Citation
The effects of anxiety-arousing messages in groups on qualitative dimensions of feedback in high and low communication apprehensive individuals

Material Information

Title:
The effects of anxiety-arousing messages in groups on qualitative dimensions of feedback in high and low communication apprehensive individuals
Creator:
Harris, James David, 1953-
Publication Date:
Language:
English
Physical Description:
xi, 93 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Anxiety ( lcsh )
Communication in small groups ( lcsh )
Dissertations, Academic -- Speech -- UF ( lcsh )
Speech thesis Ph. D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1981.
Bibliography:
Bibliography: leaves 87-91.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by James David Harris.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright James David Harris. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000295665 ( ALEPH )
07934267 ( OCLC )
ABS2013 ( NOTIS )

Downloads

This item has the following downloads:


Full Text
THE EFFECTS OF ANXIETY-AROUSING MESSAGES IN GROUPS
ON QUALITATIVE DIMENSIONS OF FEEDBACK IN HIGH AND LOW COMMUNICATION APPREHENSIVE INDIVIDUALS
By
JAMES DAVID HARRIS

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1981




ACKNOWLEDGEMENTS

There are many individuals I wish to thank for their help, advice, and support in the development of this dissertation. Very special thanks is extended to my chairman, Tony Clark. Your vigorous and sincere concern for quality in this text is indicative of your true professionalism. Thank you for the numerous critical revisions and advice from start to finish of this dissertation, without which, this project would still be in its inception period.
To Tom Abbott, who will always have my most sincere admiration and respect, thank you for your attention and counsel in this text as well as in my academic progression. I will always remember the stories, problems, laughs, disappointments and the triumphs (even if few) that we shared in Room 335.
To Paul Jensen, thank you for revealing to me what true professional p ide is all about. Your recommendations and support have been a constant inspiration not only in this manuscript but also in my personal life.




To David Robinson, thank you for the innumerable hours of counsel and guidance in the analysis of these data. I truly admire and have greatly benefited from your exceptional proficiency in design, interpretation and instruction of statistical theory and practice. I will always be indebted to you for your sincere concern for me as a past and present student.
To Marvin Shaw, special thanks are extended to you, sir, for your insightful contributions and your patient indulgence in sitting on yet another committee. You have demonstrated to me by example what it is to be a gentleman and a scholar. I will always be proud to proclaim that I had the honor of having you serve on my committee.
To Don Williams, the term "professor" has never been more aptly applied. You have instilled in me, by example, a genuine concern for conscientious scholarship and professional integrity. And above all, you have always given me your personal respect and admiration. You will always have mine.
Thanks are extended to Ted Chandler, Katie Poschall and Trev Garner for your considerable indulgence in the time consuming task of recording 990 individual ratings.

iii




Very special thanks are extended to Tom Samne and Doug BocK for lending your expertise to the analysis of these data. Your professional expediency in these analyses will always be remembered.
This dissertation is dedicated to my parents. Can
you believe I finally made it? It's been a long and hard road for all of us. Yet through your endless love and devotion you have taught me pride for personal and academic excellence. I hope that you are as proud of this dissertation as I am proud of being your son.
This dissertation is also dedicated to Melanie Ferriera, the soon to be Mrs. Harris. You've stuck by me when times were bad, pushed me and praised me at all the right moments, and you never let me give up. And most importantly, you have given me your infinite strength and limitless love, without which I would have never finished. If you can live with me through this, you can live with me through anything. Thank you Mel. I love you very much.
A special expression of thanks is extended to my
unparalleled friend, Jim Hicks. Thank you for the untold personal favors and the pleasureable experiences we have shared. Long live the Derby, and long live our friendship.




Special appreciation and thanks are extended to
Mrs. Voncile Sanders for your immense patience, understanding and quality workmanship in the preparation of the typed text.
And finally, to Don Shores, who has shown me what a
friend and Christian is; to Laurie Weiman, who is the only woman in the world to beat out Farrah Faucett as my "pin-up" girl and to the entire cast and crew of the third floow, thanks for everything.




TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS . ii
LIST OF TABLES ix
ABSTRACT x
CHAPTER
I INTRODUCTION 1
II REVIEW OF LITERATURE . . . . . 4
The Effects of Process Disrupting
Messages on Group Communication 5
History of Process Disrupting
Messages on Group Communication. 5
Methodologies and Research
Findings 7
Limitations 12
The Effects of Communication Apprehension on Group Communication. 15
Defining Communication Apprehension 17
Measurement of Communication
Apprehension 19
Observer rating scales . . 19 Physiological techniques . 21 Introspective measures . . 22 Behavioral correlates of communication apprehension. 23




TABLE OF CONTENTS (Continued)

Page

Summary of Research Findings.
Summary.. .........
RATIONALE AND HYPOTHESES ......... METHODOLOGY .... ............

32

Subjects ....... ........ 32
Design ..... ............... 32
Experimental Procedure ........ .. 33
The "Confederate" ............ 35
Independent Variable: Process Disruption Statements ........ 37
Independent Variable: Communication Apprehension .. ......... 39
Dependent Measure: Quality of Group Feedback ... ........... 45
Summary ..... ............... 46

V RESULTS AND DISCUSSION .......
Results .... ............
Interrater Reliability. .
Test of Hypotheses ....
Discussion ... ..........
Implications .. .........
Note .... ............
VI DIRECTIONS FOR FUTURE RESEARCH.

APPENDICES
APPENDIX A--FEEDBACK RATING INSTRUMENT.
APPENDIX B--McCROSKEY'S PRCA-COLLEGE. .

APPENDIX C--PRETEST OF ANXIETY AROUSING
MESSAGES .... ..........
APPENDIX D--SAMPLES OF GROUP INTERACTIONS

vii

III IV

. . 48 . 48 . 48 . 49

S. . 66

. 70
. 74




TABLE OF CONTENTS (Continued)

Page

APPENDIX E--CORRELATION COEFFICIENTS FOR
INTERRATER RELIABILITIES . . REFERENCES .
BIOGRAPHICAL SKETCH .

viii




LIST OF TABLES
Table Page
1 A Taxonomy of Process Disrupting
Message Forms................16
2 Analysis of Variance for Relevancy-Judge #4...................51
3 Analysis of Variance for Relevancy-Judge #5...................51
4 Analysis of Variance for Atomization-Judge #4...................52
5 Analysis of Variance for Fidelity-Judge #4...................52
6 Analysis of Variance for Tension-Judge #4...................54
7 Analysis of Variance for Involvement-Judge #4...................56




Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF ANXIETY-AROUSING MESSAGES IN GROUPS
ON QUALITATIVE DIMENSIONS OF FEEDBACK IN HIGH AND LOW COMMUNICATION APPREHENSIVE INDIVIDUALS By
JAMES DAVID HARRIS
August 1981
Chairman: Anthony J. Clark Major Department: Speech
The purpose of this study was to explore the effects of anxiety arousing messages on the quality of group feedback when presented in groups composed of high and low communication apprehensives. The experimental design was a 2 x 2 analysis of variance determined by the level of communication apprehension (high-low) and type of stimulus statement (unequivocal personal commitment-facetiousness). Trained confederates injected the predesignated stimulus statements into three-person groups, one of the three individuals being the confederate. Leathers' Feedback Rating Instrument (1969) was employed in the analysis of the first five feedback responses following the injection of the stimulus statement.




Three trained judges evaluated the audio-tape recordings of the group discussions. overall reliability coefficients between the three judges on each scale item across all groups were found to be quite low. The experimenter then recruited two expert judges to rate the groups' verbal behaviors. As was the case with the initial three judges, the reliability coefficients between the two expert judges were quite low. The test of hypotheses consisted of the separate evaluations of each of the two expert judges for each scale item.
Results indicated that high communication apprehensives exhibited less relevancy, atomization, fidelity and involvement as well as greater tension in their feedback responses to anxiety around messages than did low communication apprehensives. Since the interjudge reliability coefficients indicated that all the judges were not rating the same verbal behaviors in similar ways, discussion of these results focused on the validity of the Feedback Rating Instrument as a measure of small group communication process.




CHAPTER I
INTRODUCTION
Scholars who are interested in small group dynamics are constantly asking the question, "Is the system working?" In most cases, this question is approached by analyzing the quality of the group's output. Investigations have determined that the quality of the group product is directly related to the quality of the group's communication behaviors (Leathers, 1972). Therefore, communication scholars have focused on how continuing communication interchanges between group members affect the process by which groups reach decisions.
The growth of investigations into the process of group communication has prompted scholars to develop many varied styles of analysis, ranging from interaction analysis (Bales, 1950; Crowell and Scheidel, 1961; Samne, Shulman and Emerson, 1974; Stech, 1970) to orientation behaviors (Kline, 1970, 1972; Marr, 1974). These forms of analysis either record the frequency of certain types of communications, such as information giving communication, information seeking communications, procedural communications, or they analyze the
1




effects of facilitating communications injected into a group. Seldom however, have researchers considered qualitative dimensions of group communication; in particular, research is space concerning the qualitative aspects of how groups manage message forms that create tension, impede progress, or generally disrupt the process of the group.
Within this area of study, research findings indicate that certain process disrupting messages create high levels of anxiety in the receiver, as evidenced by tension, word omissions, word repetitions, and confusion in feedback responses as well as more irrelevant and digressive messages. However, no research to date has concentrated specifically on how anxiety-arousing communications affect group process. Further, researchers have not considered how anxiety-arousing communications affect individuals who reflect high levels of anxiety in their interpersonal communications.
The purpose of this study is to investigate the effects of anxiety-arousing message forms on the qualitative dimensions of group feedback in high and low communication apprehensive individuals. When presented with an anxiety-arousing message, will individuals considered high in communication




3
apprehension respond differently than those with low communication apprehension? What is the duration of the impact of these anxiety-arousing communications on the group process? And finally, what are the preferred verbal strategies by which group members cope with deviant, unexpected or disruptive communications?




CHAPTER II
REVIEW OF LITERATURE
The effect of process disrupting messages on small
group communication is a relatively new area of investigation in speech communication research, and has grown out of the acceptance within the field of speech that communication itself is a process; that communication researchers should focus primarily on the process rather than the antecedents of group dynamics to more fully understand the impact of human communication in group interaction (Sheidel and Crowell, 1966). Similarly, communication apprehension research is also a new research tradition within speech communication. Investigations of communication apprehension have ranged from identifying the apprehensive (Phillips, 1968; Clark, 1973) to the analysis of communication apprehensive's behaviors in group discussions (Wells and Lashbrook, 1970). Yet, specific aspects of these varied areas of study mandate special considerations.
The purpose of this chapter is to review the relevant literature in process disrupting message forms in group communication, as well as the definitional problems,
4




measurement and the behavioral correlates of communication apprehension.
The Effects of Process Disrupting Messages on Group Communication
This section will include a history of process disrupting messages in group communication, methodologies employed and research findings, and the limitations of investigations currently put forth.
History of Process Disrupting Messages on Group Communication
The study of process disrupting message forms on group communication grew out of the recognition that most small group investigations have been primarily product (output) oriented; that is, small group researchers have traditionally concerned themselves with such areas as task efficiency, group consensus, and solution quality. Communication scholars have recognized such a product orientation "gives no precise indication of the number or nature of factors that shape communication within the small group" (Leathers, 1969, p. 287). Communication researchers have therefore increasingly concerned themselves with a process orientation, focusing on those factors that through time impinge upon the communication patterns, flows and exchanges within the small




group which then aid in determining or shaping a group's product. As Leathers has stated, "at best, the productoriented experimenter starts with an effect, the group solution, and tries to infer the causes of that effect" (1969, p. 287).
One of the more outstanding research traditions of
group communication process is that of the effects of process disrupting statements on group interaction. Leathers (1969) conducted the initial investigation in this area studying the effects of high and low-level abstractions, unequivocal personal commitments, implicit inferences, and facetious interpolations on group process. This study has stimulated a limited yet varied amount of research into process disrupting effects.
Although no specific (operational) definition of process disrupting message forms has been accepted to date, the implications of this message type appear rather straightforward. A process disrupting message is basically any type of message which can potentially impede, impair, disorganize, confuse or in any manner contribute to disturbing the normal process of group discussion.




Methodologies and Research Findings
The primary focus of this research has been the analysis of the feedback response subsequent to a process disrupting message. Leathers (1969) developed the Feedback Rating Instrument (FRI) for such analysis. The Leathers FRI began as an investigation "to help determine whether any part of the small group process was affected by different stimulus statements" (p. 289). Transcripts from thirty discussion groups were content analyzed to determine:
(1) whether responses to various kinds of
stimulus statements possessed any common
set of characteristics; (2) whether the
strength of a given characteristic varied from response to response; and (3) whether any part of the small group communication
process was measurably affected by different
types of stimulus statements. (p. 289)
The analysis did indicate that "the responses to the stimulus statements did have common characteristics which varied in strength and that feedback was more immediately affected by the nature of the stimulus statements than by any other part of the communication process" (P. 289).
The feedback response indicated by Leathers is the "Y" response of Scheidel and Crowell's (1966) "X-Y-Z" definition of feedback sequences in which "participant (X) initiates




a comment which is followed by a comment from any other participant (Y) which in turn is followed immediately by a further comment from the first participant (X)" (p. 274). (The nine-factor Leathers FRI is found in Appendix A.) In applying the FRI to hundreds of feedback responses, Leathers (1969) found that the interobserver reliability estimates for each of the nine FRI scale items to be:
(1) deliberateness--.89; (2) relevancy-.89; (3) atomization--.82; (4) fidelity-.80; (5) tension--.62; (6) ideation--.85;
(7) flexibility--.86; (8) digression-.88; and (9) involvement--.86. (p. 291)
Similar reliability estimates have been obtained in other investigations (Leathers, 1970, Leathers, 1971).
As previously mentioned, Leathers (1969) conducted
the initial speech communication investigation of process disrupting messages in group communication. Five types of stimulus statements were investigated: high-level abstractions, low-level abstractions, unequivocal personal commitment, implicit inferences and facetious interpolations. Five expert judges unanimously agreed with two "lines" for each category, in that these statements "met their respective definitions" (p.288). Two trained confederates systematically introduced these stimulus statements at 6-minute intervals in the discussion groups, composed of four naive subjects




and the two confederates. Two trained judges applied the Leathers FRI to the feedback responses following each stimulus statement. The most important results, as reported by Leathers, were that feedback responses to high-level abstractions were confused, tense and withdrawn; feedback responses to implicit inferences were rated very unthinking, irrelevant and digressive; and feedback to facetious interpolations were especially personal and inflexible.
Leathers (1970) also investigated the effects of
trust-destroying communications on small group behavior. In the first of a two-part experiment, twenty naive subjects participated in a game with an experimenter confederate, where through cooperation and verbal reinforcement, the confederate established a high level of mutual trust with the subject, as verified in post-experiment surveys. in the second part of the experiment, the confederate and the subject were placed in a group comprised of four other students, comprising a six-person group. For the first thirty minutes of the group discussion, the confederate reinforced the trust earlier developed between him and the subject through statements which were very supportive and agreeing. However, in the last thirty minutes, the confederate injected stimulus statements at approximate




six-minute intervals that were designed to be highly trustdestroying. Two judges evaluated the feedback responses of the subjects to the trust-destroying messages. The results indicated that responses to trust-destroying messages "became particularly tense, inflexible, and personal" (p. 186).
In a subsequent investigation of trust-destroying
messages in group discussion, Prentice (1975) analyzed the degree to which nonfluencies were present in (1) the naive subject who was the target of the trust-destroying messages and (2) other group members who witnessed the trust-destroying behavior. Written transcripts were made from fifteen of the original twenty groups used in a Leathers (1970) study. Two judges verified the accuracy and completeness of the transcripts as compared with the tape-recordings of the discussions. A seven-category system was used in the analysis, consisting of (1) word repetitions, (2) stuttering, (3) sentence changes/incompletions, (4) omission,
(5) intruding incoherent sounds, (6) "you know(s)," and
(7) "I mean(s)." Additionally, a nonfluency ratio of the number of nonfluencies by the number of words spoken was conducted.




Results indicated that the nonfluency ratio increased during the trust-destroying segments of the discussions. The initial responsesof the subjects were found to have more word repetitions and more incoherent sounds following the trust-destroying messages than at other times in the discussion. Also, when the subject's verbal contributions were not a direct response to the stimulus statement, the feedback had more sentence changes, more word omissions and more incoherent sounds. Finally, the verbal contributions of those group members witnessing the trustdestroying message had more word repetitions and incoherent sounds than at other times in the discussion.
Building upon previous findings, Leathers (1971) investigated the effects of high, medium and low levels of abstraction, implicitness and facetiousness on feedback in group communication. The stimulus statements were developed in the following manner. Leathers devised ninety "potential contributions' for each of the three statement categories and varied their degree of abstraction, implicitness and facetiousness, respectively. Each statement was randomly assigned and read into a tape recorder which was then presented to six different college public speaking classes where students rated each statement as to its degree




of abstraction, implicitness and facetiousness. Two confederates memorized the statements and randomly introduced these statements at approximate 6-minute intervals into six-person groups (including the confederates). Two trained judges applied the Leathers FRI to the feedback responses following each stimulus statement.
The results indicated that as the abstraction level went from low to medium to high, that the feedback became more confused, tense and withdrawn. Also, as the level of facetiousness rose from low to medium to high, feedback became more personal and inflexible.
Limitations
The gaps in what is known about process disrupting
messages as a result of this research tradition are perhaps more glaring than any substantive contributions to knowledge in this area. For example, no data have been presented concerning similarities and differences in how these various messages influence group communication. Since most investigators have employed dependent measures to tap a single dimension of group communication, our understanding of the impact of any specific message on group communication is exceedingly limited. Few attempts have been made




to assess the full range of effects that a highly aversive message could have on subsequent communications with a group. Similarly, we have little information regarding either the duration of the impact of these various messages or the preferred verbal strategies by which group members cope with deviant, unexpected or disruptive communications.
Several methodological problems are also evident in prior research. First, researchers have instructed the confederates to inject the process disrupting statement at 6-minute intervals without regard to the particular phase in which the group was currently progressing. This could potentially lead to an injection of a process disrupting message which was awkward or inappropriate to the prevalent topic discussion. Second, most investigations have used judges to verify the validity and reliability of the stimulus statements with no further qualifications. Third, few types of process disrupting messages have been investigated. Many other forms of message types might potentially disrupt the communication in the group. Yet those that have been investigated introduce a definite level of anxiety in the receivers, as indicated by feedback




that is tense, confused, etc. Fourth, and most prominent, in all previous studies, typed transcripts of the verbal feedback responses were made from video-tapes of the groups' discussions. On these transcripts, the experimenters presented their analyses of the nonverbal behaviors elicited prior to, during and following the verbal message, such as:
Feedback response: (nervous laughter and discussants grin at each other in
an embarrassed way.) It seems that
what? What does that mean? All right,
why don't you explain cognitive aversion
or was it occlusion? (More nervous
laughter.) Maybe the problem is, ah,
maybe it is not an occlusion, maybe it is a problem of cognitive prerogatives.
Oh, hell, let's drop the subject.
(Leathers, 1971, p. 188)
The judges then evaluated the feedback responses given the verbal description in conjunction with the experimenter's evaluation of the nonverbal behaviors elicited. These verbal descriptions of certain specific nonverbal behaviors could be suspect to reliability error in the judges' ratings as well as the attempt to describe such a global concept as nonverbal communication in a few, specific words. And finally, the area of process disrupting message effects on small group communication suffers from the same theoretical flaws as other small group research in that




no efforts have been made to systemitize research findings into a conceptual taxonomy of process disrupting message forms and their various effects on group communication.
In particular reference to the final limitation cited, Table 1 represents a taxonomy of process disrupting message forms and their effects on group communication. Given the limited range of types of process disrupting statements previously investigated, several other types are included which could potentially be process disrupting. All of these various message forms fit into four conceptually distinct category types. Hence, the first attempt at creating a conceptual taxonomy of process disrupting message forms.
The Effects of Communication Apprehension on Group Communication
Within the field of speech communication, few areas
have grown as rapidly in recent years as that of communication apprehension. Three areas of communication apprehension research appear relevant to the study of group communication: defining communication apprehension, measuring communication apprehension, and the behavioral correlates of communication apprehension.




Table 1

A Taxonomy of Process Disrupting Message Forms

Category Message Descriptor Effect on Group Communication
I Determinant high level abstraction Feedback became more confused,
Interactions tense, and withdrawn
implicit inferences Feedback became more unthinking
and automatic, less relevant, and more digressive
II Social-Digres- phatic
sive Messages facetiousness Feedback became more personal and
inflexible
III Content refutational messages
Challenges
_- ideological attacks
IV Personal Feedback became more tense, inChallenges trust-destroying flexible and personal; more word
repetitions, more incoherent sounds, more word omissions, more sentence changes
.personal threat
unequivocal personal commitment




Defining Communication Apprehension
The anxiety of communicating has been referred to by a number of labels, most notably "stage fright," "speech anxiety," "communication apprehension," "unwillingness to communicate," and "reticence." It is important to distinguish these terms, for one of the major difficulties in this area of investigation is the tendency of communication researchers to use the terms interchangeable. Greenleaf (1947) defined stage fright as
An evaluative disability, occurring in
social speech situations, and characterized by anticipatory negative reactions of fear, avoidance, and various
internal and overt manifestations of
tension and behavioral maladjustments.
(p. 134)
Similarly, Clevenger (1955) defined stage fright as
Any emotional condition in which emotion
overcomes intellect to the extent that
communication is hampered, either in
audience reception or in speaker selfexpression, where the immediate object
is the speech-audience situation.
(p. 30)
As evidenced in these definitions, stage fright is only concerned with the public speaking format.
By the mid-1960's, investigators recognized that communication anxiety extended beyond the confines of the public speaking format. Exemplary of this trend was




Thorenson's (1966) account of communication anxiety when he described the communicatively anxious individual as one
Who feels uncomfortable in social group
situations, who does not perceive himself
as capable of displaying any social
initiative and who describes himself with
such adjectives as timid, awkward, and
reserved. (p. 207)
As the realization grew that an individual's reluctance to communicate might not be exclusively limited to public speaking, new terms were coined and developed to expand and amplify this concept. Giff in and Gilham (1971) defined speech anxiety as
A person's unwillingness or reluctance
to rely upon himself in a communication
situation. His self concept is at stake,
and he perceives his speech ability as inadequate to carry him through to the objective, whatever it may be, in that
setting, and the speaker is usually painfully aware of his condition. (p. 70)
Similarly, and most notable of these concepts, McCroskey (1977) defined communication apprehension "as an individual's level of fear or anxiety associated with either real or anticipated communication with another person or persons" (p. 78).
The concepts of unwillingness to communicate and
reticence refer to a more global construct of communication




anxiety (McCroskey, 1977). Burgoon (1976) defined unwillingness to communicate as a "chronic tendency to avoid and/or devalue oral communication" (p. 60). Yet Phillips (1968) recognized that many individuals have very high levels of speech anxiety/communication apprehension/ unwillingness to communicate and described these individuals as reticent, where reticence is defined as the "person for whom anxiety about participation in oral communication outweighs his projection of gain from the situation" (p. 40).
Measurement of Communication
Apprehension
A second major area of emphasis in the area of communication apprehension is the measurement techniques employed to assess the communication apprehensive. Three main categories of communication apprehension assessment have been employed: observer rating scales, devices for measuring physiological changes during speaking, and introspective or "self-report" techniques. observer rating scales
Although observer rating techniques have a long history, their use seems to produce inconsistency. Dicken,




Gibson and Prall (1950) found the true reliability of judges in observing and recording communication apprehension in a five category system to be .57. Similarly, Eckert and Key (1940) found that groups of three and four judges produced reliability coefficients of .68 on scales similar to that of Dicken et al. More recently, Mulac and Sherman (1974) developed an instrument for the behavioral assessment of speech anxiety (BASA). Even though in the initial development of the scale the reliability was shown to be quite high (.95), more research utilizing the BASA is needed to assure its widespread reliability and its internal/external validity. As Clevenger (1959) notes, most observer rating techniques (including the BASA) have several drawbacks in that observer rating techniques are not always accurate representations of the communication apprehensive speaker. Those behaviors related to communication apprehension are at best difficult to observe. Additionally, there would be a need to observe individuals across a broad spectrum of communication situations to accurately determine the level of a person's apprehension. Also, in almost every case of observer rating scales, trained observers are necessary to administer and record the behaviors, whereas other techniques do not necessitate such expertise.




Physiological techniques
Physiological techniques of measuring communication apprehension have been successfully employed (Porter, 1974). Many techniques have been devised, including measures of communication apprehension as changes in blood pressure and pulse rate before and after public speaking (Dicken and Parker, 1951), in autonomic arousal (Porter, 1974), and most prominently, in heart rate changes (Behnke and Carlile, 1971; Carlile and Behnke, 1973; Porter and Burns, 1973). Even though the reliability of these measures has been quite high, the relative expense of the mechanical devices have many of the same drawbacks as the observer rating techniques. For example, it is improbable if not impossible to obtain physiological indices of communication apprehension during every communication transaction. Accordingly, to obtain a reliable index of communication apprehension via physiological measures, a variety of communication transactions must be measured (McCroskey, 1970) As with observer rating scales, trained experts are needed in administering and recording measures of physiological changes during communication transactions.




Introspective measures
Introspective or "self-report" scales have traditionally been the most widely used index of communication apprehension. Although many authors have developed scales for the measurement of communication apprehension via self-reports, six scales have dominated the literature: (1) The Gilkenson Personal Report of Confidence as a Speaker (Gilkenson, 1942),
(2) The Watson and Friend Social Avoidance and Distress Scale (Watson and Friend, 1969), (3) McCroskey's Personal Report of Communication Apprehension-College (McCroskey, 1970, 1978), (4) Lustig's Verbal Reticence Instrument (Lustig, 1974), (5) Phillips and Erickson's Reticence Scale (Rosenfeld and Plax, 1976), and (6) Burgoon's Unwillingness to Communicate Scale (Burgoon, 1976).
Self-reports have several advantages over observer
ratings and physiological measures in that (1) self-report techniques do not require observation and measurement in a variety of communication contexts, (2) expertly trained judges are not required in the administration and recording of the results, (3) self-report techniques are much more cost efficient than observer rating or physiological measures in terms of man-hours per subject and equipment necessary for measurement, and (4) self-report techniques can be




administered to a large number of people simultaneously. Hence, self-report scales are currently the most used measures of communication apprehension. Behavioral correlates of
communication apprehension
Of particular interest to communication scholars have
been the behavioral correlates of communication apprehension. Communication apprehension has been shown to produce less self-disclosure, lower participation in small groups, less attraction to peers, less likelihood of being perceived as opinion leaders, more tension exhibition in small groups, less likelihood of being perceived as leaders, and more likelihood of choosing occupations which have lower communication demands even though they offer less income and status (McCroskey, 1976). Phillips (1968) found that high apprehensives report "shakiness" during classroom activities, felt "butterflies in their stomachs" when asked to speak to the class, found it necessary to end conversations because of their fears, reported an inability to talk to their teachers and supervisors, considered themselves as extremely quiet, felt compelled to apologize for their ideas when they were challenged, and expressed singular inability to talk to their parents. In a similar study, Phillips (1977)




found that the apprehensive did not have the ability to open conversations with strangers, to extend conversations, or to initiate friendships, to follow the group's discussion and make relevant remarks, to answer questions that arise on the job or in class, and a general inability to communicate characterized by a lack or avoidance of participation.
Recently, investigators have researched various personality correlates to communication apprehension. McCroskey, Daly and Sorenson (1976) as well as Lashbrook, Lashbrook, Bacon and Salinger (1979) found that communication apprehension was negatively correlated with tolerance for ambiguity. Additionally, McCroskey et al. (1976) found that communication apprehension was negatively correlated to self-control, adventurousness, surgence, and emotional security. Lustig (1974) found a negative correlation between communication apprehension and both self-esteem and self-acceptance. Other investigations have obtained similar results (McCroskey, Daly, Richmond and Falcone, 1977).
In investigations concerning communication apprehension in small group communication, Wells and Lashbrook (1970) found that high apprehensives interacted less in small groups than did low apprehensives and that their interactions were




less relevant than those who were not apprehensive. High apprehensives have been found to show more tension and less interest in both zero-history and intact groups (McCroskey, 1976). McCroskey (1976) reported that high apprehensives were perceived by other group members as less extroverted, composed, competent,socially attractive and task-attractive than low apprehensives. McCroskey and Richmond (1976) report similar findings concerning peer perception and apprehensives. Lustig and Grove (1975) found that groups consisting of low communication apprehensives had more speech acts, interacted more in socio-emotive areas and less in task areas than did groups consisting of both low and high communication apprehensives. In considering unwillingness to communicate, anomia-alienation and communication apprehension as predictors of small group communication, Burgoon and Burgoon (1974) concluded that "persons with a high unwillingness to communicate set will be more tense in their communication behaviors" (p. 36).
Summary of Research Findings
In summary, individuals who are considered to be high in communication apprehension generally view themselves




in less positive terms than low communication apprehensives. Similarly, high apprehensives are viewed less positively by others than are low apprehensives. High communication apprehensives are generally found to have less quality and quantity of interaction, particularly in small group discussions. Additionally, high communication apprehensives generally tend to avoid the initiation of interaction and quite often attempt to avoid interactions altogether.
There are many strengths in the area of communication apprehension research. First, the area of communication apprehension has had a steady and continuous growth of investigations from the early foundations of the field of speech communication. This growth had culminated into a highly intense research tradition. Second, there has been extensive recent inquiry into the instrumentation and methodological measurement of communication apprehension. Even though scholars are not totally satisfied with the methods of recording and evaluating communication apprehension, the amount to research and attention directed toward that question is indeed impressive. Third, communication apprehension focuses on a culturally pervasive problem in America, the fear of speaking. This strength is highlighted by the fact that communication apprehension is




is not a highly specific problem area with little applicability. Many Americans possess anxiety in their communications and therefore research in this area constitutes research on a real and consuming problem in this country.
There are several weaknesses in communication apprehension research. First, many terms have been coined in an attempt to define anxiety in interpersonal communication. Confusion can eminate from the misapplication or misinterpretation of conceptual frameworks of these various definitions. Second, there exists a need to extend the scope of investigations in communication apprehension. Most notably, further investigations are needed as to the causes of communication apprehension as well as the continuing need to further validate self-report scales purporting to measure communication apprehension. Third, and most important to this investigation, few studies have considered the effects of communication apprehension on group process. Several studies report that communication apprehensives behave differently in group contexts, but few studies focus on how these behaviors actually affect the group process.




Summary
This chapter has reviewed the relevant research regarding the effects of process disrupting message forms on group communication and the definitional problems, measurement and behavioral correlates of communication apprehension. Limitations of research to date have been highlighted in each area. The overall uniqueness of this review lies in the fact that many process disrupting statements injected in group communication process have been shown to create anxiety in small group discussions. Yet, these anxiety-arousing communications have not been studied in relation to the level of group member's communication apprehension. This study will incorporate both areas of current inquiry in an attempt to further validate process disrupting message effects on small group communication process.




CHAPTER III
RATIONALE AND HYPOTHESES
The limitations of previous research in process disrupting message effects on group communication are highlighted by the fact that communication apprehension has not been considered as a variable in the response quality of group communications following a process disrupting message. Process disrupting messages have been shown to introduce a definable level of anxiety into the feedback responses within the group. But, is the cause of this anxiety the process disrupting message or communication apprehension? The purpose of this study is to investigate the effects of anxiety-arousing message forms on the qualitative dimensions of group feedback in high and low communication apprehensive individuals.
The theoretical link between research on anxietyarousing communication and the wealth of current inquiry into communication apprehension is, however, tenuous at best. Nevertheless, the potential for expanding our knowledge about both phenomena (group communication and communication apprehension) is substantial for two 29




noteworthy reasons. First, research into communication apprehension provides us with measuring tools for identifying individuals who are thought to have either high or low thresholds for tolerating anxiety-arousing messages and their anxiety-inducing effects. Second, instruments of this sort allow us the opportunity to compare and contrast the effects of anxiety-producing statements upon individuals with extremely different tolerance levels for such anxiety, as well as their tendency toward and preferred modes of anxiety-reduction.
Previous research leads to several plausible hypotheses concerning the interrelationships of communication apprehension and the effects of anxiety arousing messages on group communication.
Hypothesis I. High communication apprehensives will demonstrate significantly less deliberateness in their feedback responses than will low communication apprehension.
Hypothesis II. High communication apprehensives will demonstrate significantly less relevancy in their feedback responses than will low communication apprehensives.
Hypothesis III. High communication apprehensives
will demonstrate significantly more atomization in their feedback responses than will low communication apprehensives.




Hypothesis IV. High communication apprehensives will demonstrate significantly less fidelity in their feedback responses than will low communication apprehensives.
Hypothesis V. High communication apprehensives will demonstrate significantly greater tension in their feedback responses than will low communication apprehensives.
Hypothesis VI. High communication apprehensives
will demonstrate significantly more ideation in their feedback responses than will low communication apprehensives.
Hypothesis VII. High communication apprehensives will exhibit significantly more flexibility in their feedback than will low communication apprehensives.
Hypothesis VIII. High communication apprehensives will exhibit significantly less digression in their feedback responses than will low communication apprehensives.
Hypothesis IX. High communication apprehensives will exhibit significantly less involvement in their feedback responses than will low communication apprehensives.




CHAPTER IV
METHODOLOGY
The purpose of this chapter is to summarize the
methodological and procedural steps employed in this investigation. Specifically, this chapter will address subject characteristics, research design, experimental procedure and measurement techniques.
Subjects
One hundred and seventy-seven undergraduates enrolled in sections of the basic public speaking course at the University of Florida volunteered to participate in this study for extra bonus points in the course. Of the total pool of volunteers, forty-four actually served as participating subjects in the experiment.
Design
The research design for this investigation was a
2 x 2 factoral with repeated and multiple dependent variables. Subjects who were variously classified as either high or low in communication apprehension (CA) were 32




I& Y-=
I
rp ck T rr TT
f

randomly assigned to one of two conditions of message disruption (MD).

Message Disruption

High Low

Communication Apprehension

Experimental Procedure

Subjects were informed, prior to volunteering for
participation, that the purpose of the study was to "examine
the quality of group decisions on problems of mutual interest
to both males and females." When subjects arrived for the
group sessions they were taken into a small room in which
the chairs were arranged in such a manner to assure equal
distance between all participants. The Experimenter provided the following instructions:
Today you are being asked to participate in a small group discussion concerning
the human relations problem just handed to you. At the bottom of that page, you will see a series of questions. These questions
were given only as a procedural guide, to allow your group common questions for discussion. Your group may choose to follow the order of the questions, but feel free
to discuss this topic in any manner the
group so wishes. I will return in approximately 25 minutes, so please begin your
discussion.




After reading the instructions to the group, the Experimenter began an audio cassette recorder visible to all
participants.
Subjects were provided a copy of the "human relations
problem" which served as the basis for group discussions.
The specific questions which followed the narrative of
the problem were designed to be both involving and timeconsuming:
Susan and John are junior Sociology majors
attending a large southern university.
They met in a freshman class and began
dating soon after. For the last six months,
they have been living together in Susan's
apartment. Last week, Susan received a phone call from her parents who told her that they are planning a visit with her
next weekend. Susan has not told her parents about her and John living together, and she does not want them to find
out (since they would not accept such an
arrangement). Susan, therefore, has asked
John to move all of his belongings out of the apartment for the weekend and to stay at a mutual friend's house. John refuses
to leave, saying that he has nothing to
hide and that he thinks it is very inconsiderate that Susan is asking him to do
SO.
A. Is Susan justified by asking John to
leave?
B. Is John being inconsiderate of Susan's
position?
C. Should Susan tell her parents about John
living with her?




D. What would you, as a parent, think of
this situation?
E. How would this situation be altered (if
at all) by the fact that John was a
graduate student (not an undergraduate)?
Upon completion of the group discussion, the participants opened the door to the experimental room. The Experimenter turned the tape recorder off, excused the "confederate," debriefed the subjects, and thanked them for their participation.
The "Confederate"
While each group was composed of three persons, only two of the group members in any group were subjects in the experiment. The third member of each group was one of the two "confederates" or "plants" who were trained by the experimenter and contributed the anxiety-arousing "stimulus statement" that served as the manipulation for each experimental condition. The "confederates" in the present study were two male undergraduate students at the University of Florida. These "confederates," although ignorant of the specific hypotheses under investigation, were given specific guidelines for verbal and nonverbal conduct during the group discussions. obviously, the contributions of the "confederates" could have posed a threat to the internal




validity of the study if uncontrolled. Consequently, "confederates" were cautioned (a) to refrain from initiating substantive themes in group discussion (i.e., proposed solutions, arguments either in favor or in opposition to a proposed solution, personal disclosures), (b) to appear interested and involved in the participation of the other group members (through forward body lean, eye contact, smiling, nodding and vocalizations which support and confirm), (c) to limit their interactions to a minimum,
(d) to utter the "stimulus statement" as close to the appropriate phase mark as was possible (between discussion on questions B and C of the problem) but to work the comment into the context of the group discussion so that the statement would not appear awkward or totally inappropriate, and (e) to permit the full and natural feedback responses of the group without either reiterating or compromising the force of the statement.
Recognizing that the aforementioned guidelines reflect subtle and important skills, the "confederates" practiced their roles with the Experimenter. only when the Experimenter determined that their group behavior met substantially the guidelines did the experimental sessions begin.




Independent Variable: Process Disruption Statements
As indicated in Chapter II, research has identified a variety of verbal statements which are potentially disrupting to the continuity of group communication. Using the taxonomy of disruptive message forms reported, messages were selected from three categories which could have a measureable impact on the level of receiver anxiety (i.e., messages which are considered anxiety producing). Consistent with these categories and drawing upon examples from previous investigations, the Experimenter composed ten statement for each of the three categories of anxietyproducing statements identified in the taxonomy. In addition, twenty statements considered non-anxiety producing were added (see Appendix C). This latter group included statements which are likely to be found in group discussions, but which do not generate anxiety by their function (i.e., they do not attack a person, refute a position or challenge group procedure).
The 50 items were rated by students in three sections of the basic Speech course at the University of Florida on a 7-point bi-polar scale. Students were asked to assess the likelihood that each statement would foster anxiety or tension when uttered in the context of a small group.




The particular scale used for this test is identical to the "tension" scale in Leathers FRI (1969). Items and responses were divided into "anxiety-producing" and ##nonanxiety producing" categories. A t-test (Williams, 1979) was performed in an attempt to validate assumed categorical differences between the two groups of statements and to assure the Experimenter that the anxiety producing statements were indeed perceived by others as likely to produce high levels of anxiety among members. Results were both statistically significant (p < .05, df = 58) and in the direction predicted as anxiety arousing were rated more likely to produce anxiety (x = 5.1) than were non-anxiety arousing statements (x = 2.7). (Because of the low subject pool of high and low communication apprehensives, only two categories of anxiety-arousing communications were used in this study.)
A further precaution was taken to ensure that the stimulus statements were anxiety producing and salient. Five experienced judges, using a Q-sort technique, divided those statements which achieved a mean rating in excess of the group mean for the anxiety arousing statements into four categories; very highly anxiety arousing, highly anxiety arousing, moderately anxiety arousing, and somewhat




anxiety arousing. Only those statements which were considered very highly/highly anxiety arousing were considered for this experiment. Then, two statements for each category were selected as the manipulation for each of the two conditions of anxiety arousing message forms. Each statement was used in three group sessions for that condition. The rationale for employing two different statements for a given condition was to provide some assurance that findings would not be specific to a single statement.
Independent Variable: Communication
Apprehension
As previously reported in Chapter II, a variety of
measures exist which purport to tap the concept of communication-bound anxiety. More than mere methodological differences, these instruments reflect important conceptual and theoretical differences in scope, situation and definition of communication-bound anxiety. The specific measure of communication-bound anxiety used in the present investigation was McCroskey's PRCA-College (Appendix B).
There are a variety of reasons for employing this particular measure. First, the PRCA-College reflects a conceptualization of communication-bound anxiety which is consistent with the philosophy of this investigation, namely,




that anxiety or apprehension can occur in a variety of communication contexts (from public speaking to small groups) and that a sensitive measuring device must be capable to tapping self-reports of anxiety across a variety of contexts.
Second, the PRCA-College has been recently revised
to include specific indicators of anxiety in a small group (McCroskey, 1978). Third, the PRCA actually includes four separate instruments which were tailored to specific age groups. The PRCA-College was specifically designed for and validated with college students, suggesting that it is highly appropriate for assessing apprehension levels among the targeted population for this study. Fourth, the PRCA has had a consistent history of high reliability with other measures as well as in test-retest reliability investigations (McCroskey, 1978).
Finally, and most prominent, a solid argument can be made for the validity of the PRCA. As McCroskey stated:
Probably the best indicator of the validity
of a measure is the degree to which it can
produce empirical results that are consistent with predictions based upon theory relating to the construct which the measure
purports to tap. (1978, p. 193)
To date, "five major theoretical propositions concerning oral communication apprehension have been set forth"




(McCroskey, 1978, p. 193). First, "people vary in the degree to which they are apprehensive about oral communication with other people" (p. 193). As McCroskey notes, "this assumption has been supported in every report of research which has employed the PRCA" (1978, p. 193).
A second proposition states that "people with high oral communication apprehension seek to avoid oral communication" (p. 194). Again, research has consistently supported this proposition. For example, McCroskey and Leppard (1975) found that high apprehensives preferred housing that through previous research had been identified as requiring less interpersonal contact than low apprehensives. Daly and McCroskey (1975) found that high apprehensives had a strong preference for occupations requiring less verbal interaction than low apprehensives. Similarly, McCroskey and Anderson (1976) found that high apprehensives preferred large lecture classes over small classes, which would require more verbal interaction. These studies, being typical of those conducted in this area, demonstrate "that the PRCA is able to predict communication avoidance behaviors that would be expected" in line with the theory of communication apprehension and "provides a strong indication of predictive validity of the instrument" (McCroskey, 1978, p. 196).




Third, "people with high oral communication apprehension engage in less oral communication than do less orally apprehensive people" (p. 196). When faced with situations that demand oral communication, communication apprehension theory would predict that individuals with high apprehension could withdraw from the interaction or remain as silent as possible. Many studies indicate that "high oral communication apprehensives engage in less oral communication behavior than do low oral communication apprehensives" (McCroskey, 1978, p. 196), again indicating support for the predictive validity of the PRCA.
Fourth, "when people with high oral communication
apprehension do communicate, their oral communication behaviors differ from those of people who are less apprehensive" (pp. 196-7). As previously noted, Wells and Lashbrook (1970) found that high communication apprehensives interacted less in small groups and that their interactions were much less relevant to the topic under discussion than were low communication apprehensive interactions. Similarly, and again previously noted, Sorenson and McCroskey (1977) found that high apprehensives exhibited more tension in small group interactions than did low communication appresives. These investigations as well as others in this




area "again provide support for a claim of predictive validity for the PRCA" (McCroskey, 1978, p. 197).
Fifth, "as a result of their oral communication behavior, high oral communication apprehensives are perceived less positively by others than are less apprehensive people" (p. 197). As previously reported, McCroskey (1976) found that high apprehensives were perceived by other group members as less extroverted, composed, competent, socially attractive and task-attractive than low apprehensives. McCroskey and Richmond (1976) have found similar results concerning peer perception and apprehension.
As summarized by McCroskey (1978):
The theoretical propositions underlying
the construct of oral communication apprehension have been consistently
supported in research employing the PRCA
to predict specific behaviors. These
results taken together provide a strong
indication of the predictive validity
of this instrument. (p. 198)
Given the results reviewed concerning the reliability of the PRCA and the evidence that the PRCA does predict those behaviors associated with the theory underlying the construct of oral communication apprehension (suggesting its validity), this investigation utilized the PRCA-College as the indicator of oral communication apprehension.




For the present study, 177 students enrolled in
sections of the basic Public Speaking course completed the 25-item PRCA-College. Subjects recorded one of five possible point values for each item ranging from "strongly agree" to "strongly disagree." The instrument has a possible range of total scores of 25-175 with a hypothetical mid-point of 75. According to McCroskey and other investigators, the standard deviation has ranged from 13 to 15. McCroskey (1978) has stated that:
Based on previous interviews with subjects scoring at the various levels, it has been
customary to consider subjects scoring
above 88 (one standard deviation above the
typical mean) to be "high" in communication
apprehension and those scoring below 58 to
be "low." While these cut-off points are
not perfect (there will be some error
particularly with subjects scoring very near
the cut-offs), they have been found to be
optimal in numerous studies. (p. 201)
The method of differentiation used in this study was to consider the individuals scoring one standard deviation below the mean to be low in communication apprehension and those who scored one standard deviation above the mean to be high in communication apprehension.




Dependent Measure: Quality
of Group Feedback
While a variety of procedures exist for measuring the quality of group communication, one measure seems particularly suited to the present investigation. Leathers (1969) FRI was selected as the most appropriate measure of group verbal feedback for several reasons. First, the intent of the instrument was to provide a multidimensional assessment of the feedback of group members to some form of contrived, process disrupting statement. Consequently, the methodology and applications of the instrument pioneered by Leathers are consistent, if not identical, to the methodology of the present investigation. Second, substantial refinement of the instrument has provided researchers with a technique for measuring verbal feedback which has been indicated to be both practical and reliable. Third, the nine scale items which comprise the rating instrument appear to tap the major qualitative dimensions of feedback which were important to the present investigation. Last, use of the Leathers' instrument provided a methodological link to a highly visible and acclaimed area of research in the field of Speech Communication. In this latter regard, results derived through use of the instrument could be




compared and contrasted with previous findings in the area of process disruption.
The procedure for applying the Leathers' instrument was rather straightforward. Using audio tapes recordings as the verbal record of the group discussion (see Appendix D), three judges independently rated each of the first five verbal utterances of group members following the injection of the designated stimulus statement into the group discussion on all nine scales of the instrument. Each scale is a seven-point bi-polar set of adjectives ranging from +3 through 0 to -3. Above each item is the scale label (i.e., deliberateness, relevancy, atomization, fidelity, tension, ideation, flexibility, digression and involvement) and beneath the numerical values which comprise the scale is a definition (in operative terms) of what behaviors are symptomatic of the adjective pairs. Leathers' FRI is contained in Appendix A.
Summary
This chapter summarized the major methodological
considerations employed in the present study, including subject demographics, research design, experimental procedure, confederate training, independent and dependent




47
variables. An analysis of variance was employed to determine the significant relationships between the variables.




CHAPTER V
RESULTS AND DISCUSSION
The purpose of this chapter is to present the results of this study and to discuss their impact upon the research hypotheses. Special attention is given to interrater reliability, test of hypotheses, and discussions of research findings.
Results
Interrater Reliability
Pearson Product Moment Correlations were computed
on the initial three judges' evaluations via: (1) overall reliability coefficients on each scale item across all groups, and (2) reliability coefficients on each cell across all nine scale items. overall reliability coefficients between judges on each scale item across all groups we-re found to be quite low (see Appendix E) .
The experimenter then recruited two expert judges,1 one in the field of small group research, the other an expert in rating error and scale implementation. As was the case with the initial three judges, the reliability 48




coefficients between the two expert judges were quite low (see Appendix E).
Since the reliability coefficients were low, and because averaging judges' ratings to test the hypotheses would be invalid die to the poor reliabilities, the subsequent evaluation of the research hypotheses consisted of two parts; each hypothesis was tested by evaluating the separate rating of each of the two expert judges, and then the experts ratings were compared/contrasted to discover how they perceived potential differences among the responses of experimental subjects.
Tests of Hypotheses
Hypothesis I--High communication apprehensives will
exhibit significantly less deliberateness in their feedback responses than will low communication apprehensives. Ratings of neither expert judge demonstrated a significant difference between high and low communication apprehensives regarding the deliberateness of their feedback responses to anxiety arousing communications.
Hypothesis II--High communication apprehensives will exhibit significantly less relevancy in their feedback responses than will low communication apprehensives. Although the ratings of the two expert judges did indicate




statistically significant differences between the responses of apprehensive and non-apprehensive subjects, the nature of the differences was not the same. The scores of Judge #4 indicated that there was a main effect for the stimulus statement (F = 4.36, p < .05, see Table 2), while the scores of Judge #5 showed a significant interaction effect between the stimulus statement and communication apprehension (F = 4.73, p < .05, see Table 3).
Hypothesis III--High communication apprehensives will exhibit significantly greater atomization in their feedback responses than will low communication apprehensives. In this instance the scores of Judge #4 indicated a significant main effect for communication apprehension (F = 4.38, p < .05, see Table 4), and also a significant main effect for the stimulus statement (F = 4.83, p < .04, see Table 4). The scores from Judge #5 did not demonstrate significant differences between the groups.
Hypothesis IV--High communication apprehensives will exhibit significantly less fidelity in their feedback responses than will low communication apprehensives. The scores from Judge #4 showed a significant main effect for the stimulus statement (F = 4.34, p < .05, see Table 5). The scores from Judge #5 did not indicate any significant differences between the groups.




Table 2
Analysis of Variance for Relevancy--Judge #4

Source df Type IV SS F Value P > F
A* 1 53.09184823 3.89 0.0640
B** 1 59.45784561 4.36 0.0512
A*B 1 22.67620577 1.66 0.2135
A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.

Table 3
Analysis of Variance for Relevancy--Judge #5

Source df Type IV SS F Value P > F
A* 1 1.96694600 0.28 0.6002
B** 1 15.78152167 2.28 0.1481
A*B 1 32.69320298 4.73 0.0432
A* denotes the two levels of communication apprehension. B* denotes the two levels of stimulus statements.




Table 4
Analysis of Variance for Atomization--Judge #4

Source df Type IV SS F Value P > F
A* 1 46.75307728 4.38 0.0508
B** 1 51.47279607 4.82 0.0414
A*B 1 25.03613128 2.35 0.1430
A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.
Table 5
Analysis of Variance for Fidelity--Judge #4
Source df Type IV SS F Value P > F
A* 1 50.20789572 3.19 0.0910
B* 1 68.35966685 4.34 0.0517
A*B 1 59.13452514 3.76 0.0685
A* denotes the two levels of communication apprehension. Bf* denotes the two levels of stimulus statement.




Hypothesis V--High communication apprehensives will
exhibit significantly greater tension in their feedback responses than will low communication apprehensives. The scores from Judge #4 showed an interaction effect between communication apprehension and the stimulus statement (F = 4.65, p < .04, see Table 6). The scores from Judge #5 did not indicate any significant differences between the groups.
Hypothesis VI--High communication apprehensives will exhibit significantly more ideation in their feedback responses than will low communication apprehensives. The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between groups.
Hypothesis VII--High communication apprehensives will exhibit significantly more flexibility in their feedback responses than will low communication apprehensives. The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between the groups.
Hypothesis VIII--High communication apprehensives will exhibit significantly less digression in their feedback responses than will low communication apprehensives. The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between the groups.




Table 6
Analysis of Variance for Tension--Judge #4
Source df Type IV SS F Value P > F
A* 1 151.78817505 11.91 0.0029
B** 1 35.99794820 2.82 0.1102
A*B 1 59.33379888 4.65 0.0447
A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.




Hypothesis IX---High communication apprehensives will exhibit significantly less involvement in their feedback responses than will low communication apprehensives. The scores recorded by Judge #4 did indicate a significant interaction effect between communication apprehension and the stimulus statement (F = 6.63, p <.01, see Table 7). The scores from Judge #5 did not indicate significant differences between the groups.
Discussion
whenever researchers employ judge's ratings of human
behavior in experimental investigations, reliability checks must be made to determine if the judges were rating the same behaviors in the same ways. Such reliability measures were conducted in this experiment and, due to the results of these reliability checks, the validity of the FRI has become an important initial focus of discussion in this analysis.
In light of the low reliability ratings of the
original three judges (all first year Ph.D. students), two expert judges were recruited to analyze the groups' behaviors. one of these two judges (who was highly recommended by Leathers, the author of this scale) has




Table 7
Analysis of Variance for Involvement--Judge #4

Type IV SS
7.21613128 19.87343076 46.34089851

F Value
1.03
2.84
6.63

P> F 0.3230
0.1090 0.0191

Source A* A*
A*B

A* denotes the two levels of communication apprehension. B **denotes the two levels of stimulus statements.




employed the scale in his own research and is an authority in the area of small group research and process evaluations. The second of the two judges is a leading authority on rating error, scale development and implementation.
The reliability ratings of the two expert judges
were strikingly similar to the reliability coefficients earlier discussed. Employing the same two evaluation techniques, and combining all five judges in the same analysis, the reliability coefficients for the two expert judges were dramatically low (see Appendix C). And, as was the case with the initial three judges, several correlations for the two expert judges were quite low with one correlation being negative. Even more surprising was the fact that no two judges demonstrated consistent mutual reliability among all nine scale items, which indicates that no two of the five judges consistently rated the same feedback responses in similar ways. These results demand direct and specific analysis of the FRI as a valid measuring tool in conjunction with rater reliability and the methodological problems in the use of the FRI.
According to the data analysis of this investigation, several questions arise as to the sensitivity of the FRI to distinguish differences in verbal messages of




individuals engaged in small group discussions. At the crux of this problem are the operational definitions for the scale items. Many, if not most of these definitions are rather ambiguous and vague. For example, scale item #1-Deliberateness is defined as "5-a7 symbol response represents a deliberate, carefully reasoned, logical response; /7a] signal response represents an immediate, unthinking, largely automatic, visceral response of Y to X." A statement could be very deliberate, carefully reasoned and logical but have nothing to do with the preceding (X) comment. How should this response be rated? Similarly, scale item #9-Involvement is defined as the "degree to which Y seeks to avoid comment on X's contribution by attempting to withdraw from the discussion of X's contribution." It is difficult to understand how judges could reasonably differentiate between whether a person is "seeking to avoid comment on X's contribution" or simply seeking to clarify, promote or continue quality progression toward the group goal.
Beyond definitional problems, the FRI may have
inherent reliability problems in analyzing the impact of a stimulus statement after the first feedback statement. Obviously, certain statements may have impact on the group




process beyond one response statement. As stated by Leathers (1969), the premise of the FRI is "to provide an accurate quantitative reading of the feedback dimensions which characteristically follow a given type of message" (p. 291). This premise of this instrument, therefore, becomes inappropriate in the consideration of analyzing progressive, contiguous feedback responses in relation to the continuing impact of the stimulus statements. If this is the case, then the instrument does not necessarily measure group process; rather, it evaluates statements in isolation, which is not commendable to group process studies. This is not, however, a condemnation of the FRI. Rather, it is an appraisal as to the implicit premise given to the quantitative evaluation of the dimensions of feedback statements following a given stimulus statement.
A causality problem may also exist in the qualitative analysis of progressive responses via the FRI. As responses are more removed in time and space from the stimulus statement, the scale becomes less viable in that judges cannot be sure whether or not the later responses are in fact due to the stimulus statement or other feedback responses. For example, feedback response #3 may demand direct attention from feedback response #4.




Judge's evaluations of these responses then is difficult. Interpretation of these evaluations also becomes difficult in that the feedback responses may not deal with the stimulus statement, but the reasons are unclear.
Training judges must also be discussed in relation
to the low reliability coefficients exhibited. This experimenter followed all the general guidelines previously set forth in the literature in the training of the initial three judges. Through individual conferences and verbal confirmation, all three judges indicated a comprehensive understanding of the FRI and its implementation. Given the low reliability ratings of these three judges in addition to the low reliability ratings of the two recruited expert judges, this experimenter believes that highly specialized training on each scale item by earlier experimenters could explain the previously reported high interobserver reliabilities. The extent to these past trainings is to date not known by this experimenter. However, given the guidelines currently put forth in the literature, consistently high judges' reliability correlations cannot be expected.




Implications
Certain implications can be drawn from this study. First, according to the expert judges evaluations, there appear to be some differences in how communication apprehensives respond to anxiety-arousing statements in a group context. It is, however, unclear exactly what these differences are. The two expert judges evaluations indicated that they perceived differences in some cases, yet they did not agree where these differences were located in relation to the FRI and to what degree these differences were exhibited.
Judge #4 indicated that high communication apprehensive individuals exhibited significantly less relevancy, atomization and fidelity in their feedback responses to anxietyarousing communications than did low communicative apprehensives. Additionally, Judge #4 indicated an interaction between communication apprehension and the stimulus statement in the tension and involvement of feedback responses to anxiety-arousing communications. Judge #4 indicated that feedback responses of high communication apprehensives to facetiousness were significantly more tense than feedback responses of low apprehensives to personal commitments, of low apprehensives to facetiousness, and of high




apprehensives to personal commitments. Lastly, Judge #4 indicated that feedback responses of high apprehensive to facetiousness were significantly less involved than the feedback responses of high apprehensives to personal commitments, of low apprehensives to personal commitments, and of low apprehensives to facetiousness.
On the other hand, the only significant difference
between communication apprehensives feedback responses to anxiety-arousing statements indicated by Judge #5 was that there was an interaction of communication apprehension and the stimulus statement in the relevancy of feedback responses. The feedback responses of high apprehensives to personal commitments were significantly less relevant than the feedback responses of high apprehensives to facetiousness, of low apprehensives to facetiousness, and of low apprehensives to personal commitment.
The only scale item in which both judges indicated
differences in the responses of high and low communication apprehensives was in the relevancy of feedback. Although Judge #4 indicated a main effect for the stimulus statement and Judge #5 indicated an interaction effect, in both cases the feedback of high apprehensives was judged less relevant than the feedback of low apprehensives. This




finding is consistent with previous research which has indicated that high communication apprehensives will exhibit less relevancy in their interactions in small groups than will low communication apprehensives (Well and Lashbrook, 1970). Even though the two expert judges did not agree about the type of effect, both indicated that high apprehensives exhibited less relevancy in their feedback responses than did low apprehensives.
As indicated above, both expert judges recorded
significant differences between high and low apprehensives on the relevancy of feedback responses. However, all other significant differences were demonstrated by only Judge #4. The implication of these interrater differences raises the issue of the FRI as a valid measure of group process. As Emmert and Brooks (1970, p. 222) point out, when a scale is developed the scholar must be careful in whether a procedure is designed to measure something about the object being judged or the individual doing the judging. In the present study, with such low reliability coefficients, it is not clear whether the FRI measured the object itself (feedback responses) or measured differences in the individuals doing the judging.




The development of measuring instruments to assess
differences in how individuals rate communication behaviors is a legitimate and needed research goal in the study of group process. The intent of the FRI is to measure the differences in the quality of feedback responses to certain stimulus statements, not differences in judges themselves. However, in this investigation the FRI manifested more differences between the judges than between the experimental groups. Therefore, investigators should be aware of the fact that the FRI may not exclusively measure differences in feedback responses to certain stimulus statements, but may also measure differences in a judge's ability to perceive differences in the object under evaluation.




65
Note
'Two expert judges were recruited to evaluate the audio tapes of the group discussions. The first judge, Dr. Thomas J. Same, is currently the Director of Cooperative Education at the University of Denver. Dr. Samne has published many articles in small group communication and rating error. He is a leading authority on the development, implementation and interpretation of the FRI. The second judge, Dr. Douglas G. Bock, is currently an Associate Professor of Speech at Utah State University, Logan, Utah. Dr. Bock has a long and highly acclaimed research record in rating error, scale use and evaluation.




CIMPTER VI
DIRECTIONS FOR FUTURE RESEARCH
As is the case with most research, this study has generated more questions than answers. Of primary importance is the continuing indepth investigations and implementation of the FRI as a measure of small group process. Interrater reliability problems were evidenced in this study and only through continuing study will these findings be explained as either internal reliability problems with the FRI or as problems within the unique combination of variables in this study. Similarly, future research should incorporate evaluations of judges from outs ide the area of speech communication to further validate the FRI.
Future research should also investigate the effects of many different types of stimulus statements on the process of group interaction. Few have been investigated to date. Similarly, the partitioning of variables such as age, sex of participant, topic saliency and egoinvolvement could give explanatory power to these effects.




Several methodological issues should be considered in future investigations. Pilot studies should be conducted to assure acceptable interjudge reliability and to assess that the stimulus statements were in fact process disruptive, primarily by post-test surveys of the subjects. Videotaping of all group discussions should be conducted to allow the judge's a full view of the group's verbal and nonverbal behavior, which could potentially increase acceptable interjudge reliability. Consideration should be given to drawing a sufficiently large sample size to allow the simultaneous analysis of three different content categories, thereby allowing comparisons of specific stimulus statement types on the same population. Additionally, scholars should incorporate stochastic models of analysis, such as Markov chain analysis and interaction process analysis, to determine if the content of feedback responses demonstrate consistent patterns through time. Finally, post-experimental interviews with the subjects could shed light on the results found and on the discussion of implication of those results.




APPENDICES




APPENDIX A
FEEDBACK RATING INSTRUMENT




APPENDIX A

FEEDBACK RATING INSTRUMENT

Symbol
3

Deliberateness
1 0 1

Signal
3

Scale #1

Symbol response represents a deliberate, carefully reasoned, logical response; signal response represents an immediate, unthinking, largely automatic, visceral response of Y to X.

Relevant
3

Relevancy

Irrelevant

Scale #2

Relevancy-extent to which Y seeks to establish the connection between X's comment and the comment wfkhch immediately preceded X's comment.

Scale #3

Unif ied
3 2
1 1___

Atomization

Atomized
3

Degree to which Y's contribution involves incomplete, fragmented or disjoined thought; includes running a number of ideas together; a number of people talking at the same time.




Fidelity
0

Confused
3

Scale #4

Extent to which Y's response to X exhibits confusion as to the meaning and/or intent of X's original message; characterized by the necessity of Y seeking clarification, definition, expansion, etc. from X.

Relaxed
3 2

Tension

Tense
3

Scale #5

Degree to which non-verbal gestures like laughter, signs, groans, etc., indicate Y's relative state of tension or relaxation.

Ideational
3 2

Ideation

Personal
3

Scale #6

Ideational responses involve an appraisal or evaluation of X's ideas; personal responses represent the degree to which Y's comments involve direct or implied criticism of X, as a person.

Flexible

Flexibility
2 1 0

Inf lexible
3

Scale #7

Inflexible response indicates Y's unwillingness to modify his position in response to X's contribution; may include a counter assertion.

Clear




Degree to which Y inhibits X's immediate response, primarily by means of lengthy and discursive utterances.

Withdrawn
3 1

Concise
3

Digression Digressive
2 1 0 1 2 3

I I

I I I

Scale :#8

Involvement
0
I I I

Scale #9

Degree to which Y seeks to avoid comment on X's contribution by attempting to withdraw from the discussion of X's contribution.

Involved
3 1




APPENDIX B
McCROSKEY'S PRCA-COLLEGE




APPENDIX B

McCROSKEY' S PRCA-COLLEGE
DIRECTIONS: This instrument is composed of 25 statements concerning your communication with other people. Please indicate the degree to which each statement applies to you by marking whether you (1) Strongly Agree, (2) Agree,
(3) Are Undecided, (4) Disagree, or (5) Strongly Disagree with each statement. There are no right or wrong answers. Work quickly, just record your first impression by circling the correct number.

1. While participating in a conversation
with a new acquaintance I feel very
nervous.
2. I have no fear of facing an audience.
3. I talk less because I'm shy.
4. I look forward to expressing my opinions
at meetings.
5. I am afraid to express myself in a
group.
6. I look forward to an opportunity
to speak in public.
7. I find the prospect of speaking
mildly pleasant.
8. When communicating, my posture feels
strained and unnatural.
9. I am tense and nervous while participating in group discussions.
10. Although I talk fluently with friends,
I am at a loss for words on the
platform.

1 2 345 12 3 45 12 34 5
12 34 5 12 34 5 12 34 5 1 2 345 1 2 345
1 2 345 12 34 5




11. 1 have no fear about expressing
myself in a group.
12. My hands tremble when I try to
handle objects on the platform.
13. 1 always avoid speaking in public if
possible.
14. 1 feel that I am more fluent when
talking to people than most people
are.
15. 1 am fearful and tense all the while
I am speaking before a group of
people.
16. My thoughts become confused and
jumbled when I speak before an
audience.
17. 1 like to get involved in group
discussion.
18. Although I am nervous just before
getting up, I soon forget my fears
and enjoy the experience.
19. Conversing with people who hold
positions of authority causes me
to be fearful and tense.
20. 1 dislike to use my body and voice
expressively.
21. 1 feel relaxed and comfortable while
speaking.
22. 1 feel self-conscious when I am
called upon to answer a question or
give an opinion in class.
23. 1 face the prospect of making a
speech with complete confidence.

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5




24. I'm afraid to speak up in conversation.
25. 1 would enjoy presenting a speech
on a local television show.

1 2 3 4 5 1 2 3 4 5




APPENDIX C
PRETEST OF ANXIETY AROUSING MESSAGES




Please read each of the following statements and record the appropriate number on the scale given as to your perception of HOW ANXIETY PRODUCING EACH STATEMENT WOULD BE IF GIVEN IN A SMALL GROUP DISCUSSION.
(non-anxiety producing) (anxiety-producing)
Relaxed Tense
1 2 3 4 5 6 7
Degree to which this statement would be anxiety-producing if given in a small group discussion.
Rating
1. Susan reminds me of an old story about Ronald Reagan--up drove an empty car and out stepped
Ronald Reagan.
2. John sounds so old fashion that he probably thinks that Hoover is still President. __3. Susan reminds me of what a friend once told me: You may be dumb but you sure are ugly. __4. Susan just needs to get her folds stoned-then they will agree to Anything. __5. Poor John, sounds like the Gator football team--3rd and long, so fall back and punt!__6. Do you suppose John wears white socks? __*7. John probably uses pot as an aphrodisiac.
I heard he seduced three guys on the way
to school last week. __*8. I hope Susan can make love better than she
treats John.
9. I haven't had so much fun in a group since the pigs ate my baby brother.




Rating
10. I guess John now has a new interpretation of the song "Another One Bites the Dust."
ii. Don't you think that this is a matter of modern social morality?
12. Don't you feel that this is more a matter of E.R.A. tenets than of personal commitment?
13. I feel the question is more concerned with religious adjudication than cohabitation,
don't you?
14. We haven't addressed the question of their interpersonal heterophily; it should be
addressed.
15. They seem so diametriacally opposed--two extremely bi-polar views of this circumstance.
*16. Isn't this really a problem of irrelevant
personal catharsis?
*17. It seems that John's stance is the result
of a cognitive occlusion.
18. Susan is really using John as a personal therapeutic tool.
19. Realistically, this is a situation of high versus low cognitive complexity.
20. Wouldn't you say that Freud would call this a problem of libido?
21. John must be totally egocentric--He thinks only of himself.
*22. Susan is totally two-faced--I don't see how
anyone could disagree with that.




Rating
23. 1 guarantee you that if the situation was reversed that John would do the same
thing.
*24. John's so completely uncaring--look at
how he is treating Susan. __25. 1 can't understand why you feel that way.
Obviously they are both uncaring about
each other's feelings. __26. John should buy Susan an engagement ring.
At least they would then have something
to fall back on. __27. I don't like John's attitude of "nothing
to hide." If he has any feelings at all,
he'd reconsider.
28. How does Susan know her parents are against
this? I strongly feel that they should just cool it and tell Susan's folks the
truth. __29. There is only one way out of this mess
and that is to face up to Susan's parents. __30. Both John and Susan are completely selfish.
There is no way around that.___31. 1 agree with that. That sounds good to me. __32. 1 really do not understand what you mean.
Could you rephrase that? __33. I am glad you feel that way. So do I. __34. 1 think I will take some notes on this
stuff, ok? __35. 1 believe we are on the right track here. __36. 1 don't know about that. What do the rest
of you think?




Rating
37. 1 like the way you said that. That is
certainly my position on the topic.
38. 1 think the involvement of all of us in
this project is commendable.
39. It sure seems that we are a hardworking
group.
40. According to my past experience, I'd say
that we are pretty much accurate in our
analysis.
41. 1 find it really interesting that you
mentioned such a proposition.
42. Then do we agree that the crux of the
matter is our lack of knowledge on this
subject?
43. 1 have found that there are a lot of people
really concerned with this.
44. Let me add one more thing to help your
answer.
45. 1 think what you have just said makes a
lot of sense.
46. It seems as if you have just stated our
central theme to this problem.
47. Yes, I do believe that our major problem
is definitional.
48. Do we have any volunteers to look into that
proposal?
49. Will you agree with me that this seems to
be the essence of the problem we've been
discussing?
50. That's correct. I concur.
* indicates those statements used in the study.




APPENDIX D
SAMPLES OF GROUP INTERACTIONS




Sample #1--High Communication Apprehensives
Stimulus Statement: John probably uses pot as an aphrodisiac. I heard he seduced three
guys on the way to school last week.

Feedback Feedback
Feedback

Feedback
Feedback

Statement #1: Gee, anything else you can think
about? I think we've covered all the ground.
Statement #2: What do you think. Does John
really care for Susan?
Statement #3: Well, he's not helping her out
any. He's putting his foot down this time
and I don't think he cares that much.
Statement #4: I think that he should.. .uh, at
first, uh...I don't know what I think.
Statement #5: Uh, I think that if, uh, John
really cares for Susan, uh, he should really plan ahead at first, and, at first he should plan. Of course, uh, they are still Juniors.

Sample#2--Low Communication Apprehensives
Stimulus Statement: I hope Susan can make love
better than she treats John.
Feedback Statement #1: Ok, well, how would you as a parent
feel about this situation? Your daughter .... the
males over there. How would you feel, cause
it's usually the male that, uh, is stereotyped
as getting the most upset and uptight. The mother calming them down, you know, so how
would you guys feel if it were your daughter
and you found out that she had been living
with somebody for six months and hadn't told
you about it for six months.
Feedback Statement #2: I would be upset that she didn't
tell me, but I.. .personally I don't see anything
wrong with it, instead of getting married and
not knowing the person.




Feedback Statement #3: Yeah, but don't forget this is
your own daughter, I mean...
Feedback Statement #4: Yeah, that's what I'm saying, and
I, you know, I thought about that. uhm, I would be upset that she didn't have enough
confidence to come and tell me what's going
on, but ... you know. If she broke it to me
in the right way, you know, she told me, "look,
I've known this guy for three years. I've
been living with him for the past three months,
six months." You know, then, you know, then
some type of feelings have emerged. It's just
not a one night stand type thing.
Feedback Statement #5: How about you. Come on, the other
male.




APPENDIX E
CORRELATION COEFFICIENTS FOR
INTERRATER RELIABILITIES




Judge #1- Judge #1- Judge #2- Judge #4Judge #2 Judge #3 Judge #3 Judge #5
Scale #1 .243 .547 .287 .346
Deliberateness
Scale #2 .044 .404 -.014 -.182
Relevancy
Scale #3 .127 .172 .392 .003
Atomization
Scale #4 -.014 -.25 .143 .272
Fidelity
Scale #5 -.096 .279 .352 .603
Tension
Scale #6 .046 -.121 .303 .049
Ideation
Scale #7 .302 .148 .079 .116
Flexibility
Scale #8 .142 -.113 -.061 .411
Digression
Scale #9 -.107 -.106 .368 .414
Involvement




REFERENCES

Bales, R. F. Interaction Process Analysis: A Method for
the Study of Small Groups. Cambridge, Mass.: AddisonWesley, 1950.
Behnke, R. R., & Carlile, L. W. Heart Rate as an Index of
Speech Anxiety. Speech Monographs, 1971, 38, 65-69.
Burgoon, J. K. The Unwillingness-to-Communicate Scale:
Development and Validation. Communication Monographs,
1976, 43, 60-69.
Burgoon, J. K., & Burgoon, M. Unwillingness to Communicate,
Anomia-Alienation, and Communication Apprehension as Predictors of Small Group Communication. Journal of
Psychology, 1974, 88, 31-38.
Carlile, L. W., & Behnke, R. R. A Report to the Criticism
of "Heart Rate as an Index of Speech Anxiety."
Speech Monographs, 1973, 40, 156-159.
Clark, A. J. Identifying the Anxious Communicator. Florida
Speech Communication Journal, 1973, 1, 17-22.
Clevenger, T. C. A Definition of Stage Fright. Central
States Speech Journal, 1955, 6, 26-30.
Clevenger, T. C. A Synthesis of Experimental Research in
Stage Fright. Quarterly Journal of Speech, 1959,
45, 135-145.
Crowell, L., & Scheidel, T. M. Categories for Analysis of
Idea Development in Discussion Groups. Journal of
Social Psychology, 1961, 54, 155-168.
Daly, J. A. & McCroskey, J. C. Occupational Choice and
Desirability as a Function of Communication Apprehension. Journal of Counseling Psychology, 1975,
22, 309-313.




Dicken, M., Gibson, F., & Prall, C. An Experimental
Study of the Overt Manifestations of Stage Fright.
Speech Monographs, 1950, 17, 37-47.
Dicken, M., & Parker, W. R. An Experimental Study of
Certain Physiological, Introspective and Rating Scale
Techniques for the Measurement of Stage Fright.
Speech Monographs, 1951, 18, 251-259.
Eckert, R. G., & Key, N. Public Speaking as a Cue to
Personality Adjustments. Journal of Applied Psychology, 1940, 24, 144-153.
Emmert, P., & Brooks, W. D. Methods of Research in Communication. Boston, Mass.: Houghton Mifflin Company, 1970.
Giffin, K., & Gilham, S. M. Relationships Between Speech
Anxiety and Motivation. Speech Monographs, 1971,
38, 70-73.
Gilkenson, H. Social Fears as Reported by Students in
College Speech Classes. Speech Monographs, 1942,
9, 141-160.
Greenleaf, F. An Experimental Study of Social Speech
Fright. (Master's Thesis, State University of Iowa,
1947).
Kline, J. A. Indices of Opinionated and Orientating
Statements in Problem-Solving Discussions. Speech
Monographs, 1970, 37, 282-286.
Kline, J. A. Orientation and Group Consensus. Central
States Speech Journal, 1972, 23, 44-47.
Lashbrook, W. B., Lashbrook, V. J., Bacon, C., & Salinger, S.
An Empirical Examination of the Relationship Between
Communication Apprehension and Tolerance of Ambiguity.
Southern Speech Communication Journal, 1979, 44,
244-251.
Leathers, D. G. Process Disruption and Measurement in
Small Group Communication. Quarterly Journal of
Speech, 1969, 40, 287-300.




Leathers, D. G. The Process Effects of Trust-Destroying
Behavior in the Small Group. Speech Monographs,
1970, 37, 180-187.
Leathers, D. G. Testing for Determinent Interactions in
the Small Group Communication Process. Speech
Monographs, 1971, 38, 182-189.
Leathers, D. G. Quality of Group Communication as a
Determinant of Group Product. Speech Monographs,
1972, 39, 166-173.
Lustig, M. W. Verbal Reticence: A Reconceptualization
and Preliminary Scale Development. Paper presented at the meeting of the Speech Communication Association, Chicago, 1974.
Lustig, M. W., & Grove, T. G. Interaction Analysis of
Small Problem-Solving Groups Containing Reticent and Non-Reticent Members. Western Speech, 1975,
29, 155-164.
Marr, T. J. Counciliation and Verbal Responses as Functions
of Orientation and Threat in Group Interaction.
Speech Monographs, 1974, 41, 6-18.
McCroskey, J. C. Measures of Communication-Bound Anxiety.
Speech Monographs, 1970, 37, 269-277.
McCroskey, J. C. The Problems of Communication Apprehension
in the Classroom. Paper presented at the meeting of
the Communication Association of the Pacific, Kobe,
Japan, 1976.
McCroskey, J. C. Oral Communication Apprehension: A
Summary of Recent Theory and Research. Human Communication Research, 1977, 4, 78-96.
McCroskey, J. C. Validity of the PRCA as an Index of Oral
Communication Apprehension. Communication Monographs,
1978, 45, 192-203.




Full Text

PAGE 1

THE EFFECTS OF ANXIETY-AROUSING MESSAGES IN GROUPS ON QUALITATIVE DIMENSIONS OF FEEDBACK IN HIGH AND LOW COMMUNICATION APPREHENSIVE INDIVIDUALS By JAMES DAVID HARRIS A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1981

PAGE 2

1 A CKNOWLED GEMENT S There are many individuals I wish to thank for their help, advice, and support in the development of this dissertation. Very special thanks is extended to my chairman, Tony Clark. Your vigorous and sincere concern for quality in this text is indicative of your true professionalism. Thank you for the numerous critical revisions and advice from start to finish of this dissertation, without which, this project would still be in its inception period. To Tom Abbott, who will always have my most sincere admiration and respect, thank you for your attention and counsel in this text as well as in my academic progression. I will always remember the stories, problems, laughs, disappointments and the triumphs (even if few) that we shared in Room 335. To Paul Jensen, thank you for revaling to me what true professional pride is all about. Your recommendations and support have been a constant inspiration not only in this manuscript but also in my personal life. 11

PAGE 3

To David Robinson, thank you for the innumerable hours of counsel and guidance in the analysis of these data. I truly admire and have greatly benefited from your exceptional proficiency in design, interpretation and instruction of statistical theory and practice. I will always be indebted to you for your sincere concern for me as a past and present student. To Marvin Shaw, special thanks are extended to you, sir, for your insightful contributions and your patient indulgence in sitting on yet another committee. You have demonstrated to me by example what it is to be a gentleman and a scholar. I will always be proud to proclaim that I had the honor of having you serve on my committee. To Don Williams, the term "professor" has never been more aptly applied. You have instilled in me, by example, a genuine concern for conscientious scholarship and professional integrity. And above all, you have always given me your personal respect and admiration. You will always have mine. Thanks are extended to Ted Chandler, Katie Poschall and Trev Garner for your considerable indulgence in the time consuming task of recording 990 individual ratings. 111

PAGE 4

Very special thanks are extended to Tom Saine and Doug Bock for lending your expertise to the analysis of these data. Your professional expediency in these analyses will always be remembered. This dissertation is dedicated to my parents. Can you believe I finally made it? It's been a long and hard road for all of us. Yet through your endless love and devotion you have taught me pride for personal and academic excellence. I hope that you are as proud of this dissertation as I am proud of being your son. This dissertation is also dedicated to Melanie Ferriera, the soon to be Mrs. Harris. You've stuck by me when times were bad, pushed me and praised me at all the right moments, and you never let me give up. And most importantly, you have given me your infinite strength and limitless love, without which I would have never finished. If you can live with me through this, you can live with me through anything. Thank you Mel. I love you very much. A special expression of thanks is extended to my unparalleled friend, Jim Hicks. Thank you for the untold personal favors and the pleasureable experiences we have shared. Long live the Derby, and long live our friendship. IV

PAGE 5

Special appreciation and thanks are extended to Mrs. Voncile Sanders for your immense patience, understanding and quality workmanship in the preparation of the typed text. And finally, to Don Shores, who has shown me what a friend and Christian is; to Laurie Weiman, who is the only woman in the world to beat out Farrah Faucett as my "pin-up" girl and to the entire cast and crew of the third floow, thanks for everything. v

PAGE 6

TABLE OF CONTENTS Page 11 IX ACKNOWLEDGEMENTS LIST OF TABLES ABSTRACT CHAPTER I INTRODUCTION 1 II REVIEW OF LITERATURE 4 The Effects of Process Disrupting Messages on Group Communication 5 History of Process Disrupting Messages on Group Communication. 5 Methodologies and Research Findings 7 Limitations 12 The Effects of Communication Apprehension on Group Communication. 15 Defining Communication Apprehension 17 Measurement of Communication Apprehension 19 Observer rating scales .... 19 Physiological techniques ... 21 Introspective measures .... 22 Behavioral correlates of communication apprehension. 23 vi

PAGE 7

TABLE OF CONTENTS (Continued) Page Summary of Research Findings. 25 Summary 28 III RATIONALE AND HYPOTHESES 29 IV METHODOLOGY 32 Subjects 32 Design 32 Experimental Procedure 33 The "Confederate" 35 Independent Variable: Process Disruption Statements 37 Independent Variable: Communication Apprehension 39 Dependent Measure: Quality of Group Feedback 45 Summary 46 V RESULTS AND DISCUSSION 48 Results 48 Interrater Reliability 48 Test of Hypotheses 49 Discussion 55 Implications 61 Note 65 VI DIRECTIONS FOR FUTURE RESEARCH 66 APPENDICES APPENDIX A — FEEDBACK RATING INSTRUMENT. ... 70 APPENDIX B— McCROSKEY S PRCA -COLLEGE 74 APPENDIX C— PRETEST OF ANXIETY AROUSING MESSAGES 78 APPENDIX D — SAMPLES OF GROUP INTERACTIONS 83 vii

PAGE 8

TABLE OF CONTENTS (Continued) Page APPENDIX E — CORRELATION COEFFICIENTS FOR INTERRATER RELIABILITIES .... 86 REFERENCES 87 BIOGRAPHICAL SKETCH 92 vm

PAGE 9

LIST OF TABLES Table Page 1 A Taxonomy of Process Disrupting Message Forms 16 2 Analysis of Variance for Relevancy — Judge #4 51 3 Analysis of Variance for Relevancy — Judge #5 51 4 Analysis of Variance for Atomization — Judge #4 52 5 Analysis of Variance for Fidelity — Judge #4 52 6 Analysis of Variance for Tension — Judge #4 54 7 Analysis of Variance for Involvement — Judge #4 56 IX

PAGE 10

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EFFECTS OF ANXIETY-AROUSING MESSAGES IN GROUPS ON QUALITATIVE DIMENSIONS OF FEEDBACK IN HIGH AND LOW COMMUNICATION APPREHENSIVE INDIVIDUALS By JAMES DAVID HARRIS August 1981 Chairman: Anthony J. Clark Major Department: Speech The purpose of this study was to explore the effects of anxiety arousing messages on the quality of group feedback when presented in groups composed of high and low communication apprehensives The experimental design was a 2 x 2 analysis of variance determined by the level of communication apprehension (high-low) and type of stimulus statement (unequivocal personal commitment-facetiousness) Trained confederates injected the predesignated stimulus statements into three-person groups, one of the three individuals being the confederate. Leathers' Feedback Rating Instrument (1969) was employed in the analysis of the first five feedback responses following the injection of the stimulus statement.

PAGE 11

Three trained judges evaluated the audio-tape recordings of the group discussions. Overall reliability coefficients between the three judges on each scale item across all groups were found to be quite low. The experimenter then recruited two expert judges to rate the groups verbal behaviors. As was the case with the initial three judges, the reliability coefficients between the two expert judges were quite low. The test of hypotheses consisted of the separate evaluations of each of the two expert judges for each scale item. Results indicated that high communication apprehensives exhibited less relevancy, atomization, fidelity and involvement as well as greater tension in their feedback responses to anxiety around messages than did low communication apprehensives. Since the inter judge reliability coefficients indicated that all the judges were not rating the same verbal behaviors in similar ways, discussion of these results focused on the validity of the Feedback Rating Instrument as a measure of small group communication process. XI

PAGE 12

CHAPTER I INTRODUCTION Scholars who are interested in small group dynamics are constantly asking the question, "Is the system working?" In most cases, this question is approached by analyzing the quality of the group's output. Investigations have determined that the quality of the group product is directly related to the quality of the group's communication behaviors (Leathers, 1972) Therefore, communication scholars have focused on how continuing communication interchanges between group members affect the process by which groups reach decisions The growth of investigations into the process of group communication has prompted scholars to develop many varied styles of analysis, ranging from interaction analysis (Bales, 1950; Crowell and Scheidel, 1961; Saine, Shulman and Emerson, 1974; Stech, 1970) to orientation behaviors (Kline, 1970, 1972; Marr, 1974). These forms of analysis either record the frequency of certain types of communications, such as information giving communication, information seeking communications, procedural communications, or they analyze the

PAGE 13

effects of facilitating communications injected into a group. Seldom however, have researchers considered qualitative dimensions of group communication; in particular, research is sparce concerning the qualitative aspects of how groups manage message forms that create tension, impede progress, or generally disrupt the process of the group. Within this area of study, research findings indicate that certain process disrupting messages create high levels of anxiety in the receiver, as evidenced by tension, word omissions, word repetitions, and confusion in feedback responses as well as more irrelevant and digressive messages. However, no research to date has concentrated specifically on how anxiety-arousing communications affect group process. Further, researchers have not considered how anxiety-arousing communications affect individuals who reflect high levels of anxiety in their interpersonal communications. The purpose of this study is to investigate the effects of anxiety-arousing message forms on the qualitative dimensions of group feedback in high and low communication apprehensive individuals. When presented with an anxiety-arousing message, will individuals considered high in communication

PAGE 14

apprehension respond differently than those with low communication apprehension? What is the duration of the impact of these anxiety-arousing communications on the group process? And finally, what are the preferred verbal strategies by which group members cope with deviant, unexpected or disruptive communications?

PAGE 15

CHAPTER II REVIEW OF LITERATURE The effect of process disrupting messages on small group communication is a relatively new area of investigation in speech communication research, and has grown out of the acceptance within the field of speech that communication itself is a process; that communication researchers should focus primarily on the process rather than the antecedents of group dynamics to more fully understand the impact of human communication in group interaction (Sheidel and Crowell, 1966) Similarly, communication apprehension research is also a new research tradition within speech communication. Investigations of communication apprehension have ranged from identifying the apprehensive (Phillips, 1968; Clark, 1973) to the analysis of communication apprehensive s behaviors in group discussions (Wells and Lashbrook, 1970) Yet, specific aspects of these varied areas of study mandate special considerations. The purpose of this chapter is to review the relevant literature in process disrupting message forms in group communication, as well as the definitional problems, 4

PAGE 16

measurement and the behavioral correlates of communication apprehension. The Effects of Process Disrupting Messages on Group Communication This section will include a history of process disrupting messages in group communication, methodologies employed and research findings, and the limitations of investigations currently put forth. History of Process Disrupting Messages on Group Communication The study of process disrupting message forms on group communication grew out of the recognition that most small group investigations have been primarily product (output) oriented; that is, small group researchers have traditionally concerned themselves with such areas as task efficiency, group consensus, and solution quality. Communication scholars have recognized such a product orientation "gives no precise indication of the number or nature of factors that shape communication within the small group" (Leathers, 1969, p. 287) Communication researchers have therefore increasingly concerned themselves with a process orientation, focusing on those factors that through time impinge upon the communication patterns, flows and exchanges within the small

PAGE 17

group which then aid in determining or shaping a group's product. As Leathers has stated, "at best, the productoriented experimenter starts with an effect, the group solution, and tries to infer the causes of that effect" (1969, p. 287) One of the more outstanding research traditions of group communication process is that of the effects of process disrupting statements on group interaction. Leathers (1969) conducted the initial investigation in this area studying the effects of high and low-level abstractions, unequivocal personal commitments, implicit inferences, and facetious interpolations on group process. This study has stimulated a limited yet varied amount of research into process disrupting effects. Although no specific (operational) definition of process disrupting message forms has been accepted to date, the implications of this message type appear rather straightforward. A process disrupting message is basically any type of message which can potentially impede, impair, disorganize, confuse or in any manner contribute to disturbing the normal process of group discussion.

PAGE 18

Methodologies and Research Findings The primary focus of this research has been the analysis of the feedback response subsequent to a process disrupting message. Leathers (1969) developed the Feedback Rating Instrument (FRI) for such analysis. The Leathers FRI began as an investigation "to help determine whether any part of the small group process was affected by different stimulus statements" (p. 289). Transcripts from thirty discussion groups were content analyzed to determine: (1) whether responses to various kinds of stimulus statements possessed any common set of characteristics; (2) whether the strength of a given characteristic varied from response to response; and (3) whether any part of the small group communication process was measurably affected by different types of stimulus statements. (p. 289) The analysis did indicate that "the responses to the stimulus statements did have common characteristics which varied in strength and that feedback was more immediately affected by the nature of the stimulus statements than by any other part of the communication process (p. 289) The feedback response indicated by Leathers is the "Y" response of Scheidel and Crowell's (1966) "X-Y-Z" definition of feedback sequences in which "participant (X) initiates

PAGE 19

8 a comment which is followed by a comment from any other participant (Y) which in turn is followed immediately by a further comment from the first participant (X) (p. 274) (The nine-factor Leathers FRI is found in Appendix A.) In applying the FRI to hundreds of feedback responses, Leathers (1969) found that the interobserver reliability estimates for each of the nine FRI scale items to be: (1) deliberateness — .89; (2) relevancy — .89; (3) atomization — .82; (4) fidelity — .80; (5) tension — .62; (6) ideation — .85; (7) flexibility — .86; (8) digression — .88; and (9) involvement — .86. (p. 291) Similar reliability estimates have been obtained in other investigations (Leathers, 1970, Leathers, 1971). As previously mentioned, Leathers (1969) conducted the initial speech communication investigation of process disrupting messages in group communication. Five types of stimulus statements were investigated: high-level abstractions, low-level abstractions, unequivocal personal commitment, implicit inferences and facetious interpolations. Five expert judges unanimously agreed with two "lines" for each category, in that these statements "met their respective definitions" (p. 288). Two trained confederates systematically introduced these stimulus statements at 6-minute intervals in the discussion groups, composed of four naive subjects

PAGE 20

and the two confederates. Two trained judges applied the Leathers FRI to the feedback responses following each stimulus statement. The most important results, as reported by Leathers, were that feedback responses to high-level abstractions were confused, tense and withdrawn; feedback responses to implicit inferences were rated very unthinking, irrelevant and digressive; and feedback to facetious interpolations were especially personal and inflexible. Leathers (1970) also investigated the effects of trust-destroying communications on small group behavior. In the first of a two-part experiment, twenty naive subjects participated in a game with an experimenter confederate, where through cooperation and verbal reinforcement, the confederate established a high level of mutual trust with the subject, as verified in post-experiment surveys. In the second part of the experiment, the confederate and the subject were placed in a group comprised of four other students, comprising a six-person group. For the first thirty minutes of the group discussion, the confederate reinforced the trust earlier developed between him and the subject through statements which were very supportive and agreeing. However, in the last thirty minutes, the confederate injected stimulus statements at approximate

PAGE 21

10 six-minute intervals that were designed to be highly trustdestroying. Two judges evaluated the feedback responses of the subjects to the trust-destroying messages. The results indicated that responses to trust-destroying messages "became particularly tense, inflexible, and personal" (p. 186) In a subsequent investigation of trust-destroying messages in group discussion, Prentice (1975) analyzed the degree to which nonfluencies were present in (1) the naive subject who was the target of the trust-destroying messages and (2) other group members who witnessed the trust-destroying behavior. Written transcripts were made from fifteen of the original twenty groups used in a Leathers (1970) study. Two judges verified the accuracy and completeness of the transcripts as compared with the tape-recordings of the discussions. A seven-category system was used in the analysis, consisting of (1) word repetitions, (2) stuttering, (3) sentence changes/incompletions (4) omission, (5) intruding incoherent sounds, (6) "you know(s)," and (7) "I mean(s)." Additionally, a nonfluency ratio of the number of nonfluencies by the number of words spoken was conducted.

PAGE 22

11 Results indicated that the nonfluency ratio increased during the trust-destroying segments of the discussions. The initial responses of the subjects were found to have more word repetitions and more incoherent sounds following the trust-destroying messages than at other times in the discussion. Also, when the subject's verbal contributions were not a direct response to the stimulus statement, the feedback had more sentence changes, more word omissions and more incoherent sounds. Finally, the verbal contributions of those group members witnessing the trustdestroying message had more word repetitions and incoherent sounds than at other times in the discussion. Building upon previous findings. Leathers (1971) investigated the effects of high, medium and low levels of abstraction, implicitness and facetiousness on feedback in group communication. The stimulus statements were developed in the following manner. Leathers devised ninety "potential contributions" for each of the three statement categories and varied their degree of abstraction, implicitness and facetiousness, respectively. Each statement was randomly assigned and read into a tape recorder which was then presented to six different college public speaking classes where students rated each statement as to its degree

PAGE 23

12 of abstraction, implicitness and facetiousness Two confederates memorized the statements and randomly introduced these statements at approximate 6-minute intervals into six-person groups (including the confederates) Two trained judges applied the Leathers FRI to the feedback responses following each stimulus statement. The results indicated that as the abstraction level went from low to medium to high, that the feedback became more confused, tense and withdrawn. Also, as the level of facetiousness rose from low to medium to high, feedback became more personal and inflexible. Limitations The gaps in what is known about process disrupting messages as a result of this research tradition are perhaps more glaring than any substantive contributions to knowledge in this area. For example, no data have been presented concerning similarities and differences in how these various messages influence group communication. Since most investigators have employed dependent measures to tap a single dimension of group communication, our understanding of the impact of any specific message on group communication is exceedingly limited. Few attempts have been made

PAGE 24

13 to assess the full range of effects that a highly aversive message could have on subsequent communications with a group. Similarly, we have little information regarding either the duration of the impact of these various messages or the preferred verbal strategies by which group members cope with deviant, unexpected or disruptive communications Several methodological problems are also evident in prior research. First, researchers have instructed the confederates to inject the process disrupting statement at 6-minute intervals without regard to the particular phase in which the group was currently progressing. This could potentially lead to an injection of a process disrupting message which was awkward or inappropriate to the prevalent topic discussion. Second, most investigations have used judges to verify the validity and reliability of the stimulus statements with no further qualifications. Third, few types of process disrupting messages have been investigated. Many other forms of message types might potentially disrupt the communication in the group. Yet those that have been investigated introduce a definite level of anxiety in the receivers, as indicated by feedback

PAGE 25

14 that is tense, confused, etc. Fourth, and most prominent, in all previous studies, typed transcripts of the verbal feedback responses were made from video-tapes of the groups' discussions. On these transcripts, the experimenters presented their analyses of the nonverbal behaviors elicited prior to, during and following the verbal message, such as : Feedback response: (nervous laughter and discussants grin at each other in an embarrassed way.) It seems that what? What does that mean? All right, why don't you explain cognitive aversion or was it occlusion? (More nervous laughter.) Maybe the problem is, ah, maybe it is not an occlusion, maybe it is a problem of cognitive prerogatives. Oh, hell, let's drop the subject. (Leathers, 1971, p. 188) The judges then evaluated the feedback responses given the verbal description in conjunction with the experimenter's evaluation of the n onverbal behaviors elicited. These verbal descriptions of certain specific nonverbal behaviors could be suspect to reliability error in the judges ratings as well as the attempt to describe such a global concept as nonverbal communication in a few, specific words. And finally, the area of process disrupting message effects on small group communication suffers from the same theoretical flaws as other small group research in that

PAGE 26

15 no efforts have been made to systemitize research findings into a conceptual taxonomy of process disrupting message forms and their various effects on group communication. In particular reference to the final limitation cited, Table 1 represents a taxonomy of process disrupting message forms and their effects on group communication. Given the limited range of types of process disrupting statements previously investigated, several other types are included which could potentially be process disrupting. All of these various message forms fit into four conceptually distinct category types. Hence, the first attempt at creating a conceptual taxonomy of process disrupting message forms T he Effects of Communication Apprehension on Group Communication Within the field of speech communication, few areas have grown as rapidly in recent years as that of communication apprehension. Three areas of communication apprehension research appear relevant to the study of group communication: defining communication apprehension, measuring communication apprehension, and the behavioral correlates of communication apprehension.

PAGE 27

16 id Eh 0] £ M fa en rd in co 2 tji C H a M en •H Q a 01 u h i E o c o X rd Eh 0) TS u 1 M o c E H 5 JJ ^ G CO (D M c en O M o G E CD H £ CO 4J • -. CO H rj H 0) rd c E M c •H E m CD T3 M u M CO CD E a E £ tX rd c u >o ^ rd 01 c BO i-i x: fl rd C E X •H H 4J V io fl •H CO c X5 •H -P 13 TS X CD c 4J 01 a 2 c 0) rH CD fa IH M CO co

PAGE 28

17 Defining Communication Apprehension The anxiety of communicating has been referred to by a number of labels, most notably "stage fright," "speech anxiety, "communication apprehension, "unwillingness to communicate," and "reticence." It is important to distinguish these terms, for one of the major difficulties in this area of investigation is the tendency of communication researchers to use the terms interchangeable. Greenleaf (1947) defined stage fright as An evaluative disability, occurring in social speech situations, and characterized by anticipatory negative reactions of fear, avoidance, and various internal and overt manifestations of tension and behavioral maladjustments, (p. 134) Similarly, Clevenger (1955) defined stage fright as Any emotional condition in which emotion overcomes intellect to the extent that communication is hampered, either in audience reception or in speaker selfexpression, where the immediate object is the speech-audience situation, (p. 30) As evidenced in these definitions, stage fright is only concerned with the public speaking format. By the mid-1960 's, investigators recognized that communication anxiety extended beyond the confines of the public speaking format. Exemplary of this trend was

PAGE 29

18 Thorenson's (1966) account of communication anxiety when he described the communicatively anxious individual as one Who feels uncomfortable in social group situations, who does not perceive himself as capable of displaying any social initiative and who describes himself with such adjectives as timid, awkward, and reserved. (p. 207) As the realization grew that an individual's reluctance to communicate might not be exclusively limited to public speaking, new terms were coined and developed to expand and amplify this concept. Giffin and Gilham (1971) defined speech anxiety as A person's unwillingness or reluctance to rely upon himself in a comtiunication situation. His self concept is at stake, and he perceives his speech ability as inadequate to carry him through to the objective, whatever it may be, in that setting, and the speaker is usually painfully aware of his condition. (p. 70) Similarly, and most notable of these concepts, McCroskey (1977) defined communication apprehension "as an individual's level of fear or anxiety associated with either real or anticipated communication with another person or persons" (p. 78). The concepts of unwillingness to communicate and reticence refer to a more global construct of communication

PAGE 30

19 anxiety (McCroskey, 1977) Burgoon (1976) defined unwillingness to communicate as a "chronic tendency to avoid and/or devalue oral communication" (p. 60) Yet Phillips (1968) recognized that many individuals have very high levels of speech anxiety/communication apprehension/ unwillingness to communicate and described these individuals as reticent, where reticence is defined as the "person for whom anxiety about participation in oral communication outweighs his projection of gain from the situation" (p. 40) Measurement of Communication Apprehension A second major area of emphasis in the area of communication apprehension is the measurement techniques employed to assess the communication apprehensive. Three main categories of communication apprehension assessment have been employed: observer rating scales, devices for measuring physiological changes during speaking, and introspective or "self-report" techniques. Observer rating scales Although observer rating techniques have a long history, their use seems to produce inconsistency. Dicken,

PAGE 31

20 Gibson and Prall (1950) found the true reliability of judges in observing and recording communication apprehension in a five category system to be .57. Similarly, Eckert and Key (1940) found that groups of three and four judges produced reliability coefficients of .68 on scales similar to that of Dicken et al. More recently, Mulac and Sherman (1974) developed an instrument for the behavioral assessment of speech anxiety (BASA) Even though in the initial development of the scale the reliability was shown to be quite high (.95), more research utilizing the BASA is needed to assure its widespread reliability and its internal/external validity. As Clevenger (1959) notes, most observer rating techniques (including the BASA) have several drawbacks in that observer rating techniques are not always accurate representations of the communication apprehensive speaker. Those behaviors related to communication apprehension are at best difficult to observe. Additionally, there would be a need to observe individuals across a broad spectrum of communication situations to accurately determine the level of a person's apprehension. Also, in almost every case of observer rating scales, trained observers are necessary to administer and record the behaviors, whereas other techniques do not necessitate such expertise.

PAGE 32

21 Physiological techniques Physiological techniques of measuring communication apprehension have been successfully employed (Porter, 1974) Many techniques have been devised, including measures of communication apprehension as changes in blood pressure and pulse rate before and after public speaking (Dicken and Parker, 1951) in autonomic arousal (Porter, 1974) and most prominently, in heart rate changes (Behnke and Carlile, 1971; Carlile and Behnke, 1973; Porter and Burns, 1973). Even though the reliability of these measures has been quite high, the relative expense of the mechanical devices have many of the same drawbacks as the observer rating techniques. For example, it is improbable if not impossible to obtain physiological indices of communication apprehension during every communication transaction. Accordingly, to obtain a reliable index of communication apprehension via physiological measures, a variety of communication transactions must be measured (McCroskey, 1970) As with observer rating scales, trained experts are needed in administering and recording measures of physiological changes during communication transactions.

PAGE 33

22 Introspective measures Introspective or "self-report" scales have traditionally been the most widely used index of communication apprehension. Although many authors have developed scales for the measurement of communication apprehension via self-reports, six scales have dominated the literature: (1) The Gilkenson Personal Report of Confidence as a Speaker (Gilkenson, 1942) (2) The Watson and Friend Social Avoidance and Distress Scale (Watson and Friend, 1969), (3) McCroskey's Personal Report of Communication Apprehension-College (McCroskey, 1970, 1978), (4) Lustig's Verbal Reticence Instrument (Lustig, 1974), (5) Phillips and Erickson's Reticence Scale (Rosenfeld and Plax, 1976), and (6) Burgoon s Unwillingness to Communicate Scale (Burgoon, 1976) Self-reports have several advantages over observer ratings and physiological measures in that (1) self-report techniques do not require observation and measurement in a variety of communication contexts, (2) expertly trained judges are not required in the administration and recording of the results, (3) self-report techniques are much more cost efficient than observer rating or physiological measures in terms of man-hours per subject and equipment necessary for measurement, and (4) self-report techniques can be

PAGE 34

23 administered to a large number of people simultaneously. Hence, self -report scales are currently the most used measures of communication apprehension. B ehavioral correlates of c ommunication apprehension Of particular interest to communication scholars have been the behavioral correlates of communication apprehension. Communication apprehension has been shown to produce less self-disclosure, lower participation in small groups, less attraction to peers, less likelihood of being perceived as opinion leaders, more tension exhibition in small groups, less likelihood of being perceived as leaders, and more likelihood of choosing occupations which have lower communication demands even though they offer less income and status (McCroskey, 1976) Phillips (1968) found that high apprehensives report "shakiness during classroom activities, felt "butterflies in their stomachs" when asked to speak to the class, found it necessary to end conversations because of their fears, reported an inability to talk to their teachers and supervisors, considered themselves as extremely quiet, felt compelled to apologize for their ideas when they were challenged, and expressed singular inability to talk to their parents. In a similar study, Phillips (1977)

PAGE 35

24 found that the apprehensive did not have the ability to open conversations with strangers, to extend conversations, or to initiate friendships, to follow the group's discussion and make relevant remarks, to answer questions that arise on the job or in class, and a general inability to communicate characterized by a lack or avoidance of participation. Recently, investigators have researched various personality correlates to communication apprehension. McCroskey, Daly and Sorenson (1976) as well as Lashbrook, Lashbrook, Bacon and Salinger (1979) found that communication apprehension was negatively correlated with tolerance for ambiguity. Additionally, McCroskey et al. (1976) found that communication apprehension was negatively correlated to self-control, adventurousness surgence, and emotional security. Lustig (1974) found a negative correlation between communication apprehension and both self-esteem and self-acceptance. Other investigations have obtained similar results (McCroskey, Daly, Richmond and Falcone, 1977) In investigations concerning communication apprehension in small group communication, Wells and Lashbrook (1970) found that high apprehensives interacted less in small groups than did low apprehensives and that their interactions were

PAGE 36

25 less relevant than those who were not apprehensive. High apprehensives have been found to show more tension and less interest in both zero-history and intact groups (McCroskey, 1976) McCroskey (1976) reported that high apprehensives were perceived by other group members as less extroverted, composed, competent socially attractive and task-attractive than low apprehensives. McCroskey and Richmond (1976) report similar findings concerning peer perception and apprehensives. Lustig and Grove (1975) found that groups consisting of low communication apprehensives had more speech acts, interacted more in socio-emotive areas and less in task areas than did groups consisting of both low and high communication apprehensives. In coni sidering unwillingness to communicate, anomia-alienation and communication apprehension as predictors of small group communication, Burgoon and Burgoon (1974) concluded that "persons with a high unwillingness to communicate set will be more tense in their communication behaviors" (p. 36) S ummary of Research Findings In summary, individuals who are considered to be high in communication apprehension generally view themselves

PAGE 37

26 in less positive terms than low communication apprehensives Similarly, high apprehensives are viewed less positively by others than are low apprehensives. High communication apprehensives are generally found to have less quality and quantity of interaction, particularly in small group discussions. Additionally, high communication apprehensives generally tend to avoid the initiation of interaction and quite often attempt to avoid interactions altogether. There are many strengths in the area of communication apprehension research. First, the area of communication apprehension has had a steady and continuous growth of investigations from the early foundations of the field of speech communication. This growth had culminated into a highly intense research tradition. Second, there has been extensive recent inquiry into the instrumentation and methodological measurement of communication apprehension. Even though scholars are not totally satisfied with the methods of recording and evaluating communication apprehension, the amount to research and attention directed toward that question is indeed impressive. Third, communication apprehension focuses on a culturally pervasive problem in America, the fear of speaking. This strength is highlighted by the fact that communication apprehension is

PAGE 38

27 is not a highly specific problem area with little applicability. Many Americans possess anxiety in their communications and therefore research in this area constitutes research on a real and consuming problem in this country. There are several weaknesses in communication apprehension research. First, many terms have been coined in an attempt to define anxiety in interpersonal communication. Confusion can eminate from the misapplication or misinterpretation of conceptual frameworks of these various definitions. Second, there exists a need to extend the scope of investigations in communication apprehension. Most notably, further investigations are needed as to the causes of communication apprehension as well as the continuing need to further validate self-report scales purporting to measure communication apprehension. Third, and most important to this investigation, few studies have considered the effects of communication apprehension on group process. Several studies report that communication apprehensives behave differently in group contexts, but few studies focus on how these behaviors actually affect the < group process.

PAGE 39

28 Summary This chapter has reviewed the relevant research regarding the effects of process disrupting message forms on group communication and the definitional problems, measurement and behavioral correlates of communication apprehension. Limitations of research to date have been highlighted in each area. The overall uniqueness of this review lies in the fact that many process disrupting statements injected in group communication process have been shown to create anxiety in small group discussions. Yet, these anxiety-arousing communications have not been studied in relation to the level of group member's communication apprehension. This study will incorporate both areas of current inquiry in an attempt to further validate process disrupting message effects on small group communication process.

PAGE 40

CHAPTER III RATIONALE AND HYPOTHESES The limitations of previous research in process disrupting message effects on group communication are highlighted by the fact that communication apprehension has not been considered as a variable in the response quality of group communications following a process disrupting message. Process disrupting messages have been shown to introduce a definable level of anxiety into the feedback responses within the group. But, is the cause of this anxiety the process disrupting message or communication apprehension? The purpose of this study is to investigate the effects of anxiety-arousing message forms on the qualitative dimensions of group feedback in high and low communication apprehensive individuals. The theoretical link between research on anxietyarousing communication and the wealth of current inquiry into communication apprehension is, however, tenuous at best. Nevertheless, the potential for expanding our knowledge about both phenomena (group communication and communication apprehension) is substantial for two 29

PAGE 41

30 noteworthy reasons. First, research into communication apprehension provides us with measuring tools for identifying individuals who are thought to have either high or low thresholds for tolerating anxiety-arousing messages and their anxiety-inducing effects. Second, instruments of this sort allow us the opportunity to compare and contrast the effects of anxiety-producing statements upon individuals with extremely different tolerance levels for such anxiety, as well as their tendency toward and preferred modes of anxiety-reduction. Previous research leads to several plausible hypotheses concerning the interrelationships of communication apprehension and the effects of anxiety arousing messages on group communication. Hypothesis I High communication apprehensives will demonstrate significantly less deliberateness in their feedback responses than will low communication apprehension. Hypothesis II High communication apprehensives will demonstrate significantly less relevancy in their feedback responses than will low communication apprehensives. Hypothesis III High communication apprehensives will demonstrate significantly more atomization in their feedback responses than will low communication apprehensives

PAGE 42

31 Hypothesis IV High communication apprehensives will demonstrate significantly less fidelity in their feedback responses than will low communication apprehensives. Hypothesis V High communication apprehensives will demonstrate significantly greater tension in their feedback responses than will low communication apprehensives. Hypothesis VI High communication apprehensives will demonstrate significantly more ideation in their feedback responses than will low communication apprehensives. Hypothesis VII High communication apprehensives will exhibit significantly more flexibility in their feedback than will low communication apprehensives. Hypothesis VIII High communication apprehensives will exhibit significantly less digression in their feedback responses than will low communication apprehensives. Hypothesis IX High communication apprehensives will exhibit significantly less involvement in their feedback responses than will low communication apprehensives.

PAGE 43

CHAPTER IV METHODOLOGY The purpose of this chapter is to summarize the methodological and procedural steps employed in this investigation. Specifically, this chapter will address subject characteristics, research design, experimental procedure and measurement techniques. Subjects One hundred and seventy-seven undergraduates enrolled in sections of the basic public speaking course at the University of Florida volunteered to participate in this study for extra bonus points in the course. Of the total pool of volunteers, forty-four actually served as participating subjects in the experiment. Design The research design for this investigation was a 2x2 factoral with repeated and multiple dependent variables. Subjects who were variously classified as either high or low in communication apprehension (CA) were 32

PAGE 44

33 randomly assigned to one of two conditions of message disruption (MD) Communication Apprehension High Low Message Disruption Type I Type II Experimental Procedure Subjects were informed, prior to volunteering for participation, that the purpose of the study was to "examine the quality of group decisions on problems of mutual interest to both males and females." When subjects arrived for the group sessions they were taken into a small room in which the chairs were arranged in such a manner to assure equal distance between all participants. The Experimenter provided the following instructions: Today you are being asked to participate in a small group discussion concerning the human relations problem just handed to you. At the bottom of that page, you will see a series of questions. These questions were given only as a procedural guide, to allow your group common questions for discussion. Your group may choose to follow the order of the questions, but feel free to discuss this topic in any manner the group so wishes. I will return in approximately 25 minutes, so please begin your discussion.

PAGE 45

34 After reading the instructions to the group, the Experimenter began an audio cassette recorder visible to all participants Subjects were provided a copy of the "human relations problem" which served as the basis for group discussions. The specific questions which followed the narrative of the problem were designed to be both involving and timeconsuming: Susan and John are junior Sociology majors attending a large southern university. They met in a freshman class and began dating soon after. For the last six months, they have been living together in Susan's apartment. Last week, Susan received a phone call from her parents who told her that they are planning a visit with her next weekend. Susan has not told her parents about her and John living together, and she does not want them to find out (since they would not accept such an arrangement) Susan, therefore, has asked John to move all of his belongings out of the apartment for the weekend and to stay at a mutual friend's house. John refuses to leave, saying that he has nothing to hide and that he thinks it is very inconsiderate that Susan is asking him to do so. A. Is Susan justified by asking John to leave? B. Is John being inconsiderate of Susan's position? C. Should Susan tell her parents about John living with her?

PAGE 46

35 D. What would you, as a parent, think of this situation? E. How would this situation be altered (if at all) by the fact that John was a graduate student (not an undergraduate) ? Upon completion of the group discussion, the participants opened the door to the experimental room. The Experimenter turned the tape recorder off, excused the "confederate," debriefed the subjects, and thanked them for their participation. The "Confederate" While each group was composed of three persons, only two of the group members in any group were subjects in the experiment. The third member of each group was one of the two "confederates" or "plants" who were trained by the experimenter and contributed the anxiety-arousing "stimulus statement" that served as the manipulation for each experimental condition. The "confederates" in the present study were two male undergraduate students at the University of Florida. These "confederates," although ignorant of the specific hypotheses under investigation, were given specific guidelines for verbal and nonverbal conduct during the group discussions. Obviously, the contributions of the "confederates" could have posed a threat to the internal

PAGE 47

36 validity of the study if uncontrolled. Consequently, "confederates" were cautioned (a) to refrain from initiating substantive themes in group discussion (i.e., proposed solutions, arguments either in favor or in opposition to a proposed solution, personal disclosures) (b) to appear interested and involved in the participation of the other group members (through forward body lean, eye contact, smiling, nodding and vocalizations which support and confirm) (c) to limit their interactions to a minimum, (d) to utter the "stimulus statement" as close to the appropriate phase mark as was possible (between discussion on questions B and C of the problem) but to work the comment into the context of the group discussion so that the statement would not appear awkward or totally inappropriate, and (e) to permit the full and natural feedback responses of the group without either reiterating or compromising the force of the statement. Recognizing that the aforementioned guidelines reflect subtle and important skills, the "confederates" practiced their roles with the Experimenter. Only when the Experimenter determined that their group behavior met substantially the guidelines did the experimental sessions begin.

PAGE 48

37 Independent Variable: Process Disruption Statements As indicated in Chapter II, research has identified a variety of verbal statements which are potentially disrupting to the continuity of group communication. Using the taxonomy of disruptive message forms reported, messages were selected from three categories which could have a measureable impact on the level of receiver anxiety (i.e., messages which are considered anxiety producing) Consistent with these categories and drawing upon examples from previous investigations, the Experimenter composed ten statement for each of the three categories of anxietyproducing statements identified in the taxonomy. In addition, twenty statements considered non-anxiety producing were added (see Appendix C) This latter group included statements which are likely to be found in group discussions, but which do not generate anxiety by their function (i.e., they do not attack a person, refute a position or challenge group procedure) The 50 items were rated by students in three sections of the basic Speech course at the University of Florida on a 7-point bi-polar scale. Students were asked to assess the likelihood that each statement would foster anxiety or tension when uttered in the context of a small group.

PAGE 49

38 The particular scale used for this test is identical to the "tension" scale in Leathers FRI (1969) Items and responses were divided into "anxiety-producing" and "nonanxiety producing" categories. A t-test (Williams, 1979) was performed in an attempt to validate assumed categorical differences between the two groups of statements and to assure the Experimenter that the anxiety producing statements were indeed perceived by others as likely to produce high levels of anxiety among members. Results were both statistically significant (p < .05, df = 58) and in the direction predicted as anxiety arousing were rated more likely to produce anxiety (x = 5.1) than were non-anxiety arousing statements (x = 2.7). (Because of the low subject pool of high and low communication apprehensives only two categories of anxiety-arousing communications were used in this study.) A further precaution was taken to ensure that the stimulus statements were anxiety producing and salient. Five experienced judges, using a Q-sort technique, divided those statements which achieved a mean rating in excess of the group mean for the anxiety arousing statements into four categories; very highly anxiety arousing, highly anxiety arousing, moderately anxiety arousing, and somewhat

PAGE 50

39 anxiety arousing. Only those statements which were considered very highly/highly anxiety arousing were considered for this experiment. Then, two statements for each category were selected as the manipulation for each of the two conditions of anxiety arousing message forms. Each statement was used in three group sessions for that condition. The rationale for employing two different statements for a given condition was to provide some assurance that findings would not be specific to a single statement. Independent Variable: Communication Apprehension As previously reported in Chapter II, a variety of measures exist which purport to tap the concept of communication-bound anxiety. More than mere methodological differences, these instruments reflect important conceptual and theoretical differences in scope, situation and definition of communication-bound anxiety. The specific measure of communication-bound anxiety used in the present investigation was McCroskey's PRCA-College (Appendix B) There are a variety of reasons for employing this particular measure. First, the PRCA-College reflects a conceptualization of communication-bound anxiety which is consistent with the philosophy of this investigation, namely,

PAGE 51

40 that anxiety or apprehension can occur in a variety of communication contexts (from public speaking to small groups) and that a sensitive measuring device must be capable to tapping self-reports of anxiety across a variety of contexts. Second, the PRCA-College has been recently revised to include specific indicators of anxiety in a small group (McCroskey, 1978) Third, the PRCA actually includes four separate instruments which were tailored to specific age groups. The PRCA-College was specifically designed for and validated with college students, suggesting that it is highly appropriate for assessing apprehension levels among the targeted population for this study. Fourth, the PRCA has had a consistent history of high reliability with other measures as well as in test-retest reliability investigations (McCroskey, 1978) Finally, and most prominent, a solid argument can be made for the validity of the PRCA. As McCroskey stated: Probably the best indicator of the validity of a measure is the degree to which it can produce empirical results that are consistent with predictions based upon theory relating to the construct which the measure purports to tap. (1978, p. 193) To date, "five major theoretical propositions concerning oral communication apprehension have been set forth"

PAGE 52

41 (McCroskey, 1978, p. 193) First, "people vary in the degree to which they are apprehensive about oral communication with other people" (p. 193). As McCroskey notes, "this assumption has been supported in every report of research which has employed the PRCA (1978, p. 193). A second proposition states that "people with high oral communication apprehension seek to avoid oral communication" (p. 194) Again, research has consistently supported this proposition. For example, McCroskey and Leppard (1975) found that high apprehensives preferred housing that through previous research had been identified as requiring less interpersonal contact than low apprehensives. Daly and McCroskey (1975) found that high apprehensives had a strong preference for occupations requiring less verbal interaction than low apprehensives. Similarly, McCroskey and Anderson (1976) found that high apprehensives preferred large lecture classes over small classes, which would require more verbal interaction. These studies, being typical of those conducted in this area, demonstrate "that the PRCA is able to predict communication avoidance behaviors that would be expected" in line with the theory of communication apprehension and "provides a strong indication of predictive validity of the instrument" (McCroskey, 1978, p. 196)

PAGE 53

42 Third, "people with high oral communication apprehension engage in less oral communication than do less orally apprehensive people" (p. 196) When faced with situations that demand oral communication, communication apprehension theory would predict that individuals with high apprehension could withdraw from the interaction or remain as silent as possible. Many studies indicate that "high oral communication apprehensives engage in less oral communication behavior than do low oral communication apprehensives" (McCroskey, 1978, p. 196), again indicating support for the predictive validity of the PRCA. Fourth, "when people with high oral communication apprehension do communicate, their oral communication behaviors differ from those of people who are less apprehensive" (pp. 196-7). As previously noted, Wells and Lashbrook (1970) found that high communication apprehensives interacted less in small groups and that their interactions were much less relevant to the topic under discussion than were low communication apprehensive interactions. Similarly, and again previously noted, Sorenson and McCroskey (1977) found that high apprehensives exhibited more tension in small group interactions than did low communication appresives. These investigations as well as others in this

PAGE 54

43 area "again provide support for a claim of predictive validity for the PRCA" (McCroskey, 1978, p. 197). Fifth, "as a result of their oral communication behavior, high oral communication apprehensives are perceived less positively by others than are less apprehensive people" (p. 197) As previously reported, McCroskey (1976) found that high apprehensives were perceived by other group members as less extroverted, composed, competent, socially attractive and task-attractive than low apprehensives. McCroskey and Richmond (1976) have found similar results concerning peer perception and apprehension. As summarized by McCroskey (1978) : The theoretical propositions underlying the construct of oral communication apprehension have been consistently supported in research employing the PRCA to predict specific behaviors. These results taken together provide a strong indication of the predictive validity of this instrument. (p. 198) Given the results reviewed concerning the reliability of the PRCA and the evidence that the PRCA does predict those behaviors associated with the theory underlying the construct of oral communication apprehension (suggesting its validity) this investigation utilized the PRCA-College as the indicator of oral communication apprehension.

PAGE 55

44 For the present study, 177 students enrolled in sections of the basic Public Speaking course completed the 25-item PRCA -College. Subjects recorded one of five possible point values for each item ranging from "strongly agree" to "strongly disagree." The instrument has a possible range of total scores of 25-175 with a hypothetical mid-point of 75. According to McCroskey and other investigators, the standard deviation has ranged from 13 to 15. McCroskey (1978) has stated that: Based on previous interviews with subjects scoring at the various levels, it has been customary to consider subjects scoring above 88 (one standard deviation above the typical mean) to be "high" in communication apprehension and those scoring below 58 to be "low. While these cut-off points are not perfect (there will be some error particularly with subjects scoring very near the cut-offs) they have been found to be optimal in numerous studies. (p. 201) The method of differentiation used in this study was to consider the individuals scoring one standard deviation below the mean to be low in communication apprehension and those who scored one standard deviation above the mean to be high in communication apprehension.

PAGE 56

45 Dependent Measure: Quality of Group Feedback While a variety of procedures exist for measuring the quality of group communication, one measure seems particularly suited to the present investigation. Leathers (1969) FRI was selected as the most appropriate measure of group verbal feedback for several reasons. First, the intent of the instrument was to provide a multidimensional assessment of the feedback of group members to some form of contrived, process disrupting statement. Consequently, the methodology and applications of the instrument pioneered by Leathers are consistent, if not identical, to the methodology of the present investigation. Second, substantial refinement of the instrument has provided researchers with a technique for measuring verbal feedback which has been indicated to be both practical and reliable. Third, the nine scale items which comprise the rating instrument appear to tap the major qualitative dimensions of feedback which were important to the present investigation. Last, use of the Leathers' instrument provided a methodological link to a highly visable and acclaimed area of research in the field of Speech Communication. In this latter regard, results derived through use of the instrument could be

PAGE 57

46 compared and contrasted with previous findings in the area of process disruption. The procedure for applying the Leathers' instrument was rather straightforward. Using audio tapes recordings as the verbal record of the group discussion (see Appendix D) three judges independently rated each of the first five verbal utterances of group members following the injection of the designated stimulus statement into the group discussion on all nine scales of the instrument. Each scale is a seven-point bi-polar set of adjectives ranging from +3 through to -3. Above each item is the scale label (i.e., deliberateness relevancy, atomization, fidelity, tension, ideation, flexibility, digression and involvement) and beneath the numerical values which comprise the scale is a definition (in operative terms) of what behaviors are symptomatic of the adjective pairs. Leathers' FRI is contained in Appendix A. Summary This chapter summarized the major methodological considerations employed in the present study, including subject demographics, research design, experimental procedure, confederate training, independent and dependent

PAGE 58

47 variables. An analysis of variance was employed to determine the significant relationships between the variables

PAGE 59

CHAPTER V RESULTS AND DISCUSSION The purpose of this chapter is to present the results of this study and to discuss their impact upon the research hypotheses. Special attention is given to interrater reliability, test of hypotheses, and discussions of research findings R esults Interrater Reliability Pearson Product Moment Correlations were computed on the initial three judges' evaluations via: (l) overall reliability coefficients on each scale item across all groups, and (2) reliability coefficients on each cell across all nine scale items. Overall reliability coefficients between judges on each scale item across all groups were found to be quite low (see Appendix E) The experimenter then recruited two expert judges, 1 one in the field of small group research, the other an expert in rating error and scale implementation. As was the case with the initial three judges, the reliability 48

PAGE 60

49 coefficients between the two expert judges were quite low (see Appendix E) Since the reliability coefficients were low, and because averaging judges' ratings to test the hypotheses would be invalid die to the poor reliabilities, the subsequent evaluation of the research hypotheses consisted of two parts; each hypothesis was tested by evaluating the separate rating of each of the two expert judges, and then the experts ratings were compared/contrasted to discover how they perceived potential differences among the responses of experimental subjects. Tests of Hypotheses Hypothesis I — High communication apprehensives will exhibit significantly less deliberateness in their feedback responses than will low communication apprehensives Ratings of neither expert judge demonstrated a significant difference between high and low communication apprehensives regarding the deliberateness of their feedback responses to anxiety arousing communications. Hypothesis II — High communication apprehensives will exhibit significantly less relevancy in their feedback responses than will low communication apprehensives Although the ratings of the two expert judges did indicate

PAGE 61

50 statistically significant differences between the responses of apprehensive and non-apprehensive subjects, the nature of the differences was not the same. The scores of Judge #4 indicated that there was a main effect for the stimulus statement (F = 4.36, p < .05, see Table 2), while the scores of Judge #5 showed a significant interaction effect between the stimulus statement and communication apprehension (F = 4.73, p < .05, see Table 3). Hypothesis III — High communication apprehensives will e xhibit significantly greater atomization in their feedback responses than will low communication apprehensives In this instance the scores of Judge #4 indicated a significant main effect for communication apprehension (F = 4.38, p < .05, see Table 4) and also a significant main effect for the stimulus statement (F = 4.83, p < .04, see Table 4) The scores from Judge #5 did not demonstrate significant differences between the groups. Hypothesis IV — High communication apprehensives will exhibit significantly less fidelity in their feedback responses than will low communication apprehensives The scores from Judge #4 showed a significant main effect for the stimulus statement (F = 4.34, p <; .05, see Table 5). The scores from Judge #5 did not indicate any significant differences between the groups.

PAGE 62

51 Table 2 Analysis of Variance for Relevancy — Judge #4 Source df Type IV SS F Value P > F A* B** A*B 1 1 53.09184823 59.45784561 22.67620577 3.89 4.36 1.66 0.0640 0.0512 0.2135 A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements. Table 3 Analysis of Variance for Relevancy — Judge #5 Source df Type IV SS F Value p > F A* 1 1.96694600 0.28 0.6002 B ** 1 15.78152167 2.28 0.1481 A*B 1 32.69320298 4.73 0.0432 A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.

PAGE 63

52 Table 4 Analysis of Variance for Atomization — Judge #4 Source df Type IV SS F Value P > F A* 1 46.75307728 4.38 0.0508 B** 1 51.47279607 4.82 0.0414 A*B 1 25.03613128 2.35 0.1430 A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements. Table 5 Analysis of Variance for Fidelity — Judge #4 Source df Type IV SS F Value P > F A* 1 50.20789572 3.19 0.0910 B** 1 68.35966685 4.34 0.0517 A*B 1 59.13452514 3.76 0.0685 A* denotes the two levels of communication apprehension, B** denotes the two levels of stimulus statement.

PAGE 64

53 Hypothesis V — High communication apprehensives will exhibit significantly greater tension in their feedback responses than will low communication apprehensives The scores from Judge #4 showed an interaction effect between communication apprehension and the stimulus statement (F = 4.65, p < .04, see Table 6). The scores from Judge #5 did not indicate any significant differences between the groups Hypothesis VI — High communication apprehensives will exhibit significantly more ideation in their feedback responses than will low communication apprehensives The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between groups. Hypothesis VII — High communication apprehensives will e xhibit significantly more flexibility in their feedback responses than will low communication apprehensives The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between the groups. Hypothesis VIII — High communication apprehensiv es will exhibit significantly less digression in their feedback responses than will low communication apprehensives The scores recorded by both Judge #4 and Judge #5 did not demonstrate significant differences between the groups.

PAGE 65

54 Table 6 Analysis of Variance for Tension — Judge #4 Source df Type IV SS F Value P > F A* 1 151.78817505 11.91 0.0029 B** 1 35.99794820 2.82 0.1102 A*B 1 59.33379888 4.65 0.0447 A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.

PAGE 66

55 Hypothesis IX — High communication apprehensives will exhibit significantly less involvement in their feedback responses than will low communication apprehensives The scores recorded by Judge #4 did indicate a significant interaction effect between communication apprehension and the stimulus statement (F = 6.63, p <.01, see Table 7). The scores from Judge #5 did not indicate significant differences between the groups. Discussion Whenever researchers employ judge's ratings of human behavior in experimental investigations, reliability checks must be made to determine if the judges were rating the same behaviors in the same ways. Such reliability measures were conducted in this experiment and, due to the results of these reliability checks, the validity of the FRI has become an important initial focus of discussion in this analysis. In light of the low reliability ratings of the original three judges (all first year Ph.D. students), two expert judges were recruited to analyze the groups behaviors. One of these two judges (who was highly recommended by Leathers, the author of this scale) has

PAGE 67

56 Table 7 Analysis of Variance for Involvement — Judge #4 Source df Type IV SS F Value P > F A* 1 7.21613128 1.03 0.3230 B** 1 19.87343076 2.84 0.1090 A*B 1 46.34089851 6.63 0.0191 A* denotes the two levels of communication apprehension. B** denotes the two levels of stimulus statements.

PAGE 68

57 employed the scale in his own research and is an authority in the area of small group research and process evaluations, The second of the two judges is a leading authority on rating error, scale development and implementation. The reliability ratings of the two expert judges were strikingly similar to the reliability coefficients earlier discussed. Employing the same two evaluation techniques, and combining all five judges in the same analysis, the reliability coefficients for the two expert judges were dramatically low (see Appendix C) And, as was the case with the initial three judges, several correlations for the two expert judges were quite low with one correlation being negative. Even more surprising was the fact that no two judges demonstrated consistent mutual reliability among all nine scale items, which indicates that no two of the five judges consistently rated the same feedback responses in similar ways. These results demand direct and specific analysis of the FRI as a valid measuring tool in conjunction with rater reliability and the methodological problems in the use of the FRI. According to the data analysis of this investigation, several questions arise as to the sensitivity of the FRI to distinguish differences in verbal messages of

PAGE 69

58 individuals engaged in small group discussions. At the crux of this problem are the operational definitions for the scale items. Many, if not most of these definitions are rather ambiguous and vague. For example, scale item #1-Deliberateness is defined as n £~ a_Z symbol response represents a deliberate, carefully reasoned, logical response; / a_J7 signal response represents an immediate, unthinking, largely automatic, visceral response of Y to X. A statement could be very deliberate, carefully reasoned and logical but have nothing to do with the preceding (X) comment. How should this response be rated? Similarly, scale item #9-Involvement is defined as the "degree to which Y seeks to avoid comment on X's contribution by attempting to withdraw from the discussion of X's contribution." It is difficult to understand how judges could reasonably differentiate between whether a person is "seeking to avoid comment on X's contribution" or simply seeking to clarify, promote or continue quality progression toward the group goal. Beyond definitional problems, the FRI may have inherent reliability problems in analyzing the impact of a stimulus statement after the first feedback statement. Obviously, certain statements may have impact on the group

PAGE 70

59 process beyond one response statement. As stated by Leathers (1969) the premise of the FRI is "to provide an accurate quantitative reading of the feedback dimensions which characteristically follow a given type of message" (p. 291) This premise of this instrument, therefore, becomes inappropriate in the consideration of analyzing progressive, contiguous feedback responses in relation to the continuing impact of the stimulus statements. If this is the case, then the instrument does not necessarily measure group process; rather, it evaluates statements in isolation, which is not commendable to group process studies. This is not, however, a condemnation of the FRI. Rather, it is an appraisal as to the implicit premise given to the quantitative evaluation of the dimensions of feedback statements following a given stimulus statement. A causality problem may also exist in the qualitative analysis of progressive responses via the FRI. As responses are more removed in time and space from the stimulus statement, the scale becomes less viable in that judges cannot be sure whether or not the later responses are in fact due to the stimulus statement or other feedback responses. For example, feedback response #3 may demand direct attention from feedback response #4.

PAGE 71

60 Judge's evaluations of these responses then is difficult. Interpretation of these evaluations also becomes difficult in that the feedback responses may not deal with the stimulus statement, but the reasons are unclear. Training judges must also be discussed in relation to the low reliability coefficients exhibited. This experimenter followed all the general guidelines previously set forth in the literature in the training of the initial three judges. Through individual conferences and verbal confirmation, all three judges indicated a comprehensive understanding of the FRI and its implementation. Given the low reliability ratings of these three judges in addition to the low reliability ratings of the two recruited expert judges, this experimenter believes that highly specialized training on each scale item by earlier experimenters could explain the previously reported high interobserver reliabilities. The extent to these past trainings is to date not known by this experimenter. However, given the guidelines currently put forth in the literature, consistently high judges' reliability correlations cannot be expected.

PAGE 72

61 I mplications Certain implications can be drawn from this study. First, according to the expert judges evaluations, there appear to be some differences in how communication apprehensives respond to anxiety-arousing statements in a group context. It is, however, unclear exactly what these differences are. The two expert judges evaluations indicated that they perceived differences in some cases, yet they did not agree where these differences were located in relation to the FRI and to what degree these differences were exhibited. Judge #4 indicated that high communication apprehensive individuals exhibited significantly less relevancy, atomization and fidelity in their feedback responses to anxietyarousing communications than did low communicative apprehensives. Additionally, Judge #4 indicated an interaction between communication apprehension and the stimulus statement in the tension and involvement of feedback responses to anxiety-arousing communications. Judge #4 indicated that feedback responses of high communication apprehensives to facetiousness were significantly more tense than feedback responses of low apprehensives to personal commitments, of low apprehensives to facetiousness, and of high

PAGE 73

62 apprehensives to personal commitments Lastly, Judge #4 indicated that feedback responses of high apprehensive to facetiousness were significantly less involved than the feedback responses of high apprehensives to personal commitments, of low apprehensives to personal commitments, and of low apprehensives to facetiousness. On the other hand, the only significant difference between communication apprehensives feedback responses to anxiety-arousing statements indicated by Judge #5 was that there was an interaction of communication apprehension and the stimulus statement in the relevancy of feedback responses, The feedback responses of high apprehensives to personal commitments were significantly less relevant than the feedback responses of high apprehensives to facetiousness, of low apprehensives to facetiousness, and of low apprehensives to personal commitment. The only scale item in which both judges indicated differences in the responses of high and low communication apprehensives was in the relevancy of feedback. Although Judge #4 indicated a main effect for the stimulus statement and Judge #5 indicated an interaction effect, in both cases the feedback of high apprehensives was judged less relevant than the feedback of low apprehensives. This

PAGE 74

63 finding is consistent with previous research which has indicated that high communication apprehensives will exhibit less relevancy in their interactions in small groups than will low communication apprehensives (Well and Lashbrook, 1970) Even though the two expert judges did not agree about the type of effect, both indicated that high apprehensives exhibited less relevancy in their feedback responses than did low apprehensives. As indicated above, both expert judges recorded significant differences between high and low apprehensives on the relevancy of feedback responses. However, all other significant differences were demonstrated by only Judge #4. The implication of these interrater differences raises the issue of the FRI as a valid measure of group process. As Emmert and Brooks (1970, p. 222) point out, when a scale is developed the scholar must be careful in whether a procedure is designed to measure something about the object being judged or the individual doing the judging. In the present study, with such low reliability coefficients, it is not clear whether the FRI measured the object itself (feedback responses) or measured differences in the individuals doing the judging.

PAGE 75

64 The development of measuring instruments to assess differences in how individuals rate communication behaviors is a legitimate and needed research goal in the study of group process. The intent of the FRI is to measure the differences in the quality of feedback responses to certain stimulus statements, not differences in judges themselves. However, in this investigation the FRI manifested more differences between the judges than between the experimental groups. Therefore, investigators should be aware of the fact that the FRI may not exclusively measure differences in feedback responses to certain stimulus statements, but may also measure differences in a judge's ability to perceive differences in the object under evaluation.

PAGE 76

65 Note *-Two expert judges were recruited to evaluate the audio tapes of the group discussions. The first judge, Dr. Thomas J. Saine, is currently the Director of Cooperative Education at the University of Denver. Dr. Saine has published many articles in small group communication and rating error. He is a leading authority on the development, implementation and interpretation of the FRI The second judge, Dr. Douglas G. Bock, is currently an Associate Professor of Speech at Utah State University, Logan, Utah. Dr. Bock has a long and highly acclaimed research record in rating error, scale use and evaluation.

PAGE 77

CHAPTER VI DIRECTIONS FOR FUTURE RESEARCH As is the case with most research, this study has generated more questions than answers. Of primary importance is the continuing indepth investigations and implementation of the FRI as a measure of small group process. Interrater reliability problems were evidenced in this study and only through continuing study will these findings be explained as either internal reliability problems with the FRI or as problems within the unique combination of variables in this study. Similarly, future research should incorporate evaluations of judges from outside the area of speech communication to further validate the FRI. Future research should also investigate the effects of many different types of stimulus statements on the process of group interaction. Few have been investigated to date. Similarly, the partitioning of variables such as age, sex of participant, topic saliency and egoinvolvement could give explanatory power to these effects. 66

PAGE 78

67 Several methodological issues should be considered in future investigations. Pilot studies should be conducted to assure acceptable interjudge reliability and to assess that the stimulus statements were in fact process disruptive, primarily by post-test surveys of the subjects. Videotaping of all group discussions should be conducted to allow the judge's a full view of the group's verbal and nonverbal behavior, which could potentially increase acceptable interjudge reliability. Consideration should be given to drawing a sufficiently large sample size to allow the simultaneous analysis of three different content categories, thereby allowing comparisons of specific stimulus statement types on the same population. Additionally, scholars should incorporate stochastic models of analysis, such as Markov chain analysis and interaction process analysis, to determine if the content of feedback responses demonstrate consistent patterns through time. Finally, post-experimental interviews with the subjects could shed light on the results found and on the discussion of implication of those results.

PAGE 79

APPENDICES

PAGE 80

APPENDIX A FEEDBACK RATING INSTRUMENT

PAGE 81

Scale #1 Scale #2 Scale #3 + Symbol 3 APPENDIX A FEEDBACK RATING INSTRUMENT Deliberateness 10 1 Signal 3 Symbol response represents a deliberate, carefully reasoned, logical response; signal response represents an immediate, unthinking, largely automatic, visceral response of Y to X. Relevant 3 2 Relevancy 10 1 Irrelevant 2 3 Relevancy-extent to which Y seeks to establish the connection between X's comment and the comment which immediately preceded X's comment. Unified 3 Atomization 10 1 Atomized 3 Degree to which Y's contribution involves incomplete, fragmented or disjoined thought; includes running a number of ideas together; a number of people talking at the same time. 70

PAGE 82

Scale #4 Scale #5 Scale #6 71 Clear 3 Fidelity 10 1 Confused 3 Extent to which Y's response to X exhibits confusion as to the meaning and/or intent of X's original message; characterized by the necessity of Y seeking clarification, definition, expansion, etc. from X. Relaxed 3 Tension Tense 3 Degree to which non-verbal gestures like laughter, signs, groans, etc., indicate Y's relative state of tension or relaxation. Ideational 3 2 Ideation 1 Personal 3 Ideational responses involve an appraisal or evaluation of X's ideas; personal responses represent the degree to which Y's comments involve direct or implied criticism of X, as a person. Scale #7 Flexible 3 2 Flexibility 10 1 Inflexible 2 3 Inflexible response indicates Y's unwillingness to modify his position in response to X's contribution; may include a counter assertion.

PAGE 83

Scale #8 Concise 3 Digression 10 1 Digressive 2 3 Degree to which Y inhibits x's immediate response, primarily by means of lengthy and discursive utterances. 72 Scale #9 Involved 3 2 Involvement 10 1 Withdrawn 2 3 Degree to which Y seeks to avoid comment on X's contribution by attempting to withdraw from the discussion of X's contribution.

PAGE 84

APPENDIX B McCROSKEY'S PRCA -COLLEGE

PAGE 85

APPENDIX B McCROSKEY'S PRCA-COLLEGE DIRECTIONS: This instrument is composed of 25 statements concerning your communication with other people. Please indicate the degree to which each statement applies to you by marking whether you (1) Strongly Agree, (2) Agree, (3) Are Undecided, (4) Disagree, or (5) Strongly Disagree with each statement. There are no right or wrong answers. Work quickly, just record your first impression by circling the correct number. 1. While participating in a conversation with a new acquaintance I feel very nervous 12 3 4 5 2. I have no fear of facing an audience. 12 3 4 5 3. I talk less because I'm shy. 12 3 4 5 4. I look forward to expressing my opinions at meetings. 12 3 4 5 I am afraid to express myself in a group. 12 3 4 5 6. I look forward to an opportunity to speak in public. 12 3 4 5 7. I find the prospect of speaking mildly pleasant. 12 3 4 5 8. When communicating, my posture feels strained and unnatural. 12 3 4 5 9. I am tense and nervous while participating in group discussions. 12 3 4 5 10. Although I talk fluently with friends, I am at a loss for words on the platform. 12 3 4 5 74

PAGE 86

75 11. I have no fear about expressing myself in a group. 12 3 4 5 12 My hands tremble when I try to handle objects on the platform. 12 3 4 5 13. I always avoid speaking in public if possible. 12 3 4 5 14. I feel that I am more fluent when talking to people than most people are. 12 3 4 5 15. I am fearful and tense all the while I am speaking before a group of people. 12 3 4 5 16. My thoughts become confused and jumbled when I speak before an audience. 12 3 4 5 17. I like to get involved in group discussion. 12 3 4 5 18. Although I am nervous just before getting up, I soon forget my fears and enjoy the experience. 12 3 4 5 19. Conversing with people who hold positions of authority causes me to be fearful and tense. 12 3 4 5 20. I dislike to use my body and voice expressively. 12 3 4 5 21. I feel relaxed and comfortable while speaking. 12 3 4 5 22. I feel self-conscious when I am called upon to answer a question or give an opinion in class. 12 3 4 5 23. I face the prospect of making a speech with complete confidence. 12 3 4 5

PAGE 87

76 24. I'm afraid to speak up in conversation. 12 3 4 5 25. I would enjoy presenting a speech on a local television show. 12 3 4 5

PAGE 88

APPENDIX C PRETEST OF ANXIETY AROUSING MESSAGES

PAGE 89

Please read each of the following statements and record the appropriate number on the scale given as to your perception of HOW ANXIETY PRODUCING EACH STATEMENT WOULD BE IF GIVEN IN A SMALL GROUP DISCUSSION. (non-anxiety producing) (anxiety-producing) Relaxed Tense 1 2 3 4 5 6 7_ Degree to which this statement would be anxiety-producing if given in a small group discussion. 1. Susan reminds me of an old story about Ronald Reagan — up drove an empty car and out stepped Ronald Reagan. 2. John sounds so old fashion that he probably thinks that Hoover is still President. 3. Susan reminds me of what a friend once told me: You may be dumb but you sure are ugly. 4. Susan just needs to get her folds stoned — then they will agree to Anything 5. Poor John, sounds like the Gator football team — 3rd and long, so fall back and punt I 6. Do you suppose John wears white socks? *7. John probably uses pot as an aphrodisiac. I heard he seduced three guys on the way to school last week. *8. I hope Susan can make love better than she treats John. 9. I haven't had so much fun in a group since the pigs ate my baby brother. 78 Rating

PAGE 90

79 Rating 10. I guess John now has a new interpretation of the song "Another One Bites the Dust." 11. Don't you think that this is a matter of modern social morality? 12. Don't you feel that this is more a matter of E.R.A. tenets than of personal commitment? 13. I feel the question is more concerned with religious adjudication than cohabitation, don't you? 14. We haven't addressed the question of their interpersonal heterophily; it should be addressed. 15. They seem so diametriacally opposed — two extremely bi-polar views of this circumstance. *16. Isn't this really a problem of irrelevant personal catharsis? *17. It seems that John's stance is the result of a cognitive occlusion. 18. Susan is really using John as a personal therapeutic tool. 19. Realistically, this is a situation of high versus low cognitive complexity. 20. Wouldn't you say that Freud would call this a problem of libido? 21. John must be totally egocentric--He thinks only of himself. *22. Susan is totally two-faced — I don't see how anyone could disagree with that.

PAGE 91

80 Rating 23. I guarantee you that if the situation was reversed that John would do the same thing. *24. John's so completely uncaring — look at how he is treating Susan. 25. I can't understand why you feel that way. Obviously they are both uncaring about each other's feelings. 26. John should buy Susan an engagement ring. At least they would then have something to fall back on. 27. I don't like John's attitude of "nothing to hide." If he has any feelings at all, he'd reconsider. 28. How does Susan know her parents are against this? I strongly feel that they should just cool it and tell Susan's folks the truth. 29. There is only one way out of this mess and that is to face up to Susan's parents. 30. Both John and Susan are completely selfish. There is no way around that. 31. I agree with that. That sounds good to me. 32. I really do not understand what you mean. Could you rephrase that? 33. I am glad you feel that way. So do I. 34. I think I will take some notes on this stuff, ok? 35. I believe we are on the right track here. 36. I don't know about that. What do the rest of you think?

PAGE 92

81 Rating 37. I like the way you said that. That is certainly my position on the topic. 38. I think the involvement of all of us in this project is commendable. 39. It sure seems that we are a hardworking group. 40. According to my past experience, I'd say that we are pretty much accurate in our analysis 41. I find it really interesting that you mentioned such a proposition. 42. Then do we agree that the crux of the matter is our lack of knowledge on this subject? 43. I have found that there are a lot of people really concerned with this. 44. Let me add one more thing to help your answer. 45. I think what you have just said makes a lot of sense. 46. It seems as if you have just stated our central theme to this problem. 47. Yes, I do believe that our major problem is definitional. 48. Do we have any volunteers to look into that proposal? 49. Will you agree with me that this seems to be the essence of the problem we've been discussing? 50. That's correct. I concur. indicates those statements used in the study.

PAGE 93

APPENDIX D SAMPLES OF GROUP INTERACTIONS

PAGE 94

Sample #1 — High Communication Apprehensives Stimulus Statement: John probably uses pot as an aphrodisiac. I heard he seduced three guys on the way to school last week. F eedback Statement #1 : Gee, anything else you can think about? I think we've covered all the ground. Feedback Statement #2 : What do you think. Does John really care for Susan? F eedback Statement #3 : Well, he's not helping her out any. He's putting his foot down this time and I don't think he cares that much. Feedback Statement #4 : I think that he should... uh, at first, uh...I don't know what I think. F eedback Statement #5 : Uh, I think that if, uh, John really cares for Susan, uh, he should really plan ahead at first, and, at first he should plan. Of course, uh, they are still Juniors. Sample #2 — Low Communication Apprehensives Stimulus Statement : I hope Susan can make love better than she treats John. Feedback Statement #1 : Ok, well, how would you as a parent feel about this situation? Your daughter ,.. .the males over there. How would you feel, cause it's usually the male that, uh, is stereotyped as getting the most upset and uptight. The mother calming them down, you know, so how would you guys feel if it were your daughter and you found out that she had been living with somebody for six months and hadn't told you about it for six months. Feedback Statement #2 : I would be upset that she didn't tell me, but I .. .personally I don't see anything wrong with it, instead of getting married and not knowing the person. 83

PAGE 95

84 Feedback Statement #3 ; Yeah, but don't forget this is your own daughter, I mean... Feedback Statement #4 : Yeah, that's what I'm saying, and I, you know, I thought about that. Uhm, I would be upset that she didn't have enough confidence to come and tell me what's going on, but... you know. If she broke it to me in the right way, you know, she told me, "look, I've known this guy for three years. I've been living with him for the past three months, six months." You know, then, you know, then some type of feelings have emerged. It's just not a one night stand type thing. Feedback Statement #5 : How about you. Come on, the other male.

PAGE 96

APPENDIX E CORRELATION COEFFICIENTS FOR INTERRATER RELIABILITIES

PAGE 97

Judge Judge #1#2 Judge Judge #1#3 Judge Judge #2#3 Judge #4Judge #5 Scale #1 Deliberateness .243 .547 .287 .346 Scale #2 Relevancy .044 .404 -.014 -.182 Scale #3 Atomization .127 .172 .392 .003 Scale #4 Fidelity -.014 -.25 .143 .272 Scale #5 Tension -.096 .279 .352 .603 Scale #6 Ideation .046 -.121 .303 .049 Scale #7 Flexibility .302 .148 .079 .116 Scale #8 Digression .142 -.113 -.061 .411 Scale #9 Involvement -.107 -.106 .368 .414 86

PAGE 98

REFERENCES Bales, R. F. Interaction Process Analysis: A Method for the Study of Small Groups Cambridge, Mass.: AddisonWesley, 1950. Behnke, R. R., & Carlile, L. W. Heart Rate as an Index of Speech Anxiety. Speech Monographs 1971, 38, 65-69. Burgoon, J. K. The Unwillingness-to-Communicate Scale: Development and Validation. Communication Monographs 1976, 43, 60-69. Burgoon, J. K., & Burgoon, M. Unwillingness to Communicate, Anomia-Alienation, and Communication Apprehension as Predictors of Small Group Communication. Journal of Psychology 1974, 88, 31-38. Carlile, L. W. & Behnke, R. R. A Report to the Criticism of "Heart Rate as an Index of Speech Anxiety. Speech Monographs 1973, 40, 156-159. Clark, A. J. Identifying the Anxious Communicator. Florida Speech Communication Journal 1973, 1, 17-22. Clevenger, T. C. A Definition of Stage Fright. Central States Speech Journal 1955, 6_, 26-30. Clevenger, T. C. A Synthesis of Experimental Research in Stage Fright. Quarterly Journal of Speech 1959, 45, 135-145. Crowell, L. & Scheidel, T. M. Categories for Analysis of Idea Development in Discussion Groups. Journal of Social Psychology 1961, 54, 155-168. Daly, J. A. & McCroskey, J. C. Occupational Choice and Desirability as a Function of Communication Apprehension. Journal of Counseling Psychology 1975, 22, 309-313. "" 87

PAGE 99

88 Dicken, M., Gibson, F., & Prall, c. An Experimental Study of the Overt Manifestations of Stage Fright. Speech Monographs 1950, 17_, 37-47. Dicken, M. & Parker, W. R. An Experimental Study of Certain Physiological, Introspective and Rating Scale Techniques for the Measurement of Stage Fright. Speech Monographs 1951, 18, 251-259. Eckert, R. G. & Key, N. Public Speaking as a Cue to Personality Adjustments. Journal of Applied Psychology 1940, 24, 144-153. Emmert, P., & Brooks, W. D. Methods of Research in Communication Boston, Mass.: Houghton Mifflin Company, 1970. Giffin, K., & Gilham, S. M. Relationships Between Speech Anxiety and Motivation. Speech Monographs 1971, 38, 70-73. Gilkenson, H. Social Fears as Reported by Students in College Speech Classes. Speech Monographs 1942, 9, 141-160. Greenleaf, F. An Experimental Study of Social Speech Fright. (Master's Thesis, State University of Iowa, 1947) Kline, J. A. Indices of Opinionated and Orientating Statements in Problem-Solving Discussions. Speech Monographs 1970, 37_, 282-286. Kline, J. A. Orientation and Group Consensus. Central States Speech Journal 1972, 2_3, 44-47. Lashbrook, W. B. Lashbrook, V. J., Bacon, C, & Salinger, S An Empirical Examination of the Relationship Between Communication Apprehension and Tolerance of Ambiguity. Southern Speech Communication Journal 1979, 44 244-251. Leathers, D. G. Process Disruption and Measurement in Small Group Communication. Quarterly Journal of Speech 1969, 40, 287-300.

PAGE 100

89 Leathers, D. G. The Process Effects of Trust-Destroying Behavior in the Small Group. Speech Monographs 1970, 37, 180-187. Leathers, D. G. Testing for Determinent Interactions in the Small Group Communication Process. Speech Monographs 1971, 38, 182-189. Leathers, D. G. Quality of Group Communication as a Determinant of Group Product. Speech Monographs 1972, 39, 166-173. Lustig, M. W. Verbal Reticence: A Reconceptualization and Preliminary Scale Development Paper presented at the meeting of the Speech Communication Association, Chicago, 1974. Lustig, M. W., & Grove, T. G. Interaction Analysis of Small Problem-Solving Groups Containing Reticent and Non-Reticent Members. Western Speech 1975, 29, 155-164. Marr, T. J. Counciliation and Verbal Responses as Functions of Orientation and Threat in Group Interaction. Speech Monographs 1974, 4_1, 6-18. McCroskey, J. C. Measures of Communication-Bound Anxiety. Speech Monographs 1970, 3_7, 269-277. McCroskey, J. C. The Problems of Communication Apprehension in the Classroom Paper presented at the meeting of the Communication Association of the Pacific, Kobe, Japan, 1976. McCroskey, J. C. Oral Communication Apprehension: A Summary of Recent Theory and Research. Human Communication Research 1977, 4_, 78-96. McCroskey, J. C. Validity of the PRCA as an Index of Oral Communication Apprehension. Communication Monographs 1978, 45, 192-203.

PAGE 101

90 McCroskey, J. C, & Anderson, J. F. The Relationship Between Communication Apprehension and Academic Achievement Among College Students Human Communication Research 1976, _3, 67-72. McCroskey, J. C, Daly, J., Richmond, V., & Falcone, R. L. Studies of the Relationship Between Communication Apprehension and Self-Esteem. Human Communication Research 1977, 3, 269-277. McCroskey, J. C, Daly, J. A., & Sorenson, G. Personality Correlates of Communication Apprehension: A Research Note. Human Communication Research 1976, 4_, 376380. McCroskey, J. C, & Leppard, T. The Effects of Communica tion Apprehension on Nonverbal Behavior. Paper presented to the Eastern Communication Association Convention, New York, 1975. McCroskey, J. C, & Richmond, V. P. The Effects of Communication Apprehension on the Perception of Peers. Western Speech 1976, 40, 14-21. Mulac, A., & Sherman, A. R. Behavioral Assessment of Speech Anxiety. Quarterly Journal of Speech 1974, 60, 135-143. Phillips, G. M. Reticence: Pathology of the Normal Speaker. Speech Monographs 1968, 3J7, 39-49. Phillips, G. M. Rhetoritherapy versus the Medical Model: Dealing with Reticence. Communication Education 1977, 2_6, 34-43. Porter, D. T. Self-Report Scales on Communication Apprehension and Autonomic Arousal: A Test of Construct Validity. Speech Monographs 1974, 4_1, 267-276. Porter, D. T., & Burns, G. P. A Criticism of "Heart Rate as an Index of Speech Anxiety. Speech Monographs 1973, 40, 156-159. Prentice, D. S. The Effects of Trust-Destroying Communication on Verbal Fluency in the Small Group. Speech M onographs 1975, 42_, 262-270.

PAGE 102

91 Rosenfeld, L. B. & Plax, T. Personality Discriminants of Reticence. Western Speech 1976, 3_7, 22-31. Saine, T. J., Shulman, L. S., & Emerson, L. C. The Effects of Group Size on the Structure on Interaction in Problem-Solving Groups. Southern Speech Communication Journal 1974, 3_9, 333-345. Scheidel, T. M., & Crowell, L. Feedback in Small Group Communication. Quarterly Journal of Speech 1966, 52, 273-278. Sorensen, G. A., & McCroskey, J. C. The Prediction of Interaction Behavior in Small Groups: Zero History vs. Intact Groups. Communication Monographs 1977, 44, 73-80. Stech, E. L. An Analysis of Interaction Structure in the Discussion of a Ranking Task. Speech Monographs 1970, 37, 249-256. Thorenson, C. E. A Study of Characteristics. American Education Research Journal 1966, 3_, 207-216. Watson, D., & Friend, R. Measurement of Social-Evaluative Anxiety. Journal of Consulting and Clinical Psychology 1969, 33_' 448-457. Wells, J., & Lashbrook, W. A Study of the Effects of Systematic Desensitization on the Communication Anxiety of Individuals in Small Groups Paper presented at the meeting of the Speech Communication Association, New Orleans, 1970. Williams, Frederick. Reasoning with Statistics New York, New York: Holt, Rinehart, and Winston, 1979.

PAGE 103

BIOGRAPHICAL SKETCH James David Harris was born on July 29, 1953, in Bowling Green, Kentucky, to David L. and Mary E. Harris. He graduated from Bardstown High School in Bardstown, Kentucky, in 1971. In the fall of 1971, Jim began his undergraduate career at Western Kentucky University in Bowling Green, Kentucky, majoring in psychology and speech. He received his B.A. degree in the fall of 1975 and then began his master's program at Western Kentucky in speech communication. He received his M.A. in 1977. Upon graduating from Western Kentucky, he began his Ph.D. program at the University of Florida in speech communication. Jim was indeed fortunate enough to have presented at the SSCA convention in Austin, Texas, April of 1981, the #1 ranked paper in the Vice President's Spotlight Program for Debut Papers. He concentrated his efforts in his graduate program in small group research, research methodologies, and interpersonal communication. He will be taking a position this fall as the Placement Officer for Graduate Students in the Department of Cooperative Education at the University of Denver. 92

PAGE 104

93 The day after he graduates from the University of Florida, Jim will marry Miss Melanie Jeanne Ferreira of Fernandina Beach, Florida, a former student of his. Jim assures everyone that this relationship began after the class was completed.

PAGE 105

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. CtiUlUUii V ^ biiuk Anthony J. Clark, Chairman Associate Professor of Speech I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Thomas B. Abbott Professor of Speech I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. 'Paul J Professor I certify that I' have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. n r// >/ ... Marvin E. Shaw Professor of Psychology

PAGE 106

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. j5^uM^Z>^nM^Aj^> Donald E. Williams Professor of Speech This dissertation was submitted to the Graduate Faculty of the Department of Speech in the College of Liberal Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1981 Dean for Graduate Studies and Research

PAGE 107

UNIVERSITY OF FLORIDA 3 1262 08554 2719


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E1577OZBI_WKHAPF INGEST_TIME 2017-07-11T22:09:24Z PACKAGE AA00004904_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES