ll

4

 

‘EHE FERFGRMAHCSE S'WLE TEE"? AS A
FEEDI'QTQR GE CLASERQCM
EEHAW SE PAHEQNS

Tim?“ {‘09 {423‘- Dane a? M. 15%
magi: 3&1fo STE-3‘8 WRITERS???
$03111 3?. 5011.633 Jr.

2968

THES\9

 

4"" l
1 “I
E
‘
._ Punh.
g .
a“ L’n'i

ABSTRACT

THE PERFORMANCE STYLE TEST AS A PREDICTOR OF
CLASSROOM BEHAVIOR PATTERNS

BY

John R. Jones Jr.

This study was undertaken with the hOpe of further
validating the typology developed by Ring (1966) concerning
performance styles of peOple as they are distinguished by
their level of social competence. The concept of performance
styles is based on a dramaturgical approach to interpersonal
interaction advocated by Goffmann (1959). Three styles are
arranged in a typology: the p_is incapable of good perform—
ance giving and consequently avoids social interaction, he
may be described as an introvert; the £_is socially competent
and is motivated to give a good performance, he may be
described as an extrovert; finally, the g_who is not competent
socially, but who has learned to imitate the styles of those
around him, may be termed a conformist.

One hundred and eighty-five students took the Performance
Style Test, developed by Ring to distinguish between E, g,
and 2. Of these, eleven students were selected for each group

on the basis of their scores. Each of these persons were in

John R. Jones Jr.

turn observed in the classroom setting, and concrete measures
of behavior patterns were utilized, such as frequency of
conversation, duration of conversation, seating position in
the classroom, etc.

It was hypothesized that the socially competent group,
:fs, would be distinguished from the other two groups on all
of the measures; for example, they would converse more fre-
quently than pfs or gfs, etc. Results indicate that the dif-
ference between the £_and p_groups on various behavior patterns
is reliable and consistent. The relation between gfs
(conformists) and the other two groups was not consistent in
the hypothesized direction.

Suggestions for sharpening the distinctions between
groups are offered and the potential use of the Performance

Style Test and related typology are discussed.

///;3/::Q

THE PERFORMANCE STYLE TEST AS A PREDICTOR OF

CLASSROOM BEHAVIOR PATTERNS

By

M
1.“

Aqfi
John R.“Jones Jr.

A THESIS

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

MASTER OF ARTS

Department of Psychology

1968

DEDICAT ION

To all who have come before,
that my work may help those
who come after.

ii

ACKNOWLEDGMENTS

It is an opportune moment to express gratitude for the
unrepayable assistance offered by Dr. Gary Stollak who pro-
vided the basic ideas for this study, for the more than
adequate assistance in the statistical treatment of the data
offered by Dr. Lawrence Messe', and for the patient criticism
and helpful advice of Dr. Dozier Thornton. A special word
of appreciation and thanks is due to each of ten un-named
undergraduate students who diligently collected the data.

Finally, thanks are in order to Miss Geraldine Waters

and Mrs. Andrew Wilson for their technical assistance.

iii

TABLE OF CONTENTS

 

Page

ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . iii
LIST OF TABLES AND APPENDICES . . . . . . . . . . . . V
INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 1
The Performance Style of the p_. . . . . . . . . 2

The Performance Style of the r . . . . . . . . . 4

The Performance Style of the g_. . . . . . . . . 5

Use of Naturalistic Observation. . . . . . . . . 7
METHOD. . . . . . . . . . . . . . . . . . . . . . . . 11
Selection and Assignment of Ss to Groups . . . . 11
Testing Procedure. . . . . . . . . . . . . . . . 11
Observational Procedure. . . . . . . . . . . . . 14
RESULTS . . . . . . . . . . . . . . . . . . . . . . . 17
DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 25
REFERENCES. . . . . . . . . . . . . . . . . . . . . . 52
APPENDICES. . . . . . . . . . . . . . . . . . . . . . 55

iv

LIST OF TABLES AND APPENDICES

Table Page

1. Criteria for entry into group, based on respec-
tive percentile scores. . . . . . . . . . . . . . 13

2. Mean Scores on each of seven behavioral measure—

ments . . . . . . . . . . . . . . . . . . . . . . 15
3. Analyses of Variance. . . . . . . . . . . . . . . 18
Appendix
1. A Test to measure performance styles in inter-
personal relations (Ring) . . . . . . . . . . . . 55
2. Data Compilation from running log . . . . . . . . 40

INTRODUCTION

A concept of performance styles, based on the drama-
turgical approach advocated by Goffmann(1959), has recently
been developed by Ring et a1. (1966). This view defines
interpersonal behavior as “performance-giving" and the
individuals involved as actors, each controlling the concep—
tion which others form of them, The idea is not based on
special people who read scripts written by others, nor is it
a new mode of relating to others. It is simply a new way of
describing what goes on when two or more people meet, a
description written by a third person observer. The under-
lying principle is similar to that of Berne (1964). Attempt-
ing to distil the myriad of behavior patterns characteristic
of interpersonal interaction, Ring et a1. originally formu-
lated a typology of three performance styles, with a
description of a model individual associated with each of the
styles (Ring 1966). According to Goffmann, three variables
control any performance: the actor and his attributes, those
of the setting for the performance, and those of the audience
for whom the performance is given. Of these, Ring is only
presently concerned with the first, which he bases on three

components: 1) motivation to perform, 2) knowledge of the

"script", 5) and competence in performing. Competence is
considered to be the base for the other components and the
first in time. Thus, a person unskilled in acting will not
be motivated to act and his knowledge of the scripts will be
necessarily poor.

Returning to a previous statement, it is hoped that
the variety of combinations of these components for the
attributes of the actor may be synthesized into a few basic
patterns. Part of the reason for this hope is that the com-
ponents themselves are structurally interdependent. Thus we
have mentioned the dependence of motivation on competence;
the converse is also true; furthermore, knowledge follows
from both and is necessary for both. Secondly, we emphasize
that types, as such, represent reference points; they may
be useful in prediction of behavior and further theory con-
struction. The validity of such a typology rests on
empirical investigation; this is the reason for our study.
Ring describes each of the three performance styles and the

individual characteristics of each of them.

The Performance Style of the p

 

"The p_(i.e., the sort of person who displays the per-
formance style we are about to describe) is conceived to be
'low' on all three actor components. Motivationally, he
dislikes and prefers to avoid those interpersonal contexts

which, in his own eyes, call for him to 'act' or 'play a

role'. He wants merely to 'be himself', and for others to
'be themselves‘ too. With respect to skill factors, it is
assumed that a p lacks the social agility required for
successful performance-giving. He is also wanting in his
knowledge of the scripts appropriate to a wide variety of
social settings.

"Interpersonally, then a p_is likely to be somewhat less
than impressive; whenever he feels that he cannot 'be himself',
he behaves ineptly and feels restively uncomfortable and
ill-at-ease. In Goffman's terms, a p_is a person who is
deficient in the 'arts of impression management' and knows it."

It is perhaps important to consider some implications
of this performance style. Any interpersonal behavior may be
conceived of as a performance; to navigate in the world of
human affairs, we must control the information which others
have of us. It should only take a moment's introspection to
assent to the fact that we act in different ways with differ—
ent people, we control the conception which others form of
us. According to Goffmann,such acting is work, demanding
close attention to scripts, staging, etc.; being oneself is
easy. The p_is not a competent actor; so, he avoids situ-
ations which demand performance—giving (an unrealistic moti-
vation, as will be explained later). The p_would rather
§g_;;§, throw away the script; he knows he couldn't handle
it if he didn't. For this reason, he demands an informal

atmosphere, places high value on "sincerity", and seeks cues

to others evaluation of him as a person. He cannot act, so

he emphasizes spontaneous, "non-phony" living.

The Performance Style of.£

 

"The £_is an individual who is typically skilled in
interpersonal relations. In fact, in virtually every respect,
£;§_and p;§_constitute antipodal types. An £_enjoys inter-
personal relationships which make a p_feel ill-at—ease; an‘g
knows what to do in interpersonal contexts where a p_is at a
loss as to how to behave; finally, an £_can do what needs to
be done, a p_frequently cannot. From this contrast with pLg,
it can be seen that we assume £;§_are 'high' on each of the
three actor components on which the p;§_were assumed 'low'.
Finally, whereas p;§_may be regarded as being motivated
primarily by self-expressive needs, £L§ seem better described
in terms of a somewhat manipulative interpersonal orienta-
tion. Their behavior may often be tinged with a certain
Machiavellian quality (Christie, 1962) and, when dependent
on others, they resemble what Jones (1964, pp. 73-74) has
labeled 'acquisitive ingratiators'. These instrumental moti—
vations, coupled with an adroitness in interpersonal relations,
enable £;§_to exploit social situations with an effectiveness
and versatility that p;§_can only find depressing.

"In order to imagine an r one need only think of the
skilled hostess, the engaging salesman, or the appropriately

doleful funeral director. These examples illustrate roles,

to be sure, but we believe that there are many individuals
who characteristically perform interpersonally in such
fashion reminiscent of role-playing itself. That is, such
persons seem always to be putting on a performance for
others in much the same way stage actors do for audiences."

The £.is motivated to give a good performance; and
because he is a competent actor, he seeks others evaluation
of his performing ability. The £_knows how to handle the
script, he accepts its necessity in the majority of inter-
personal relations and then improves on it, stamping it
with his own personality (Ring, 1966). While it is true
that he may manipulate others for his own ends, the same
may be said for the p_who demands that others come to him
on his own terms; he demands the informal, "real self'I

atmosphere.

The Performance Style of the g

 

"The g_is an individual whose behavior is dictated
almost completely by the nature of interpersonal relations.
Such an individual becomes the person the "script " calls
for; and when the "script" changes, so does he, for there
is no internal resistance to change. Such a person is called
a chameleon.

"In terms of our three actor components, a 2 may be
characterized as follows: a.g is motivated to give good

performances (like an E), but only performances of a passive,

conventional, approval-seeking sort (unlike r). A g_has
sufficient skill to enact successfully only such passive
non-innovative performances--on the skill component,
therefore he falls somewhere between the p_and the E,
but probably closer to the former. Because we assume that
a §_has a strong need to achieve social approval, one would
expect him to be knowledgeable about the kind of performance
required in a wide range of situations; at the least, he
can be assumed to be as informed about the script-demands
of his preferred kind of performance as an.£ is about his."
The g_is a person in the same boat as the 9; he lacks
social competence. However, instead of withdrawing from
people, he moves toward them. He in motivated to give good
performances, but only for social approval. Because the g
is socially inept, he plays relatively passive, conventional
roles which appear as "good" performances. He conforms and
uses this conformity to manipulate others (Jones, 1964);
by it he receives the social approval which he demands.
After originating the performance style typology and a
paper and pencil test for distinguishing the individual's
particular typology, Ring has begun research on this founda-
tion with the hope of stronger validation and greater
usefulness. It is the hope of this author to employ methods
of naturalistic observation with a similar endeavor to provide

further validation of the performance style typology.

Use of Naturalistic Observation

Several recent articles expound the advantages and
disadvantages of naturalistic observation (Rausch, 1967;
Sechrest, 1967). Only a few points need be made. Research
questions involving behavioral patterns of persons left to
their own resources should involve low manipulation and low
imposition by the experimenter (Willems, 1967). This point
especially is relevant to the present study of performance
styles, since these, by definition, are affected by
"audience" and "setting" variables. Presence of the experi-
menter and the laboratory setting are salient features and
quite possibly affect the performance styles of the subjects,
perhaps differentially between groups. Since we are con-
cerned with phenomena of human behavior as such (Manzel,
1967), it seems wise to consider them in a context involving
the least interference. We may at the same time avoid
artificial tying or untying of variables involved in the
performance styles, thus, the behavior patterns of the g,
for example, might vary from everyday life when situated in
the laboratory setting.

Of perhaps most importance is the extent to which we
may generalize findings. As Willems points out, it is easy
to carry the results of controlled laboratory experiments
to everyday life, supposing that caution is the only require-

ment (1967). Rather, it seems that naturalistic methods are

needed to optimize generalization and to check laboratory
findings.

Before we specify the locus of our observation, we
should consider several assumptions which must be made.
First, we assume that behavior, experiences, thoughts, etc.
and/or reports of behavior, experience, thoughts, etc. as
these occur in actual life situations are legitimate sources
of data (Reusch, 1967). Secondly, we assume that attitudes
(in this case, a particular performance style) must be in-
ferred from some behavior, that there cannot be “inconsistency"
between attitude and behavior (Sechrest, 1967). Thus, we
assume that behavior patterns serve as data and that these
are good means of isolating attitudes.

It seems then that a valuable source of verification of
the Ring typology may occur in almost any natural setting.
For several reasons, we have selected the classroom area.
The setting is quite common to the $5 of our study (elementary
psychology students); and we have here a kind of situational
sampling (Sechrest, 1967) in which we make observations only
in those situations having a reasonably high probability of
yielding data in which we may be interested. Furthermore,
we find here an opportunity to assess attitudes without
exclusive dependence on questionnaires and interviews. There
are a variety of ways in which an attitude may manifest itself
and verbal statements are only one of many (Sechrest, 1967).

We have selected several behavior patterns typical of the

classroom. Then, in the context of the personality typology,
we should be able to predict the respective behavior
patterns of a 9, an E, or a‘g. In line with current think-
ing on naturalistic observation and its advantages, we have
isolated measures which do not require the cooperation of
the subject: such as seating position, time of arrival for
class, postural and gestural cues, etc. Finally we have
chosen a setting in which the undergraduate subject will
behave as an accurate informant about himself (Jones, 1964).
Summing in the words of Marvin Dunnette, ”. . . labora-
tory findings usually lead to elaborate theories or behavioral
taxonomies, entirely consistent within themselves but lacking
the acid test of contact with reality. . . . Psychologists
who choose to partake of the advantages of the more rigorous
controls possible in the psychometric of experimental labora-
tories must accept responsibility for assuring the day to
day relevance of the observations they make," (1966).

Particular areas of classroom behavior patterns are

utilized in the following hypotheses:

(1) With respect to conversation before, during, and
after the class period, rfs should speak more
frequently than pfs; gfs should fall in between.

(2) With respect to the number of people involved in
conversation, gfs should speak with more people
than pfs do; again, gfs should fall in between.

(5) Students of the £_group should arrive earlier for

class than students of the p_or 9 group doo

10

(4) Students of the £_group should remain longer in
the classroom area after class than students of
either other group do.

(5) With respect to the length of conversation, rfs
should speak for a longer duration during each
conversation than pfs; gfs should fall in between.

(6) With respect to participation in class, gfs should
ask questions or enter into group discussion more
often than do students of either other group.1

(7) With respect to seating position in the classroom,
gfs should sit somewhere near the front and center
of the classroom, pfs should sit near the rear and

sides of the classroom, gfs should fall in between.2

 

1Three other hypotheses contingent on the data related
to class participation were deleted for reasons discussed
later.

2If gfs require social approval and recognition, they
should sit up front and near the teacher. If pfs wish to
avoid interaction, they should sit at the extremities.

METHOD

Groups representing each of the performance styles were
chosen on the basis of their scores on the Performance Style
Test. Individuals in these groups were observed during
their classroom periods and the results were compared with

the predictions.

Selection and Assignment of Ss to Groups

One hundred and eighty—five Ss participated by taking
the Performance Style Test (App.1). All Ss were drawn on a
volunteer basis from a subject pool comprised of students
enrolled at the time of this study in the introductory
psychology course at Michigan State University. Participa-
tion in the psychological experiments is a requirement of
the course; no financial incentives or reimbursements were

involved.

TestingiProcedure

Early in the Spring quarter of 1968, 185 students en-
rolled in the elementary psychology course, who have volun-
teered for the study in order to fulfill research credit

requirements, will be asked to complete a paper-and—pencil

11

12

test (the Performance Style Test, designed by Ring et al.
1966, cf. App. 1). It was designed by the authors to
measure an individual's characteristic performance style.
The Performance Style Test consists of 55 true-
false statements. The content of these items reflects
the characteristics attributed to p's, r's, and c's.
Three scoring keys were devised (one for each perform-
ance style) on the basis of the authors' intuitive
judgements concerning how each item would be answered
by a person of a given typological status. Each
respondent, then, gets a score (in the form of a per-
centile rank) on each of the three performance style
dimensions. The test takes approximately 10 minutes
to complete.
Administration of the test for all Ss took place at
the same time. Instructions were simple and were included
with each copy of the test. Questions were answered in a
way which did not disclose the purpose of the test. 83 were
asked to include a schedule of their classtime and locations
(for the purported reason that the present study was compar—
ing test outcomes of students who attend early classes with
those of students who attend classes later in the day). As
the Ss left the testing room, they were divided into groups
of five and photographed, and the photographs were labeled
with their respective names. (The purpose of the class
schedules and photographs was facility of identification
during the later observation period, cf. below.) Finally,
the tests were scored, and three groups, each comprised of
approximately 11 83, were formed on the basis of their scores

on respective typologies. In this investigation, we re-

stricted ourselves to relatively "pure" types (of. Table 1).

15

Table 1

Criteria for entry into group, based on
respective percentile scores.

 

 

Categorical Scores

 

 

p r c
p above 90% below 40% below 40%
c
O
-a
+1:
()0
3f: r below 35% above 90% below 55%
a)m
m4J
H
0434
gt) c below 45% below 40% above 85%
(D

 

(cf. App.1)

 

Note: Based on these criteria, the groups finally consisted
of: £_= 9, p_= 9, g.= 10.

14
Observational Procedure

The study begins with the instructions to the co-workers
(10 undergraduate seniors) who collected the data of natural—
istic observation in the classroom setting. Each co-worker
was assigned a number of 85 from the three groups as deter-
mined by the test. Since class schedules were requested and
pictures taken, the co-workers were able to identify the Ss
assigned to them. The co-workers were given data sheets (cf.
Table 2) and the following instructions:

The plan is simply to observe and record several behavior
patterns of each student involved in the experiment. The
original observations will be recorded in the form of a
running log; the material will then be transcribed onto a
data sheet. With the running log we are attempting to record
the observable behavior of the student as it occurs. The data
sheet is used to Operationalize these patterns for statistical
use. In recording the running log, we will be looking for
the following behavior patterns:

1. Time of arrival into the classroom area (i.e., the
classroom itself and the hallway outside).

2. Conversation between arrival and departure times
a. number of people involved in conversation
b. duration of conversation (in secs.)

5. participation during actual classtime
a. duration of participation (in secs.)
b. manner of presentation (rate 1-5)
c. content of presentation (rate 1-5)
4. seating position (mark on chart)

5. time of departure.

15

Table 2

Mean Scores on each of seven behavioral measurements.

 

 

Mean scores

 

 

Hypothesis Behavioral measure p_ g lg

(1) Number of conversations per 1.55 5.39 1.70
class period.

(2) Number of people involved per .835 1.28 1.50
conversation per class.

(3) Minutes arrived before class- 6.55 6.66 4.20
time.

(4) Minutes departed after class .72 .94 1.89
ends.

(5) Number of seconds per con- 21.16 51.61 54.50
versation.

(6) Number of participations .55 .58 .45
per class period.

(7) proportionate distance from 61.12% 56.56% 42.85%

the front of the classroom.

 

16

Note of explanation for criteria:

--criteria for manner of presentation
Watch for: rate of speech, stammering, pitch of
voice, use of gestures, excuses about
the quality of speech, blushing,
posture, etc.

-—criteria for content of material

Watch for: does the material pertain to homework

; assignments, dates for exams, grading
procedures, etc.; or to meaning of the
text or lecture material. Does the
question or statement reflect thought,
consideration; or is it a parroting of
previous statements?

--rating: rate 1 = calm; 5 = nervous
1 = meaningful; 5 = superficial

Seating position: On the large chart, mark the student's
desk in relation to the entire class. When transpos-
ing it onto the data sheet's smaller chart, mark the
student's desk in relation to the class seating
arrangement, i.e., take the cluster of seating of
the class from the large chart and transpose it onto
the smaller one. Then state the percentage of seats
from the sides and from the front on the smaller
chart.

Example of the running log: I first saw Charlie Brown
enter the hallway outside of the classroom at about
10:00 AM (class begins at 10:00 AM). He talked to
two people for about 50 secs. just before entering
the classroom area; hung up his coat, and then took
a seat all the way over on the other side of the
room. He sat down, picked up a newspaper and began
to read for about 4 minutes. He than talked to the
girl sitting behind him for about 2 minutes. He
turned around and put his head on the desk and
"slept" for about 10 minutes. He was awakened by
raucous laughter and began to rub his eyes. He then
raised his hand and asked a question about a reading
assignment. He turned around and talked to the girl
behind him for another 1 min. (it should be mentioned
that Charlie was calm during his question, I would
rate him 1). Class ended-at 10:50 and Charlie re-
mained to talk with four fellows for about five min.
Then he left. (Seating position is marked on the
chart.)

The compilation of data included in the running log was

recorded on individual data sheets (cf. App. 2).

RESULTS

The experimental hypotheses based on measures of class-
room behavior were divided into those related to interpersonal
interaction, classroom participation, and seating arrangement.
Nine hypotheses, then, were develOped, each with a specific
behavioral measure.

With respect to interpersonal interaction in the class—
room, it was first hypothesized (1) that students of the g
group would converse more frequently than students of the
other two groups. Table 2 summarizes the number of conversa-
tions carried on by students in each of the three groups.

The specific measure for this hypothesis was the number of
times each student spoke in the classroom situation to another
person or group of persons. Thus, students of the g group
spoke an average of three or more times before, during, or
after each class period, while students classified in the
other groups spoke only one or more times. Analysis of
variance (Table 3) applied to these results indicates a sig-
nificant difference between groups (p < .05). And finally,
the Studentized Range Statistic indicates that the difference
is in the expected direction (p < .01). Thus, students who

were assigned to the £_group based on responses to the

17

Analyses of Variance

18

Table 5

 

 

 

 

Hypothesis SS df MS F
Between Groups
(1) 44.57 2 22.18 4.82 (p<.05)
(2) 2.626 2 1.515 5.55 (p<.06)
(5) 49.62 2 24.81 .269 (p>.10)
(4) 14.44 2 7.22 .218 (p>.10)
(5) 6587.51 2 4195.5 1.24 (p<.10)
(6) .281 2 .141 .012 (p>.10)
(7) 7927.69 2 5964.00 5.25 (p<.06)
Between Observations

(1) .09 1 .09 .0006

(2) .657 1 .422 1.088

(5) .644 1 .84 .040

(4) 1.97 1 1.97 .585

(5) 589.27 1 589.27 .577

(6) .094 1 .094 .50

(7) 2055.96 1 2055.96 .027

 

19

Performance Style Test did, in fact, speak reliably more
often than students of either of the other two groups.

The second hypothesis (2) about interpersonal inter-
action in the classroom states that £_type students will join
in conversation with more peOple than the p_or g_type students
will. Thus, it would be expected that an £_might join in
conversation with two, three, or more students in or around
the classroom, while pfs or c's would probably stick with one
or two people. The specific measure, then, is simply the
number of people usually involved when a student of the g,

p, or 9, groups enters into conversation. It should be noted
that some students of the p_group did not speak at all during
the class period but were still included for compilation of
the mean scores. An analysis of variance of these results
indicated a borderline significant difference between groups
(p < .06). The Studentized Range Statistic indicates sig-
nificantly (p < .01) that the difference between p_and both
the g and £_groups is in the expected direction; thus, gfs
and gfs on the whole, join in conversation with a greater
number of people than pfs do. We find no significant differ-
ence between E and g_type students on this measure.

The third hypothesis concerning interpersonal inter-
action is actually composed of two hypotheses (5 and 4).

The first (5) states that £_type students should arrive
earlier than the scheduled commencement time for each class,

that pfs will arrive later, and that gfs will fall somewhere

20

in between. The Specific measure is simply the number of
minutes that a student arrives before class time. Thus,
if a student arrives ten minutes before class time, he is
scored 10; if exactly on time, he is scored 0, if two
minutes late, he is scored -2, etc. Table 1 indicates that
students of the p_and £_groups tend to arrive about 6%
minutes before class, while a g_type student usually arrives
4 minutes early. Analysis of variance did not indicate
significance; and the simple mean scores were obviously not
in the hypothesized direction. The second part of this
hypothesis (4) states that r type students will tend to wait
longer after class ends before they leave the actual class-
room area, than students of the other groups. The measure
is basically the same. Table 2 indicates that no one waits
around too long after class ends. Analysis of variance,
of course, did not find a significant difference between
groups. Again, the direction was not that of the hypothesis.
The fifth and final hypothesis (5) concerning inter—
personal interaction states that those students in the 3
group, when conversing, will tend to speak for longer intervals
than members of either of the other groups. Here the measure
is in seconds of talking time; thus, the individual conver—
sations of rfs and gfs tend to last for about a minute, while
those of the pfs lasts only about twenty seconds. Analysis
of variance surprisingly did not indicate a significant dif-

ference between groups (cf. Discussion). The simple means,

21

however, do indicate a noticeable, if not reliable difference
between rfs and gfs on the one hand and pfs on the other.

The next set of hypotheses are grouped together insofar
as they refer to classroom participating behavior. However,
for reasons to be explained in the Discussion, the results
of only one of these are listed in Table 2. This hypothesis
(6) states that Efs will participate more frequently during
class time than will pfs or gfs. Thus, an r would be
expected to answer or ask more questions than students of
either other group. The specific measure is simply the
number of classroom participations (i.e., anytime the stu-
dent speaks to the teacher or class as a student) which each
student offers during classtime. Table 2 patently indicates
a lack of participation. Analysis of variance indicated no
differences; and directions were not according to the
hypothesis.

Finally, the hypothesis (7) concerning seating arrange-
ment states that students of the g_group will tend to sit
towards the front of the classroom and students of the B
group will tend to sit towards the rear. The specific measure
here is a given percentage, based on proportionate distance
from the front of the classroom. Thus, if a student sits in
the very first row or in the location proximate to the teacher,
his "score" would be 0%; someone sitting in the back of the
classroom would be scored 100%; someone in the middle, 50%,

etc. Mean scores for the groups in Table 2 indicate that pfs

22

tended to sit in the back two-thirds of the classroom, while
students of the other two groups usually sit in the front
half of the classroom. Analysis of variance indicated a
borderline significant difference (p < .06) between groups.
The Studentized Range Statistic indicated that both £_and g
groups sat reliably closer to the front of the classroom than
did students of the p_group (p < .01).

Analysis of variance measures of the seven above hypothe-
ses indicate that while there is often significant differences
between groups, i.e., p, g, or 9, there was never a significant
difference between observations, i.e., results from the first
classroom observation for each student were fairly consistent

with those of the second classroom observation (cf. Table 5).

DISCUSSION

The results indicate that there are, in fact, consistent
differences in simple classroom behavior patterns; but, more
importantly, that these differences exist when people are
grouped along a dimension such as social competence. Hence,
a firmer basis for a typology classifying peOple according
to their performance styles in social interaction is estab-
lished. Consideration of those hypotheses indicating sig-
nificant differences between groups will help to assess how
great the differences are, how consistent, and consequently,
how reliable future predictions might be.

Both p_and £_groups were consistently antipodal (Ring,
1966). Out of the seven hypotheses finally considered,
these two groups were at Opposite ends of the continuum on
four of them; and on three of these, the differences were
significant in the predicted direction. The behavioral pat-
terns on which these predictions were based are both easily
observable and important in their implications. Thus, gfs
tend to speak more often, with more people, and for a greater
length of time than pfs do. And Efs tend to sit more toward
the front of the class than pfs do.

In one sense, the present study has indicated the prob-

ability of consistent differences between E and g groups on

25

24

behavioral patterns in the classroom thus heightening the
value of the Ring typology as based on the Performance Style
Test. But let us consider the data concerning the g_group.
It is variant from the predictions and not nearly so con-
sistent as p and £_patterns, and possibly suggests a partial
revision of the Ring typology.

It will be recalled from the Introduction that Ring sets
up the p_and £_styles as Opposite. This trend, as we have
seen, has been generally confirmed by our data. When discuss-
ing the g_group, “the chameleons," he states vaguely that
they fall somewhere in between p_and 3 groups on several
measures (1967). His original postulation (1966) however,
states that both gfs and pfs lack social competence, but
compensate in different ways. The logical theme of prediction,
reflected in our hypotheses, stated that gfs would fall in
between the other two groups on most measures, but that they
would generally be more similar to pfs than to rfs. In one
out of four hypotheses with notable differences, this basis
of prediction was borne out significantly and in the expected
direction. Students of the E group converse reliably more
often before, during, and after class than do students of
the other groups. The results of the remaining three hypothe—
ses indicate g2 significant differences between £_and 9
groups.

Obviously, some explanation for the unexpected results

could be so derived as to accord essentially with the theme

25

of the hypotheses. Such explanations are frequently found
in discussions of results when the results plainly did not
support the experimental hypothesis; such explanations,
then, should not be considered as conclusions, but as modifi—
cations of the original hypothesis, subject to further
research. In the present study, validation of the basic
typology is well supported; but consideration of unpredicted
results may assist a clearer understanding of Ring's theory.
The g_category presents something of a problem to Ring;
he is unable to specify its position on various measures and
does not mention it specifically in any of his hypotheses
(1966, 1967). Instead, it is placed "in between." Closer
examination of the description of the g_reveals that he is,
in fact, a hard man to "pigeon-hole." Thus, he may feel
socially incompetent and become anxious in situations demand-
ing a well-handled performance; but he also has the ability
to put on a good show. He has "studied" the styles of
apparently more competent people, memorized the scripts, and
is able to repeat it, probably verbatim. Let us clarify with
an example: The £_enjoys conversation before class, he is
confident in his "acting" ability and makes it a point to
demonstrate it every day, no matter whom he is talking to.
The p_scrupulously avoids conversation, makes no bones about
it, and at best utilizes one script. Both can be distinguished
by their observable behavior patterns: one talks, the other

doesn't, etc. But the g_does not exemplify such consistency

26

between conscious appraisal of the circumstances and overt
behavior. He may not feel up to talking, but he also knows
from experience that social approval is contingent upon
talking. He does not know what to say but feels that he must
say something. He resolves the conflict neither by develop-
ing social competence nor by remaining silent, but by becom-
ing a conformist, i.e., he imitates the socially competent
role. He is neither p_(remains silent) nor 3 (socially
competent), but an admixture, the chameleon, who takes a little
from.both.

Perhaps this may shed some light on the unexpected re-
sults, the inconsistent directions of the g_group differences.
The E is, in point of fact, socially incompetent, if we under-
stand this as the ability to handle a variety of interpersonal
situations well. Obviously, this concept of social competence
is rather vague; it is certainly difficult to specify which
behavior patterns measure it and which do not. Since we are
fairly able to distinguish between the two antipodal types,
p_and r, on the assumed continuum of social competence, there
is probably some basis for distinction; but why does the
introduction of the g_aspect of the typology throw in a monkey
wrench? Let us speculate. The 9 may differ from the p_in
his ability to "pose" as a socially competent performer; he
utilizes more of the socially competent script; he is, in
some sense, an imposter. But, if the gfs use of the script

is imperfect, if he is an imposter, his performance should

27

breakdown in a crucial area. Considering again the results
from various hypotheses, it is possible that the one hypothesis
with results which distinguished rfs from gfs was based on
one of these "crucial" areas. Specifically, a person's actual
frequency of conversation tells more about his level of social
competence than do such measures as number of people involved
in the conversation or duration of conversation, etc. To stand
in the presence of a large group of people is one thing, to
speak to them is another. To speak at length with one person
is one thing, to speak many times to many peOple is another.
It is possible then, that some measures of behavioral patterns
may be able to distinguish "real" social competence and others
not.

This reasoning suggests a modification of the theme of
the hypotheses to accommodate for areas where the Q will be
able to "bluff" the audience. Suppose that we distinguish
between two types of conversational content: meaningful and
superficial. Each of us has his own opinion about the criteria
for such a distinction, but some consensus could probably be
reached, possibly along the lines suggested in the Method
section of this paper for rating of class participation con-
tent. On the basis of Ring's typology (1966) and the results
of this paper's hypotheses we could say: the p_will limit
himself to "meaningful" conversation, the g_will usually
utilize "small—talk", while the £_will be comfortable with

either mode of communication. The p_avoids superficialities

28

as representing phoniness; the g_is mimicking another person's
role and should avoid too much talk coming from the "heart";
the £_realizes the necessity of both types of communication
for versatile social interaction. An arithmetical rendition
of this hypothesis would account for rfs noticeably more fre-
quent conversation than either pfs or gfs. Validation would
require investigation of the actual conversational material
utilized by the three types. Thus, a more specific definition
of social competence might be developed, and more critical
measures of behavioral patterns utilized.

Do our results then represent "sound and fury, signify-
ing nothing"? By no means. This further validation of Ring's
typology and his use of the Performance Style Test indicate
that we have begun to isolate a concept such as social
competence. But, the really important upshot of this paper
is that the validation was based on unobstrusive measures of
overt behavior patterns, and simple patterns at that.
Concretely, we will be able to administer the paper and pencil
test, and in some instances be able to predict several be-
havioral patterns fairly accurately. If we find "pure" 3
and 2 types, we will be able to easily distinguish them on
several counts (cf. Table 1).

The findings have potential use in psychotherapy. If the
£_type is seen as basically desirable, we could suggest be-
havioral modification on the part of the p, Thus, a p_com-

plains that he is socially rejected, that he feels lonely,

29

and contemplates suicide. Hoping that his disturbance is
not too deeply rooted, we may suggest that he endeavor to
converse more often at class, that he join in larger conver-
sational groups, that he sit towards the front of the class-
room, etc.‘ Relying, then, on a tenet of behavioral therapy,
we hope that this overt behavior change will favorably
affect other patterns of behavior. Again, we assume that
this sort of social competence reflects a better "adjusted"
personality, a happier person, etc.

A brief look at the non-significant results is warranted.
The measure of participation in class along with its several
dependent measures indicates little or no difference between
groups. This last mentioned variable with its results was
the most disappointing. In each of the groups, there were
no more than five instances of participation in class by the
whole group! Consequently, measures of duration, manner,
and content of participation were rendered useless. If any
conclusion may be drawn, it is a critical one concerning a
system of education which elicits such a parsimonious response
from students.

Non-significant results of the measures of arrival and
departure probably reflect faulty estimation by observers
who were working under the difficulty of not knowing exactly
what they were looking for; and this was magnified by the
fact that only two observations were being made. Other limi-

tations to the present study entered in the form of lack of

30

reliability on the part of the observers. Again, not enough
time or variety of observation made cross-checks impossible.
Finally, there was no male/female discrimination.
Modification of the experiment would take the form of
an increased observation period (including perhaps five or six
actual classroom periods for each student). This would pre-
sumably work to improve the reliability of all measures, but
particularly those of rating and the arrival and departure
times. Recording of arrival times would not begin until after
the first two observation periods. The extension of the
present hypothesis discussed above could be measured by some
implementation of a study of actual conversational material.
Furthermore, this type of behavioral observation follow-
ing Ring's typology and implementation of the Performance
Style Test may well be extended to other areas, thus sharpen-
ing the concepts on which the test is based and the attention
to cues involved in the behavior patterns. For example, the
test could be administered to graduate students in clinical
psychology undergoing supervision in actual psychotherapy.
Observations of behavioral patterns in a given situation,
e.g., a perceptual isolation situation, could be made at
different times during the course of training. In this way,
we might be able to determine correlations between a person's
type of social competence and say, his length of time spent
in therapy with a given client. Again, such an example only

indicates the varied potential of the test and the numerous

51

settings for behavioral observations or experimental manipu-
lations which are possible.

How far may the results of the present study be general-
ized? Are they applicable beyond the elementary psychology
student population of the actual testing situation? One
factor may influence the extent of generalization, specifical—
ly generalization across age groups. Due to the qualities
being measured and discriminated by the performance style
test, it seems that people's styles may change with age.

In general, there seems to be a slight evolving deve10pment
of composure in the majority of personalities which seems to
be reflected in a transition from the E type of social inter-
action to that of an £_or a 9, Thus, the self-conscious and
easily embarrassed adolescent will probably become, to some
extent, the self-confident adult. Use of the Performance
Style Test itself might be used to check such a possibility,
simply by comparing proportions of the typology across age

groups.

REFERENCES

Berne, E. Games People Play. New York: Grove Press, Inc.,
1964.

Christie, R. "Impersonal interpersonal orientations and
behavior," Unpublished research proposal, Columbia
University, 1962. ‘

Dunnette, M. D. "Fads, Fashions, and Folderol,“ Amer.
Psychol. 21, 545-52, 1966.

Goffmann, E. The Presentation of Self in Everyday Life.
New York: Doubleday & Company, Inc. 1959.

Jones, E. E., Gergen, K. J., and Davis, K. E. Psych. Mono.
76. No. 2 Whole No. 521, 1962.

Jones, E. E. Ingratiation. New York: Appleton-Century—
Crofts, 1964.

Manzel, E. W. "Naturalistic and experimental research on
primates," Human Development 10: 170—86, 1967.

Rausch, H. L. "Naturalistic aspects of the clinical
research method," Human Development 10: 155-69, 1967.

Ring, K., Braginsky, D., and Braginsky, B. "Performance
styles in interpersonal relations," Psychological
Reports, 18, 205-20, 1966.

Ring, K., Braginsky, D., and Levine, L. "Performance styles
in interpersonal behavior: an experimental validation
of a typology," J. of Experimental Social Psychol.

3: 140-59, 1967.

Sechrest, L. "Naturalistic methods in the study of social
attitudes," Human Development 10: 199—211, 1967.

 

Willems, E. P. "Toward an explicit rationale for naturalis-
tic research methods," Human Development 10: 155-69,
1967.

52

APPENDICES

APPENDIX 1

Instructions, items, and scoring keys

for the Performance Style Test

For each of the statements presented below you are to decide

whether it is true pp applied pp_you p£_false §§_ip pg

 

applied pp_you.

You are to mark your answers on the answer sheet provided.

If a statement is true or mostly true, as applied to you,
blacken in the first column space opposite the corresponding
item number. If a statement is false or not usually true,

as applied to you, blacken in the second column space opposite

the corresponding item number.

Make sure, in using the answer sheet, that the number of the
statement agrees with the number on the answer sheet.

Please do not omit any items.

HID

"le

1. I would be uncomfortable in anything other than -
conventional dress.

2. If given the chance I would make a good leader F T F
of people.

5. I have skill in influencing others. F T F

55

4.

10.

11.

12.

15.

14.

15.

54

I must admit that I enjoy trying to manipulate
others for my own purposes.

I like to do things that other people regard
as unconventional.

I often find it's difficult to get peOple to
do me favors, even when I have a right to
expect them.

When in a group of people I have trouble
thinking of the right things to talk about.

I find it easy to get along with people.

I dislike having to behave according to rules
of etiquette.

In most social situations, I feel tense and
constrained.

I can fit in pretty easily with any group of
people.

It's usually easy for me to persuade others
to my own point of view.

I like to conform to custom and to avoid
doing things that people I respect might
consider unconventional.

I think I could be a successful businessman,
if I wanted to.

I like to avoid situations where I am expected

to do things in a conventional way.

IH

In

16.

17.

18.

19.

20.

21.
22.

25.

24.

25.

26.

27.

28.

35

I usually find it difficult to change someone
else's opinions.

When serving on a committee, I like to be
appointed or elected chairman.

I must admit I try to see what others think
before I take a stand.

I can easily make other people afraid of me,
and sometimes do for the fun of it.

A person should adapt his ideas and his
behavior to the group that happens to be with
him at the time.

I do not mind meeting strangers.

I think I'd enjoy being an actor (or actress).
At parties I am more likely to sit by myself
or with just one other person than to join

in with the crowd.

I can usually get people to do what I want.

I usually have trouble making myself heard

in an argument.

I like to be the center of attention in a group.

PeOple can pretty easily change me even though
I thought my mind was already made up on a
subject.

Even the idea of giving a talk in public makes

me afraid.

IH

IO

29.
50.
51.

52.

35.

34.

55.

56.

57.

58.

59.

40.

41.

42.

45.

44.

56

I think I would enjoy being a salesman.

I like to meet new people.

I don't like participating in formal ceremonies.
If I'm with someone I don't like, I usually
don't express my real feelings to him.

I like to follow instructions and do what is
expected of me.

I find it hard to talk when I meet new people.
I frequently feel intense sympathy for others.
I enjoy being with people who are suave and
sophisticated.

I think it's important to learn how to obey.

I think most people would lie to get ahead.
When in a new situation, it's best to watch
what others do.

I enjoy being the host (or hostess) of a party.
I feel I can handle myself pretty well in most
social situations.

I sometimes enjoy misleading peOple just for the
fun of it.

I can deceive people, if I have to, without
feeling guilty about it.

I don't mind pretending to like someone when I

really don't if there's a good reason to do so.

['1

H *3

'13 "HID

'11

29.
50.
51.
32.

55.

54.

55.

56.

57.

58.

59.

40.
41.

42.

45.

44.

56

I think I would enjoy being a salesman.

I like to meet new people.

I don't like participating in formal ceremonies.
If I'm with someone I don't like, I usually
don't express my real feelings to him.

I like to follow instructions and do what is
expected of me.

I find it hard to talk when I meet new people.
I frequently feel intense sympathy for others.
I enjoy being with people who are suave and
sophisticated.

I think it's important to learn how to obey.

I think most people would lie to get ahead.
When in a new situation, it's best to watch
what others do.

I enjoy being the host (or hostess) of a party.
I feel I can handle myself pretty well in most
social situations.

I sometimes enjoy misleading peOple just for the
fun of it.

I can deceive people, if I have to, without
feeling guilty about it.

I don't mind pretending to like someone when I

really don't if there's a good reason to do so.

IH.

I0

45.

46.

47.

48.

49.

50.

51.

52.

55.

54.

55.

57

I like people to notice and to comment upon

my appearance when I am out in public.

I often feel like telling peOple what I really
think of them.

I feel ill at ease with people I don't know.

I have no dread of going into a room by myself
where other people have already gathered and
are talking.

I am a good mixer.

I like to go to parties.

In general, I find that I dislike non-
conformists.

I don't like to be too conspicuous at social
gatherings.

I should like to belong to several clubs or
lodges.

I often find that my wishes conflict with those
of others.

I feel guilty whenever I have done something

I know is wrong.

T

IH

IO

58

Percentile norms, means, standard deviations,
and ranges for the three scales of

the PST, according to sex of respondent

p scale £_scale g_scale
Score Males Females Males Females Males Females
41 99
4O 98 MAXIMUM POSSIBLE
SCORE

59 99 99 97 99 p = 54
38 98 96 96 98 r = 50
57 97 97 95 97 c = 47
56 97 97 90 95
55 96 95 87 95 99
54 94 94 85 89 99 98
33 95 95 80 85 97 97
52 92 90 76 80 96 94
51 90 87 70 76 95 88
50 88 85 63 70 89 83
29 85 81 56 63 86 75
28 80 76 51 55 78 68
27 76 71 46 50 69 58
26 75 67 58 44 60 48
25 67 62 51 58 55 58
24 62 55 26 52 46 51
25 55 49 21 28 33 23

22 50 40 17 23 26 16

Score
21
20
19
18
17
16
15
14
13
12
11

Mean

Range

p.scale

Males Females

45 55

55 50

29 25

21 18

14 15

10 9

8 6

4 4

2 2

1 2

1

23.44 24.24
6.17 6.13
8-42 9-42
591 461

59

£,scale
Males Females
15 18
11 14
10 12
8 10
6 8
5 6
4 5
2 5
2 2
2 2
1 1
28.18 27.01
6.57 6.25
9-45 7-45
591 461

g_scale
Males Females
21 12
15 10
11 8
6 5
4 5
2 2
2 1
1
25.15 26.45
4.56 4.52
15-36 10-57
391 461

FRONT

APPENDIX 2

DATA COMPILATION FROM RUNNING LOG

Class begins __;__ Student arrived .ppa__
Class ends __3___ Student departed __g__

CONVERSATION PARTICIPATION
Frequency

(no. of times)

PeOple involved
(X no. for each time)

Durgtien
(X no. of secs.)

Manper
(x rating)

Congent
(X rating)

*************

SEATING CHART

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

40

BACK

”7:11:77 11': 11111111171517 11111111111711“