion versus Sensing and Feeling versus Thinking.

Four 30 minute videotapes were made of the interaction of each
of two groups of female target persons, each containing a representa—
tive of Intuitive Feeler (IF), Intuitive Thinker (IT), Sensing Feeler
(SF), and Sensing Thinker (ST) types on the Myers-Briggs. All fell
within the middle range on Extraversion-Introversion. Following each
videotaped session, targets individually monitored a replay and
stopped the tape to identify feelings and thoughts they had had during
the original meetings which had been structured to elicit typical

examples of personality functioning. From these data four types of

 

Martha Lynn Aldenbrand

multiple-choice questions were constructed, those assessing prediction
of (l) target feelings, (2) target thoughts, (3) target behaviors, and
(4) inter-target sociometric relationships.

Female observers were also selected for Myers-Briggs scores to
represent 9 categories: IF, IT, SF, ST, plus observers scoring in the
middle range on Intuition-Sensing and Feeling-Thinking.

Observers were informed which target person was the focus of each
question, shown a short videotaped sequence, and when the tape was
stopped, requested to answer the relevant multiple-choice question.

A second measure of empathic accuracy and similarity was given after
the final taped session of each group, when observers completed an
abbreviated form of the Myers-Briggs test the way she thought each
individual target would respond.

Hypotheses that empathic accuracy would be facilitated by Intui-
tion and Feeling orientations as well as similarity between subject
and target were tested. Findings revealed that empathic accuracy was
only slightly above a chance level (approximately 38% versus 25%) across
all observers. There were no significant differences related to
Intuition, Feeling, or similarity.

The present investigation's design fell into the common error
in this research area that the products of empathy, i.e., correct
answers, were obtained without attention to the underlying process of
empathy. Empathy being a discontinuous process of attending and inte—
grating, adequate research must allow both sub-processes to emerge.

This study was directed toward the products of the attentional aspect,

 

Martha Lynn Aldenbrand

where the observer actively selects the cues she deems most relevant to
understanding. It ignored, however, periods of integration where the
observer was minimally attending to new cues, and, instead, internally
deciphering and organizing already received cues. Thus, when the
experimenter stopped the videotape during moments of integration,
natural minimal attending may have reduced the chances of responding
accurately. It was suggested that empathy research allow both natural
timing and the personal interpretation of the stimulus event if the
integrative aspect of empathy is to be expressed.

Finding differential accuracy toward targets who supposedly repre-
sented the same Myers-Briggs personality category obfuscated a meaning-
ful evaluation of similarity. The expression of desired stimulus
qualities was unduly influenced by other confounding personality traits,
such as dogmatism, extraversion-introversion, etc. An approach which

assures the prominence of appropriate target behaviors was suggested.

l

THE EFFECTS OF
INTUITION, FEELING, AND SIMILARITY
ON PREDICTIVE EMPATHY

By
Martha Lynn Aldenbrand

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

l974

 

Primary:

Secondary:

Tertiary:

DEDICATION

To my parents, Marjorie and Godfrey,
who taught me the importance of'empathy
and attending to feelings.

To my husband, John, whose patience,
warmth, and caring made writing easier.

2b Melanie, who sang "Look What They’ve
Done to My song, Ma" during those times
I failed to see the logic of'my committee
members suggestions.

ii

ACKNOWLEDGMENTS

I wish to express my gratitude to my committee members: Dozier
Thornton, Elaine Donelson, Joel Aronoff and Dave Wessel. Dozier's
empathic understanding has always intrigued me and has made working
with him enjoyable. Listening to my ideas as they hatched and knowing
exactly when to add his own made the research fun. Later, his creative
suggestions concerning the organization and readability of the disser-
tation were very helpful. And his thoughts about future research modi-
fications made me think of this research as a continuing project.

Elaine contributed a substantial knowledge of the research
literature which saved me considerable time and broadened my perspec—
tives on empathy as a concept. Her valuable and detailed criticism of
the design and manuscript helped me over many rough spots. She was
also a much used source of warmth and comfort during trying times.

I needed and appreciate Joel's dedication to simplifying my
original design. His gentle, but persistent,_way of pointing out
things I was overlooking helped anticipate and correct many potential
problems. I am grateful for the considerable time and effort he spent
proofreading as his comments were useful and thought-provoking.

Dave Wessel gave useful statistical advice, especially concerning
pre-design issues.

 

iii

TABLE OF CONTENTS

 

Page

LIST OF TABLES . . .. ..................... vi
LIST OF FIGURES ........................ viii
I. INTRODUCTION ...................... l
Choosing a Definition of Empathy .......... 3

II. THE PROBLEM ....................... 8
III. REVIEW OF THE LITERATURE ................ 12
Empathy as a General or Specific Ability ...... l3
Recognition of Emotions .............. l6
Veridicality .................... l7

Implicit Personality Theories ........... l9

IV. DEFINITIONS, HYPOTHESES, AND RATIONALE FOR HYPOTHESES. . 22

V. MEASURES ........................ 27
Personality Assessment Measures .......... 27

The Myers-Briggs Type Indicator ........ 27

Empathy Measures .................. 33

l. The Videotape Empathy Series ........ 33

2. The Personality Empathy Test ........ 50

VI. PROCEDURE ........................ 52
Selection of Subjects and Group Members ...... 52

Procedure for Subjects ............... 54
Statistical Procedures ............... 57

VII. RESULTS ......................... 59
Main Effects .................... 6O
Differential Effects ................ 70

iv

TABLE OF CONTENTS—«continued Page

Measurement Effects ................. 74
Post-experiment Data ................ 85
Summary of the Results ............... 86
Main Effects .................. 86
Differential Effects .............. 87
Measurement Effects .............. 87
VIII. DISCUSSION ........................ 90
The Stimulus: The Videotaped Group Discussion . . . 90
The Response: The Empathy Measures ......... 94
Assumptions Concerning the Videotape Empathy
Series ..................... 95

Measurement Error gua_Videotape Empathy Series. 97
Measurement Error Related to the Personality

Empathy Test .................. 100

Selective Videotape Empathy Series versus Person—
ality Empathy Test Error .............. 102
Implications for Further Research .......... 103
Assumptions About Empathy as a Process ..... 103
The Complexity of the Design .......... 105

Specific Design Problems and Alternative Soluv
tions ..................... 106
Summary ....................... 110
APPENDICES

A. THE ASSESSMENT MEASURES ................. 112

B. DATA SUPPORTING THE USE OF THE MYERS-BRIGGS SCALES AS
CONTINUA ........................ 126
C. THE EMPATHY MEASURES .................. 129
D. THE AWARENESS MATERIALS ................. 142

E. FEELING SUMMARY USED BY TARGET TO IDENTIFY FEELINGS. . . 146
F. POST-EXPERIMENT DATA .................. 148
BIBLIOGRAPHY .......................... 150

 

LIST OF TABLES

TABLE

10.

11.

12.

13.

. Outline for constructing the Thinking and Feeling Empathy

Tests ...........................

. Summary of Analysis of Variance for Videotape Empathy

Series ..........................

. Summary of Analysis of Variance for Personality Empathy

Test ...........................

. Analyses of Variance from Table 2 with Special Relevance

to Hypothesis I ......................

. Analyses of Variance from Table 3 with Special Relevance

to Hypothesis I ......................

. Analyses of Variance from Table 2 with Special Relevance

to Hypothesis II .....................

. Analyses of Variance from Table 3 with Special Relevance

to Hypothesis II .....................

. Means from Judgers-Measures Interaction (J x M) with Rela-

tion to Empathy Measured via a Videotape Source ......

. Means from Judgers-Measures-Perceiver Targets Interaction

(J x M x X) with Relation to Empathy Measured via a Video-
tape Source ........................

Means from Judgers-Measures-Perceiver Targets-Judger
Targets Interaction (J x M x X x Y) with Relation to
Empathy Measured via Personality Test Source .......

Analyses of Variance from Table 2 with Special Relevance
to Hypothesis III .....................

Analyses of Variance from Table 3 with Special Relevance
to Hypothesis III .....................

Analyses of Variance from Table 2 with Special Relevance
to Hypothesis IV .....................

vi

Page

34

61

62

64

64

66

66

67

67

69

7O

71

71

LIST OF TABLES-«continued

TABLE

14.

15.

16.
17.

18.

19.

20.

21.

22.

Analyses of Variance from Table 3 with Special Relevance
to Hypothesis IV .....................

Analyses of Variance from Table 2 with Special Relevance
to Hypothesis V ......................

Summary of Analysis for Order of Presentation Effects. . .

Means from Films (F) Analysis of Variance Reported in
Table 16 .........................

Means from the Presentation x Films x Measures (M : F x P)
Interaction Reported in Table 16 .............

Summary of Analysis of Variance for Congruence of Individu

ual Target Representatives on Videotape Empathy ......
Means from the Measures x Perceiver Targets x Judger
Targets x Alter Targets Interaction (M x X x Y x A)
reported in Table 18 ...................

Summary of Analysis of Variance for Personality versus

Videotape Empathy Measures ........... . . . . .

Means from the Form x Perceiver Targets X Judger Targets

Interaction Reported from Table 21 ............
vii

Page

72

73
75

77

77

80

81

83

84

LIST OF FIGURES
FIGURE Page
1. Time schedule for group meetings and individual recall

sessions ......................... 37

2. Eye direction of group members using split-screen
cameras ......................... 47

3. Videotape screen configuration using one member as

primary focus ...................... 47
4. An example of a Social-interaction rating scale ..... 49
5. Selection procedure according to distribution of scores

on experimental variables ................ 53
6. Classification of subjects by scores on independent vari-

ables .......................... 53
7. Arrangement of subjects and subject-names on screen . . . 55
8. Order of presentation of empathy tests per group ..... 57

9. Distribution of means from Measures x Perceiver Targets x
Judger Targets x Alter Targets Interaction (M x X x Y x A)
reported in Table 19 ................... 82

viii

I. INTRODUCTION

When one person speaks to another, he sends a verbal message,
but he also watches to see whether, or how much, the other person
understood what he meant. Likewise, the other listens to the verbal
content of the message, but also watches for situational and non-verbal
cues, like voice quality, posturing and the presence of others, to
interpret the meaning of the content. It seems natural that with the
number and complexity of cues available to the listener, some listeners
understand cues better than others. This ability to understand the
thoughts and feelings of another person is called empathy.

Sullivan says the rudiments of empathy arise during infancy.
The infant apprehends anxiety in the mother and in turn becomes anxious.
However, early infant empathy is a phenomenon which the infant cannot
control. He feels anxious, but his anxiety does not lead to construc-
tive actions. He cannot soothe the mother or alter the environment.
The usual result is that the mother soon apprehends the infant's
anxiety and thereby increases her own. Since the infant is totally
dependent on the mother, this leads to an intolerable situation. The
only defense that works, according to Sullivan, is that the child falls
asleep. As the infant grows, however, apprehending the emotional
state of the mother does lead to increased skill, the child learns

specific mechanisms of approach and avoidance.

One would wonder, then, why people are not perfectly empathic,
since empathy begins so early and people seem to grow more intelligent
with age. There is one important altering factor, however, which con-
founds the steady growth of empathy, that is language. As the child
grows he must encode the meaning of feelings into words. But the
source of the words, primarily the mothering-figure, does not always
want the child to know her feelings. She may fail to label some feel-
ings, leaving them less manageable, especially as the world is becoming
more verbal to the child. She may mislabel them, either by intention
or through her own distorted learning, calling anger "sadness" or fear
”confusion". The child must then cope with having some "admissible"
feelings, some distorted feelings, and some non-verbal feelings.

A study by Feigenbaum (1967) showed children to lag two to four years
in their understanding of negative versus positive feelings in others.

The need to encode basic experience into verbal labels, and the
difficulty of attaining a complete translation, may be related to
another factor often connected with empathy, namely intuition. Since
intuition is commonly defined as the non-verbal apprehension of the
primary meaning of a situation, an intuitive ability may be advantageous
to understanding those blurred situations in which the child was in-
tentionally or unintentionally mistaught the characteristics of certain
feelings. An intuitive person seems to have left many basic experiences
uncategorized, or freely flexible, so that verbal labels have not com-
mitted him to a single rigid definition. This intuitive "regression in
the service of the ego," matching present data to non-verbal data

seemingly extant in all its original complexity, has long been a factor

 

attributed to empathic skill (Katz, 1963; Rommetveit, 1960; Wescott,
1968). However, because of the elusive nature of non-verbal data and
because the logical processes behind intuitively derived labels (if
such logical processes exist) are seldom conscious in all their com-
plexity, scientists have hesitated to measure intuition and use it as
a variable in research. Thus, it has been long asserted that empathy
and intuition are highly correlated, but it has rarely been studied.
The process of matching the experience expressed by another
with one's own experience leads to the consideration of a final factor
related to empathy, that of similarity. Similarity between perceiver
and perceived facilitates empathy because translation across experi-
ences is easier. And in on-going situations, where a continual flow
of information must be assimilated, ease of translation may be directly
related to the amount of empathy produced. Similarity in terms of
attitudes, beliefs, socioeconomic class, and perhaps more basically,
cognitive-perceptual style, then, should tend to increase the likeli-
hood of empathy. The present research employs the three concepts
discussed above (attending to the other's feelings, intuition, and

similarity) to assess the qualities of an empathic person.

Choosing a Definition of Empathy

Relating intuition, feeling-orientation, and similarity to empathy
is not as easy as one would initially assume, however. One immediately
must face the problem of the definition of empathy. At least five well
recognized, but different definitions of empathy exist in the field.

Directly related to the discussion of similarity is Smith's (1966)

definition that empathy is the correct inference that similarity exists
between the self and other. A second broader definition includes dis-
similarity in the accurate prediction of the thoughts, feelings or
behaviors of another person. A third definition emphasizes increased
responsiveness to the feelings of another. A fourth adds the ability
to communicate to another that one understands his feelings. And,
finally, a fifth defines empathy vaguely as the capacity to adopt a
broad moral perspective.

The first approach to empathy considers similarity so vital to
empathy that it defines empathy in terms of similarity. Thus, Smith
(1966) defines empathy as “the tendency of a perceiver to assume that
another person's feelings, thoughts, and behavior are similar to his
own (p. 93)." According to this view, empathy is derived from identifi-
cation with another person. Through the process of generalization one
assumes the other is like himself in many ways. To the extent that one
is correct in assuming similarity, he is empathic. With respect to
dissimilarity, empathy is an irrelevant term.

The second definition, that empathy is the ability to accurately
predict the feelings, thoughts, or actions of another person is the
definition focused upon in this paper and will be discussed in greater
detail later. Since empathy is seen as the process of viewing the world
as another person views it, similarity as well as dissimilarity are
considered important discriminations in assessing the other's personal
space, hence the departure from Smith.

Empathy viewed as emotional responsiveness paralleling the feel—

ings of the other was perhaps the earliest systematic treatment of

empathy (Lipps, 1907). It was strongly related to Darwin's position

in The Expression of Emotions in Man and Animals (1872) that certain
universal expressions, such as sneering, result from common emotions,
and that common emotions, e.g., fear, are elicited by the stimulus
expressions. Following this tradition, role-playing may be viewed

as the empathic awakening of emotions connected with adopting differ-
ing role positions which results in deeper understanding of the experi-
ences of the other. In fact, Kerr and Speroff (1951) refer to empathy
as role—playing. Speroff (1953) further suggests that empathy is
facilitated by role-reversal, in which the subject restates the other's
point of view until the other verifies it as correct.

Another branch of this tradition is the work of Stotland (1969),
Stotland, Sherman, and Shaver (1971), and Mehrabian (1972). Stotland
measures empathy by recording palmar sweating, vasoconstriction and
basal skin conductance of subjects viewing an emotional scene.
Mehrabian has devised an empathy test based on emotional lability.
Those who report increased emotional responding in diverse social situ-
ations are considered empathic.

In assessing the Lipps tradition, however, the primary considera-
tion will be whether concomitant emotional experiences are necessary
for empathy to occur, not simply whether they are advantageous to
empathic understanding. It is at this point where this author departs
from Lipps' tradition. While emotional responses to the situation of
another are assumed to benefit assimilating the other's experience,
there appear to be many instances when one understands but does not

share the experience of the other. In fact, Rogers (1951) differentiates

 

 

between shared feeling (sympathy) and understood but non-shared feel-
ings (empathy). In therapy this differentiation may be crucial-—one

might understand that a client's mother had nagged him once too often
and as a result he killed her, but if the therapist felt as angry as

the client, we might have two murderers on our hands.

A fourth view, usually preferred by counseling psychologists,
defines empathy as effectively communicating that one understands the
feelings of the other. Thus, inarticulate individuals are rarely
empathic. Rogers (1951) was one of the first to add communication as
a second step in empathy following understanding the other's frame of
reference, although Murray (1938) did describe an interactive process,
called ”recipathy”, through which a therapist might understand his
client. Recipathy involved the therapist watching the feelings invoked
in himself as the client talked. These feelings were then treated as
feelings the client needed to invoke in others, not merely as unrelated
therapist arousal.

Truax and Carkhuff (1967), closely aligned with Rogers, have done
extensive investigation into empathy viewed from the communication per-
spective. Part of this work has included a scale for the measurement
of counseling empathy (Truax, 1961). Empathy according to this approach
includes responding with words, gestures, and tone of voice that matches
the emotional intensity expressed by the other, responding concretely
rather than abstractly to the messages of the other, and responding
frequently enough that the other is assured that understanding continues
to exist. While this paper agrees that the process of communicating

understood messages is important, it takes as its focus the more

 

fundamental process of understanding pg: sg without confounding under-
standing with its communication.

Finally, empathy has been defined as making moral choices (Hogan,
1969). However, this seems a distant relative from the definitions
mentioned above. The scale derived by Hogan seems to be measuring a
conglomerate of values, e.g., standing up for ”what is right”, feeling
sorry for others, enjoying the company of diverse types of people.
Empathy defined this way may be considerably correlationally based,
i.e., people who are apt to be courageous, sympathetic, and openminded
are also apt to be empathic. Since the present research attempts to
measure empathy directly, this definition must be considered irrelevant.

This research will use the second definition of empathy, that
empathy is the ability to accurately predict the feelings, thoughts, or
actions another person describes as his own. It is closely related but
more comprehensive than the first definition in which empathy is defined
as the correct prediction of similarity between self and other.
Dissimilarity, as well as similarity, may be correctly predicted accord-
ing to the definition adopted. This definition is also closely related
but more comprehensive than the third definition in which empathy is
defined as the concomitant arousal of feelings in the subject as in
the other. In the present definition, the subject may respond to his
own thoughts as well as feelings in attempting to ascertain the state of
the other. The present definition arises from its own tradition and is
often found in the literature called "person perception” or “interpersonal
perception”. These terms will be used interchangeably with "empathy"

when citing literature denoted as such.

 

II. THE PROBLEM

This research attempts to answer three major questions: (1) Does
an individual's perceptual style with respect to whether he handles
incoming information by using Intuition as opposed to Sensation, effect
his empathic ability? (2) Does an individual's style for making deci-
sions with respect to whether he decides primarily by Feeling as

opposed to Thinking, effect his empathic ability? And (3) Does simi-

 

larity between the target and the subject on the two orientations effect
empathic ability?

The hypothesized relationship of intuitive and feeling orienta-
tions with empathic skill has lengthy common sense and clinical support,
but, as yet, little experimental justification. Lack of research justi-
fication is most likely the result of what seems to be the intangible
nature of the process of intuition and the extensive amount of non-
verbal interpretation necessary in identifying feelings. Only recently
have some authors tried to measure the process of intuition (Rommetveit,
1960). These attempts, however, rely on making judgments when the
premises behind the judgments remain out of the awareness of the judger.
To assume that one cannot be aware of the intuitive process, or that all
non-aware judgments are intuitive,seems absurd. Some studies have given
minimal cues to subjects and ”forced“ intuitive judgments (Hathaway,

1955; Valentine, 1929).

 

A few studies have defined intuition vaguely and explored it as
a trait manifested in some people more than others. Estes (1937)
found that the most accurate judges of feelings during motion picture
segments were those who worked at intuitive occupations or had intui-
tive avocations, such as painting and drama. Allport (1961) reports

that people who adopt an analytical, reflective approach to perception

are less accurate than those who adopt a more global, intuitive approach.

Similarly, very little research has studied whether those who are
oriented toward attending to the feelings of others and making judgments
according to feelings as opposed to logic are more empathic than those
who do not. Most research concerning feelings and empathy has focused
on whether and under what conditions it was possible to recognize the
feelings of others. They have attended less frequently to perceiver
characteristics of good predicters of feelings. Although, Halpern
(1954) did find a significant correlation between predictive accuracy
on a personality inventory and femininity of attitude--a scale loaded
heavily on the experience and expression of intense emotion.

In order to study these two orientations, this research has
employed a theoretical framework which definitively describes these
processes, that is Jung's theory. According to Jung, each human being
uses one of two alternative perceptual modes (called “functions”) from
which he garners information about the world. He also uses one of two
judgmental functions with which he makes decisions about how to use the
information perceived.

Jung calls the alternative perceiving functions “intuition“ and

”sensing”. Intuition is an unconscious organization of stimuli

 

resulting in the individual only experiencing ”a complete whole,
without ... being able to explain or discover in what way this content
has been arrived at. (It) ... is a kind of instinctive apprehension,
irrespective of the nature of its contents” (Jung, 1923, p. 263). On
the other hand, Sensation relies on conscious experience gained by the
sense organs, denying unconscious Intuition. Sensing is directed toward
the objective, external environment, responding most decisively to the
strongest stimuli impinging on the individual. According to Jung,
Intuition and Sensation are opposite functions of perception, one uncon-
scious the other conscious.

The judging functions are ”thinking" and ”feeling”. They relate
to how the individual makes decisions or comes to conclusions about the
stimulus data he has acquired. The function of Thinking is based on
certain ”laws” or logical structures, which may be either conscious or
unconscious, and consequently may be either rational or irrational.

An example Jung uses of the latter are thoughts arising during dreams,
which seem reasonable in the dream, but unreasonable when awake. The
function of Feeling relates to forming value judgments with reference to
acceptance or rejection. Feeling may refer to specific objects, such

as a beautiful rose, or whole situations, such as an ugly mood. Feeling
and Thinking may also have contradictory components, an individual may
think_a dress is ugly, but notice that her friend is wearing it and feel
that the dress on the friend is beautiful.

Jung views most people as operating with one dominant perceiving
and one dominant judging function. Thus, most people are either

Intuitive-Thinkers (NT), Intuitive—Feelers (NF), Sensing-Thinkers (ST),

 

 

 

or Sensing-Feelers (SF). Moreover, they will display one of two primary
attitudes toward the world, Introversion or Extraversion. These atti-
tudes represent the degree to which the individual openly uses his
functions. An Introvert keeps to himself, is wary about the environ-
ment, and relies on his own resources for fulfillment. An Extravert is
out—going, trustful, and dependent on the external world for fulfillment.
Finally, Jung says that although most people can be assigned to one or
the other of the perceiving or judging functions, those who become most
fully functioning move toward greater differentiation and balance, thus
tending to blend Introversion-Extraversion and the perceiving and judg-
ing functions in their everyday lives. From Jung's perspective, then,
the most empathic person should be one who does not exclusively use
either judging or perceiving dimensions.

To determine whether empathy is higher among Intuiters and Feelers,
or middle perceivers and judgers as Jung would hypothesize, a simple
design was employed which measured the empathic accuracy of subjects
designated Intuiter, Senser, Middle Intuiter—Senser, and Feeler, Thinker,
Middle Feeler—Thinker toward targets presented via a videotaped discus-
sion. To determine whether similarity influences empathic accuracy tar-
gets were chosen who represented each combination of Jung's perceiving
and judging orientations. Thus, accuracy scores toward each NF, NT, SF,
ST target were analyzed with respect to the subject's own perceiving or
judging orientation. In this research, empathic accuracy was measured
two ways: (1) by correctly predicting what targets said they were think—
tngrorufeeling.during the videotaped discussions, and (2) by correctly

predicting how targets would answer questions on a personality test.

 

 

III. REVIEW OF THE LITERATURE

Comprehensive literature reviews were made early by Bruner and
Tagiuri (1954) and Taft (1955) and again more recently by Tagiuri
(1969). These have been augmented by more circumscribed reviews by
Shrauger and Altrocchi (1964) and Kanekar (1972). From these and other
sources it can be noted that predictive empathy research has taken two
primary directions, one investigating whether empathy was a general or
specific ability, the other assessing the extent to which an individual
could understand the emotions of a target person. This paper will dis-
cuss these divergent paths as well as two additional topics affecting
the outcome of empathy research: veridicality and the subjective forma-
tion of implicit personality theories.

It should be pointed out at the onset that a unified tradition
of research has not been established in this area. In fact, experi-
mental stimuli and responses have been so varied that few generaliza-
tions can be inferred when comparing data. Stimuli have included photo-
graphs, handwritihg samples, graphic displays of human faces, written
personality descriptions and autobiographical accounts, projective test
protocols, tape recordings, motion pictures, videotapes, role plays
(usually of emotional situations), and other live interactions.
Responses have included recognition of the emotion or personality dis-

played by free responding or multiple choice answers, by sociometric

12

 

13

rank ordering, by predicting future behaviors or present occupations,
and by predicting responses on personality or attitude tests. Subjects
have been directed to interpret how specific targets would respond and
how certain classes of persons, e.g., college students, would respond.
In general, much of the literature agrees that stimuli and responses
that accurately allow the complexity of the target and the subject to
emerge are more likely to show significant differences among the vari-
ables studied than are the most simplified measures (Gage and Cronbach,
1955, Jackson and Messick, 1963; Cline, 1964; Shrauger and Altrocchi,
1964, Tagiuri, 1969).

Empathy as a General or Specific Ability

One of the first problems predictive accuracy research tackled was
to determine whether certain individuals had a heightened capacity to be
empathic in all situations, i.e., in general, or whether empathy was
more a function of the situation or the transparency of the specific
target individual involved. Two of the earliest studies reported in
this area established the format for later research. Steinmetz (1945)
had subjects predict a target individual's answers on a personality test
as a measure of "psychological perception'l or awareness of others.
Dymond (1949), first using the term "empathy“ for this method, had group
members predict how others in the group would rank themselves on certain
personality traits, e.g., friendly-unfriendly, sense of humor. Both
Steinmetz and Dymond report differences supporting empathy as a general

ability.

14

The literature is equivocal on this issue, however. Allport

(1937) suggests that judging ability is neither entirely general nor

'Tfimtirely specific. Some individuals are usually more accurate in
general, but specific situations play a contributing role. Cline (1955)
and later Cline and Richards (1960 and 1961) demonstrated that some
generality in judging ability does exist apart from similarity of sub-
ject and target. However, Gage and Cronbach (1955), Purcell, Modrick
and Yamahiro (1960) and Stone and Leavitt (1954) suggest that little
generality exists. Crow and Hammond (1957) and Sechrest and Jackson
(1961) attribute whatever generality that exists to response set and
other response biases. Reviews of the literature by Bruner and Tagiuri
(1954) and Taft (1955) side with the generality view, believing that
"good" and "poor" judges do exist.

The methodology used with the generality-specificity issue has
been criticized from two viewpoints which may partially account for
the discrepant results cited above. It has been noted that success in
predicting others ratings on personality tests or questionnaires may
result from knowledge of a stereotype of the individual, e.g., male
factory worker, as well as knowledge of the particular target involved.
Most research has not differentiated between these two sources.

Another criticism concerns the similarity versus dissimilarity of the
targets and judges. Failure to systematically assess the effects of

this dimension may unduly bias success toward target and judge pairs

which are similar.

Cronbach (1955) first pointed to the influence of the sterotyping

factor. He demonstrated that accurate knowledge of the stereotype of

 

the target was usually the best predictor of his performance. There-
fore, one did not need a stimulus at all, other than a classification
label, to "empathically" predict target responses. Thus, Cronbach
devised elaborate scoring procedures for factoring out stereotype
accuracy, differential accuracy (how the target differs from the stereo-
type) and other response set tendencies. Bronfenbrenner, Harding and
Gallwey (1958) found that two similar abilities: accuracy toward a
generalized other versus sensitivity toward a specific person, appear

to be independent. Cline and Richards (1960) found similar results.

Another issue was confronted by Bender and Hastorf (1950) and
Gage and Cronbach (1955). They distinguished accuracy on the basis
of real versus assumed similarity and dissimilarity. Real similarity
between subject and target has long been recognized as facilitating
empathy. Allport (1937) reports this effect for sex, age, background,
complexity, and personality characteristics. Similar results have
been found by Notcutt and Silva (1951); Suchman (1956); Gage, Leavitt
and Stone (1956); and Bronfenbrenner, Harding and Gallwey (1958).

This beneficial effect may be related to the relative ease of assimila—
tion and reliance on oneself as referent (Blanchard, 1966 and 1967)
versus differentiation from oneself as referent (Lundy, 1956).

Assumed similarity and dissimilarity were terms used by Gage and
Cronbach (1955) to assess instances where the subject assumes the tar-
get is similar or dissimilar to himself when he is not. Bender and
Hastorf (1950) found that some subjects consistently tend to be
”empathizers” (assumed similarity) and some ”projectors“ (assumed dis-

similarity). Bieri (1955) notes that cognitively simple subjects tend

 

to assume similarity between themselves and targets, whereas cognitively

complex subjects tend to assume dissimilarity and differentiation.

Recognition of Emotions

The literature on the recognition of emotions has shown that
ease of recognition is related to at least three factors: the method
of stimulus presentation, the degree of discrimination demanded of the
subject, and the extent to which labels are self-generated. As described
earlier, person perception experiments have used varied presentation

modes, including character drawings, photographs, and motion pictures

 

as stimuli. It is obvious that the amount of information presented,

as well as its naturalness, may greatly differ among these methods. It
is not surprising, then, to find that accurate labeling is more often
found in complex presentations, such as motion pictures, than in simple,
acontextual examples, such as photographs of real emotions elicited in

a laboratory (Tagiuri, 1969).

The difficulty of the discrimination demanded of the subject is

 

also important. Since the work of Woodworth (1938) and Schlosberg
(1952) who devised, respectively, scalar and orthogonally dimensional
models for delineating the similarity of emotions, we have had a method
to test the accuracy of discriminating emotional stimuli. Thus, it was
shown to be harder to discriminate between love and happiness, than
between love and anger (Woodworth, 1938). Research in this area has
generally supported the degree of difficulty reported to exist between
two given emotions on both Woodworth's and Schlosberg's scales (Davitz

and Davitz, 1959b, Abelson and Sermat, 1962; and Engen, Levy and

 

Schlosberg, 1938). Thompson and Meltzer (1964) did find, however, that
some emotions: happiness, fear, love and determination, were easier to
recognize overall than was disgust, contempt, and suffering. Thus,
research which fails to account for the differential discriminability
of emotional stimuli may confound the real results concerning accurate
perception by making some tasks or items more difficult than others.

A final factor affecting the recognition of emotions is the label
itself. Subjects have been shown to be more often correct in labeling
an emotion when they use their own labels than when they are supplied
multiple-choice labels (Munn, 1940). However, free responding also
produces methodological problems. Which label, the experimenter's,
the target's, or the subject's, is correct? Fernberger (1928) found
that ”false“ interpretations, i.e., those not generated by the target,
were nevertheless acceptable to the targets unless they grossly dif-
fered from the emotion originally named. On the one hand the new label
may represent a simple substitution of a synonym, but it may also allow
another more diffuse, non-contradictory emotion to pass for correct.
Thus, the degree of difficulty of the empathy response desired should be

reflected in the latitude allowed in labeling.

Veridicality

One central problem confronting empathy research is the determina-
tion of what the target individual was actually experiencing as the
stimulus was presented. Experimenters who use role players as stimuli
have predetermined the emotion or behavior intended to be presented.

The role player acts angry, fearful or surprised and the subject must

 

guess the feeling imitated. One problem with this method, however, is
that the emotion expressed is often exaggerated and/or stereotyped.
In real life people may express their emotions subtly, only sometimes
in a stereotyped fashion.

But the problem is even more difficult for researchers who are
trying to assess empathy by observing natural behaviors. Sometimes
an individual is vividly aware of his emotions and behaviors, but some-
times he may have only unconscious or non-verbalized awareness of his
feelings. An individual may not know he is angry for three or four
minutes after he has been displaying non-verbal cues signaling his
anger. The knotty issue, then, is: who really knows what the target
is feeling? The target himself? A seasoned observer? Who is correct?

Most predictive accuracy research has avoided the issue by de-
fining empathy as the prediction of answers on a personality test or
attitude questionnaire. While this does have the appeal of objectivity,
i.e., the answers can easily be verified, the flavor of the decision
process or the actual experience of the target taking the test is un—
known. Predictive accuracy via personality tests, then, is a valid
but highly circumscribed method for assessing empathy. When we try to
assess those circumstances most common to our usual meaning of empathy:
the interpretation of live experiences, we are again faced with the
veridicality dilemma.

Research in counseling psychology, where empathy consists of
recognizing the feelings of a client, has tended to favor the use of
”experts“, i.e., people assumed to be empathic by others, Such as

counselors and advanced graduate students in counseling (Buchheimer,

 

 

l-

Goodman and Sircus, 1965; Campbell, 1967; Chapman, 1966; Rank, 1966).
Subjects, in these cases, are actually trying to be empathic with the
labels of empathic persons. When the expert and the target agree on
the label, no grave problems arise. But despite their skills, being
removed from the actual experience, often tends to make experts rely
on more stereotyped or overly simplified answers than would be true
with direct experience as reported by the target (Allport, 1961).

At least one group of counseling psychologists (Kagan, Schauble,
Resnikoff, Danish, and Krathwohl, 1969) have attempted to breech the

gap in interpretation by employing a technique designed to help the

 

target discover and elucidate the feelings he was having during the

taped session. This ”process recall” consists of replaying the video—
taped session before the target and an interrogator. The interrogator
asks open-ended questions designed to elicit clear and vague feelings,

thoughts and body cues of the target. While this method does not

 

facilitate recall of truly unconscious material, it does help the tar-
get focus on minimally aware cues. Kagan and Krathwohl (1967) and
Kagan §t_pl, (1969) have shown substantial results using this method
with counselor training. A more detailed description of the process

recall method can be found in the Measures section.

Implicit Personality Theories

Implicit personality theories are inferential linkages which
people assume to exist between personality traits. Among the earliest
research in this area was the work of Asch (1946), who found the

importance of a central trait, "warm-cold“, to significantly effect

 

20

the formation of impressions made by subjects, and Kelly (1955) whose
Role Construct Theory gave central importance to the notion that people
form idiosyncratic personality theories. Other psychologists have
pointed to the importance of cognitive complexity in the differential
assessment of persons (literature reviewed by Crockett, 1965; and
Hastorf, Schneider, and Polefka, 1970), and the importance of the
strength of a subject's evaluative dimension in judging others (Steiner,
1954; Warr and Simms, 1965). In addition, Zajonc (1960) and Cohen
(1961) have shown that pieces of information are related more complexly
when subjects expect to receive rather than to communicate the informa-
tion.

Research in this area has been of two sorts: (1) trying to un-
cover commonly held personality theories derived from stereotypes or
language based correlations, and (2) trying to link idiographic impli-
cit theories to personality traits like cognitive complexity or authori-
tarianism. Concerning the latter, Schneider (1973) concludes ”it is
relatively easy to show that individuals differ in their implicit per-
sonality theories, but there has been limited success relating such dif-
ferences to traditional personality variables."

The methodology of empathy research has dealt inconsistently with
the effects of implicit personality theories, sometimes maximizing
their influence, sometimes minimizing it. Methods employing the judg—
ment of emotions from photographs, diagrams, or brief vignettes imply
that empathy is merely the interpretation of specific stimuli, and
thereby minimize the implicit personality effect. Methods utilizing

personality tests, sociometric measures, and handwriting analyses imply

 

 

21

that from a few cues the subject extrapolates an image of a cohesive
personality structure, maximizing the implicit personality influence.
The present research recognizes the discrepancy and employs each of

the two methods, one designed to enhance the influence of implicit per—
sonality theories--a personality test measure, and one designed to

minimize the effect--a videotape measure.

 

 

IV. DEFINITIONS, HYPOTHESES, AND RATIONALE
FOR HYPOTHESES

Hypothesis I: With regard to Perception, subjects with high Intui-
tive orientation will show more empathic accuracy than subjects
with middle Intuitive-Sensing or Sensing orientations.

Definitions:

Perception (P): According to Jung, the method by which sensory
stimulus data are translated into information by the sub-
ject. The two extreme types of Perception are Intuition
and Sensation.

 

Intuition (1): According to Jung indirect awareness by the un-
conscious, accompanied by conscious ideas or associations
(hunches) which become attached to the unconscious material.

Sensation (S): According to Jung direct awareness through the
five senses.

Middle Intuition-Sensation (N/S): Awareness which alternates
between the use of Intuition and Sensation.

Hypothesis II: With regard to Judgment, subjects with high Feeling
orientation will show more empathic accuracy than subjects with
middle Feeling—Thinking or Thinking orientations.

Definitions:

Judgment (J): According to Jung, the method by which conclusions
and/or decisions are reached. The two extreme types of
Judgment are Feeling and Thinking.

Feeling (F): According to Jung, an emotional procedure culminat-
ing in an appraisal on a subjective basis, usually involving
some form of a good-bad dimension.

 

Thinking (T): According to Jung,a procedure aimed at making an
impersonal finding on the basis of logic or reasoning.

Middle Feeling-Thinking (F/T): A decision making process which
alternates between feeling and thinking.

22

23

Rationale for Hypothese_>I and II: These hypotheses are based on the
conception that empathy is an alogical, intuitive process aimed prim-
arily at understanding the feelings of others. Theoretical support for
these hypotheses can be linked to Sullivan's developmental approach
where empathy is prototaxic and parataxic, i.e., intuitive, originating
when anxiety in the mother is apprehended by the child.

Hypothesis III: Similarity of Perceiver orientation between subjects
and targets will result in more empathic accuracy than between
subjects and targets dissimilar in Perceiver orientation.

Hypothesis IV: Similarity of Judger orientation between subjects and
targets will result in more empathic accuracy than between subjects
and targets dissimilar in Judger orientation.

Definitions:
Similarity of Orientation: The presence of the same Perception

or Judgment function in both target and subject, e.g., a target
categorized as a Senser and a subject categorized as a Senser.

 

Dissimilarity of Orientation: The presence of opposite or par—
tially opposite Perception or Judgment functions in both target
and subject, e.g., a target categorized as Senser and a subject
categorized as either Intuiter or Middle Intuiter-Senser.

 

 

Target: A person viewed via videotape with whom subjects are
asked to try to be empathic.

Rationale for Hypotheses III and IV: These hypotheses are related to

 

the assumption that we understand best those who are similar to us.
According to this reasoning, an individual discriminates more stimuli in
areas which are familiar to him, and thus can be more accurate, than in
areas in which he is unfamiliar. Allport (1937) and others report this
effect for age, sex, background, complexity, and personality character—

istics. Smith (1966) defines empathy in terms of similarity.

 

24

Hypothesis V: With regard to empathy viewed via a videotape source,
Feeling oriented subjects will be most accurate on a feeling-
oriented measure, while Thinking oriented subjects will be most
accurate on a thinking-oriented measure.

Definitions:

Feeling-oriented measure: A multiple—choice measure whereby sub-
jects choose which of four alternative feelings a particular
target experienced immediately prior to the completion of a video-
tape sequence.

Thinking-oriented measure: A multiple-choice measure whereby
subjects choose which of four alternative thoughts a particular
target experienced immediately prior to the completion of a video-
tape sequence.

Rationale for Hypothesis V: Hypothesis V is a logical extension of the

 

similarity hypotheses related to employing the variables of Judgment:
those designated as Thinkers should discriminate thinking in others
better than those designated as Feelers, and Xi££.!§£§§: If this
hypothesis is not supported, one of two assumptions may be true:
(1) although individuals indicate that they use one process of attending
more than the other, they do not focus on that process in others, or
(2) an error source has interfered with assessing this effect.
Hypothesis VI: Neither the order in which the target groups were pre-
sented, nor the order in which the films within the target groups
were presented, will result in significant increases in empathic
ability as the research progresses.
Rationale for Hypothesis VI: Increased empathic accuracy over sessions
can be a confounding variable denoting practice effects or reduction of
anxiety. Since empathy has been considered a relatively permanent, or
slowly changeable trait, an increase in empathy might undercut this
assumption.

However, increased empathy can originate from two sources:

 

25

(1) learning to be more empathic in general, and (2) learning to be
more empathic toward a particular target. If (1), then learning becomes
a confounding influence on ability. The effect of unequal learning
rates, pre-test abilities, etc., cannot be assessed, and the results
must be considered unstable, capable of changing at the next practice.
If (2), then the results may represent a natural, less distorted process
of empathic understanding related to familiarity with the stimulus.
Hypothesis VII: The two target individuals representing each Intui-
ting—Sensing, Feeling-Thinking orientation will draw similar
empathy scores from the subjects.

Rationale for Hypothesis VII: Since two targets have been used to

 

represent each extreme combination of Perceiving and Judging functions
such that there are two Intuitive-Feelers, two Intuitive-Thinkers, two
Senser-Feelers, and two Senser-Thinkers, the two individuals represent-
ing each type should draw similar scores. If this occurs, we can
assume that the desired target characteristics were salient factors in
the empathic discrimination of the subjects. If this fails to occur,
the conclusion that the targets actually presented discriminable data
according to their label is questionable. This particularly effects
Hypotheses III and IV, since similarity unrecognized in targets can
hardly be assumed to influence empathic accuracy.
Hypothesis VIII: A different pattern of mean responses will exist on
the videotape measures of empathy than on the personality test
measure.

Rationale for Hypothg§is VIII: While both personality test and tests

 

based on videotape stimuli have been used to assess empathy, it is

reasonable to assume that they are not measuring precisely the same

26

qualities. Personality tests measure global abstractions in response
to a stimulus presentation, while videotape tests measure the meaning
of specific behaviors. The personality test assumes consistency, self-
knowledge, and honesty on the part of the target, while specific
behaviors on the videotape are more related to the situational context,
may be straightforward or camoflaging, and may be ”out of character“
from time to time.

In addition, certain empathy processes may be detected by one

measure, hidden by others. Strategy behavior may be facilitated by per-

 

sonality tests where diverse information must be related to hitherto
unexplored questions. On the other hand, extrapolation may cause con—
fusion and error in videotape tests where specific situational and

behavioral cues are to be interpreted.

 

V. MEASURES

The measures used in this research may be divided into “personal-
ity assessment measures" and ”empathy measures”. The personality
assessment measures were the Extravert-Introvert, Sensing-Intuition,
and Thinking-Feeling scales from the Myers-Briggs Type Indicator. The
empathy measures were the Videotape Empathy Series and the Personality

Empathy Test both constructed by the experimenter.

Personality Assessment Measures

 

The Myers-Briggs Type Indicator

 

The Myers-Briggs Type Indicator is a 166 item forced-choice persi
ality test which purports to measure personality type from a Jungian
perspective. The MBTI measures four central dimensions: Extraversion»
Introversion (E-I), Sensing-Intuition (S-N), Thinking—Feeling (T-F), a1
Judging-Perceiving (J-P). Each person is seen as functioning primaril,
within the framework of one side of each dichotomy.

While each of the dichotomies seemed relevant to the functioning
of empathy, the Judging-Perceiving dimension was eliminated when searc
of the literature (Stricker and Ross, 1964; and Sundberg and Mendelsoh
in Buros, 1961) indicated that the J-P dimension seemed to be measurin
preference for order and planning versus preference for spontaneity an
novelty, rather than preference for making decisions versus preference

for open-ended experiencing as the manual suggests. Correlations betw

27

28

the J-P and S-N dimensions have also been found to range from 0.26 to
0.47. In this context, these moderately strong correlations provide
good reason to discount the use of both scales concurrently.
Intercorrelation among the other MBTI scales were cited in the
MBTI manual. They include eight studies of high school and college
populations. Correlations between the E—I and S—N scales ranged from
-.14 (N = 184) to 0.06 (N = 2511). Correlations between the E—I and
T-F scales ranged from —.19 (N = 240) to 0.04 (N = 184). Correlations
between the S-N and T-F scales ranged from 0.02 (N = 541) to 0.10
(N = 614). In sum, the scales appear to be largely independent of one
another.

a) Extraversion-Introversion: Myers (1962) describes E-I in the

 

following way:
The introvert's main interests are in the inner world of concepts
and ideas, while the extravert's main interests are in the outer
world of people and things. Therefore, when circumstances permit,
the introvert directs both perception and judgment upon ideas,
while the extravert likes to direct both upon his outside environ-
ment.... Introverts are harder to understand than extraverts for
two reasons. They are not merely less communicative; they are

also a good deal more complicated. (P. 57)

Although Myers and Briggs conceived of all four dimensions to be
dichotomous, few investigators agree that statistical analyses of the
regression curves of the dichotomies support this contention. More—
over, Jung (1928) describes as less differentiated the “normal man, who
is normal partly because nothing excessive is allowed." Following this
thinking, and the substantial amount of research that has disconfirmed
the dichotomous nature of the scales, this study will consider extra—

version-introversion, sensing—intuition, and thinking-feeling each to be

29

continua. See Appendix B for confirmation of the non-dichotomous nature
of the scores of the initial sample in this study.

In this research, the middle range of the E-I scale was used as
a control for (l) the effect of extreme extravert-introvert response
styles grossly influencing behavior, and (2) the reduction of adjust-
ment as a confounding variable.

Validational data presented in the manual shows the E-I scale to
correlate positively (p< .01 with Deference and Abasement and nega—
tively with Exhibition, Affiliation and Dominance on the Edwards Person-
ality Preference Schedule (EPPS).1 0n the Allport-Vernon-Lindzey Scale
of Values (AVL), E-I correlated positively (p< .01) with Theoretical
and Aesthetic scales, while correlating negatively with Economic and
Political scales. 0n the Personality Research Inventory (PRI), E-I
correlated positively (p< .01) with Free-floating Anxiety, Impulsivee
ness, and Gregariousness. Introverts have less people-oriented values
(Theoretical and Aesthetic), and seem to react to others with a flight
rather than fight reaction (Abasement and Deference). While introverts
seem to experience more anxiety, they also seem to be more independent.
The extraverts, on the other hand, present themselves more to others
(Exhibition, Affiliation), have more people-interactive values
(Economics and Politics), and fight rather than flee when confronted
(Dominance). These results seemingly correlate accurately with tradi-

tional definitions of extraversion-introversion.

 

1Extraverts score high on the E-I scale.

30

Additional correlational data collected by Stricker and Ross
(1964) found the E-I scale to correlate positively with the Minnesota
Multiphasic Personality Inventory (MMPI) extraversion-introversion scale
(Si, p_= 0.63) and the Gray—Wheelwright Psychological Type Questionnaire,
an earlier questionnaire measuring Jung's psychological types (p.= 0.79).
Stricker and Ross also cite previous research reporting highly signifi—
cant correlations with the E-I scale on the Maudsley Personality Inven—
tory (Eysenck, 1959 and Howarth, 1962), and an extraversionvintroversion
factor identified by Ross (1963). Stricker and Ross conclude that the
MBTI E-I scale measures extraversion-introversion as it is commonly
defined: "interest and facility in social relations, frequently in-
volving talkativeness“ (Stricker and Ross, 1964).

Stricker and Ross also note the possibility that the E-I scale

may be partially measuring adjustment. The E-I correlated (Ef1.01) with

a neurotic scale (0, p_ .39) and three of four psychotic scales (Pt,

p.= .30; Ma, y_= -.29; and Sc, p_= .23) on the MMPI. While it did corre-
late consistently with relevant Maudsley scales, and correlated nega-
tively with the CPI Self-Assurance scale, previous studies cited by
Stricker and Ross have found the E-I scale to correlate consistently with
measures of adjustment: the Maudsley Neuroticism scale (Howarth, 1962),
the PRI Free-Floating Anxiety scale (Myers, 1962b), ratings of Needs
Psychologist's Attention and Low Stamina (Ross, 1961).

b) Sensingrlntuition: Myers (1962b) describes intuition and sens-

ing in the following way:

31

When people prefer sensing, they find too much of interest in

the actuality around them to spend much energy listening for

ideas out of nowhere. When people prefer intuition, they are too

much interested in all the possibilities that occur to them to
give a whole lot of notice to the actualities. For instance, the
reader who confines his attention strictly to what is said here
on the page is following the habit of the people who prefer sens-
ing. One who reads between the lines and runs ahead to the pos-
sibilities which arise in his own mind is illustrating the way

of the people who prefer intuition. (Pp. 51-52)

Validational data presented in the manual shows the S—N scale to
correlate positively (Ef:.01) with Autonomy and negatively with Defer-
ence and Order on the EPPS.1 0n the AVL, S-N correlated positively
(ps .01) with the Aesthetic and Theoretical, but negatively with the
Economic and Political scales. On the PRI, the S-N scale correlated
positively (p< .01) with Liking to Use Mind,* Artistic versus Practical,

Tolerance of Complexity, Impulsiveness,*

and Progressive versus Conserva-
tive. The S-N scale correlated negatively with the Gregariousness,
Masculine Vigor and Social Know-How scales on the same test. Inspection
of these positive and negative lists shows the Intuitive end of the con—
tinuum to be aesthetic, theoretical and autonomous, while the Senser end
to be gregarious, ordered and economic. Although “hunch” oriented be—
havior, usually closely associated with intuitive behavior was neither
pin-pointed by the MBTI questions nor has it been expressly studied with
relation to the S-N scale, behavior contingent on making hunches does
seem to correlate positively, supporting the conceptual definitions

stated by the manual.

 

1Intuiters score high on the S-N scale.

*
Denotes more than one study in which the scale has correlated
significantly at the .01 level with the attribute indicated.

 

32

Data by Stricker and Ross (1964) found a significant positive
correlation (pf:.01) between the Myers-Briggs S—N scale and a similar
S-N scale on the Gray-Wheelwright Psychological Type Questionnaire.
However, they also note that non-conformity seems positively correlated
to the intuition end of the scale. In this case, the scale may have a
heavy conformity-non—conformity loading, or since Stricker and Ross
were using the N score on a non-continuous basis with the S, the N
scale may be measuring non-conforming intuitives rather than conforming
ones. The chief drawback of the S-N scale is its lack of breadth in
covering the diverse aspects of intuition, such as valuing hunch be-
havior versus valuing empiricism. It is, however, the only scale which
measures preference for intuition and gives validational data to sup-
port its claim to do so.

c) Thinking-Feeling: Myers (1962b) describes the Thinking-

Feeling dimension in the following way. There are

two distinct and sharply contrasting ways of coming to conclusions.

One way is by the use of thinking, which is a logical process,

aimed at an impersonal finding. The other way is by the use of

feeling, which is a process of appreciation, equally reasonable in
its fashion, bestowing on things a personal, subjective value....

If, when one judges these ideas, he concentrates on whether or not

they are true, that is thinking-judgment. If one is conscious

first of like or dislike, of whether these concepts are sympathetic
or antagonistic to the ideas he prizes, that is feeling-judgment.

(P. 52)

Validational data presented in the manual shows the T—F scale to corre-
late positively (p< .01) with the Nurturance, Affiliation, Succorance,
and Abasement scales, and negatively with the Endurance, Order, Autonomy,

1

Dominance, and Achievement scales on the EPPS. On the AVL, the T-F

 

1Feeling scores high on the T-F scale.

33

scale correlated positively (pf:.01) with the Social and Religious
scales, and negatively with the Theoretical, Economic, and Political
scales. On the PRI, the T-F scale correlated positively (BK .01) with
Free-Floating Anxiety, Spiritual vs. Material, and Tolerance for Com-
plexity scales, while it correlated negatively with Masculine Vigor,
Attitude toward Work, and Self~Sufficiency. Combined, these results
seem to indicate that the Feeler end of the continuum is related to the
reconciliation of positive and negative feelings, and social and re—
ligious life preferences. The Tolerance for Complexity correlation is
mystifying. The Thinker end of the continuum can be described as
vigorous, achievement oriented, and self—sufficient. With reference to
the manual descriptions, the subjective and evaluative aspects seem
largely fulfilled by the validational results. The impersonal, objec-
tive orientation of the Thinker seems likewise substantiated. Again

a highly significant correlation (p< .01) was found between the Myers-
Briggs T-F scale and the Gray Wheelwright Psychological Type Question-

naire by Stricker and Ross (1964).

Empathy Measures

1. The Videotape Empathnyeries (V.E.S.)

The Videotape Empathy Series consists of four multiple-choice sub-
scales used to measure empathic accuracy of subjects toward group members
viewed via videotape. The four measurements include: Thinking Empathy,
Feeling Empathy, Behavior Empathy, and Social-interaction Empathy.
Defined operationally, Thinking Empathy is the ability to accurately

predict, given four alternative choices, what a target individual would

 

 

34

say he was thinking at a particular moment during a videotaped group
session. Similarly, Feeling Empathy consists of the ability to accur-
ately predict what a target individual would say he was feeling at a
particular moment during a videotaped group session. Behavior Empathy
is defined as the ability to predict which behaviors a target individual
will perform immediately following an experimenter stopped segment.
Social—interaction Empathy is the ability to choose which of three other
group members a target individual will say he places closest or farthest
from himself on selected sociometric continua, e.g., trust, liking,

similarity to self.

Table 1. Outline for constructing the Thinking and Feeling Empathy
tests.

 

 

1. 1/2 hour group session videotaped.

II. Awareness training.
III. Recall session, group members interviewed separately.
IV. Construction of test items by experimenter.

V. Evaluation of face validity of items by group members.

VI. Omission or rewording of test items by experimenter with
reference to group members input.

VII. Pilot study to determine level of difficulty for items and
weight of alternative responses within items.

VIII. Omission or substitution of items or alternatives according
to level of difficulty and/0r unbalanced weight of alterna—
tives.

 

 

35

a) The Thinking and Feeling Empathy Tests

The procedure for constructing the Thinking and Feeling Empathy
tests is outlined in Table 1, on the preceding page. Since essentially
the Thinking and Feeling Empathy tests ask subjects to predict what
group members would say they were experiencing during intermittent seg-
ments of their group sessions, it is imperative that the group members
relate as accurately as possible what they were experiencing as the
group session progressed. In order to facilitate the recognition of
subtle as well as obvious experiential data, a special awareness train-
ing technique was instituted after the first group session and before
the first recall session.1 The awareness training session consisted of
arranging feelings presented in the form of a jigsaw puzzle (see
Appendix 0) into dichotomies ”experience frequently“ and ”experience
rarely” and later “feel comfortable experiencing” and “feel uncomfort-
able experiencing”. From each of the dichotomies the group members were
asked to select a feeling they would like to explore in greater depth.
Using an adaptation of Gendlin's focusing technique (see Appendix 0),
each group member was instructed to silently attend to the bodily
sensations, emotional responses and mental meanderings accompanying the
experience of the original feeling. For approximately 10 minutes, the
individual was guided toward discovering the many gradations of experi-

ences occurring concurrently. After a short rest, another feeling was

 

1The awareness training session was given following the first
group session in order to reduce contamination or carry over to the
videotaped group session. Awareness training was not instituted on re-
call days in order to reduce the chance of bringing in extraneous
feelings which might interfere with recall.

36

chosen and the procedure was repeated. A total of six feelings were
selected for focusing. The awareness training session lasted one hour.
A technique closely related to the Interpersonal Process Recall
developed by Kagan and Krathwohl (1967) was used to elicit feeling and
thinking responses during the replay of the videotaped session. Recall
data was collected on the odd days following the group sessions (see
Figure 1), each group member being interviewed individually for
approximately 1 1/2 hours. During the recall session, the group member
was asked to watch the videotape and say ”stop” whenever she recognized
feelings she had had about herself or the other group members, thoughts
she had had but did not express, bodily sensations she had experienced,
or expectations she had had about herself or the situation.1 Statements
which the group member made after she stopped the tape were written
down by the interviewer, either by capturing the general tone and mean-
ing of the answer or by an exact quotation if the response was short.
In addition to the verbal explication by the group member, she was given
a list of 70 feeling words grouped by similarity, e g., angry, frus-
trated, disgusted (see Appendix E), and asked to check off all the rele-
vant feelings associated with the interaction just interrupted. She was
permitted to check the feelings before or after she commented on the

interaction, depending on which was more functional. (She might have

 

1A remote control switch by which the group member could stop the
tape was originally planned for use instead of saying ”stop”. The video-
tape technician suggested, however, that stopping the tape for as long
as a minute might have a detrimental effect on the quality of the tape,
so the ”stop“ method was instituted instead. It is recognized by the
experimenter that a remote control switch has advantages over the "stop"
method especially with regard to making it easier to stop anxiety toned
segments.

37

.mcowmmwm Fqumc Fascw>wvcv use mmcwpmms azocm Low o_:uw;om weak

.P mczmwm

LmnEmE asogm dz u a mamasme azogm Fm u o memnEmE azogm #2 u m mgmnsme asocm mm u <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

garage;
mmmcmLmZm
o copmmmm . . . . a cormmmm . . . . o comemm . . . . o comewm . . . .
. a u m < . a o m < . a u m < . a u m <
FFmomL a cowmmmm Fpmumg m cowmmmm Ppmumg N cowmmmm F_mumc P :owmmmm
u commmmm azocm u cowmmmm azogm u sewmmwm azocm o :owmmmm asocm
Ppmomg Ffimomc __mowc Fpmum;
m :owmmmm m cowmmmm m cowmmwm m cowmmmm
__muwg FFmomL __mumc Freumc
< covmmmm < cowmmmm < commmmm < cowmmmm
FPmomL Ffimumc __momg Prmumc
m sea in see 0 see m sea e sag m sea N sag _ sac

 

 

 

 

 

 

 

 

 

38

wanted to state clear or strong feelings first, or check the list first,
if feelings were less clear or hard to verbalize.) Finally, the inter-
viewer sometimes stopped the videotape to elicit feeling and thinking
responses which she had summised from cues in the interaction, such as
abrupt shifts in theme, body posture, and voice level or tone.

After the recall session, the experimenter organized the recall
data into feeling and thinking responses from which 6 feeling and 6
thinking multiple choice questions were constructed. Each feeling item
contained a verbalized and a clarifying statement about the feeling,
e.g., "careful--not to say anything that would hurt ______fs feelings.”
This is in accord with Chapman's (1966) observation that “adjectives
(describing feelings), without clarifying phrases, have different mean-
ings for different people.... The use of phrases might be more specific
and therefore might give more consistency of interpretation than the
adjectives alone“ (p. 56). Thinking items are also stated in sentences,
e.g., “I hope they ask me what instrument I play." Three “incorrect“
alternatives were constructed by the experimenter by using either feel-
ing words left unchecked by the group member or by thoughts not indi-
cated as present during recall.

To ascertain the accuracy of the multiple choice items, the video-
tape was replayed for the group members after all the group sessions
had been completed. The tape was stopped after each interaction for
which a multiple choice item had been constructed, and the designated
group member was given the item and asked to mark whether the alterna-
tive was "true”, "false”, or "somewhat true". Alternatives which the

experimenter had intended to be "incorrect", but which the group member

39

marked “true” or “somewhat true” were either omitted or reworded to
make them false.

In order to test the likelihood to choose each ”incorrect"
alternative per item, a pilot study was conducted. Twenty subjects
viewed each videotape in series and indicated which feeling or thinking
response they thought the targeted individual would say she was experi-
encing just before the videotape stopped. Results of an item analysis
showed the average percent correct to be 34%, a little above chance
prediction. This meager score was thought to result from two sources:
(1) the difficulty of some items, and (2) the unusual drawing power of
some ”incorrect" alternatives. Items diagnosed to be suffering from
either of these two error sources were either omitted, reworded, or
new alternatives were substituted.

The instructions given to the subjects before taking each empathy
test are similar to those used by both Chapman (1966) and Campbell
(1967). They are:

The following is a multiple choice test used in conjunction
with a videotape. Prior to today a series of videotapes were
made in which four women participated in group discussions. Sub—
sequently each group member individually watched a replay of each
videotape and recalled what she was thinking or feeling during
certain segments of the tape. One correct and three false altern-
atives were constructed for many of the segments recalled.

In this part of the experiment, the first videotape of Group
A will be replayed and stopped at the segments mentioned above.
When the videotape stops you are to turn the page and read a ques—
tion about one of the group members. You will be given ten
seconds to think about what that individual was thinking or feel—
ing just before the videotape stopped. While you are thinking,
the alternatives to the questions which begin in the middle of
the page, should be kept covered with the sheet of paper provided.
After ten seconds, the experimenter will signal you to remove the
paper and choose which of the four alternatives you think the indi-
cated group member would say she was feeling or thinking just
before the videotape stopped. After you have circled the answer

40

to the question, 00 NOT TURN THE PAGE. Wait until the videotape

has been started and stopped again at a new segment to turn the

page. IMMEDIATELY COVER THE BOTTOM HALF OF THE PAGE WITH THE

PAPER PROVIDED WHEN YOU TURN THE PAGE.

Remember, you are to choose which alternative the indicated
group member WOULD SAY she was thinking or feeling just before
the videotape is stopped.

Before each videotaped segment was begun, the subject was told
the name of the group member who was the target of the next question.
This was done to insure that the subject was paying attention to the
behavior of the targeted member and not, by chance, to another member.
The names of the group members were posted near their image on the
screen.

The ten-second reflection period was designed to enable the sub-
ject to get a clear picture of her own view of the situation before
reading the four logical possibilities presented in the multiple choice
alternatives. This was performed to avoid the possibility that the
cleverness or persuasibility of the "wrong“ alternatives would be a

crucial error factor in determining the subject's response.

Results of similar tests. Although the procedure for construct-

 

ing the Videotape Empathy Series (VES) is unique, two empathy scales
constructed by Chapman (1966) and Campbell (1967) have many similari-
ties to the VES. Both Chapman's and Campbell's Affective Sensitivity
scales (AS I and AS II, respectively) were designed to measure the
empathy of persons viewing certain videotaped counselor interviews.

The subject was presented with short counseling segments and asked to
choose which of three alternative responses most accurately stated what

the counselee was feeling about either himself or about his counselor.

41

Correct and incorrect responses in Chapman's empathy test were
determined by two criterion groups, judged high and low in empathic
regard who were asked to view short counseling vignettes and "try to
:feel as the client felt" when the tape was stopped. From a list of 57
feeling adjectives, each individual selected those feelings which he
thought were most clearly revealed by the counselee in the vignette.

The item selected most frequently by the high criterion group and not
by the low criterion group, and vice versa, were selected as items for
the AS I. Four to seven items were chosen for each vignette.

The subjects, master's candidates and counseling educators at two
universities, were ranked on an empathy continuum by their peers or
superiors. The upper and lower thirds of the distribution were shown
the counseling vignettes and asked to ”try to feel as the client last
felt.“ For each vignette they were given the 4 to 7 feeling words
described in the previous paragraph and asked to rate each on a con-
tinuum for “I have this kind of feeling strongly,‘l to "I have this kind
of feeling not at all.” No significant difference was found between the
means of the two groups. Among the explanations for these results were
that (l) the feeling adjectives were too imprecise and needed clarifica—
tion, and (2) the rankings used to select the criterion groups and the
subjects according to their empathy skills, by the admission of some of
the rankers, was sketchy at best, faulty at worst.

The present research has taken into account Chapman's first ob-
servation by adding explanatory statements to the feeling adjectives for
clarification. Since no outside judges are used, the second problem is

not applicable.

42

Campbell's AS II also followed Chapman's suggestion and added
explanatory statements to clarify feelings described. Campbell's
method for ascertaining the true feelings presented in the videotaped
segments was more complicated, however. He used three sources for
identifying feelings: (1) judges who watched the videotapes and
described the feelings they observed; (2) judges supplied with back-
ground data regarding previous counseling interviews between the
counselor and counselee; and (3) actual statements of feelings made by
the counselee during an Interpersonal Process Recall of the counseling
session. In addition three levels of sophistication of judges were
used: practicing counselors and doctoral candidates in counseling;
M.A. counseling and guidance candidates, and non-counseling and guid-
ance individuals. The latter two groups were used to construct
“distractors”, i e., “wrong" answers.

Based on item analyses of total scores, peer ratings and staff
ratings, Campbell found no significant differences between the three
methods of item selection, although he found that some counselees
verbalized more about their feelings than others, and therefore pro«
duced more data from which to garner type III items. Concerning the
AS II as a whole, low positive correlations occurred between scale
scores and peer and staff ratings of counselor effectiveness (average
correlation = .26), and between scale scores and peer and staff ratings
of affective sensitivity (average correlation = .38).

Although suggestions for improving the AS II were not offered
by Campbell, at least two alterations seem necessary to this experi-

menter. First, some method for verification that the target individual

43

was actually feeling the feelings attributed to him by the judges
should be implemented. Second, a method for analyzing the quality of
the "wrong” answers should be constructed.

With relation to the use of outside judges, in the opinion of
this experimenter outside judges often confound the results of the
experiment. The subject must be empathic with an empathic person,
rather than with the target person himself. Often outside judges are
employed, however, because they are more willing to identify negative
or anxiety producing feelings than are the target individuals. To
overcome this predicament in the present research, it was decided that
helping the group members recognize and accept their feelings would, if
successful, alleviate most of the need for outside judges. Therefore
the series of awareness exercises described previously were constructed
to facilitate this learning.

To insure that the “wrong“ answers on the multiple choice test
were not the greatest variant in determing the subject's empathic
ability, or lack of it, either because some of the "wrong” answers were
“right” or because the level of difficulty of the items was extreme, a
method for analyzing the multiple choice items was constructed. As
described previously, each group member reviewed the videotape and the
multiple choice items related to her, commenting on the right and wrong
alternatives and rating each item.

ng Videotaped Group Discussion. The videotaped discussions were
designed to provide several samples of everyday behavior. While it was
thought that examples of everyday conversations could be generated by

allowing the group members simply to talk among themselves about

44

whatever seemed relevant at the time, in order to reduce anxiety which
might arise from the special circumstance of being videotaped, two
discussion topics were supplied to the group members at the beginning
of each session. Specifically, the group members were told that their
primary job was to get to know each other as well as they could. They
could do this by talking about either of the two topics supplied at the
beginning of each group session, or by discussing whatever they thought
would facilitate their understanding of each other.

The choice of topics supplied for the videotaped discussion are
based on observations from two theories. The first is Eric Berne's,
who observes that individuals communicate on certain levels of inter-
action during their everyday lives. They begin at simple levels and
move to more intricate ones. His levels: the ritual, pastime, game and
intimate interaction, reflect how committed the individual is to the
interaction and how safe he feels in the interaction.

The second theoretical observation, Maslow's, states that people
move through their lives in a hierarchical fashion, first solving basic
needs then moving on to more complex interpersonal ones, and finally to
self-enhancing ones. His levels: the physiological, safety, belonging-
ness and love, esteem, and self-actualization, reflect the needs the
individual has satisfied, what problems the individual is focussing on,
and which needs are still vaguely defined because he has not entertained
them yet. With these two observations in mind, an order of topical
discussion was designed that would encounter the safest and least commit-
ting discussions first and would move into deeper, more complex, and

riskier subjects later.

45

The topics and order of discussion are: (1) What are the advant-
ages and disadvantages of dorm life? (pastime); (2) How do you usually
spend your time? (pastime--allows predominant hierarchy level to
emerge); (3) As a child, how did you get what you wanted from your
parents? (game—-safety, belongingness and love); (4) How do you react
when you"re angry? (game, intimate-—safety); (5) If someone thought
something about you which might hurt you, would you rather have him
tell you about it, or keep it to himself? Why? (game, intimate--
safety, esteem); (6) What are some of the things you've noticed about
how your group functions? (game, intimate--safety, belongingness and
love, esteem); (7) Are you more like a Parent, an Adult or a Child?
(game, intimate--esteem); (8) What would you have liked to have said
in this group, but you didnt? (intimate--safety, belongingness and love,
esteem). In fact, the group did discuss the two topics supplied each
session, but often the discussion of these was minimal, lasting between 5
and 10 minutes. The actual group discussions, then, usually reflected
topics generated by the group members. The topics characteristically
remained at Berne's pastime level and Maslow's belongingness level,
although when the supplied topics were discussed, the level of conversa-
tion usually attained that previously signified in parentheses.

Visual organization of the videotaped group. The visual format

 

tZY'which the videotaped members are arranged on the screen is an
lrnportant factor in the degree of empathic accuracy that is technically
airtainable. Three methods of recording group sessions via remOte-
(XJritrol cameras were considered; two being encorporated, one being dis-

car~ded. Factors seeming most essential were: (1) placing four

 

46

individuals on the screen without appearing so distant as to miss
subtle behavioral cues needed for empathic judgment, (2) obtaining a
front view, as opposed to a side view, of all the target individuals,
and (3) retaining a sense of the group, i.e., knowing who is talking
to whom.

The first and most natural method was simply to videotape all
four group members sitting in a half-circle. This method satisfied
condition 3, i.e., it allowed a sense of group functioning, but the
image of the target individuals on the screen was too small to allow
the easy interpretation of subtle, but important behavioral cues.
Moreover, only profiles of the two outside members were usually attain-
able. A second split-screen method solved the latter two concerns, but
created the first, the absence of a feeling for group cohesiveness, as
the two inside members no longer faced each other but the outside of
the screen (see Figure 2).

A final method, utilizing one camera to focus on one group
member in a corner of the screen while shooting the total group as a
backdrop (see Figure 3), was discarded because of the unpredictable
nature as to who would finally be the targeted individual of a particu-
lar series. (Empathy sequences were not determined until after recall
sessions were complete.) Also, while this quarter-screen method
allowed the total group to be seen, it reduced attentiveness to the
other members of the group as they appeared smaller and more recessed.

A compromise to maximize satisfaction of the factors initially
mentioned was accomplished by taping the group as a whole for the first

five minutes, followed by 10 minutes of split-screen close-up focussing,

 

Figure 2. Eye direction of group members using split-
screen cameras.

 

 

 

 

O

 

 

00
OO

 

 

 

Figure 3. Videotape screen configuration using one
member as primary focus.

five minutes of full group, 10 minutes of split-screen. Seating posi-
tions were rotated each session to disallow for side view distortions

being distributed unequally.

b) The Behavior Empathy Test

 

The Behavior Empathy Test (B.E.) was designed to measure how
accurately a subject could predict which behaviors a target individual

would engage in following the cessation of a videotaped segment.

48

It answered the question, “how well does an individual know what another
individual is going to do?"

B.E. was constructed by noting instances in each group session
where target individuals accompanied their verbal messages with clearly
describable behavior, e.g., asked others for their opinions, laughed at
a remark, changed the subject. From these instances, an equal number
of behavioral responses were chosen per target person, and "incorrect“
behavioral responses were constructed by the experimenter. The problem
of the subjective validity of “incorrect" responses did not arise with
this test, since the actual behavior of the subject could be objectively
ascertained by watching the tape, i.e., it was obvious from watching
the tape which behaviors the individual did or did not perform. The
relative power of the “incorrect" alternatives was examined by the same
procedure mentioned in the previous section. l'Incorrect" responses

with exaggerated drawing power were omitted or reworded.

c) The Social-interaction Test

The Social-interaction Empathy test (SIE) was designed to measure
how accurately a subject could predict which of the group members a
targeted individual would feel closest to or farthest from on various
sociometric continua: trust, liking, similarity to self. It was de-
signed to determine how well an individual understood the social ties
between group members.

SIE was constructed from double-layered scales (see Figure 4)
given to each group member following every videotape recall session.

One level of the scale asked each group member to rate the remaining

49

 

Sue Gail Betty

x x x

\ V I I , \ I V

\ \ \

LIKE \ \\ q_IKE
MOST \\ \ \ LEAST
\ \ \
\\ \\ \‘
\ \

 

\
PERSON I HAVE KNOWN PERSON I HAVE KNOWN
AND LIKED MOST AND LIKED LEAST

Figure 4. An example of a Social-interaction rating scale.

group members on a continuum from ”most" to "least” according to a
specific sociometric trait. Because this level was an intragroup
comparison, both ends of the continuum were instructed to be used.

To balance this forced extreme rating, a second level was employed
which instructed the individual to draw a line from each person on the
first level to a position on the second level which represented where
that person belonged with respect to all the people he had known
regarding that trait. The top rating was the only rating used in the
research. The second level was designed to reduce anxiety concerning
extreme judgments and to balance the top rating according to his own
perspective.

The SIE test was composed of multiple choice questions which re—
flected the ratings of the sociometric scales, e.g., ”Who would Mary
say she liked best?" Because the items were constructed directly from
the scales, rewording was unnecessary. An item analysis was used to
determine difficulty level, and items deemed too easy or too difficult

were omitted.

50

2. The Personality Empathy Test

The Personality Empathy Test (PET) is a variant of a widely used
empathy test format in which subjects and target individuals both take
a personality, interest, or attitude inventory. The subjects view a
limited sample of the target person's behavior, e.g., interview, dis-
cussion, photograph, tape-recording, handwriting, and retake all or
part of the pre-test inventory as they think the target person would
take it. The Personality Empathy Test is a 100 item test composed of
questions from the Myers-Briggs E-I, S-N and T-F scales, and the Com-
plexity scale of the Omnibus Personality Inventory. The subject is asked
to answer each item the way he thinks specified videotaped group members
would answer that item. The test was given after the final taped
session of each videotaped group.

Often multiple-choice empathy tests are constructed by merely ask-
ing the subjects to fill out a complete personality test as a target
person would. One major drawback of this method is that it includes
items which the target person himself may answer with uncertainty.

Thus the variability in the test-retest reliability is built into the
new method. During the pre-experiment testing in this research, each
individual is asked to mark an "X“ by every question which "importantly
describes some aspect of her personality." She is instructed to mark
between 30 and 40 of 150 responses. The PET was constructed from these
designated items of each videotaped group member.

Cronbach (1955) cites another drawback of this method of empathy
testing. Some questions may have a sterotyped answer, such as "Some-

times I just like to relax and listen to music," which reflects more

 

51

knowledge of the average person, or a stereotype like ”college student",
than empathic understanding of the specific target person. To reduce
the effect of these non—discriminatory items, only two—alternative
items which are answered one way by more than 35% but less than 65% of
a sample of college students were used.

The PET is scored by summing the correct predictions made by each
subject about each group member. A total of 25 points is possible
concerning each group member, and a cumulative score of 100 is possible
toward the total videotaped group. Thus, both person-specific and

total empathy scores are available.

 

 

IV. PROCEDURE

Selection of Subjects apd Group MEEPEEE

Eight group members (targets), 20 pilot study subjects, and 36
experimental subjects were selected from similar populations. The
targets and pilot study subjects were selected from a group of 383
female students enrolled in introductory and personality classes at
Michigan State University during the Spring term, 1972. Following com-
pletion of videotaping and test construction, the experimental subjects
were selected from another sample of 157 female students enrolled in
various level psychology classes during the Summer term, 1972.

Female students were asked to volunteer to take a general per-
sonality interest test for which they received 2 ”experimental credits"
which could be used as extra credit toward their final grade in the
course. They were also told that they might be chosen to participate
in a second experiment involving empathy toward videotaped group
members, for which various amounts of experimental credit and money
would be given depending upon their function in the experiment. Video—
taped group members received $20 plus 4 extra credit points. Pilot
study subjects received $10 plus 4 extra credit points. Experimental
subjects received $8 plus 4 extra credit points. The difference in the
amount of payment received reflects the proportional amount of hours

worked.

52

 

 

 

53

Scores for potential subjects and targets were first screened on
the control variable, Extraversion-Introversion. Scorers falling outside
the middle fiftieth percentile, i.e., extreme scorers, were dropped from
consideration. From the remaining pool, 2 targets were chosen to repre-
sent each combination of the four extremes on the Sensing-Intuition and
Thinking-Feeling scales, i.e., ST, SF, NT, NF. Extremes were defined as
the top and bottom 25 percentiles on each scale (see Figure 5).

z 25% : 121/2%: 25% : 121/,%: 25%
Extreme Middle ' Extreme

Figure 5. Selection procedure according to distribution of

scores on experimental variables.

Similarly, potential subjects were screened first on the Extra-
version-Introversion scale with only the middle fiftieth percentile con-
sidered further. Then, four representatives of each possible extreme
and middle scoring combination were selected (see Figure 6). Pilot study
subjects had been selected at random from the distribution of experi-

mental variables.

 

 

 

' Thinking--T
Feeling (F) Feeling (F) Thinking (I)

Intuition NF NTr NT

m) (4) 44f (4)
Intuition- NF NT NT
Sensing S SF S

(151) (4) <4) (4)
Sensing SF ST ST

(S) (4) (4)F (4)

 

 

 

Total = 36 subjects

Figure 6. Classification of subjects by scores on independent
variables.

54

Procedure for Subjects

Subjects were divided into two groups. One group viewed one
videotaped group series, Group A, first. The second group viewed the
remaining videotaped group series, Group B, first. This was to assure
that any improvement in empathic understanding that accrued to practice
was equally distributed among the two groups videotaped. To determine
whether such a difference did occur, an analysis of variance
(Hypothesis VI) was performed on the data with respect to accuracy
versus first or last viewing of the series. Between 8 to 10 subjects
viewed a group series at a time.

As instructions, the subjects were told that they were partici«
pating in a study designed to measure empathic ability. Empathic abil—
ity was defined as the ability to understand what another person was
thinking or feeling the way the other person experienced it. For
instance, if a person assumed a happy stance, to cover up sadness he
really felt, it would be empathic to say he was sad. On the other hand,
if he acted happy and did not experience sadness, even if the judge
thought he was really sad underneath, it would be empathic to say he
was happy. Empathy, therefore, is the ability to understand what another
person is experiencing at a particular moment.

The format was explained: the subjects would watch two groups of
four people who were strangers to each other discuss various topics.
One group would be seen each week. In order that the group members
could be viewed more closely on the screen, two methods of presentation
would be used: sometimes the group would be shown in a semi—circle on

the screen just as they had been sitting originally, and sometimes by

 

55

using two cameras, one-half of the semi-circle would be shown on the
top half of the screen, the other half on the bottom. To make this
switch less confusing, the names of the subjects were pasted on the
top of the screen in the order they sat in the semi-circle, and each
name was pasted on the side of the screen next to where they were

sitting when the screen was split (see Figure 7).

 

L
M I J
A N P O
R D A A
Y A M N
MARY LINDA
PAM JOAN

 

 

Figure 7. Arrangement of subjects and subject-names on screen.

The subjects were told that at the beginning of each segment,
they would be advised as to which particular group member would be the
subject of the next test question. The subjects were instructed to pay
close, but not total, attention to this group member, particularly with
regard to what the individual was thinking, feeling, or about to do.

As the videotaped session progressed, the subject was told that the
experimenter would stop the tape, after which she was to turn the page
of the test booklet and cover the bottom half of the page with the
cover sheet provided, i.e., cover the alternative answers. The subject

and the experimenter would then read the question together (the

 

56

experimenter reading aloud so all subjects would be sure of the ques-
tion), and the subject would be given 10 seconds to think about her own
answer to the question. Following this 10 second period, the subject
would be instructed to remove the cover sheet, look down to the bottom
of the page, and answer which of the four alternatives she thought was
correct.

The importance of the 10 second interval to clearly answer the
question in the subject's own mind was stressed. It was explained that
all the alternatives could be logical answers for the question, and that
sometimes one alternative might fool them if they did not have their own
answer firmly in mind. The subjects were admonished to remain silent
throughout the session, so as not to influence others, and not to be
rushed when answering. It was explained that the research would be
completed in three weeks or less and they could compare notes with
others, or ask questions in a follow-up session concerning their scores,
the content of the sessions, or the research design.

Stimuli for measuring feeling, thinking and behaving empathy were
presented by means of videotape stops. The social interaction items,
concerning the total session, were asked at the end of each session.

The Personality Empathy Test was given at the end of the last session

for each group (see Figure 8).

57

 

 

 

 

 

 

 

 

 

 

 

 

Session 1 Session 2 Session 3 Session 4
Film 1: Film 2: Film 3: Film 4:
Feeling, Feeling, Feeling, Feeling,
Thinking, Thinking, Thinking, Thinking,
Behavior Behavior Behavior Behavior
Empathy Empathy Empathy Empathy
Social Social Social Social
Interaction Interaction Interaction Interaction
Empathy Empathy Empathy Empathy
Personality
Empathy
Figure 8. Order of presentation of empathy tests per group.

Statistical Procedures

 

Five repeated measures analyses of variance were performed to
determine whether the differences predicted in the hypotheses were sup—
ported or rejected by the null hypothesis. To minimize Type I errors
resulting from chance rejection, a total confidence level of .10 was
used. This seemingly high confidence leVel was divided among the five
analyses, however, such that .025 levels were acceptable for the two
primary designs and .0167 levels were acceptable for the three minor
designs. It was felt that the number and complexity of the variables,
in addition to the exploratory nature of the minor designs, warranted a
more lenient .10 confidence level. Note that the two designs measuring
the main effects have a combined confidence level of .05 as is standard

in most research.

58

Design 1 explored the differences between subject variables with
respect to videotape empathy measures, while Design 2 determined dif-
ferences between subject variables on the personality empathy measure.
Design 3 determined the effects of order of presentation on level of
empathic accuracy. Design 4 analyzed the effects of utilizing two
target individuals per target category on similarity of response
toward the two category members. Design 5 compared subject response on

videotape versus personality empathy test measures.

VII. RESULTS

Five analyses of variance using a repeated measures design were
employed to test the hypotheses. Two analyses, considered to be the
major analyses, were used to test the main effects and differential
effects. Three analyses, considered to be minor analyses, were used to
test measurement effects. Because of the complex and exploratory
nature of some of the hypotheses, a total confidence level of .10 was
used over the five analyses. The major two analyses utilized a combined
confidence level of .05 to attain significance, or .025 per design.

The three minor analyses also utilized a combined confidence level of
.05 to attain significance, requiring a .0167 level per design.

The procedure for the analysis of results is to consider the two
major designs, or main and differential effects, first, then to consider
the measurement effects. Overall interpretation of the major results
tables will proceed before more detailed inspection of the specific
analyses related to each hypothesis will be made.

Because some analyses are relevant to more than one hypothesis, a
left hand column in each major results table indicates the analyses
relevant to each hypothesis. It should be noted that careful attention
has been paid to the possibility that one significant analysis, being
related to more than one hypothesis, could result in overestimating the

strength of that analysis. The chance for this Type I error has been

59

60

decreased by (1) anticipation of its occurrence, (2) careful inspec-
tion of the direction of the means differences with respect to each
hypothesis, and (3) use of a statistic (wz) which estimates the percent
of the total variance attributable to a particular analysis. Thus,
only analyses contributing to a large percent of the variance will be

considered worthy of supporting more than one hypothesis.

Main Effects
The main effects are concerned with the total empathy displayed
by subjects regarding two personality orientations: Perceiving,
measured on a continuum from Intuition to Sensing, and Judging, measured
on a continuum from Feeling to Thinking. Hypothesis I states that
among all the Perceivers, those with an Intuitive orientation will be
most empathic. Hypothesis II states that among all the Judgers, those
with a Feeling orientation will be most empathic. Tables 2 and 3
assess these effects, Table 2 concerning empathy via a videotape source,
Table 3 concerning empathy measured via a written personality test.
While not specifically related to Hypothesis I and II, it may be
worthwhile to note the outstanding features of Tables 2 and 3 in order
to clarify their overall meaning. In Table 2, 6 of 31 analyses, or
19%, are significant at the prescribed .025 level. Chance predicts
that 1% of the analyses should be significant at this level, so it can
be assumed that some significant differences are evident beyond chance.
Table 2 suggests a strong measurement (M) component as the factor creat-
ing most of the significant variance, i.e., all significant analyses

contain an “M” variable. This factor might be interpreted as an

 

 

61

Table 2. Summary of Analysis of Variance for Videotape Empathy Series

 

 

 

 

 

 

 

 

 

 

Hypotheses Source df MS F p w2
I Perceivers (P) 2 .000 <1 n.s
II Judgers (J) 2 .007 <1 n.s
I, II P x J 4 .007 <1 n.s

Subjects (S) : P x J 27 .026

Measures (M) 3 .316 33.79 .0001 .10
I P x M 6 .006 <1 n.s.
II J x M 6 .025 2.67 .021 .01
I, II P x J x M 12 .009 <1 n.s.

S x M : P x J 81 .009

Perceiver Targets (X) 1 .034 1.75 n.s
I, III P x X 2 .018 <1 n.s
II, IV J x 2 .039 1.98 n.s
I,II,III,IV P x J x X 4 .007 <1 n.s

S x X : P x J 27 .009

M x X 3 .139 11.09 .0001 .04
I, III P x M x X 6 .010 <1 n.s.
II, IV J x M x X 6 .035 2.76 .017 .01
I,II,III,IV P x J x M x X 12 .006 <1 n.s.

S x M x X : P x J 81 .013

Judger Targets (Y) 1 .000 <1 n.s
I, III P x Y 2 .003 <1 n.s
II, IV J x Y 2 .003 <1 n.s
I,II,III,IV P x J x Y 4 .011 <1 n.s

S x Y P x J 27 .015

M x Y 3 .126 10.99 . 001 .04
I, III P x M x Y 6 .007 <1 n.s.
II, IV J x M x Y 6 .013 1.17 n.s.
I,II,III,IV P x J x M x Y 12 .009 <1 n.s.

S x M x Y P x J 81 .011

X x Y 1 .006 <1 n.s
I, III P x X x Y 2 .001 <1 n.s
II, IV J x X x Y 2 .001 <1 n.s
I,II,III,IV P x J x X x Y 4 .015 <1 n.s

S x X x Y P x J 27 .015

M x X x Y 3 .062 4.44 .006 .02
I, III P x M x X x Y 6 .004 <1 n.s.
11, IV J x M x X x Y 6 .017 1.26 n.s.
I,II,III,IV P x J x M x X x Y 12 .014 1.04 n.s.

S x M x X x Y ' P x J 81 .014

Total 575 .016

 

62

Table 3. Summary of Analysis of Variance for Personality Empathy Test

 

 

 

 

 

 

 

 

 

 

Hypotheses Source df MS F p w2
I Perceivers (P) 2 .006 <1 n.s
II Judgers (J) 2 .056 1.27 n.s
I, II P x J 4 .044 <1 n.s

Subjects (S) : P x J 27 .044

Measures 3 .293 12.41 . 001 .05
I P X M 6 .025 1.05 n.s.
II J x M 6 .026 1.09 n.s.
I, II P x J x M 12 .030 1.28 n.s.

S x M : P x J 81 .024

Perceiver Targets (X) l .804 24.35 . 001 .05
I, III P X X 2 .032 <1 n.s.
II, IV J X 2 .053 1.61 n.s.
I,II,III,IV P x J X X 4 .018 <1 n.s.

S x X ' P x J 27 .033

M x X 3 .082 3.64 .016 .01
I, III P X M x X 6 .019 <1 n.s.
II, IV J X M x X 6 .041 1.80 n.s.
I,II,III,IV P x J x M x X 12 .020 <1 n.s.

S x M x X : P x J 81 .023

Judger Targets (Y) 1 .090 6.09 .020 .01
I, III P X Y 2 .013 <l n.s.
II, IV J X Y 2 .049 3.36 .05,n.s.
I,II,III,IV P x J X Y 4 .024 1.61 n.s.

S x Y P X J 27 .015

M x Y 3 .137 7.80 . 001 .03
I, III P x M x Y 6 .005 <l n.s.
II, IV J x M X Y 6 .008 <1 n.s.
I,II,III,IV P X J X M x Y 12 .006 <1 n.s.

S x M x Y P x J 81 .018

X x Y 1 .554 22.83 . 001 .03
1,111 P X X x Y 2 .023 <1 n.s.
II, IV J x X X Y 2 .020 <1 n.s.
I,II,III,IV P X J x X x Y 4 .007 <1 n.s.

S X X x Y P x J 27 .024

M x X x Y 3 .299 19.00 .0001 .05
I, III P X M x X X Y 6 .020 1.26 n.s.
II, IV J X M X X X Y 6 .044 2.80 .016 .01
I,II,III,IV P x J x M x X x Y 12 .025 1.57 n.s.

S x M X X X Y ' P X J 81 .016

Total 575 .029

 

63

indication that the measures do in fact differ, however, an artifactual
difference is probably truer. Specifically, differences in scoring

the scales and the difficulty of the measures probably resulted in the
differences in the percentage of items correct.

In Table 3, 9 of 31 analyses, or 27%, are significant at the
prescribed .025 level. This again far exceeds the 1% chance occurrence
predicted. Three components stand out in the significant analyses of
Table 3: Measures (M), Perceiver Targets (X), and Judger Targets (Y).

A similar artifactual explanation seems relevant for the measures com—
ponent. Significant Perceiver-target and Judger—target analyses indi-
cate that all the videotaped group members (targets) were perceived
differently. The least this data indicates is that all the targets

were not grouped together in one category. But normal expectation

would produce the results that, in fact, personality differences emerged.
Only inspection of the data with reference to HYPOTHESIS VII, which
compares the two individuals representing each target dimension, can
indicate whether additional significant differences between target~pairs
were evident. In this case, significant differences would not be desire
able as they would indicate that individual differences would be stronger
than common personality dimension bonds. This issue will be further
discussed with reference to Hypothesis VII.

Looking now at Hypothesis I, results can be clearly viewed in
'hables 4 and 5 which collapse relevant data from Tables 2 and 3. With
rwaspect to differentiation of all Perceivers across all measures and
tat~gets (P), no significant differences are observed on interactional

anéilyses in either table. Thus, Hypothesis I was not confirmed:

Intzuiters were not most empathic among the Perceivers.

64

Table 4. Analyses of Variance from Table 2 with Special Relevance to
Hypothesis I

 

 

 

Source df MS F p

P 2 .000 <1 n.s
P x J 4 .007 <1 n.s
P x M 6 .006 <1 n.s
P x J x M 12 .009 <1 n.s
P x X 2 .018 <1 n.s
P x J x X 4 .007 <1 n.s
P x M x X 6 .010 <1 n.s
P x J x M x X 12 .006 <1 n.s
P x Y 2 .009 <1 n.s
P x J x Y 4 .011 <1 n.s
P x M x Y 6 .007 <1 n.s
P x J x M x Y 12 .009 <1 n.s
P x X x Y 2 .001 <1 n.s
P x J x X x Y 4 .015 <1 n.s
P x M x X x Y 6 .004 <1 n.s
P x J x M x X x Y 12 .014 1.04 n.s

 

Table 5. Analyses of Variance from Table 3 with Special Relevance to
Hypothesis I

 

 

 

Source df MS F p

P 2 .006 <1 n.s.
P x J 4 .044 <1 n.s.
P x M 6 .025 1.05 n.s.
P x J x M 12 .030 1.28 n.s.
P x X 2 .032 <1 n.s.
P x J x X 4 .018 <1 n.s.
P x M x X 6 .019 <1 n.s.
P x J x M x X 12 .020 <1 n.s.
P x Y 2 .013 <1 n.s.
P x J x Y 4 .024 1.61 n.s.
P x M x Y 6 .005 <1 n.s.
P x J x M x Y 12 .006 <1 n.s.
P x X x Y 2 .023 <1 n.s.
P x J x X x Y 4 .007 <1 n.s.
P x M x X x Y 6 .020 1.26 n.s.
P x J x M x X x Y 12 .025 1.57 n.s.

 

65

Regarding Hypothesis II, scant results supporting the hypothesis
can be found on both Tables 6 and 7, where relevant variables have
been collapsed from Tables 2 and 3 respectively. Checking, first, the
most global indicator (J), no significant differences are found on
either Table 6 or 7. Interactional analyses in Table 6 show 2 of 15
analyses significant at the .025 level: Judgers x Measures (JxM) and
Judgers x Measures x Perceiver Targets (JxMxX). These interactions,
however, only account for an estimated 2% of the variance from Table 2
(w2 = 1% each for J x M and J x M x X). Table 8 further shows two
factors mainly accounting for the significant ‘J x M interaction.
First, the Behavioral Empathy measure has lower means and the Social
Interaction Empathy measure has higher means than the two relatively
equally scoring Feeling and Thinking empathy measures. As previously
suggested, this is probably due to artifactual scoring and difficulty
inequities between the measures. Secondly, with respect to Behavioral
Empathy, the Thinkers score much lower than the other two orientations.
A likely explanation for this is that since the B.E. means are only
slightly above chance expectation (.25 for four alternatives per ques-
tion), intrinsic measurement error relates to this effect.

With respect to Table 9, two factors seem to account for the sig—
nificant J x M x X interaction. Again artifactual measurement differ-
ences described above are evident. And secondly, larger differences are
found between empathic accuracy toward Intuiters and Sensers on Behavior
and Social Interaction Empathy than are found on the other measures,
Sensers attracting consistently more accuracy on SIE. The greater B.E.

differences again may be attributed to measurement error, but the SIE

66

Analyses of Variance from Table 2 with Special Relevance to

Hypothesis II

Table 6.

 

 

MS

df

Source

 

.1
0

.01

.5 5 .........
SSZSSSZSSSSSSSSS

0 ..0 .........
nnononn.

0000000000000000

2462246224622462
.1 .1 .1 1|.

VI

X

X Y YYX

X X XXX

M XXM YYMYXXM

X XXX XXXXXXX
JMJXJMJYJMJXJMJ
XXXXXXXXXXXXXXX
JPJPJPJPJPJPJPJP

 

Table 7.

Analyses of Variance from Table 3 with Special Relevance to

HypotheSis II

 

 

MS

df

Source

 

1
0

.
s
........ n.....5
SSSSSSSS’SSSSSZS
........ 5..oo.0
nnnnnnnnonnnnn.n

8.1 O 6.1
26 00 3r0. 85
111.111.13.1..1.1.1.12.Ino

oooooooooooooooo

24.102246224622462
.1 1| .1 .1

V:
X

VA VI VIVIVA

X X XXX

M VAVAM VnVIMVIVAVAM

X XXX XXXXXXX
JMJXJMJYJMJXJMJ
XXXXXXXXXXXXXXX
JPJPJPJPJPJPJPJP

 

Table 8.

to Empathy Measured via a Videotape Source

Means from Judgers-Measures Interaction (J x M) with Relation

 

 

 

 

 

 

 

 

 

 

 

 

F.E.* T.E.* B.E.* S.I.E.*
Feeling .39 .39 .34 .41
Thinking—Feeling .39 .36 .33 .44
Thinking .38 .39 .28 .43
*F.E. = Feeling Empathy
T.E. = Thinking Empathy
B.E. = Behavior Empathy
S.I.E. = Social Interaction Empathy
Table 9. Means from Judgers-Measures-Perceiver Targets Interaction
J x M x X) with Relation to Empathy Measured via a Videotape
Source
F.E. T.E. B.E. S.I.E.
N S N S N S N S
Feeling .38 .40 .38 .40 .37 .32 .35 .49
Thinking-Feeling .39 .39 .34 .37 .40 .25 .42 .47
Thinking .39 .38 .40 .38 .26 .30 .38 .49

 

 

 

 

 

 

 

68

effect is puzzling. One possibility is that Feeling, Thinking and
Behavior Empathy are all directed to specific situational events—-a
target's behavior at a moment in time, whereas Social Interaction
Empathy is determined by more global assessments over the entire half-
hour session. It may be that some targets are easier to understand

than others when viewed overall, even though their specific behaviors
are difficult to interpret. As a whole, then, the significance of 2

of 12 analyses related to Hypothesis IV on Table 6, since they repre-
sent only 2% of the variance from Table 2, are not considered sufficient
to support confirmation of the hypothesis.

Likewise viewing Table 7, one of 15 interactional analyses is sig-
nificant at the prescribed .025 level. An estimated 1% of the variance
from Table 3 is accounted for by this significant analysis. Table 10
gives the means for this significant Judgers x Measures x Perceiver
Targets x Judger Targets (J x M x X x Y) interaction. It will be noted
that the most discrepant means lie within the first two measures:
understanding the Thinking-Feeling orientation of targets (T-F) and
understanding the Intuition—Sensing orientation of the targets (N-S).
Thinker subjects seem to understand targets on these two measures some-
what better than the other two groups, especially on the Intuition—
Sensing dimension. These results tend to directly disconfirm Hypothe-
sis II: Feelers are not more empathic. These results are consistent,
however, with the earlier suggestion that written empathy measures are

more likely to stress a logic component than are observational measures.

69

 

Hm. mm. mm. m¢. we. we. Ho.

 

ow. mo. Hm. mm. mm. HH. Hm.

 

932::

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mv. mm.
mm. on. NH. Hm. mm. ov. mm. mm. Hm. mo. No. we. Hm. mm. mm. mm. mcwwaJlmcchng
Hm. mm. mm. mu. mm. Mm. Ho. we. mm. mm. mm. mm. mm. Hm. mm. om. mcw_mmJ
H J H J H J H J H J H J H J H J
m z m z m z m z
u JIH muz JIH

 

 

 

wucsom pmoH HpHHmCOmeJ mw> vwgzmmmz HcpmJEJ 0p Coprme cpwz
A> x x x z x av cowpumcmch mpmeaH LomuzalmpmmLmH Lm>mmongummcsmmmzimcmmvzn EOLJ «com: .o_ mHan

 

70

Differential Effects

Three hypotheses assess the differential effects. Hypothesis III
states that similarity of Perceiver orientation between subjects and
targets will facilitate empathy. Similarly, Hypothesis IV states that
similarity of Judger orientation between subjects and targets will
facilitate empathy. Hypothesis V states that among Judgers, subjects
with Feeling orientations will score higher on Feeling measures of
empathy, while subjects with Thinking orientations will score higher on
Thinking measures of empathy. Tables 11 and 12 collapse relevant data
concerning Hypothesis III from Tables 2 and 3. And Tables 13 and 14
collapse data concerning Hypothesis IV from Tables 2 and 3. Table 15

relates to Hypothesis V.

Table 11. Analyses of Variance from Table 2 with Special Relevance to
Hypothesis III

 

 

 

Source df MS F p

P x X 2 .018 <1 n.s.
P x J x X 4 .007 <1 n.s.
P x M x X 6 .010 <1 n.s.
P x J x M x X 12 .006 <1 n.s.
P x Y 2 .003 <1 n.s.
P x J x Y 4 .011 <1 n.s.
P x M x Y 6 .007 <1 n.s.
P x J x M x Y 12 .009 <1 n.s.
P x X x Y 2 .001 <1 n.s.
P x J x X x Y 4 .015 <1 n.s.
P x M x X x Y 6 .004 <1 n.s.
P x J x M x X x Y 12 .014 1.04 n.s.

 

71

Table 12. Analyses of Variance from Table 3 with Special Relevance to
Hypothesis III

 

 

 

Source df MS F p

P x X 2 .032 <1 n.s.
P x J x X 4 .018 <1 n.s.
P x M x X 6 .019 <1 n.s.
P x J x M x X 12 .020 <1 n.s.
P x X x Y 2 .023 <1 n.s.
P x J x X x Y 4 .007 <1 n.s.
P X M x X x Y 6 .020 1.26 n.s.
P x J x M x X x Y 12 .025 1.57 n.s.

 

 

Inspection of Tables 11 and 12 shows no significant differences
on any analyses related to Hypothesis III. Thus, subjects did not have
greater empathic accuracy toward targets with similar perceiver orienta-

tions.

Table 13. Analyses of Variance from Table 2 with Special Relevance to
Hypothesis IV

 

 

 

Source df MS F p
J x Y 2 .003 <1 n.s
P x J x Y 4 .011 <1 n.s
J x M x Y 6 .013 1.17 n.s
P x J x M x Y 12 .009 <1 n.s
J x X x Y 2 .001 <1 n.s
P x J x X x Y 4 .015 <1 n.s
J x M x X x Y 6 .018 1.26 n.s
P x J x M x X x Y 12 .014 1.04 n.s

 

72

Table 14. Analyses of Variance from Table 3 with Special Relevance to
Hypothesis IV

 

 

 

Source df MS F .p . w2
J x Y 2 .049 3.36 .05, n.s

P x J x Y 4 .024 1.61 n.s.

J x M x Y 6 .008 <1 n.s.

P x J x M x Y 12 .006 <l n.s.

J x X x Y 2 .020 <1 n.s.

P x J x X X Y 4 .007 <l n.s.

J x M x X x Y 6 .004 2.80 .025 .01
P x J x M x X x Y 12 .025 1.57 n.s.

 

 

With respect to Hypothesis IV, Table 13 shows none of eight inter—
actions significant, while Table 14 shows one of eight significant at
the required .025 level. The one significant analysis accounts for an
estimated 1% of the variance from Table 3. Means for the significant
Judgers x Measures X Perceiver Targets X Judger Targets interaction have
been presented in Table 10. Perusal of the means shows that Thinker
subjects outscore Feeler subjects when interpreting Thinker targets
7 out of 8 times. However, Thinker subjects also outscore Feeler sub—
jects 5 out of 8 times when interpreting Feeler Targets. Empathic
accuracy toward Senser-Thinker targets is also consistently lower than
toward other subjects. This finding points to differences in clarity
regarding presentation of target characteristics and will be discussed
further under Hypothesis VII. Differences with respect to better per-
formance by Thinkers, then, partially support Hypothesis IV, but in
light of other non—significant results in Table 13, are not sufficient
to support Hypothesis IV that judgers best understand targets that are

similar to them.

73

Table 15. Analyses of Variance from Table 2 with Special Relevance to
Hypothesis V

 

 

 

Source df MS F p w2
P x M 6 .006 <1 n.s.
J x M 6 .025 2.67 .021 .01
P x J x M 12 .009 <1 n.s.
P x M x X 6 .010 <1 n.s.
J x M x X 6 .035 2.76 .017 .01
P X J x M x X 12 .006 <1 n.s.
P x M x Y 6 .007 <1 n.s.
J x M x Y 6 .013 1.17 n.s.
P x J x M x Y 12 .009 <1 n.s.
P x M x X x Y 6 .004 <1 n.s.
J x M x X x Y 6 .018 1.26 n.s.
P x J x M x X x Y 12 .014 1.04 n.s.

 

Hypothesis V states that feeling-oriented subjects will do better
on feeling-oriented measures of empathy, while thinking—oriented sub—
jects will do better on thinking-oriented measures via the videotape
source. Table 15 collapses the analyses from Table 2 relevant to this
hypothesis. It can be noted that 2 of 12 analyses are significant.

The means for the Judgers x Measures interaction have been presented in
Table 8 and the means for Judgers x Measures x Perceiver Targets inter«
action have been presented in Table 9. As was noted previously the
variation in measures on these two tables are primarily related to the
Behavioral Empathy and Social Interaction Empathy measures, not the
measures relevant to this hypothesis. Feeling and thinking oriented
individuals do not differ appreciably on the thinking and feeling
empathy measures. Thus these significant interactions do not support

Hypothesis V. In conjunction with the other non-significant analyses,

 

74

Hypothesis V is not confirmed. Thinkers do not do better than Feelers

on thinking measures, nor is the converse true.

Measurement Effects

Three hypotheses are related to measurement effects. Hypothesis
VI tests the effects of order of presentation, specifically whether
empathic accuracy is increased as the research progresses. Hypothesis
VII assesses whether the two target individuals representing each
target dimension drew similar empathic accuracy scores from the sub—
jects, as would be predicted if the target characteristics were salient.
Hypothesis VIII tests whether the method used to measure empathy,
specifically via videotape or personality test source, provides similar
accuracy scores. The null hypothesis is predicted.

Hypothesis V1 is based on the analysis described in Table 16.
Looking first at the overall results of Table 16, it is evident that
2 of 7 analyses, or 29%, are significant at the prescribed .0167 level.
This exceeds the less than 1% chance expected overall at this level.
The two significant analyses point to differences in responses to film
order (F) and to the measures within films as the sessions progressed
(M : F x P). The lack of significance when the group (G) factor was
present indicates that, although the first group of subjects viewed one
videotaped group first, while the second group of subjects viewed the
other videotaped group first, it was the order of the films irrespective
of the specific film involved which created the significant variance.
The almost significant presentation effect (P) supports the importance

of ordering.

 

 

 

 

 

 

omo. _mF_ _mp0H
mmo. o_m w x a x a n m x z
.m.: _ mwo. eN a x a u a x z H>
__. _ooo. km.k Nw_. em a x a ” sz maasmma: H>
nu 0N0. eoN g x a u m x a
.m.: .oeo. mN.~ mmo. o a n w x a H>
_o. FPO. mm.N mac. 0 Adv E_wa H>
w_o. em a u a x m
.m.: _ _oo. F a x a H>
.m.c .m_o. mF.o e_F. _ Aav cowpapcamaaa H>
eeo. em w ” Amv mpuaflnsm
.m.: _ moo. _ Amy mazoaw H>
N3 a m m: we mogzom mummgpoamz
mpuwmmm cowpmpcwmwgm we Lmvgo Low mwmz—m:< we zgmaazm .o_ wpnmh

76

Hypothesis VI predicts that empathic accuracy will not increase
over testing sessions. As explained in the rationale for the hypothe—
ses, consistent differences could arise from at least two primary
sources: (1) learning related to experimental practice and (2) perform-
ance increments related to increased information and/or strategy
testing. If the first were true, accuracy would increase constantly
according to presentation (P) and film-order (F). If the second source
held, accuracy would increase from Film 1 to 4, then decrease as the
new Film 1 is introduced, repeating the pattern of increase from the
first film series. Thus F, but not P, would be significant. These
trends may be detected from Tables 17 and 18 which show the means of the
significant analyses from Table 16.

From Table 17 it can be observed that neither trend is clear.
Presentation I shows instability over testing, while Presenataion II
shows an increase in accuracy. It is unclear whether the inconsistency
in Presentation I is related more to lack of accommodation to the experi-
mental procedure, to failure in trial and error learning or developing
adequate strategies, or to the fact that empathy is primarily situation—
al and not substantially increased over short intervals. Due to lack of
consistency over Presentation 1, it is difficult to assess whether a
meaningful decrement occurred between Film 4 of Presentation I and Film
1 of Presentation II. Therefore, learning related to increased ability
or to increased information is impossible to assess from these data.

Table 18 provides more information with respect to the measures
influence. 0n the Feeling Empathy measure, no increase in accuracy was

shown between Presentation I and II, and a slightly decreasing trend was

77

Table 17. Means from Films (F) Analysis of Variance Reported in Table
16

 

 

 

 

Presentation I Presentation II
F1 F2 F3 F4 F1 F2 F3 F4
Subjects .37 .39 .34 .37 .36 .36 .42 .41

 

 

 

 

Table 18. Means from the Presentation x Films x Measures (M : F x P)
Interaction Reported in Table 16

 

 

 

 

 

 

Presentation I Presentation II
F.E. T.E. B.E. S.E. F.E° T.E. B.E. S.E.
Film 1 .45 .42 - .25 .37 .37 .43 .25 .41
Film 2 .40 .32 .38 .45 .40 .34 .31 .40
Film 3 .34 .29 .29 .45 .34 .45 .36 .51
Film 4 .43 .40 .31 .33 .36 .41 .32 .53

 

 

 

 

 

 

 

 

 

 

found within each presentation. Thus, it seems unlikely that either
increased empathic skill or adequate informational predictors were
gained with respect to feeling empathy. It should be remembered that
much feeling recognition is situation specific, although, of course,
some dispositional behavior does exist. With respect to Thinking
Empathy, some increase does exist in Presentation 11 over Presentation
1, although dips in accuracy occur in the middle of both presentations.

Thus, some empathic gains due to learning or practice, although not

 

 

 

78

consistent, are evident. Behavior Empathy shows a pattern of initial
chance responding with slight increments as the film progresses.
Empathic carry over seems to be negligible, probably due to the complex
nature of open-ended predictions of behavior and because of the great
emphasis on situation versus disposition in answering accurately.

Thus, learning Behavior Empathy seems minimal. Finally, empathy
related to understanding social interactions increases from presenta-
tion to presentation and within presentations to a degree. It is un-
clear whether the large decrement in Film 4 of Presentation I should be
attributed to faulty strategy changing, or whether the apparent gains
evident on Films 3 and 4 are illusory.

Overall, then, some measurement-specific effects can be noted
across films and presentations. The near significant differences on
the presentation dimension (P) coupled with the significant differences
in the films within presentations, and measures within films and pre—
sentations leads to a possible rejection of Hypothesis VI. However,
while films on Presentation II showed fairly consistent increases on
Thinking Empathy and Social Interaction Empathy, scores hardly improved
on Feeling Empathy and Behavior Empathy. In addition when looking from
Film 1 to Film 4, no constant increases are observed, although Social
Interaction Empathy in Presentation 11 is almost consistently gaining.
With regard to accepting or rejecting Hypothesis VI, then, a condi-
tional rejection seems most reasonable. While empathy does not increase
over sessions on Feeling and Behavior Empathy, inconsistent empathic
increases are observed on Thinking and Social Interaction Empathy.

Thus, on two measures either learning or measures-adaptation increases

79

are likely and measurement of a natural responding pattern has not been
achieved.

Hypothesis VII tests the effects of utilizing two target repre-
sentatives for each target dimension. It is hypothesized that no sig-
nificant differences will occur between the two representatives (A1 and
A2). Table 19 gives a summary of the analysis of variance results

relating to this hypothesis. As has been true, measurement variance is

again prominent (M, M x X, M x Y, M x X x Y). With regard to the

 

critical A variable, however, a complicated picture is presented. Five
of eight analyses relevant to Hypothesis VII are significant at the
required .0167 level. However, cumulatively these significant analyses
are estimated to account for only 5% of the variance in Table 18.
The interaction most sensitive to the complicated effects of the A di-
mension is the M x X x Y x A interaction, and the means of this inter-
action are listed and graphed in Table 20 and Figure 9 respectively.
While similar patterns of mean responding toward the targets can
be seen for each measure, gross differences between the means can be
seen toward NT targets on Feeling Empathy, ST targets on Behavior
Empathy, and SF targets on Social-interaction Empathy. Thus, gross dif-
ferences toward alternative targets seem evenly distributed. Further
observations of Figure 9 would seem to indicate that while mean differ-
ences between A's are significant at the required level, similar pat-
terns seem to exist to responding to A's, e.g., on T.E., SF targets
draw most empathic accuracy, followed by NF, NT and ST targets in that
order. While not totally consistent, these patterns do seem substantial.

In addition it may be noted that in 2 target pairs (NF and ST), one

80

 

 

 

Table 19. Summary of Analysis of Variance for Congruence of Individual
Target Representatives on Videotape Empathy

Hypotheses Source MS df p w2
Subjects (S) .040 35
Measures (M) .682 3 41.15 .0001 .06
S x M .017 105
Perceiver Targets (X) .055 l 1.60 n.s.
S x X .035 35
M x X .274 3 10.30 .0001 .02
S x M x X .027 105
Judger Targets (Y) .007 1 <1 n.s
S x Y .024 35
M x Y .251 3 11.19 .0001 .02
S x M x Y .022 105
X x Y .019 1 <1 n.s
S x X x Y .023 35
M x X x Y .133 3 5.04 .003 .01
S x M x X x Y .026 105

VII Alter Targets (A) .168 1 10.20 .003 .01
S x A .016 35

VII M x A .149 3 5.44 .002 .01
S x M x A .027 105

VII X x A .334 1 10.37 .003 .01
S x X x A .032 35

VII M x X x A .013 3 <1 n.s.
S x M x X A .030 105

VII Y x A .079 1 3.35 .076, n.s.
S x Y x A .024 35

VII M x Y x A .015 3 <1 n.s
S x M x Y A .022 105

VII X x Y x A .293 l 10.81 .002 .01
S x X x Y A .027 35

VII M x X x Y A .090 3 3.64 .015 .01
S x M x X Y x A .025 105
Total .030 1151

 

81

Table 20. Means from the Measures x Perceiver Targets x Judger Targets
x Alter Targets Interaction (M x X x Y x A) reported in
Table 18

 

 

NF NT SF ST

 

Feeling Empathy .37 .40 .48 .30 .36 .32 .42 .46
Thinking Empathy .36 .41 .36 .37 .39 .45 .32 .39
Behavior Empathy .28 .31 .42 .36 .29 .29 .23 .33

Social-inter- .39 .44 .35 .35 .44 .58 .41 .49
action Empathy

 

representative consistently scored higher than the other. This may
indicate greater transparency of feelings and thoughts by some target
representatives than others. With regard to the rationale behind
assessing Hypothesis VII, then, that differences in A would not create
enough variance to confound the determination of main effects, it seems
best to conditionally reject Hypothesis VII on the basis of five out of
eight relevant analyses attaining significance. However, because tar-
get pairs did often seem to attract similar accuracy patterns, consider-
able reservation must be attached to the rejection.

Hypothesis VIII predicts that differences will occur between
videotape and personality test empathy. It is evident from Table 21
that 5 of 31, or 16%, of the analyses are significant at the prescribed
.0167 level. Again, this exceeds the 1% chance of significance ex-
pected when this many analyses are conducted requiring this level of

significance. It will be noted that only analyses devoid of the

 

60
58
56
54
52
50
48
46
44
42
4o
38
36
34
32
3o
28
26
24
22

 

 

 

82

 

 

 

 

 

\ l
\ I
\“ .
\‘II
NF NT SF ST NF NT SF ST NF NT SF ST NF NT SF ST
Feeling E. Thinking E Behavior E. Social-int. E.
Key: —»——~——-= Target Group A
- — — - = Target Group B
Figure 9. Distribution of Means from Measures X Perceiver

Targets x Judger Targets x Alter Targets Interaction
(M x X x Y x A) reported in Table 19.

 

83

Table 21. Summary of Analysis of Variance for Personality versus Video-
tape Empathy Measures

 

 

 

Hypotheses Source df MS F p w
Perceivers (P) 2 .003 <1 n.s.
Judgers (J) 2 .007 <1 n.s.
P x J 4 .009 1.11 n.s.
Subjects (S) : P x J 27 .008

 

 

 

 

 

 

 

VIII Form of Test (F) 1 3.752 382.61 .0001 .65
VIII P x F 2 .000 <1 n.s.
VIII J x F 2 .007 <1 n.s.
VIII P x J x F 4 .008 <1 n.s.
S x F : P x J 27 .010
Perceiver Targets (X) l .022 2.99 n.s.
P x X 2 .009 1.19 n.s.
J x X 2 .007 <1 n.s.
P x J x X 4 .007 <1 n.s.
S x X P x J 27 .007
VIII F x X 1 .124 14.08 .001 .02
VIII P x F x X 2 .012 1.34 n.s.
VIII J x F x X 2 .017 1.90 n.s.
VIII P x J x F x X 4 .001 <l n.s.
S x F x X : P x J 27 .009
Judger Targets (Y) 1 .035 12.84 .002 .01
P x Y 2 .005 1.60 n.s.
J x Y 2 .008 2 99 n.s.
P x J x Y 4 .003 1 26 n s
S x Y . P x J 27 .003
VIII F x Y 1 .010 3.35 n.s
VIII P x F x Y 2 .001 <1 n.s
VIII J x F x Y 2 .003 l 15 n.s
VIII P x J x F x Y 4 .004 1.39 n.s
S x F x Y P x J 27 .003
X x Y 1 .074 26.36 . 001 .01
P x X x Y 2 .001 <1 n.s.
J x X x Y 2 .003 1.07 n.s.
P x J x X x Y 4 .004 1.41 n.s.
S x X x Y P x J 27 .003
VIII F x X x Y 1 .070 9.50 .005 .01
VIII P x F x X x Y 2 .005 <1 n.s.
VIII J x F x X x Y 2 .001 <1 n.s.
VIII P x J x F x X x Y 4 .001 <1 n.s.
S x F x X x Y P x J 27 .007
Total 287 .020

 

 

 

 

84

Perceivers or Judgers factors have maintained significance. Although
this trend has been evident throughout the study, it reaffirms that
neither method was adequate for finer empathic discrimination. Part

of this is related to low level discrimination toward Perceiver targets
and Judger targets-~only 50% of the time are the differences significant
(F x X and Y, but not X and F x Y). Part also related to measurement
error.

While the mean difference between Forms (F) appears highly sig-
nificant, it should be recalled that .50 is the chance expectation for
the two—alternative Personality Empathy Test, while .25 is the chance
expectation for the four—alternative Videotape Empathy Series. Thus,
neither form shows high accuracy when chance expectations are accounted
for. And the F difference must be considered illusory.

Closer examination of the means of the most discriminating signifi-

cant interaction in Table 21, F x X x Y, can be seen in Table 22.

Table 22. Means from the Form x Perceiver Targets x Judger Targets
Interaction Reported from Table 21

 

 

Videotape Test |[ Personality Test

 

NF NT SF ST H NF NT SF ST

 

 

 

 

 

 

 

 

 

Subjects .38 .37 .41 .40 H .63 .66 .63 .54

 

Two factors stand out. Dissimilar accuracy patterns exist across
measures, subjects scoring best toward Sensers on the V.E.S., but worst

toward ST's on the P.E.T. And, while V.E.S. scores are uniform, only 4

85

percentage points difference existing between high and low means, one
P.E.T. target score is considerably distant from the others. The first
finding supports the contention that empathy as variously measured is
not the same trait, some targets drawing higher empathy on some measures
than on others. The second finding may point to response-set differ-
ences between the measures. The P.E.T., being more simplified and
therefore strategy encouraging, results in highly inter-dependent

scores yielding gross differences, while the V.E.S.‘s complexity and
emphasis on specific behaviors is strategy discouraging and does not

show the effect of inter—dependence.

Post-experiment Data

Since empathy was explored in a straightforward manner in this
research, and since enhancing motivation to be empathic seemed consist-
ent with the design, the subjects were told before they participated in
the research that they would be told the results of their efforts upon
completion of the total project. They were also told that at this
time they were free to discuss their opinions with others. (This
served as a reasonable and predictable time-limit for inter—subject
silence.) After the results were communicated and discussed, subjects
were asked ”to help the experimenter determine why the results came
out as they did‘I by writing any ideas they might have on pieces of
paper and handing them in. A list of the ideas submitted by the subjects
are presented in Appendix F.

The comments centered around three main issues: (1) problems in-
volving the quality of videotaped information, especially the defensive-

ness of the targets, (2) problems experienced by subjects in uniformly

86

attending to the videotape, and (3) problems involving difficulty of
the test items. It should be noted that subjects made their comments
after receiving their individual and total group scores. Since the
scores were universally low, subjects' comments may have unduly re-
flected problems they remembered encountering rather than instances
where they believed they were correct. Further, comments reflected
sources consciously attributed to difficulties, not sources, such as

dependent variable differences, not recognized by the subjects.

 

Summary of the Results

Main Effects»

Hypothesis 1: With regard to Perception, subjects with high

Intuitive orientation will show more empathic accuracy than

subjects with middle Intuitive-Sensing or Sensing orientations.

Hypothesis II: With regard to Judgment, subjects with high

Feeling orientation will show more empathic accuracy than subv

jects with middle FeelingeThinking or Thinking orientations.

No analyses related to Hypothesis I were found to be significant,
therefore clearly this hypothesis was not confirmed. Scant results were
found to support Hypothesis II, Thinking—oriented subjects scoring lower
than others on Behavior Empathy, and all subjects showing much variation
toward Intuiting and Sensing targets on Behavior and Social Interaction
Empathy. These results were interpreted as arising from measurement erre
or with relation to Behavior Empathy since means on this measure were
close to chance prediction. The Social Interaction Empathy variance was
attributed to the grossly larger stimulus configuration (involving all bee
havior over the half-hour session) which had to be synthesized for accuracy

on Social Interaction Empathy, making error more probeble. Thus Hypothesis

II, since no other relevant analyses were significant, was rejected.

87

Differential Effects

 

Hypothesis III: Similarity of Perceiver orientation between sub—
jects and targets will result in more empathic accuracy than be-
tween subjects dissimilar in Perceiver orientation.

Hypothesis IV: Similarity of Judger orientation between subjects

and targets will result in more empathic accuracy than between

subjects dissimilar in Judger orientation.

No analyses related to Hypothesis III were found to be significant,
thus Hypothesis III was rejected. 0f 16 analyses related to Hypothesis
IV, one was significant. The significant interaction showed that while
Thinkers showed more accuracy toward Thinker Targets 7 out of 8 times,
Thinkers also showed more accuracy toward Feeler Targets 5 out of 8
times. A tendency toward scoring less well toward the Senser-Thinker
targets was also noted. Because of the equivocal nature of these re-
sults and the lack of significance of any of the other 15 relevant analy-
ses, Hypothesis IV was rejected.

Hypothesis V: With regard to empathy via a videotape source,

Feeling oriented subjects will be most accurate on a Feeling-

oriented measure, while Thinking-oriented subjects will be most

accurate on a Thinking-oriented measure.

Two of twelve analyses related to HypotheSIS V showed significant
differences among the Judgers across measures. However, no differences
among the Judgers on Thinking or Feeling measures were found. Therefore,
Hypothesis V was rejected.

Measurement Effects

Hypothesis VI: Neither the order in which the target groups were

presented, nor the order in which the films within the target

groups were presented, will result in significant increases as

the research progresses.

Order effects were found to be present most clearly with regard to

measures within films and presentations, presenting a complicated picture.

88

Empathic gains were interpreted in terms of two probable sources: gains
resulting from learning to be empathic in general and gains resulting
from learning to be empathic toward each target, i.e., strategy confirm-
ation. No consistent evidence was found to support learning to be
empathic, while learning to be empathic toward specific targets was
found with regard to Behavior, Social Interaction, and guardedly with
regard to Thinking Empathy. Order effects were found to account for
considerably more variance than did the specific target groups being
watched, i.e., the pattern of responding over time was more prominent
than the pattern of responding to a particular target group.

Hypothesis VII: The two target individuals representing each

Intuiting-Sensing, Thinking-Feeling orientation will draw

similar empathy scores from the subjects.

While differences between the means toward alternative targets
were significant, examples of gross differences occurred in three of the
four target categories (Intuition-Thinking, Sensing—Feeling, and Sensing-
Thinking) across three of four measures (Feeling Empathy, Behavior
Empathy, and Social Interaction Empathy). Thus gross differences seemed
randomized. In addition, although mean differences were significantly
large, similar patterns of responding to alternate targets existed across
measures. The interpretation offered points to similar trends in under-
standing target pairs across measures. Instances where one target
representative scored consistently higher than her target pair were
noted.

Hypothesis VIII: A different pattern of mean responses will exist
on the videotape measures than on the personality test measure.

89

Although personality differences between the two forms of empathy
measurement were found to be significant, these were attributed to
differences in number of item-alternatives (.25 chance expectation for
the videotape empathy test and .50 chance expectation for the personality
empathy test). Scores across targets on the videotape test were less
variant than scores across targets on the personality test, the Senser-
Thinker target on the personality test being considerably closer to
chance than the other scores. This latter occurrence was attributed to
the greater opportunity to use correct or incorrect strategies in assess-
ing target qualities on the personality test. The videotape test was

thought to be more a measure of empathy toward situational behaviors.

VIII. DISCUSSION

Since the results of this research point to significant measure-
ment effects overshadowing and interfering with any possible dependent
variable differences, the format of this discussion will center upon
interpreting the important sources of measurement error with regard to
the results. The major characteristics of the results will be listed
first, then their relationship to the sources of error will be explored.

With regard to the results, five characteristics seem most
salient. (1) There were no significant dependent variable effects.

(2) The percentage of items correctly attained by the subjects across

all items was low. (3) Significant differences existed between the two
representatives of each target dimension. (4) Conditionally significant
order effects were found between the filming sessions. And (5), greater
variance was displayed in the personality test measure with regard to
accurate empathy than in the videotape measure. All the results are seen

as being related to measurement sources.

The Stimulus: The Videotaped Group Discussion

 

Several factors related to the videotaped group discussion could
have caused error: (1) the method of choosing individuals to represent
the target dimensions could have been faulty, (2) the technical presenta«
tion of the videotape may have failed to adequately capture subtle

stimulus behaviors, (3) knowledge of being videotaped could have produced

90

91

confounding effects, (4) the content level of the videotaped group dis-
cussion could have influenced the difficulty of interpretation, or

(5) the differential ability of the targets to express themselves,
besides those related to expected intrinsic target differences, may have
decreased accuracy scores toward some targets.

At least two methodological alternatives existed with regard to
choosing targets: targets could have been chosen at random within each
specified category, or targets could have been carefully screened
within each category. The former would decrease the effects of experi—
menter-created sets in the target. The latter would intentionally reduce
the effects of extraneous "natural" variables in order to increase the
power of the experimental effect. This study employed the former method,
but perhaps to its detriment in the following ways: (1) one Senser—
Thinker target, while scoring strongly Sensing-Thinking, seemed to dis-
play many intuitive behaviors, e.g., a lively sense of humor, especially
plays on words; (2) one Intuiter-Feeler target spoke louder and more
definitively than the others in her group, drawing strong feelings from
other targets and subjects (as communicated privately by the targets and
subjects). The inconsistency of the Senser-Thinker target seems explain-
able when considering a reservation Stricker and Ross (1964) pointed out
about the Sensing-Intuition scale on the Myers-Briggs: it correlated
significantly with non—conformity in their validational studies. It is
felt that the particular Senser—Thinker in question was more repelled by
the non-conformity implications on the Intuition end of the scale than
attracted to the creativeness depicted, thus scoring anti-Intuition or

Sensing. The experimenter now believes that preventing these instances

92

of questionable construct validity and unbalanced target dominance could
have reduced measurement error.

Although efforts were taken to maximize visual acuity and group
feeling on the videotape screen, these efforts were not entirely success-
ful. It remained difficult to tell specifically who each target was
looking at, and when the full group angle was presented, the two targets
on the end could only be viewed in profile.* Since many non-verbal
expressions are subtle, e.g., squinting the eyes of slightly frowning,
the lack of visual clarity toward the targets may have been crucial in
misinterpreting feelings or non-verbal thoughts.

The stimulus situation itself may have produced a powerful set to
respond cautiously. Knowing that others would be watching their behavior,
the targets probably reduced statements related to anger, fear or hurt,
although many came through non-verbally. The experimenter, however,
will never forget the target who, during a recall session, responded to
seeing herself smiling and nodding her head up and down by exclaiming,
”I totally disagreed with everything she had to say." Whether the
target usually masked her disagreement with others, or whether her mask-
ing was heightened by the videotape was impossible to ascertain, but
whichever was true, masking behaviors were not infrequent. This points
to an important difference between one popular form of person perception
research where targets role play stimulus behaviors and the form studied
here where natural masking behaviors are included. Doubtlessly the

latter form of empathy is more difficult, although more realistic.

 

*
No empathy questions were asked about targets when they could
be viewed only in profile.

93

An experimenter should, however, seriously consider the intricacy which
her design can profitably allow. For more global or untested designs,
as this one is, perhaps the role playing format would afford more
initially discriminating data. An alternative would be to extend the
group discussion in the videotape situation until the group members
were relaxed. The final test of an hypothesis, however, should be pre-
pared to deal with the rigor needed to test humans in all their complex
behaviors.

The content of the videotaped group discussions also could have
been influential in the clarity of behavioral information presented.
A11 behavior is not unidimensional, i.e., more than one thought or
feeling can occur simultaneously. Moreover, a hierarchy is likely to
exist concerning the unidimensionality of expression. For instance,
with regard to Berne's conceptualization of levels of interaction,
responding at an intimate level by allowing feelings and thoughts to
freely flow should be close to unidimensional; whereas, responding at
the ritual or pastime level is less ego involving and allows divergent
thoughts to occur simultaneously. While talking about football, one‘s
mind may wander to other things he wants to do, to chores he has to
finish, or places he would rather be, but when one feels hurt or angry,
his mind seldom wanders to more mundane topics. Since, in fact, most
of the videotape content remained at the pastime level, despite the
attempts of the discussion topics provided to elevate them, the likeli-
hood that most target behavior was unidimensional is slim. The added
complexity of the targets saying one thing while thinking another

surely increased difficulty and added to the subject's errors.

 

A

94

Interpreting these real life masking behaviors is one reason empathy
is not more common in most people's behavior.

Finally, while overall differences were evident among the target
pairs, in two of four cases one individual drew consistently higher
accuracy scores than the other. This trend may be related to the rela-
tive openness of the two individuals compared to the target they were
paired with. In addition, information from post-experiment comments
indicate subjects did feel that they understood some targets better
than others. It is difficult, of course, to control for equal open-
ness, but it is also evident that error caused from inflated or de-

flated scores could well have confounded experimental results.

The Response: The Empathy Measures

 

Error related to the empathy measures can arise primarily from
two sources: faulty assumptions about the nature of empathy incorporated
into the measurement procedures and inadequacy of the measurement device
itself to obtain the information sought. Among the faulty assumptions
pertaining to the nature of empathy, the Videotape Empathy Series is
particularly vulnerable in the following areas: (1) it assumes an indi-
vidual is equally empathic over time, i.e , one is not variously tuned
into and tuned out of his environment, any error therefore results from
misinterpretation, not failure to having witnessed the stimulus event;
(2) it assumes an individual can attend to thinking, feeling, behaving
and social interaction cues simultaneously and can move between these
modes with facility; and (3) it assumes only one empathic response

exists.

 

95

Failures in the Videotape Empathy Series qua device can be related
to: (l) the construction of correct and incorrect responses with rela«
tion to possible multiple correct responses and the difficulty of the
items; and (2) the effect of feedback between alternative test responses
and subsequent videotape verification, specifically with respect to
strategy use and its concomitant success or failure.

Assumptions Concerninggthe Videotape Empathy Series

One assumption concerning the nature of empathy underlying the
Videotape Empathy Series which is probably faulty is that subjects are
equally attuned from moment to moment to the entire range of inter-
personal stimuli confronting them. Empathy is probably more likely a
discontinuous process, requiring attending to stimulus behavior
(external attention) and retreating into thinking about what has been
observed (reduced external attention). Empathy concerning externally
attended behavior should be high, while empathy during more internal«
ized periods having less access to data, should be lower. Thus, since
the intervals between questions on the videotape tests are experimenter
controlled, empathy may function on a hit or miss basis, sometimes
coinciding with the attention of the subject, sometimes occurring dur-
ing a period of reduced attention to the external environment. Several
subjects reported this privately after filming sessions. It would
seem, then, that subject-controlled empathy stops would yield more
relevant empathy scores. Problems related to subject controlled stops
do arise, however, and will be discussed later.

Another assumption underlying the Videotape Empathy Series is

that subjects can, in fact, attend to and integrate many types of empathy

96

concurrently. This may be a fallacious assumption. A cognitive overload
may result from dealing with material in Such diverse ways. Personal
coping mechanisms (or defense mechanisms) may then take over. Some sub-
jects may concentrate on one form of measurement, e.g., feeling empathy,
ignoring the others, some subjects may try to attend to all forms with
reduced capacity. Some may try to deal with forms they are least
familiar with to compensate for inadequacy. Some may have increased
anxiety impairing their attention span. Whichever methods subjects
choose, confounding data results. This is particularly relevant to
Hypothesis V where Thinkers and Feelers were hypothesized to do better
on thinking and feeling measures respectively. The lack of significant
results may have largely arisen from diverse solutions to this problem.
A final assumption implicitly maintained by the Videotape Empathy
Series is that only one empathic response exists, i.e., the target
individual was only thinking one thought or feeling one feeling in
response to the question asked. Yet, more probably a complex series of
feelings or thoughts were occurring. For instance, one target may have
been feeling angry_at another group member, but apprehensive about ex—
pressing it, while at the same time feeling hurt that the others did not
seem to be on her side. While some of these thoughts seem to be more
recessive than dominant, if another person asked her if she were feeling
one of the more recessive feelings the target might reply affirmatively.
This happened, in fact, during the construction of the Videotape
Empathy Series items. Items were devised by constructing logical altern-

atives to the feelings not checked as present by each target when she

97

stopped the videotape during recall. When the ”wrong” alternatives
were presented to the targets in the validational session, however,
sometimes targets marked "true” or "somewhat true“ to these alterna-
tives indicating more than one feeling or thought was present. Subjects
tuned into a "minor" feeling or thought when confronted with alterna-
tives lacking her pet choice could very well become confused or anxious,
especially if this occurred often. Operationally, the definition of
empathy neatly skirts the issue of more than one correct response by
defining only the original spontaneous target recall as empathic. Yet,
leaving other acknowledged responses as confounding variables increases
error.

Measurement Error ggg Videotape Empathy Series

 

One primary source of error related to the Videotape Empathy
Series as a measurement device has been discussed above concerning the
number of correct empathic responses possible. While the test purports
one correct and three incorrect alternatives, other correct responses,
although omitted, are possible. This problem is exascerbated by the
10 second focusing period where the subject, having read the qUestion,
firmly sets in his mind what his own answer is. Despite having been
utilized to prevent the logic of alternatives outweighing their stimulus
value, this firm establishment of the felt true—response may be an
actual obstacle if the response is not presented.

Another problem arises concerning the difficulty of the items.
Three types of empathy behaviors may be focused upon: (1) obvious
examples of well represented behaviors, e.g., crying at being hurt,

swearing when angry; (2) subtle examples from peripheral or non-verbal

98

behaviors, e.g., looking down to show non-compliance when disagreeing,

talking faster when anxious; or (3) abstruse examples usually resulting
from lack of overt variability of expression or masking behavior, e.g.,
remaining straightfaced when one's viewpoint has been derogated, smil-

ing when angry.

The Videotape Empathy Series was constructed to deal with all
three levels of difficulty. However, it may have unduly focused on the
latter two levels as a result of a pilot study item analysis. In the
pilot study, alternative selections chosen very frequently or very in-
ferquently were either omitted or altered in an effort to increase the
discrimination index by equalizing the drawing power of the alterna-
tives. Since another pilot study was not conducted (because of time
and expense) to ascertain whether these changes had had their desired
effect, the result may have been to exclude or increase the difficulty
of the obvious items and to make alternative selections on the subtle
and abstruse items more plausible, and thus, more difficult. The low
percentage of items correct across all the tests as well as post—
experiment data from the subjects points to the test being too difficult
to supply enough information concerning the dependent variables. This
may reveal a greater likelihood to error on the side of difficulty with
respect to person perception experiments dealing With natural or near-
natural settings. The opposite may be true with experiments utilizing
role-playing: target behaviors may be exaggerated, and thus, too easy.

Another area of concern with the Videotape Empathy Series is the
possible lack of independence of its items due to individual differences

among subjects regarding the need to verify the accuracy of their

99

responses. Not only did certain subjects report privately that they
spent a good deal of effort checking to see if their answers were cor-
rect, but following Behavior Empathy questions, where subjects could
determine whether their predictions were correct, many subjects dis-
played facial grimaces or smiles indicating they were checking the
accuracy of their predictions.

The subjects verified responses in two ways: checking subsequent
videotape information for confirmation, and analyzing the alternatives
of subsequent test items. A subject confronted with a behavioral
sequence might label it "anger” and mark the "anger” alternative. Then,
as the tape started anew the subject might try to confirm the veracity
of his previous anger answer by looking for anger clues in the target's
new behavior. In addition, he might check subsequent item alternatives
for references to anger still lingering.

Two effects of this verification process were evident: (l) certain
strategies arose from the selective attention paid to verification
material versus other natural stimuli, and (2) the subject failed to
attend to information she might attend to if independence were guaranteed.
Strategies may have had a positive or negative effect. Positively, they
may have reflected the normal process of self—verifying hunches, a
realistic part of the empathic process. Negatively, they may have acted
as interference to normal responding. Two instances occurred where
strategy behavior reflected the natural empathic process: (1) when the
strategy was initially correct and did not interfere with future attend—
ing, and (2) when verification did not include follow—up checking of

subsequent test-alternatives. In the latter case, a non—natural source

100

would provide the verification, since usually one does not have a small
finite set of alternatives to check against when making an empathic
judgment.

Strategy formation became confounding when the experimental situ—
ation interfered with or punished natural strategy responding. This
could have occurred when the subject did not have time to explore her
strategies, or when the subject's strategies were mistaken. In the
first instance, the subject was confronted with two tasks: confirming
her previous response, and attending to information which may have
been relevant to her next response. If she attended to the new informa-
tion or her strategy was correct, no interference resulted. But if she
attended to her strategy and it was wrong, e.g., no anger alternative
existed in the next question, additional negative results could have
occurred. The subject might have become anxious, adding emotional
interference to her poor performance. She might have become confused,
unsure which strategies have worked, which have failed. Or she might
have employed her typical defensive stance: giving up, getting bored,
answering at random, etc. Whatever the case, unpredictable results
would have occurred. Since this study cannot ascertain how many subjects
used strategies, whether successful or unsuccessful, and since a strong
probability exists that strategy use was often punished, a mystery re-
sults as to the confounding nature of this effect.

Measurement Error Related to the Personality
Empathy Test

 

Measurement error with respect to the Personality Empathy Test

originated from two sources: the generality of the items which increased

101

their difficulty and the complexity of the videotape test response which
distracted subjects from attending to material on a level appropriate
to the Personality Test. With respect to the first source, although the
Myers-Briggs' items used in Personality Empathy Test were those the
target indicated “importantly described her,” only those items marked
"true“ by 35% to 65% of the population were included. As explained
earlier in the methods section, this was performed to reduce the effect
of knowing common stereotypes as the source of empathic accuracy. By
eliminating the extremes, however, the empathic task became more diffi-
cult and the percentage of correct responses was reduced. While knowl-
edge of stereotypes is not synonymous with empathy, it may be a valid
part of it. Those who can use their knowledge of stereotypes to under—
stand others should, in fact, be more empathic than those who cannot.
Thus, an important source of empathic accuracy may have been eliminated
when removing items commonly answered “true" or “false” by more than
65% of the population.

The second source of error related to the Personality Empathy
Test derived from primary emphasis of the project being placed on the
videotape tests, the personality test being administered on the final
day of each group meeting after all videotape testing procedures were
finished. Since the nature of the videotape test questions was
specific, i.e , asking for an interpretation of a particular behavioral
sequence, and since the nature of the questions on the personality test
was general, subjects may have paid less overall attention to behaviors
described in the personality test. Moreover, this is a significant loss,

since the personality test relies heavily on the subjects ability to

102

integrate behavioral information, the part of the empathic process sig-
nificantly underplayed in the videotape tests.

Selective Videotape Empathy Series versus

Personality Empathy Test Error

When considering both measuring techniques, it is clear that dif-
fering measurement error arises from each source and handicaps each in
terms of failing to produce scores well above chance. The Videotape
Empathy Series by its complexity and lack of flexibility for the sub-
ject, i.e., by requiring diversified responses at short externally-
imposed intervals,increases error if and when the subject cannot readily
respond. The subject does not have time to reconstruct previous strate—
gies nor check out current alternatives without missing information
likely to be valuable to the next response. Increased anxiety is often
the result.

The Personality Empathy Test, on the other hand, derives its error
from the fact that, if anything, it is too simple and thereby confusing.
It is more likely to suffer from the effects of introspective circum-
locutions by the subject trying to describe complex dispositions in a
two—alternative format. Moreover, one cannot determine whether the sub—
ject recognized and weighed the inevitable contradictory behavior in the
target and made a complex choice, or whether she viewed the question
in its simplified form and chose the only alternative she saw as obvious.
Empathy in these terms is not well articulated nor interpretable.

In addition, the Personality Empathy Test asks only four basic

questions: Is the target a thinker or feeler, intuiter or senser,

103

introvert or extravert, cognitively simple or complex? Each question on
the Videotape Empathy Series asks an independent question: what was
Target A feeling just then? What will she do next? As a result, a few
strategy errors on the Personality Empathy Test can lead to gross dif-
ferences between subjects, while many errors on the Videotape Empathy
Series are needed to produce the same result. In one instance, a Senser-
Thinker target had markedly lower accuracy scores directed toward her
than the remaining targets did. This kind of error is serious because
too much error attributable to target idiosyncracies confounds dependent

variable differences.

Implications for Further Research

 

Problems with the present research seem to focus around three
areas: assumptions about empathy as a process, complexity of the design,
and specific measurement faults. These three areas will be the focus
of this section.

Assumptions About Empathy as a Process

 

The operational definition offered in the beginning of this paper
is a statement about the products of empathy, i.e., correct answers,
without making implications concerning the process by which the answers
can be found. This is an important omission because it does not dis—
criminate between methods which enhance the empathic process and methods
that undercut or at least diminish the effect of the process.

The process of empathy is an active, not a passive, one. The sub—
ject searches for visual and verbal cues present in the target and then

integrates the selectively perceived stimuli into an hierarchy of

 

104

probabilities with relation to experiences she has had and recognized

 

in the past. But time must be allowed for both searching and‘inte-
grating behaviors if empathy is to be attained. The present research
facilitated the subject‘s searching behavior by providing videotaped

data about real functioning, but it handicapped integrative behavior

by reducing the time needed to integrate and by not tapping the subject‘s
true, naturally derived responses either with respect to choosing the
behaviors with which to be empathic or choosing the mode of communica«
tion felt most natural. Thus, the process aspect of empathy was dimin—
ished at worst, confounded at best.

Unfortunately, data produced via videotape seems impossibly con—
founded as well if the proposed process is really allowed to emerge.
Under these circumstances, the subject would watch the videotape and
stop it whenever she had an empathic comment to make. She might say,
“Right here she is feeling angry." The problem arises when one tries to
validate the target's true feelings at that moment. If the same research
procedure used in this research is employed, the target may well have
failed during recall to stop the tape at the point where the subject has
stopped it, since the average number of stops made by the targets during
recall was about once every minute and a half. And, since one may have
as many as twenty slight facial expressions per minute, it obviously
would be difficult to capture enough simultaneous responses by the sub»
jects and targets to ascertain whether many empathic responses were
correct.

A second alternative might be to tape a session, collect the subs

ject's responses, and have the target validate the responses pest hoc.

105

This runs into more serious flaws: the interval between taping and vivid
recall is short (about 48 hours), so the experiment would have to be run
soon after the tape was made. Moreover, targets are easily influenced,
after their first response, by the seeming logic of other responses. In
addition, an informational overload could develop if the targets had
many responses to validate. Thus, while grievous problems exist when
limiting the empathic process of the subjects, equally grievous problems
arise when subjects are freed from normal experimental constraints.

While this writer does not have an answer to the methodological
problem of adequately tapping the process of empathy as it occurs, she
does think that this is the most important problem facing empathy
research today. In the past studies have implied, more probably hoped,
that the subjects could be more flexible than the experimental design
and communicate their empathic knowledge at the times and using the
concepts appointed by the experimenter. While some important informa-
tion can be gained through these methods, far more rigorous and subject-
freeing methods will have to be devised if the most subtle and perplex-
ing questions about the actual process of empathy are to be explored.

The Complexity of the Design

While the design attempted to assess the relationship between per-
ceiving and judging orientations on differing measures of empathy, the
design was too complex to ascertain these effects. Attending to the
various behaviors related to each measure interfered with attending to
behaviors related to the other measures, and thereby increased diffi-
culty. Future studies then, should employ a series of independent tests

to assess these complex questions. Feeling empathy may be explored at

 

106

one setting, thinking empathy at another, behavior empathy at still
another, etc. Moreover, the complexity of stimulus variables should
likewise be determined separately, obvious or straightforward stimulus
behaviors being separated from subtle stimulus behaviors and from
abstruse or masking behaviors. The effects of role-playing versus
true-to-life stimulus situations could again be explored separately.
These are all variables which could profitably be compared with one
another if separate testing situations could reduce measurement error,
or at least make more clear the source from which the majority of the
error originated.

Specific Design Problems and Alternative
Solutions

There are five specific areas noticed by the writer where improve-
ments in design-related techniques could have reduced error of measure-
ment. Methods of choosing targets as well as training targets to attend
to their internal experiencing, and specifications concerning the recall
technique need to be improved. In addition, the construction of
alternative answers on the videotape test and assumptions concerning the
ascription of similarity between targets and subjects need to be con—
sidered in more depth.

As has been pointed out earlier, targets were selected who met
certain testing criteria, but beyond that, no effort was made to ascerv
tain whether additional circumstantial qualities might interfere with
their performance as targets. Selection only on the basis of test
scores was performed to assure "free responding" as opposed to altered

or biased responding. The reasoning did not anticipate the effects of

107

one of the targets being considerably more dominant than the others and
one target consistently displaying characteristics more closely associ—
ated with a different target dimension. Thus, it became clear that in
future research some method of trial screening both with regard to
videotaping and recall should be carried out prior to target group form—
ation. For instance, two assistants could simulate a group interaction
and the interviewer could run through a typical recall session. This
allows for investigation of three factors: (1) whether the potential
target behaves like her target dimension predicts she should behave;
(2) whether some extraneous quality, such as dominance or submission,
interfers with either group functioning or stimulus clarity; and

(3) whether the potential target can adequately recall her experiences
during the videotape session. This takes considerably more money and
time than was available to this experimenter, but from hindsight this
writer now considers it crucial.

In addition to screening out targets who either produce a paucity
of recollected material or who distort the material that is recalled, it
is still felt that some additional exercises specifically concerned with
enhancing the ability to focus on feelings are helpful for recall ses-
sions. It is further suggested that as a complement to the initial
focusing exercise conducted before the first recall session begins, the
first 10 minutes of each recall session consist of a focusing exercise
to acclimate the targets to the recall session itself which is consider—
ably different from their everyday life where it may be important to

guard their privacy.

 

108

A note about the recall technique should also be made. It is
felt by this experimenter that the initial recall made by the target
is usually the most sound. Following the initial statement, some
targets have attempted to discuss what happened in the future as a
result of this response, what happened in the past to cause it, etc.

It is felt that these comments are most likely to be irrelevant or mis—
placed than the initial recall statement. Therefore, although this
experimenter implicitly followed this rule, that only the initial recall
would be used as the source of empathy questions, it may be an important
point to state explicitly for the benefit of future research.

This brings us again to the knotty problem of the test alterna-
tives in videotape tests using multiple choice answers. Having estab-
lished the correct alternative by the first recall to the videotape
stimulus, “incorrect" alternatives pose a problem. In this research,
the incorrect alternatives were experimenter derived and target con—
firmed. In similar tests (Chapman, 1966; Campbell, 1967) incorrect
alternatives have been derived from responses of those scoring lowest
on separate empathy criteria. Targets did not confirm the “wrong”
answers as wrong. The writer feels that while confirmation of inaccuracy
was an important addition, certain problems arose from it. Namely, since
the taping sessions lasted two weeks, and time had to be allowed for
constructing incorrect alternatives, the confirmation sessions did not
occur immediately following videotaping (2 weeks being the longest
interval). By the end of this time, however, whatever traces existed
of some of the recalled material often became confused with the logic of

other responses. Sometimes the target did not even remember what her

109

recall had been. Thus, in future research, especially if less sessions
are used, confirmation sessions should occur sooner.

It is also possible in the future to use ”wrong" answers in a
more meaningful way. In this research wrong answers were defined as
any answer the targets confirmed they were not experiencing at a
particular sequence on the tape. While not suggesting further com-
plexity be added to this particular design, it might be interesting to
construct each “wrong” answer to represent a particular personality
dimension. For instance, a group of Intuiter-Thinker individuals might
be asked for their spontaneous empathic responses to videotape segments
and from those found to be inaccurate, a "wrong“ alternative could be
constructed. The same could be accomplished for each perceiver-judger
dimension--one incorrect response representing each dimension. Patterns
of correct as well as incorrect responding might then be determined.
Although great pains would have to be taken to assure near-equal diffi—
culty of wrong responses, more information about patterns of responding
would be available than in typical experiments where wrong answers are
superfluous and unanalyzable.

Finally, this experiment made certain assumptions about l‘similarity“
between targets and subjects, and within targets and subjects, which
could have been better validated. While the targets did meet certain
test-scoring criteria, we do not know that the factors measured on these
tests were outstanding personality characteristics in these individuals.
For instance, an individual may be an extreme Intuiter, but the salient
factor about her personality is her extreme dogmatism, a factor not

studied in this experiment. Since experiments which attempt to make

110

relevant statements about certain personality dimensions would most
likely have more significant results if they reduced the error attribute
able to more over-riding personality characteristics present in the
target, this writer suggests using some measure to assess the strength
of the personality factor involved. One method to do this could be to
utilize an hierarchical measure of personality to confirm the strength
of the independent variable. Kelly‘s Role Construct Repertory Test
could be used in this way, or descriptions or rankings from others who
know the potential subject might be utilized. In the past, and in this
research, the assumption of similarity on weak evidence may have con—
tributed to much of what is usually a very large sOurce of error:
subjects within variable error. Taking more time assuring that subjects

are really representative of their label may reduce this greatly.

Summary

The lack of significance of the dependent variables and the low
percentage of correct answers were attributed to measurement error both
with respect to the videotape stimulus and to the responses deemed
empathic. Problems related to the stimulus were: (1) the lack of clear
construct validity across the targets; (2) the failure to capture
subtle target behaviors on the screen; (3) the possible inhibiting ef-
fect of the videotape situation per_se; (4) the superficiality of the
videotaped group discussions; and (5) the differential ability of the
targets to express themselves.

The response, as measured by the Videotape Empathy Series, was

confounded by faulty assumptions regarding the nature of empathy,

111

specifically that subjects are equally attentive over time, can handle
four methods of empathic judgment with facility, and will be attending
to only one correct response. As a measuring device, the Videotape
Empathy Series was faulted for its method of constructing correct and
incorrect alternatives and for the possible lack of independence of its
items for those utilizing empathic strategies. The Personality Empathy
Test's generality and limited alternative selection as well as its
seemingly secondary importance with respect to the Videotape Empathy
Series were thought to be contributions to the overall measurement
error.

It was suggested that in future research the process of empathy
should be facilitated, the complexity of the design should be reduced,
and specific precautions should be taken regarding choosing and training
targets, constructing test alternatives, and assessing the notion of

"similarity”.

APPENDICES

APPENDIX A

THE ASSESSMENT MEASURES

112

113

COMBINED PERSONALITY INVENTORY

This is a test designed to show your style of life. The answer
you choose to any question is neither “right“ nor "wrong“. It simply
helps to point out what type of person you are. Therefore, for each
question choose the answer which comes closest to how you usually feel
or act. Mark your choice on the separate answer sheet, as shown in
the samples below.

 

151. People who work hard are unhappy. (A) Basically true.
(B) Basically false.

152. Do you usually (A) get up early? (8) sleep in late?
15] 1 ==: 2 ::: 3 ::: 4 =:: 5 =::
X 152 1 === 2 === 3 === 4 2:: 5 ===

 

 

 

If you think it is basically true that “people who work hard are
unhappy,“ you would mark answer “1“ as it is marked above. Likewise, if
you usually “sleep in late,“ you would mark answer "2" as it is marked
above.

Finally, each question you come across that you feel represents
you in an important way, mark an “X“ in the column by that number. In
the sample above, if you thought getting up late represented an import—
ant fact about you, you would mark an “X" in the column to the left of
the answer as it is marked above.

Although you may feel conflicted about some of the answers to
questions, please try to answer all the questions. If you have a ques~
tion at any time, raise your hand and the test administrator will answer
your question.

—l

10.

11.

12.

13.

14.

15.

114

PART I

. I dislike following a schedule.

(A) Basically true. (B) Basically false.

. With many people you don't know how you stand.

(A) Basically true. (B) Basically false.

. If I can see how a person reacts to one situation, I have a good

idea of how he will react to other situations.
(A) Basically true. (B) Basically false.

. Are you naturally (A) a good mixer?” (B) Rather quiet and

reserved in company?

. Do you think it more important to be able (A) to see the possi-

bilities in a situation, (B) to adjust to the facts as they are?

. Would you rather work under someone who is (A) always kind,

(B) always fair?

. I would be uncomfortable in anything but fairly conventional dress.

(A) Basically true. (8) Basically false.

. I wish people would be more honest with you.

(A) Basically true. (B) Basically false.

. Often a person's basic personality is altered by such things as a

religious conversion, psycho-therapy, or a charm course.
(A) Basically true. (B) Basically false.

I usually take an active part in the entertainment at parties.
(A) Basically true. (B) Basically false.

In your crowd, are you (A) one of the last to hear what's going'
on, (B) full of news about everybody?

Are you inclined (A) to value sentiment above logic, (8) to
value logic above sentiment?

People are too complex to ever be understood fully.
(A) Basically true. (B) Basically false.

Everybody tries to be nice. ,
(A) Basically true. (B) Basically false.

I am a better talker than listener.
(A) Basically true. (B) Basically false.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

115

Would you rather be considered (A) a practical person, (B) an
ingenious person?

00 you think it is a worse fault (A) to show too much warmth,
(B) not to have enough warmth?

For most questions there is one right answer, once a person is able
to get all the facts.
(A) Basically true. (B) Basically false.

I have no fear of spiders.
(A) Basically true. (B) Basically false.

I like to have a place for everything and everything in its place.
(A) Basically true. (8) Basically false.

Can the new people you meet tell what you are interested in
(A) right away, (B) only after they really get to know you?

Would you rather have as a friend someone who (a) is always coming
up with new ideas, (B) has both feet on the ground?

A strong person does not show his emotions and feelings.
(A) Basically true. (B) Basically false.

My way of doing things is apt to be misunderstood by others.
(A) Basically true. (B) Basically false.

You have probably got to hurt someone if you are going to make
something out of your self.
(A) Basically true. (B) Basically false.

I like to fool around with new ideas, even if they turn out later
to have been a total waste of time.
(A) Basically true. (B) Basically false.

It is hard for me to start a conversation with strangers.
(A) Basically true. (B) Basically false.

Most people are consistent from situation to situation in the way
they react to things.
(A) Basically true. (B) Basically false.

I get disgusted with myself when I can not understand some problem
in my field, or when I can not seem tOcmake any progress on a
research problem.

(A) Basically true. (B) Basically false.

I have always hated regulations.
(A) Basically true. (B) Basically false.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

116

Most people are pretty smug about themselves, never really facing
their bad points.
(A) Basically true. (B) Basically false.

Are you more careful about (A) people's feelings, (B) people's
rights?

Would you judge yourself to be (A) more enthusiastic than the
average person, (B) less excitable than the average person?

Perfect balance is the essence of all good composition.
(A) Basically true. (B) Basically false.

All people can talk about these days, it seems, is movies, TV, and
foolishness like that.
(A) Basically true. (B) Basically false.

Politically I am probably something of a radical.
(A) Basically true. (B) Basically false.

I like to talk before groups of people.
(A) Basically true. (B) Basically false.

When you get right down to it, people are quite alike in their
emotional makeup.
(A) Basically true. (B) Basically false.

I always see to it that my work is carefully planned and organized.
(A) Basically true. (B) Basically false.

Straightforward reasoning appeals to me more than metaphors and
the search for analogies.
(A) Basically true. (B) Basically false.

I am certainly lacking in self-confidence.
(A) Basically true. (B) Basically false.

I am embarrassed with people I do not know well.
(A) Basically true. (B) Basically false.

Some people are too complicated for me to figure out.
(A) Basically true. (B) Basically false.

People are too self-centered.
(A) Basically true. (B) Basically false.

You can not classify everyone as good or bad.
(A) Basically true. (B) Basically false.

When you are in an embarrassing spot, do you usually (A) change the
subject, (B) turn it into a joke, (C) days later, think of what you
should have said.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

117

It does not bother me when things are uncertain and unpredictable.
(A) Basically true. (B) Basically false.

Different people react to the same situation in different ways.
(A) Basically true. (B) Basically false.

I must admit that it makes me angry when other people interfere
with my daily activity.
(A) Basically true. (B) Basically false.

I much prefer friends who are pleasant to have around to those who
are always involved in some difficult problem.
(A) Basically true. (B) Basically false.

Criticism or scolding makes me very uncomfortable.
(A) Basically true. (B) Basically false.

Give me a few facts about a person and I will have a good idea of
whether I will like him or not.
(A) Basically true. (B) Basically false.

I enjoy myself most when I am alone, away from people.
(A) Basically true. (B) Basically false.

I can enjoy being with people whose values are very different from
mine.
(A) Basically true. (B) Basically false.

I find it hard to set aside a task that I have undertaken, even
for a short time.
(A) Basically true. (B) Basically false.

I have never been made especially nervous over trouble that any
members of my family have gotten into.
(A) Basically true. (B) Basically false.

A child who is popular will be popular as an adult, too.
(A) Basically true. (B) Basically false.

In your scheme of living, do you prefer to be (A) original,
(B) conventional?

Do you think it is a worse fault to be (A) unsympathetic,
(B) unreasonable?

When I work on a committee I like to take charge of things.
(A) Basically true. (B) Basically false.

I enjoy going with a crowd.
(A) Basically true. (B) Basically false.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

118

I want to know something will really work before I am willing to take
a chance on it.
(A) Basically true. (B) Basically false.

It is a good rule to accept nothing as certain or proved.
(A) Basically true. (B) Basically false.

I find that my first impressions of people are frequently wrong.
(A) Basically true. (B) Basically false.

I usually do not like to talk much unless I am with people I know
very well.
(A) Basically true. (B) Basically false.

I try to remember good stories to pass them on to other people.
(A) Basically true. (B) Basically false.

I like to listen to primitive music.
(A) Basically true. (B) Basically false.

People these days have pretty low moral standards.
(A) Basically true. (B) Basically false.

I dislike having others deliberate and hesitate before acting.
(A) Basically true. (8) Basically false.

I would like to be an actor on the stage or in the movies.
(A) Basically true. (B) Basically false.

I like to go alone to visit new and strange places.
(A) Basically true. (B) Basically false.

People are pretty much alike in their basic interests.
(A) Basically true. (B) Basically false.

I like people I get to know.
(A) Basically true. (B) Basically false.

I find that a well~ordered mode of life with regular hours is con—
genial to my temperament.
(A) Basically true. (B) Basically false.

I must admit I try to see what others think before I take a stand.
(A) Basically true. (B) Basically false.

119

PART III
WHICH WORD IN EACH PAIR REPRESENTS YOU BETTER?

76. (A) build (B) invent 97. (A) statement (B) concept
77. (A) benefits (B) blessings 98. (A) justice (B) mercy
78. (A) facts (B) ideas 99. (A) sensible (B) fascinating
79. (A) calm (B) lively 100. (A) party (B) theater
80. (A) forgive (B) tolerate 101. (A) production (B) design
81. (A) imaginative (B) matter-of— 102. (A) three (B) nine
82. (A) Hzgity (B) quiet 103. (A) analyze (B) sympathize
83. (A) peacemaker (B) judge 104. (A) theory (B) experience
84. (A) complex (B) simple 105. (A) checkers (B) chess
85. (A) determined (B) devoted 106. (A) convincing (B) touching
86. (A) sign (B) symbol 107. (A) make (B) create
87. (A) uncritical (8) critical 108. (A) sociable (B) detached
88. (A) who (B) what 109. (A) firm-minded (B) warm-
hearted

89. (A) speak (B) write 110. (A) complicated (B) uncompli-
90. (A) compassion (B) foresight 111. (A) 53:23 (B) discuss
91. (A) forest (B) field 112. (A) foundation (B) spire
92. (A) literal (B) figurative 113. (A) reserved (B) talkative
93. (A) soft (B) hard 114. (A) wary (B) trustful
94. (A) theory (B) certainty 115. (A) intricate (B) simple
95. (A) concrete (B) abstract 116. (A) thinking (B) feeling
96. (A) gentle (B) firm 117. (A) accept (B) alter

118. (A) tangled (B) untangled

120

PART III

ANSWER THESE QUESTIONS USING THE DIRECTIONS FOR PART I, ON THE FRONT
COVER

119. Would you rather (A) support the established methods of doing good,
(B) analyze what is still wrong and attack unsolved problems?

120. At parties, do you (A) sometimes get bored, (B) always have fun?

121. Do you admire more the person who is (A) conventional enough never
to make himself conspicuous, (B) too original and individual to
care whether he is conspicuous or not?

122. Do you think the people close to you know how you feel (A) about
most things, (B) only when you have had some special reason to
tell them?

123. In a large group, do you more often (A) introduce others, (B) get
introduced?

124. I do not like to see women smoke.
(A) Basically true. (B) Basically false.

125. People are quite different in their basic interests.
(A) Basically true. (B) Basically false.

126. You can not accurately describe a person in just a few words.
(A) Basically true. (B) Basically false.

127. I get very tense and anxious when I think other people are disap-
proving of me.
(A) Basically true. (B) Basically false.

128. If you once start doing favors for people, they will just walk all
over you.
(A) Basically true. (B) Basically false.

129. I am very slow in making up my mind.
(A) Basically true. (B) Basically false.

130. Some of my friends think that my ideas are impractical if not a
bit wild.
(A) Basically true. (B) Basically false.

131. I must admit that I would find it hard to have for a close friend
a person whose manners or appearance made him somewhat repulsive,
no matter how brilliant or kind he might be.

(A) Basically true. (B) Basically false.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

144.

145.

146.

121

Each person's personality is different from the personality of
every other person.
(A) Basically true. (B) Basically false.

Usually I prefer known ways of doing things rather than trying out
new ways.
(A) Basically true. (B) Basically false.

People get ahead by using "pull“, and not because of what they
know.
(A) Basically true. (B) Basically false.

I have had very peculiar and strange experiences.
(A) Basically true. (B) Basically false.

In the matter of friends, do you tend to seek (A) deep friendship
with a very few people, (B) broad friendship with many different
people?

Do you get more annoyed at (A) fancy theories, (8) people who do
not like theories?

It is hard for me to act natural when I am with new people.
(A) Basically true. (B) Basically false.

I think I get a good idea of a person's basic nature after a brief
conversation with him.
(A) Basically true. (B) Basically false.

I always follow the rule: business before pleasure.
(A) Basically true. (B) Basically false.

People really need a strong, smart leader.
(A) Basically true. (B) Basically false.

I prefer to engage in activities from which I can see definite
results rather than those in which no tangible or objective results
are apparent.

(A) Basically true. (B) Basically false.

I can be comfortable with nearly all kinds of people.
(A) Basically true. (B) Basically false.

The average person is not very well satisfied with himself.
(A) Basically true. (B) Basically false.

I find it difficult to give up ideas and opinions which I hold.
(A) Basically true. (B) Basically false.

I would like the job of a foreign correspondent for a newspaper.
(A) Basically true. (B) Basically false.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

158.

159.

160.

161.

122

Most people inwardly dislike putting themselves out to help other
people.
(A) Basically true. (B) Basically false.

People can be described accurately by one term, such as
“introverted“, or ”moral”, or “sociable“.
(A) Basically true. (B) Basically false.

I think you can never really understand the feeling of other
people.
(A) Basically true. (B) Basically false.

I do not like things to be uncertain and unpredictable.
(A) Basically true. (B) Basically false.

I would certainly enjoy beating a crook at his own game.
(A) Basically true. (B) Basically false.

Many of the girls I knew in college went with a fellow only for
what they could get out of him.
(A) Basically true. (B) Basically false.

Novelty has a great appeal to me.
(A) Basically true. (B) Basically false.

People are too easily led.
(A) Basically true. (B) Basically false.

I usually feel nervous and ill at ease at a formal dance or party.
(A) Basically true. (B) Basically false.

If I could ask a person three questions about himself (and assum-
ing he would answer them honestly), I would know a great deal
about him.

(A) Basically true. (B) Basically false.

In your crowd, are you usually (A) one of the first to try a new
thing, (B) one of the last to fall into line?

Do you more often let (A) your heart rule your head, (B) your head
rule your heart?

Is it higher praise to call someone (A) a man of vision, (B) a man
of common sense?

Can you (A) talk easily to almost anyone for as long as you have
to, (B) find a lot to say only to certain people or under certain
conditions?

I do not like to work on a problem unless there is a possibility of
coming out with a clear-cut and unambiguous answer.
(A) Basically true. (B) Basically false.

162.

163.

164.

165.

166.

167.

168.

169.

170-

171.

172.

173.

174.

175.

1760

123

When I meet a person, I look for one basic characteristic through
which I try to understand him.
(A) Basically true. (B) Basically false.

It would not make me nervous if any members of my family got into
trouble with the law.

(A) Basically true. (B) Basically false.

In my experience, people are pretty stubborn and unreasonable.
(A) Basically true. (B) Basically false.

At times I feel that I can make up my mind with unusually great
ease.

(A) Basically true. (B) Basically false.

I feel nervous if I have to meet a lot of people.
(A) Basically true. (B) Basically false.

I find that my first impression of a person is usually correct.
(A) Basically true. (B) Basically false.

It bothers me when something unexpected interrupts my daily routine.
(A) Basically true. (B) Basically false.

People are pretty different from one another in ”what makes them
tick".
(A) Basically true. (B) Basically false.

I keep out of trouble at all costs.
(A) Basically true. (B) Basically false.

People are always dissatisfied and hunting for something new°
(A) Basically true. (B) Basically false.

In reading for pleasure, do you (A) enjoy odd or original ways of
saying things, (B) wish writers would say exactly what they mean?

As a guest, do you more enjoy (A) joining in the talk of the
group, (B) talking separately with people you know well?

If you have a good idea about how several people will react to a
certain Situation, you can expect most other people to react the
same way.

(A) Basically true. (B) Basically false.

I do not like to undertake any project unless I have a pretty good
idea as to how it will turn out.
(A) Basically true. (B) Basically false.

People are basically similar in their personalities.
(A) Basically true. (B) Basically false.

177.

178.
179.

180.

181.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

192.

124

I have had strange and peculiar thoughts.
(A) Basically true. (B) Basically false.

Are you (A) easy to get to know, (B) Hard to get to know?

Do you usually get on better with (A) imaginative people,
(B) realistic people?

The unfinished and the imperfect often have greater appeal for me
than the completed and polished.
(A) Basically true. (B) Basically false.

It is not hard to understand what really is important to a person.
(A) Basically true. (B) Basically false.

A person's reaction to things differs from one situation to another.
(A) Basically true. (B) Basically false.

Many of my friends would probably be considered unconventional by
other people.
(A) Basically true. (B) Basically false.

Do you (A) show your feelings freely as you go along, (B) keep
them to yourself?

If you were a teacher, would you rather teach (A) fact courses,
(B) theory courses?

Which of these two is the higher compliment (A) he is a person.of
real feeling, (B) he is consistently reasonable?

When you have to meet strangers, do you find it (A) pleasant, or
at least easy, (B) something that takes a good deal of effort?

People are unpredictable in how they will react from one situation
to another.
(A) Basically true. (B) Basically false.

Trends toward abstractionism and the distortion of reality have
corrupted much art in recent years.
(A) Basically true. (B) Basically false.

People are so complex it is hard to know what “makes them tick”.
(A) Basically true. (B) Basically false.

In doing something which many other people do, does it appeal more
to you (A) to do it in the accepted way, (B) to invent a way of
your own?

At a party, do you like (A) to help get things going, (B) to let
the others have fun in their own way?

125

193. I am against giving money to beggars.
(A) Basically true. (B) Basically false.

194. I show individuality and originality in my schoolwork.
(A) Basically true. (B) Basically false.

195. Would you have liked to argue the meaning of (A) a lot of these
questions, (B) only a few.

THIS QUESTION IS NOT PART OF THE TEST. IT REFERS TO RESEARCH THAT IS
BEING CARRIED ON THIS SUMMER.

200. If selected, would you like to participate in research concerning
videotaped small discussion groups? (A) Yes. (B) Maybe.
(C) No.

NOW GO BACK AND MAKE SURE YOU HAVE MARKED AN ”X” BY ALL THE ITEMS WHICH
ARE IMPORTANT REPRESENTATIONS OF YOU.

WRITE ANY ADDITIONAL COMMENTS YOU MIGHT HAVE ON THE BACK OF THE ANSWER
SHEET.

APPENDIX B

DATA SUPPORTING THE USE OF THE MYERS-BRIGGS SCALES
AS CONTINUA

126

127

Table 23. Distribution of Sample Population on Extravertalntroyert

Dimension

 

 

SCORES

O—‘Nw-DU‘IO‘ANW

 

 

 

3 6 912 1518 2124 2730 3336 3942 4548 5154 57
INTROVERT —€>-EXTRAVERT

 

Table 24.

Distribution of Sample Population on Sensing-Intuition
Dimension

 

 

SCORES

_.a 3.0 .—I
KO 0 =3 [\3

O—‘Nw-bU'IOWNm

 

 

 

3 6 912 1518 2124 2730 3336 3942 4548 5154 57
SENSING —€>'INTUITION

 

128

Table 25. Distribution of Sample Population on Thinking-Feeling
Dimension

 

 

._I_J._a
ROD—4N

 

SCORES

 

 

 

O—‘NMJ>U‘IO\\IOO

3 6 912 1518 21 24 27303336 394245485154 57
FEELING —>- THINKING

 

 

APPENDIX C

THE EMPATHY MEASURES

129

130

The following is a multiple choice test used in conjunction with
a videotape. Prior to today a series of videotapes were made in which
four women participated in group discussions. Subsequently each group
member individually watched a replay of each videotape and recalled
what she was thinking or feeling during certain segments of the tape.
One correct and three false alternatives were constructed for many of
the segments recalled.

In this part of the experiment, the first videotape of Group A
will be replayed and stopped at the segments mentioned above. When the
videotape stops you are to turn the page and read a question about one
of the group members. You will be given ten seconds to think about
what that individual was thinking or feeling just before the videotape
stopped. While you are thinking, the alternatives to the questions which
begin in the middle of the page, should be kept covered with the sheet of
paper provided. After ten seconds, the experimenter will signal you to
remove the paper and choose which of the four alternatives you think the
indicated group member would say she was feeling or thinking just before
the videotape stopped. After you have circled the answer to the ques-
tion, DO NOT TURN THE PAGE. Wait until the videotape has been started
and stopped again at a new segment to turn the page. IMMEDIATELY COVER
THE BOTTOM HALF OF THE PAGE WITH THE PAPER PROVIDED WHEN YOU TURN THE
PAGE.

Remember, you are to choose which alternative the indicated group
member WOULD SAY she was thinking or feeling just before the videotape
is stopped.

21.

21.

131

GROUP A, SESSION 1

. How did Mary feel as she said she had to be older than her sisters

did to get the same privileges?

. How did Linda feel when she said most of her friends were going to

U of M or State or out of state?

What was Pam thinking as Linda described lying to her mother about
the time?

. How did Mary feel as she said she had to be older than her sisters

did to get the same privileges?

A. Embarrassed—-the others might think she was not as responsible
as her sisters.

B. Protected-~her parents seemed to watch over her more, even
though they did not give her privileges early.

C. Guilty-—for putting her family in a bad light.

D. Resentful——she always seemed to get the raw end of the deal.

. How did Linda feel when she said that most of her friends were going

to U of M or State or out of state?

A. Lonely--most of her friends seemed to be elsewhere.

B. Satisfied-—to go to State. She did not have to apply anywhere
else.

C. Regretful--that she did not apply to U of M or out of state.

D. Cynical——because her friends used going out of state to gain
status.

What was Pam thinking as Linda described lying to her mother about
the time? .

A. I agree, you have to lie to your parents sometimes.

B. Lying is stupid, there is no sense to it!

C. Her mother must be pretty dumb to believe it.

D. I remember the time I lied to my parents and got caught.

22.

22.

132

. What was Pam thinking as Linda talked about Nassau?

. What was Linda thinking as she said she paid half her college

expenses?

What will Mary do next?

. What was Pam thinking as Linda talked about Nassau?

A. I have never been out of Michigan!

B.
C.

B.

C.
D

Linda is trying to impress us.
I would like to go to Nassau too!

D. I do not want to go to Nassau.

. What was Linda thinking as she said she paid half her college
expenses?
A. I did not want to pay anything, but my parents made me pay half.

I do not want to give the wrong impression, it was my idea to
pay half.
I do not like situations where you have to prove you are poor.

. It is a lie, but maybe they will like me better.

What will Mary do next?

A.

B.
CG
D.

Say, "Yeah, lots of times I didn‘t know.“

Say, “I really knew though.“

Say, “They should be happy I came home.“

Say, “Sometimes I‘d try to make a joke out of it.”

23.

23.

133

. What was Pam feeling when Linda said she got away with more than her

sister ever tried to?

. How did Pam feel after she told about her roommate asking about her

grades?

What was Pam thinking as Mary described her dorm?

. What was Pam feeling when Linda said she got away with more than her

sister ever tried to?

A. Amused——she usually got away with more too.

B. Disgusted——she saw Linda as proud and headstrong.

C. Regretful——that she had teased her older brother about how much
she got.

D. Impatient—-she wanted to get on to another topic.

. How did Pam feel after she told about her roommate asking about her

grades?

A. Close-—the others seemed to agree that asking what your grades are
is wrong.

B. Amused——at how scared she had been to tell anyone.

C. Depressed—-it seemed like her grades would never get better.

D. Guilty——she always seemed to put her roommate down.

What was Pam thinking as Mary described her dorm?

A. She must be very unhappy up here!

B. She seems like she is acting—«her expressions are not real.
C. She is making me look bad for liking the dorm.

D. I never looked at it that way. But that is what it is like.

24.

24.

134

. How did Mary feel as Pam told how her brother usually drives her

places?

. How was Ceci feeling after Pam told about her roommate asking about

her grades.

What was Ceci thinking as Mary described her dorm?

. How was Mary feeling as Pam told how her brother usually drives her

places?

A. Amused--she thought it was funny.

B. Impatient--she wanted to go on to deeper things.

C. Anxious—-she was afraid she would have to come up with a story
too.

D. Surprised--that Pam let her brother get away with that.

. How was Ceci feeling after Pam told about her roommate asking about

her grades?

A.
B.
C.
D. Capable——her grades were good, she did not mind telling.

Ashamed-~her grades were not too good either.
Surprised-—she wondered what was so wrong with asking about
grades.

Angry--what a dumb thing for the roommate to ask!

What was Ceci thinking as Mary described her dorm?

A.
B.

C.
D.

Why doesn‘t she move out?

I like the dorm--but I had to move out because it was too expen-
sive.

That is exactly what it is like-~very clinical.

1 am glad Mary got more into the discussion.

25.

25.

135

. What will Ceci do next?

. What was Mary thinking as Linda said that classes were hard up here?

What was Ceci thinking as Pam described finding friends that are
like you?

. What will Ceci do next?

A. Tell what Linda's brother might have been thinking when he would'
not let her drive.

B. Say that she would not let anyone else drive her car either.

C. Acknowledge that similar things had happened to her.

D. Ask what reasons Linda‘s brother gave.

. What was Mary thinking as Linda said that classes were hard up here?

A. I disagree-—classes are easy for me.

B. Yep--I have to work hard here for the average grades I get!
C. I wonder if I am the dumbest one here?

D. Probably I went to a better high school than she did.

What was Ceci thinking as Pam described finding friends that are like
you?

A. It is not so hard for me to find friends.

B. Pam likes a lot of friends; I like only a few.

C. I agree, most of my friends are pretty much like me.

D. I disagree, opposites attract, not those who are just like you.

 

26.

26.

136

. What was Linda thinking as she told about getting the keys to her

boyfriend's car?

. What will Pam do next?

How did Linda feel as Mary talked about moving winter term?

. What was Linda thinking as she told about getting the keys to her

boyfriend's car?

A. I am used to things like that. I guess they are not.

B. They think I am spoiled.

C. I love telling others this. They really think I am good.
D. I wonder what they think of my boyfriend?

. What will Pam do next?

A. Say that her parents never helped her with school work before, so
they do not supervise now.

B. Say that she paid for all of first term, so her parents could not
get down on her for her grades.

C. Ask how the other‘s parents react to grades.

D. Tell about her roommate's parents reaction.

How did Linda feel as Mary talked about moving winter term?

A. Annoyed--she could not understand why Mary would have moved out
in the first place.

B. Sorry for Maryembecause she must have been very lonely.

C. Surprised--that Mary could not make friends easier.

D. Helpful--she usually made it easier for new people to fit in.

27.

27.

137

. What was Ceci thinking as Linda told about getting the keys to her

boyfriend's car?

. How did Linda feel about Mary's parents rejoicing when her grades

were good?

How did Pam feel as she commented on the room looking antique.

. What was Ceci thinking as Linda told about getting the keys to her

boyfriend's car?

A. I would like to have a car to borrow!

B. I would not want the keys to my boyfriend‘s car.
C. I could not even afford the gas.

D. I wonder where she found a boyfriend like that.

. How did Linda feel about Mary's parents rejoicing when her grades

were good?

A. Skeptical--that Mary's parents really did that to control her.
B. Guilty--her grades are never good enough to be rejoiced about.
C. Disappointed--her parents never praise her for her grades.

D. Bored-—everybody always talks about grades.

How did Pam feel as she commented on the room looking antique?

A. 0ptimistic--she liked to talk about antiques and old places.

B. Impatient--she wanted to get this group over with.

C. Disappointed--she wanted the group to be more interesting than it
turned out to be. .

D. Comfortable--she could tell she was getting to know and like the
others when she started talking about the surroundings.

 

28.

28.

138

. What will Linda do next?

. What was Mary thinking as Ceci told about not drinking water with

meals?

How did Mary feel after saying that everybody dried out this
winter?

. What will Linda do next?

A. Talk about the responsibility she feels driving her boyfriendfs
car.

B. Say that his insurance will cover her accidents.

C. Tell an incident about almost getting in a crash.

D. Change the subject.

. What was Mary thinking as Ceci told about not drinking water with

meals?

A. Ceci's father must be a health nut.

B. She is making me thirsty.

C. I should pay more attention to my weight.

D. That is interesting. I would like to hear more about it.

How did Mary feel after saying that everybody dried out this
winter?

A. Comfortable-—she felt like she was more a part of the group now.
B. Awkward-—she felt like she was talking to fill the silences.

C. Guilty-—she thought she was not saying enough Of substance.

D. Annoyed——she thought others in the group should have said more.

139

9. How did Ceci feel after she mentioned hitch-hiking?

19. How did Ceci feel when Linda said, ”That's great!”?

9. How did Ceci feel after she mentioned hitch-hiking?

A.
B.
C.
D.

Foolish--she thought the others considered hitch-hiking dangerous.
Guilty--for putting Linda down.

Superior--she could do things for herself.

Discouraged--her idea had not seemed to go over well.

 

19. How did Ceci feel when Linda said, ”That's great!”?

A.
B.

C.

Uneasy-—it really was not that great. Anybody could do it.
Bitter——she really would like to live in a plush off—campus
apartment.

Embarrassed--she thought she was beginning to sound poverty
stricken.

. Happy--Linda was saying she had good sense.

140

10. What was Linda thinking after Ceci mentioned hitch-hiking?

20. What was Mary thinking as Ceci described how her father brought up
the family?

l0. What was Linda thinking after Ceci mentioned hitch—hiking?

A. Hitch hikers get all wet or splashed. I would not like it.
B. I have hitch-hiked, but it takes too long.

C. I would not hitch-hike. I am more dependent.

D. I do not think Ceci likes me.

20. What was Mary thinking as Ceci described how her father brought up
the family?

A. It sounds pretty cold to me!

8. That is what I hate about families-—they test you.

C. I would rather have a car than know why I could not have one.

D. That is a good way to raise kids-—to let them decide for them-
selves.

141

NOW THAT YOU HAVE VIEWED THE VIDEOTAPED GROUP SESSION, ANSWER THE
FOLLOWING GENERAL QUESTIONS ABOUT THE GROUP MEMBERS.
Questions on Linda

29. Who did Linda like best during the group session?
B) Mary B) Pam C) Ceci

30. Who did Linda like least during the group session?
A) Mary B) Pam C) Ceci

31. Who did Linda think was most similar to her?
A) Mary B) Pam C) Ceci

Questions on Mary

32. Who did Mary like best during the group session?
A) Pam B) Ceci C) Linda

33. Who did Mary like least during the group session?
A) Pam B) Ceci C) Linda

34. Who did Mary think was most similar to her?
A) Pam B) Ceci C) Linda

Questions on Pam

35. Who did Pam like best during the group session?
A) Ceci B) Linda C) Mary

36. Who did Pam like least during the group session?

A) Ceci B) Linda C) Mary
37. Who did Pam think was most similar to her?
A) Ceci B) Linda C) Mary

Questions on Ceci

38. Who did Ceci like best during the group session?
A) Linda B) Mary C) Pam

39. Who did Ceci like least during the group session?
A) Linda 8) Mary C) Pam

40. Who did Ceci think was most similar to her?
A) Linda B) Mary C) Pam

 

APPENDIX D

THE AWARENESS MATERIALS

142

143

GENDLIN'S FOCUSING TECHNIQUE

This is going to be just to yourself. What I will ask you to do
will be silent, just to yourself. Take a moment to relax ... (5 seconds).
All right now, just to yourself, inside you, I would like you to pay
attention to a very special part of you....

Pay attention to that art where you usually feel sad, glad, or
scared ... (5 seconds). Pay attention to that area in you and see how
you are now.

See what comes to you when you ask yourself, ”How am I now?” "How
do I feel?" “What is the main thing for me right now?”
Let it come in whatever way it comes to you, and see how it is.

 

30 seconds or less

 

If among the things you have just thought of, there was one major
problem which felt important, continue with it. Otherwise, select a
meaningful personal problem to think about. Make sure you have chosen
some personal problems of real importance in your life. Choose the thing
which seems most meaningful to you.. . (19 seconds).

Of course, there are many parts to that one thing you are thinking
about-eto many to think of each one alone. But, you can feel all of
these things together. Pay attention there where you usually feel things,
and in there you can get a sense of what all of the problem feels like.
Let yourself feel all of that.

 

 

 

30 seconds or less

 

As you pay attention to the whole feeling of it, you may find that
one special feeling comes up. Let yourself pay attention to that one
feeling.

 

l minute

 

Keep following one feeling. Do not let it be just_words or pictures,
wait and let words or pictures come from the feeling.

 

l minute

 

If this one feeling changes, or moves, let it do that. Whatever it
does, follow the feeling and pay attention to it.

 

l minute

 

Now, take what is fresh, or new, in the feel of it now ... and go
very easy. Just as you feel it, try to find some new words or pictures
to capture what your present feeling is all about. There does not have

144

to be anything that you did not know before. New words are best, but
old words might fit just as well. As long as you now find words or
pictures to say what is fresh to you now.

 

1 minute

 

If the words or pictures that you now have make some fresh differ-
ence, see what that is. Let the words or pictures change until they
feel just right in capturing your feeling.

 

1 minute

 

Now I will give you a little while to use in any way you want to,
and then we will stop.

  

' af'ANxmus 7

 

APPENDIX E

FEELING SUMMARY USED BY TARGET TO
IDENTIFY FEELINGS

146

147

FEELING INDICATOR

Check all the feelings you were having in the last few seconds before

the videotape stopped.

in response to.

surprised
awe

talkative
amused

happy

hopeful

optimistic
enthusiastic
courageous--daring

sensible--reasonable

patient
flexible
helpful
capable
determined
proud

relieved

comfortable--at ease

satisfied
refreshed

protected
appreciated

ambivalent
confused-~mixed up
baffled

erratic--disorganlzed.:::::

impatient

listless--indifferent::::::

bored
daydreaming
exhausted

reluctant
careful
cautious

stubborn--resistant:::

forced
controlled
trapped

frustrated
annoyed-~angry
disgusted--fed up

upset-disturbed
anxious
embarrassed
foolish-ridiculous
guilty

worried

scared

disappointed
abused
bitter

regretful
discouraged
depressed
defeated
lonely
helpless
hopeless

other

 

 

 

 

Explain, if you can, what these feelings were

III»

 

APPENDIX F

POST-EXPERIMENT DATA

148

149

SUBJECTS' OBSERVATIONS CONCERNING THE EXPERIMENT

Observations

 

I. Problems with quality of videotaped information.

A.

Some videotaped sessions were more interesting than
others.

B. Group II was easier to understand than Group I.
C. Targets did not say their real feelings. The situ-

ation was artificial.

. Thought targets were trying to be safe.

II. Problems with attending.

E. Subject was more awake some days than others.

.31

KLHICD

. Took awhile to become accoustomed to the testing

situation.

. Had a hard time following participants.
. Had a hard time seeing the videotape.

Too complex.
Felt pressured during the testing situation.

. Felt empathized differently with some than with

others.

III. Problems with tests.

L.

Information on videotape not adequate to answer
questions.

M. Had different alternatives than those given.

. Questions too difficult, too specific; better if

more general.

. Thinking questions were more flexible than the

others.

. Subject reacted to the same words on the Myers-

Briggs differently.

No. of Ss
Responding

—l-—l—J|\)—J

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Abelson, R. P., and Sermat, V. Multidimensional scaling of facial
expressions. Journal of Experimental Psychology, 1962, 63,
546-54.

Allport, G. W. Personality; A Psychological Interpretation.
New York: Henry Holt, 1937.

 

Allport, G. W. Pattern and Growth in Personality. New York:
Harper and Row, 1961.

Asch, J. E. Forming impressions of personalities. Journal of Abnormal
and Social Psychology, 1946, 41, 258—290.

Bender, I. E., and Hastorf, A. H. On measuring generalized empathic
ability (social sensitivity). Journal of Abnormal and Social
Psychology, 1953, fig, 503-506.

Bender, I. E., and Hastorf, A. H. The perception of persons: Fore-
casting another person's responses on three personality scales.
Journal of Abnormal and Social Psychology, 1950, 45, 556-61.

Berne, E. Transactional Analysis in Psychotherapy. London: Evergreen
Books Ltd., 1961.

Bieri, J. Cognitive complexity-simplicity and predictive behavior.
Journal of Abnormal and Social Psychology, 1955, 51, 263-268.

Blanchard, W. A. Relevance of information and assimilation-contrast
in interpersonal prediction. Perceptgal and Motor Skills, 1967,
24, 515-25.

 

Blanchard, W. A. Assimilation and contrast in interpersonal predic-
tion with control for the interaction of real similarity and dif-
ferential accuracy. Journal of Personality and Social Psychology,
1966, 3, 567—73.

 

Bronfenbrenner, U., Harding, J., and Gallwey, M. O. The measurement of
skill in social perception. In A. L. Baldwin et al. (eds.),
Talent and Society. Princeton, N. J.: Van Nostrand, 1958.

 

Bruner, J. 5., and Tagiuri, R. The perception of people, in Lindzey,
C. (ed.), Hardbook of Social Psychology. Cambridge, Mass.:
AddiSOn-Wesley, 1954, 634—54°

150

 

151

Buchheimer, A., Goodman, J., and Sircus, G. Videotapes and kinescope
recordings as situational tests and laboratory exercises in
empathy for the training of counselors. Hunter College of the
City University of New York, 1965. A technical report submitted
to the U. S. Office of Education, NDEA Grant No. 7-42-0550-1670.

Buros, 0. K., Ed. The Sixth Mental Measurements Yearbook. Highland
Park, N. J.: The Gryphon Press, 1965, pp. 321-26.

Campbell, R. J. The development and validation of a multiple-choice
scale to measure affective sensitivity (empathy). Unpublished
doctoral dissertation, Michigan State University, 1967.

Chapman, J. L. The development and validation of a scale to measure
empathy. Unpublished doctoral dissertation, Michigan State
University, 1966.

Cline, V. B. Ability to judge personality assessed with a stress inter-
view and sound-film technique. Journal of Abnormal and Social *

Psychology, 1955, £9, 183—7.

Cline, V. B. Interpersonal Perception. In B. A. Maher (ed ), Progress
in Experimental Personality Research, 1964, v01. 1, 221-284

 

 

Cline, V. B., and Richards, J. M. Accuracy of interpersonal perception--

a general trait. Journal of Abnormal and Social Psychology,
1960, 99, 1—7.

Cline, V. B. and Richards, J. M. A comparison of individuals versus
groups in judging personality. .Journal Abnormal and Social
Psychology, 1961, gg, 446—49.

Cohen, A. R. Cognitive tuning as a factor affecting impression forma-
tion. Journal of Personality, 1961, g2, 235-245.

Cook, M. Interpersongl Perception. London: Penguin Books Inc., 1971.

 

Cottrell, L. S., and Dymond, R. F. The empathic responses: A neglected
field for research. Psychiatry, 1949, 1g, 355-59.

Crockett, W. H. Cognitive complexity and impression formation. In B. A. l
Maher (ed.), Progress in Experimggtal Personality Research, Vol. ‘

2, New York: Academic Press, 19 5

Cronbach, L. J. Processes affecting scores on ”understanding of others”
and “assumed similarity". Psychological Bulletin, 1955, gg,
177—193.

 

sets in interpersonal perception. Journal of Abnormal and Social
Psychology 1957, 54_, 384- 390.

 

l
Crow, W J , and Hammond, K. R. The generality of accuracy and response i

152

Darwin, C. The Expressions of the Emotions in Man and Animals.
London: J. Murray, 1872.

Davitz, J., and Davitz, L. Correlates of accuracy in the communication
of feelings. Journal of Communication, 2; 5—13.

Dymond, R. F. A scale for the measurement of empathic ability.
Journal of Copsulting Psychology, 1949, 15, 127—133.

Ekman, P., and Friesen, W. V. Constants across cultures in the face
and emotion. Journal of Personality 55d Social Psychology, 1];
124-129.

Engen, T., Levy, N., and Schlosberg, H. The dimensional analysis of a
new series of facial expressions. Journal of Experimental
Psychology, 1958, 55, 454-458.

Estes, S. G. Judging personality from expressive behavior. Journal of
Abnormal and Social Psychology, 1938, 55, 217-36.

Estes, S. G. The judgment of personality on the basis of brief periods
of behavior. Unpublished doctoral dissertation, Harvard Univer-
sity, 1937.

Eysenck, H. J. Manual of the Maudsley Personality Inventory. London:
University of London Press, 1959.

Feigenbaum, K. D. Egocentrism in young children. Paper presented to
Psychology Department Colloquium, Michigan State University, 1967.

Fernberger, S. W. False suggestion and the Piderit model. American
Journal of Psychology, 1928, 59, 562-568.

Gage, N. L., and Cronbach, L. J. Conceptual and methodological problems
in interpersonal perception. Psychological Review, 1955, 523
411-422.

 

Gage, N. L., Leavitt, G. S., and Stone, G. C. The intermediary key in
the analysis of interpersonal perception. Psychological Bulletin,
1956, 55, 258—266.

 

Gendlin, E. T. Experiencing and the Cregtion of Meaning. New York:
The Free Press of Glencoe, 1962.

Halpern, H. M. Some factors involved in empathy. Dissertation
Abstract, 1954, 143 l458~9.

Hastorf, A. H., Schneider, D. J., and Polefka, J. Person Perception.
Reading, Mass.: Addison-Wesley, 1970.

 

in.

153

Hathaway, S. Clinical intuition and inferential accuracy. Journal of
Personality, 1955, 24, 223-250.

Heist, P., and Yonge, G. Omnibus Persopglity Inveptory, Form F.
New York: The Psychological Corporation, 1962.

Hogan, R. Development of an empathy scale. Journal of Consulting and
Clinical Psychology, 1969, 55, 307—16.

Howarth, E. Extroversion and dream symbolism: an empirical study.
Psychological Reports, 1962, 19, 211-214.

Jackson, D. N., and Messick, S. J. Individual differences in social
perception. British Journal of Social gpd Clinical Psychology,
1963, 2, 1-10.

Jung, C. G. Psychological Types. Trans. H. G. Baynes. New York:
Harcourt, Brace and Company, 1923. (First German edition, 1921.)

 

Jung, C. G. Contributions to Analytical Psychology. Trans. C. F.
Baynes and H. G. Baynes. New York: Harcourt, Brace and Company,
1928.

Kagan, N., Schauble, M. S., Resnikoff, A., Danish, S. J., and Krathwohl,
D. R. Interpersonal Process Recall, Journal of Nervous and Mental
Disease, 1969, 135, 365-374.

Kagan, N., and Krathwohl, D. R. Studies in Human Interaction. (Final

report, Grant number 0ET-32-0410-270) East Lansing, Michigan:

Michigan State University, Educational Publication Services, 1967.

Kanekar, S. Accuracy in person perception. Psychologia: An Inter-
national Journal of Psychology in the Orient, 1972, 153 76—88.

 

Katz, R. L. Empathy: Its Nature and Uses. London: The Free Press of
Glencoe, 1963.

Kelly, G. A. The Psychology of Persopgl Constructs. New York: Norton,
955, 2 Vols.

Kerr, W. A., and Speroff, B. T. The Measurement of Empathy. Chicago:
Psychometric Affiliates, 1951.

Kirk, R. E. Experimental design: procedures for the bepgvioral sciences.
Belmont, Ca1if.: Brooks/Cole, 1968.

Lipps, T. Das Wissen von fremden Ichen. Psychologische Untersuchungen,
1907, 1, 694-722.

 

Lundy, R. M. Assimilative projection and accuracy of prediction in inter-
personal perception. Journal of Abnormgl and Social Psychology,
1956, 52, 33-38.

 

154

Maslow, A. H. Motivation and Personality, New York: Harper and Row,
1954.

Mehrabian, A., and Epstein, N. A measure of emotional empathy.
Journal of Personality, 1972, 45, 525-543.

Mendelsohn, G. A. Myers-Briggs Type Indicator. In 0. K. Buros, The Sixth

 

Mental Measurements Yearbook. Highland Park, N. J.: Gryphon Press,
1965, 331-332.

Munn, N. L. The effect of knowledge of situation upon judgment of emo-
tion from facial expressions. Journal of Abnormal and Social
Psychology; 1940,_55, 324-338.

Murray, H. A. Explorations in Personality. New York: Oxford University
Press, 1938.

Myers, 1. B. The Myers-Briggs Type Indicator. Princeton, N. J.:‘
Educational Testing Service, 1962a.

Myers, I. B. Manual(196g),the Myers-Briggs Type Indicator. Princeton,
N. J.: Educational Testing Service, 1962b.

Notcutt, B., and Silva, A. L. M. Knowledge of other people. Journal
of Abnormal and Social Psychology, 1951, 45, 30-37.

Purcell, K., Modrick, J. A., and Yamahiro, R. Item vs. trait accuracy
in interpersonal perception. Journal of General Psychology,
1960, 55, 285-292.

 

Rank, R. C. Counseling competence and perceptions. The Personnel and
Guidance Journal, 1966, 45, 359-365.

 

 

Rogers, C. R. Client-Centered Therapy. Boston: Houghton Mifflin,
1951.

Rommetveit, R. Selectivity, Intuition, and Halo Effects in Social
Perception. Oslo: Oslo University Press, 1960.

 

Ross, J. Progress Report on the College Student Characteristics Study:
June, 1961. Research Memorandum 61-11. Princeton, N. J.:
Educational Testing Service, 1962.

 

Schlosberg, H. The descriptions of facial expressions in terms of two
dimensions. Journal of Experimental Psychology, 1952, 44, 229-
237.

Schneider, 0. J. Implicit personality theory: A review. Psychological
Bulletin, 1973, Z5, 296-311.

 

155

Sechrest, L., and Jackson, 0. N. Social intelligence and accuracy of
interpersonal predictions, Journal of Personality, 1961, 55,
167-182.

 

Shrauger, S., and Altrocchi, J. The persOnality of the perceiver as a
factor in person perception. Psyghological Bulletin, 1964, 55_
289—308.

 

Smith, H. C. Sensitivity to People.) New York: McGraw-Hill, 1966.

Speroff, B. J. Empathy and role reversal as factors in industrial
harmony. Journal of Social Psychology, 1953, 55, 117-120.

Steiner, I. Interpersonal behavior as influenced by accuracy of social
perception. Psychological Review, 1955, 55, 268-274.

Steinmetz, H. C. Directive psychotherapy: V. Measuring psychological
understanding. Journal of Clinical Psychology, 1945, 4, 331—335.

 

Stone, G. C., and Leavitt, G. S. Generality of accuracy in perceiving
standard persons. Chicago: University of Illinois, 1954.
(Mimeo.)

Stotland, E. Exploratory investigations of empathy. In L. Berkowitz
(ed.), Advanced Experimental Social Psychology, 1969, Vol. 4,
271-314.

 

Stotland, E., Sherman, S. E., and Shaver, K. G. Empathy and birth
order. Lincoln, Neb.: University of Nebraska Press, 1971.

 

Stricker, L. J., and Ross, J. Intercorrelations and reliability of the
Myers—Briggs Type Indicator scales. Psychological Reports,
1963, 15, 287-293.

 

Stricker, L. J., and Ross, J. Some correlates of a Jungian personality
inventory. Psychological Reports, 1964, 44, 623-643.

 

Strunk, Orlo. Empathy: A review of theory and research. Psycho-
1ogig§1_News1etter, 1957, 5, 47-57.

Suchman, J. R. Social sensitivity in the small task-oriented group.
Journal of Abnormal and Social Psychology. 1956, 55, 75-83.

 

Sullivan, H. S. The Interpepspnal Theory of Psychiatry. New York:
Norton, 1953.

 

Sundberg, N. D. Myers-Briggs Type Indicator. In 0. K. Buros, The Sixth

 

Mental Measurements Yearbook. Highland Park, N. J.: Gryphofi‘Tl ‘
Press, 1965, 332-335.

 

156

 

Taft, R. The ability to judge people. Psychological Bulletin, l955,
5 l-23.

_9

Tagiuri, R. Person perception. In G. Lindzey (ed.), Handbook of
Social Psychology. Reading, Mass.: Addison-Wesley, 1968, Vol.
4, 395-449.

Thompson, D. F., and Meltzer, L. Communication of emotional intent by

facial expression. Journal of Abnormal and Social Psychology,
I964, 55, 129—l35.

Truax, C. B. A scale for the measurement of accurate empathy.
Psychiatric Institute Bulletin, Wisconsin Psychiatric Institute,
University of Wisconsin, l96l, l) 12.

 

Truax, C. B., and Carkhuff, R. R. Toward Effective Counseling and

Psychotherapy: Training and Practice. Chicago: Aldine Publishing

 

Company, 1967.

Valentine, C. W. The relative reliability of men and women in intuitive

judgments of character. British Journal of Psychology, l929, 12,
213-238.

 

Warr, P. B., and Sims, A. A. A study of cojudgment processes. Journal

of Persona1ity, 1965, 553 598-604.

Wescott, M. R. Toward a Contgmporary Psychology of Intuition.
New York: Holt, Rinehart and Winston, 1968.

 

 

Vernon, P. E. Personality Assessment: A Critical Survey. London:
Methune, 1963.

Woodworth, R. S. Experimental Psychology. New York: Henry Holt,
l938

Zajonc, R, B. The process of cognitive tuning in information.
Journgl of Abnormal and Social Psychology, l960, 51, 159-167.

 

, 77,,_ __ .7 _ _ ___.___ 7. _._. i III I _

"I7'11lii’iiiliilliiii1E5