THE EFFECTS OE‘MAN‘IPULATING
NONVERBAL BEHAVIORS 0F INTERVIEWERS
ON THE NONVERBAL BEHAVIOR OF
INTERVTEWEES

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
ROBERT A. KAPLAN
1971

H~1tll¥

“an at

University

 

This is to certify that the

thesis entitled

THE EFFECTS OF MANIPULATING '
NONVERBAL BEHAVIORS OF INTERVIEWERS
ON THE NODWERBAL BEHAVIORS

OF INTERVIEWIiES

presented by
ROBERT A. KAPLAN

has been accepted towards fulfillment
of the requirements for ,

Ph.D. degreein Psychology

 

 

0-7639

          

. nus, _
HDAu & SUNS’
800K BINDERY IND.

LEEPJLRY SINDERS

Cull! III' “I-..

 

 

 

 

 

 

ABSTRACT

THE EFFECTS OF MANIPULATING NONVERBAL
BEHAVIORS OF INTERVIEWERS ON THE
NONVERBAL BEHAVIORS OF INTERVIEWEES

BY

Robert A. Kaplan

The purpose of the present investigation was to
study the nonverbal interaction between members of a dyad;
an interviewer and an interviewee. Interviewers were
trained to manipulate certain nonverbal behaviors, and
the effects of these manipulations on the nonverbal beha-
viors of subjects were recorded and studied. In addition,
measures of affiliation were gathered from subjects, and
these measures were studied in relationship to observed
nonverbal behaviors.

Eight individuals, four males and four females,
were trained to conduct fifteen minute interviews in which
they varied their nonverbal behavior in three discrete time
periods. During the first and third time periods inter-
viewers were to control the movements of four body areas
(head and neck, hands and arms, legs and feet, and body

and postural shifts). During the second time period

Robert A. Kaplan

interviewers displayed movements in these four areas
according to a specific program.

Forty subjects, twenty males and twenty females,
were selected from introductory psychology courses. Each
subject was interviewed by one of the trained interviewers,
and afterwards answered a questionnaire designed to assess
how well the subject like the interviewer, as well as more
general aSpects of social affiliation.

Each interview was recorded on video tape, and was
then rated by a pair of trained raters. The behaviors of
both interviewers as well as subjects were then subjected
to various statistical analyses. In addition, five separate
nonverbal variables (positive head nodding, forward leaning,
Space closing leg movement, arms crossed postures and legs
crossed postures) were studied in relationship to the
assessed measures of affiliation.

The results of the statistical analyses indicated
the following: (1) subject nonverbal behavior of the head
and hands was significantly different across time periods,
although these differences were not in the predicted
direction; (2) measures of affiliation did not seem to be
related to the nonverbal variables studied, nor to the
degree to which subjects were influenced by the interviewers;
(3) no sex differences were found in relationship to the
degree by which subjects were influenced by the interviewers;

and (4) there were significant sex differences in the

Robert A. Kaplan

self-reported measures of affiliation. Females tended to
like the interviewers more, seemed to be more Open in
terms of new social situations, and reported themselves as
being more outgoing as compared to males. Various explan-
ations of the above results were discussed, as well as
their implications. Results broadly indicate that certain
nonverbal behaviors of subjects vary in a significant
manner when these subjects are in the presence of inter-
viewers who vary their own nonverbal behaviors according
to a specific program. Unlike previous research, these
changes did not seem to be related to differences in sex
or affiliation. Implications of these general relationships

were discussed, as well as suggestions for future research.

 

 

THE EFFECTS OF MANIPULATING NONVERBAL
BEHAVIORS OF INTERVIEWERS ON THE
NONVERBAL BEHAVIOR OF

INTERVIEWEES

BY

,1
Robert Aeraplan

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Psychology

1971

Now what is first to be considered is that the
organism always works as a whole. We have not a liver
or a heart. We are liver and heart and Brain and so on,
and even this is wrong. We are not a summation of parts,
but a coordination of all these different bits that go
into the making of the organism.

(F. Perls, 1969, p. 5)

"But animals don't always speak with their
mouths," said the parrot in a high voice, raising her
eyebrows. "They talk with their ears, with their feet,
with their tails--everything" . . . "You have to notice
all those little things if you want to learn animal-
language. That is the first thing to remember: bein a
good noticer is terribly important in learning animaT-
language."

(Advice from Polynesia the Parrot.
H. Lofting, 1967, p. 7, p. 99)

TO THOSE WHO HAVE BROUGHT ME JOY
AND CONTENTMENT OVER THE LONG HAUL.

ii

 

IN ACKNOWLEDGMENT

OF THE GOOD DR. STOLLAK, WHOSE MIND AND OFFICE HAVE ALWAYS
BEEN OPEN. ONE OF BUT A FEW WHO TEACH BY BEING.

OF LARRY MESSE FOR GUIDANCE THROUGH THE MURKY CHASMS OF
STATISTICAL ANALYSES.

OF ART SEAGULL AND BILL CRANO FOR COMMENTS AND IDEAS.

OF CONRAD COLE, FOR.A GERMINAL IDEA IN THE RIGHT PLACE
AT THE RIGHT TIME.

OF LARRY BLUMBERG AND DICK MILLER WHO DON'T ANSWER.WHY
QUESTIONS.

OF MARGIE, FOR BEING AND BECOMING HERSELF.

iii

 

TABLE OF CONTENTS

Page
ACMOWLEDGMNTS O O O O O O I O O O O O O O O O I O O O . iii

LIST OF TABLES O O O O O O I O O I O O O O O O O O O O Vii

LIST OF APPENDICES O O O O O O O O O O O O O O I O O O x
Chapter
I O INTmDUG‘ION I O O O O O O O O 0 O O 0 O O O O l
Purposes and Goals . . . . . . . . . . . . . 5
Interpersonal Influence: Verbal Aspects . . ‘ 6

Interpersonal Influence: Language and

Cultural Functions . . . . . . . . . . . _ lO
Interpersonal Influence: Context Analysis 7‘13
Interpersonal Influence: Body Congruence i916
Interpersonal Influence: Proxemic Vari-

ables of Hall . . . . . . . . . . . . . . 5 l7

Interpersonal Influence: Work of Mehrabian 18
Interpersonal Attraction . . . . . . . . . 21
Sex Differences . . . . . . . . . . . . . . 21

Summary and Hypotheses . . . . . . . . . . . 24
Hypothesis I . . . . . . . . . . . . . . . 25

HYPOtheS is II 0 O I O O O O O O O O O I O O 2 6
HYPOtheSiS I II I O O O O O O I O O O O O O 2 6

II 0 METHOD 0 o o o o o o o o o o o o o o o o' o o o 28

subjects 0 O O O O O O O O O O O O O O O O O 28

Interviewers . . . . . . . . . . . . . . . . 29
Training of the Interviewers . . . . . . ._ 29
Basic Skills for Conducting Interviews. . 30
Nonverbal Interview Training . . . . . . 31

segments A, A. O I O O O O O O O O O O 31
segment B I O O O O O O O O O O O O O O 32

iv

Chapter Page

Experimental Procedure . . . . . . . . . . . 34
Questionnaire . . . . . . . . . . . . . . . . 35
Rating of the Data Questionnaire . . . . . . 36
Video Tapes of the Interviews: Rating Pro-

cedure . . . . . . . . . . . . . . . . . 36

Rating of the Studied Variables . . . . . . . 38
Training of Raters . . . . . . . . . . . . . 39

III. ESUIITS O O O O O O O O C O O O I O O O O O O O 41

Results of the Analyses of Interviewer

Behavior . . . . . . . . . . . . . . . . . 41
Interviewer Head Movement Frequency . . . . 42
Interviewer Head Movement Duration . . . . 42
Interviewer Hand Movement Frequency . . . . 43
Interviewer Hand Movement Duration . . . . 43
Interviewer Leg Movement Frequency . . . . 45
Interviewer Leg Movement Duration . . . . . 45
Interviewer Body Movement Frequency . . . . 47
Interviewer Body Movement Duration . . . . 47
Summary of Interviewer Behavior . . . . . . 48
Results of the Analyses of Subject Behavior 48

Subject Head Movement Frequency . . . . . 48

Subject Head Movement Duration . . . . . . 49
Subject Hand Movement Frequency . . . . . . 50
Subject Hand Movement Duration. . . . . . . 50
Subject Leg Movement Frequency . . . . . . 51
Subject Leg Movement Duration . . . . . . . 52
Subject Body Movement Frequency and

Duration . . . . . . . . . . . . . . . . 52
Summary of Subject Behavior . . . . . . . . 52

Questionnaire Data: Correlations of Responses
with Five Kinesic Variables . . . . . . . . 53

Analysis of the Eight Questions . . . . . . 55

Question One: "How do you feel about the

interview you have just finished? . . . 55
Question Two: "How did you feel about the
person who interviewed you? . . . . . . 56

Question Three: "If the interviewer lived
in or near your dorm, or building, what
chance would there be of becoming

friends with the interviewer? . . . . . 56
Question Four: "How do you think the
interviewer felt about you? . . . . . . 57

Chapter Page

Question Five: "How much do you like

being emotionally close to other

persons?. . . . . . . . . . . . . . . . 57
Question Six: "When you meet strangers in

a social situation (for example, in a

class, at a party, at the grill) how do

you feel? . . . . . . . . . . . . . . . 58
Question Seven: "If you enter a room, and

a group is laughing and talking, how

strong a desire do you have to enter the

group? . . . . . . . . . . . . . . . . 58
Question Eight: "Ar you an outgoing

person? . . . . . . . . . . . . . . . . 58

Summary and Discussion of Questionnaire
Arlalyses I O O O O O O O O C O O C I O I 59

Overall Summary of Results . . . . . . . . . 60
IV. DISCUSSION 0 O O O O O O O O O O I O O O I I O 61

overView O O O O O 0 O O O O O O 0 O O O O O 61
Methodological Considerations . . . . . . . . 63

Comments Concerning the Nonverbal Variables
Rated 0 O O O O O O C O O O I O I O O O O 66
Comments Concerning the Length of the

InterView O O O O O O O O O O I O O C O O 67
Racial ASpects of Interviewer and Subject
Populations I O I O O O O I O O O I O O O 68

Implications of the Findings of the Present

Research . . . . . . . . . . . . . . . . . 69
Time Period Effects . . . . . . . . . . . . 69
Discussion of Affiliation Measures and Non-
verbal Correlates . . . . . . . . . . . . 79
Sex Differences . . . . . . . . . . . . . . 80
Questionnaire Variables . . . . . . . . . . 81

Implications of the Above Findings for Future
Research . . . . . . . . . . . . . . . . . 83
Clinical Implications of the Present Research 85

LIST OF MFERENCES O O I O O O O O O O O O O I O O O O ' 87

APPENDICES O O O O O O O I O O O O O O O O O O O O O O 96

vi

LIST OF TABLES

Page

1. Means of interviewer hand movement duration . . . . 44
2. Summary of simple effects for the three-way inter-

action of sex of subject x sex of interviewer x

time period for interviewer hand movement duration 46
3. Sex of interviewer x sex of subject interaction

means for subject head movement duration . . . . . 49
4. Results of recombining data . . . . . . . . . . . . 51
5. Sex of subject x sex of interviewer interaction

means for question one. . . . . . . . . . . . . . . 56

6. Pair I test for agreement . . . . . . . . . . . . . 96
7. Pair II test for agreement . . . . . . . . . . . . 96
8. Pair III test for agreement . . . . . . . . . . . . 97

9. Summary of analysis of variance for interviewer
he ad frequency I I I I I I I I I I I I I I I I I I 9 8

10. Summary of interviewer head frequency means . . . . 98

11. Summary of analysis of variance for interviewer
head duration . . . . . . . . . . . . . . . . . . . 99

12. Summary of interviewer head movement duration means
in seconds I I I I IAIvI IvI I I'IvI-IvI I I I I I 99

13. Summary of analysis of variance for interviewer
hand frequency . . . ... . . . . . . . . . . . . . 100

14. Summary of interviewer hand movement frequency means 100

15. Summary of analysis of variance for interviewer
hand duration I I I I I I I I I I I I I I I I I I I 101

16. Summary of interviewer hand movement duration means 101

vii

17.

18.
19.

20.
21.

22.

23.

24.
25.

26.

27.
28.

29.

30.

31.

32.

33.
34.

35.

Page

Summary of analysis of variance for interviewer leg
frequency I I I I I I I I I I I I I I I I I I I I I 102

Summary of interviewer leg movement frequency means 102

Summary of analysis of variance for interviewer leg
movement duration . . . . . . . . . . . . . . . . . 103

Summary of interviewer leg movement duration means 103

Summary of analysis of variance for interviewer
body movement frequency . . . . . . . . . . . . . . 104

Summary of interviewer body movement frequency
means I I I I I I I I I I I I I I I I I I I I I I I 104

Summary of analysis of variance for interviewer
bOdy duration I I I I I I I I I I I I I I I I I I I 10 5

Summary of interviewer body movement duration means 105

Interviewer means for head, hand, leg and body
variables for each interviewer and each time period 106

Summary of analysis of variance for subject head
movement frequency . . . . . . . . . . . . . . . . 108

Summary of subject head movement frequency means. . 108

Summary of analysis of variance for subject head
movement duration . . . . . . . . . . . . . . . . . 109

Summary of subject head movement duration means . . 109

Summary of analysis of variance for subject hand
movement frequency I I I I I I I I I I I I I I I I l l 0

Summary of subject hand movement frequency means. . 110

Summary of analysis of variance for subject hand
movement duration . . . . . . . . . . . . . . . . . 111

Summary of subject hand movement duration means . . 111

Summary of analysis of variance for subject leg
movement frequency . . . . . . . . . . . . . . . . 112

Summary of subject leg movement frequency means . . 112

viii

36.

37.
38.

39.
40.

41.

42.

43.

44.

45.

46.

47.
48.
49.
50.
51.
52.
53.

54.

Summary of analysis of variance for subject leg
movement duration . . . . . . . . . . . . . . .

Summary of subject leg movement duration means .

Summary of analysis of variance for subject body
movement frequency . . . . . . . . . . . . . . .

Summary of subject body movement frequency means

Summary of analysis of variance for subject body
movement duration . . . . . . . . . . . . . . .

Summary of subject body movement duration means.

Subject means for head, hand, leg and body vari-
ables for each interviewer in each time period .

Summary of analysis of variance for recombined
data, subject hand duration, no. 1 . . . . . . .

Summary of analysis of variance for recombined
data, subject hand duration, no. 2 . . . . . . .

Summary of analysis of variance for recombined
data, subject hand duration, no. 3 . . . . . . .

Summary of analysis of variance for recombined
data, subject hand duration, no. 4 . . . . . . .

Summary of analysis of variance for question
Summary of analysis of variance for question

Summary of analysis of variance for question
Summary of analysis of variance for question

Summary of analysis of variance for question

1
2
3

Summary of analysis of variance for question 4 .
5
6
Summary of analysis of variance for question 7
8

Summary of analysis of variance for question

ix

Page

113

113

114

114

115

115

116

118

118

119

119
120
120
120
121
121
121
122
122

LIST OF APPENDICES

Page
A. Rater Training and Agreement Data . . . . . . . . . 96
B. Summaries of Statistical Analyses . . . . . . . . . 98
C. Questionnaire . . . . . . . . . . . . . . . . . . . 123

CHAPTER I

INTRODUCTION

It is no accident that the last sentence of

Finnegans Wake leads naturally back into the first. The

 

theme is that of unity, of totality, the continuous flow
from end to end and back again. Communication is much

like Joyce's book, a continuous flow, often inexplicable

at first glance (and at last glance, for that matter) but
indeed an onrushing stream in which we are totally immersed
at all times. Generally we are blind, at least on a con-
scious level, to the things that rush and flow past. When
most persons think of communication, they think of talking,
of words. Perhaps this is a natural concomitant of the
written language tradition. Oral tradition, on the other
hand, is not only rich in the lore and wisdom it contains,
but also in the added wealth of information that it carries.
For example, if these words were printed in a brilliant
shade of blue upon a yellow page, perhaps you, the reader,
would be formulating impressions somewhat different from
those that are now being shaped. Surely, if the margins
were not so straight, nor the typist so accurate, your

reaction to these words and ideas would be different.

Take, for example, the following poem by William Carlos
Williams:

Well God is love so love me God is love
so love me God is love so love me well

The poem has been jumbled, above, for a reason which may

become apparent when it is displayed as written, below:

Well God is
love

so love me
God

is love so
love me God
is

love so love
me well

(Calypsos I, W.C. Williams,
Pictures from Breughel, p.56)

There are subtleties in communication which, for the most
part, are overlooked and not dealt with. This work does
have nice margins, and the pages are crisp and white. Thus,
it is communicated that this is a serious work to be
regarded as such by those who pick it up and read it. But
you, the reader, obtain only a partial sense of what is
truly being communicated. You do not know the look on this
writer's face when these lines were composed, not the dis-
comfort displayed when dealing with a difficult concept, nor
the smile, when in the midst of the verbiage, a good sentence
or paragraph is produced. Communication is not only words
on a typed page. Nor is it only sound waves that strike

the tympanic membrane carrying messages on their way to

the brain, at which time specific verbal referents are
decoded and encoded. Communication is also the fact that
Speaker A has a tic in his left eye, that he hesitates
for 0.3 seconds before each phrase, that he looks down
when he begins a sentence and that he looks up at the end
of each sentence or phrase to assess the effect his words
have had on his audience. In other words, organisms convey
messages, or communicate through a multitude of different
channels. Messages carried through these various channels
often complement each other in terms of the gestalt of
what is being expressed. Very often, however, these
channels are at odds with one another, each channel con-
veying slightly, or very different messages.

Since the verbal channel is usually perceived as
the dominant means of conveying and receiving messages,
the less easily perceived and/or understood messages of
the nonverbal channels tend to be overlooked, or ignored.
This is the case in normal day to day interaction and, as
well, in the history of the study of communication in the
behavioral sciences. It took the work of Darwin (1965)
and Freud (1933) who, through excellent use of naturalistic
observation, Sparked a great deal of interest in the further
study of nonverbal behavior. It is clear that Freud
integrated a great deal of his knowledge of nonverbal
behavior into his theory and practice of psychoanalysis

(Reik, 1948; Sherman, 1965). Duncan (1969) offers an

excellent overview of the more recent noverbal literature.
"Nonverbal behavior" has been interpreted to include any
form of behavior that is not strictly related to verbal
content. Thus Duncanlists some six separate areas of
nonverbal research. These are
(a) body motion or kinesic behavior; gestures and other
body movements, including facial expression, eye
movement, and posture;
(b) paralanguage: voice qualities, speech non-
fluencies and such nonlanguage sounds as laugh-

ing, yawning, and grunting;

(c) proxemics: use of social and personal Space and
man's perception of it. . . ;

(d) olfaction;
(e) skin sensitivity to touch and temperature; and

(f) use of artifacts: such as dress and cosmetics.
(Duncan, 1969, p. 118)

All of the above mentioned behaviors play a part in the
total process of communication. These are not the only
categories of nonverbal behavior, nor does Duncan make any
such claims. However, the field is too large to deal with
unless one narrows his focus to some specific area. This
choice can be made by many different processes, but this
writer has tried to make this choice based on its
relevancy to interpersonal influence. Assuming that the
content of verbal behavior is the most important aspect

of normal interpersonal communication, it is difficult to
select other variables in a hierarchy of importance. In

terms of the six categories that Duncan mentions above,

perhaps the first two, kinesics and paralanguage, play the
next most important role in communication. The focus of
this research will be on the nonverbal area of kinesics,
or body and gestural movement. Ekman and Friesen (1968),
Mahl (1968), Mehrabian (1970) and Krout (1935) offer com-
prehensive reviews of this literature.

In summary, it can be said that the gestalt of
communication covers a very broad Spectrum of behaviors.
It is often assumed that the primary mode of communication
between humans is through the content of verbal channels.
However, it is apparent that nonverbal modes of communica-
tion play an important role in the communicative activities
of every day life. Often the information that is conveyed
in or through nonverbal channels is minimally perceived,
at least on a conscious level, and thus awareness of its
importance has been long in coming. Research concerns and
technological developments (such as imprOved audio and
video-tape recording equipment) have fostered much research
in the nonverbal areas within the past 15-20 years, and
thus there are adequate foundations from which one can begin

research in this area.

Purposes and Goals

 

This research deals with nonverbal, kinesic behavior
as it is displayed in a dyadic relationship. There has

been ample precedent set in using the dyad, as many of the

works reviewed by Duncan (1969), Ekman (1968) and Mahl
(1968) follow this framework. The dyad is the smallest
unit in which true interaction between individuals can
occur, and it is assumed that relationships that can be
established within this smaller behavioral unit will apply
to larger groups as well. Within the framework of the
dyad, this research will focus on three major factors.

These are

(a) the degree, nature, and manner of nonverbal
influence that one member of the dyad exerts
upon the other member of the dyad;

(b) the effect of differential degrees of affili-
ative states that may occur between members of
a dyad, and the manner in which this affects
the nonverbal behaviors of the members of the
dyad; and

(c) differential effects in nonverbal behavior that
can be attributed to sex differences in members
of the dyads.

Most of the nonverbal literature deals with interactive
effects stemming from various experimental conditions.

These works apply directly to factor (a) above, and will

be dealt with first.

Interpersonal Influence:
Verbal Aspects

The original impetus for this research came from
the work of Matarazzo and his colleagues (Matarazzo, 1962;
Matarazzo, Saslow, Matarazzo, 1956; Matarazzo, Weitman,
Saslow, 1963; Matarazzo, Wiens, Matarazzo, Saslow, 1968;

Matarazzo, Wiens, 1967; Saslow, Matarazzo, 1959; Wiens,

Saslow, Matarazzo, 1966). These studies all focused upon
the patterns and relationships of temporal aSpects of verbal
behavior between members of a dyad. Their work followed
that of Chapple and others (see Matarazzo, et. al., 1956,
for a review of this earlier work) and made use of Chapple's
"Interaction Chronograph" and a later, modernized adaptation
of the chronograph, which allowed them to gather a great
deal of information about verbal relationships by merely
recording speech and pause durations of any two individuals.
The primary emphasis of this work was the manner in which
temporal aSpectS of the verbal productions of one individual
affected similar verbal productions of another. Matarazzo
has been collecting data in this area since 1954, and since
that time has arrived at some very sound observations and
propositions. Behaviors that have been studied include
length of Speech utterances, length of reaction time after
speech utterances, frequency of interruption, effect of
head nodding on speech utterances, and the effect of saying
"Mn'Hmn" on the length of speech utterances, to give but a
partial accounting.

Matarazzo found that persons typically speak in
short utterances; one third of all utterances are less than
12 seconds in duration, and three quarters of all utterances
are less than 36 seconds in duration. These figures hold
up for patients in therapy, for therapists, and for inter-

viewees during one session interviews. In a recent study

(Matarazzo, et. al., 1968) which is quite relevant to the
present research, Matarazzo began to manipulate the
utterances of an interviewer in a dyad to see if the
utterances of the interviewee would be affected by these
manipulations. An experimental condition was instituted
in which the interviewer would speak for 15 minutes in 5
second utterances, 15 minutes in 10 second utterances, and
then 15 minutes in 5 second utterances again. The results
were clear cut. Interviewees had a mean length of Speech
utterance of 24.3 seconds for the first period, 46.9
seconds for the second period, and 26.6 seconds for the
third period. The actual mean durations of the interviewer's
utterances were 5.3, 9.9 and 6.1 seconds, respectively.
This work was repeated many times with many variations.

In another condition of this same basic research scheme
(reported in Matarazzo, et. al., 1968) the interviewer
spoke in a 10-5-10 second pattern, and the mean length of
utterances for interviewees went from 41.1 to 22.8 back

to 48.2 seconds, reSpectively.. All of these studies seem

to point to the fact that the verbal actions of one member

 

of a dyad are closely related in a positive manner to the
verbal actions of the other member of the dyad.

It should be mentioned that Matarazzo and his
colleagues did not gather information on other nonverbal
behaviors that occurred during their interviews. This

fact does not negate their findings, but one can speculate

upon the influence that the nonverbal behaviors may have
had in producing the changes in verbal rates of responding.
It is within the realm of possibility to assume that some
nonverbal components, such as the qualities of speech tone,
or certain gestural or facial patterns exhibited by the
experimenter may have accounted, in part, for the increases
in verbal behavior of the subjects. Rosenthal (1967) has
commented upon this type of "unintended communication" in
research projects, and although he has not isolated exact
variables, he has speculated that certain experimenters,
for example, tip off their subjects as to what results are
expected. Rosenthal reports that he can Spot these experi-
menters after observing them for only 30 seconds. In Mata-
razzo's case, however, it is reasonable to assume that
there may be some combination of verbal and nonverbal
factors in Operation that combine to produce the obtained
results. In regard to these studies, it appears that there
are marked interactive relationships that occur between
individuals. If one approaches the sphere of nonverbal
behavior as a type of language behavior, it would seem

that the same, or at least similar types of relationships

should be found.

10

Interpersonal Influence:
Language and Cultural Functions

The notion that nonverbal behavior constitutes a
type, or pattern of language is put forth strongly by
Birdwhistle (1952, 1957, 1962, 1963a, 1963b). Birdwhistle
breaks nonverbal behavior down into component parts, just
as a linguist would dissect a spoken or written bit of
verbal behavior. Using traditional language as a model,
Birdwhistle draws many parallels between the building
blocks, or elements of verbal and nonverbal "language."
For example, "linguistics" itself has a parallel in
"kinesics." "Sounds" are paralleled by "muscular and
skeletal shifts." "phonemes" by "kines," "allophones" by
"allokines," "morphemes" by "kinemorphs," and so on. Not
only does Birdwhistle make these theoretical comparisons,
be he also supplies a transcription scheme (1952) so that
there can exist a written nonverbal language.

Birdwhistle and Hall (1963, 1964) both believe in
the specificity of nonverbal language, in terms of culture
and patterns of learning. Thus, it is their contention
that one cannot go into a tribe, for example, in Chicago,
and expect that the patterns of nonverbal language will be
the same as were found in a tribe in Moscow. To Bird-
whistle and Hall, any study of nonverbal behaviors that
is not within a cultural context is not seen as a valid

exercise. Efron (1941), working from this same point of

11

view, dramatically illustrates the differences in nonverbal
gestural behavior between various American subcultures.
Efron found that Jews and Italians in New York City
differed significantly in the behaviors utilized in every-
day interpersonal behavior. In addition, there seemed to
be certain patterns of behavior that typified each sub-
culture. This type of finding raises the question of the
applicability of results obtained from the study of non-
verbal behaviors. Indeed, if such results are so cultur-
ally determined, then any study of them might have only
minimal interest or relevance.

There are others, however, who hold ideas quite
different from those of Birdwhistle and Hall. Darwin
(1965) and Ekman and Friesen (1968) propose a universality
of at least certain nonverbal behaviors. Darwin felt that
humans react in the same manner to certain emotional or
situational states, at least in terms of facial expressions.
Ekman (personal communication) has adOpted this view,
although in a somewhat modified form. Ekman insists that
there are certain "micro-expressions" of 1/125th second
duration which reveal universal reactions, again in terms
of facial expressions, and that these universal expressions
exist between and across various cultures. When exposed
to a specific stimulus, such as a terrifying sight or
sound, it is Ekman's contention that all persons react

with very Similar facial displays of fear. However,

12

immediately after this "micro-expression" has been dis-
played, the culturally determined facial expressions come
into play, thus accounting for the reported cultural
differences in facial expression.

The point to be emphasized here is that somewhere
between the ends of the continuum as expressed by Bird-
whistle and Hall on one hand, and Darwin and Ekman on the
other, lies the degree to which relationships that are
learned in one subculture are applicable to individuals
in other subcultures. In other words, if the findings of
the present research are applicable only to members of a
midwestern university subculture, then there is little
reason, if any, for embarking on such an endeavor. A way
to deal with this dilemma is to examine studies in which
members of varying cultures and/or subcultures distinguish
and interpret various nonverbal behaviors. Mahl (1959,
1968), Ekman and Friesen (1967, 1968), Ekman (1965), and
Ekman and Winters (1965) deal with just this problem.
Utilizing both trained and untrained observers of nonverbal
behaviors, it was found that Specific nonverbal communica-
tions, either in the context of positions, acts, or other
variables, could be reliably interpreted. Although no
demographic data are reported in these studies, it can be
assumed that all of the subjects and/or observers did not
come from the same subcultures. Thus, at least within a

large culture, such as is found in the United States, one

13

can assume that there will be sufficient reliability in
the interpretation of the meaning of Specific behaviors,
and therefore findings should be applicable to large seg-

ments of the United States pOpulation.

Interpersonal Influence:
Context Analysis

Scheflen (1963, 1964a, 1964b, 1965, 1966, 1967) is
closely allied with Birdwhistle, and applies many of Bird-
whistle's principles to the study of the dyadic relation-
ship of therapist and patient. He makes use of a method
which he calls "context analysis."

Context analysis theory sees the communica-

tional behaviors as more likely to be mutual,

often Simultaneous, and highly interdependent.

(Scheflen, 1963, p. 128)

Scheflen contends that nonverbal acts are not random, nor
are they loosely tied to specific happenings, but that they
comprise a very specific language which regulates and
facilitates the flow of on going interpersonal behavior.
In other words, individuals use their nonverbal language
either consciously or unconsciously to let others know
just where they stand in relationship to these others.
For example, persons use their nonverbal language to keep
the verbal flow from becoming too personal, to let others
know they are pleased with what is going on, or a number

of other variables. The following passage offered by

Scheflen will clarify this point:

14

In a session involving . . . two psychothera-
pists, the schiZOphrenic girl, and her mother,
every few minutes the young woman crossed her
legs seductively, exposing her thigh and present-
ing her breast to the older of the two physicians.
Immediately afterward she would Sit close to him
and form a lexical alliance with him against her
mother. On each such occasion, the mother made a
characteristic gesture, sweeping the side of her
index finger across her nostrils.

This gesture has been studied in many situ-

ations with other subjects, and is regularly

associated with negation, or disapproval. Each

time the older woman wiped her nose the daughter

uncrossed her legs, returned them to a more "lady-

1ike" position, and disassociated herself from the

doctor. The mother sometimes crossed her ankles

in a particular way with or without wiping her

nose and the effect was the same.

(Sheflen, 1963, pp.l31-l32)

There are two elements of importance here. One is the
actual nonverbal language that seems to be transPiring in
this type of transaction (and according to Scheflen, in
every transaction). It does indeed seem that certain indiv-
iduals have very specific nonverbal methods of conveying
Specific meanings. Scheflen points out that these nonverbal
acts are used again and again in the repertoire of all
individuals, and that these acts (which Scheflen calls
"points" or "markers") seem to spell out the boundaries of
any particular interaction. In addition, Scheflen (1964b)
has studied body posture and position, and has come up with
hypotheses Similar to those formulated concerning gestural
acts.

The second element to be considered here is the

degree to which intricate nonverbal interaction takes place

15

between individuals. This is one of the major concerns of
the current research, and Scheflen's studies emphasize the
degree and complexity of the interaction that does occur.
It can be seen that the implications of Scheflen's
work point in the direction of a language of nonverbal
behaviors; a language occurring with the frequency and
regularity of verbal language. Similar to verbal language,
nonverbal language behaviors tend to play an intimate
relationship in the patterns of interaction that exist
between individuals. Unfortuantly, Scheflen does not
report any "hard data" concerning his many observations.
Thus, while the relationships that Scheflen reports seem
to be valid, no empirical evidence, other than the case
study observations such as the one cited above, is offered.
Thus, it is difficult.for one to observe just what type of
interrelationships do exist in the interactions reported.
It is one thing to point to the presence of a language of
nonverbal behavior, but this reveals nothing about the
relationships that may exist within that language framework.
Matarazzo has Shown this to be the case with verbal language.
Other investigators (e.g., Charny, 1966) working in a
tradition similar to that of Sheflen, have begun to explore
some of the basic relationships that seem to exist within

the context of nonverbal language.

16

Interpersonal Influence:
Body Congruence

Charny (1966) explored a behavior that Scheflen
(1964) calls "congruent body positions." Congruency in
body posture, or body position, deals with the manner in
which one individual sits in relationship to another.
For example, if one member of a dyad sits with his right
leg crossed over his left, and the other member of the dyad
assumes this same position, there is "lower body congruency."
Charny studied body congruency of a therapist and a patient
during a thirty minute therapy session. He found that
across eight successive time periods, the degree of body
congruency increased. Lexical analysis of the interview
revealed that during periods of congruency, tOpics dis-
cussed by the patient were interpersonally oriented,
positive, and specific, while during periods of non-congruency,
topics were self-centered, negative, and non-specific.
Charny concluded that body congruency is an indicator of
greater relatedness, or rapport, during a therapy session.

Charny's work, like that of Scheflen, offers support
to the notion that nonverbal interactive behaviors are an
important aSpect of the dyadic relationship. What is not
_known from this study, however, is the manner in which
greater congruency occurred. It might be speculated that
either the therapist or the patient began to institute a

certain type of body posture, and that as this posture was

17

displayed more and more frequently it elicited similar
postural responses from the other member of the dyad. Note
that this explanation does not affect Charny's original
interpretation of the same data. He stated that at times
of body congruency, there is a greater sense of rapport

or relatedness. It may well be that at times of body
congruence, the nonverbal channels mesh in such a manner

so as to increase feelings of closeness, or relatedness.

In this manner individuals may feel more comfortable in
relating through verbal channels, thus the differences in

verbal content.

Interpersonal Influence:
Proxemic Variables of Hall

Hall (1959, 1963, 1964) is another investigator who
has placed a great deal of importance upon nonverbal vari-
ables that affect interpersonal relationships. Working in
much the same framework as Birdwhistle, Hall prefers to
call his work "Proxemics" as opposed to "Kinesics." Unlike
Birdwhistle (but more like Scheflen), Hall attends primarily
to the interactive aspects of nonverbal factors. He dis-
tinguishes some eight distinct proxemic variables which
include differences in posture, angle of orientation, near-
ness to other persons, degree or amount of-eye contact,
loudness, touch, smell, and thermal qualities. Hall contends

that each of these variables affects the quality of any

18

relationship. These factors were important in the study
of the dyad that was undertaken in the current research.
In other words, as many of these variables as possible were
controlled for so that differences attributable to these
variables did not obscure differences obtained as a result
of the experimental procedure. Many investigators have
begun to deal with the variables that Hall has isolated,

and these studies are summarized below.

Interpersonal Influence:
Work of Mehrabian

Mehrabian and his colleagues (1967, 1968a, 1968b,
1969a, 1969b, 1969c, 1970; Mehrabian and Friar, 1969, and
Beakel and Mehrabian, 1969) have done extensive work using
many of the variables that Hall (1963) outlined. Most of
the work done follows a similar format. In a typical study
(Mehrabian and Friar, 1969) subjects were asked to imagine
themselves in various social Situations with persons of
different social status, different sex, and persons who
they liked or disliked. The dependent variables studied
were the distance between the subject and the imagined
other, posture, body position, eye contact, and other
similar variables. A major drawback of the work of Mehr-
abian and his colleagues has been that they do not deal
with an interacting dyad in almost all of their studies,

but deal instead with a subject imagining himself in a

19

relationship. They have, however, made a number of observa—
tions that seem to coincide with.observations reported by
others using interacting dyads. Relationships that Mehr-
abian has found include findings such as the following:

(1) The physical distance between individuals decreases
as the amount of positive regard, or positive affect increases.

(2) Eye contact tends to increase when persons tend to
like each other more, although he finds that this is not a
constant relationship. There seems to be the least amount
of eye contact between persons who do not like each other,
or who like each other the least, and the most eye contact
between persons who are neutral towards each other. With
persons liked most, there is slightly less eye contact than
in the neutral condition.

(3) With females, "body orientation" (the degree to
which persons either face each other or turn away from each
other) follows a pattern Similar to that of eye contact.
That is, least liked females turn away from a.face-to-face
orientation the most, while neutral females tend to have
the most face-to-face, or near face-to-face orientation.
Most liked females turn away from each other slightly.
Males, however, tend to show little differences in body
orientation with disliked or neutral others, while there
is a slight decrease of the face-to-face orientation with

most liked males.

20

Mehrabian explains the lack of eye contact and decrease in
face-to—face orientation with most liked others by the fact
that "intimate" friends tend to stand closer to each other,
looking slightly past each other. In other words, intimates
seem to have a separate pattern of behavior which they
follow.

(4) A body position with arms akimbo (hands on hips,
elbos out) seems to communicate negative feeling or intent.

(5) Relaxation of the body tends to communicate a
sense of more positive regard than a tense, tight configura-
tion.

(6) Persons who lean forward tend to communicate more
positive regard than persons who sit in upright positions,

or who lean away from the other member of the dyad.

An important aspect of these findings is the effect of sex
differences upon nonverbal behavior, and the effect of dif-
ferential internal states (e.g., differences in status,
affiliative states, etc.). As was stated in the onset of
this discussion, the study of sex differences and differences
in interpersonal attraction are issues which are considered
in the present research. Mehrabian found sex differences
in almost all of his studies, and although he makes little
or no attempt to explain such differences, their existence
alone is an important consideration. Others have found

sex differences in the expression of various nonverbal

behaviors, and these studies will be discussed below.

21

Interpersonal Attraction

The effects of differences in interpersonal attrac-
tion between members of interacting dyads has not received
much attention. The few studies conducted raise important
considerations that have direct bearing upon the present
research.

Rosenfeld (1965a, 1965b, 1966a, 1966b, 1967) and
Rosenfeld and Franklin (1966) have dealt with the aspect of
interpersonal attraction in terms of approval seeking and
approval avoiding situations. It was found that in approval
seeking situations, subjects used twice as many hand ges-
tures and smiles as Opposed to approval avoiding situations,
in which subjects used more negative head nods. It was
also found that approval seekers stand closer to other mem-

bers of the dyad than approval avoiders.

Sex Differences

Exline (1963), Exline and Winters (1965) and Exline,
Gray, and Schutte (1965) have studied sex differences as
well as the differential effects of interpersonal attraction.
The dependent variables were the frequency and duration of
eye contact between the members of dyads. One very signi-
ficant finding was that women tend to engage in more eye
contact than men in a number of situations. Exline (1963)
ties this to a significantly higher need for affiliation

and inclusion found for his female subjects. In one study

22

(Exline and Winters, 1965), eye contact was related to how
much the subject liked the person who was interviewing him
(or her). Interviewers assumed one of three attitudes
after several minutes of the interview had passed. They
became either obnoxious, more friendly, or remained neutral.
In the obnoxious condition, eye contact drOpped 8.8 percent,
remained essentially the same in the neutral condition, and
rose 3.3 percent in the more friendly condition. All-
subjects in the obnoxious condition showed a decrease of
eye contact. In a variation of the above experimental
condition, subjects were allowed to choose preferred, or
most liked interviewers. While females showed an increase
in eye contact, men did not. However, the men did Show a
decrease in eye contact with the least preferred inter-
viewers.

Rubin (1970) studied eye contact with couples who
displayed differing amounts of love for each other via a
Love Scale that he developed. As predicted, he found that
couples who scored as being strongly in love engaged in
significantly more eye contact than couples who scored as
being weak in love. In addition, it was found that
individually, women engage in more gazing, or attempts at
eye contact than do men, regardless of the love component
of their relationship.

The studies of Mehrabian, Rosenfeld, Exline, and

Rubin all point to the manner in-which nonverbal behavior

23

is modified through the influence of differing states of
interpersonal attraction, and differences in sex. Their
work had particular bearing on the present research, in
that these studies laid the basic framework for the second
and third major considerations that were dealt with in
this study. Exline and Mehrabian (as well as Argyle,
Salter, Nicholson, Williams and Burgess, 1970) all
indicate that vomen are more prone to certain nonverbal
behaviors than are men. Exline attempted to explain this
phenomenon by speculating that women have greater needs
for affiliation and inclusion than do men. Another
possible explanation lies in extrapolating from the find-
ings of Witkin and his colleagues (Witkin, Lewis, Hertzman,
Machover, Meissner, and Wapner, 1954). They report that
in terms of perceptual measures, women are more field
dependent than are men. In other words, women are more
influenced by the perceptual field that surrounds them
than men are, and will tend to make use of the information
gained via the field as opposed to using information from
internal reference points, whatever these points may be.
Men, on the other hand, tend to disregard the information
gathered via the field, and depend more upon these inter-
nal mechanisms. It may be inferred that the situation of
the dyad represents a social field, and that women carry
their perceptual field dependence into some type of social

field dependence. In this case, women would tend to be

24

more influenced by the nonverbal components of the social
field than men. This, then, could account for the fact
that women are more affected by changes in nonverbal
behavior, and therefore exhibit more of these nonverbal
behaviors than do men.

In terms of the variable of interpersonal attrac-
tion, or the degree of affiliation one has in regard to
the other member of the dyad, the above studies under-
score the differences in nonverbal behavior that seem to
result from differences in interpersonal attraction. The
present research attempted to study variables affected
by differences in affiliative states and attempted to
designate other specific nonverbal variables, if any, that

were also affected by such differences.

Summary and Hypotheses

Within the context of this review, it has been
shown that there are many interactive aspects of nonverbal
behavior. While most studies have pinpointed specific
aspects of behavior, or have pointed to a generalized lan-
guage of nonverbal behavior, no one has investigated the
more global relationships of nonverbal behavior as Mata-
razzo and his colleagues have done with regard to verbal
behavior. It was the main purpose of this research to
investigate and establish the nonverbal relationships that
exist between members of an interacting dyad. Thus, the

first hypothesis:

25

Hypothesis I

The nonverbal behaviors of one member of a
dyad are intimately related to the nonverbal
behaviors of the second member of that same dyad
such that an increase or decrease in specific
nonverbal acts of one member will bring about
a significant corresponding increase or decrease
in the corresponding nonverbal acts of the
second member of that same dyad.

The concept of "nonverbal acts" was borrowed from.Ekman-
and Friesen (1968). They defined an act in the following
manner, and this same definition was used for the current
research.

An act begins when a movement is first
detectable and ends when a movement is no longer
apparent or when another visually distinctive
type of movement commences. A still osition
begins and ends when movement stOps or Begins.

. . . A nonverbal act is defined as a movement
within any Single body area (head, face, shoulder,
hands, or feet) or across multiple body areas,
which has visual integrity and is visually dis-
tinct from another act. The beginning of an act
is determined as the point at which the part of
the body under scrutiny begins to move from the
still body position, or by the addition of
another distinct act. . . The classification of
acts and positions is thus based upon what is
easily recognizable to any observer.

(Ekman and Friesen, 1968, pp. 193-194)

Data were collected from four distinct body areas which
included

(1) the head and neck,

(2) the hands and arms,

(3) the feet and legs, and

(4) global body acts, i.e., changes of posture or
position.

26

The second hypothesis dealt with the area Of affili-
ation or the degree to which one member of the dyad liked

the other member.

Hypothesis II

 

The degree to which one member Of a dyad is
influenced by the nonverbal acts Of the other
member of the dyad is related in a positive
manner to the degree of positive regard one
member feels towards the other, such that the
more one likes another, the more he will be
influenced by the other's nonverbal acts, and thus
will incorporate these same acts in his own non-
verbal behaviors. In addition, persons who Show
more positive regard towards the other member Of
the dyad will also demonstrate more of these
specific behaviors: (a) positive head nods,

(b) smiles, (c) open arm postures, (d) Open leg
postures, (e) more leaning towards the other
person, (f) more leg movement which tends to
close the space between the members Of the dyad,
and (9) more eye contact with the other member
Of the dyad.

The third hypothesis dealt with sex differences as

they applied to nonverbal behaviors.

Hypothesis III

 

Female subjects will tend to display more
nonverbal behaviors, and will tend to be influenced
more by the nonverbal manipulations of the other
member Of the dyad than will male subjects.

Hypothesis II actually has two separate parts. The first
part deals with changes in frequency Of nonverbal acts.

That is, persons who diSplay more positive regard for the

other member of the dyad should engage in closer modeling

27

of the behaviors of the other member, as stated in the
first hypothesis.

The second aspect of this hypothesis deals with
the diSplay of Specific nonverbal behaviors that are not
necessarily related to the acts Of the other member Of the
dyad. Exline (1963), Mehrabian (1967-1970), Argyle (1965)
and Perls (1969) have related all Of the behaviors men-
tioned in Hypothesis II to feelings Of positive regard.
It was hOped that this research would provide a way in
which to validate these findings in terms of positive
regard as measured by a questionnaire to be described
below.

Hypothesis III states that female subjects will
tend to display more nonverbal behaviors than men, and

thus will Show a greater modeling effect than will men.

CHAPTER II

METHOD

Subjects

Forty subjects, 20 male and 20 female, were
selected from the undergraduate population Of Michigan
State University. Selection was accomplished by placing
Sign up sheets in the large lecture classes Of the
introductory psychology courses. These sheets indicated
that subjects were needed for research in interviewing
techniques, and that all subjects would receive course
credit as well as $1.50 for participation in this research.

Each subject took part in a fifteen minute inter-
view in which the interviewer behaved according to a
programmed schedule. After the completion of the inter-
view, each subject filled out short questionnaires regard-
ing their feelings and thoughts about the interview experi-
ence and the person who interviewed them. The subject was
then given course research credit and $1.50 as payment for
his participation. Subjects were told nothing Of the
nature of the experiment, and those who asked about the
study were told they would receive a notice describing the

experiment at a later date. At the completion Of the

28

29

collection of all data, such notices were sent to all

subjects.

Interviewers

 

Eight individuals,l four males and four females,
were selected to carry out the interviews with the above
mentioned subjects. Six Of these persons were actors and
actresses, members of the theater department. It was
assumed that theater students would have had greater
training in the control Of their bodies, and therefore
would be better able to program their nonverbal behaviors
in accordance to a standard nonverbal repertoire. The
remaining two individuals were not actors, but did display
skill in interviewing and in controlling nonverbal
behaviors. Each of the eight individuals was paid $30.00

for the work carried out for this experiment.1

Training Of the Interviewers

Interviewers were trained along two separate lines:

(1) Basic Skills in how to conduct an interview.
(2) DevelOpment Of a standard nonverbal repertoire
to be used during the 15 minute interview.

 

1A great deal of appreciation is due to the eight inter-
viewers, Lynn Kosy, Nancy Pulk, FritO Suarez, Cindy Barber,
Mike Maurice, Ron Herrman, Dave Caldwell, and Jim Allen for
their OOOperation and help in the completion Of this
research.

30

Basic Skills for Conducting_Interviews

The interviewers met with the experimenter and the
goals Of the study were explained to them. Since each
interviewer was to carry out five separate 15 minute inter-
views, each interviewer had tO develop basic interviewing
skills. The experimenter discussed techniques Of non-
directive, non-threatening interviewing, and modeled an
interview for them. At that time there was discussion
concerning the interview modeled, with the experimenter
giving feedback as necessary. The interviewers then broke
into groups of two and began to practice interviewing each
other. Practice continued until all were sufficiently
skilled and at easein the role of the interviewer.

The content of each interview followed a similar
format. Each interviewer was given a standard list of
tOpics to discuss with each subject. The list included
the following items:

(1) School status and history: what is the 338

year in school, how did the S. choose M.S.U.,

how does the S. like M.S.U., what is the S.'s

major, what courses does the S. find interest-

ing, what plans or goals does the S. have, does

S. want to go on for further schooling?

(2) What is the past history of the S.: where

did S. grow up, what was it like growing up

there, how is it different from living in

E. Lansing, what types Of things did S. do

before coming tO M.S.U., what are S.'s work

experiences, how is M.S.U. the same or different
from prior schooling?

31

(3) What are S.'s interests in terms of sports,

recreation, hobbies, music, what does S. do

for relaxation, what does S. enjoy the most,

what activities is S. good at, what would S.

like to be able to develOp, or do?
This served as a basic outline Of tOpics for the interviewers.
Subjects were encouraged to talk at length about all or
any tOpics that interested them. There was no stipulation
that all Of the above tOpics were to be covered; the main
idea was for the interviewer to have a ready supply of
questions tO keep conversation flowing. By keeping the
interview on this rather superficial level, there was
little Opportunity for very personal kinds Of information
to be dealt with which might have caused the subject to

become unduly anxious, thus resulting in a possible increase

in nonverbal behaviors.

Nonverbal Interview Training

Each of the interviewers learned a standard, non-
verbal repertoire which, unlike the verbal aspects Of the
interview, was designed to effect the greatest possible
change in the nonverbal behavior of the subjects. The
repertoire consisted Of three five minute segments, the
first and third segment being the same. Thus, an A—B-A'

design.

Segments A. A'

During A and A', nonverbal behaviors were kept at

a bare minimum. This was especially true for the-areas

32

that were to be rated, namely the head and neck, arms and
hands, legs and feet, and postural and positional shifts.
The interviewer was instructed to assume a neutral stance
in terms of facial expression, smiling, and head nodding.
Although the experimenter had planned to control the amount
of eye contact during these first and last periods, the
interviewers found it impossible to do so and at the same

time conduct a natural, flowing interview.

Segment B

During the second five minute period, all inter-
viewers were to follow a similar schedule of nonverbal
movements. For each minute of this segment the following
schedule was to be maintained:

(1) two body shifts, which involved shifting of the
body posture or position in the chair.

(2) two leg shifts, which involved a major motion
with the legs, such as crossing, uncrossing,
extending forward, tucking them under the chair,
etc.

(3) six hand gestures, such as scratches, moving the
hand to the head or hair area, as well as normal
gestures used with speaking.

(4) unrestricted and amplified head nodding. Inter-
viewers were told to nod frequently to all
statements the subject made.

Specific nonverbal movements were not programmed for each
interviewer. During practice sessions it was found that
this was too hard to accomplish while still looking natural.
The requirement that only a certain number of each movement

be displayed enabled the interviewers to maintain a more

33

natural flow in both the verbal and nonverbal components of
the interview. It should be added that all movements were
practiced so that they did not look unusual or out of place.

Thus, during the first five minutes, the inter-
viewer was to maintain a motionless posture. After five
minutes had passed the interviewer was to proceed naturally
into the movements described above. After the next five
minutes, or the tenth minute of the interview, the inter-
viewers stopped moving, and again were to maintain a
motionless posture.

The interviewers practiced the varying conditions
of the interview with each other and the experimenter. In
addition, each interviewer was video-taped while in the
process of conducting an interview. These taped inter-
views were viewed by the group as a whole, and thus ample
feedback was available. Practice continued until all inter-
viewers could meet a similar criterion in terms of the
number of kinesic behaviors actually demonstrated during
the three five minute periods.

Of the four male interviewers, two of them each
saw five male subjects, and two of them each saw five
female subjects. Thus the male interviewers as a group
saw twenty subjects, ten males and ten females. Of the
four female interviewers, two of them each saw five male
subjects, and two of them each saw five female subjects.
Thus the female interviewers as a group also saw twenty

subjects, ten males and ten females.

34

Experimental Procedure

 

Each interview took place in the same 10 x 15 foot
room. The room was equipped with a microphone that hung
from the ceiling, and two television cameras which were
behind small windows on the wall opposite from where the
interviewer and subject sat. The subject and the inter-
viewer sat in identical chairs, each having two arms. The
chairs faced each other at a 20 to 30 degree angle, and
were about two feet apart. On the wall behind the sub-
ject's head was a large 12 inch clock with a sweep second
hand. This clock was within the line of sight of the
interviewer and was used by the interviewer to keep track
of the three five minute time periods. When subjects
arrived at the experimental room, the door was closed and
they were asked to wait outside until called. At that
time the interviewer would set the clock to the nearest
hour, half hour or quarter hour marking. This was done
to eliminate time keeping errors on the part of the
interviewer. Then the interviewer opened the door, asked
the subject to come in, and showed him which chair to sit
in. At that time each interviewer made the following
statement:

This is a study of different interview
techniques. Since I am a student, and not a
professional, many of the things we talk about
will not be of a personal nature. Behind those

windows are television cameras which will record
our entire interview. In addition this

35

micrOphone will record the things we say. If

you have any questions we will be happy to

answer them after we have finished this stage

of our research. Now let's see if we can get

started. . . .
The interviewer then began to go into the list of standard
questions, described above. Each interviewer used a clip-
board with a list of the interview questions so that it
was possible to refer to the list when questioning
became difficult. At the termination of the interview,
the interviewer made the following statement:

Well, I think that about does it for this

part of our study. If you will step into the

other room, we have a short form we would

like you to fill out. At that time you will

receive your credit and pay.
The subject was then led into another room, and was asked

to fill out the questionnaire, described below.

Questionnaire

 

This questionnaire was designed to assess several
factors. Questions 2, 3, and 4 dealt with Hypothesis II,
that is, the degree to which the subject reported he liked
the interviewer. Question 1 assessed an attitude towards
the interview in general, and questions 5, 6, 7, and 8
dealt with the needs of the subject to be close to and be
included by others. Question 9 dealt with the manner in
which subjects reacted to interviewers of the same, or
Opposite sex, while questions 10 and 11 dealt with the

apparent validity of the experimental situation. That is,

36

were the gestures and behaviors of the interviewers so
obviously different in the three experimental conditions
that the subjects reacted to them on a conscious level.
The questionnaire can be found in Appendix C. After the
subjects completed their questionnaire, they were given
their credit and pay. Questions concerning the nature

of the experiment were deferred, and a printed explanation

was sent to each subject who requested such information.

Rating of the Data Questionnaire

The questions from the questionnaire were scored
according to the following system. Each of the first
eight questions had seven possible answers. Answer "a"
was given a value of "1," answer "b" a value of "2" and
so on to answer "g" which was equal to "7." In this
manner it was possible to obtain a numerical score for
each of the eight questions.

Video Tapes of the Interview:
Ratinngrocedure

 

 

The video tapes of each interview portrayed the
interviewer and subject in a "split screen" arrangement.
Each tape was rated by pairs of raters.2 During the

actual rating procedure, only half of the screen was

 

2Very special thanks and consideration are due for
my six raters,Jody and Dave Stewart, Paul Richards, Kathy
Vinson, Lavon Leisz, and Whitney Emmerson, who performed
beautifully while suffering through incredible technical
difficulties.

37

visible, and the audio portion of the tape was not turned
on. In all cases, the half of the tape dealing with the
subject was shown first.

All raters worked in pairs. Each member of the
pair rated two separate variables at the same time, and
each member rated variables different from those of the
other member. In this manner, four variables could be
rated at the same time, with each rater rating variables
different from the other. Two separate sets of variables
were rated from each tape. The first set consisted of
the four categories of nonverbal acts defined above.
These were for the four major body regions:

(1) head and neck;

(2) hands and arms;

(3) legs and feet; and

(4) general body and postural shifts.
The second set of variables rated were those described in
Hypothesis II, with two exceptions. These are described

below.

(a) Positive head nods: these were up and down
motions of the head, clearly demarcated.

(b) Open arm postures: any position in which the
arms were not crossed. Any touching of the
hands and/or arms was considered to be crossed.

(c) Open leg postures: any position in which the
legs and/or feet were uncrossed. Any crossing
of the feet and/or legs was considered a
closed position.

(d) Forward leaning: any leaning towards the other
member, in which the subject's back was moved
away from the back of the chair.

38

(e) Space closing leg movement: leg movement or
position in which the subject's legs (or leg)
were extended in some way towards the inter-
viewer, as Opposed tO being in a neutral posi-
tion, or moved away from the interviewer.

It was impossible to Obtain reliable data concerning eye
contact and smiling behavior from the video tapes. Thus
these two variables described in Hypothesis II were not

included in the study.

Rating Of the Studied Variables

The four major kinesic variables (head and neck
movement, arm and hand movement, leg and feet movement,
and body and postural shifts) were rated for both frequency
and duration. In addition, both subjects and interviewers
were rated for these four variables. Raters used small
event recording panels which held two push buttons. When
these buttons were depressed, they set Off counters and
started clocks running. Thus, by holding the buttons down
for the duration Of a nonverbal act, frequency and dura-
tion measures could be Obtained. Raters viewed the tapes
in five minute segments. At the end Of each segment the
tape was stOpped, and the frequency and duration scores
were recorded. After the completion of a video-taped seg-
ment, the tape was rewound, and the raters viewed the same
segment again, but each rated different variables. Thus,
two independent measures were Obtained for each variable

for both frequency and duration. These scores were then

39

averaged, and the average was used as the raw data in the
statistical analyses.

The second set of variables was rated in a slightly
different manner. The subject only was rated for these
variables. In addition, there was no duplication of
rating functions. The reason for this was that the five
variables that were rated during this segment were so
Obvious, and the rater agreement during the training phase
so high, that the experimenter felt that duplication was
unnecessary. For the five variables rated in this segment,
only duration measures were taken for (1) Open arm pos-
tures, (2) Open leg postures, and (3) space closing leg
movement, while frequency and duration measures were taken
for (4) positive head nodding and (5) forward leaning.
These variables were rated in one 15 minute running of the
video tape, since the experimenter was not interested in

time period changes in these particular variables.

Training Of Raters

 

The experimenter met with the raters, explained to
them the concept of "nonverbal acts," and methods used to
discriminate between various acts. The rationale of the
experiment was not explained to them, nor was information
given as to who was a subject and who was an interviewer.
The experimenter made a sample video tape which contained

all the motions and acts that would be rated and the

40

raters practiced with these tapes, using the rating equip-
ment until they were able to reach a sufficiently high
level of agreement on a test video tape.

Each pair Of raters viewed a five minute test
segment, and rated it for the variables described above.
The tables in Appendix A give the actual scores and per-
centage Of agreement for each variable. To summarize
those results, Pair I had a range of agreement of 79-100
percent, with an average agreement score Of 93 percent.
Pair II had a range of 76-100 percent, with an average
agreement Of 90 percent. Pair III had a range of 78-100

percent with an average agreement Of 91 percent.

CHAPTER III

RESULTS

The design Of this study was double nested, in
that interviewers were nested within cells and subjects
nested within interviewers.

Data were gathered on the four major kinesic
variables for both frequency and duration and for inter-
viewers as well as subjects. These data generated
sixteen separate analyses Of variance with a 2 x 2 x 2
x 5 x 3 mixed-model design. The factors were sex of
interviewers (2) x sex of subjects(2) x number of
interviewers per cell (2) x number Of subjects per cell
(5) x time periods (3). Because Of the nested factors,
preliminary tests had to be made to determine the appro-
priate error terms. The analysis of variance summary
tables in Appendix B indicate which error terms were

used.

Results of theAnalyses
of Interviewer BehaviOr

Before studying the effects of the interviewer's
change in behavior on the subjects, it was first neces-

sary to test whether, in fact, the interviewers were

41

42

able to control and manipulate their own nonverbal behaviors.

Summary tables in Appendix B present the means obtained

for interviewer behavior. These results are now discussed

in detail below.

Interviewer Head Movement Frequency

There was a significant time period main effect
(F=15.9, df=2,8, p<.01). The means for time periods I, II,
and III were 15.4, 26.2, and 13.2, respectively. Newman-
Keuls analyses indicated that these time period means
differed in the predicted manner, with period I differing
significantly from periods II and III (p<.05), while

periods I and III did not differ significantly from each

other.

Interviewer Head Movement Duration

There was a significant time period main effect
(F=l4l.4, df=2,8, p<.OOl). The means for time periods I,
II, and III were 57.3, 188.3, and 55.4 seconds, respectively.
The Newman-Keuls analysis indicated that these means
differed in the predicted manner, with periods I and III
differing significantly (p<.05) from period II, while

they did not differ significantly from each other.

43

Interviewer Hand Movement Frequency

There was a time period main effect (F=18.0,
df=2, 8, p<.01). The means for time periods I, II, and
III were 6.0, 20.1, and 5.4, respectively. The Newman-
Keuls analysis indicated that these means differed in the
predicted direction, with periods I and III differing
significantly from period II (p<.05), while they did not

differ significantly from each other.

Interviewer Hand Movement Duration

There was a time period main effect (F=l66.0,
df=2,8, p<.001). The means for time periods I, II, and
III were 41.4, 188.1, and 42.6 seconds respectively.
Newman-Keuls analysis indicated that these means differed
in the predicted direction, with periods I and III differ-
ing significantly from period II (p<.05), while they did
not differ significantly from each other.

For this variable there were significant nesting
effects, resulting in a significant time period x sex Of
subject interaction (F=7.57, df=2,8, p<.05). Table 1
presents the means which illustrate this interaction.

Newman-Keuls analyses revealed that these means
differed as expected across time periods, with periods I
and III being significantly different (p<.05) from period
II, while they did not differ significantly from each

other. In addition, it was found that interviewers who

44

saw male subjects differed significantly from interviewers
who saw female subjects in time periods I and III. Closer
examination Of the means represented in Table 1 indicates
wide individual differences between interviewers. The
extremely high durations recorded for Male Interviewer

NO. l are sufficient to throw Off the total mean, thus
causing a significant interaction forsmx Of subject x time

period.

Table 1. Means of interviewer hand movement duration.

 

 

Interviewers Time Periods
I II III

 

Male Subjects

Male I No.1 180.0 263.6 192.6
Male I No.2 7.2 69.8 0.2
Female I No.1 51.4 171.6 54.4
Female I No.2 12.6 207.2 9.6
Female Subjects
Male I No.3 63.4 250.6 54.8
Male I No.4 13.8 199.4 12.4
Female I No.3 6.4 190.2 6.0
Female I No.4 1.2 152.8 9.0

 

There was a significant three-way interaction for
sex of interviewer x sex Of subject x time period (F=5.6,
df=2,8, p<.05), also stemming from significant nesting

effects. These means are also represented in Table 1.

45

An analysis Of simple effects was carried out for this
interaction and is summarized in Table 2. Other than the
expected effects due to time periods, there were no other
significant differences. Newman-Keuls analyses for the
four significant time period interactions revealed that
the means of the time periods differed in the expected
direction, with periods I and III differing significantly
from period II (p<.05) while they did not differ signifi-
cantly from each other. Examination Of the F statistics
displayed in Table 2 reveals that the cause for the signi-
ficant three-way effect stems from the biasing effect of

Male Interviewer NO. 1.

Interviewer Leg Movement Frequency

There was a significant time period main effect
(F=52.6, df=2,8, p<.001). The means for time periods I,
II, and III were 2.53, 19.15, and 2.85, respectively.
Newman-Keuls analysis of the time period means revealed
that the differences were in the predicted direction with
periods I and III differing significantly from period II
(p<.05), while they did not differ significantly from each

other.

Interviewer Leg Movement Duration

There was a significant time period main effect

(F=l7.0, df=2,8, p<.01). The means for time periods I,

46

Hoo.vm
«

 

 

III NH m.mmHv~ HOHHm
m.o H m.mmmm HHH .Hmm msHH x H mHmsmH :Hsqu muommnsm
H.o H o.mnm~ HH .Hmm mEHa x H mHmEOm OH£HH3 muomnnsm
m.o H v.mmmh H .Hmm mEHB x H OHOEOH GHnqu muommnsm
m.H H ¢.mmvmm HHH .Hmm mEHB x H OHmz stqu muomnnsm
H.H H m.mmmmm HH .Hmm OsHe x H OHM: cHnuHs muomflnsm
H.H H m.~wmbm H .Hmm msHH x H mez cHnqu muumnnsm
m.o H ¢.vmmw HHH .Hmm OaHB x m OHOEOH cquHz mHm3OH>HOucH
¢.m H m.m~mmm HH .Hmm mEHB x m OHMEOH cHsuHs mH03OH>HmucH
m.o H o.HHHNH H .Hmm OEHB x m mHmEOm GHSUHz mHm3OH>HmucH
>.H H m.mmHo¢ HHH .Hmm mEHB x m mmHmz cHnuH3 mumzmH>HmucH
m.o H o.ooo¢ HH .Hmm mEHB x m mHmz stuH3 mum3mH>HOucH
¢.H H N.mmmvm H .Hmm OEHB x m mHmz cHnuH3 mnm3mH>HmusH
III m m.MHhH HOHHm
«¢.mH~ m H.Hnnmmm m mHmsmH x H mHmsmm cHauHs mooHHmH msHe
«m.HmH m o.vmmmHm m mHmz x H OHmEOm cHnqu mOOHHOm mEHB
«v.hnm N m.-mmhv m mHmEmm x H mez cHnuH3 mOOHHmm mEHB
«v.Hv N o.NvOH~ m OHM: x H mHmz cHnuHs mOOHHmm mEHB

m HO m2 OOOMHHO> HO wousom

 

 

.GOHDMHOO usmEm>OE pawn Hw3mH>HmucH HOH OOHHOQ 08H» x HO3OH>HOHGH HO xmm x
pomnnsm HO xmm HO soHuomHmucH >m3|wmnnu on» HOH mpommmm mHmEHm HO mumfifidm

.N OHQMB

47

II, and III were 6.4, 102.5, and 7.6 seconds, respectively.
Newman—Keuls analysis of the time period means revealed
that the differences were in the predicted direction,

with periods I and III differing significantly from period

II (p<.05), while they did not differ significantly from

each other.

Interviewer Body Movement Frequency

There was a significant time period main effect
(F=43.9, df=2,8, p<.001). The means for time periods I,
II and III were 0.25, 5.58, and 0.38, respectively.
Newman—Keuls analysis of the time period means revealed
that the differences were in the predicted direction, with
periods I and III differing significantly from period II

(p<.05), while they did not differ significantly from each

other.

Interviewer Body Movement Duration

There was a significant time period main effect
(F=35.0, df=2,8, p<.01). The means for periods I, II,
and III were 0.38, 8.93, and 0.53 seconds, respectively.
Newman-Keuls analysis Of the time period means revealed
that the differences were in the expected direction, with
periods I and III differing significantly from period II

(p<.05),while they did not differ significantly from each

other

48

Summary Of Interviewer Behavior

In terms Of time period means the results are quite
clear. In each nonverbal category, the interviewers per-
formed as they were trained. Analyses revealed that the
time period differences were in the predicted direction,
that is with periods I and III differing significantly
from period II, while they did not differ significantly
from each other. This does not imply that all interviewers
performed perfectly in all cases, but that on the whole,
they did perform well, and in the expected direction.

Only on one variable, hand movement duration, did indivi-
dual interviewer differences give rise tO confounding

interactions.

Results of the Analyses of Subject Behavior

 

Subjects were rated and scored on the same kinesic
variables as the interviewers. Summary mean tables and
analysis of variance summary tables are found in Appendix
B. The results of these analyses are now discussed in

detail, below.

Subject Head Movement Frequency

This analysis revealed a significant time period
main effect (F=12.1, df=2,8, p<.01). Newman-Keuls analysis
revealed that these means did not differ in the predicted

direction, but that period I differed significantly from

49

periods II and III (p<.05). Periods II and III did not
differ significantly from each other. The means for time
periods I, II, and III were 27.7, 24.5, and 23.4, respec-

tively.

Subject Head Movement Duration

There were no significant main effects revealed by
this analysis, although there was a significant interaction
between sex Of interviewer x sex of subject (F=4.5, df=1,32,
p<.05). The means of this interaction are displayed in

Table 3.

Table 3. Sex of interviewer x sex Of subject interaction
means for subject head movement duration.

 

 

 

Subjects Interviewers
Male Female

Male 147.9 185.6

Female 205.7 178.0

 

An analysis of the simple effects revealed that the mean

for male subjects seeing male interviewers was significantly
different (p<.05) from the mean of female subjects seeing
male interviewers. There were no other significant dif-
ferences. In other words, females who saw male interviewers
had more total head movements than males who saw male inter-
viewers. Subjects seeing female interviewers did not show

such differences.

50

Subject Hand Movement Frequency

There was a significant time period main effect
(F=17.4, df=2,72, p<.01). The means for time periods I,
II, and III were 19.9, 16.9, and 15.9, respectively.
Newman-Keuls analysis of these means revealed that they
did not differ in the predicted direction, but that period
I differed significantly from periods II and III (p<.05),
while periods II and III did not differ significantly

from one another.

Subject Hand Movement Duration

There were no significant effects demonstrated in
this analysis. The main effect of time period approached
significance, however (F=3.0, df=2,8, p>.10). Additional
analyses were run on this variable in which two of the
eight interviewers were eliminated from the analysis.

These two interviewers seemed to do the poorest job of
keeping within the schedule of programmed movements, and

it was thought that the additional movements they made might
have affected the overall results for hand movement duration.
Thus the data from six interviewers were run in a 2 x 2 x 5
x 3 mixed-model design in all four possible combinations.

The analysis Of variance summary is presented in Appendix

B. For each analysis, there was a significant time period

main effect, as presented in the table below.

51

Table 4. Results Of recombining data for subject hand
movement duration.

 

 

 

Recombination
Of Time Periods
Data I II III F df
Recombination I 157.3 177.8 176.1 3.7* 2,32
Recombination II 163.3 180.8 190.0 4.4* 2,32
Recombination III 136.5 167.9 163.7 7.3** 2,32
Recombination IV 142.9 170.9 177.5 7.4** 2,32
*p<.05
**p<.01

Newman-Keuls analyses in each case indicated that period I
differed significantly (p<.05) from periods II and III,
while periods II and III did not differ from each other.
The reason why data were recombined only once was due to
the fact that the interviewers tended to perform.most poorly
on this particular variable. One may recall that when the
results for the interviewers were reviewed, hand movement
duration was the only variable that had significant effects
other than time period main effects. Reviewing the recom-
bined analyses for each Of the other variables revealed
that generally, the trends Of the main analyses were

reflected in each of the recombined analyses.

Subject Leg Movement Frequency

The main effect for sex of interviewer approached
significance (F=4.0, df=1,36, p<.10). The means indicate

that subjects who saw male interviewers had significantly

52

more leg movements (mean frequency=12.5) than subjects with

female interviewers (mean frequency=8.08).

Subject Leg Movement Duration

As above, the main effect for sex of interviewer
approached significance (F=3.08, df=1,36, p<.10). Subjects
who saw male interviewers had significantly longer periods
Of leg movement (mean duration=69.5 seconds) than persons

who saw female interviewers (mean duration=33.2 seconds).

Subject Body Movement
Frequency and Duration

There were no significant effects revealed by

either Of these analyses.

Summary Of Subject Behavior

Although there was a significant time period main
effect for subject head nodding frequency, and subject hand
movement frequency, in both cases the effect was not in the
predicted direction. Hypothesis I called for an increase
in the nonverbal behavior Of the subject during period II
and then a subsequent decrease in period III. For these
two variables, above, frequency means were highest in
period I, while periods II and III were lower in frequency
and about equal.

Changes in the duration Of the subject's hand

movements were only minimal except for the additional

53

analyses in which the two interviewers who performed poorly
on this variable were drOpped from consideration. In

these subsequent analyses, however, the duration means did
not follow the pattern of the hand and head frequency
means, but followed the predicted pattern, in that period
II means were higher than period I means.

Other findings included a significant interaction
effect between sex Of interviewer x sex Of subject in
head movement duration. It was found that males seeing
male interviewers displayed less head movement as compared
to females seeing males.

For leg movement frequency and duration there were
marginally significant effects (p<.10) related to the sex
Of the interviewer. Subjects seeing male interviewers
tended to have more leg movement than those subjects seeing

female interviewers.

Questionnaire Data: Correlations Of
Responses with Five Kinesic Variables

 

Scores from eight separate questions from the
questionnaire were correlated with five secondary kinesic
variables (positive head nodding, forward leaning, Open arm
postures, Open leg postures, and Space closing leg movement).
Only one correlation was significant at the .05 level.

This was between Question 8 (Are you an outgoing person?)

and positive head nod duration, r=.33.

54

Other correlations that were significant at the
.10 level were Question 3 (If the interviewer lived in or
near your dorm, or building, what chance would there be of
becoming friends with the interviewer?) with legs closed
(crossed). For this variable the correlation was r=-.29,
p<.07. One other minimally significant correlation was
Question 4 (How do you think the interviewer felt about
you?) with legs closed (crossed) which yielded a correla-
tion of r=-.30, p<.06.

It must be noted that the number of statistically
significant correlations was not above chance expectations.
For the most part, the correlations between the five secon-
dary kinesic variables and the eight questions revealed no
significant results. This suggests that the five variables
chosen for study were not related in any way to self reports
of "degree of liking" or "degree of positive regard."
Forward leaning, one Of the five variables, occurred with
little and irregular frequency. In addition, space closing
leg movement and leg closing showed little variability, in
that a subject would usually assume a leg position, e.g.,
stretched out, and/or crossed, and would then keep that 'F
position for the remainder Of the fifteen minute interview.
In other words, in an experiment of short duration, which
does not stress promotion or creation Of extreme psycho-
logical states, variables which entail measures Of body
areas that do not move a great deal will probably not tend

to be meaningful and/or significant.

55

The results Of these correlations indicated that if
variables that correlate with positive regard or liking do
indeed exist, such as eye to eye contact (Exline, 1963),
they are not the five variables chosen in the present
study. It was unfortuante that eye contact and smiling
behavior could not be rated (due to the technical limitations
Of the video system utilized), for these variables seem to
be more related to variables such as those presented in

the eight questions.

Analysis Of the Eight Questions

A 2 x 2 x 2 x 5 mixed-model analysis Of variance
was run on each Of the eight questions. The summary analysis
Of variance tables of each analysis can be found in Appen—

dix B. The results of each analysis are as follows:

Question One: "How do you feel about the interview you
have just finiShed?“

 

There was a marginally significant sex of actor x
sex Of subject interaction revealed by this anaIysis (F=3.13,
df=1,36, p<.10). The means Of this interaction are found
in Table 5. Tests of simple effects revealed that the
mean Of male subjects who saw male interviewers (5.3)
differed significantly (p<.05) from the means of male
subjects who saw female interviewers (6.2) and female

subjects who saw male interviewers (6.3). That is,

56

subjects who saw male interviewers liked the interview
situation significantly less than females who saw male

interviewers, and males who saw female interviewers.

Table 5. Sex of subject x sex of interviewer interaction
means for question one.

 

 

 

Subjects Interviewers
Male Female

Male ' 5.3 6.2

Female 6.3 6.1

 

Question Two: "How did you feel about thegperson who
interviewedyou?1r

 

Analysis Of this question revealed a significant
main effect for sex of subject (F=4.7, df=1,36, p<.05).
The mean for male subjects was 5.50, and the mean for female
subjects was 6.10. Females tended to report higher liking
towards their interviewer, regardless Of sex, as compared

to male subjects.

Question Three: "If the interviewer lived in or near your
dorm, or building, what chance would
there be Of becoming friends with the
interviewer?-1r

 

 

There were no significant differences revealed by

this analysis.

57

Question Four: "How do you think the interviewer felt
aboutyou?"

 

 

There was a marginally significant main effect for
sex Of interviewer on this variable (F=4.0, df=1,36, p<.10).
The mean for male subjects was 4.7, and the mean for
female subjects was 5.3. Subjects, regardless of sex,
felt that female interviewers liked them more than did the

male interviewers.

Question Five: "How much doyyou like being emotionally
close to other persons?“

 

 

There were two significant main effects in this
analysis. One was due to the sex of the interviewer (F=8.l7,
df=1,36, p<.01). The mean for subjects who saw male inter-
viewers was 5.65 while the mean for subjects who saw female
interviewers was 6.45. Subjects who saw female interviewers,
regardless Of sex, reported that they liked being more
emotionally close than those subjects who saw male inter—
viewers.

The second main effect was for sex of subject
(F=8.l7, df=1,36, p<.01). The mean for male subjects was
5.65, while the mean for female subjects was 6.45. These
means indicate that female subjects, on the whole, reported
that they enjoyed being more emotionally close than did

male subjects.

58

Question Six: "When you meet strangers in a social
situatiOn (for exampleL in a claserat
a party, at the grill),‘how do you feel?"

 

There was a sex Of subject main effect (F=6.75,
df=1,36, p<.05). The mean for male subjects was 4.95,
while the mean for female subjects was 5.90. These means
indicate that females reported they enjoyed entering into
new social situations, or enjoyed meeting persons in these

situations significantly more than did male subjects.

Question Seven: "If you enter a room, and a group is
laughing and taIking, how strong a
desire do you have to enter the group?"

 

 

 

 

There was a sex of subject main effect for this
variable (F=6.04, df=1,36, p<.05). The mean for male sub-
jects was 4.85, while the mean for female subjects was
5.90. These means indicate that females reported more
willingness to enter into strange social situations than

did males.

Question Eight: "Are you an outgoing person?"

 

 

There was a significant sex of subject for this
variable (F=22.34, df=1,36, p<.01). The mean for male
subjects was 4.00, while the mean for female subjects was
6.05. These means indicate that females tend to report
themselves as more outgoing than do males, as the results

Of the previous two questions would also indicate.

59

Summary and Discussion
Of Questionnaire Analyses

The results of these analyses indicate that there
were definite sex differences in the manner in which sub-
jects answered these questionnaires. In all cases in
which these differences occurred, females always scored
higher, that is, more positively than males. Females
reported themselves to be more Open (question 8) more at
ease in a strange social situation (question 6), more
eager or willing to engage in social interchange (question
7), more willing to admit emotional closeness (question 5),
and felt more strongly that the interviewer liked them
(question 2).

The results Of these analyses have importance for
several reasons. First, in terms Of the original hypo-
theses of this study, these results tend to disprove the
aspect Of Hypothesis II which stated that persons scoring
higher on the questionnaire variables would show a greater
degree Of influence in terms Of the four major kinesic
variables discussed above. There were no significant sex
differences reported for the four major kinesic variables,
and thus this aspect of Hypothesis II was not confirmed

in view of these results.

60

Overall Summary Of Results

Drawing together these results in terms of the
three original hypotheses, the following Observations can
be presented:

Hypothesis I was not supported by the data collected.
Although there were definite changes in the behavior of the
subjects, these changes were not in the direction predicted
by the first hypothesis.

Hypothesis II was not supported by the results of
this study. Although some Of the results pointed in the
direction predicted by the second hypothesis, the majority
of these results indicated that the degree of liking, or
positive regard as measured by the questionnaire,was not
related to the nonverbal behaviors specified, nor was it
related to the degree to which subjects were influenced by
the nonverbal manipulations Of the interviewers.

Hypothesis III was notsupported by the data
collected. There were no significant sex differences
related to the rated kinesic variables. There were signi—
ficant sex differences found in the responses to the ques-
tionnaire, but these results were not predicted by this

third hypothesis.

CHAPTER III

DISCUSSION

Overview

 

This study develOped out Of a personal interest in
the nonverbal aspects of communication. To a therapist,
or any person who deals on an intimate level with other
human beings, clear and concise understanding Of behavior
is an ever present goal. Historically, man has been
primarily concerned with verbal communication in its
spoken and written forms. Perhaps this interest in the
verbal aspects of communication was due to the ease of
its study. That is, one can easily count the words of a
written page, or Obtain agreement about what a person has
said. Or, perhaps there are greater motives for studying
the verbal aspects of human behavior, for these qualities
are among the few that set man apart from the rest of non—
human world; the profoundly complex ability to communicate
thoughts and feelings.

There are times, however, when traditional modes
of communication seem somewhat inadequate and ineffectual.
And it is at these times when other aspects of behavior

tend to complete the gestalt of communication. These other

61

62

aspects are, of course, the nonverbal components Of behavior.
In the introduction to the present study, it was shown how
others, conducting research in this same area, have dealt
with the field of nonverbal behavior. It was shown that
persons have the ability to communicate feelings and
thoughts through nonverbal channels. In addition, many
aspects of nonverbal communication seem to be affected by
differences in sex, affiliative states, and internal states
Of anxiety, fear, etc. Studies mapping and coding minute .
aSpects Of behavior have been conducted in order to bring
meaning into vague nonverbal patterns by trying to tie
certain patterns to certain referents. The purpose Of such
work is to establish a language Of nonverbal communication
much like a verbal language. All in all, many facets Of
nonverbal behavior have been studied, providing a large
foundation for later research to build upon.

Surprisingly, the impetus for the present study
did not come from the studies Of nonverbal behavior, but
from a study of verbal behavior in which it was shown that
manipulations in the pattern of communication Of one indi-
vidual brought about certain changes in the communication
patterns Of other individuals. The present study set out
to examine whether certain manipulations of nonverbal
variables in one individual, an interviewer, would indeed
bring about changes in the behavior of another individual,

a subject.

63

In the discussion that follows, this writer reviews
some Of the methodological aspects of the present study,
and then proceeds to discuss the implications of the find-
ings Of the present study, both in terms of present

knowledge and in terms of future or planned research.

Methodological Considerations

 

A major premise Of this study was the assumption
that individuals could be trained to manipulate their non-
verbal behavior while interacting with another in a more
or less normal manner. In determining the validity Of
this assumption, two separate sets Of factors had to be
taken into account. The first was the actual validity, in
terms of the raw data collected by the raters, and the
second was the degree to which the interviewers were
believable in the actual interview situation. The results
Of the analyses Of the nonverbal behavior of the eight
interviewers indicated in general that they did success-
fully perform in the expected manner. However, closer
examination revealed considerable individual differences,
at least with one of the eight interviewers, differences
which may well have affected the overall data. It may be
added that overall, these differences do not appear signi-
ficant, but since each interviewer saw one eighth Of the
population, the effects of individual differences seem tO

mount rapidly. Thus, while it does seem that individuals

64

can be trained to manipulate behavior in a prescribed
manner, one must exert a good deal Of care in the final
selection of these individuals, or improve training until
"control" is achieved.

How believable were the interviewers themselves?
In other words, did the subjects really believe that the
interviewers were doing what they said they were doing,
i.e., "conducting an experiment on interviewing tech-
niques," or were the shifts in nonverbal movement so
Obvious that the subjects could tell what was going on,
and thus respond to the manipulations on a conscious level?
Several questions were built into the questionnaire which
assessed the subject's response to the interviewer and the
study itself. Question 10 stated "If there was anything
that bothered you about this experience, we would appreci-
ate your comments,“ and Question 11 asked, "What things
could the interviewer have done to make you feel more at
ease?" It was felt that if things appeared too contrived,
or obvious, that the subjects would report these kinds Of
Opinions in response to the questions. Out Of all subjects,
only one reported that he was consciously aware Of the
interviewer's movements, and felt that these were a part
of the experiment. Thus, it would seem that while one is
engaged in conversation with another, a very wide range
of nonverbal behaviors can be tolerated at a less than

conscious level. Six subjects reported that their

65

interviewer seemed somewhat tense and nervous. Whether the
subjects were responding to the interviewer's nonverbal
patterns, and using these to define the nervous behavior,
or whether the nervousness was ascertained in other ways

is not known. It did seem that several Of the interviewers
were somewhat nervous during their first interview, but
this seemed to drOp out with later subjects.

Since the subjects were in an Obvious laboratory
situation, i.e., they were shown the micrOphone and the
video cameras, this writer was interested in what effect
this might have had upon the results Obtained. Only five
subjects reported feeling somewhat nervous about the record-
ing process, and all five stated that this nervousness
drOpped out soon after the interview started.

Finally, one subject out of the forty commented
on the fact that his interviewer kept looking at the large
clock (to keep track Of the five minute intervals), and
seemed to be quite angry at the interviewer for being so
wrapped up in his clock watching. This subject did not
seem aware Of any change in nonverbal behavior.

The above findings tend to indicate that the inter-
viewers and the experimental situation did seem to have
validity, both in terms Of the raw data collected on the
interviewers' behavior, and the apparent validity of the
exPerimental situation as determined by the comments Of

the subjects.

66

Comments Concerning the
Nonverbal Variables Rated

From the onset Of this experiment, the design was
such that the variables to be studied would be very broadly
based. All hand movements were grouped into one type Of
variable called hand acts, all head movements were called
head acts, etc. The reason for such a definition Of
variables was due to the supposition that if nonverbal
behavior was indeed subject to influence, it should be
noticeable in these broad categories. Upon reflection, it
is the Opinion Of this writer that part of the reason why
certain variables such as leg movement were not significant
was due to the fact that either the wrong variables were
rated, or that many separate types of variables were rated
together as one. Perhaps with some areas, such as hand or
head movements, this was not so important. However, it
was also felt that the manner in which hand movement was
rated detracted from a clear demonstration of the desired
effect. Ekman and Friesen (1968) mention that they studied
eight separate hand and leg acts, acts which were decided
upon after EQEE Egg Observation of the interaction patterns
Of one individual. This specific rating procedure would
not be suitable for a study such as this, although it
would be possible to View several samples Of interaction,
and determine from these samples typical modes of inter-

action that could be charted for frequency and duration.

67

In this way it might be possible to determine specific
interactive influences,if any, as Opposed to the more
general trends found in this study. It may be assumed

that such pre-experimental behavioral Observations would
have resulted in the drOpping of the body movement variable
from the present study. This variable occurred with such
little frequency that very little could be said about its

capacity for nonverbal influence.

Comments Concerning the
Length Of the Interview

For several reasons, the somewhat arbitrary period
Of fifteen minutes was chosen as the standard length Of
each interview. In addition, experimental time period
lengths were restricted to five minutes. Matarazzo, et.
a1. (1968), during the laboratory phase Of his verbal
influence work, utilized a 45 minute interview and three
fifteen minute intervals. It is unknown to this writer
how many different variations Matarazzo utilized before
he came up with the three fifteen minute periods. It is
the Opinion of this writer that variations in the time
periods chosen could indeed add to or perhaps detract from
the results Obtained. For example, are five minutes enough
time for the actions of the interviewer to influence most
fully the actions Of the subject? Or, would longer time

periods merely reflect the present findings? Were the

68

results Obtained in this study a mere artifact Of the time
periods themselves, artifacts which would disappear if
conditions were changed? These questions cannot, of
course, be answered by the results of the present study.
Yet they are valid questions that could usefully be
answered in other similar research.

In addition to the length of the time periods, the
question of order is a particularly important issue. Since
there were no real significant differences between time
period II and time period III means, would periods Of
activity during the first time period result in no
significant effects at all? Or might a change of the order
of manipulations, coupled with longer time period lengths,
show different significant patterns? This particular issue

will be discussed below.

Racial Aspects of Interviewer
and Subject POpulations

In order to assure homogeneity of interviewers,
all were Caucasian. One female interviewer was Mexican-
American, and thus had slightly darker features than the
others, but Spoke with no accent or any "non-standard"
idioms or inflections. The subject population, which was
a random sample Of students enrolled in introductory
Psychology courses, included several black students.

Since the total number of blacks was only about 20% Of

69

the total pOpulation, separate analyses of black vs. white
behavior were not carried out. Efron (1941) reports
differences in gesturing behavior due to differences in
ethnic backgrounds, and thus it would be interesting to
assess differences, if any, in black and white populations
in response to interviewers of different racial character-
istics. Since the design of the present experiment did
not take into account such differences, it is impossible
to speculate as to the effect Of such manipulation from
the results Of this research. However, future research

may profit from such procedures.

Implications Of the Findings
Of the Present Research

 

 

Taking into account the above-mentioned limitations
and characteristics Of the present research, there are
basic findings which did result from the experimental

manipulations and these implications shall now be discussed.

Time Period Effects

Birdwhistle and Scheflen present a strong case for
an actual language Of nonverbal, interpersonal behavior.
In his Introduction to Kinesics, Birdwhistle sets down
several basic rules to be considered when evaluating

kinesics, or nonverbal behaviors:

70

A. NO motion is a thing in itself. It is always
a part Of a pattern. There is no "meaningless"
motor activity.

B. Until otherwise demonstrated, body motion patterns
should be regarded as socially learned.

C. NO unit of motion carries meaning per s2, Meaning
arises in context.

(Birdwhistle, 1952, p. 10)
Birdwhistle makes a strong point for studying nonverbal
behavior within the framework Of the total gestalt in which
it occurs. In addition, behaviors and patterns of behavior
are not seen as mere chance occurrences, but as learned
patterns emitted in response to specific stimuli.
Scheflen, working out Of the same frame of reference
as that of Birdwhistle, makes the following statement about

nonverbal communication:

Communication, then, includes all behaviors
by_which a group_forms, sustains, mediates,
corrects and integrates its reIationships. In
the fIOw Of an interaction, communicative beha-
viors serve to give continuous notification of
the states of each participant and Of the rela-
tionships that Obtain between them. Individuals
growing up in a culture must learn these com-
municative behaviors and perform them correctly
if they are to be comfortable, perhaps even if
they are to survive in their world.

(Scheflen, l964b,p. 318)

 

Again, the emphasis is upon the total gestalt Of the pat-
terns Of nonverbal behavior used in the communicative
process. Behaviors are used, Scheflen states, not only
to communicate various referents, but also to establish

the on going state Of affairs between individuals.

71

Haley (1963) deals with the concept Of communica-
tion in terms of a reciprocal process between individuals:

When any two people meet for the first time
and begin to establish a relationship, a wide
range Of behavior is potentially possible between
them . . . As the two people define their relation-
ship with each other, they work Out together what
type of communicative behavior is to take place
in this relationship. From all the possible mes-
sages they select certain kinds and reach agree-
ment that these shall be included. This line
they draw which separates what is and what is not
to take place in a relationship can be called a
mutual definition of the relationship. Every
message they interchange by its very existence
either reinforces this line or suggests a shift
in it to include a new kind of message. In this
way the relationship is mutually defined by the
presence or absence Of messages interchanged
between two peOple.

(J. Haley, 1963, p. 6-7)

As Haley suggests, communication is an on going process
that is used by individuals to define their relationship
to one another.

The present study was designed to see whether an
interviewer could manipulate the behavior of a subject via
the manipulation Of his own nonverbal behaviors. Looking
at the interaction from a slightly different vantage
point, the interviewer was effecting changes in the
"normal" or "expected" communication pattern of an inter-
view situation. As a result, the subject was forced to
respond to an individual who was noncommunicative on the
verbal level.

The results of this study indicated that there were

significant differences in the frequency and duration Of

72

certain nonverbal behaviors. It may be postulated that
these differences were related to the nonverbal manipula—
tions Of the interviewers.

As will be recalled, the frequency and duration
measures did not change in the same direction, as pre-
dicted. Instead frequency scores for head and hand move—
ment were significantly higher during period I, while
means for head and hand duration revealed a pattern in
which period I means were lower than period II and III
means. These results suggest that frequency and duration
measures are not interchangeable, but can have quite
different characteristics. It could be speculated that
duration measures are somewhat more useful in represent-
ing the nonverbal behavior that transpired in the inter-
view situations. For example, a hand movement of one
second has the same frequency Of occurrence as a con-
tinual hand motion of twenty to thirty seconds. Yet it
is clear that a very different amount Of nonverbal
behavior has taken place. This does not imply that
frequency scores should, or can be ignored, for while the
frequency scores do not depict the total interaction situ-
ation, they may be more representative of "universal" or
"micro-momentary" states.

The most clearly significant results Obtained in
this study were those dealing with head and hand frequency

scores. Yet in the time periods in which frequency means

73

were highest, duration means tended to be the lowest. What
could be happening is that the emotional state Of the
interviewer during the initial period Of non-movement
caused the initiation of significantly more head and hand
movements, although these movements were Of shorter dura-
tion than movements occurring later in the interview.
Perhaps the subject, responding on some level to the
inactivity Of the interviewer, felt more "aroused" or
"anxious" and thus initiated movements of his own more
frequently than he normally would. Finding that the
interviewer did not reciprocate these movements, there
was seemingly no "reinforcement," and as a result these
movements were Of shorter duration.

During period II, the subject was "reinforced"
by the reciprocal movements Of the interviewer, and thus
did not have to make as many initiating movements. Thus,
the frequency rate fell during this time period. In
addition, it could be speculated that because of the
reciprocated movements of the interviewer, the subject
felt more comfortable, and thus produced longer gestures
to match those Of the interviewer.

Returning to the comments Of Birdwhistle, Scheflen,
and Haley, mentioned above, one can apply their comments
to the results Obtained from the present study. In dis-
cussing the relevance Of communication theory in human

interaction, Haley feels that during the initial states

74

Of a relationship, the members spend time and effort in
defining the relationship itself. Haley contends that
individuals communicate on all levels, and certainly
nonverbal interaction would fit into his definition. In
other words, individuals utilize nonverbal means to estab-
lish their relationship. This is much in line with
Scheflen's statement that "Communication . . . includes
all behaviors by which agroup_forms, sustains, mediates,
corrects, and integrates its relationship_." (Scheflen,
l964b,p. 318). Returning to the results of the present
study, it is the contention Of this writer that period I
correSponds to the time in which the members Of the inter-
view were beginning tO formulate and establish their
relationship. It can be assumed that this "establishing"
was taking place on many different levels. Whether the
pattern established on all levels was the same as that
established for the nonverbal levels was not studied by
the present research. The nonverbal pattern, however,
seemed to indicate that during the first period, the
subject was trying to establish some sort Of nonverbal
relationship with the interviewer, but was thwarted in

his efforts. In the second period, however, when the
interviewer began to move, and thus reciprocated the sub-
ject's movements, some sort of nonverbal relationship came
into being. Thus, it would seem that the subject no

longer felt compelled to establish the relationship,

75

since it was already established, and therefore the
frequency rate fell during period II.

It will be recalled that period III scores for
both frequency and duration were very similar to the
scores Obtained in period II. This would indicate that
the final manipulation Of the interviewer had no apparent
effect upon the nonverbal behaviors Of the subjects.

These results are not, however, incongruent with the

model Haley sets forth. Returning to the interaction
process, Haley feels that once the relationship has been
established, as would be the case Of period II, there is
then more room for flexibility and variability on the

part of the members Of the relationship. Thus, when the
interviewer stOpped moving during the third period, the
subject did not feel compelled to reestablish the relation-
ship, since that relationship already existed.

This explanation collapses the relationship process
into a very short fifteen minute interval. This does not
imply that the relationship between subject and inter-
viewer had been worked out on all levels. Nor does it
imply that merely because some sort Of relationship had
been established during the fifteen minute interval that
this relationship was then impervious to change. Although
there was no apparent change in the subjects' behavior
during the third time period, it must be mentioned that

these periods were not very long, and it might be that

76

longer intervals of movement or nonmovement, fifteen
minutes, for example, might have caused changes which would
require subjects to redefine their relationship with the
interviewer several times.

The preceding discussion Offers one interpreta-
tion Of the data. However, due to the nature of the
results, other interpretations and explanations should
also be considered. One such explanation is that the
results Obtained across the fifteen minute time period
had nothing to do with the manipulations Of the inter-
viewers. It may be that in any strange, or new situation
individuals are somewhat nervous at first, and until they
feel more at ease with the situation, they emit more non-
verbal behaviors such as those studied in the present
research. This would imply that no matter what the inter-
viewer did, there would be higher frequency scores during
the first few minutes Of any interivew. A simple way Of
checking this hypothesis would be to change the order Of
the experimental manipulation, or to run control inter-
views in which there would be no manipulations at all.

One Of the basic assumptions of the present
research was that there exists a language of nonverbal
behavior, and that relationships that exist in nonverbal
language parallel relationships found in verbal language.
It was assumed that manipulations Of interviewer behavior

would cause reciprocal changes in behavior on the part of

77

the subjects. As the results clearly state, this was not
the case in the present research. However, the implications
Of these results are not clear. It may be that there are
not parallel languages Of verbal and nonverbal behavior.

Or, it may be that the manipulations utilized in the present
study did not correctly utilize nonverbal language channels.
There is very high agreement in terms Of verbal language

as to what is being said, and communicated. However, in
nonverbal language, these issues are not as clear. Thus,
the interviewers, while they were increasing the frequency
of certain gross categories of nonverbal behavior, may

have been sending conflicting messages in terms Of the
specific nonverbal referents that they used. For example,
while an interviewer increased the degree of hand movement
displayed, he might have begun to utilize specific non-
verbal referents (Scheflen, 1963) which tended to shut

down the nonverbal communication Of the subject. Clearly,

a better understanding Of the specific nonverbal behaviors
and their referents must come about before one can make
accurate statements about this hypothesis.

Assuming that there are specific referents associ-
ated with specific behaviors, the questions of universality
is still unanswered. That is, are the nonverbal behaviors
that two intimate persons use in their communication under-
stood by others? Mahl (1968) would tend to agree that

universality exists. However, it would seem that as

78

persons became more and more intimate, they would relate
to more and more subtle nonverbal cues, while persons in

a first meeting situation would only respond to more gross
cues. This in itself may explain the lack Of predicted
results of the present study. Perhaps the cues that the
interviewers were manipulating were so subtle that they
did not cause any changes in the subjects' behavior, sub-
jects who were responding primarily to gross cues, perhaps
such as eye contact.

In addition to the time period main effects, dis—
cussed above, there were also other significant inter-
actions revealed by the analyses Of subject behavior.

One such interaction was that beween sex Of interviewer x
sex of subject for head movement duration. As the results
indicate, there were significant differences between the
means Of females who saw male interviewers vs. males who
saw male interviewers. There were no differences between
subjects who saw female interviewers. These results
indicate that at least in some areas Of body movement
there appear tO exist differences that are attributable
to sex. Perhaps the female, due to her role, or her
imagined role in the society of the college undergraduate,
feels on some level that certain behaviors should be
increased. Or perhaps she is sufficiently aroused in the

presence of a male but not in the presence of a female.

 

— ——"——' -- ..1 -~

79

An interesting relationship was discovered with leg
movement for both frequency and duration. It was found
that subjects who saw male interviewers tended to display
somewhat more leg movement than those who saw female
interviewers. A possible explanation for this tendency
might be found in the nature Of the leg movements made by
interviewers of each sex. Since all of the female inter-
viewers were wearing skirts, their leg movements were
somewhat restricted in comparison to those Of the male
interviewers. Viewing the results in terms of this assump-
tion, it would seem that the more Obvious the movement,
the greater the effect in influencing nonverbal behavior.
Thus the more Obvious movements made by the male inter-
viewers might be reSponsible for the increases found in
subjects who saw males. It is interesting to note that
female interviewers had higher means for leg movement for
both frequency and duration. Thus it appears that sheer
number Of movements may not be as important as the
Obvious nature of movements in effecting change in the

behaviors of others.

Discussion Of Affiliation Measures
and Nonverbal Correlates

Although Mehrabian (1967, 1968, 1969, 1970),
Rosenfeld (1965, 1966, 1967) Exline (1963, 1965) and
Rubin (1970) have all related nonverbal behaviors to

degrees Of affiliation, or liking, the results Of the

80

present study revealed no such trends. One of the most
widely studied variables in this area is that Of eye con-
tact. Because Of limitations in the experimental situation,
it was impossible to Obtain accurate measures Of this vari-
able in the present study. The same was true for smiling
behavior. However, some of the variables that have been
reported as being related to affiliation, such as forward
leaning, or extended leg postures, were not related to

the measures of affiliation utilized in the present study.
These differences may be explained, in part, by the fact
that researchers such as Mehrabian have typically used
imagined dyads, as Opposed to interacting dyads as used in
the present study. The results, above, would indicate

that at least on some variables, such as forward leaning,
there are significant differences between real and

imagined dyads. In addition, measures Of affiliation differ,
and it is very difficult to Obtain agreement between
researchers as to what really constitutes an affiliative
state. This confusion can, of course, give rise to much

variance in terms of results Obtained in different studies.

Sex Differences

As indicated in the results, the predicted sex
differences in terms of total nonverbal activity as well
as the degree to which subjects would be influenced by the

nonverbal manipulations were not Obtained by this study.

81

For the most part, male and female subjects behaved in a
quite similar manner. There were some exceptions, Of
course, such as head movement duration, discussed above.
In general, however, there were no significant differences
found in the results for the four major kinesic variables
rated.

It should be mentioned that the results Of this
study contradict somewhat those of Exline (1963), Mehra—
bian (1967-1970) and Rubin (1970) in which definite sex
differences were Obtained on certain nonverbal variables.
This may be due to differences in the size of subject
pOpulation. Most studies reporting such sex differences

had higher sample sizes than the present study.

Questionnaire Variables

In addition to the prediction Of Hypothesis II
that certain nonverbal variables would correlate with the
results of the questionnaire, it was also predicted that
persons with higher scores on the questionnaire would also
show a greater tendency to be influenced by the interviewer,
and thus show more of the predicted patterns Of behavior.
The results of the questionnaire do not support this
assumption.

Perhaps the most interesting aspect uncovered by
the questionnaires were the consistent sex differences

that were discovered. It was found that females tended to

82

see themselves as more self-assured, Open, and confident
about social situations than did males. In addition, this
more sociable attitude carried over to questions concern-
ing the interviewer, such that females tended to like the
interviewer more than males did, and also reported that
they felt the interviewer liked them to an extent greater
than that reported by males.

In addition to the subject sex differences, one
question (question 5) revealed significant sex of inter-
viewer effects, in which subjects seeing female inter-
viewers reported that they were more confortable being
close than those subjects who saw male interviewers. A
possible implication Of such a finding is that female
interviewers were able to communicate some sense Of
psychological well-being, or security, such that subjects
who saw females felt more comfortable about being emotion-
ally close to others.

In trying to assess the meaning Of these results,
one must bear in mind that there is no check on the
validity Of these self reports. As has been shown, the
significant differences found in the questionnaires did
not reflect any differences in the observable behaviors
of the subjects. However, although these questionnaires
do not seem to be related to observable behavior, it may
be assumed that they are related to the internal attitudes

held by the subjects. If one assumes that these

83

questionnaires correctly reflect such internal attitudes,
then one may ask why females should feel more at ease in
a variety of social situations, report themselves as more
outgoing, etc. One possible explanation is the belief
that females in late adolescence are more socially mature
than males, and thus this added maturity has given them a
more aggressive social outlook on life in general.

An alternative explanation may be that females in
the undergraduate society Of Michigan State University are
more conscious of their social stance and status than are
males, and may have responded to some social pressure to
become more outgoing, etc., while males, who either are
less interested in the social pressures, or who feel it
somewhat differently, tend to think Of themselves as less
outgoing, etc.

. A third explanation may revolve around the imagined
expectations that females may have concerning their social
status, and thus in response to these expectations, report
themselves to be more outgoing, etc., while in reality
they may not behave any differently from their male

counterparts .

Implications of the Above Findings
for Future Research

 

As indicated in the beginning of this chapter,
there are many ways in which the methodology Of the present

study might be modified. The results seem to indicate that

84

by the manipulation of certain nonverbal variables, one
can indeed influence the nonverbal behaviors of another.
At this stage, these relationships are still somewhat
vague. Many questions should be answered before any type
of general rule or statement can be made. The actual
length of the experimental situation could be varied, most
likely with profitable results. In addition, the order of
presentation of certain types or patterns Of behavior
could also be varied. Rating systems should be refined

so that more specific variables are rated, as Opposed to
the broader nonverbal areas rated in the present study.

In addition,persona1ity variables, or other interpsychic
variables might be collected on each subject, so that
additional information may be gathered concerning relation-
ships Or patterns that may emerge.

It is almost axiomatic that when one begins to
explore new concepts or new relationships within the con-
text of a relatively new field, there will be a great deal
Of wrong turns and false starts. The present study was no
exception to this general rule. Many Of the variables
tested proved to be of little worth. Many Of the relation-
ships that "should" have held true did not. However, the
relationships discovered do provide new directions that

can be fruitfully followed in future research.

85

Clinical Implications
of the Present Research

 

 

Persons who work within the interpersonal framework,
either as clinicians or in related areas, Often place a
great deal Of importance upon the nonverbal behaviors Of
their clients. However, very few of them consider that
they are engaged in two way nonverbal communication with
these same clients; a process in which both they and the
client are working together to formulate an on going
relationship. The results of the present study seem to
indicate that even in a period as brief as fifteen minutes,
such definition is taking place, and that this relationship
has a tendency to persevere, even in light of subsequent
behavior changes.

Thus, in face to face interchanges, it becomes
essential to the clinician, if he is making assumptions
concerning the motives, thoughts, and feelings of another
through the other's nonverbal behaviors, to become more
acutely aware of his own pattern and system Of nonverbal
behavior, since he, the clinician, is exerting certain
nonverbal influences upon that other person. For as the
client and therapist define their relationship, each can
(and probably does) set boundaries for each other, until
some compromise is worked out.

Persons working with schizophrenic patients Often

are amazed at the sensitivity these patients demonstrate

86

in regard to minute changes in affect, tone of voice, or
other nonverbal aspects of communication. It may be that
the schiZOphrenic population relies more heavily upon
nonverbal modes of communication than on verbal modes,
since persons tend to be more consistent in their nonverbal
modes of communication. Thus, persons working with this
pOpulation must be extremely aware of their own nonverbal
behaviors if the resulting relationship is to be inter-

preted and responded to accurately.

LIST OF REFERENCES

LIST OF REFERENCES

Allport, G. W. Studies in Expressive Movement. New

Argyle,

Argyle,

York: The MacMiIIan Company, 1933.

M., Salter, V., Nicholson, H., Williams, M., and“
Burgess, P. "The Communication Of Inferior and
Superior Attitudes by Verbal and Non-verbal
Signals," British Journal of Social and Clinical
Psychology, 1970, 9, 222-231.

 

M. "Eye-contact, Distance and Affiliation,"‘
Sociometry, 1965, 28, 289-304.

 

 

Bandura, A., and Walters, R. Social Learnin and Per-
sonalitprevelopment. New York: HOIt, Rinehart,

Baxter,

Beakel,

 

and WihSton, Inc., 1963.

J., Winters, E., and Hammer, R. "Gestural Beha-
vior During a Brief Interview as a Function of

Cognitive Variables," Journal of Personality and
Social Psychology, 1968, 8, (3)] 303-307.

 

 

 

N., and Mehrabian, A. "Inconsistent Communica-
tions and PsychOpathOlogy," Journal of Abnormal
Psychology, 1969, 74, 126-130}

 

 

Birdwhistle, R. Introduction to Kinesics. University

Of Louisville, 1952.

. "Kinesics in the Context of Motor Habits,"

 

Paper presented before American Anthr0pological
Association, December 28, 1957.

. "Field Methods and Techniques: Body Motion

 

Research and Interviewing," Human Organization,
1962, 2, (1), 37-38.

 

"The Kinesic Level in the Investigation of the
Emotions," Expressions of the Emotions in Man, '
Edited by P. Knapp. New York: International
Universities Press, l963a,pp. 123-139.

87

88

Birdwhistle, R. "Body signals: Normal and Pathological,"

Boomer,

Borden,

Paper presented to the American Psychological
Association, September, l963b.

D. "Speech Disturbance and Body Movement,"
Journal Of Nervous and Mental Disease, 1963, 136,
263-366.

 

E. "Commentary: Patient Therapist Interaction,"
Research in Psychotherapy, Vol. III, Edited by

J. Shlien, Washington, D.C.: American Psychological
Association, Inc., 1968.

 

Bruning, J. L., and Kintz, B. L. Computational Handbook

Charny,

Condon,

Darwin,

Decker,

 

of Statistics. Glenview, Ill.: Scott, Foresman
and Company, 1968.

 

E. J. "Psychosomatic Manifestations of Rapport in
Psychotherapy," Psychosomatic Medicine, 1966, 28,
305-315.

 

W., Ogston, W. "Sound Film Analysis of Normal and
Pathological Behavior Patterns," Journal of
Nervous and Mental Diseases, 1966, 143, 338-347.

 

 

C. The Expressions of the Emotions in Man and
Animals. Chicago: University of Chicago Press,

 

J., Maccoby, N., Breitrose, H. "Improving Accuracy
in Interpreting Non-verbal Cues of Comprehension,"
Psychology in the Schools, 1965, 2, (3), 239-244.

 

Dittman, A. "The Relationship between Body Movement and:”

Moods in Interviews," Journal of Consulting
Psychology, 1962, 26, (5), p. 480.

 

 

. "Kinesic Research and Therapeutic Process:
Further Discussion," Expression of the_Emotions

in Man, Edited by P. Knapp, New York: International
Universities Press, 1963, pp. 140-147.

, Parloff, M., and Boomer, D. "Facial and Bodily~
Expression: A Study of Receptivity of Emotional
Cues," Psychiatry, 1965, 28, 239-244.

 

, Stein, 8., and Shakow, D. "Sound Motion Picture
Facilities for Research in Communication," Methods
Of Research in Psychotherapy, Edited by L. Gott-
schalk and A. Auerbach, New York: Appleton-Century-
Crofts, 1966, pp. 25-34.

 

89

Dittman, A., and Wynne, L. "Linguistic Techniques and
the Analysis of Emotionality in Interviews,"
Methods Of Research in Psychotherapy, Edited by
L. Gottschalk and A. AuerbaEh, New York: Appleton-
Century-Crofts, 1966, pp. 146-153.

, Llewellyn, L. "Body Movement and Speech Rhythm
in Social Conversation," Journal of Personalit
and Social Psychology, 1969, ll, (2), pp. 98-1 6.

 

Duncan, Starkey, Jr. "Nonverbal Communication," Ps cho-«*
logical Bulletin, 1969, 72, 2, pp. 118-137.

Edwards, A. Statistical Methods. New York: Holt, Rine-
hart and Winston, Inc., 1967.

 

Efron, D. Gesture and Environment. New York: Kings
Crown Press, 1941.

Ekman, P. "A Methodological Discussion of Nonverbal
Behavior," The Journal of Psychology, 1957, 43,
pp. 141-149.

. "Body Position, Facial Expression, and Verbalb‘
Behavior During Interviews," Journal of Abnormal
and Social Psychology, 1964, 68, (3), pp. 295-301.

 

. "Communication through Nonverbal Behavior: A
Source of Information about an Interpersonal
Relationship," Affect, Cognition, and Personalit ,
Edited by S. Tompkins, C. Izard. Springer PEBIish-
ing Company, Inc., 1965, pp. 390-443.

, and Friesen, W. "Head and Body Cues in the
Judgement of Emotion: A Reformulation," Perceptual
and Motor Skills, 1967, 24, pp. 711-724.

 

, and Friesen, W. "Nonverbal Behavior in Psycho-
therapy Research," Rgsearch in Psychotherapy,
Vol. III. Edited by J. Shlien, Washington, D.C.:
American Psychological Association, Inc., 1968,
pp. 179-216.

 

Exline, R. "Explorations in the Process of Person Percep-
tion: Visual Interaction in Relation to Competition,
Sex, and Need for Affiliation," Journal of Person-
ality, 1963, 31, pp. 1-20.

 

90

Exline, R., and Winters, L. "Affective Relations and
Mutual Glances in Dyads," Affect, Cognition and
Personalify, Edited by S. Tompkins and C. Izard,
New York: Springer Publishing Company, Inc.,
1965, pp. 319-351.

 

 

Exline, R., Gray, D., and Schuette, D. "Visual Behavior
in a Dyad as Affected by Interview Content and
Sex of Respondent," Journal of Personality, 1965,
1, pp. 201-209.

 

Freedman, N., and Hoffman, S. "Kinesic Behavior in
Altered Clinical States: Approach to Objective
Analysis Of Motor Behavior During Clinical Inter-
Views," Perceptual and MOtor Skills, 1967, 24,
pp. 527-539.

 

Freud, S. The Basic Writings Of Sigmund Freud. Edited
and translated by A. A. Brill. New York: Modern
Library, 1938.

 

Giedt, F. "Comparison of Visual, Content, and Auditory,z
Cues in Interviewing," Journal Of Consulting
Psychology, 1955, 19, pp. 407-416.

 

 

. "Cues Associated with Accurate and Inaccurate,
Interview Impressions," Psychiatfy, 1958, 21,

 

Grant, E. "An Ethological Description of Non-verbal
Behavior During Interviews," British Journal of
Medical Psychology, 1968, 41, pp. I77-183.

 

 

Haley, Jay. Strategies of Psychotherapy. New York:
Grune and Stratton, 1963.

Hall, E. T. The Silent Langua e. Garden City: Doubleday
and Company, Inc., 195 .

 

 

Hays, W. Statistics for Psycholo ists. New York: Holt,
Rinehart and Winston, Inc., I963.

Hodgman, C.D., Ed. C.R.C. Standard Mathematical Tables.
Cleveland: ChemicaI Rfihher Publishing Co., 1961.

 

Krout, M. "Autistic Gestures, an Experimental Study in
Symbolic Movements," Psychological Monographs,
1935, 46:208.

 

Leventhal, H., and Sharp, E. "Facial Expressions as
Indicators of Distress," Affect, Cognition and
Personalify, Edited by S. Tompkins and C. Izar ,
New York: Springer Publishing Co., Inc. 1965,
pp. 296-318.

 

 

91
Lofting, Hugh. Doctor Dolittle, A Treasufy. Philadelphia:
J. B. Lippincott Company, 1967.

Luft, J. "On Nonverbal Interaction," The Journal of
Psychology, 1966, 63, pp. 261-268.

 

Luria, A. R. The Nature of Human Conflicts. New York:
Liveright, Inc., 1932.

 

Mahl, G. "Disturbances and Silences in the Patient's
Speech in Psychotherapy," Journal of Abnormal
and Social Psychology, 1956, 53, pp. 1-15}

 

, Danet, B., and Norton, N. "Reflection of Major
Personality Characteristics in Gestures and Body
Movements," Paper presented at American Psycho-
logical Association Annual Meeting, 1959.

. "The Lexical and Linguistic Levels in the
Expression of the Emotions," Expressions Of the
Emotions in Man, Edited by P. Knapp, New York:
International Universities Press, 1963, pp. 77-105.

 

 

, and Schulze, G. "Psychological Research in the
Extralinguistic Area," Approaches to§emiotics.
Edited by T. Sebeok, A. Hayes, and M. Bateson.
Hague: Mouton and Company, 1964, pp. 51-124.

"Nonverbal Anticipation Of Verbalization,"
Paper presented on symposium: New Approaches
to the Study of Facial Expression and Body Move-
ment, Annual Meeting of the American Psychological
Association, 1966.

. "Gestures and Body Movements in Interviews,"-
Research in Psychotherapy, VOl. III. Edited by
J. Shlien. Washington, D.C.: American Psycho-
logical Association, Inc., 1968, pp. 295-346.

 

Matarazzo, J., Saslow, G., and Matarazzo, R. "The Interaction
Chronograph as an Instrument for Objective Measure-
ment of Interaction Patterns during Interviews,"

The Journal of Psychology, 1956, 41, pp. 347-367.

 

. "Prescribed Behavior Therapy: Suggestions

from Non-content Interview Research," Experimental
Foundations of Clinicalgsycholo , Edited by

A. Bachrach. New York: Basic BOOEs, Inc., 1962,
pp. 471-509.

 

92

Matarazzo, J., Weitman, M., and Saslow, G. "Interview
Content and Interviewee Speech Durations,"
Journal of Clinical Psychology, 1963, 19,
pp. 463—472.

, and Wiens, A. "Interviewer Influence on Dura—L»
tions of Interviewee Silence," Journal of Experi-
mental Research in Personality, 1967, 2, pp. 56969.

, Wiens, A., Matarazzo, R., and Saslow, G. "Speech
and Silence Behavior in Clinical Psychotherapy and
Its Laboratory Correlates," Research in Psycho-
therapy, Vol. III. Eidted by J. Shlien. Washing—
ton, D.C.: American Psychological Association,
Inc., 1968, pp. 347-394.

 

Mehrabian, A. "Orientation Behaviors and Nonverbal Atti-
tude Communication," Journal of Communication,
1967, 17, 324-332.

. "Relationship of Attitude to Seated Posture,
Orientation, and Distance," Journal of Personalify
and Social Psychology, 1968, 10, pp. 26-39}

 

. "Inference of Attitudes from the Posture,
Orientation and Distance of a Communicator,"

Journal of Consulting and Clinical Psychology,
I ’ pp. - O

. "Significance of Posture and Position in the
Communication of Attitude and Status Relation-

ships," Psycholggical Bulletin, l969a, 71, (5),
pp. 359-372.

 

 

, and Williams, M. "Nonverbal Concomitants of
Perceived and Intended Persuasiveness," Journal
of Personality and Social Psychology, 19695, I3,
pp. 37-580

 

. "Some Referents and Measures Of Nonverbal
Behavior," Behavioral Research, Methods and
Instruments, 1969c, l, (6), pp. 203-207.

 

, and Friar, J. "Encoding of Attitude by a

Seated Communicator via Posture and Position Cues,"
Journal of Consultin and Clinical Psychology,
l969d, 33, (3). PP. 30-336.

. "A Semantic Space for Nonverbal Behavior,"
Journal of Consulting and Clinical Psychology,
1970, 35, (2). PP. 248-2572

 

93

Pande, S., and Garti, J. "A Method to Quantify Reciprocal
Influence between Therapist and Patient in Psycho-
therapy," Research in Psychotherapy, VOl. III,
Edited by J. Shlien. Washington, D. C.: American
Psychological Association, Inc., 1968, pp. 395- 415.

Perls, F. Gestalt Therapy Verbatim. Lafayette, Ca.: Real
People; Press, 1969.

 

Pittenger, R., Hockett, C., and Danehy, J. The First Five
Minutes, Ithica: Paul Martineau, 1966.

Reik, T. Listening with the Third Ear. New York: Farrar,
Strauss and Company, 1949.

 

Rosenfeld, H. "Effect Of an Approval-seeking Induction on
Interpersonal Proximity," Psychological Rsports,
1965a, 17, pp. 120-122.

. "Temporal Mediation of the Similarity-Attraction
Hypothesis," Journal of Personalify, 1965b, 33,

 

 

"Approval-seeking and Approval-inducing Func-
tions Of Verbal and Nonverbal Responses in the
Dyad," Journal of Personality and Social Psycho-
logy, l966a, 4, pp. 5974605.

 

. "Instrumental Affiliative Functions of Facial
and Gestural Expression," Journal of Personality
and Social Psychology, 1966b, 4, pp. 65-72:

 

 

, and Franklin, S. "Arousal of Need for Affili-
ation in Women," Journal Of Personality and
Social Psychology, 1966c, 3, pp. 2459248.

 

 

Rosenthal, R. Experimenter Effects in Behavioral Research.
New York: Appleton-Century-Crofts, 1966.

. "Unintended Communication of Interpersonal
Expectations," American Behavioral Scientist,
1967, 10' (8) ' pp. 24-260

 

, and Rosnow, R. aftifact in Behavioral Research.
New York: Academic Press, 1969.

 

Rubin and Zick, "Measurement of Romantic Love," Journal of
Personality and Social Psychology, 1970, 16,
pp. 265—2730

 

94

Rubinstein, L., and Cameron, D. "Electronic Analysis Of
Nonverbal Communication," Comprehensive Psychiatry,
1968' 9' pp. 200-2080

Sainsbury, P. "Gestural Movement during Psychiatric
Interview," Psychosomatic Medicine, 1955, 17,

Saslow, G., and Matarazzo, J. "A Technique for Studying
Changes in Interview Behavior," Research in
Ps chothera , VOl. I. Edited by E. Rfihenstein
ané M. ParIogf. Washington, D.C.: American Psycho-
logical Association, Inc., 1959, pp. 125-159.

Scheflen, A. "Communication and Regulation in Psycho-
therapy," Psychiatry, 1963, 26, pp. 126-136.

 

. "On the Nature Of Human Communication," Paper
presented to American Association for Group
Psychotherapy, January, l964a.

 

. "The Significance of Posture in Communication
Systems, Psychiatry, l964b, 27, pp. 316-331.

 

. "Quasi-courtship Behavior in Psychotherapy,"
Psychiatry, 1965, 28, 245-257.

 

 

. "Natural History Method in Psychotherapy,"
Metncds Of Research in Psychotherapy, Edited by

L. Gottschalk and A. Auerbach. New York: Appleton-
Century-Crofts, 1966, pp. 263-287.

 

. "On the Structuring of Human Communication,"
American Behavioral Scientist, 1967, 10, (8),
pp. 8L12.

 

Schlosberg, H. "Three Dimensions of Emotion," Psychological
Review, 1954, 61, pp. 81-88.

Sherman, M. "Peripheral Cues and the Invisible Counter-
transference," American Journal of Psychotherapy,
1965, 19, pp. 280-292.

Wachtel, P. "An Approach to the Study Of Body Language in
Psychotherapy," Psychotherapy: Theory, Research,
and Practice, 1967, 4, (3), pp. 97-100.

Whalen, C. "Effects of a Model and Instructions on Group
Verbal Behaviors," Journal of Consultin and
Clinical Psychology, 1969, 33, (5), pp. 559-521.

95

Wiens, A., Saslow, G., and Matarazzo, J. "Speech Inter-
ruption Behavior during Interviews," Psychotherapy:
Theory, Research, and Practice, 1966, 3, (4),
pp. 153-I68.

Winer, B. J. Statistical Principles ianxperimental
Design. New York: McGraw-Hill BoOk Company, 1962.

Witkin, H., Lewis, H., Hertzman, M., Machover, K., Meiss-
ner, P., and Wapner, S. Personalifythrough
Perception. New York: Harper and Brothers, 1954.

 

APPENDIX A

RATER TRAINING AND AGREEMENT DATA

RATER.TRAINING AND AGREEMENT DATA

Table 6. Pair I test for agreement.

 

 

 

Rater A Rater B % Agreement
Head Frequency 44 45 98%
Head Duration 195 192 98%
Hand Frequency 28 22 79%
Hand Duration 141 139 98%
Leg Frequency 26 28 93%
Leg Duration 86 84 97%
Body Frequency 0 0 100%
Body Duration 0 0 100%
Positive Head Nods F 25 26 96%
Positive Head Nods D 25 29 86%
Space Closing Legs 3 3 100%
Forward Lean 0 0 100%
Arms Crossed 420 435 96%
Legs Crossed 51 60 85%

 

Table 7. Pair II test for agreement.

 

 

 

Rater A Rater B % Agreement
Head Frequency 27 23 85%
Head Duration ~ 224 252 88%
Hand Frequency 13 17 76%
Hand Duration 156 180 86%
Leg Frequency 22 23 95%
Leg Duration 130 142 91%
Body Frequency 1 l 100%
Body Duration 1 l 100%
Positive Head Nods F 34 31 91%
Positive Head Nods D 42 40 95%
Space Closing Legs 0 0 100%
Forward Lean 2 2 100%
Arms Crossed 35 40 87%
Legs Crossed 870 853 99%

 

96

97

Table 8. Pair III test for agreement.

 

 

 

Rater A Rater B % Agreement
Head Frequency 28 34 82%
Head Duration 195 187 90%
Hand Frequency 17 17 100%
Hand Duration 121 126 95%
Leg Frequency 10 13 76%
Leg Duration 44 40 90%
Body Frequency 0 0 100%
Body Duration 0 0 100%
Positive Head Nods F 21 23 91%
Positive Head Nods D 25 23 92%
Space Closing Legs 4 5 80%
Forward Lean l l 100%
Arms Crossed 365 377 .96%

Legs Crossed 680 642 94%

 

APPENDIX B

SUMMARIES OF STATISTICAL ANALYSES

SUMMARIES OF STATISTICAL ANALYSES

Table 9. Summary of analysis of variance for interviewer
head frequency.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 520.83 1 520.83 1.285
C (Sex of Subject) 177.63 1 177.63 .438
AC 124.03 1 124.03 .306
Error 1620.93 4 405.23 --
D (Time Period) 3893.39 2 1946.69 15.909*
AD 0.46 2 0.23 .001
CD 13.06 2 6.53 .053
ACD 356.46 2 178.35 1.456
Error 978.86 8 122.35 --
Total 7685.72 23

*

p<.01

Table 10. Summary of Interviewer Head Frequency Means.

 

Time Periods

 

I II III
A11 Interviewers 15.4 26.2 13.2
Male I. 13.4 24.1 11.1
Female I. 17.4 28.4 15.3
I. with Male Subjects 16.6 27.1 14.8
I. with Female Subjects 14.3 25.4 11.5

 

98

99

Table 11. Summary of analysis of variance for interviewer
head duration.

 

 

 

Source of variance SS df MS F
A (Sex Of Interviewer) 1755.67 1 1755.67 .071
C (Sex of Subject) 1261.00 1 1261.00 .051
AC 32242.40 1 32242.40 1.311
Error 98319.23 4 24579.80 --
D (Time Period) 464540.06 2 232270.03 l41.39*
AD 545.00 2 272.50 .165
CD 8333.26 2 4166.63 2.536
ACD 992.26 2 496.13 .302
Error 13142.06 8 1642.75 --
Total 621131.00 23

*

p<.01

Table 12. Summary of interviewer head movement duration

 

 

 

means.
Time Periods
I II III
All Interviewers 57.3 188.3 55.4
Male I. 56.5 183.3 49.8
Female I. 58.2 193.4 61.0
I. with Male Subjects 62.1 180.7 68.0

I. with Female Subjects 52.6 195.9 42.8

 

100

Table 13. Summary of analysis of variance for interviewer
hand frequency.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 2.13 1 2.13 .026
C (Sex of Subject 58.80 1 58.80 .732
AC 40.83 1 40.83 .508
Error 321.00 4 80.25 --
D (Time Period) 5536.79 2 2768.39 18.011*
AD 236.46 2 118.23 .769
CD 19.40 2 9.70 .063
ACD 206.66 2 103.33 .672
Error 1229.59 8 153.70
Total 7651.71 23

*

p<.01

Table 14. Summary of interviewer hand movement frequency
means.

 

 

Time Periods

 

I II III
All Interviewers 5.9 20.0 5.4
Male I. 7.2 18.2 6.4
Female I. 4.8 21.9 4.3
I. with Male Subjects 7.1 20.2 6.2
I. with Female Subjects 4.9 19.9 4.5

 

101

Table 15. Summary of analysis of variance for interviewer
hand duration.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 38341.87 1 38341.87 .555
C (Sex of Subject) 13846.00 1 13846.00 .200
AC 185.00 1 185.00 .002
Error 276202.69 4 69050.67 --
D (Time Period) 569042.06 2 284521.03 l66.003**
AD 6144.79 2 3072.39 1.792
CD 25947.46 2 12973.73 7.569*
ACD 19400.06 2 9700.03 5.659*
Error 13711.59 8 1713.94 --
Total 962821.60 23

*p<.05

**p<.001

Table 16. Summary of interviewer hand movement duration

 

 

 

means.
Time Periods
I II III
All Interviewers 41.4 188.1 42.6
Male I. 66.1 195.9 65.0
Female I. 17.9 180.9 20.3
I. with Male Subjects 62.8 178.1 64.7

I. with Female Subjects 21.2 198.7 20.6

 

102

Table 17. Summary of analysis of variance for interviewer
leg movement frequency.

 

 

 

 

Source of Variation SS df MS F
A (Sex of Interviewer) 35.20 1 35.20 .210
C (Sex of Subject) 15.40 1 15.40 .092
AC 134.40 1 134.40 .803
Error 669.23 4 167.30 --
D (Time Period) 7229.14 2 3614.57 52.607*
AD 5.01 2 2.50 .036
CD 27.71 2 13.85 .201
ACD 31.51 2 15.75 .229
Error 549.66 8 68.70 --
Total 8697.33 23

*p<.01

Table 18. Summary of interviewer leg movement frequency
means.

 

 

Time Periods

 

I II III
A11 Interviewers 2.5 19.2 2.9
Male I. 2.0 18.4 2.6
Female I 3.1 20.0 3.2
I. with Male Subjects 3.5 18.9 3.3
I. with Female Subjects 1.6 19.4 2.5

 

 

103

Table 19. Summary of Analysis of variance for interviewer
leg movement duration.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 12261.40 1 12261.40 3.362
C (Sex of Subject) 437.00 1 437.00 .119
AC 9381.00 1 9381.00 2.572
Error 14585.90 4 3646.47 --
D (Time Period) 243360.46 2 121680.23 l7.002*
AD 14271.66 2 7135.83 .997
CD 539.46 2 269.73 .037
ACD 15231.66 2 2615.83 .365
Error (57252.60 8 7156.57 --
Total .357321.60 23

*p<.01

Table 20. Summary of interviewer leg movement duration
means.

 

 

Time Periods

 

I II III
All Interviewers 6.4 102.5 7.6
Male I. 3.7 77.0 5.6
Female I. 9.2 128.0 9.6
I. with Male Subjects 10.6 101.6 10.1
I. with Female Subjects 2.3 103.5 5.1

 

 

104 :”

Table 21. Summary of analysis of variance for interviewer
body movement frequency.

 

 

 

 

Source of Variation SS df MS F 1P]
A (Sex of Interviewer) 20.83 1 20.83 1.596
C (Sex Of Subject) 2.13 1 2.13 .163
AC 0.03 l 0.03 .002 .
Error 52.20 4 13.05 -- _
D (Time Period) 738.81 2 369.40 43.977*
AD 30.21 2 15.10 1.798
CD 0.81 2 0.40 .004
ACD 0.01 2 0.00 .000
Error 67.20 8 8.40 --
Total 912.27 23
*p<.01

Table 22. Summary of interviewer body movement frequency
means.

 

 

Time Periods

 

I II III
All Interviewers 0.25 5.58 0.28
Male I. 0.35 6.70 0.20
Female I. 0.15 4.40 0.35
I. with Male Subjects 0.40 5.80 0.40
I. with Female Subjects 0.10 5.30 0.35

 

105

Table 23. Summary of analysis of variance for interviewer
body movement duration.

 

 

 

 

Source Of Variance SS df MS F
A (Sex of Interviewer) 39.67 1 39.67 1.079
C (Sex of Subject) 0.00 l 0.00 .000
AC 0.07 1 0.07 .002
Error 146.96 4 36.74 --
D (Time Period) 1915.80 2 957.90 35.003*
AD 66.20 2 33.10 1.209
CD 4.46 2 2.23 .081
ACD 0.20 2 0.10 .003
Error 218.93 8 27.36 --
Total 2392.33 23

*p<.01

Table 24. Summary of interviewer body movement duration

means .

 

 

Time Periods

 

I II III
All Interviewers 0.38 8.93 0.53
Male I. 0.45 10.60 0.55
Female I. 0.30 7.30 0.50
I. with Male Subjects 0.65 8.80 0.45
I. with Female Subjects 0.10 9.10 0.60

 

 

106

 

 

 

 

N.~ ~.~ m.m m.m ~.~ G.OH «.0 m.HH HHH mocmsemnm
N.- m.mm ¢.o~ N.MH v.o~ ~.HN O.w~ N.MH HH usmfim>oz
v.H v.~ ~.m e.OH N.m o.~H w.m N.OH H can:
«.mm m.m~ e.vH m.mm m.mw ~.HHH m.~ m.~m HHH «coHumnso
m.mmH H.5mH N.omH ¢.mH~ m.~om m.mmH e.w0H o.hH~ HH usmfio>oz
o.mm 0.0m v.- o.N~H o.vm o.NOH m.m m.~h H comm
o.OH ~.~H ¢.h v.mH m.mH o.m~ ~.~ «.mH HHH hosoovmum
v.om v.5m «.NN N.HN «.mm N.om «.mw N.vm HH ucmEo>oz
v.vH N.¢H m.m m.mH O.MH o.m~ ~.m o.o~ H comm
v.02 n.02 v.02 n.02 m.oz H.Oz «.02 H.Oz mOOHHmm
OHmEom OHMEOH OHmz OHM: OHOEOH OHMEOH mHmz mHmz mEHB
muomflnsm mHmemH muomflnsm mHmz
musmm mHm3OH>HouoH msHOOm mH030H>HoucH
.OOHHOQ wEHu room can H03OH>HOHGH
comm HOH mmHanHm> woon can OOH .Ocms .Omon How momma HO3OH>HOHGH .mm wHoma

NmO

13.8
199.4
12.4

63.4
250.6
54.8

12.6
207.2
9.6

51.4
171.6
56.4

7.2
0.2

69.8

180.0
263.6
192.6

I
II
III

Movement
Duration

Hand

NOQ'

OOO
I—INO

NOO
HON
I-l

NNO
Mr-lln

\OV‘O
COP-l
N

\ooom

mmoo
H

I
II
III

Leg
Movement
Frequency

107

NQO
OOV'
0"

GOV

6.0
110.8
11.6

II
III

Movement
Duration

Leg

OOQ‘
O
ONO

OON
O
OI-nO

ONQ‘

OOO
Q'OQ'

OV'O

ONO
OmO

\DNN
OVO

I
II
III

Body
Movement
Frequency

OmO
o
emf-4

OON
OO‘O

COO
OVO
r-l

«we:
OkDO

OQ'O
COO

NON

HKDO

II
III

Body
Movement
Duration

 

in seconds.

*Duration scores are

 

108 ‘

Table 26. Summary of analysis of variance for subject
head movement frequency.

 

 

 

 

 

Source of Variation SS df MS F
A (Sex Of Interviewer) 16.13 1 16.13 .061
C (Sex of Subject) 8.53 l 8.53 .032
AC 1.20 l 1.20 .004
Error 9471.45 36 263.09 --
D (Time Period) 422.71 2 211.35 12.188*
AD 10.41 2 5.20 .300
CD 18.11 2 9.05 .522
ACD 69.35 2 34.67 1.999
Error 138.73 8 17.34 --
Total 10156.66 55

*p<.01

Table 27. Summary of subject head movement frequency

 

 

 

means.
Time Periods
I II III
All Subjects 27.7 24.5 23.4
Male S. 27.9 24.2 22.6
Female S. 27.5 24.8 24.0
S. with Male I. 28.3 24.5 23.9

S. with Female I. 27.2 24.6 22.7

 

109

Table 28. Summary of analysis of variance for subject
head movement duration.

 

 

 

 

 

 

 

Source of Variation SS df MS F
A (Sex of Interviewer) 785.40 1 785.40 .111
C (Sex of Subject) 18975.67 1 18975.67 2.702
AC 32176.87 1 32176.87 4.582*
Error 252763.49 36 7021.20 --
D (Time Period) 1446.19 2 723.09 .929
AD 34.06 2 17.03 .021
CD 2553.80 2 1276.90 1.641
ACD 99.80 2 49.90 .064
Error 6223.73 8 777.96 —-
Total 315059.06 55
*p<.05
Table 29. Summary Of subject head movement duration means.
Time Periods

I II III
.All Subjects 175.9 184.1 178.0
Male 8. 157.7 177.2 165.5
Female S. 194.2 191.1 190.7
S. with Male I. 173.1 181.1 176.3
S. with Female I. 178.7 187.2 179.9

 

 

110

Table 30. Summary of analysis of variance for subject

hand movement freque

ncy.

 

 

 

 

 

 

 

Source of Variation SS df MS F
A (Sex of Interviewer) 35.20 1 35.20 .248
C (Sex of Subject) 48.87 1 48.87 .330
AC 60.09 1 60.20 .425
Error 5099.68 36 141.65 --
D (Time Period) 349.40 2 174.70 17.442*
AD 24.46 2 12.23 1.221
CD 1.80 2 .90 .089
ACD 2.46 2 1.23 .123
Error 721.19 72 10.16 --
Total 6341.32 119
*p<.01
Table 31. Summary of subject hand movement frequency
means.
Time Periods

I II III
All Subjects 19.9 16.9 15.9
Male S. 19.2 16.3 15.5
Female S. 20.7 17.5 16.4
S. with Male 1. 19.9 15.8 15.5
S. with Female 1. 19.9 18.0 16.4

 

 

111

Table 32. Summary of analysis of variance for subject
hand movement duration.

 

 

 

Source of Variation SS df MS F
A (Sex of Interviewer) 300.83 1 300.83 .027
C (Sex of Subject) 3898.80 1 3898.80 .035
AC 1702.53 1 1702.53 .153
Error 399869.79 36 11107.48 --

D (Time Period) 10790.59 2 5395.29 3.003
AD 1018.46 2 509.23 .283
CD 375.80 2 187.90 .106
ACD 353.26 2 176.63 .099
Error 14157.06 8 1769.63 --
Total 432467.16 55

 

 

Table 33. Summary of subject hand movement duration means.

 

Time Periods

 

I II III
All Subjects 160.3 177.2 182.6
Male S. 168.5 181.5 187.1
Female S. 152.1 172.9 178.0
S. with Male I. 157.8 181.3 185.7
S. with Female I. 162.8 173.1 179.4

 

112

Table 34. Summary of analysis of variance for subject
leg movement frequency.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 576.40 1 576.40 4.078
C (Sex of Subject) 54.67 1 54.67 .386
AC 25.20 1 25.20 .178
Error 5087.63 36 141.32 --

D (Time Period) 5.60 2 2.8 .160
AD 74.86 2 37.43 2.144
CD 14.60 2 7.30 .418
ACD 0.26 2 0.13 .007
Error 1256.66 72 17.45 --
Total 7095.92 119

 

 

Table 35. Summary of subject leg movement frequency

 

 

 

means.
Time Periods

I II III
All Subjects 10.4 10.5 9.9
Male 5. 9.7 9.4 9.8
Female S. 11.1 11.6 10.2
S. with Male 1. 11.5 13.2 12.3
S. with Female I. 9.3 7.6 7.2

 

113

Table 36. Summary of analysis of variance for subject leg

movement duration.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 35123.40 1 35123.40 3.081
C (Sex of Subject) 4600.40 1 4600.40 .403
AC 106.40 1 106.40 .009
Error 410282.75 36 11397.02 --

D (Time Period) 503.21 2 251.60 .389
AD 1353.11 2 676.55 1.046
AC 1992.61 2 996.30 1.540
ACD 1562.31 2 781.15 1.154
Error 46566.73 72 646.75 --
Total 502090.99 119

 

 

Table 37. Summary of subject leg movement duration means.

 

Time Periods

 

I II III
All Subjects 50.3 51.5 55.1
Male S. 61.3 57.9 56.3
Female S. 39.3 45.2 54.1
S. with Male I. 62.8 72.2 73.3
S. with Female I. 37.8 31.0 37.0

 

114

Table 38. Summary of analysis of variance for subject
body movement frequency.

 

 

 

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 0.008 1 0.008 .031
C (Sex of Subject) 0.075 1 0.075 .297
AC 0.075 1 0.075 .297
Error 9.100 36 0.252 --
D (Time Period) 0.516 2 0.258 2.224
AD 0.016 2 0.008 .068
CD 0.350 2 0.175 1.508
ACD 0.050 2 0.025 .215
Error 8.400 72 0.116 --
Total 18.591 119
Table 39. Summary of subject body movement frequency
means.
Time Periods

I II III
All Subjects 0.05 0.16 0.20
Male S. 0.10 0.30 0.10
Female S. 0.00 0.10 0.30
S. with Male I. 0.05 0.20 0.20

S. with Female 1. 0.05 0.15 0.20

 

 

115

 

 

 

 

 

 

Table 40. Summary of analysis of variance for subject
body movement duration.
Source of Variance SS df MS F
A (Sex of Interviewer) 0.675 1 0.675 1.503
C (Sex of Subject) 0.675 1 0.675 1.503
AC 0.408 1 0.408 .908
Error 16.160 36 0.448 --
D (Time Period) 1.216 2 0.608 1.652
AD 0.350 2 0.175 .475
CD 0.350 2 0.175 .475
ACD 0.216 2 0.108 .293
Error 26.520 72 0.368 --
Total 45.591 119
Table 41. Summary of subject body movement duration.

 

 

Time Periods

 

I II III
All Subjects 0.05 0.25 0.28
Male S. 0.10 0.40 0.30
Female S. 0.00 0.00 0.30
S. with Male I. 0.05 0.35 0.40
S. with Female I. 0.05 0.15 0.15

 

L

116

 

 

 

m.mH «.mH m.mH o.mH v.MH ~.o~ mme m.NH HHH mocmskum
v.H~ ¢.vH m.hH o.hH m.¢H «.HN o.mH N.MH HH ucmsm>oz
~.mm o.hH m.HN m.om m.¢H w.v~ m.mH m.mH H comm
o.me v.HmH m.mom m.mom m.va v.¢bH m.mHH o.th HHH «coHumuso
m.th «.mbH ¢.Hom m.¢o~ ~.~m~ w.me o.omH o.moH HH ucmsm>oz
m.th m.HmH w.0Hm o.mo~ «.mmH ~.mmH «.mmH m.¢vH H comm
¢.H~ o.m~ ~.- v.n~ m.>H ~.>m m.¢~ o.H~ HHH mocmsvmnm
m.mm ~.mm m.H~ m.m~ m.hH m.m~ o.v~ o.m~ HH ucosm>oz
m.mm o.hm m.n~ o.mm o.mN o.Hm o.Hm m.vm H com:
«.02 m.oz ¢.oz m.oz N.oz H.oz m.oz H.oz moOHHmm
mHmem mHmEmm mHmz mHmz mHmEmm mHmEmm mHmz mHmz mEHB
Ammono< ommmv mumsmw>9mucH
muomnnsm mHmsmm muomnnsm mHmS;
.GOHHmm mEHu comm cH Hm30H>HmucH
comm How mmHanHm> hoon can 00H .ocmc .omms How momma Domnnsm .Nv mHnma

117

 

 

.mocoomm cH who mmuoom COHumnso .33.0
«.0 «.0 0.0 0.0 0.0 «.0 0.0 0.H HHH :oHumusa
0.0 «.0 «.0 0.0 «.0 «.0 0.0 0.0 HH unmEm>oz
0.0 0.0 0.0 0.0 0.0 «.0 0.0 «.0 H swam
«.0 0.0 0.0 0.0 0.0 «.0 0.0 0.0 HHH scamsvmum
0.0 «.0 «.0 0.0 «.0 «.0 «.0 0.0 HH unmam>oz
0.0 0.0 0.0 0.0 0.0 «.0 0.0 «.0 H zoom
0.00 0.0H «.00 «.00 «.00 0.0H 0.00 0.00 HHH coHumuso
0.00 0.0 0.00 0.00 0.00 «.0H 0.00 0.00 HH usmfim>oz
0.00 0.0« 0.«0 0.00 0.00 0.H0 0.H0 0.00H H 00H
0.0 0.0 0.0H 0.0H 0.0 0.0 «.0H 0.0H HHH mocmsvmum
0.0 0.0 0.0H 0.0H 0.0 0.0 «.0H «.0 HH newsm>oz
«.HH 0.0 0.0H 0.0 0.0 0.«H «.«H «.0 H 00H
0.HOH «.H0« 0.00« 0.«0H 0.00« 0.00H 0.00H 0.00« HHH coHumuso
0.00H «.00H 0.00H 0.00H 0.0HH «.00H 0.00H 0.00« HH usmfim>oz
0.0HH 0.00« 0.00H 0.«0H «.00H «.00H 0.0«H 0.0H« H 0:00

118

Table 43. Summary of analysis of variance for recombined
data, subject hand duration, no. 1.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 25585.34 1 25585.34 2.270
B (Sex of Subject) 920.41 1 920.41 .081
AB 799.34 1 799.34 .070
Error 180273.73 16 11267.10 --
C (Time Effect) 5203.43 2 2601.71 3.781*
AC 5927.70 2 2963.85 4.307*
BC 875.83 2 437.91 .636
ABC 1247.69 2 623.84 .906
Error 22016.66 32 688.02 --
Total 242850.18 59

*p<.05

Table 44. Summary of analysis of variance for recombined
data, subject hand duration, no. 2.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 10010.41 1 10010.41 .985
B (Sex of Subject) 8143.35 1 8143.35 .801
AB 1000.00 1 1000.00 .098
Error 162493.33 16 10115.83 --

C (Time Effect) 7159.23 2 3579.61 4.46*
AC 3284.63 2 1642.31 2.049
BC 1342.29 2 671.14 .837
ABC 4943.43 2 2471.71 3.084
Error 25641.06 32 801.28 --
Total 224018.18 59

 

*p<.05

 

119

Table 45. Summary of analysis of variance for recombined
data, subject hand duration, no. 3.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 2356.26 1 2356.26 .227
B (Sex of Subject 6573.06 1 6573.06 .634
AB 6912.26 1 6912.26 .667
Error 165743.73 16 10358.98 --

C (Time Period) 11583.89 2 5791.94 7.31*
AC 3344.43 2 1672.21 2.111
BC 192.03 2 96.01 .121
ABC 2210.83 2 1105.41 1.395
Total 224264.00 59

*p<.01

Table 46. Summary of analysis of variance for recombined
data, subject hand duration, no. 4.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer 129.06 1 129.06 .013
B (Sex of Subject) 448.26 1 448.26 .048
AB 20461.06 1 20461.06 2.212
Error 147963.33 16 9247.70 --

C (Time Period) 13525.63 2 6762.81 7.46*
AC 715.43 2 357.71 .395
BC 644.43 2 322.21 .355
ABC 5920.63 2 2960.31 3.269
Error 28971.86 32 905.37

Total 218779.73 59

 

*p<.01

120

Table 47. Summary of analysis

of variance for question 1.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 2.025 1 2.025 2.100
C (Sex of Subject) 1.225 1 1.225 1.270
AC 3.025 1 3.025 3.138
Error 34.700 36 0.963 --
Total 40.974 39

 

Table 48. Summary of analysis

of variance for question 2.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 1.600 1 1.600 2.087
C (Sex of Subject) 3.600 1 3.600 4.696*
AC 1.600 1 1.600 2.087
Error 27.600 36 .766 --
Total 34.400 39

*p<.05

Table 49. Summary of analysis

of variance for question 3.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 0.900 1 0.900 .625
C (Sex of Subject) 0.100 1 0.100 .069
AC 1.600 1 1.600 1.110
Error 51.800 36 1.438 --
Total 54.400 39

 

121

 

 

 

 

 

 

 

 

Table 50. Summary of analysis of variance for question 4.
Source of Variance SS df MS F
A (Sex of Interviewer) 3.6 1 3.6 4.00
C (Sex of Subject) 0.0 1~ 0.0 .00
AC 0.0 1 0.0 .00
Error 32.4 36 0.9 --
Total 36.0 39

Table 51. Summary of analysis of variance for question 5.
Source of Variance SS df MS F
A (Sex of Interviewer) 6.4 1 6.4 8.17*
C (Sex of Subject) 6.4 1 6.4 8.17*
AC 0.9 l 0.9 1.14
Error 28.2 36 0.78 --
Total 41.9 39

*p<.01
Table 52. Summary of analysis of variance for question 6.

 

 

 

Source of Variance SS df MS F

A (Sex of Interviewer) 0.025 1 0.025 .018
C (Sex of Subject) 9.025 1 9.025 6.757*
AC 0.625 1 0.625 .467
Error 48.100 36 1.338 --
Total 57.775 39

 

*p<.05

122

Table 53. Summary of analysis of variance for question 7.

 

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 0.625 1 0.625 .342
C (Sex of Subject) 11.025 1 11.025 6.041*
AC 0.025 1 0.025 .013
Error 65.700 36 1.825 --
Total 77.375 39

*p<.05

Table 54. Summary of analysis of variance for question 8.

 

 

 

Source of Variance SS df MS F
A (Sex of Interviewer) 1.225 1 1.225 .651
C (Sex of Subject) 42.025 1 42.025 22.34*
AC 0.025 1 0.025 .013
Error 67.700 36 1.880 --
Total 110.974 39

 

*p<.01

APPENDIX C

QUES TI ONNAI RE

 

' V 'kfl..-. .

QUESTIONNAIRE

This form has been designed to investigate your
reaction to the interview you have just finished.
In addition, we have asked some questions about your-
self. Please answer all the questions as well as you
are able.

The first few questions each have seven answers.
Please check or circle the answer that is closest to
your own feelings about the question.

How do you feel about the interview you have just
finished?

(a) disliked it very much.

(b) disliked it.

(c) vaguely disliked it.

(d) have no feelings about it, one way or another.

(e)
(f)
(g)

moderately liked it.
liked it.
liked it very much.

HHHHHHH

How did you feel about the person who interviewed you?

(a)
(b)
(c)
(d)
(e)
(f)
(g)

disliked the interviewer very much.

disliked the interviewer.

vaguely disliked the interviewer.

felt rather neutral towards the interviewer.
moderately liked the interviewer.

liked the interviewer.

liked the interviewer very much.

HHHHI—IHH

If the interviewer lived in or near your dorm, or
building, what chance would there be of becoming
friends with the interviewer?

(a) No chance at all that we would become friends.

(b) Most likely no chance that we would become friends.

(c) It would be doubtful that we would become friends.

(d) I'm not sure, one way or another.

(e) There would be a chance that we would become friends.

(f) There would be a very good chance that we would
become friends.

(9) I'm sure we would become friends.

123

 

4.

5.

6.

124

How do you think the interviewer felt about you?

(a) The interviewer disliked me very much.
(b) The interviewer disliked me.

(c) The interviewer vaguely disliked me.
(d) The interviewer was neutral towards me.
(e) The interviewer moderately liked me.
(f) The interviewer liked me.

(9) The interviewer liked me very much.

How much do you like being emotionally close to other

persons?

(a) I strongly dislike being close to other persons.

(b) I dislike being close to other persons.

(c) I vaguely dislike being close to other persons.

(d) I have rather neutral feelings about being close
to other persons.

(e) I moderately like being close to other persons.

(f) I like being close to other persons.

(9) I like very much being close to other persons.

When you meet strangers in a social situation (for
example, in a class, at a party, at the grill) how
do you feel?

(a) I strongly dislike talking to strangers.
(b) I dislike talking to strangers.
(c) I vaguely dislike talking to strangers.

(d) I feel

neutral about talking, or when talking to

strangers.
(e) I moderately enjoy talking to strangers.
(f) I enjoy talking to strangers.
(g) I enjoy very much talking to strangers.

If you enter a room, and a group is laughing and talking,

how strong

(a) I have
(b) I have
(c) I have
(d) I have
way.
(e) I have
(f) I have
(9) I have

a desire do you have to enter the group?

no desire to enter the group.

a very weak desire to join the group.

a weak desire to join the group.

no feelings about joining the group, either

a moderate desire to join the group.
a strong desire to join the group.
a very strong desire to join the group.

125

8. Are you an outgoing person?

(a) No, not at all.

(b) No, not very much.

(c) No, only a little bit.

(d) I can't say, one way or another.

(e) Yes, a little bit.

(f) Yes, I'm somewhat of an outgoing person.
(g) Yes, I am truly an outgoing person.

9. Do you think you would have acted differently with an
interviewer of the Opposite sex?

(a) YES (b) NO

If you have answered YES to the last question, in what way
would you have acted differently?

10. If there was anything that bothered you about this
experience, we would appreciate your comments.

11. What things could the interviewer have done better to
make you feel more at ease?

12. You are (a) MALE Your interviewer was (a) MALE
(b) FEMALE (b) FEMALE

If you want to be sent a short summary of the results of
this study, please write your name and address on the back
of this page.

Thank you for your time.

"ITHJIEEIEIJ'WMWEWMW