Citation
The use of emotive words in intercollegiate debate

Material Information

Title:
The use of emotive words in intercollegiate debate
Creator:
Kish, Gerald R., 1957- ( Dissertant )
Markel, Norman ( Thesis advisor )
Jensen, Paul J. ( Reviewer )
Clark, Anthony J. ( Reviewer )
Bennet, Gerald T. ( Reviewer )
Randles, Ronald H. ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
1991
Language:
English
Physical Description:
vii, 175 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Applied statistics ( jstor )
Debate ( jstor )
Modeling ( jstor )
Nondestructive testing ( jstor )
Rebuttal testimony ( jstor )
Speeches ( jstor )
Statistical models ( jstor )
Statistics ( jstor )
Strong arguments ( jstor )
Words ( jstor )
Communication Processes and Disorders Thesis, Ph. D.
Debates and debating ( lcsh )
Dissertations, Academic -- UF -- Communication Processes and Disorders
Rhetoric ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
This study investigated the relationship between emotive words and five variables of academic debate: winning, speaker position, side of topic, year of debate, and gender mixture of the teams. Debate is an art that dates to the beginning of civilization. Research has to determined the keys to successful advocacy. Several scholars suggest that emotive words are vital to successful communication. No one has investigated emotive words in debate. Transcripts of nineteen years of intercollegiate debate from the National Debate Tournament were anlyzed by a computer aided content analysis. The empirical focus utilized emotive words as the depended variable. The data were analyzed by the use of six SAS regression procedures. The results of this study's five research questions, related to academic debate and the use of emotive language, are: 1. Do winning and losing debate teams differ in their use of emotive words during rebuttal speeches in a debate round? No support was found. 2. do affirmative and negative debaters differ in their use of emotive words in rebuttal speeches? No support was found. 3. Do the first and second rebuttal speakers for each debate team differ in their use of emotive words? No support was found. 4. Does the quantity of emotive words used by debaters vary over a period of years? The results suggested that the use of emotive words was related to the year variable. 5. Does the quantity of emotive words used differ between mixed gender debate teams and single gender debate teams? The results found that mixed-gender teams use greater emotive words. The significant result related to research question number four is the most important finding of this study. The statistical analysis that debaters utilized increasingly greater number of emotive words over the nineteen years of debates studied. The most important contribution of this study is that academic debate is placed in an empirical context. That is, this research is a significant step in closing the breach between the academic debate community and the quantitative researchers in the communication field.
Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 155-174).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Gerald R. Kish.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. 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:
001753184 ( alephbibnum )
AJG6147 ( notis )
26585639 ( oclc )

Downloads

This item has the following downloads:


Full Text














THE USE OF EMOTIVE WORDS IN INTERCOLLEGIATE DEBATE


BY


GERALD R. KISH





















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







UNIVERSITY OF FLORIDA

1991


UNIVERSITY OF FLORIDA LE:AoRIES














ACKNOWLEDGMENTS


I would first like to thank my parents, Gerald and Sue

Kish, for their constant support and encouragement.

I would also like to express my sincere appreciation to

my committee chairman, Dr. Norman Markel, for his continued

support, guidance, and encouragement throughout the course

of this study. His enthusiasm and dedication have greatly

contributed to making both the completion of this study and

my graduate program enjoyable and worthwhile experiences.

I would like to also express my thanks to Dr. Paul

Jensen for his rigorous editorial assistance and his never-

ending enthusiasm.

Special thanks are extended to Dr. Anthony J. Clark for

twice recruiting me to the University of Florida. I

greatly appreciate what he has taught me, both in the field

of debate and in the conducting of research.

Many thanks to Dr. Ronald Randles for his assistance in

the generation of the experimental design and for providing

me with many statistical methods.

It is with great appreciation that I acknowledge

Professor Gerald Bennett for providing me with an

opportunity to develop my quantitative research skills

ii







in a significant research project. I also appreciate his

relentless support and encouragement.

I would like to offer special thanks to Associate in

Psychiatry Lynn Robbins for both her unending help with my

statistics and her friendship.

I would also like to offer special thanks to Chris

Morris for all his help with this project.

I cannot thank Linda Harris enough for all her help.

I thank Dr. Marsha Vanderford, Dr. George Barnard, Dr.

Gus Newman, Dr. Anita Raghavan, Dr. Ringo Ma, Mrs. Jane

Milam, Kelly Roberts, Teresa Gudaitis, Roberta Bell, Lise

Kent, Trish Sample, Donna Arents, Ann Allison, Roseanna

Rutledge, Clem Sepulveda, Sara McDaniel and Sam Walch.

Most especially, I would like to express my deepest

thanks to Lynn Kish for sharing in everything I do with

enthusiasm, friendship, and love. Also, my thanks for

getting her doctorate while waiting on mine.


iii

















TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS . . . ....... ii

ABSTRACT . . . . . .vi


CHAPTER ONE

INTRODUCTION . . . . 1

Overview . . . . . 1
Literature Review . . . . 3
Brief History of Debate . . . 4
Importance of the Rebuttal Speeches. . 7
Judging in Academic Debate . . . 9
Emotive Words . . . . 12
Quantitative Research in Debate . .. 17
Word Choice and Debate . . . .. 23
Word Choice and Legal Research . . .. 27
Empirical Focus .. . . . . 32
The Dependent variable . . . 32
Gottschalk-Gleser Content Analysis Scale 32
The Semantic Differential Scale . .. 34
Validity and Reliability . . .. 35
Research Questions . . . .. 36
Notes . . . . .. . 38

CHAPTER TWO

METHODOLOGY . . . . . 53

Research Design . . . . .. 53
Variables . . . . .. .. 53
Procedure . . . .. 55
Transcripts for Analysis. ..... . .. 55
Subjects . . . .. . 56
Materials . . . . . 56
Identifying Emotive Words . . .. 57
Statistical Model . . . .. 59
Notes . . . . . 61








CHAPTER THREE


RESULTS . . . .

The Study . . .
Linear Regression Analysis .
Regression Analysis . .
Research Question Results .
Summary . . . .


S . . 64

S . . 64
S . . 115
S . . 119
S . . 138
S . . 139


CHAPTER FOUR


DISCUSSION . . .

Overview of the Study . .
Interpretation of the Results .
Implications for Future Research
Limitations of the Study .



REFERENCES . . .

BIOGRAPHICAL SKETCH . .


. . . 141


. 141
. 143
* 152
. 153


S. . ... 155

S. . ... 175


j j
















A Dissertation Presented to the Graduate School of the
University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE USE OF EMOTIVE WORDS IN INTERCOLLEGIATE DEBATE


by

Gerald R. Kish

May, 1992

Chair: Norman Markel, Ph.D

Major Department: Communication Processes and Disorders

This study investigated the relationship between

emotive words and five variables of academic debate:

winning, speaker position, side of topic, year of debate,

and gender mixture of the teams.

Debate is an art that dates to the beginning of

civilization. Research has not determined the keys

to successful advocacy. Several scholars suggest that

emotive words are vital to successful communication. Noone

has investigated emotive words in debate.

Transcripts of nineteen years of intercollegiate

debate from the National Debate Tournament were analyzed by

a computer aided content analysis. The empirical focus

utilized emotive words as the dependent variable.

The data were analyzed by the use of six SAS regression







procedures. The results of this study's five research

questions, related to academic debate and the use of emotive

language, are: 1. Do winning and losing debate teams differ

in their use of emotive words during rebuttal speeches in a

debate round? No support was found. 2. Do affirmative and

negative debaters differ in their use of emotive words in

rebuttal speeches? No support was found. 3. Do the first

and second rebuttal speakers for each debate team differ in

their use of emotive words? No support was found. 4. Does

the quantity of emotive words used by debaters vary over a

period of years? The results suggested that the use of

emotive words was related to the year variable. 5. Does the

quantity of emotive words used differ between mixed gender

debate teams and single gender debate teams? The results

found that mixed-gender teams use greater emotive words.

The significant result related to research question

number four is the most important finding of this study. The

statistical analysis that debaters utilized increasingly

greater number of emotive words over the nineteen years of

debates studied. The most important contribution of this

study is that academic debate is placed in an empirical

context. That is, this research is a significant step in

closing the breach between the academic debate community and

the quantitative researchers in the communication field.


vii















CHAPTER ONE


INTRODUCTION

Overview




"Ideally, all judges and all debaters should

proceed from a commonly understood set of rules for making

the decision. Unfortunately, different types of judges

have different standards (Thomas, 1987c, p.123)."

Researchers in the field of contemporary academic debate

have attempted to discover what overall factors contribute

to the debate decision. Among the factors that researchers

have studied are characteristics of the judge's ballot

(Berthold, 1970), judging bias (McCroskey & Camp, 1966;

Brooks, 1971), stock issues in debate, the ability of

debaters to adapt (Markgraf, 1966; Verderber, 1968; Pearce,

1969), gender of the debater (Hayes & McAdoo, 1972; Hill,

1973b; Rosen, Dean, & Willis, 1978), side of the issue

(Hill, 1973b; Rosen, Dean, & Willis, 1978), competitiveness

of the debaters (Wilson, 1979), rate of speech (Vasilius &

DeStephen, 1979), evidence use (Larson & Giffin, 1964;

Dresser, 1964; Newman & Sanders, 1965; Dresser, 1966; Rieke

& Smith, 1968), qualities of arguments (Allen & Kellermann,

1988), debate jargon (Vasilius & DeStephen, 1979), and the








use of praise and derogation (Burgoon, Wilkinson, &

Partridge, 1979). None of these studies have demonstrated a

clear relationship between any one specific factor and

winning in intercollegiate debate.

Only a few studies have examined the use of language

as a factor in winning a debate. Rouse and Thomas (1987)

found that debaters changed their word usage over a span of

13 years. Newman (1939) argued that while language appears

to have a rational, or logical, basis, communication

(language) often functions through irrational, or illogical,

means. In general, research has also found that certain

types of language are related to psychological

characteristics of the speaker (Markel, 1990; Gibson &

Felkins, 1974; Gottschalk, 1961; Newman, 1939; Newman &

Mather, 1938). In addition, other research (Parkinson,

Geisler, & Pelias, 1983; Parkinson, 1981; Conley, 1979;

O'Barr & Conley, 1976) has found that word choice can have

an influence on success in an actual trial setting. Further

research has demonstrated the existence of emotive words

(Clore, Ortony, and Foss, 1987; Irene, 1990). Each of these

conclusions may be applied in an investigation of the issue

of the effect of emotive language on debate decisions.

To better understand the importance of language in

debate, one must first consider the functions of language.

The emotive function of language has been described as a

fundamental dimension of human life (Sapir, 1921; Jakobson,

1960; Irvine, 1990). The emotive function of language

communicates a speaker's emotions and attitudes (Fiske,








1982). The Gottschalk and Gleser (1969) hostility scale

and the Osgood, Suci, and Tannenbaum (1957) semantic

differential scale both purport to measure the speaker's

emotions and attitudes. A modified emotive word list was

developed from these two scales by Markel (1989). It

allows analysis by categorizing words into positive and

negative words.

The purpose of this study was to investigate the

use of emotive words in collegiate debate. While the

emotive function of language is important, the use of

emotive words in academic debate has never been

investigated. Initially, the significance of debate will be

discussed, followed by the history of debate,

characteristics of the national debate tournament, speaker

duties and the importance of the rebuttal speech, and debate

judges and their judging paradigms in debate. Following

this, a discussion of function of emotive words and debate

will be considered. Next, a discussion of the

quantitative research in debate, and of research of language

effects in both debate and the field of law will be

presented. Finally, a discussion of the dependent variable

will be offered. The methodology for the study will be found in

chapter two. The results will be found in chapter three and

the discussion in chapter four.

Literature Review

From the earliest times in recorded history, thoughtful

people have recognized the importance of debate for both

society and the individual (Freeley, 1976, p.2). Debate has








a "long and honored place in academia" (Colbert & Biggers,

1985, p.237).

Debate is a process of inquiry and argumentation that

seeks a reasoned judgment of a proposition. Patterson and

Zarefsky (1983) defined debate as argumentation that occurs

in a formal setting. To find a resolution of two or more

differing positions involves argumentation and, very often,

debate.

Differences between people or between groups of people

can be resolved by five different means (Patterson &

Zarefsky, 1983, p.309): first, a resolution may be reached

by force; second, decisions may be made based on impulse or

caprice; third, resolutions may be decided by chance;

fourth, one can allow another party to make the decision,

and, finally, each side can argue the issue and attempt to

persuade the other. The most logical and reasonable method

to reach a resolution is the fifth means, that of

argumentation.

Our society provides many opportunities for a wide

variety of types of argumentation and debate, including

political and judicial debates, parliamentary debates and

academic debate.

Brief History of Debate

An examination of the history of debate is important in

understanding the evolution of the activity and providing a

foundation from which debate can be evaluated. Debate has

always been viewed as a process of rational decision making

(Petrie, 1969); understanding the factors involved in such








decision making may clarify the importance of rationality in

such decisions. Briefly, debate appears to have first

been recorded by scholars of the Chou Dynasty some 3000

years ago (Freeley, 1976 p.17). Debate also appears in

Homer's Iliad and Odyssey (Ovid, 1955; Sproule, 1974) and in

Aristotle's Rhetoric (Murphy, 1964; Sproule, 1974).

Medieval educators required disputations of their students

and intercollegiate debate began as early as the fifteenth

century (Murphy, 1964; Freeley, 1976). From these early

intercollegiate contests, debate organizations were formed

and debate tournaments organized (Coburn, 1972). In the

United States, these organizations and debate tournaments

have undergone many changes (McBath & Aurbach, 1967).

The debates investigated in this study were those of

championship rounds at the National Debate Tournament. The

National Debate Tournament (NDT) uses a national resolution

that is chosen in the fall of each year and debated by

college debaters through the year. The NDT is the final

tournament of the year and usually occurs in late April or

May. The NDT resolution is not the only debate topic that

college debaters can debate. There have always been other

resolutions, but the NDT resolution was considered the most

important for many years. Matlon and Keele (1984) surveyed

703 participants in the NDT and reported that some

of the most intellectually gifted students have participated

in the National Debate Tournament" (p.195). After obtaining

the advanced degrees, the debaters were found having

advanced professional positions of significant








responsibility, such as cabinet members, congresspersons,

ambassadors, judges, and other high status professional

positions (p.195). Several respondents compared the NDT to

the Super Bowl, the Master's Golf Tournament, or to the

World Series (p.200). In the early 1970s, another type of

resolution was created by the Cross-Examination Debate

Association (CEDA). The CEDA topic was a non-policy

resolution and was created to counterbalance perceived

abuses of NDT, including excessive speaking rates. CEDA

started slowly, but by the middle 1980s, CEDA had replaced

NDT as the debate format of choice for most debaters.

Patterson and Zarefsky (1983) wrote that NDT

resolutions have four characteristics: First, is

resolutions have policy implications (p.90), Secondly, is

resolutions are worded broadly (p.91), Thirdly,

resolutions propose actions that are unlikely to be taken

soon (p.91), fourth and finally, the meaning should not be

self-evident (p.92). The characteristics were designed to

increase argumentation by debate teams on both sides of the

debate resolution. In NDT, or policy debate, the

affirmative team has traditionally had three burdens to

prove: (1) that a significant problem exists, (2) that the

current system cannot solve the problem because of an

inherent barrier, and (3) that the affirmative's solution of

the problem will be efficacious or solvent. These

requirements have been known as the "stock issues" in a

debate. Just as the affirmative has certain traditional







requirements, each of the speeches in a debate has

requirements and duties.


Importance of the Rebuttal Speeches

The speakers all have different duties (See note 1) and

while all the speeches in a debate round are important, the

most important speeches are the rebuttal speeches, for

several reasons. First, the rebuttal speeches are only one-

half as long as the constructive speeches. The debater's

time in the rebuttal speech is thus twice as valuable.

Mistakes in time usage have been critical in debate rounds.

The debater's time usage is most critical for the first

affirmative rebuttal because that speech must respond to

both the second negative constructive and the first negative

rebuttal. The first affirmative rebuttal must deal with two

negative speeches that total three times the affirmative

speaker's time. Any dropped issues are crucial as it is

generally considered that the second affirmative speaker can

not rectify the first affirmative rebuttal's mistakes (See

note 2).

Second, while new arguments can be made in either of

the constructive speeches, arguments can only be extended in

the rebuttal speech. Once a speaker fails to cover an

argument, it is presumed to be lost to that team.

Third, a recency effect might possibly allow a judge to

better remember rebuttal speeches. Fourth, debaters vary in

the effectiveness of their rebuttal strategy. Thomas (1979)

explained that in championship debates, it is fair to say

that more debates are won--or lost--in the








rebuttals than in the constructive speeches. All too many

debaters, who have a good grasp of debate in general, go

into their rounds without a clear picture of what they are

going to do in their rebuttals" (p.288). A key to success

in rebuttals is the ability of a debater to cover the other

team's arguments. Rebuttal speakers typically speak rapidly

in order to respond to their opponent's arguments and extend

their own arguments. A successful rebuttal requires

fluency and, more importantly, a gift of economical

language" (Thomas, 1979, p.294). The importance of language

in the rebuttal speech is highlighted in a study by Rouse

and Thomas (1987) which found that first affirmative

debaters steadily declined in their elaboration and

rhetorical amplification over the years. They concluded

that this strongly suggests that rhetoric and

psychological tactics may have given way to more efficient,

compacted argumentation" (p.109).

Scholars differ in their assessment of deciding which

rebuttal speech is paramount. Rouse and Thomas (1987) wrote

that the first affirmative rebuttal speeches often decide

the outcome of the debate. However, William Southworth

(1984) wrote in his NDT final round ballot that ". as

is so often the case, one need only turn to the last two

rebuttals to isolate what issues) the teams considered

necessary to warrant a ballot" (p.53). Mayer and Meldrum

(1987) concluded that the last rebuttal was the most

valuable speech. Of course, one reason that the last







rebuttal might be the most important is because no other

speaker can subsequently influence the judgess.


Judging in Academic Debate

Since the the time of the early Greek rhetoricians, the

question of "who has won" (Sayer, 1974, p.6) has been the

focus of every forensic debate. Academic debaters entrust

the decision to a third party, the debate judge. The judge

listens to both sides and then decides which team won the

round. Understanding the elements that contribute to

winning academic debates has been the focus of some

research, and will be considered here.

It is the debate judge who always makes the decision as

to which team wins each round, with no ties allowed. David

Thomas' (1987c) explanation of the basic judging paradigms

is important:


If the judge is sincere and conscientious, he
or she will make the decision on the basis of
some model of an ideal debate and of what a
debate team must do in order to deserve to
win the debate. In adapting to a judge,
therefore, it is important for the debater to
realize what the judge's model of the ideal
debate consists of. A debate is like any
other human experience in that the
participants sees it from their own personal
perspective, and they may honestly vary from
each other in describing what happened -- thus
the saying that 'debaters never lose debates,
only decisions.' Like the debaters, the judge
also sees the debate from a personal
perspective. (p.116)



Debaters usually strive to understand their judge's

perspective and adapt their presentation to that

perspective. However, the existence of competing paradigms








has been controversial throughout the history of debate

(Dunne, Mack, & Pruett, 1971). The judging paradigm

controversy was argued in academic circles as early as 1917,

(Dunne, Mack, & Pruett, 1971) with one paradigm arguing that

" the task of the judge, therefore, is to place

himself in the position of one who has no opinions or

knowledge of the subject, other than what has been

presented, and to make the decision which any reasonable and

intelligent person would predicate upon the premises"

(p.201). An opposing paradigm argued that the

decision should always be given to the team which shows

superior attainment" (p.201) or speaking skills.

These competing paradigms are meta-perspectives that

serve the debate judge as a mechanism for decision-making in

debate rounds. Research has demonstrated that there are

five different judging paradigms in current academic debate

(Cross & Matlon,1978). The five are: the hypothesis-testing

model, the chooser of policy systems, the tabula rasa judge,

the stock issues judge, and the evaluator of argument

skills. Others (Hample, 1979; Crable, 1976) would argue

that a sixth, a legal judging paradigm, exists as well. (See

note 3).

Whatever the paradigm, judges possess a great deal of

power in academic debate. Hufford (1965) wrote that the

debate judge is like the Supreme Court, infallible.

There is no appeal (p.120)." The power of the debate judge

has provoked controversy for at least the last fifty years.

Laase (1942) wrote that because of the criticism leveled








against judges, debate should switch to a peer debater

quality rating system. This proposal was never implemented

on a wide scale and criticisms of judges have continued.

The majority of judges in academic debate are usually

debate coaches or speech communication teachers, but other

faculty and laymen judges are also used (Klopf, 1964).

Judges at the National Debate Tournament are typically

debate coaches. Judges who are considered to be "good"

judges, those who are considered to be unbiased and

competent in the conventions of debate, are sought after by

debaters. The good judge's viewpoints are valued because

debaters wish to be able to predict how their judges will

evaluate the debate. "Bad" judges are considered to be

incompetent or biased or both. However, debaters do

typically attempt to adapt to judges, even bad ones.

Unlike trial judges, but similar to appellant judges,

academic judges usually provide a written explanation to the

participating teams after the end of the round. Not

surprisingly, good judges usually write comprehensive

ballots, discussing the reasons for the decision in greater

detail (See footnote 4).

Ideally, all judges would utilize the same standards

for judging debate and making the decision (See footnote 5).

That would allow debaters to expect consistent judging and

they could thus concentrate on the topic and not worry about

idiosyncratic differences in judges. Because of the lack of

uniform standards for debate judging, there has been a

" continuing disagreement within the debate community








in regard to methods of judging debate. It is little wonder

that our debaters are at times dismayed with the decisions

rendered by their judges" (Dunne, Mack, & Pruett, 1971,

p.207). The resulting confusion has helped undermine the

logical and rational processes of debate and has contributed

towards research that has investigated other factors that

would explain success in debate. One of these other factors

is the debater's word choice of emotive words.


Emotive Words


Jakobson (1960), Ogden and Richards (1923), Berger and

Bradac (1982), and Littlejohn (1983) all expressed the

notion that words have an emotive function. Each of these

scholars described the emotive function of words as the

expression of a speaker's feelings. The goal of this

function is explained by Littlejohn (1983) who claimed that

" with emotive discourse the communicator hopes to

elicit similar feelings and attitudes" (p.96).

Jakobson (1960) identified six primary functions of

words -- emotive, referential, poetic, phatic,

metalingual, and conative. Jakobson (1960) identified the

emotive function of words as being the communication of

the attitudes and emotions of the addresser. Jakobson

(1960) further explained that:

(the) emotive or 'expressive' function [of
words], focused on the addresser, aims a
direct expression of the speaker's attitude
toward what he [/she] is speaking about. It
tends to produce an impression of a certain







the term 'emotive', .has proved to be
preferable to 'emotional.' (p.354)


In elaborating Jakobson's definition of the emotive

function of words, Fiske (1982) stated the term

"expressive" is also used to describe this emotive function

(p.37). Fiske goes on to say that the emotive function

serves to communicate characteristics of the speaker such as

emotions, feelings, status and other items that make the

message uniquely related to the speaker (Fiske, p.37).

The present study focuses on the emotive function of

words. That the emotive function of words may be

central to the study of debate, and could be a factor in

persuading the judges to accept the debater's arguments is

indicated by Jakobson (1960) if we analyze language

from the standpoint of the information it carries, we cannot

restrict the notion of information to the cognitive aspect

of language" (p.354).

The major thrust of the present research has been to

examine debaters' use of emotive words in their presentation

of arguments. The emotive function of words also serves to

persuade the listener. Bateson (quoted in Kobayshi, 1988)

stated that through their work great artists

communicated complex emotive messages and provided

perceptual perspectives that pushed or pulled people towards

what they did" (p.352). That is to say, a central factor in

emotive words is that the message conveys the addresser's

emotions and attitudes, and in addition, as indicated by







Bateson, and germane to debate, is the fact that emotive

words attempt to push the listener in a given direction.

Berger and Bradac (1982), in a similar vein, have

indicated that a primary function of expressive or emotive

words is to convey information other than the message

content to the listener. Furthermore, they state that the

emotive function's information is communicated or 'given

off' by word choice (p.53). They wrote that there are three

primary functions of words -- referential,

instrumental, and emotive.

The emotive function of words has been investigated

by a small number of scholars. One reason for the lack of

research in emotive language was offered by Beeman (1988)

who explained that linguists of all breeds seem to

develop cold feet when it comes to discussion of the

expression of emotion in language" (p.9). She suggested

that the problem was the perception of emotive language as

"soft" and "idiosyncratic," as contrasted with rule-governed

structures in linguistics. Irvine (1990) concluded that

while many scholars have ignored the emotive function,

" two of linguistics' most stellar figures, Jakobson

(1960) and Sapir (1921, 1927), thought otherwise: Affect, or

emotion, according to them, was a fundamental dimension of

human life and a factor cross-cutting all levels of

linguistic organization" (p.126).

Some researchers who have investigated emotive

word usage include: Allport and Odbert (1936), Dahl and

Stengel (1978), Bolton (1979), Berger and Bradac (1982),







Clore, Ortony, and Foss (1987), and Irvine (1990). Allport

and Odbert (1936) created a lexicon of trait-names that were

divided into those words that were emotionally neutral,

evaluative, and emotionally active words.

Dahl and Stengel (1978) divided emotive words into

three primary dimensions: (1) Subject-Object or It-Me, (2)

Attraction-Repulsion, and (3) Extensor-Contractor or To-

From. Dahl and Stengel also divided the attraction-repulsion

words into positive and negative sub-types. They utilized 58

judges to classify 371 words and the attraction-repulsion/

positive-negative words were significantly agreed upon by

the judges. Bolton (1979) wrote that "emotions are the

key to vital communication" (p.92) and offered a short list

of "feeling words." Bolton (1979) wrote that speakers should

use feeling words that match the listener's experience.

Bolton provided an example of a couple who failed to do so.

They are looking at the Grand Canyon for the first time and

after a long silence one partner says to the other, "This is

magnificent. It is sublime" (p.92). The other responds;

"You think it's pretty" (p.92).

Berger and Bradac (1982) wrote that language could be

classified into subtypes such as "powerful and powerless"

language (p.60) or into "familiarity and goodness" (p.57).

Familiarity was related to a concept of a lexical

"goodness," (p.57) in which predictable or familiar words

were rated positively. They also discussed "linguistic

charms," in which lexical items typically produce

very positive evaluations" (p.57). The reverse of the good







word list contained words that denote excretory and

sexual processes, profanation of scared objects, and death

or decay" (p.57). The language associated with powerless

speech contained hedges, intensifiers and tag questions.

Words could be categorized as being either positive or

negative.

Clore, Ortony and Foss (1987) created an emotive word

list, which they called an affective lexicon. They defined

" affective to refer to the positive or negative

evaluation, or valence, inherent in the meaning of a term"

(p.751). They also found that 234 English words were

empirically classified as having affective conditions. A

subject pool of 435 undergraduate students was used to rate

words and multiple contexts were used. One conclusion of

this study also found that these words were related to

psychological conditions of the speaker.

Irvine (1990) wrote that a growing body of research has

supported the importance of emotive words. Irvine

investigated the Wolof population in the African nation of

Senegal and found distinctive differences in the use of

emotive speech.

The research on the emotive function of communication

supports the notion that emotive words are inherent in

persuasive language. Research also supports the notion that

emotive words can be divided into subtypes, such as the

division into positive and negative emotive words. Clore,

Ortony and Foss (1987) wrote that a positive or negative

evaluation was "inherent" (p.751) to the concept.







Considering the inherent nature of emotive words, it is

interesting to note that there has been no investigation of

the use of emotive words in academic debate. This study

will investigate the use of emotive words in academic debate.

While the influence of emotive words in academic debate has

not been investigated, there is a body of research in

debate, some of which is relevant to this study.


Quantitative Research in Debate


Anderson (1966) reported that the status of an academic

field can be determined by the nature of the research and

that the future of a field can be predicted from the field's

research trends; unfortunately, quantitative research in

academic debate is sparse. Vasilius and DeStephen (1979)

wrote that there are many empirical studies

relating to communication strategies, but fewer on debate

strategies and even fewer on style in debate" (p.197). Some

existing research has even been described by Thompson (1966)

as ". banal and provincial" (p.109) (See note 6).

In particular, there has been little research

investigating language use by debaters. The existing

quantitative research that has examined the relationship of

other variables to debate has been generally inconsistent

(See note 7). Of the research which has a direct bearing on

the subject of this study, the studies (Hayes & McAdoo,

1972; Hill, 1973b; Rosen, Dean, & Willis, 1978) that

investigated the influence of gender are important, and the







study of word use by Rouse and Thomas (1987) provides a

foundation for the present study.

Three studies have investigated the issue of gender in

debate. One examined the effect of gender and evaluation

(Hayes & McAdoo, 1972); the second also examined the effect

of gender on the debate results (Rosen, Dean & Willis,

1978). Both of these studies found that mixed-gender two

person teams did better than all male or all female debate

teams (See footnote 20). However, a third study by Hill

(1973b) found that all male teams did better than all female

or mixed-gender teams.

Rouse and Thomas (1987) conducted a content analysis of

the transcripts of the first affirmative rebuttal speeches

of the NDT from 1972 until 1984. This is an important study

because they conducted a content analysis of 13 years of

NDT speeches. They examined speaking rates and noted that

three first affirmative rebuttals with the fastest speaking

rates were on losing teams and the average rate of winning

speakers was 270 words per minute, compared to an overall

average of 286. Rate of speech varied from 234 in 1972 to

355 in 1984. Despite this finding, the Rouse and Thomas

study did not conclude that there was a direct correlation

between a rapid rate and losing.

Rouse and Thomas also wrote that several trends were

noticed. First, they reported an increasing trend of

debaters offering "roadmaps" or signposting the material

that their speech would cover. Second, they found a

consistent decline in elaboration and rhetorical







amplification. They cited two rebuttal responses from their

transcripts, both in which a single argument is advanced.

The earlier example, from 1972, contained one hundred words

exactly, while a 1979 example contained only seven words.

This "streamlining" they saw as a natural response to the

requirements of a successful first affirmative rebuttal,

i.e., fluency, and, more importantly a gift of

economical language" (p.101). As they considered the

outcome of a debate to be frequently determined by this

rebuttal, it may be concluded that language use is

crucial to decisions (See footnote 8).

That study, however, had some limitations. They made

no analysis of negative speeches for comparison. They

examined the first affirmative rebuttal speech. They did

not conduct any statistical analysis because the study was

"descriptive" (p.110). The study did not test for changes

over a time dimension even though "there have been some

notable differences observed over the time span, such as the

speeding up of speakers' rates" (p.110). The study

ultimately concluded that other factors besides the speaking

technique affect the winning or losing of a debate round.

Other quantitative research in academic debate that is

relevant to the current study has investigated judging

in academic debate. This research has focused on

examining judging practices, alternative methods of judging,

and evaluations of the quality of judging. Some of these

studies include an investigation of the judging of

persuasive speech (Benson & Friedley, 1982), a study of







demographic characteristics influencing debate judging

(Hill, 1973b), and three studies that conducted an analysis

of judges' criteria that included the debater's language

(Williams & Webb, 1964; Williams, Webb, & Clark, 1966;

Williams, Clark, & Wood, 1966). Benson and Friedley

(1982) investigated the judging of persuasive speaking and

concluded that little is known about how judges arrive at

their decisions. They noted that a major purpose of their

study would be to stimulate quantitative research, stating

that only with an established body of empirical

research to describe and evaluate its contributions" will

forensics be able to withstand budgetary pressures (p.12)

(See note 9).

One empirical investigation of non-debater variables on

the judges' decisions was conducted by Hill. Hill's (1973b)

dissertation investigated the influence of variables not

related to the debater's ability on the judge's decisions.

He investigated the influence of the topic, speaker

position, sex, prestige of the school, and proximity. The

research concluded that both the topic and the speaker

position had no significant influence on the debate outcome.

Hill also reported that the geographical proximity of the

judge's school to the debater's schools did not influence

the debates' outcome. He did find that gender had an effect

with all male debate teams having a greater chance at

winning than all female or mixed debate teams. The study

established that prestige also had an effect, with junior

college debaters being most negatively affected. Hill







defined high prestige teams as those schools that had

previously participated in the National Debate Tournament.

Williams and Webb (1964) also investigated factors in

debate evaluation. The study asked judges to respond to 37

bipolar descriptive scales of debaters, and used factor

analysis to determine that the dominant factor was

"argument." This factor included scales for supporting

material, concreteness, logic, relevance of evidence,

analysis, reasoning, pertinency, refutation, persuasiveness,

use of motive appeals, and organization. A second factor

was "vocal-correctness" and it included articulation,

pronunciation, grammar and vocal quality. The third factor

was "overall-delivery" and included scales for eye contact,

rate, spontaneity, interestingness, facial expression and

intelligibility. The fourth was "apparent-character" and

included scales for courtesy, sportsmanship and ethics. They

reported that the first factor accounted for 36% of the

total variance. The amount of the accounted variance for

the other factors was not offered. They also reported in an

endnote that terms in addition to those reported in

the results to this study included: Word choice, Use

of figurative language" (p.128). Williams and Webb did not

explain why these items were ignored.

Two follow-up studies (Williams, Webb, & Clark,1966;

Williams, Clark & Wood,1966) supported the initial results

of Williams' and Webb's (1964) study. Williams, Webb, and

Clark (1966) conducted a follow-up study to the Williams and

Webb (1964) study. The results were similar to the earlier







study. Argument was again the dominant factor. They

concluded that "relatively few dimensions of evaluation

underlie judges' use of a relatively large number of rating

scales" (p.20). While word choice and the use of figurative

language were not discussed in the study, the "vocabulary"

(p.16) scale did not have a high loading on any of the

factors in the study.

Williams, Clark and Wood (1966) conducted yet another

follow-up study and again obtained similar results. The

argument dimension was found to predict the overall rating

as well as the other factors combined (p.100). They

concluded that judges offer overall assessments of

argumentative skill rather than judging on individual

factors such as logic or evidence.

Overall, quantitative studies in debate which focus on

judges has produced some consistent results. (1) Judges

currently use five different paradigms in judging debate:

hypothesis tester, policy evaluator, tabula rasa critic,

stock issues judge, and evaluator of argument skills. (2)

Studies of judging criteria report that the factor of

argument is important to the judge's decision in the round.

Other studies reported that the need issue was paramount.

(3) Judges and debaters usually do not view the round in the

same fashion. (4) At least one study suggested that

debaters' word choices do influence the debate decision.

(5) Bias of the judges regarding the topic was not found,

but some geographical bias was found. (6) Debaters become







confused because of inconsistency in judging. (7) Debaters

were able to adapt to judges over time.

The question of what influenced a judge to vote for a

given team has been of interest for quite some time and has

still not been resolved. Studies that examined the use of

logical arguments found that logic was not solely related to

success in debate. Studies also demonstrated that evidence

was not solely related to debate success either. Despite

debate's continued reliance on evidence and logical appeal

as the basis of success in debate, research has failed to

support either evidence or logic as being significantly

related to success in debate. Some of the studies that

investigated academic debate in the area of logical

processes included research (Allen & Kellermann, 1988) that

found that disadvantages were not persuasive in real world

terms. Yet, disadvantages were also seen as being decisive

in deciding the round (Allen, 1987). Similarly, rate of

speech, use of jargon and the amount of evidence was found

to be unrelated to academic debate success (Vasilius &

DeStephen, 1979). These questions will require equally an

empirical, quantitative effort to understand what transpires

in the debate round.


Word Choice and Debate


In examining the existing research specific to academic

debate, only a minor amount examining the use of language

exists; nevertheless, it has provided interesting results.

Four published articles (Giffin,1959; Williams and Webb,







1964; Williams, Webb, & Clark, 1966; and Rouse & Thomas,

1987) have dealt with the use of language in debate.

Giffin (1959) reported that the debater's use of

language accounted for only 5.29% of the factors involved in

the debate judge's decision; however, this author believes

there were several methodological problems with Giffin's

study. First, he used the debate ballot as the basis of the

judging criteria. The 34 judges were instructed to "make

your decision first, giving appropriate ratings and

rankings; then we would like to have you indicate those

criteria which you used in making your decision and to what

extent each criterion was considered" (p.70). Giffin failed

to explicitly state that the criteria were those from the

debate ballot, but an examination of his methodology

revealed that the criteria were from the ballot. The ballot

constrained the judges to use only the criteria listed.

Second, if language effects are subtle, there is no reason

that the judges would self report those effects. Third, the

study asked the judges to report the criteria that were

used. Giffin only used the judge's self-report. The study

did not examine any other variables. Fourth, Giffin

explained that language was defined as only the "phrasing of

concepts clearly and concisely" (p.70).

Giffin's failure to report any significant language

effect might have resticted further research if other

scholars concluded that Giffin had demonstrated a lack of

importance of language effects in debate. While Williams

and Webb (1964) included the variables of word choice and







the use of figurative language in their study, the results

of the two variables were never reported. Williams, Webb,

and Clark (1966) reported that the debater's vocabulary

failed to load highly on any of the study's evaluative

factors.

Rouse and Thomas (1987) reported differences in the

first affirmative rebuttal speakers' use of evidence and in

their conformity with a supposedly ideal argument pattern.

The study also discovered differences in the debaters' use

of "road maps" (p.108) and word economy, but only examined

the first affirmative rebuttals in each of thirteen debates

and did not use any statistical analysis.

Debate textbooks have paid little attention to the

importance of language in debate. Only five out of twelve

debate textbooks examined (Wood & Goodnight, 1989; Fryar,

Thomas and Goodnight, 1989; Sanders, 1983; Freeley, 1976;

and Patterson & Zarefsky, 1983) were found to discuss

the issue of language effects and debate.

Wood and Goodnight (1989) only suggested that debaters

use language that would be most appropriate for their judge

(p.280). Fryar, Thomas and Goodnight (1989) suggested that

speaking clearly was much more important than speaking

quickly. They also suggested practicing speeches to cut

down the number of nonessential arguments. Practice would

also allow the debaters to cut down on the nonessential

words.

Sanders (1983) wrote that the wording of issues was

" .. extremely important and yet, something that







is more often than not overlooked" (p.57). For example, he

discussed the statement that, ... the states need more

revenue" (p.57) and concluded that it had no impact. The

effective alternative was the states can no longer

finance a viable public education system" (p.57). He also

noted that the wording had to be both succinct and

persuasive. Sanders also wrote that debaters should seek

to avoid ambiguous wording of issues.

Freeley (1976) discussed logical, ethical and emotional

appeals without discussing the differences in the language

that would make up such appeals. Freeley (1976) also

stressed the importance of defining terms, stating that the

" advocates must carefully consider all possible

definitions of all terms" (p.43). An extension of this

argument would be for the advocates to equally consider

their choice of language.

Patterson and Zarefsky (1983) wrote that debaters have

choices in language use, for example, the use of

connotation. Patterson and Zarefsky offered the example of

the verb "said," which is neutral. A positive connotation

can be created by replacing "said" with "exclaimed," or it

can be given a negative connotation by replacing "said" with

"confessed." Patterson and Zarefsky noted that in

constructing a case, therefore, advocates should consider

whether positive, negative, or neutral connotations are the

most desirable and should word their arguments accordingly"

(p.67). They also noted that debaters have a choice in the

precision of the language, and the intensity of the







language. Debaters have a choice of being precise or vague,

and debaters can use irony or hyperbole, thus ranging from

stating the reverse to large overstatement. However,

Patterson and Zarefsky noted that either tactic was likely

to backfire (p.67).

The remaining choice that Patterson and Zarefsky

discussed was style. Debaters could use straightforward

speech but they could also utilize analogies, similes, and

metaphors. Patterson and Zarefsky's discussion of language

concluded that:

the basic idea underlying all presentational
choices is that language is not a neutral
instrument, a vehicle in which the contents
of argument are conveyed. Rather, language is
an inseparable part of the argument itself
and one that exerts great influence on how
listeners perceive and react to the case. (p.68)


Word Choice and Legal Research

In terms of the use of language, the associated field

of law provides some interesting research. For example,

research (O'Barr & Conley, 1976; Parkinson & Parkinson,

1979; Conley, 1979; Parkinson, 1981; and Parkinson, Geisler

& Pelias, 1983) investigating language choices and language

effects in legal situations has found a relationship

between language choices and success in trials. Some of

these findings indicate that seemingly insignificant

differences in language can be important. O'Barr and Conley

(1976) noted that:

some differences in courtroom language may be
so subtle as to defy precise description by
all but those trained in linguistic analysis
.New research on language used in trial
courtrooms reveals that the subliminal







messages communicated by seemingly minor
differences in phraseology, tempo, length of
answers and the like may be far more
important than even the most perceptive
lawyers have realized. (p.8)

O'Barr and Conley classified language along a "power

language continuum" in which a powerless speaker's language

contained a high frequency of hedges, such as "Perhaps .

or "I think ," repetition, intensifiers, such as "very

close friends" instead of simply "friends," and a greater

use of direct quotations (p.9). The article did not fully

identify the characteristics of the powerful speaker. The

study found that speakers speaking in the powerless mode

were found to be significantly viewed as having less

competence, believability, intelligence, assertiveness and

likeability.

Another study that investigated language in the

courtroom (Conley, 1979) also reported a statistically

significant difference between powerful and powerless

speakers. This article reported the finding that speakers

who utilized "hypercorrect speech" were found to be viewed

as less competent, less convincing, less intelligent and

less qualified than a speaker who used a more natural form

of language.

Additional research on speech tactics and success in

trials has been conducted by the Duke University Law and

Language Project. One resulting article (Parkinson &

Parkinson, 1979) reported that a computerized language

analysis of actual jury trials found that successful defense

attorneys used fewer adverbs. They also used more vague or








abstract language. The study also examined the speech of

the prosecution and the defendants. Prosecution attorneys

who won their cases were found to speak longer, make more

indicative statements and ask more questions referring

directly to a witness. Losing attorneys were found to have

used more conditional language. These attorneys also used

more careful and polite language. In contrast, the

defendants who were successful were those who used more

courteous language. The successful defendants also made

fewer references to themselves and they made more

grammatically complete sentences.

Another study (Parkinson, 1981) from the Duke

University Law and Language Project reported that a number

of different speech behaviors were co-occurring with either

success or failure in courtroom trials. The study focused

on "message style, not message meaning" (p.31). The speech

behaviors differed for the speaker. What was successful for

some speakers were not successful for other speakers.

Prosecution attorneys were successful when they were

verbose, verbally assertive, and when they referred

specifically to the witness and used the pronoun "you."

Defense attorneys were successful when using ambiguous

or abstract language. They also used fewer adverbs, more

legal jargon, and fewer afferent words relating to the five

senses. Successful defendants were found to use more polite

and courteous language. They also used more demonstrative

language such as "this" or "that." The successful

defendants used grammatically complete sentences.







The same speech behaviors that were successful for the

defendants were found to be associated with failure for the

attorneys. Both demonstrative language and grammatically

complete sentences were associated with failure in defense

attorneys. Politeness and and hypercorrect grammar were

also present when prosecution attorneys failed. In addition

to the speech behaviors that were successful for defendants

but unsuccessful for the attorneys, Parkinson (1981) found

that prosecution attorneys who used conditional language

such as "might" or "could" failed. Defense attorneys who

used concrete words also failed. That finding was not

surprising since successful defense attorneys used abstract

language. Parkinson (1981) found statistically

significant differences associated with the word choice

and success/failure in trials. The study used a stepwise

discriminant analysis and concluded that the method .

was able to predict trial outcome from language behavior

variables with an overall success rate of 77.08%" (p.27).

Another study (Parkinson, Geisler, & Pelias, 1983) that

also investigated language choices and trial success

reported results similar to the Parkinson (1981) study. One

difference was discovered in the use of nouns without

physical referents such as "honor" and nouns with physical

referents such as "automobile." Successful defendants and

their attorneys used more nouns with physical referents

while successful attorneys for the plaintiff used more nouns

without physical referents. Plaintiffs were found to be

successful when they were verbose and when they used more







adverbs and adjectives. They were unsuccessful when they

used more nouns without physical referents. The results

were found to be significant through the use of multiple

discriminant analysis. The study found that "the difference

between a successful and unsuccessful courtroom performance

may be only a few words per thousand" (p.21) and concluded

that "there are significant, though subtle, speech

characteristics which co-occur with trial success" (p.21).

Thus, while a few studies have investigated word use in

debate rounds, the results have been generally ignored. One

exception was the study by Rouse and Thomas (1987) that did

report some language differences. Research in legal studies

has demonstrated clear-cut and strong effects of certain

types of language on the winning and losing of trials. The

fact that minute differences in word use were significantly

related to success in courtroom trials should provide a

mandate for investigating word use in debates. Both

hinge upon the use of words to persuade a judge (or jury)

to accept only one of two opposing sides.

The present study examined the emotive words used

by all four debaters in their rebuttal speeches. The

purpose of this study is to conduct an analysis of the

transcripts from nineteen of the final rounds of the NDT and

determine the existence of and relative usage of emotive

words and the relationship, if any, to five variables

including result, side of resolution, speaker position, year

of debate, and gender composition of the team.








Empirical Focus

The Dependent Variable

Transcripts of nineteen years of intercollegiate

debate from the National Debate Tournament were converted

into word-processed form and analyzed by a software called

"wordscan" that can count and extract words from sub-

dictionaries. The transcripts were compared to a list of

negative and positive emotive words that were drawn from

Gottschalk and Gleser's content analysis scale and from

Osgood's semantic differential scale. These scales purport

to reflect the attitudes and emotions of the respondents.

Exemplars of emotive words are presented in tables 3-5

through 3-80.

The Gottschalk-Gleser Content Analysis Scale

The original scales were first developed by Gottschalk

and Gleser in 1969. The scales have been widely used by

scholars in a range of different fields (Gottschalk, 1986).

The content analysis scales have been used to measure the

affective states in African-American children (Uliana,

1979). The scales have also has been used to measure

hostility (Gift, Cole, & Wynne, 1986), depression

(Gottschalk & Hoigaard, 1986), alienation of drug addicts

(Viney, Westbrook, & Preston, 1986), and the speech of

mentally ill patients (Lebovits & Holland, 1986). Further,

the scales have been used to conduct textual analysis

(Nimicks, 1985).

One problem with the use of the scale is that it takes

considerable time to score the results, but Gottschalk








(1986) noted that a computer should solve that problem. A

number of scholars have used modified versions of the scales

(Gottschalk, 1986).

The scale has been found to be both reliable and valid

(Gottschalk, Eckard, & Feldman, 1979). Viney (1986)

surveyed the research on Gottschalk and Gleser's content

analysis scale and reported that the range of reported

interjudge reliability for the hostility scale was from 0.76

to 0.98 and that [the scales] are also consistently

high (p.60). Viney (1986) also reported that since

construct validity best reflects whether the scale is

successful, that the scale seems likely to have

inherent content validity" (p.60) because the content is

directly derived from the subject's communication. Viney

(1986) also surveyed the literature of twenty years dealing

with the scale's validity and with the hostility scale, and

found that it was independent of age, educational level and

sex. The scale was found to be significantly correlated

with a number of emotive states (p.65) and was correlated

with other measures of the same construct .

(such as) self-reports and observations of behaviors "

(p.63).

Gottschalk (1979) explained that the content

analysis scale had a set of construct-validation

studies [that] had to be carried out to ascertain exactly

what this content analysis procedure was measuring ."

(p.548). He noted that the procedure could be applied

to a variety of contexts such as literature, public








speeches, and any other type of language material" (p.550)

as long as the samples are based on a response to standard

instructions and with equal temporal units. Debate could be

easily analyzed through the procedure since the debaters

operate within a framework of standard instructions and have

identical temporal units. The scales have also been

demonstrated to be valid and reliable for non-American

subject groups including Germans (Schofer, Koch, & Balck,

1979; Koch, 1986), demonstrating a cross-cultural utility

for the instrument. Gottschalk (1979) wrote that the use of

the scale in other languages was evidence of the

validity and universality of the procedure (p.39).

Gottschalk (1979) wrote that content analysis was the

best system available for measuring emotive words:

Reliable and valid measurement of affects, emotions,
and moods have posed a problem for psychiatric and
psychophysiological research as the demand has grown
for more sensitive, precise, and objective assessment
methods than the method of clinical impressionistic
evaluation. There are three major methods in current
use for assessing these psychological variables: self-
report scales, behavioral rating scales, and the
content analysis of verbal behavior [which] can
avoid most of the shortcomings of the self-report and
observer rating methods (p.541)

The Semantic Differential Scale

Charles Osgood, George Suci, and Percy Tannenbaum

(1957) created the semantic differential scale as an

instrument to describe major dimensions along with varying

judgments and meanings. The scale was composed of word

pairs such as "good-bad," "kind-cruel," "peaceful-

ferocious" and "happy-sad." Subjects were asked to select

one word from each pair. Staats and Staats (1969) pointed








out that the word's meaning involves a psychological process

that is different from word association processes. The

scale has been widely used and it has also been modified.

Weinreich (1969) expanded the scale with the use of Roget's

Thesaurus (p.118). Cliff (1969) utilized adverbs as

multipliers.

The scale has been reported as reliable (Solarz, 1969)

and valid across cultures (Tanaka, Oyama & Osgood, 1969).

Osgood, Suci, and Tannenbaum (1957) reported a .85 test-

retest correlation coefficient. They also reported that the

scale had face validity and that ". throughout our work

with the semantic differential we have found no reason to

question the validity of the instrument on the basis of its

correspondence with the results to be expected from common

sense (p.141).

Validity and Reliability

The reliability of the current study is easily answered

by the computerized emotive word list(s) and the

computerized Wordscan software that counts the presence of

the words. For example, "opportunity" is an emotive word.

"Opportunity" was used five times by the first affirmative

side speaker in 1979. Wordscan would always count five uses

of "opportunity" when checking that speech. Reliability

becomes perfect, as Lewis-Beck in Weber (1990) explained

" once the text is computerized, say with an optical

scanner, it is relatively easy to make a classification from

more than one dictionary. Moreover, with computers, the

coding rules are necessarily made explicit, allowing for








perfect 'intercoder reliability'" (p.5). The present study

did use an optical scanner, an emotive word dictionary, and

software that allowed the presence of emotive words to be

consistently counted. The reliability was 1.00.

The validity is based on the original validities of the

content analysis scale and the semantic differential scale.

Since both of those scales had established validities, the

use of the emotive word list should be valid as well, based

on the notion of concurrent validy. Concurrent validity

uses an established validity of an existing measure to

provide validity for a new measure (Walizer & Wienir, 1978).


Research Questions


The present study examined the use of emotive

language in academic debate. The central research question

was:

Research Question One

Do winning and losing debate teams differ in their

use of emotive words during rebuttal speeches in a debate

round?

Related research questions are:

Research Question Two

Do affirmative and negative debaters differ in

their use of emotive words in rebuttal speeches?

Research Question Three

Do the first and second rebuttal speakers for each

debate team differ in their use of emotive words?




37


Research Question Four

Does the quantity of emotive words used by debaters

vary over a period of years?

Research Question Five

Does the quantity of emotive words used differ between

mixed-gender debate teams and single-gender debate teams?

(See note 10)







Notes




1 The first affirmative speaker presents the
affirmative case. The first negative speaker usually
presents the negative philosophy and often directly clashes
with the affirmative case structure. The second affirmative
speaker responds to the negative attacks and rebuilds the
affirmative case. The second negative speaker usually offers
the off-case attacks, so-called because the arguments do not
directly clash with the affirmative. Disadvantages are
examples of off-case arguments. The first negative rebuttal
responds to the second affirmative responses. The first
affirmative rebuttal must respond to both second negative
speaker and the first negative rebuttal speaker. This speech
is crucial because the rebuttal speaker has five minutes to
respond to fifteen minutes of attacks. The second negative
rebuttal speaker responds to the first affirmative rebuttal
speaker and offers the final rational for a negative ballot.
The second affirmative speaker responds to the second
negative speaker and offers the final rationale for an
affirmative ballot.

2 Scott Deatherage (1985) explained in his ballot
in the final round of the NDT that he voted negative as a
result of a brilliantly conceived"(p.55) second
affirmative rebuttal that fell just short of
covering for earlier time allocation mistakes" (p.55).

3 The hypothesis-testing model is centered on the
belief that the debate process is similar to the testing of
a hypothesis in a laboratory test. The debate is a test of
the resolution alone, which would allow the negative to
"defend anything or everything that is non-propositional"
(Cross & Matlon, 1978, p.111).
The chooser of policy systems is a debate paradigm in
which the debate teams are similar to a legislative body
that evaluates competing systems which solve the same basic
problems. The policy system paradigm has been very common in
debate. The affirmative team usually offers a problem and
solution while the negative usually would attempt to
demonstrate that the current existing system could solve the
problem and that the affirmative solution would not work
and/or that it bring about worst harms.
The evaluator of argument skills paradigm would have the
judge decide the round on the basis of which team utilized
better analysis, evidence, reasoning, organization,
refutation and delivery.
Tabula rasa, Latin for blank slate, assumes that the
judges who use this paradigm are completely open to any
theory or approaches in the debate round.
The stock issues paradigm dates from ancient times
(Cross & Matlon, 1978, p.112). This paradigm contains four
elements: (1) The significance of the issuess; (2) The







inherency of the issue; (3) The issue of solvency or
solution; (4) The issue of disadvantage. The four issues are
approximately the issues of ill, blame, cure, and cost
(Cross & Matlon, p.112).
The legal paradigm assumes that the judge would evaluate
the debate by the use of the legal principles such as
precedent. However, a study (Cross & Matlon,1978) that
examined the judging philosophies at the most important
debate tournament, the National Debate Tournament (NDT), did
not find any judges who used the legal model.

4 The written explanations of the judge are called
ballots and are usually on standard forms. Most debate
tournaments and therefore most debate judges use ballots
obtained from the American Forensic Association called the
Form C ballots.

5 However, there exist different standards that the
debaters must adapt to. The nature or personality of the
judge is also critical to debate. The judge must "decide on
the substantive merits of the question in dispute"
(Patterson & Zarefsky, 1983, p.105). The judge must be
unbiased; he or she will need to decide the question based
on their prior background and beliefs as well as the
specifics of the debate itself. This is also influenced by
the judging paradigm in the debate round. The tabula rasa
(blank slate) judge tries to remove the prior background and
beliefs and only judge the round on what actually has
transpired in the round.
6 The first explanation for the sparse research is
that the existing behavioral or quantitative research is
fragmented. One survey (Anderson, 1974) found that almost
30 percent of all existing bibliographical entries on
behavioral research in debate were unavailable at the
nation's third largest library at the University of
Illinois. In 1991, The Chronicle of Higher Education
Almanac ranked the University of Illinois at Urbana-
Champaign as the fifth largest library in the United States
and Canada.This unavailability was the result of lack of
indexing for both important journals and the forensic
honoraries. As Anderson points out, many of the
early empirical studies and most of the current ones are
published in these journals" (Anderson, 1974, p.148). A
related problem is that different operational definitions
are often utilized for similar phenomena, so that
comparisons or generalizations across different studies is
difficult.
The second explanation for the lack of research is that
the field is not oriented towards its own meta-research.
There appears to be little interest within the field in
conducting research involving debate or debate processes.
One suggestion (Cox, 1976) is that we are so
intensely involved in participation that seldom do we study
and professionally evaluate the discipline for the benefit
of others"(p. 57). Debate coaches might respond that their







preparation of their teams for the rigors of the debate
circuit would consume their time. Klopf and Rives (1965)
wrote that 85 percent of the debate coaches felt that their
professional and academic advancement was restricted .
by directing forensics activities. The additional hours
required to coach, administer, and attend tournaments
prevents them from engaging in activities such as research
and publishing" (p.36). One result of the restriction is
that frequently debate coaches or the directors of forensic
programs will leave those positions to concentrate on non-
debate research or teaching. Often, the debate job is
considered unimportant. Kully (1972) explained that .
at some institutions responsibility for the debate squad is
considered a menial task for a junior faculty member or
graduate assistant" (p.195). Academics in the field are
confronted with two almost mutually exclusive choices,
research and publications or debate tournament success. As
Andersen (1974) explained, those conducting
behavioral research in speech communication and persuasion
are moving in one direction while those active in forensics
and competitive debate are moving in a different if not an
opposite direction" (p.149.) Research thus needs to focus
more on bringing the two directions together.
The third explanation for the lack of research is that
while the field of speech communication has undergone
significant changes, debaters have remained much the same.
When departments of speech communication were heavily
centered on rhetoric, debate was more central to the
departments. But as departments became more behavioral and
experimental, debate became less important. As debate
becomes less important, the need to conduct research is
reduced.
The direction of those active in competitive debate is
readily shown by the field's foremost academic journal. The
Journal of The American Forensic Association was founded in
1963 and originally focused on both debate and the
individual events in forensics, such as persuasive or
impromptu speaking. The journal, usually known by its
initials, JAFA, deals with the whole range of interest in
forensics, including descriptive articles, theory articles
and some quantitative or behavioral research articles.
However, debate theory articles came to dominate the journal
at the expense of research. After a questionnaire surveyed
the membership of the American Forensics Association, the
journal was renamed in 1988 as Argumentation and
Advocacy:The Journal of The American Forensic Association.
The professional associations have ignored research as
well. The American Forensic Association waited 20 years
before creating a committee on research (Walwik, 1969). The
focus of the professionals in the debate field is largely on
theoretical articles that further develop the practices in
the debate round, not on research that might help explain
the results of the debate round.

7 The existing quantitative research in academic
debate is best categorized in three areas (Anderson, 1974) :







(1) logical processes; (2) personality or characteristics of
debate participants; and (3) judging procedures. Each of
the three areas will next be discussed.
Logical processes research examinines the relationship
of the addresser to the addressee. Debate has been based on
the belief that it is a logical process. The research in
the area of logical processes has examined the effects of
evidence, investigations of logical and/or emotional
appeals, research on reasoning, one-sided versus two-sided
appeals, and receiver variables. Stanley Rives (Dresser,
1964) was quoted as explaining in his address to the Midwest
Forensic Association that, debate is primarily an
exercise in reasoning and evidence The debater, coach
or judge who concerns himself with academic debate agrees to
concern himself primarily with reasoning and evidence, to
accept these as the basis of an intelligent
decision"(p.101). Some of the research that would be
included under logical processes includes debater's use of
evidence (Larson & Giffin, 1964), the use of evidence in
ten championship debates (Dresser, 1964), the use of
evidence in one championship debate (Newman & Sanders,
1965), the dilemma of ethics in the use of evidence (Rieke &
Smith, 1968), the impact of evidence on decision making
(Dresser, 1966), the rhetoric of evidence (Gregg, 1967), the
relation of logic to argumentation (Petrie, 1969), the logic
of evidence (Kellermann, 1980), the effectiveness of NDT
final round disadvantages (Allen & Kellermann, 1988), and
the effect of various time limits on the quality or
effectiveness of rebuttals (Mayer & Meldrum, 1987).
Three different studies (Larson & Giffin, 1962;
Dresser, 1964; Newman & Sanders, 1965) examined the evidence
used in debates and have found a large proportion of such
evidence is unverifiable, misrepresented or fabricated.
Larson and Giffin (1964) examined the evidence used
in four randomly selected debates from the prestigious Heart
of America Tournament at the University of Kansas in 1962.
Larson and Giffin found that only half the evidence was
validly represented. They reported that forty-two percent
of the evidence were unverifiable; five percent was out of
context; only three percent was manufactured or quoted from
nonexistent sources.
Dresser (1964) examined the use of evidence in ten
championship debates. He examined the final rounds of the
National Debate Tournament for 1950, 1952, and 1955 through
1962. Dresser investigated the evidence for clarity of
documentation, recency of evidence, competency of sources
and the closeness of the evidence to the debater's claim.
The study found that most debaters used "evidence of
opinion" (p.106) and that the debaters were clear about the
qualifications of their sources but much less so about the
recency of the sources.
Another study (Newman & Sanders, 1965) investigated the
use of evidence in a single final round of the National
Debate Tournament. They found that of 71 pieces of evidence
read in the round, three were fabrications, 23 were
misrepresentations, and 6 were unverified. Such findings







were troublesome to the authors and created a demanded for
improvement in the use of evidence. Fabricated or
misrepresentated evidence could easily change the results of
a debate round. The importance of accurate evidence has been
well known for a long time. Carney Smith (1937) wrote in the
Quarterly Journal of Speech that "a single word changed or
omitted in a quotation may go unnoticed by the other team,
yet may alter the meaning of the entire statement" (pg.83-
84)
A study that focused on the use of evidence and ethics
was conducted by Rieke and Smith (1968). They wrote that the
competitive nature of academic debate was both its strength
and weakness because the debater is expected to be both a
fervent champion of their cause and ". a judicious
participant in rational decision-making" (p.228). Rieke and
Smith argued that these two duties are contradictory. They
suggested that it is difficult for debaters to advocate
their position with maximum energy and to also be .
committed to rational methods rather than victory at all
costs" (p. 229). One result of the contradictory duties was
an increase in unethical or questionably ethical practices.
This study conducted a survey of debaters regarding ethical
practices. The authors concluded that an
extraordinary contradiction exists between what is done and
what debaters perceive as being done" (p.227). They wrote
that academic debate had failed because of the unethical
practices. The impact of the results were argued to go
beyond academic debate because so many future lawyers and
other professionals are student debaters.
Several studies (Dresser, 1966; Petrie, 1969;
Kellermann, 1980; Allen & Kellermann, 1988) have examined
the relationship of evidence useage in debate to logic and
decision making in debate. The consensus of the studies is
that evidence does not influence the decision process in
debate and the use evidence is not logical. Debate
arguments have been found to be not logical as well.
Arguments such as disadvantages that have been successful in
debate were found to be not persuasive in real world terms
Dresser (1966) examined the impact of evidence on
decision making. This article reported that the common sense
view that a speaker who uses facts to support assertions
would be more likely to influence his audience was not true.
Research indicated that the quality of evidence, amount of
evidence or the identification of evidence simply
did not make any significant difference to their listeners"
(p.40). Dresser did not suggest what elements would make any
significant difference to the listeners.
Gregg (1967) amplified this argument when Gregg examined
the rhetoric of evidence and concluded that logical pattern
of argumentation fails to consider the personalities of the
interactants, it neglects the subject matter's values, and
it assumes that the judge would make a careful examination
of the evidence.
Petrie (1969) investigated the relationship of
argumentation to logic. Petrie noted that while no one would
deny that logic and formal argument might sometimes change







belief, the question remains, however, as to the
precise manner of such causal efficacy or logic may
cause belief through external 'accidental' features of
formal argument such as style or impressiveness, regardless
of the formal correctness of the argument" (p.55). Petrie
never resolved either the issue of the mechanism of logic or
the issue of any external features such as style.
Kellermann (1980) noted that the research on evidence
" has produced such inconsistent results that no
coherent theoretical perspective on the usefulness of
evidence can be extracted" (p.159). Kellermann concluded
that the poor results are possibly due to variations in
audience, source and message variables.
Allen and Kellermann (1988) investigated the
effectiveness of four NDT final round negative
disadvantages. This study noted that negative disadvantages,
which are those arguments that the negative suggests would
result if we believed the thesis of the affirmative case.
The judge's belief is the result of the judge's paradigm and
the interaction of the four debaters and their
argumentation. Judges sometimes have found themselves voting
for arguments that they personally do not believe but that
the judge "believes" in the context of the debate round. The
disadvantages were described as having a common
characteristic of being "high impact/low probability",
defined as having severe consequences but a rather low
likelihood. Allen and Kellermann (1988) explained that ". .
[the] reason such arguments have popularity is that judges
vote for them"(p.93). The popularity of these disadvantages
is due to the fact that they often present clear cut choices
in weighing the debate round. Judges are often faced with
both sides in the debate round claiming global destruction
which, for example, might force the judge to compare full
scale nuclear war with a runaway greenhouse effect.
The disadvantages are themselves the subject of controversy.
They were described by Allen (1987) as having a significant
value because they represent "real world" arguments and
debates over real policy decisions. But other scholars
(Hollihan, Riley & Baaske, 1985; Rowland, 1986) suggest that
these arguments fail to reflect real world policy decisions,
and that they have application only in academic debate. Howe
(1981) argued that these types of arguments have no truth
value and that judges should vote against them because they
are bad and not truthful arguments. The Allen and Kellermann
(1988) study presented four disadvantages from NDT final
rounds to 229 undergraduate students. The disadvantages were
taken from the transcripts in JAFA. Three of the four
disadvantages were the deciding issue in those rounds by the
majority of the judges. Allen & Kellermann (1988), noted
that these arguments were selected because they had
been evaluated as good arguments presented by good teams.
These should be high quality arguments and represent the
best of academic debate; in other words, the debate
community had already established that these arguments had
worth"(p.101). However, the study found that while the
disadvantages had some level of argumentative acceptability,







the disadvantages were not judged to be persuasive in real
world terms (p.104) when presented to students.
An investigation of the effect of various time limits
on the quality of rebuttals (Mayer & Meldrum, 1987) found
that the time limit for a rebuttal did not affect the
rebuttal's quality. This study utilized five minute rebuttal
speech debates and compared them with seven minute
rebuttals. Debates were found to not give any more explained
qualitative arguments, rather they gave more unexplained
quanitative arguments. It was also found that the rebuttal
speeches were different in quality from each other. They
found that the last rebuttalist gave the better speech. One
possible reason is that the final speaker has a better, more
holistic view of the debate (p.164). Quality in this study
was measured as the effectiveness of the speech.
Overall, the logical processes research demonstrates
that logic and evidence are only part of the reason for
arguments being accepted by judges. Another explanation
might be message variables. The research also demonstrates
that the dual roles of passionate advocate and rational
decision-maker were contradictory and lead to ethical abuses
that included fabricated evidence. Finally, disadvantages
that debate judges had used to justify their advantage were
not found to be persuasive in real world terms.
Consistently, the research indicates that there are other
factors that influenced decision making besides evidence or
logic. One of these factors is related to the debater's
personality and characteristics of the individual debate
round.
Several research studies in academic debate have
examined the personality and other related variables of
debaters. This area of research has also examined the
interface between personality characteristics and
participation, the descriptive studies of debate rounds, and
comparison of the different activities within forensics
(Anderson, 1974). Not all of the research that is described
under the personality category is explicitly related to
personality.
The range of this research is another example of the
scattered research in debate. Some research involves the
changes that debaters undergo or the difference between
debaters and non-debaters. This research includes a survey
of the participants in the NDT from 1947 to 1980 (Matlon &
Keele, 1984), a study of the differences between debaters
and nondebaters (Stewart & Merchant, 1969), a study on the
image of the debater (King & Phifer, 1968), a study on the
effect of debate students in the bicentennial debates
(Semlak & Shields, 1977), a study (Colbert, 1987) that
examined the effects of debate training on critical thinking
ability, a study of the competitiveness of the debaters
(Wilson, 1979), and a study that investigated debaters'
ability for interpersonal and concept compatibility (Allen,
1963).
Other research hag examined characteristics of the
debaters including the debater's nonverbal decoding
abilities (Barker, 1965; Sayer, 1974). Three studies have







also investigated gender differences (Hayes & McAdoo, 1972;
Hill, 1973b; Rosen, Dean, & Willis, 1978).
Several studies examined behavior or characteristics of
the debaters while in the debate round. These studies
include one which examined the evaluation criteria as
predictors of debate success (Burgoon, 1975), one that
investigated the debater's dimensions of credibility
(Burgoon and Montgomery, 1976), one study that examined the
relative effectiveness of praise and derogation as
persuasion strategies (Burgoon, Wilkinson, & Partridge;
1979), one investigation of the relationship between debate
success and rate, jargon, and evidence (Vasilius &
DeStephen, 1979) and one content analysis of 13 years of
the first affirmative rebuttal speeches from the NDT (Rouse
& Thomas, 1987).
To characterize debaters, a survey of 703 debate
participants in the National Debate Tournament from 1947 to
1980 (Matlon & Keele, 1984) found that the two most common
major fields of study debaters engaged in were political
science, with 24 percent, and speech communication, with 20
percent. The next most common were history at 12 percent,
economics, at ten percent, English, at six percent,
philosophy, at five percent, and business administration, at
four percent. These debaters went on to obtain advanced
degrees with 633 having at least one advanced degree and
209 having two or more advanced degrees of the 703 people
surveyed. Matlon and Keele reported that they found .
that some of the most intellectually gifted students have
participated in the National Debate Tournament" (p.195).
After obtaining the advanced degrees, the debaters were
found having advanced professional positions of significant
responsibility, such as cabinet members, congresspersons,
ambassadors, judges, and other high status professional
positions (p.195). Several respondents compared the NDT to
the Super Bowl, the Master's Golf Tournament, or to the
World Series (p.200). The most salient finding of this study
was that there was a significant relationship between the
goals of higher education and the participation in
competitive debate. The authors concluded that "We are
almost hesitant, therefore, to conclude that the results
support debate participation almost without exception"
(p.205).
Several studies (Stewart & Merchant, 1969; King &
Phifer, 1968; Semlak & Shields, 1977; Colbert, 1987; Wilson,
1979; Allen, 1963) have examined the perceived differences
between debaters and non-debaters. Some of the findings
indicate that debaters have been found to be more
competitive, have better delivery skills, and demonstrate
superior critical thinking skills.
Stewart and Merchant (1969) examined perceived
differences between debaters and non-debaters. An audience
that observed both debaters and non-debaters speaking on the
same topics was able to identify the debaters 90% of the
time. The debaters were scored higher on every speech scale
except "fair-unfair"(p.72). Stewart and Merchant concluded
that experienced debaters overwhelm non-debate audiences.







King and Phifer (1968) examined the public image of
debaters. The study used a semantic differential scale that
was administered to both debaters and nondebaters. A factor
analysis indicated that debaters were viewed as .
effective, interesting, original, and convincing" (p.51).
The study also found no significance difference between the
debaters and nondebaters.
Semlak and Shields (1977) examined the effect of debate
training on those who participated in the bicentennial youth
debates and found that the students who received debate
training significantly improved their abilities in analysis,
delivery and organization (p. 195). Delivery consisted of
both the manner of the speaker as well as the speaker's
language choice.
Colbert (1987) examined the effects of debate training
on critical thinking ability and found that debaters scored
higher than non-debaters on the Watson-Glaser Critical
Thinking Appraisal, a measure of critical thinking. However,
the question of whether debate produced such changes or
merely attracted those students who already would score
higher in critical thinking was not answered. Colbert found
a difference between the debaters trained in NDT debate and
those trained in the newer CEDA (Cross-Examination Debate
Association) debate, but the quality of the difference was
not present (p.200). Colbert found a difference but he could
not identify it.
Wilson (1979) conducted research into the
competitiveness of intercollegiate debaters and found that
debaters are significantly more competitive than non-
debaters and that most of the variance was due to one
factor, toughmindedness. This study used Cattell's Sixteen
Personality Factor Questionnaire, a self-report instrument.
Allen (1963) assessed two different variables of
debaters. First, the debater's "interpersonal
compatibility"(p.67), the ability to work well with another,
was measured by the use of Schutz's FIRO-B scale. The
second variable was the debater's "concept
compatibility"(p.72), the ability to translate a single
concept into an internally consistent debate. This was
measured by the use of a semantic differential. Allen
concluded that debate teams in which the debaters were both
interpersonally and conceptually compatible would .
more efficiently interact in the preparation of debate
cases" (p.26). The finding that debaters who get along with
their partners and who approach the topic in a compatible
fashion are more effective would not be very surprising to
many debate coaches.
Two different studies (Barker, 1965; Sayer, 1974)
investigated debater's ability to decode nonverbal messages.
Debaters have been found to be poor at decoding nonverbal
cues but they have the potential to make correct assessments.
Barker (1965) examined debater-judge ratings and
found that judges and debaters do not rate the
debaters' performances the same" (p.19). The results
offered empirical proof that debaters are poor at
deciphering nonverbal communication from the judges. Sayer







(1974) wrote that Barkers' findings supported the old debate
maxim that debaters possess the worst perception of
what has happened during any round of debate" (p.5). One
reason for the debater's inability to correctly understand
the judges nonverbal cues is that debaters are usually too
busy, either preparing for their next speech or writing
down the other team's speech.
However, debaters do have the potential ability to
understand the judges' nonverbals. Sayer (1974) conducted a
study in which debaters were asked to predict the judge's
decision based upon their observation and
interpretation of the nonverbal cues" (p.4). The debaters
were correct at predicting the decision 66.5% of the time.
Sayer concluded that debaters have the potential to
correctly make nonverbal assessment. This study worked
because it asked the debaters to make a prediction that
forced them to pay more attention to the nonverbal cues.
Another group of studies examined the behavior of the
debater during the debate round. These studies examined the
evaluation criteria (Burgoon, 1975), debater's credibility
(Burgoon & Montgomery, 1976), use of praise and derogation
(Burgoon, Wilkinson, & Partridge, 1979), use of rate,
evidence and jargon (Vasilius & DeStephen, 1979) and a
content analysis of thirteen years of the first affirmative
rebuttal speeches from the NDT (Rouse & Thomas, 1987).
A study that examined the evaluation criteria as
predictors of debate success (Burgoon, 1975) examined the
six criteria on the American Forensics Association Form C
ballot, the most widely used debate ballot. The six criteria
were: organization, refutation, analysis, reasoning,
evidence, delivery, and success. The six criteria were
thought to be a quantitative measure of the debater's
personality. The study found that all six factors were
related to the debaters winning the debate. The analysis
demonstrated that the judges did not make discrete
differentiations among the six factors. The suggested
reasons for this result were that the judges were making a
gross or global evaluation or that there were other factors
that influenced the results. Burgoon did not speculate
regarding what the other factors could be, but one factor
could be the language used by the debaters; the debaters
might use language containing emotionally loaded words which
would influence the judge's emotive response.
Burgoon, along with Montgomery, (1976) also
investigated the debater's dimensions of credibility. They
received 186 completed questionnaires from debate coaches
throughout the United States. The questionnaire had 47
items, utilizing a seven interval semantic differential
instrument which asked the coaches to give their opinions of
an ideal debater's credibility. They found that three
factors accounted for 51 percent of the variance. The first
factor, which accounted for 48 percent of the explainable
variation, was "task competencies" (p.174). This factor
included qualities such as expert, active, articulate,
informed, coherent, competent and impressive. This factor is
a combination of the general quality of competence along







with good delivery skills that would deliver the competence.
Burgoon and Montgomery's second factor was "social
competencies" (p.174), which accounted for 26 percent of the
explainable variations and included qualities such as the
ability to be open-minded, relaxed, organized, poised, good-
natured, perceptive, and sociable. Burgoon and Montgomery
(1976) explained that this factor is related to the image of
the person who controls both social relationships and
himself (p.175). The debater is seen as both relaxed and
self-assured.
Burgoon and Montgomery's third factor was
"assertiveness" (p.174), which also accounted for 26 percent
of the variance, including the qualities of aggression,
headstrongness, criticalness, extrovertedness, verbal
ability, and experience. This third factor is based
primarily on the quality of extroversion along with the
quality of being critical (p.175). The verbal quality could
include the use of language by the debaters.
The finding that the debater's credibility is based on
three factors or dimensions was significant (p.175).
Burgoon and Montgomery (1976) argued that judges view
debaters through these three factors rather than the six
factors along the Form C ballots (p.176). A second
significant finding was that character, honesty and
ethicalness were not significant factors (p.176). The
third finding by Burgoon and Montgomery was that there was
little common agreement among the judges for a single
salient factor (p. 176). This finding would suggest that
judges view debates in a number of different ways and that
there are almost no universally accepted qualities that are
regarded as paramount. The overall finding of Burgoon and
Montgomery (1976) would serve to undermine much of the
research directed at personality and debate.
A study that examined the relative effectiveness of
praise and derogation as persuasion strategies (Burgoon,
Wilkinson, & Partridge; 1979) found that language of praise
was more favorable than derogation. This study also examined
the gender of the speaker as it interfaced with the type of
language that was chosen. They found that it was least
expected for a male to derogate another male and most
expected for a male to praise another male. The study found
that females derogating males with nontraditional receivers
were more effective than males derogating females, which was
contrary to the researcher's expectations. Also contrary to
their expectations, males were not rated higher on
credibility than female speakers.
Vasilius and DeStephen (1979) conducted an
investigation into the relationship between debate
tournament success and rate of speech, evidence, and jargon.
Vasilius and DeStephen (1979) found that faster- speaking
teams did not have greater success than slower speaking
teams; teams that used a greater amount of evidence were not
found to be more successful than those teams using a smaller
amount of evidence, and teams using a greater amount of
jargon were not more successful than teams using little
jargon. Jargon was defined as the specialized debate







language. A relationship between rate and jargon was found;
those teams who spoke faster were more likely to use jargon.
The study suggested that these factors do not contribute to
debate success and concluded that further research would be
needed to identify those factors that do contribute towards
success.
Overall, the research in debate regarding personality
and characteristics of debaters has been varied. There have
been a number of important results: (1) Debate produced
changes in the participants. They become significantly
improved in their analysis, delivery and organization. They
demonstrated better critical thinking ability and better
debaters demonstrated better interpersonal and concept
compatibility. These changes proved to be valuable for
debaters in future professional careers. (2) Debaters have
poor nonverbal decoding skills but they can learn to improve
those skills. (3) While there have been no clear
differences between male and female debaters, the results of
two of three studies suggests that mixed gender teams do
better than single gender teams. (4) Variables such as task
competency, social competence and assertiveness, were found
to be related to debater's credibility. Faster speaking
debaters and debaters who used more evidence and more jargon
were not necessarily more successful in debate.
(5) A study of first affirmative rebuttal speakers reported
that the debaters' arguments varied from the ideal, varied
in rate and varied in substantiation, and that the speeches
changed over time.
8 They quoted Clough in 1972 as making the argument
that "Solvency. One, lack of flexibility and defense. This
is not a plan meet need; it is a disadvantage couched as
one. It has nothing to do with the case we present to you.
But I will suggest that we can get the military. First, the
volunteers are right now" (p.109). This was contrasted with
Dripps who argued in 1979 that "Three, empirically it's
false, twenty-eight amendments; Constitution not destroyed"
(p.109).

9 Judges have been found to use different judging
styles or paradigms to evaluate debate. One study (Dunne,
Mack, & Pruett, 1971) suggested that judges utilize either a
logical (issues) or a proficiency (skills) paradigm to
evaluate the debate. Two other studies (Cox, 1974; Cross &
Matlon, 1978) reported that judges utilize five different
paradigms. Those two studies agreed on: (1) chooser of
policy system, (2) evaluator of argument skills; (3) tabula
rasa; and (4) evaluator of stock issues. One study (Cox,
1974) suggested the fifth paradigm to be a critic of
rhetorical proposition while the other study (Cross &
Matlon, 1978) suggested that it was a hypothesis tester.
Other studies (Hample, 1979; Crable, 1976) have suggested a
legal model.
(Also see footnote 3). Dunne, Mack, and Pruett
(1971) conducted an analysis of judging styles in debate.
Judges were classified as either logical (issues) or







proficiency (skills) judges. Only about half the judges
sampled agreed with the logical-proficiency dichotomy. Of
those judges who did agree, a majority favored the logical
judging paradigm. One possible reason for the large number
of judges that rejected the logical-proficiency paradigm was
that five different judging paradigms had emerged (Cox,
1974) where the logical paradigm overlapped with four
different paradigms and the proficiency paradigm became the
evaluator of argument skills.
Cox (1974) conducted an analysis of judging philosophies
in academic debate and reported that judges viewed
themselves primarily as reflecting one of five judging
philosophies: (1) chooser of policy systems with 42.9% of
the judges reporting that they saw themselves as primarily
using this philosophy, (2) evaluator of issues with 32.1%,
(3) evaluator of argument skills with 15.5%, (4) critic of
rhetorical proposition with 3.6%, and (5) tabula rasa with
5.9 % (p.62). Cox also found that many judges expressed
willingness for the debate to include theoretical issues,
including the judging paradigm to be used in the debate
round.
Similar to Cox's study, Cross and Matlon (1978)
conducted an analysis of judging philosophies in academic
debate and concluded that five philosophies prevail,
including: (1) The hypothesis testing model; (2) The chooser
of policy systems; (3) The evaluator of argument skills; (4)
The Tabula Rasa judge; and (5) The stock-issues judge. The
stock-issues is much the same as the evaluator of issues in
Cox's study. It is also noteworthy that Cox's critic of
rhetorical proposition was replaced by the hypothesis
tester. Cross and Matlon reached two important conclusions.
First, they found that the majority of judges in the
academic debate community view debates with extraordinary
consensus regardless of their stated judging philosophies"
(p. 123). This finding is in contrast to the results of the
Burgoon and Montgomery study. The second finding was that
judges appear to accept new theory arguments if the theory
is forcibly and convincingly defended by the team that
originated the argument (p. 123).
Four different studies (McCroskey & Camp, 1964
;Williams & Webb, 1964; Williams, Webb, & Clark, 1966; and
Williams, Clark, & Wood, 1966) investigated the criteria
employed by the debate judge in making a decision.
McCroskey and Camp (1964) examined stock issues, judging
criteria, and the debate decisions. The results included the
finding that the "need" issue or the importance of the
problem was paramount. A second finding was that the judge
usually determines the central issue during the constructive
speech. The successful debaters were those who came to an
agreement on the stock or central issues. McCroskey and Camp
further noted that debaters were unable to render an
objective decision about the outcome of the rounds they had
participated in. This final finding supports the forty
percent disagreement that King and Clevenger (1960)
reported.







Bauer and Colburn (1966) wrote that judges differ in
their decisions. Four reasons for the judges' differing were
offered: (1) the contest was a close debate; (2) the judges
used different evaluation criteria; (3) the judge was less
competent due to a lack of knowledge; and (4), an unethical
decision was made (pg. 23-24). One measure of determining
the rationale for the judge's decision is through the
judge's ballot. The ballot serves to provide direct feedback
to the debaters and their coach.
Ellis and Minter (1967) wrote that one can
never directly test the validity of debate decisions, since
there are no absolute criteria against which to judge them,
but the validity of decisions can be indirectly investigated
by seeing how consistent the decisions are from judge to
judge" (p.53). One of their findings regarding consistency
was that 27% of the judges might be inconsistent because of
travel fatigue (p.56).
Lewis and Larsen (1981) conducted a study of forensic
judges in individual events that examined interrater
agreement and found that additional training significantly
increased inter-judge agreement, and also found that .
the present study indicates that practical results can be
achieved by applying empirical research methods to the
judging of forensics activities" (p.16). Of course,
interjudge agreement may be affected by bias.
Three studies (McCroskey & Camp, 1966; Brooks, 1971;
and Hill, 1973) investigated judging bias. Investigations
regarding bias has centered around either a bias regarding
the topic or a geographic bias.
McCroskey and Camp (1966) asked 95 debate judges of
their bias regarding the topic. They found no bias by
judges regarding the topic. They reported an earlier study
by Scott found no bias and they concluded that .
Finally, Scott's conclusion that judges' bias on the topic
has no effect on their decisions was supported" (p. ).
Brooks (1971) noted that "an integral part of learning
is evaluation and feedback" (p.197), and in debate the
ballot is the feedback. Brooks attempted to find
geographical bias in judges' decisions. The results
supported some bias due to geographical distance and
concluded that more research was needed in investigating the
psychological and social distance. Part of the social and
psychological distance between the debaters and the judge is
that the judge usually has a holistic view of the debate
round while the debaters only see their viewpoints. It is
not uncommon for debate partners to have misperceptions
about each other's stands in a debate round.
Hill's dissertation (1973b) found no bias towards
either the topic or the geographical proximity of the
judge's and debater's schools. No study has supported any
bias towards the topic by debate judges. The existing
evidence on any geographical bias is less clear. One study
(Brooks, 1971) suggests that some bias exists while a second
study (Hill, 1973) found no support for a geographic bias.
One reason for investigations into potential judging bias is
because debaters rarely agree with the judge's decision.







Three studies (King & Clevenger, 1960; Klopf, Evans &
DeLozier, 1965; Hill, 1973) conducted comparisons of the
debater and the judge's assessment. Not surprisingly, King
and Clevenger (1960) found that debaters disagreed with the
judges over the result of a given round almost 40 percent of
the time. Klopf, Evans and De Lozier (1965) found that
laymen, students, and other faculty could judge debate if
their sole function was to make the win-loss decision.
Debaters were found to not agree with the judges.
One study, however, which did find a correlation
between the judges' evaluations and the debaters'
evaluations was Hill's (1973). Hill found that debaters had
similar assessments to those assessments of the judges.
Hill noted that his results were different from other
research and suggested the use of classroom debaters might
have caused the different result. Hill also speculated ". .
. that involvement in tournament debating tends to destroy
the individual's ability to accurately assess his own
performance"(p.377).
Three studies (Markgraf, 1966; Pearce, 1969; and
Berthold, 1970) examined the debaters' ability to adapt to
judges. Markgraf (1966) wrote that debaters feel confusion
because of the absence of uniform standards of judging.
Another problem was the great discrepancy between
professed ideals and actual debate judging" (p.38).
Markgraf argued that the solution was for the judges to
adapt to the debaters and not the other way around. While
he did not give his perspective a name, Markgraf was
advocating a tabula rasa judging paradigm.
One possible method to reduce the disagreement between
the debaters and judges would be for the judges to state
their viewpoints. In support of this method, Verderber
(1968) found that when judges state their criteria and
preferences before the debate, the majority of debaters
would adapt to the judges. Debaters also can learn the
judge's viewpoints over time. Pearce (1969) reported that
repeated exposure to a judge made debaters more
receptive to communication"(p.77) because the debaters could
learn the judge's criteria. Debaters improved because they
learned the judge's preferences and paradigm.
Berthold (1970) noted that most debate coaches
and debaters would agree that the single most important item
in any debate tournament is the judge's ballot. The ballot
not only reveals which team won a given debate, but it also
presumably indicates in a number of ways the participants'
debate skills" (p.30). Berthold analyzed the C-form ballots
that Burgoon (1975) also utilized. Berthold found that
judges wrote more in the first round of a tournament but
found that this was not related to the judges' overall
performance.
10 No all female teams in the sample.














CHAPTER TWO
METHODOLOGY


Research Design


The study investigated any effect of the use of

emotive words by four speakers involved in NDT

championship debate rounds. Operationally, the

relationships between three dependent variables and five

independent variables will be investigated. The three

dependent variables will be the overall use of emotive

words, and the uses of positive and negative emotive words.

The five independent variables are: result of the debate

round (winner or loser); side of the debate team

(affirmative or negative); position of the debate speaker in

rebuttal speeches (first rebuttal, second rebuttal);

the year of the debate final (1967 to 1985); and gender

make-up of the team (mixed or single-gender).



Variables



The dependent variable in this study was emotive words.

Those words were selected from two original instruments, the

Gottschalk-Gleser content analysis scale and the Osgood,

Suci, and Tannenbaum semantic differential scale. Three

dependent variables were identified in terms of the emotive









word list: positive emotive words, negative emotive words,

and the overall use of emotive words. Positive emotive

words, negative emotive words, and the overall use of

emotive words were further divided into the following

categories:

1. The emotive words used by the first negative

speaker in the first negative rebuttal speech, including

the use of positive words, negative words, and total

emotive words.

2. The emotive words used by the second negative

speaker in the second negative rebuttal speech, including

the use of positive words, negative words and total

emotive words.

3. The emotive words used by the first affirmative

speaker in the first affirmative rebuttal speech,

including the positive words, negative words, and total

emotive words.

4. The emotive words used by the second affirmative

speaker in the second affirmative rebuttal speech,

including the positive words, negative words, and total

emotive words.

The five independent variables will be:

1. Result of the debate round (winner or loser) as

measured by having a majority of judges vote for the

team.

2. Side of the debate team (affirmative or

negative).







3. Position of the debate speaker in rebuttal

speeches (first rebuttal, second rebuttal).

4. Year of the debate final (1967 until 1985).

5. Gender make-up of the team (mixed or single-gender

teams).



Procedures


Transcripts for Analysis

The transcripts of the final rounds of the National

Debate Tournament were published annually by the American

Forensic Association in JAFA from 1967 until 1985. This

study will analyze the transcripts of all of the rebuttal

speeches from those debates. The transcripts of the rounds

were made from audio tape recordings of the rounds. Rives

and Boaz (1977) explained that the debate was

edited from a tape recording. Except for the correction

of obviously unintended errors this is as close to

a verbatim transcript as was possible to obtain from the

recording" (p.10).

Thirty-eight debate teams representing twenty different

schools participated in the final round of the NDT between

1967 and 1985. Several schools made multiple appearances in

the final round, including six by Harvard and four each by

Dartmouth and Northwestern. A total of seventy-six debaters

were involved in these debates but because five debaters

appeared in two different final rounds, there was a total of

seventy-one different debaters involved. Negative teams won

eleven of the nineteen debates. Of the one hundred and nine








judges, fifty-seven voted for the negative teams and fifty-

one voted for the affirmative teams. Of the topics, five

were international in scope and fourteen were national

topics. The sides, teams, topics and results of the final

rounds from 1967 to 1985 are displayed in note 1.


Subjects


The study utilized the existing published record of

nineteen years of the transcripts of the final round of the

National Debate Tournament. A total of seventy-one debaters

took part in these debates. Four women and sixty-six men

debated in the final rounds between 1966 and 1985. A total

of one hundred and nine judges were used to judge the

nineteen debates.


Materials


The study utilizes three pieces of hardware: the

optical scanner; the computer that will utilize the wordscan

and other content analysis software, a Macintosh PC; and the

computer that will conduct the multiple analysis of

regression upon the results of the data, a mainframe VAX.

The study utilizes the wordscan and browser software

that was discussed above and the study will also utilize

SAS, a statistical package available from the Center for

Instructional and Research Computing Activities (CIRCA) at

the University of Florida. The wordscan program examined

the "text only" debate transcripts for emotive words. Those

emotive words were taken from Gottschalk and Gleser (1969)








content analysis scales (the hostility scale) and the semantic

differential scales of Osgood, Suci and Tannenbaum (1957).

Both scales contain bipolar terms that fall into positive

and negative categories. Items from both scales were

combined and additional words were derived from the original

by the use of a thesaurus and grammatical derivations.


Identifying Emotive Words

The transcripts were processed by an optical scanner

(Model: Apple Scanner 300dpi) that will convert the

transcripts into a word processed form (MS Word 4.0). The

scanned document was manually checked against the original

to ensure its accuracy. The resulting document was divided

into smaller documents: one document for each of the

rebuttal speeches and two or three documents for each of the

constructive speeches. This division was necessary for ease

of analysis and to accommodate limitations of the Wordscan

software. All documents were placed into a "text only"

format that have all non-word elements such as punctuation

removed. The resulting documents) were examined by a word

processing program called Wordscan that was developed by

Durkee and distributed through the public domain by EduComp

Services (Baez, 1989).

The program compared the "text only" document to a

separate dictionary and provided a display of any dictionary

ideas that were contained in the document and the number of

uses. In this study the dictionary is a list of emotive

words that were derived from two other instruments, the







hostility scale (Gottschalk, Winget & Gleser, 1969) and the

semantic differential (Osgood, Suci, & Tannenbaum, 1957).

The hostility scale was designed to measure attitudes and

the semantic differential was designed to measure emotions.

This study investigated the use of emotive words in academic

NDT debate. From Gottschalk's hostility scale and Osgood's

semantic differential scale, an aid has emerged to examine

emotive words (See note 2). Another public domain word

processor, Browser-developed by Zimmerman (Baez, 1989)

allowed any word from a document to be displayed in context.

Together, Wordscan and Browser examined the processed

transcripts to determine the existence, types, amount and

context of both negative and positive emotive words. The

examination of context indicated whether or not the word use

was anomalous. For example, the word was not used in a

quote or the word was not used as a filled pause, one

example is the word well.

Content analysis scales such as the Gottschalk-Gleser

content analysis method can be utilized through the use of a

computer. Gottschalk, Hausmann, and Brown (1979) wrote that

the use of the computer can save time, increase the

uniformity of the analysis and aid in efficiency and

rapidity. Viney (1986) stated that computerization is both

feasible and cost-beneficial. Viney reported on three

different computer-based scoring systems based on the

Gottschalk-Gleser scales. Viney also underscored an ethical

value to the use of content analysis:







Insofar as content analysis is a technique
for analyzing data and not for collecting it,
it raises no ethical issues. Yet, because it
can be applied to data collected in ethically
appropriate ways, it has value. It provides a
means of making sense of and rendering
useful/ the information from a simple
question posed to research participants such
as 'I am interested in X. I'd like you to
tell me about your experience of X.' (pp. 75-
76)


The participant's statement that Viney described could

be examined for use of emotive language. The Wordscan

software produced lists of the positive and negative words

that the debaters utilize.

Statistical Model

The best statistical model for conducting the analysis

is the use of the multiple regression model. Lewis-Beck

(1980) explained that multiple regression has great

range and its mastery will enable the researcher to analyze

virtually any set of quantitative data" (p. 47). This model

allows formulation of the relationship between the many

independent variables and the dependent variable(s).

Agresti and Agresti (1979) also explained the value of

the multiple regression model; it enables one to

develop a better predictor of a dependent variable than can

be obtained by using only one independent variable. Also,

these models allow one to analyze partial relationships

between two variables, controlling for other variables"

(p.322). Since the study utilized a number of different

variables, the possibility that partial relationships

existed between the variables might explain the existing

inconclusive results in the literature.




60



The formula that would be utilized for the multiple

regression was:

Y = Ln (P/l-P) = BO + BIX1 + B2X2 + B3X3 + B4X4

+ B5X5 + E


The variables deal with the result (X1), side (X2),

rebuttal position (X3), year of the debate final (X4), and

gender make-up of the team (X5) (See note 3).








Notes
SThe participating debate teams, result, strength of
the judges' decisions and the topic/resolution are listed:

1967 Dartmouth (negative) defeated Wayne State
(affirmative) on a six to one judges' decision.
Topic- Resolved: That the United States should
substantially reduce its foreign policy commitments.

1968 Wichita State (affirmative) defeated Butler
(negative) on a six to one judges' decision.
Topic- Resolved: Should the federal government
guarantee a minimum annual cash income to all citizens?

1969 Harvard (affirmative) defeated Houston (negative)
on a five to two judges' decision.
Topic- Resolved: Should executive control of United
States foreign policy be significantly curtailed?

1970 Kansas (negative) defeated Canisius (affirmative)
on a four to three judges' decision.
Topic- Resolved: Should the federal government grant
annually a specific percentage of its income tax revenue to
the states?

1971 U. California at Los Angeles (affirmative)
defeated Oberlin (negative) on a four to three judges'
decision.
Topic- Resolved: Should the federal government adopt a
program of compulsory wage and price controls?

1972 U. California at Santa Barbara (affirmative)
defeated Southern California (negative) on a four to three
judges' decision.
Topic- Resolved: Should greater controls be imposed on
the gathering and utilization of information about U.S.
citizens by governmental agencies?

1973 Northwestern (negative) defeated Georgetown
(affirmative) on a four to three judges' decision.
Topic- Resolved: Should the federal government provide
a program of comprehensive medical care for all United
States citizens?

1974 Harvard (affirmative) defeated Augustana of
Illinois (negative) on a three to two judges' decision.
Topic- Resolved: Should the federal government control the
supply and utilization of energy in the United States?

1975 Baylor (negative) defeated Redlands (affirmative)
on a three to two judges' decision.
Topic- Resolved: Should the power of the presidency be
significantly curtailed?







1976 Kansas (negative) defeated Georgetown
(affirmative) on a four to one judges' decision.
Topic- Resolved: Should the federal government adopt a
comprehensive program to control land use in the United
States?

1977 Georgetown (affirmative) defeated Southern
California (negative) on a four to one judges' decision.
Topic- Resolved: Should the federal government
significantly strengthen the guarantee of consumer product
safety required of manufacturers?

1978 Northwestern (negative) defeated Southern
California (affirmative) on a three to two judges' decision.
Topic- Resolved: Should United States law enforcement
agencies have significantly greater freedom in the
investigation and/or prosecution of felony crime?

1979 Harvard (negative) defeated Northwestern
(affirmative) on a three to two judges' decision.
Topic- Resolved: That the federal government should
implement a program which guarantees employment
opportunities for all United States citizens in the Labor
force.

1980 Northwestern (affirmative) defeated Harvard
(negative) on a five to zero judges' decision.
Topic- Resolved: That the federal government should
significantly strengthen the regulation of mass media
communication in the United States.

1981 Pittsburgh (negative) defeated Dartmouth
(affirmative) on a five to zero judges' decision.
Topic- Resolved: That the United States should
significantly increase its foreign military commitments.

1982 Louisville (negative) defeated Redlands
(affirmative) on a three to two judges' decision.
Topic- Resolved: That the Federal Government should
significantly curtail the powers of labor unions in the
United States.

1983 Kansas (affirmative) defeated Dartmouth
(negative) on a four to one judges' decision.
Topic- Resolved: Should United States military
intervention into internal affairs of any foreign nation or
nations in the Western Hemisphere be prohibited?

1984 Dartmouth (negative) defeated Louisville
(affirmative) on a four to one judges' decision.
Topic- Resolved: Should any and all injury resulting
from the disposal of hazardous waste in the United States be
the legal responsibility of the producer of that waste?





63


1985 Harvard (negative) defeated Iowa (affirmative) on
a five to zero judges' decision.
Topic- Resolved: Should the United States government
significantly increase exploration and/or development of
space beyond the Earth's mesosphere?

2 Norman Markel combined the hostility and semantic
differential scales into an emotive word list.

3 The value, P, is the percentage of emotive
words in each rebuttal speech.
The actual value choices for each independent
variable is displayed below:


= 1 if team won the round
= 0 if team lost the round

= 1 if affirmative
= 0 if negative

= 1 for first speaker of that team.
= 2 for second speaker of that team.


where X1


where X2


where X3


where X4


1967
for 1968
for 1969
for 1970
for 1971
for 1972
for 1973
for 1974
for 1975
for 1976
for 1977
for 1978
for 1979
for 1980
for 1981
for 1982
for 1983
for 1984
for 1985


where X5 = 1


if single gender team
= 0 if mixed gender team


1 for
= 2
= 3
= 4
= 5
= 6
= 7
= 8
= 9
=10
=11
=12
=13
=14
=15
=16
=17
=18
=19


=















CHAPTER THREE
RESULTS


Results of the study, comparison of the emotive

word use in the rebuttal speeches, results of the linear

regression, findings of the research questions, and a

summary of results are included.

The Study

The study was designed to examine the use of emotive

words by academic debaters in the National Debate

Tournament's championship round for each year from 1967 to

1985. The use of emotive words varied between debaters.

The first four tables (3-1, 3-2, 3-3, 3-4) present the

numbers of emotive words used by each debater. Table 3-1

displays the use of positive, negative, and overall use of

emotive words by the rebuttal speakers in the NDT from 1967

to 1969. Table 3-2 displays the use of positive, negative,

and overall use of emotive words by the rebuttal speakers in

the NDT from 1970 to 1974. Table 3-3 displays the use of

positive, negative, and overall use of emotive words by the

rebuttal speakers in the NDT from 1975 to 1979 and table 3-4

displays the use of positive, negative, and overall use of

emotive words by the rebuttal speakers in the NDT from 1980

to 1985. These tables display the number of emotive words

spoken by each debater.










Table 3-1: Use of Emotive Words in the NDT Rebuttal
Speakers from 1967 to 1969:

Year/Speaker Negative Positive Overall

1967 1NR 36 58 94

1AR 48 71 119


2NR

2AR

1968 1NR

1AR

2NR

2AR

1969 1NR

1AR

2NR

2AR


12

9

68

51

67

46

65

81

70

110


21

90

127

132

145

122

133

147

125

196









Table 3-2: Use of Emotive Words in the NDT Rebuttal
Speakers from 1970 to 1974:

Year/Speaker Negative Positive Overall

1970 1NR 52 60 112

1AR 74 73 147

2NR 80 61 141

2AR 76 95 171

1971 1NR 78 62 140

1AR 71 55 126

2NR 90 114 204

2AR 105 78 183

1972 1NR 83 44 127

1AR 59 41 100

2NR 93 47 140

2AR 95 64 159

1973 1NR 69 44 113

1AR 69 87 156

2NR 98 82 180

2AR 65 75 140

1974 1NR 94 80 174

1AR 53 42 95

2NR 87 55 142

2AR 91 55 146










Table 3-3: Use of Emotive Words in the NDT Rebuttal
Speakers from 1975 to 1979:

Year/Speaker Negative Positive Overall

1975 1NR 107 47 154

1AR 79 92 171

2NR 80 66 146

2AR 70 93 100

1976 1NR 103 65 168

1AR 70 62 132

2NR 106 85 191

2AR 104 61 165

1977 1NR 91 72 163

1AR 83 137 220

2NR 139 61 200

2AR 82 69 151

1978 1NR 81 35 116

1AR 21 79 100

2NR 98 65 163

2AR 88 88 176

1979 1NR 99 49 148

1AR 95 83 178

2NR 92 68 160

2AR 90 55 145


i









Table 3-4: Use of Emotive Words in the NDT Rebuttal
Speakers from 1980 to 1985:

Year/Speaker Negative Positive Overall

1980 1NR 79 86 165

1AR 105 122 227

2NR 82 98 180

2AR 89 109 198

1981 1NR 92 61 153

1AR 84 79 163

2NR 114 85 199

2AR 73 63 136

1982 1NR 96 103 199

1AR 107 120 227

2NR 103 111 214

2AR 73 91 164

1983 1NR 94 73 167

1AR 129 75 204

2NR 82 72 154

2AR 79 95 174

1984 1NR 98 97 195

1AR 88 140 228

2NR 86 100 186

2AR 106 75 181

1985 1NR 105 74 179

1AR 182 108 290

2NR 99 98 197

2AR 128 103 231







The debaters varied in their choices of emotive

words as well as the amount of emotive language that they

used. Tables 3-5 through 3-80 display the most frequently

used emotive language for each speaker. For each speaker,

the five most often used choices are presented, assuming

that each of those words were used more than two times. If

the speaker used identical numbers of some choices, then

all those choices are also displayed. The number of word

choices varies from zero (when a speaker never used an

emotive word choice more than two times) to ten choices

(when a speaker had six ties among those words used three

times).

Certain emotive words were used more than other words.

This was clear for the negative words (tables 3-5 through 3-

80). The debaters used a greater number of different

positive words than negative words. The debaters were

generally consistent in their use of the negative words.

The word "not" was used more than any other negative word.

"Not" was the most used negative word in all nineteen years

of the National Debate Tournament studied, for both the

second negative and second affirmative speaker. It was the

most used word for the first negative and first affirmative

speakers for eighteen of the nineteen debates. Both times

that the word "not" was not the most used word, it was the

second most used word. This happened in 1985 for the first

negative rebuttal and in 1983 for the first affirmative

rebuttal. In both debates, "argument" was used more than

any other negative emotive word.










Table 3-5:


Most Often Used Emotive Words for First
Negative Rebuttal 1967


Positive words


united
maintain
certainly
strong


Table 3-6:


Negative words


not


Most Often Used Emotive Words for First
Affirmative Rebuttal 1967


Positive words


gentleman
united
know
free
fact


Negative words


not
breaking








Table 3-7:


Most Often Used Emotive Words for Second
Negative Rebuttal 1967


Positive words


Negative words


not


Table 3-8:


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1967


positive words


negative words


not
argument










Table 3-9:


Positive words


kind
present
solution
advantage
social
solving
welfare


Table 3-10: Me
AJ

Positive words

gentleman
advantage
new
assistance
care
important
security
significant
social


Most Often Used Emotive Words for First
Negative Rebuttal -1968


negative words


not
no
poor
problems
without
deny


ost Often Used Emotive Words for First
affirmative Rebuttal 1968

negative words


not
poor
no
never





73




Table 3-11: Most Often Used Emotive Words for Second
Negative Rebuttal 1968


Positive words


negative words


present
gentleman
assistance
social
able
better
kind
solving
welfare


Table 3-12:


Most Often Used Emotive Words for Second
Affirmative Rebuttal -1968


Positive words

gentleman
advantage
conclude
security
significant
social


Negative words


not
argument
poor
problem
bit


not
poor
problem










Table 3-13:


Positive word

gentleman
intelligence
value
true
judgment
united


Table 3-14:


Most Often Used Emotive Words for First
Negative Rebuttal 1969

is Negative words


not
war
without
threat


Most Often Used Emotive Words for First
Affirmative Rebuttal 1969


Positive words

gentleman
intelligence
value
true
united


Negative words


not
harms


war









Table 3-15: Most Often Used Emotive Words for Second
Negative Rebuttal 1969


Positive words


intelligence
gentleman
important
like














Table 3-16:


Negative words


not
never
deny


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1969


Positive words

gentleman
respect
intelligence
united
judgment


Negative words


not
harm
harms


I









Table 3-17: Most Often Used Emotive Words for First
Negative Rebuttal 1970


Positive words

health
care
aid
advantage
increasing













Table 3-18: M4
A

Positive words


Negative words


not
poor
deny
problem
critical
cut


ost Often Used Emotive Words for First
affirmative Rebuttal 1970


Negative words


health
gentleman
care
advantage
fine


not
poor
no
never









Table 3-19: Most Often Used Emotive Words for Second
Negative Rebuttal 1970


Positive words


Negative words


care
health
solve
even
like
right


Table 3-20:


not
poor


problem
cut


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1970


Positive words


Negative words


advance
gentleman
solve
health
care
please


not


cut
problem










Table 3-21: Most often Used Emotive Words for First
Negative Rebuttal 1971


Positive words


Negative words


care
fact
power
advantage


Table 3-22:


Most Often Used Emotive Words for First
Affirmative Rebuttal 1971


Positive words


Negative words


gentleman
health
care
just
well


not
deny
poor
lose


not
no
harms
denies
hurts
problem









Table 3-23: Most Often Used Emotive Words for Second
Negative Rebuttal 1971


Positive words


Negative words


solve
whole
care
health
present


not
problem
problems
poor


Table 3-24:


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1971


Positive words


Negative words


advantage
help
increase
even
right


not
deny


critically










Table 3-25: Most Often Used Emotive Words for First
Negative Rebuttal 1972


Positive words


Negative words


well
might
win
ok
strong


not


argument
war


Table 3-26: Most Often Used Emotive Words for First
Affirmative Rebuttal 1972


Positive words


Negative words


united
guard
right
security
greater


not
argument
lose









Table 3-27: Most Often Used Emotive Words for Second
Negative Rebuttal 1972


Positive words


united
know
join
gentlemen


Table 3-28: MC
Aj


Positive words
well
win
right
guard
like
constructive
winning


Negative words


not
no
argue
death
argument
die
suffering


ost Often Used Emotive Words for Second
affirmative Rebuttal 1972


Negative words
not
argument


war
never










Table 3-29: Most Often Used Emotive Words for First
Negative Rebuttal 1973


Positive words


well
might
strong
win
ok


Table 3-30:


Negative words


not


drug
drugs
nothing
poor


Most Often Used Emotive Words for First
Affirmative Rebuttal 1973


Positive words


Negative words


health
care
increase
please
well


not


blood
problem





83



Table 3-31: Most Often Used Emotive Words for Second
Negative Rebuttal 1973


Positive words


Negative words


health
increase
advantage
advantages
clearly
free
persists
true


Table 3-32:


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1973


Positive words


Negative words


advantage
care
present
good
save
absolutely


not
strike
bad
drug


not


threat
blood
deny
destroy
never










Table 3-33: Most Often Used Emotive Words for First
Negative Rebuttal 1974


Positive words


Negative words


protection
clean
know
win
advantage
increase


not
pollution
no
harms
argument


Table 3-34: Most often Used Emotive Words for First
Affirmative Rebuttal 1974


Positive words


Negative words


gentleman
protection


not


problem
pollution





85




Table 3-35: Most Often Used Emotive Words for Second
Negative Rebuttal 1974


Positive words


Negative words


advantage
clean
win
well


Table 3-36:


pollution
argument
harm


Most Often Used Emotive Words for Second
Affirmative Rebuttal 1974


Positive words

increase
advantage
protection
please


Negative words


not
harm
pollution
no
against









Table 3-37: Most Often Used Emotive Words for First
Negative Rebuttal 1975


Positive words


special
please
know
just


Table 3-38:


Negative words


not


harm
deny


Most Often Used Emotive Words for First
Affirmative Rebuttal 1975


Positive words


Negative words


justice
right
advantage
better
new


not


never
argument










Table 3-39: Most Often Used Emotive Words for Second
Negative Rebuttal 1975


Positive words


Negative words


justice
please
power
true
good


not
no


Table 3-40: Most Often Used Emotive Words for Second
Affirmative Rebuttal 1975


Positive words


Negative words


justice
advantage
present
better
value


not


argument
never










Table 3-41: Most Often Used Emotive Words for First
Negative Rebuttal 1976


Positive words


Negative words


new
absolutely
aid
allow
just
united


not


problem
destroy
argument


Table 3-42:


Positive word

increase
adequate
reasonable
respect
increased


Most Often Used Emotive Words for First
Affirmative Rebuttal 1976

Is Negative words


not
argue


arguments
failure










Table 3-43: Most Often Used Emotive Words for Second
Negative Rebuttal 1976


Positive words


Negative words


fact
increase
absolutely
present
great


not


argument
pollution
negative
wrong


Table 3-44: Most Often Used Emotive Words for Second
Affirmative Rebuttal Speech 1976


Positive words

win
constructive
please
even
solve
trust


Negative words


not
argument
arguments
no










Table 3-45:


Positive word

present
safety
please
certainly
resolution


Most Often Used Emotive Words for First
Negative Rebuttal 1977

Is Negative words


not


argument
deny
negative


Table 3-46:


Most Often Used Emotive Words for First
Affirmative Rebuttal 1977


Positive words


Negative words


present
safety
respect
increase
new


not


argument
negative










Table 3-47: Most Often Used Emotive Words for Second
Negative Rebuttal 1977


Positive words


Negative words


even
just
safety
new
win


not


argument
argue
arguments


Table 3-48: Most Often Used Emotive Words for Second
Affirmative Rebuttal Speech 1977


Positive words


Negative words


well
present
please
right
just
safety


not


argument
wrong









Table 3-49: Most Often Used Emotive Words for First
Negative Rebuttal 1978


Positive words


Negative words


justice
know


not


argue
pollution


Table 3-50:


Most Often Used Emotive Words for First
Affirmative Rebuttal 1978


Positive words


Negative words


even


increase
heart
solve
solves










Table 3-51: Most Often Used Emotive Words for Second
Negative Rebuttal 1978


Positive words


Negative words


increase
fact
significant
well
increased


not
no
pollution
cut


Table 3-52: Most Often Used Emotive Words for Second
Affirmative Rebuttal Speech 1978


Positive words


Negative words


right
advantage
necessary
soft
increase
please
reasonable
significant
solve


not


negative
cuts
pollution




Full Text

PAGE 1

THE USE OF EMOTIVE WORDS IN INTERCOLLEGIATE DEBATE BY GERALD R. KISH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1991 UNIVER SITY OF F LORIDA U BRA RIES

PAGE 2

ACKNOWLEDGMENTS I would first like to thank my parents, Gerald and Sue Kish, for their constant support and encouragement. I would also like to express my sincere appreciation to my committee chairman, Dr. Norman Markel, for his continued support, guidance, and encouragement throughout the course of this study. His enthusiasm and dedication have greatly contributed to making both the completion of this study and my graduate program enjoyable and worthwhile experiences. I would like to also express my thanks to Dr. Paul Jensen for his rigorous editorial assistance and his never ending enthusiasm. Special thanks are extended to Dr. Anthony J. Clark for twice recruiting me to the University of Florida. I greatly appreciate what he has taught me, both in the field of debate and in the conducting of research. Many thanks to Dr. Ronald Randles for his assistance in the generation of the experimental design and for providing me with many statistical methods. It is with great appreciation that I acknowledge Professor Gerald Bennett for providing me with an opportunity to develop my quantitative research skills ii

PAGE 3

in a significant research project. I also appreciate his relentless support and encouragement. I would like to offer special thanks to Associate in Psychiatry Lynn Robbins for both her unending help with my statistics and her friendship. I would also like to offer special thanks to Chris Morris for all his help with this project. I cannot thank Linda Harris enough for all her help. I thank Dr. Marsha Vanderford, Dr. George Barnard, Dr. Gus Newman, Dr. Anita Raghavan, Dr. Ringo Ma, Mrs. Jane Milam, Kelly Roberts, Teresa Gudaitis, Roberta Bell, Lise Kent, Trish Sample, Donna Arents, Ann Allison, Roseanna Rutledge, Clem Sepulveda, Sara McDaniel and Sam Walch. Most especially, I would like to express my deepest thanks to Lynn Kish for sharing in everything I do with enthusiasm, friendship, and love. Also, my thanks for getting her doctorate while waiting on mine. iii

PAGE 4

-----TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . ii ABS TRACT . . . . . . . . . . . Vi CHAPTER ONE INTRODUCTION. 1 Overview . . . . . . . 1 Literature Review . . . . . . . 3 Brief History of Debate . . . . . 4 Importance of the Rebuttal Speeches. . . . 7 Judging in Academic Debate. . . . . . 9 Emotive Words ................... 12 Quantitative Research in Debate . . . . 17 Word Choice and Debate .............. 23 Word Choice and Legal Research. ...... 27 Empirical Focus. . . . ....... 32 The Dependent variable. . . . . .. 32 Gottschalk-Gleser Content Analysis Scale .... 32 The Semantic Differential Scale ......... 34 Validity and Reliability . . . . 35 Research Questions. . . . . . 36 Notes . . . . . . . . . . . 38 CHAPTER TWO METHODOLOGY Research Design Variables . Procedure. . .. Transcripts for Analysis. Subjects ...... . Materials ....... Identifying Emotive Words Statistical Model .... Notes . . . . . iv 53 53 53 55 . . . . 55 . . 56 56 . 57 . . . . . 59 . . . 61

PAGE 5

CHAPTER THREE RESULTS The Study .... Linear Regression Analysis Regression Analysis ... Research Question Results Summary . . . . CHAPTER FOUR DISCUSSION Overview of the Study ... Interpretation of the Results Implications for Future Research .. Limitations of the Study .... REFERENCES BIOGRAPHICAL SKETCH. V 64 64 115 119 138 139 141 . . 141 143 . 152 153 155 17 5

PAGE 6

A Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE USE OF EMOTIVE WORDS IN INTERCOLLEGIATE DEBATE by Gerald R. Kish May, 1992 Chair: Norman Markel, Ph.D Major Department: Communication Processes and Disorders This study investigated the relationship between emotive words and five variables of academic debate: winning, speaker position, side of topic, year of debate, and gender mixture of the teams. Debate is an art that dates to the beginning of civilization. Research has not determined the keys to successful advocacy. Several scholars suggest that emotive words are vital to successful communication. Noone has investigated emotive words in debate. Transcripts of nineteen years of intercollegiate debate from the National Debate Tournament were analyzed by a computer aided content analysis. The empirical focus utilized emotive words as the dependent variable. The data were analyzed by the use of six SAS regression vi

PAGE 7

procedures. The results of this study's five research questions, related to academic debate and the use of emotive language, are: 1. Do winning and losing debate teams differ in their use of emotive words during rebuttal speeches in a debate round? No support was found. 2. Do affirmative and negative debaters differ in their use of emotive words in rebuttal speeches? No support was found. 3. Do the first and second rebuttal speakers for each debate team differ in their use of emotive words? No support was found. 4. Does the quantity of emotive words used by debaters vary over a period of years? The results suggested that the use of emotive words was related to the year variable. 5. Does the quantity of emotive words used differ between mixed gender debate teams and single gender debate teams? The results found that mixed-gender teams use greater emotive words. The significant result related to research question number four is the most important finding of this study. The statistical analysis that debaters utilized increasingly greater number of emotive words over the nineteen years of debates studied. The most important contribution of this study is that academic debate is placed in an empirical context. That is, this research is a significant step in closing the breach between the academic debate community and the quantitative researchers in the communication field. vii

PAGE 8

CHAPTER ONE INTRODUCTION Overview "Ideally, all judges and all debaters should proceed from a commonly understood set of rules for making the decision. Unfortunately, different types of judges have different standards (Thomas, 1987c, p.123) ." Researchers in the field of contemporary academic debate have attempted to discover what overall factors contribute to the debate decision. Among the factors that researchers have studied are characteristics of the judge's ballot (Berthold, 1970), judging bias (Mccroskey & Camp, 1966; Brooks, 1971), stock issues in debate, the ability of debaters to adapt (Markgraf, 1966; Verderber, 1968; Pearce, 1969), gender of the debater (Hayes & McAdoo, 1972; Hill, 1973b; Rosen, Dean, & Willis, 1978), side of the issue (Hill, 1973b; Rosen, Dean, & Willis, 1978), competitiveness of the debaters (Wilson, 1979), rate of speech (Vasilius & Destephen, 1979), evidence use (Larson & Giffin, 1964; Dresser, 1964; Newman & Sanders, 1965; Dresser, 1966; Rieke & Smith, 1968), qualities of arguments (Allen & Kellermann, 1988), debate jargon (Vasilius & Destephen, 1979), and the 1

PAGE 9

2 use of praise and derogation (Burgoon, Wilkinson, & Partridge, 1979). None of these studies have demonstrated a clear relationship between any one specific factor and winning in intercollegiate debate. Only a few studies have examined the use of language as a factor in winning a debate. Rouse and Thomas (1987) found that debaters changed their word usage over a span of 13 years. Newman (1939) argued that while language appears to have a rational, or logical, basis, communication (language) often functions through irrational, or illogical, means. In general, research has also found that certain types of language are related to psychological characteristics of the speaker (Markel, 1990; Gibson & Felkins, 1974; Gottschalk, 1961; Newman, 1939; Newman & Mather, 1938). In addition, other research (Parkinson, Geisler, & Pelias, 1983; Parkinson, 1981; Conley, 1979; O'Barr & Conley, 1976) has found that word choice can have an influence on success in an actual trial setting. Further research has demonstrated the existence of emotive words (Clore, Ortony, and Foss, 1987; Irene, 1990). Each of these conclusions may be applied in an investigation of the issue of the effect of emotive language on debate decisions. To better understand the importance of language in debate, one must first consider the functions of language. The emotive function of language has been described as a fundamental dimension of human life (Sapir, 1921; Jakobson, 1960; Irvine, 1990). The emotive function of language communicates a speaker's emotions and attitudes (Fiske,

PAGE 10

1982). The Gottschalk and Gleser (1969) hostility scale and the Osgood, Suci, and Tannenbaum (1957) semantic differential scale both purport to measure the speaker's emotions and attitudes. A modified emotive word list was developed from these two scales by Markel (1989). It allows analysis by categorizing words into positive and negative words. The purpose of this study was to investigate the 3 use of emotive words in collegiate debate. While the emotive function of language is important, the use of emotive words in academic debate has never been investigated. Initially, the significance of debate will be discussed, followed by the history of debate, characteristics of the national debate tournament, speaker duties and the importance of the rebuttal speech, and debate judges and their judging paradigms in debate. Following this, a discussion of function of emotive words and debate will be considered. Next, a discussion of the quantitative research in debate, and of research of language effects in both debate and the field of law will be presented. Finally, a discussion of the dependent variable will be offered. The methodology for the study will be found in chapter two. The results will be found in chapter three and the discussion in chapter four. Literature Review From the earliest times in recorded history, thoughtful people have recognized the importance of debate for both society and the individual (Freeley, 1976, p.2). Debate has

PAGE 11

4 a "long and honored place in academia'' (Colbert & Biggers, 1985, p.237). Debate is a process of inquiry and argumentation that seeks a reasoned judgment of a proposition. Patterson and Zarefsky (1983) defined debate as argumentation that occurs in a formal setting. To find a resolution of two or more differing positions involves argumentation and, very often, debate. Differences between people or between groups of people can be resolved by five different means (Patterson & Zarefsky, 1983, p.309): first, a resolution may be reached by force; second, decisions may be made based on impulse or caprice; third, resolutions may be decided by chance; fourth, one can allow another party to make the decision, and, finally, each side can argue the issue and attempt to persuade the other. The most logical and reasonable method to reach a resolution is the fifth means, that of argumentation. Our society provides many opportunities for a wide variety of types of argumentation and debate, including political and judicial debates, parliamentary debates and academic debate. Brief History of Debate An examination of the history of debate is important in understanding the evolution of the activity and providing a foundation from which debate can be evaluated. Debate has always been viewed as a process of rational decision making (Petrie, 1969); understanding the factors involved in such

PAGE 12

5 decision making may clarify the importance of rationality in such decisions. Briefly, debate appears to have first been recorded by scholars of the Chou Dynasty some 3000 years ago (Freeley, 1976 p.17). Debate also appears in Homer's Iliad and Odyssey (Ovid, 1955; Sproule, 1974) and in Aristotle's Rhetoric (Murphy, 1964; Sproule, 1974). Medieval educators required disputations of their students and intercollegiate debate began as early as the fifteenth century (Murphy, 1964; Freeley, 1976). From these early intercollegiate contests, debate organizations were formed and debate tournaments organized (Coburn, 1972). In the United States, these organizations and debate tournaments have undergone many changes (McBath & Aurbach, 1967). The debates investigated in this study were those of championship rounds at the National Debate Tournament. The National Debate Tournament (NOT) uses a national resolution that is chosen in the fall of each year and debated by college debaters through the year. The NOT is the final tournament of the year and usually occurs in late April or May. The NOT resolution is not the only debate topic that college debaters can debate. There have always been other resolutions, but the NOT resolution was considered the most important for many years. Matlon and Keele (1984) surveyed 703 participants in the NOT and reported" ... that some of the most intellectually gifted students have participated in the National Debate Tournament" (p.195). After obtaining the advanced degrees, the debaters were found having advanced professional positions of significant

PAGE 13

6 responsibility, such as cabinet members, congresspersons, ambassadors, judges, and other high status professional positions (p.195). Several respondents compared the NDT to the Super Bowl, the Master's Golf Tournament, or to the World Series {p.200). In the early 1970s, another type of resolution was created by the Cross-Examination Debate Association {CEDA). The CEDA topic was a non-policy resolution and was created to counterbalance perceived abuses of NDT, including excessive speaking rates. CEDA started slowly, but by the middle 1980s, CEDA had replaced NDT as the debate format of choice for most debaters. Patterson and Zarefsky (1983) wrote that NDT resolutions have four characteristics: First, is resolutions have policy implications (p.90), Secondly, is resolutions are worded broadly (p.91), Thirdly, resolutions propose actions that are unlikely to be taken soon (p.91), fourth and finally, the meaning should not be self-evident (p.92). The characteristics were designed to increase argumentation by debate teams on both sides of the debate resolution. In NDT, or policy debate, the affirmative team has traditionally had three burdens to prove: (1) that a significant problem exists, (2) that the current system cannot solve the problem because of an inherent barrier, and (3) that the affirmatives solution of the problem will be efficacious or solvent. These requirements have been known as the "stock issues" in a debate. Just as the affirmative has certain traditional

PAGE 14

requirements, each of the speeches in a debate has requirements and duties. Importance of the Rebuttal Speeches 7 The speakers all have different duties (See note 1) and while all the speeches in a debate round are important, the most important speeches are the rebuttal speeches, for several reasons. First, the rebuttal speeches are only one half as long as the constructive speeches. The debater's time in the rebuttal speech is thus twice as valuable. Mistakes in time usage have been critical in debate rounds. The debater's time usage is most critical for the first affirmative rebuttal because that speech must respond to both the second negative constructive and the first negative rebuttal. The first affirmative rebuttal must deal with two negative speeches that total three times the affirmative speaker's time. Any dropped issues are crucial as it is generally considered that the second affirmative speaker can not rectify the first affirmative rebuttal's mistakes (See note 2). Second, while new arguments can be made in either of the constructive speeches, arguments can only be extended in the rebuttal speech. Once a speaker fails to cover an argument, it is presumed to be lost to that team. Third, a recency effect might possibly allow a judge to better remember rebuttal speeches. Fourth, debaters vary in the effectiveness of their rebuttal strategy. Thomas (1979) explained that in championship debates, it is fair to say that more debates ... are won--or lost--in the

PAGE 15

8 rebuttals than in the constructive speeches. All too many debaters, who have a good grasp of debate in general, go into their rounds without a clear picture of what they are going to do in their rebuttals" (p.288). A key to success in rebuttals is the ability of a debater to cover the other team's arguments. Rebuttal speakers typically speak rapidly in order to respond to their opponent's arguments and extend their own arguments. A successful rebuttal" ... requires fluency and, more importantly, a gift of economical language" (Thomas, 1979, p.294). The importance of language in the rebuttal speech is highlighted in a study by Rouse and Thomas (1987) which found that first affirmative debaters steadily declined in their elaboration and rhetorical amplification over the years. They concluded that" ... this strongly suggests that ... rhetoric and psychological tactics may have given way to more efficient, compacted argumentation" (p.109). Scholars differ in their assessment of deciding which rebuttal speech is paramount. Rouse and Thomas (1987) wrote that the first affirmative rebuttal speeches often decide the outcome of the debate. However, William Southworth (1984) wrote in his NDT final round ballot that .. as is so often the case, one need only turn to the last two rebuttals to isolate what issue(s) the teams considered necessary to warrant a ballot" (p.53). Mayer and Meldrum (1987) concluded that the last rebuttal was the most valuable speech. Of course, one reason that the last

PAGE 16

rebuttal might be the most important is because no other speaker can subsequently influence the judge(s). Judging in Academic Debate 9 Since the the time of the early Greek rhetoricians, the question of "who has won" (Sayer, 1974, p.6) has been the focus of every forensic debate. Academic debaters entrust the decision to a third party, the debate judge. The judge listens to both sides and then decides which team won the round. Understanding the elements that contribute to winning academic debates has been the focus of some research, and will be considered here. It is the debate judge who always makes the decision as to which team wins each round, with no ties allowed. David Thomas' (1987c) explanation of the basic judging paradigms is important: If the judge is sincere and conscientious, he or she will make the decision on the basis of some model of an ideal debate and of what a debate team must do in order to deserve to win the debate. In adapting to a judge, therefore, it is important for the debater to realize what the judge's model of the ideal debate consists of ... A debate is like any other human experience in that the participants sees it from their own personal perspective, and they may honestly vary from each other in describing what happened -thus the saying that 'debaters never lose debates, only decisions.' Like the debaters, the judge also sees the debate from a personal perspective. (p.116) Debaters usually strive to understand their judge's perspective and adapt their presentation to that perspective. However, the existence of competing paradigms

PAGE 17

--------------~ 10 has been controversial throughout the history of debate (Dunne, Mack, & Pruett, 1971). The judging paradigm controversy was argued in academic circles as early as 1917, (Dunne, Mack, & Pruett, 1971) with one paradigm arguing that II the task of the judge, therefore, is to place himself in the position of one who has no opinions or knowledge of the subject, other than what has been presented, and to make the decision which any reasonable and intelligent person would predicate upon the premises" (p.201). An opposing paradigm argued that" ... the decision should always be given to the team which shows superior attainment" (p.201) or speaking skills. These competing paradigms are meta-perspectives that serve the debate judge as a mechanism for decision-making in debate rounds. Research has demonstrated that there are five different judging paradigms in current academic debate (Cross & Matlon,1978). The five are: the hypothesis-testing model, the chooser of policy systems, the tabula rasa judge, the stock issues judge, and the evaluator of argument skills. Others (Hample, 1979; Crable, 1976) would argue that a sixth, a legal judging paradigm, exists as well. (See note 3) Whatever the paradigm, judges possess a great deal of power in academic debate. Hufford (1965) wrote that the debate judge is 11 like the Supreme Court, infallible. There is no appeal (p.120)." The power of the debate judge has provoked controversy for at least the last fifty years. Laase (1942) wrote that because of the criticism leveled

PAGE 18

11 against judges, debate should switch to a peer debater quality rating system. This proposal was never implemented on a wide scale and criticisms of judges have continued. The majority of judges in academic debate are usually debate coaches or speech communication teachers, but other faculty and laymen judges are also used (Klopf, 1964). Judges at the National Debate Tournament are typically debate coaches. Judges who are considered to be "good" judges, those who are considered to be unbiased and competent in the conventions of debate, are sought after by debaters. The good judge's viewpoints are valued because debaters wish to be able to predict how their judges will evaluate the debate. "Bad" judges are considered to be incompetent or biased or both. However, debaters do typically attempt to adapt to judges, even bad ones. Unlike trial judges, but similar to appellant judges, academic judges usually provide a written explanation to the participating teams after the end of the round. Not surprisingly, good judges usually write comprehensive ballots, discussing the reasons for the decision in greater detail (See footnote 4). Ideally, all judges would utilize the same standards for judging debate and making the decision (See footnote 5). That would allow debaters to expect consistent judging and they could thus concentrate on the topic and not worry about idiosyncratic differences in judges. Because of the lack of uniform standards for debate judging, there has been a II continuing disagreement within the debate community

PAGE 19

12 in regard to methods of judging debate. It is little wonder that our debaters are at times dismayed with the decisions rendered by their judges" (Dunne, Mack, & Pruett, 1971, p.207). The resulting confusion has helped undermine the logical and rational processes of debate and has contributed towards research that has investigated other factors that would explain success in debate. One of these other factors is the debater's word choice of emotive words. Emotive Words Jakobson (1960), Ogden and Richards (1923), Berger and Bradac (1982), and Littlejohn (1983) all expressed the notion that words have an emotive function. Each of these scholars described the emotive function of words as the expression of a speaker's feelings. The goal of this function is explained by Littlejohn (1983) who claimed that ... with emotive discourse the communicator hopes to elicit similar feelings and attitudes" (p.96). Jakobson (1960) identified six primary functions of words -emotive, referential, poetic, phatic, metalingual, and conative. Jakobson (1960) identified the emotive function of words as being the communication of the attitudes and emotions of the addresser. Jakobson (1960) further explained that: (the) emotive or 'expressive' function [of words), focused on the addresser, aims a direct expression of the speaker's attitude toward what he (/she) is speaking about. It tends to produce an impression of a certain

PAGE 20

the term 'emotive', ... has proved to be preferable to 'emotional (p.354) In elaborating Jakobson's definition of the emotive function of words, Fiske (1982) stated the term 13 "expressive" is also used to describe this emotive function (p.37). Fiske goes on to say that the emotive function serves to communicate characteristics of the speaker such as emotions, feelings, status and other items that make the message uniquely related to the speaker (Fiske, p.37). The present study focuses on the emotive function of words. That the emotive function of words may be central to the study of debate, and could be a factor in persuading the judges to accept the debater's arguments is indicated by Jakobson (1960) 11 if we analyze language from the standpoint of the information it carries, we cannot restrict the notion of information to the cognitive aspect of language" (p.354). The major thrust of the present research has been to examine debaters' use of emotive words in their presentation of arguments. The emotive function of words also serves to persuade the listener. Bateson (quoted in Kobayshi, 1988) stated" that through their work great artists communicated complex emotive messages and provided perceptual perspectives that pushed or pulled people towards what they did" (p.352). That is to say, a central factor in emotive words is that the message conveys the addresser's emotions and attitudes, and in addition as indicated by

PAGE 21

Bateson, and germane to debate, is the fact that emotive words attempt to push the listener in a given direction. 14 Berger and Bradac (1982), in a similar vein, have indicated that a primary function of expressive or emotive words is to convey information other than the message content to the listener. Furthermore, they state that the emotive function's information is communicated or 'given off' by word choice (p.53). They wrote that there are three primary functions of words -referential, instrumental, and emotive. The emotive function of words has been investigated by a small number of scholars. One reason for the lack of research in emotive language was offered by Beeman (1988) who explained that linguists of all breeds seem to develop cold feet when it comes to discussion of the expression of emotion in language" (p.9). She suggested that the problem was the perception of emotive language as "soft" and "idiosyncratic," as contrasted with rule-governed structures in linguistics. Irvine (1990) concluded that while many scholars have ignored the emotive function, two of linguistics' most stellar figures, Jakobson (1960) and Sapir (1921, 1927), thought otherwise: Affect, or emotion, according to them, was a fundamental dimension of human life and a factor cross-cutting all levels of linguistic organization" (p.126). Some researchers who have investigated emotive word usage include: Allport and Odbert (1936), Dahl and Stengel (1978), Bolton (1979), Berger and Bradac (1982),

PAGE 22

15 Clore, Ortony, and Foss (1987), and Irvine (1990). Allport and Odbert (1936) created a lexicon of trait-names that were divided into those words that were emotionally neutral, evaluative, and emotionally active words. Dahl and Stengel (1978) divided emotive words into three primary dimensions: (1) Subject-Object or It-Me, (2) Attraction-Repulsion, and (3) Extensor-Contractor or To From. Dahl and Stengel also divided the attraction-repulsion words into positive and negative sub-types. They utilized 58 judges to classify 371 words and the attraction-repulsion/ positive-negative words were significantly agreed upon by the judges. Bolton (1979) wrote that "emotions are the key to vital communication" (p.92) and offered a short list of "feeling words." Bolton (1979) wrote that speakers should use feeling words that match the listener's experience. Bolton provided an example of a couple who failed to do so. They are looking at the Grand Canyon for the first time and after a long silence one partner says to the other, "This is magnificent. It is sublime" (p.92). The other responds; "You think it's pretty" (p.92). Berger and Bradac (1982) wrote that language could be classified into subtypes such as "powerful and powerless" language (p.60) or into "familiarity and goodness" (p.57). Familiarity was related to a concept of a lexical "goodness," (p.57) in which predictable or familiar words were rated positively. They also discussed "linguistic charms," in which lexical items" ... typically produce very positive evaluations" (p.57). The reverse of the good

PAGE 23

----16 word list contained words that 11 denote excretory and sexual processes, profanation of scared objects, and death or decay" (p.57). The language associated with powerless speech contained hedges, intensifiers and tag questions. Words could be categorized as being either positive or negative. Clore, Ortony and Foss (1987) created an emotive word list, which they called an affective lexicon. They defined II affective to refer to the positive or negative evaluation, or valence, inherent in the meaning of a term" (p.751). They also found that 234 English words were empirically classified as having affective conditions. A subject pool of 435 undergraduate students was used to rate words and multiple contexts were used. One conclusion of this study also found that these words were related to psychological conditions of the speaker. Irvine (1990) wrote that a growing body of research has supported the importance of emotive words. Irvine investigated the Wolof population in the African nation of Senegal and found distinctive differences in the use of emotive speech. The research on the emotive function of communication supports the notion that emotive words are inherent in persuasive language. Research also supports the notion that emotive words can be divided into subtypes, such as the division into positive and negative emotive words. Clore, Ortony and Foss (1987) wrote that a positive or negative evaluation was "inherent" (p.751) to the concept.

PAGE 24

17 Considering the inherent nature of emotive words, it is interesting to note that there has been no investigation of the use of emotive words in academic debate. This study will investigate the use of emotive words in academic debate. While the influence of emotive words in academic debate has not been investigated, there is a body of research in debate, some of which is relevant to this study. Quantitative Research in Debate Anderson (1966) reported that the status of an academic field can be determined by the nature of the research and that the future of a field can be predicted from the field's research trends; unfortunately, quantitative research in academic debate is sparse. Vasilius and Destephen (1979) wrote that" .. there are many empirical studies relating to communication strategies, but fewer on debate strategies and even fewer on style in debate" (p.197). Some existing research has even been described by Thompson (1966) as banal and provincial" (p.109) (See note 6). In particular, there has been little research investigating language use by debaters. The existing quantitative research that has examined the relationship of other variables to debate has been generally inconsistent (See note 7). Of the research which has a direct bearing on the subject of this study, the studies (Hayes & McAdoo, 1972; Hill, 1973b; Rosen, Dean, & Willis, 1978) that investigated the influence of gender are important, and the

PAGE 25

study of word use by Rouse and Thomas (1987) provides a foundation for the present study. 18 Three studies have investigated the issue of gender in debate. One examined the effect of gender and evaluation (Hayes & McAdoo, 1972); the second also examined the effect of gender on the debate results (Rosen, Dean & Willis, 1978). Both of these studies found that mixed-gender two person teams did better than all male or all female debate teams (See footnote 20). However, a third study by Hill (1973b) found that all male teams did better than all female or mixed-gender teams. Rouse and Thomas (1987) conducted a content analysis of the transcripts of the first affirmative rebuttal speeches of the NOT from 1972 until 1984. This is an important study because they conducted a content analysis of 13 years of NOT speeches. They examined speaking rates and noted that three first affirmative rebuttals with the fastest speaking rates were on losing teams and the average rate of winning speakers was 270 words per minute, compared to an overall average of 286. Rate of speech varied from 234 in 1972 to 355 in 1984. Despite this finding, the Rouse and Thomas study did not conclude that there was a direct correlation between a rapid rate and losing. Rouse and Thomas also wrote that several trends were noticed. First, they reported an increasing trend of debaters offering "roadmaps" or signposting the material that their speech would cover. Second, they found a consistent decline in elaboration and rhetorical

PAGE 26

19 amplification. They cited two rebuttal responses from their transcripts, both in which a single argument is advanced. The earlier example, from 1972, contained one hundred words exactly, while a 1979 example contained only seven words. This ''streamlining" they saw as a natural response to the requirements of a successful first affirmative rebuttal, i e I II fluency, and, more importantly a gift of economical language'' (p.101). As they considered the outcome of a debate to be frequently determined by this rebuttal, it may be concluded that language use is crucial to decisions (See footnote 8). That study, however, had some limitations. no analysis of negative speeches for comparison. examined the first affirmative rebuttal speech. They made They They did not conduct any statistical analysis because the study was "descriptive" (p.110). The study did not test for changes over a time dimension even though "there have been some notable differences observed over the time span, such as the speeding up of speakers' rates" (p.110). The study ultimately concluded that other factors besides the speaking technique affect the winning or losing of a debate round. Other quantitative research in academic debate that is relevant to the current study has investigated judging in academic debate. This research has focused on examining judging practices, alternative methods of judging, and evaluations of the quality of judging. Some of these studies include an investigation of the judging of persuasive speech (Benson & Friedley, 1982), a study of

PAGE 27

20 demographic characteristics influencing debate judging (Hill, 1973b), and three studies that conducted an analysis of judges' criteria that included the debater's language (Williams & Webb, 1964; Williams, Webb, & Clark, 1966; Williams, Clark, & Wood, 1966). Benson and Friedley (1982) investigated the judging of persuasive speaking and concluded that little is known about how judges arrive at their decisions. They noted that a major purpose of their study would be to stimulate quantitative research, stating that" ... only with an established body of empirical research to describe and evaluate its contributions" will forensics be able to withstand budgetary pressures (p.12) (See note 9). One empirical investigation of non-debater variables on the judges' decisions was conducted by Hill. Hill's (1973b) dissertation investigated the influence of variables not related to the debater's ability on the judge's decisions. He investigated the influence of the topic, speaker position, sex, prestige of the school, and proximity. The research concluded that both the topic and the speaker position had no significant influence on the debate outcome. Hill also reported that the geographical proximity of the judge's school to the debater's schools did not influence the debates' outcome. He did find that gender had an effect with all male debate teams having a greater chance at winning than all female or mixed debate teams. The study established that prestige also had an effect, with junior college debaters being most negatively affected. Hill

PAGE 28

21 defined high prestige teams as those schools that had previously participated in the National Debate Tournament. Williams and Webb (1964) also investigated factors in debate evaluation. The study asked judges to respond to 37 bipolar descriptive scales of debaters, and used factor analysis to determine that the dominant factor was "argument." This factor included scales for supporting material, concreteness, logic, relevance of evidence, analysis, reasoning, pertinency, refutation, persuasiveness, use of motive appeals, and organization. A second factor was "vocal-correctness" and it included articulation, pronunciation, grammar and vocal quality. The third factor was "overall-delivery" and included scales for eye contact, rate, spontaneity, interestingness, facial expression and intelligibility. The fourth was "apparent-character" and included scales for courtesy, sportsmanship and ethics. They reported that the first factor accounted for 36% of the total variance. The amount of the accounted variance for the other factors was not offered. They also reported in an endnote that ... terms in addition to those reported in the results to this study included: ... Word choice, Use of figurative language" (p.128). Williams and Webb did not explain why these items were ignored. Two follow-up studies (Williams, Webb, & Clark,1966; Williams, Clark & Wood,1966) supported the initial results of Williams' and Webb's (1964) study. Williams, Webb, and Clark (1966) conducted a follow-up study to the Williams and Webb (1964) study. The results were similar to the earlier

PAGE 29

22 study. Argument was again the dominant factor. They concluded that "relatively few dimensions of evaluation underlie judges' use of a relatively large number of rating scales" (p.20). While word choice and the use of figurative language were not discussed in the study, the "vocabulary" (p.16) scale did not have a high loading on any of the factors in the study. Williams, Clark and Wood (1966) conducted yet another follow-up study and again obtained similar results. The argument dimension was found to predict the overall rating as well as the other factors combined (p.100). They concluded that judges offer overall assessments of argumentative skill rather than judging on individual factors such as logic or evidence. Overall, quantitative studies in debate which focus on judges has produced some consistent results. (1) Judges currently use five different paradigms in judging debate: hypothesis tester, policy evaluator, tabula rasa critic, stock issues judge, and evaluator of argument skills. (2) Studies of judging criteria report that the factor of argument is important to the judge's decision in the round. Other studies reported that the need issue was paramount. (3) Judges and debaters usually do not view the round in the same fashion. (4) At least one study suggested that debaters' word choices do influence the debate decision. (5) Bias of the judges regarding the topic was not found, but some geographical bias was found. (6) Debaters become

PAGE 30

confused because of inconsistency in judging. were able to adapt to judges over time. 23 (7) Debaters The question of what influenced a judge to vote for a given team has been of interest for quite some time and has still not been resolved. Studies that examined the use of logical arguments found that logic was not solely related to success in debate. studies also demonstrated that evidence was not solely related to debate success either. Despite debate's continued reliance on evidence and logical appeal as the basis of success in debate, research has failed to support either evidence or logic as being significantly related to success in debate. Some of the studies that investigated academic debate in the area of logical processes included research (Allen & Kellermann, 1988) that found that disadvantages were not persuasive in real world terms. Yet, disadvantages were also seen as being decisive in deciding the round (Allen, 1987). Similarly, rate of speech, use of jargon and the amount of evidence was found to be unrelated to academic debate success (Vasilius & Destephen, 1979). These questions will require equally an empirical, quantitative effort to understand what transpires in the debate round. Word Choice and Debate In examining the existing research specific to academic debate, only a minor amount examining the use of language exists; nevertheless, it has provided interesting results. Four published articles (Giffin,1959; Williams and Webb,

PAGE 31

1964; Williams, Webb, & Clark, 1966; and Rouse & Thomas, 1987) have dealt with the use of language in debate. 24 Giffin (1959) reported that the debater's use of language accounted for only 5.29% of the factors involved in the debate judge's decision; however, this author believes there were several methodological problems with Giffin's study. First, he used the debate ballot as the basis of the judging criteria. The 34 judges were instructed to "make your decision first, giving appropriate ratings and rankings; then we would like to have you indicate those criteria which you used in making your decision and to what extent each criterion was considered" (p.70). Giffin failed to explicitly state that the criteria were those from the debate ballot, but an examination of his methodology revealed that the criteria were from the ballot. The ballot constrained the judges to use only the criteria listed. Second, if language effects are subtle, there is no reason that the judges would self report those effects. Third, the study asked the judges to report the criteria that were used. Giffin only used the judge's self-report. The study did not examine any other variables. Fourth, Giffin explained that language was defined as only the "phrasing of concepts clearly and concisely" (p.70). Giffin's failure to report any significant language effect might have resticted further research if other scholars concluded that Giffin had demonstrated a lack of importance of language effects in debate. While Williams and Webb (1964) included the variables of word choice and

PAGE 32

25 the use of figurative language in their study, the results of the two variables were never reported. Williams, Webb, and Clark (1966) reported that the debater's vocabulary failed to load highly on any of the study's evaluative factors. Rouse and Thomas (1987) reported differences in the first affirmative rebuttal speakers' use of evidence and in their conformity with a supposedly ideal argument pattern. The study also discovered differences in the debaters' use of "road maps" (p.108) and word economy, but only examined the first affirmative rebuttals in each of thirteen debates and did not use any statistical analysis. Debate textbooks have paid little attention to the importance of language in debate. Only five out of twelve debate textbooks examined (Wood & Goodnight, 1989; Fryar, Thomas and Goodnight, 1989; Sanders, 1983; Freeley, 1976; and Patterson & Zarefsky, 1983) were found to discuss the issue of language effects and debate. Wood and Goodnight (1989) only suggested that debaters use language that would be most appropriate for their judge (p.280). Fryar, Thomas and Goodnight (1989) suggested that speaking clearly was much more important than speaking quickly. They also suggested practicing speeches to cut down the number of nonessential arguments. Practice would also allow the debaters to cut down on the nonessential words. Sanders (1983) wrote that the wording of issues was II .. extremely important ... and yet, something that

PAGE 33

26 is more often than not overlooked" (p.57). For example, he discussed the statement that, ... the states need more revenue" (p.57) and concluded that it had no impact. The effective alternative was" the states can no longer finance a viable public education system" (p.57). He also noted that the wording had to be both succinct and persuasive. Sanders also wrote that debaters should seek to avoid ambiguous wording of issues. Freeley (1976) discussed logical, ethical and emotional appeals without discussing the differences in the language that would make up such appeals. Freeley (1976) also stressed the importance of defining terms, stating that the ... advocates must carefully consider all possible definitions of all terms" (p.43). An extension of this argument would be for the advocates to equally consider their choice of language. Patterson and Zarefsky (1983) wrote that debaters have choices in language use, for example, the use of connotation. Patterson and Zarefsky offered the example of the verb "said," which is neutral. A positive connotation can be created by replacing "said" with "exclaimed," or it can be given a negative connotation by replacing "said" with "confessed." Patterson and Zarefsky noted that" ... in constructing a case, therefore, advocates should consider whether positive, negative, or neutral connotations are the most desirable and should word their arguments accordingly" (p.67). They also noted that debaters have a choice in the precision of the language, and the intensity of the

PAGE 34

27 language. Debaters have a choice of being precise or vague, and debaters can use irony or hyperbole, thus ranging from stating the reverse to large overstatement. However, Patterson and Zarefsky noted that either tactic was likely to backfire (p.67). The remaining choice that Patterson and Zarefsky discussed was style. Debaters could use straightforward speech but they could also utilize analogies, similes, and metaphors. Patterson and Zarefsky's discussion of language concluded that: the basic idea underlying all presentational choices is that language is not a neutral instrument, a vehicle in which the contents of argument are conveyed. Rather, language is an inseparable part of the argument itself and one that exerts great influence on how listeners perceive and react to the case. (p.68) Word Choice and Legal Research In terms of the use of language, the associated field of law provides some interesting research. For example, research (O'Barr & Conley, 1976; Parkinson & Parkinson, 1979; Conley, 1979; Parkinson, 1981; and Parkinson, Geisler & Pelias, 1983) investigating language choices and language effects in legal situations has found a relationship between language choices and success in trials. Some of these findings indicate that seemingly insignificant differences in language can be important. O'Barr and Conley (1976) noted that: some differences in courtroom language may be so subtle as to defy precise description by all but those trained in linguistic analysis ... New research on language used in trial courtrooms reveals that the subliminal

PAGE 35

messages communicated by seemingly minor differences in phraseology, tempo, length of answers and the like may be far more important than even the most perceptive lawyers have realized. (p.8) 28 O'Barr and Conley classified language along a "power language continuum" in which a powerless speaker's language contained a high frequency of hedges, such as "Perhaps or "I think ... ," repetition, intensifiers, such as "very close friends" instead of simply "friends," and a greater use of direct quotations (p.9). The article did not fully identify the characteristics of the powerful speaker. The study found that speakers speaking in the powerless mode were found to be significantly viewed as having less competence, believability, intelligence, assertiveness and likeability. Another study that investigated language in the courtroom (Conley, 1979) also reported a statistically significant difference between powerful and powerless speakers. This article reported the finding that speakers who utilized "hypercorrect speech" were found to be viewed as less competent, less convincing, less intelligent and less qualified than a speaker who used a more natural form of language. Additional research on speech tactics and success in trials has been conducted by the Duke University Law and Language Project. One resulting article (Parkinson & Parkinson, 1979) reported that a computerized language analysis of actual jury trials found that successful defense attorneys used fewer adverbs. They also used more vague or

PAGE 36

29 abstract language. The study also examined the speech of the prosecution and the defendants. Prosecution attorneys who won their cases were found to speak longer, make more indicative statements and ask more questions referring directly to a witness. Losing attorneys were found to have used more conditional language. These attorneys also used more careful and polite language. In contrast, the defendants who were successful were those who used more courteous language. The successful defendants also made fewer references to themselves and they made more grammatically complete sentences. Another study (Parkinson, 1981) from the Duke University Law and Language Project reported that a number of different speech behaviors were co-occurring with either success or failure in courtroom trials. The study focused on "message style, not message meaning'' (p.31). The speech behaviors differed for the speaker. What was successful for some speakers were not successful for other speakers. Prosecution attorneys were successful when they were verbose, verbally assertive, and when they referred specifically to the witness and used the pronoun "you." Defense attorneys were successful when using ambiguous or abstract language. They also used fewer adverbs, more legal jargon, and fewer afferent words relating to the five senses. Successful defendants were found to use more polite and courteous language. They also used more demonstrative language such as "this" or "that." The successful defendants used grammatically complete sentences.

PAGE 37

30 The same speech behaviors that were successful for the defendants were found to be associated with failure for the attorneys. Both demonstrative language and grammatically complete sentences were associated with failure in defense attorneys. Politeness and and hypercorrect grammar were also present when prosecution attorneys failed. In addition to the speech behaviors that were successful for defendants but unsuccessful for the attorneys, Parkinson (1981) found that prosecution attorneys who used conditional language such as "might" or "could" failed. used concrete words also failed. Defense attorneys who That finding was not surprising since successful defense attorneys used abstract language. Parkinson (1981) found statistically significant differences associated with the word choice and success/failure in trials. The study used a stepwise discriminant analysis and concluded that the method" ... was able to predict trial outcome from language behavior variables with an overall success rate of 77.08%" (p.27). Another study (Parkinson, Geisler, & Pelias, 1983) that also investigated language choices and trial success reported results similar to the Parkinson (1981) study. One difference was discovered in the use of nouns without physical referents such as "honor" and nouns with physical referents such as "automobile." Successful defendants and their attorneys used more nouns with physical referents while successful attorneys for the plaintiff used more nouns without physical referents. Plaintiffs were found to be successful when they were verbose and when they used more

PAGE 38

31 adverbs and adjectives. They were unsuccessful when they used more nouns without physical referents. The results were found to be significant through the use of multiple discriminant analysis. The study found that "the difference between a successful and unsuccessful courtroom performance may be only a few words per thousand" (p.21) and concluded that "there are significant, though subtle, speech characteristics which co-occur with trial success" (p.21). Thus, while a few studies have investigated word use in debate rounds, the results have been generally ignored. One exception was the study by Rouse and Thomas (1987) that did report some language differences. Research in legal studies has demonstrated clear-cut and strong effects of certain types of language on the winning and losing of trials. The fact that minute differences in word use were significantly related to success in courtroom trials should provide a mandate for investigating word use in debates. Both hinge upon the use of words to persuade a judge (or jury) to accept only one of two opposing sides. The present study examined the emotive words used by all four debaters in their rebuttal speeches. The purpose of this study is to conduct an analysis of the transcripts from nineteen of the final rounds of the NDT and determine the existence of and relative usage of emotive words and the relationship, if any, to five variables including result, side of resolution, speaker position, year of debate, and gender composition of the team.

PAGE 39

32 Empirical Focus The Dependent Variable Transcripts of nineteen years of intercollegiate debate from the National Debate Tournament were converted into word-processed form and analyzed by a software called "wordscan" that can count and extract words from sub dictionaries. The transcripts were compared to a list of negative and positive emotive words that were drawn from Gottschalk and Gleser's content analysis scale and from Osgood's semantic differential scale. These scales purport to reflect the attitudes and emotions of the respondents. Exemplars of emotive words are presented in tables 3-5 through 3-80. The Gottschalk-Gleser Content Analysis Scale The original scales were first developed by Gottschalk and Gleser in 1969. The scales have been widely used by scholars in a range of different fields (Gottschalk, 1986). The content analysis scales have been used to measure the affective states in African-American children (Uliana, 1979). The scales have also has been used to measure hostility (Gift, Cole, & Wynne, 1986), depression (Gottschalk & Hoigaard, 1986), alienation of drug addicts (Viney, Westbrook, & Preston, 1986), and the speech of mentally ill patients (Lebovits & Holland, 1986). Further, the scales have been used to conduct textual analysis (Nimicks, 1985) One problem with the use of the scale is that it takes considerable time to score the results, but Gottschalk

PAGE 40

33 (1986) noted that a computer should solve that problem. A number of scholars have used modified versions of the scales (Gottschalk, 1986). The scale has been found to be both reliable and valid (Gottschalk, Eckard, & Feldman, 1979). Viney (1986) surveyed the research on Gottschalk and Gleser's content analysis scale and reported that the range of reported interjudge reliability for the hostility scale was from 0.76 to 0.98 and'' .. that [the scales] are also consistently high ... (p.60). Viney (1986) also reported that since construct validity best reflects whether the scale is successful, that the" ... scale seems likely to have inherent content validity" (p.60) because the content is directly derived from the subject's communication. Viney (1986) also surveyed the literature of twenty years dealing with the scale's validity and with the hostility scale, and found that it was independent of age, educational level and sex. The scale was found to be significantly correlated with a number of emotive states (p.65) and was correlated with" ... other measures of the same construct ... (such as) self-reports and observations of behaviors .. (p.63). Gottschalk (1979) explained that the content analysis scale had" a set of construct-validation studies (that] had to be carried out to ascertain exactly what this content analysis procedure was measuring ... (p.548). He noted that the procedure could be applied to a variety of contexts such as" ... literature, public

PAGE 41

------------------34 speeches, and any other type of language material" (p.550) as long as the samples are based on a response to standard instructions and with equal temporal units. Debate could be easily analyzed through the procedure since the debaters operate within a framework of standard instructions and have identical temporal units. The scales have also been demonstrated to be valid and reliable for non-American subject groups including Germans (Schofer, Koch, & Balck, 1979; Koch, 1986), demonstrating a crqss-cultural utility for the instrument. Gottschalk (1979) wrote that the use of the scale in other languages was" ... evidence of the validity and universality of the procedure ... (p.39). Gottschalk (1979) wrote that content analysis was the best system available for measuring emotive words: Reliable and valid measurement of affects, emotions, and moods have posed a problem for psychiatric and psychophysiological research as the demand has grown for more sensitive, precise, and objective assessment methods than the method of clinical impressionistic evaluation. There are three major methods in current use for assessing these psychological variables: self report scales, behavioral rating scales, and the content analysis of verbal behavior ... [which] can avoid most of the shortcomings of the self-report and observer rating methods ... (p.541) The Semantic Differential Scale Charles Osgood, George Suci, and Percy Tannenbaum (1957) created the semantic differential scale as an instrument to describe major dimensions along with varying judgments and meanings. The scale was composed of word pairs such as "good-bad," "kind-cruel," "peaceful ferocious" and "happy-sad." Subjects were asked to select one word from each pair. Staats and Staats (1969) pointed

PAGE 42

35 out that the word's meaning involves a psychological process that is different from word association processes. The scale has been widely used and it has also been modified. Weinreich (1969) expanded the scale with the use of Roget's Thesaurus (p.118). Cliff (1969) utilized adverbs as multipliers. The scale has been reported as reliable (Solarz, 1969) and valid across cultures (Tanaka, Oyama & Osgood, 1969). Osgood, Suci, and Tannenbaum (1957) reported a .85 test retest correlation coefficient. They also reported that the scale had face validity and that" throughout our work with the semantic differential we have found no reason to question the validity of the instrument on the basis of its correspondence with the results to be expected from common sense ... (p.141). Validity and Reliability The reliability of the current study is easily answered by the computerized emotive word list(s) and the computerized Wordscan software that counts the presence of the words. For example, "opportunity" is an emotive word. "Opportunity" was used five times by the first affirmative side speaker in 1979. Wordscan would always count five uses of "opportunity" when checking that speech. Reliability becomes perfect, as Lewis-Beck in Weber (1990) explained once the text is computerized, say with an optical scanner, it is relatively easy to make a classification from more than one dictionary. Moreover, with computers, the coding rules are necessarily made explicit, allowing for

PAGE 43

36 perfect 'interceder reliability"' (p.5). The present study did use an optical scanner, an emotive word dictionary, and software that allowed the presence of emotive words to be consistently counted. The reliability was 1.00. The validity is based on the original validities of the content analysis scale and the semantic differential scale. Since both of those scales had established validities, the use of the emotive word list should be valid as well, based on the notion of concurrent validy. Concurrent validity uses an established validity of an existing measure to provide validity for a new measure (Walizer & Wienir, 1978). Research Questions The present study examined the use of emotive language in academic debate. The central research question was: Research Question One Do winning and losing debate teams differ in their use of emotive words during rebuttal speeches in a debate round? Related research questions are: Research Question Two Do affirmative and negative debaters differ in their use of emotive words in rebuttal speeches? Research Question Three Do the first and second rebuttal speakers for each debate team differ in their use of emotive words?

PAGE 44

Research Question Four Does the quantity of emotive words used by debaters vary over a period of years? Research Question Five 37 Does the quantity of emotive words used differ between mixed-gender debate teams and single-gender debate teams? (See note 10)

PAGE 45

38 Notes 1 The first affirmative speaker presents the affirmative case. The first negative speaker usually presents the negative philosophy and often directly clashes with the affirmative case structure. The second affirmative speaker responds to the negative attacks and rebuilds the affirmative case. The second negative speaker usually offers the off-case attacks, so-called because the arguments do not directly clash with the affirmative. Disadvantages are examples of off-case arguments. The first negative rebuttal responds to the second affirmative responses. The first affirmative rebuttal must respond to both second negative speaker and the first negative rebuttal speaker. This speech is crucial because the rebuttal speaker has five minutes to respond to fifteen minutes of attacks. The second negative rebuttal speaker responds to the first affirmative rebuttal speaker and offers the final rational for a negative ballot. The second affirmative speaker responds to the second negative speaker and offers the final rationale for an affirmative ballot. 2 Scott Deatherage (1985) explained in his ballot in the final round of the NDT that he voted negative as a result of 11 a brilliantly conceived"(p.55) second affirmative rebuttal that fell" ... just short of covering for earlier time allocation mistakes" (p.55). 3 The hypothesis-testing model is centered on the belief that the debate process is similar to the testing of a hypothesis in a laboratory test. The debate is a test of the resolution alone, which would allow the negative to "defend anything or everything that is non-propositional" (Cross & Matlon, 1978, p.111). The chooser of policy systems is a debate paradigm in which the debate teams are similar to a legislative body that evaluates competing systems which solve the same basic problems. The policy system paradigm has been very common in debate. The affirmative team usually offers a problem and solution while the negative usually would attempt to demonstrate that the current existing system could solve the problem and that the affirmative solution would not work and/or that it bring about worst harms. The evaluator of argument skills paradigm would have the judge decide the round on the basis of which team utilized better analysis, evidence, reasoning, organization, refutation and delivery. Tabula rasa, Latin for blank slate, assumes that the judges who use this paradigm are completely open to any theory or approaches in the debate round. The stock issues paradigm dates from ancient times (Cross & Matlon, 1978, p.112). This paradigm contains four elements: (1) The significance of the issue(s); (2) The

PAGE 46

----. ---39 inherency of the issue; (3) The issue of solvency or solution; (4) The issue of disadvantage. The four issues are approximately the issues of ill, blame, cure, and cost (Cross & Matlon, p.112). The legal paradigm assumes that the judge would evaluate the debate by the use of the legal principles such as precedent. However, a study (Cross & Matlon,1978) that examined the judging philosophies at .the most important debate tournament, the National Debate Tournament (NDT), did not find any judges who used the legal model. 4 The written explanations of the judge are called ballots and are usually on standard forms. Most debate tournaments and therefore most debate judges use ballots obtained from the American Forensic Association called the Form C ballots. 5 However, there exist different standards that the debaters must adapt to. The nature or personality of the judge is also critical to debate. The judge must "decide on the substantive merits of the question in dispute" (Patterson & Zarefsky, 1983, p.105). The judge must be unbiased; he or she will need to decide the question based on their prior background and beliefs as well as the specifics of the debate itself. This is also influenced by the judging paradigm in the debate round. The tabula rasa (blank slate) judge tries to remove the prior background and beliefs and only judge the round on what actually has transpired in the round. 6 The first explanation for the sparse research is that the existing behavioral or quantitative research is fragmented. One survey (Anderson, 1974) found that almost 30 percent of all existing bibliographical entries on behavioral research in debate were unavailable at the nation's third largest library at the University of Illinois. In 1991, The Chronicle of Higher Education Almanac ranked the University of Illinois at Urbana Champaign as the fifth largest library in the United States and Canada.This unavailability was the result of lack of indexing for both important journals and the forensic honoraries. As Anderson points out, 11 many of the early empirical studies and most of the current ones are published in these journals" (Anderson, 1974, p.148). A related problem is that different operational definitions are often utilized for similar phenomena, so that comparisons or generalizations across different studies is difficult. The second explanation for the lack of research is that the field is not oriented towards its own meta-research. There appears to be little interest within the field in conducting research involving debate or debate processes. One suggestion (Cox, 1976) is" ... that we are so intensely involved in participation that seldom do we study and professionally evaluate the discipline for the benefit of others"(p. 57). Debate coaches might respond that their

PAGE 47

40 preparation of their teams for the rigors of the debate circuit would consume their time. Klopf and Rives (1965) wrote that 85 percent of the debate coaches felt that their professional and academic advancement was restricted" by directing forensics activities. The additional hours required to coach, administer, and attend tournaments prevents them from engaging in activities such as research and publishing" (p.36). One result of the restriction is that frequently debate coaches or the directors of forensic programs will leave those positions to concentrate on non debate research or teaching. Often, the debate job is considered unimportant. Kully (1972) explained that 11 at some institutions responsibility for the debate squad is considered a menial task for a junior faculty member or graduate assistant" (p.195). Academics in the field are confronted with two almost mutually exclusive choices, research and publications or debate tournament success. As Andersen (1974) explained, ... those conducting behavioral research in speech communication and persuasion are moving in one direction while those active in forensics and competitive debate are moving in a different if not an opposite direction" (p.149.) Research thus needs to focus more on bringing the two directions together. The third explanation for the lack of research is that while the field of speech communication has undergone significant changes, debaters have remained much the same. When departments of speech communication were heavily centered on rhetoric, debate was more central to the departments. But as departments became more behavioral and experimental, debate became less important. As debate becomes less important, the need to conduct research is reduced. The direction of those active in competitive debate is readily shown by the field's foremost academic journal. The Journal of The American Forensic Association was founded in 1963 and originally focused on both debate and the individual events in forensics, such as persuasive or impromptu speaking. The journal, usually known by its initials, JAFA, deals with the whole range of interest in forensics, including descriptive articles, theory articles and some quantitative or behavioral research articles. However, debate theory articles came to dominate the journal at the expense of research. After a questionnaire surveyed the membership of the American Forensics Association, the journal was renamed in 1988 as Argumentation and Advocacy:The Journal of The American Forensic Association. The professional associations have ignored research as well. The American Forensic Association waited 20 years before creating a committee on research (Walwik, 1969). The focus of the professionals in the debate field is largely on theoretical articles that further develop the practices in the debate round, not on research that might help explain the results of the debate round. 7 The existing quantitative research in academic debate is best categorized in three areas (Anderson, 1974)

PAGE 48

41 (1) logical processes; (2) personality or characteristics of debate participants; and (3) judging procedures. Each of the three areas will next be discussed. Logical processes research examinines the relationship of the addresser to the addressee. Debate has been based on the belief that it is a logical process. The research in the area of logical processes has examined the effects of evidence, investigations of logical and/or emotional appeals, research on reasoning, one-sided versus two-sided appeals, and receiver variables. Stanley Rives (Dresser, 1964) was quoted as explaining in his address to the Midwest Forensic Association that, ... debate is primarily an exercise in reasoning and evidence ... The debater, coach or judge who concerns himself with academic debate agrees to concern himself primarily with reasoning and evidence, to accept these as the basis of an intelligent decision"(p.101). Some of the research that would be included under logical processes includes debater's use of evidence (Larson & Giffin, 1964), the use of evidence in ten championship debates (Dresser, 1964), the use of evidence in one championship debate (Newman & Sanders, 1965), the dilemma of ethics in the use of evidence (Rieke & Smith, 1968), the impact of evidence on decision making (Dresser, 1966), the rhetoric of evidence (Gregg, 1967), the relation of logic to argumentation (Petrie, 1969), the logic of evidence (Kellermann, 1980), the effectiveness of NOT final round disadvantages (Allen & Kellermann, 1988), and the effect of various time limits on the quality or effectiveness of rebuttals (Mayer & Meldrum, 1987). Three different studies (Larson & Giffin, 1962; Dresser, 1964; Newman & Sanders, 1965) examined the evidence used in debates and have found a large proportion of such evidence is unverifiable, misrepresented or fabricated. Larson and Giffin (1964) examined the evidence used in four randomly selected debates from the prestigious Heart of America Tournament at the University of Kansas in 1962. Larson and Giffin found that only half the evidence was validly represented. They reported that forty-two percent of the evidence were unverifiable; five percent was out of context; only three percent was manufactured or quoted from nonexistent sources. Dresser (1964) examined the use of evidence in ten championship debates. He examined the final rounds of the National Debate Tournament for 1950, 1952, and 1955 through 1962. Dresser investigated the evidence for clarity of documentation, recency of evidence, competency of sources and the closeness of the evidence to the debater's claim. The study found that most debaters used "evidence of opinion" (p.106) and that the debaters were clear about the qualifications of their sources but much less so about the recency of the sources. Another study (Newman & Sanders, 1965) investigated the use of evidence in a single final round of the National Debate Tournament. They found that of 71 pieces of evidence read in the round, three were fabrications, 23 were misrepresentations, and 6 were unverified. Such findings

PAGE 49

42 were troublesome to the authors and created a demanded for improvement in the use of evidence. Fabricated or misrepresentated evidence could easily change the results of a debate round. The importance of accurate evidence has been well known for a long time. Carney Smith (1937) wrote in the Quarterly Journal of Speech that "a single word changed or omitted in a quotation may go unnoticed by the other team, yet may alter the meaning of the entire statement" (pg.8384) A study that focused on the use of evidence and ethics was conducted by Rieke and Smith (1968). They wrote that the competitive nature of academic debate was both its strength and weakness because the debater is expected to be both a fervent champion of their cause and .. a judicious participant in rational decision-making" (p.228). Rieke and Smith argued that these two duties are contradictory. They suggested that it is difficult for debaters to advocate their position with maximum energy and to also be" ... committed to rational methods rather than victory at all costs" (p. 229). One result of the contradictory duties was an increase in unethical or questionably ethical practices. This study conducted a survey of debaters regarding ethical practices. The authors concluded that" ... an extraordinary contradiction exists between what is done and what debaters perceive as being done" (p.227). They wrote that academic debate had failed because of the unethical practices. The impact of the results were argued to go beyond academic debate because so many future lawyers and other professionals are student debaters. Several studies (Dresser, 1966; Petrie, 1969; Kellermann, 1980; Allen & Kellermann, 1988) have examined the relationship of evidence useage in debate to logic and decision making in debate. The consensus of the studies is that evidence does not influence the decision process in debate and the use evidence is not logical. Debate arguments have been found to be not logical as well. Arguments such as disadvantages that have been successful in debate were found to be not persuasive in real world terms Dresser (1966) examined the impact of evidence on decision making. This article reported that the common sense view that a speaker who uses facts to support assertions would be more likely to influence his audience was not true. Research indicated that the quality of evidence, amount of evidence or the identification of evidence 11 simply did not make any significant difference to their listeners" (p.40). Dresser did not suggest what elements would make any significant difference to the listeners. Gregg (1967) amplified this argument when Gregg examined the rhetoric of evidence and concluded that logical pattern of argumentation fails to consider the personalities of the interactants, it neglects the subject matter's values, and it assumes that the judge would make a careful examination of the evidence. Petrie (1969) investigated the relationship of argumentation to logic. Petrie noted that while no one would deny that logic and formal argument might sometimes change

PAGE 50

43 belief, ... the question remains, however, as to the precise manner of such causal efficacy .. or logic may cause belief through external accidental' features of formal argument such as style or impressiveness, regardless of the formal correctness of the argument" (p.55). Petrie never resolved either the issue of the mechanism of logic or the issue of any external features such as style. Kellermann (1980) noted that the research on evidence .. has produced such inconsistent results that no coherent theoretical perspective on the usefulness of evidence can be extracted" (p.159). Kellermann concluded that the poor results are possibly due to variations in audience, source and message variables. Allen and Kellermann (1988) investigated the effectiveness of four NDT final round negative disadvantages. This study noted that negative disadvantages, which are those arguments that the negative suggests would result if we believed the thesis of the affirmative case. The judge's belief is the result of the judge's paradigm and the interaction of the four debaters and their argumentation. Judges sometimes have found themselves voting for arguments that they personally do not believe but that the judge "believes" in the context of the debate round. The disadvantages were described as having a common characteristic of being "high impact/low probability", defined as having severe consequences but a rather low likelihood. Allen and Kellermann (1988) explained that" . [the] reason such arguments have popularity is that judges vote for them"(p.93). The popularity of these disadvantages is due to the fact that they often present clear cut choices in weighing the debate round. Judges are often faced with both sides in the debate round claiming global destruction which, for example, might force the judge to compare full scale nuclear war with a runaway greenhouse effect. The disadvantages are themselves the subject of controversy. They were described by Allen (1987) as having a significant value because they represent "real world" arguments and debates over real policy decisions. But other scholars (Hollihan, Riley & Baaske, 1985; Rowland, 1986) suggest that these arguments fail to reflect real world policy decisions, and that they have application only in academic debate. Howe (1981) argued that these types of arguments have no truth value and that judges should vote against them because they are bad and not truthful arguments. The Allen and Kellermann (1988) study presented four disadvantages from NDT final rounds to 229 undergraduate students. The disadvantages were taken from the transcripts in JAFA. Three of the four disadvantages were the deciding issue in those rounds by the majority of the judges. Allen & Kellermann (1988), noted that" ... these arguments were selected because they had been evaluated as good arguments presented by good teams. These should be high quality arguments and represent the best of academic debate; in other words, the debate community had already established that these arguments had worth"(p.101). However, the study found that while the disadvantages had some level of argumentative acceptability,

PAGE 51

44 the disadvantages were not judged to be persuasive in real world terms (p.104) when presented to students. An investigation of the effect of various time limits on the quality of rebuttals (Mayer & Meldrum, 1987) found that the time limit for a rebuttal did not affect the rebuttal's quality. This study utilized five minute rebuttal speech debates and compared them with seven minute rebuttals. Debates were found to not give any more explained qualitative arguments, rather they gave more unexplained quanitative arguments. It was also found that the rebuttal speeches were different in quality from each other. They found that the last rebuttalist gave the better speech. One possible reason is that the final speaker has a better, more holistic view of the debate (p.164). Quality in this study was measured as the effectiveness of the speech. Overall, the logical processes research demonstrates that logic and evidence are only part of the reason for arguments being accepted by judges. Another explanation might be message variables. The research also demonstrates that the dual roles of passionate advocate and rational decision-maker were contradictory and lead to ethical abuses that included fabricated evidence. Finally, disadvantages that debate judges had used to justify their advantage were not found to be persuasive in real world terms. Consistently, the research indicates that there are other factors that influenced decision making besides evidence or logic. One of these factors is related to the debater's personality and characteristics of the individual debate round. Several research studies in academic debate have examined the personality and other related variables of debaters. This area of research has also examined the interface between personality characteristics and participation, the descriptive studies of debate rounds, and comparison of the different activities within forensics (Anderson, 1974). Not all of the research that is described under the personality category is explicitly related to personality. The range of this research is another example of the scattered research in debate. Some research involves the changes that debaters undergo or the difference between debaters and non-debaters. This research includes a survey of the participants in the NDT from 1947 to 1980 (Matlon & Keele, 1984), a study of the differences between debaters and nondebaters (Stewart & Merchant, 1969), a study on the image of the debater (King & Phifer, 1968), a study on the effect of debate students in the bicentennial debates (Semlak & Shields, 1977), a study (Colbert, 1987) that examined the effects of debate training on critical thinking ability, a study of the competitiveness of the debaters (Wilson, 1979), and a study that investigated debaters' ability for interpersonal and concept compatibility (Allen, 1963) Other research has examined characteristics of the debaters including the debater's nonverbal decoding abilities (Barker, 1965; Sayer, 1974). Three studies have

PAGE 52

45 also investigated gender differences (Hayes & McAdoo, 1972; Hill, 1973b; Rosen, Dean, & Willis, 1978). Several studies examined behavior or characteristics of the debaters while in the debate round. These studies include one which examined the evaluation criteria as predictors of debate success (Burgoon, 1975), one that investigated the debater's dimensions of credibility (Burgoon and Montgomery, 1976), one study that examined the relative effectiveness of praise and derogation as persuasion strategies (Burgoon, Wilkinson, & Partridge; 1979), one investigation of the relationship between debate success and rate, jargon, and evidence (Vasilius & Destephen, 1979) and one content analysis of 13 years of the first affirmative rebuttal speeches from the NDT (Rouse & Thomas, 1987). To characterize debaters, a survey of 703 debate participants in the National Debate Tournament from 1947 to 1980 (Matlon & Keele, 1984) found that the two most common major fields of study debaters engaged in were political science, with 24 percent, and speech communication, with 20 percent. The next most common were history at 12 percent, economics, at ten percent, English, at six percent, philosophy, at five percent, and business administration, at four percent. These debaters went on to obtain advanced degrees with 633 having at least one advanced degree and 209 having two or more advanced degrees of the 703 people surveyed. Matlon and Keele reported that they found" ... that some of the most intellectually gifted students have participated in the National Debate Tournament" (p.195). After obtaining the advanced degrees, the debaters were found having advanced professional positions of significant responsibility, such as cabinet members, congresspersons, ambassadors, judges, and other high status professional positions (p.195). Several respondents compared the NDT to the Super Bowl, the Master's Golf Tournament, or to the World Series (p.200). The most salient finding of this study was that there was a significant relationship between the goals of higher education and the participation in competitive debate. The authors concluded that "We are almost hesitant, therefore, to conclude that the results support debate participation almost without exception" (p.205). Several studies (Stewart & Merchant, 1969; King & Phifer, 1968; Semlak & Shields, 1977; Colbert, 1987; Wilson, 1979; Allen, 1963) have examined the perceived differences between debaters and non-debaters. Some of the findings indicate that debaters have been found to be more competitive, have better delivery skills, and demonstrate superior critical thinking skills. Stewart and Merchant (1969) examined perceived differences between debaters and non-debaters. An audience that observed both debaters and non-debaters speaking on the same topics was able to identify the debaters 90% of the time. The debaters were scored higher on every speech scale except ''fair-unfair"(p.72). Stewart and Merchant concluded that experienced debaters overwhelm non-debate audiences.

PAGE 53

46 King and Phifer (1968) examined the public image of debaters. The study used a semantic differential scale that was administered to both debaters and nondebaters. A factor analysis indicated that debaters were viewed as" .. effective, interesting, original, and convincing" (p.51). The study also found no significance difference between the debaters and nondebaters. Semlak and Shields (1977) examined the effect of debate training on those who participated in the bicentennial youth debates and found that the students who received debate training significantly improved their abilities in analysis, delivery and organization (p. 195). Delivery consisted of both the manner of the speaker as well as the speaker's language choice. Colbert (1987) examined the effects of debate training on critical thinking ability and found that debaters scored higher than non-debaters on the Watson-Glaser Critical Thinking Appraisal, a measure of critical thinking. However, the question of whether debate produced such changes or merely attracted those students who already would score higher in critical thinking was not answered. Colbert found a difference between the debaters trained in NOT debate and those trained in the newer CEDA (Cross-Examination Debate Association) debate, but the quality of the difference was not present (p.200). Colbert found a difference but he could not identify it. Wilson (1979) conducted research into the competitiveness of intercollegiate debaters and found that debaters are significantly more competitive than non debaters and that most of the variance was due to one factor, toughmindedness. This study used cattell's sixteen Personality Factor Questionnaire, a self-report instrument. Allen (1963) assessed two different variables of debaters. First, the debater's "interpersonal compatibility"(p.67), the ability to work well with another, was measured by the use of Schutz's FIRO-B scale. The second variable was the debater's "concept compatibility"(p.72), the ability to translate a single concept into an internally consistent debate. This was measured by the use of a semantic differential. Allen concluded that debate teams in which the debaters were both interpersonally and conceptually compatible would" ... more efficiently interact in the preparation of debate cases" (p.26). The finding that debaters who get along with their partners and who approach the topic in a compatible fashion are more effective would not be very surprising to many debate coaches. Two different studies (Barker, 1965; Sayer, 1974) investigated debater's ability to decode nonverbal messages. Debaters have been found to be poor at decoding nonverbal cues but they have the potential to make correct assessments. Barker (1965) examined debater-judge ratings and found that judges and debaters" ... do not rate the debaters' performances the same" (p.19). The results offered empirical proof that debaters are poor at deciphering nonverbal communication from the judges. Sayer

PAGE 54

47 (1974) wrote that Barkers' findings supported the old debate maxim that" ... debaters possess the worst perception of what has happened during any round of debate" (p.5). One reason for the debater's inability to correctly understand the judges nonverbal cues is that debaters are usually too busy, either preparing for their next speech or writing down the other team's speech. However, debaters do have the potential ability to understand the judges' nonverbals. Sayer (1974) conducted a study in which debaters were asked to predict the judge's decision" ... based upon their observation and interpretation of the nonverbal cues" (p.4). The debaters were correct at predicting the decision 66.5% of the time. Sayer concluded that debaters have the potential to correctly make nonverbal assessment. This study worked because it asked the debaters to make a prediction that forced them to pay more attention to the nonverbal cues. Another group of studies examined the behavior of the debater during the debate round. These studies examined the evaluation criteria (Burgoon, 1975), debater's credibility (Burgoon & Montgomery, 1976), use of praise and derogation (Burgoon, Wilkinson, & Partridge, 1979), use of rate, evidence and jargon (Vasilius & Destephen, 1979) and a content analysis of thirteen years of the first affirmative rebuttal speeches from the NDT (Rouse & Thomas, 1987). A study that examined the evaluation criteria as predictors of debate success (Burgoon, 1975) examined the six criteria on the American Forensics Association Form C ballot, the most widely used debate ballot. The six criteria were: organization, refutation, analysis, reasoning, evidence, delivery, and success. The six criteria were thought to be a quantitative measure of the debater's personality. The study found that all six factors were related to the debaters winning the debate. The analysis demonstrated that the judges did not make discrete differentiations among the six factors. The suggested reasons for this result were that the judges were making a gross or global evaluation or that there were other factors that influenced the results. Burgoon did not speculate regarding what the other factors could be, but one factor could be the language used by the debaters; the debaters might use language containing emotionally loaded words which would influence the judge's emotive response. Burgoon, along with Montgomery, (1976) also investigated the debater's dimensions of credibility. They received 186 completed questionnaires from debate coaches throughout the United States. The questionnaire had 47 items, utilizing a seven interval semantic differential instrument which asked the coaches to give their opinions of an ideal debater's credibility. They found that three factors accounted for 51 percent of the variance. The first factor, which accounted for 48 percent of the explainable variation, was "task competencies" (p.174). This factor included qualities such as expert, active, articulate, informed, coherent, competent and impressive. This factor is a combination of the general quality of competence along

PAGE 55

48 with good delivery skills that would deliver the competence. Burgoon and Montgomery's second factor was "social competencies" (p.174), which accounted for 26 percent of the explainable variations and included qualities such as the ability to be open-minded, relaxed, organized, poised, good natured, perceptive, and sociable. Burgoon and Montgomery (1976) explained that this factor is related to the image of the person who controls both social relationships and himself (p.175). The debater is seen as both relaxed and self-assured. Burgoon and Montgomery's third factor was "assertiveness" (p.174), which also accounted for 26 percent of the variance, including the qualities of aggression, headstrongness, criticalness, extrovertedness, verbal ability, and experience. This third factor is based primarily on the quality of extroversion along with the quality of being critical (p.175). The verbal quality could include the use of language by the debaters. The finding that the debater's credibility is based on three factors or dimensions was significant (p.175). Burgoon and Montgomery (1976) argued that judges view debaters through these three factors rather than the six factors along the Form C ballots (p.176). A second significant finding was that character, honesty and ethicalness were not significant factors (p.176). The third finding by Burgoon and Montgomery was that there was little common agreement among the judges for a single salient factor (p. 176). This finding would suggest that judges view debates in a number of different ways and that there are almost no universally accepted qualities that are regarded as paramount. The overall finding of Burgoon and Montgomery (1976) would serve to undermine much of the research directed at personality and debate. A study that examined the relative effectiveness of praise and derogation as persuasion strategies (Burgoon, Wilkinson, & Partridge; 1979) found that language of praise was more favorable than derogation. This study also examined the gender of the speaker as it interfaced with the type of language that was chosen. They found that it was least expected for a male to derogate another male and most expected for a male to praise another male. The study found that females derogating males with nontraditional receivers were more effective than males derogating females, which was contrary to the researcher's expectations. Also contrary to their expectations, males were not rated higher on credibility than female speakers. Vasilius and Destephen (1979) conducted an investigation into the relationship between debate tournament success and rate of speech, evidence, and jargon. Vasilius and Destephen (1979) found that fasterspeaking teams did not have greater success than slower speaking teams; teams that used a greater amount of evidence were not found to be more successful than those teams using a smaller amount of evidence, and teams using a greater amount of jargon were not more successful than teams using little jargon. Jargon was defined as the specialized debate

PAGE 56

49 language. A relationship between rate and jargon was found; those teams who spoke faster were more likely to use jargon. The study suggested that these factors do not contribute to debate success and concluded that further research would be needed to identify those factors that do contribute towards success. Overall, the research in debate regarding personality and characteristics of debaters has been varied. There have been a number of important results: (1) Debate produced changes in the participants. They become significantly improved in their analysis, delivery and organization. They demonstrated better critical thinking ability and better debaters demonstrated better interpersonal and concept compatibility. These changes proved to be valuable for debaters in future professional careers. (2) Debaters have poor nonverbal decoding skills but they can learn to improve those skills. (3) While there have been no clear differences between male and female debaters, the results of two of three studies suggests that mixed gender teams do better than single gender teams. (4) Variables such as task competency, social competence and assertiveness, were found to be related to debater's credibility. Faster speaking debaters and debaters who used more evidence and more jargon were not necessarily more successful in debate. (5) A study of first affirmative rebuttal speakers reported that the debaters' arguments varied from the ideal, varied in rate and varied in substantiation, and that the speeches changed over time. 8 They quoted Clough in 1972 as making the argument that "Solvency. One, lack of flexibility and defense. This is not a plan meet need; it is a disadvantage couched as one. It has nothing to do with the case we present to you. But I will suggest that we can get the military. First, the volunteers are right now" (p.109). This was contrasted with Dripps who argued in 1979 that "Three, empirically it's false, twenty-eight amendments; Constitution not destroyed" (p.109). 9 Judges have been found to use different judging styles or paradigms to evaluate debate. One study (Dunne, Mack, & Pruett, 1971) suggested that judges utilize either a logical (issues) or a proficiency (skills) paradigm to evaluate the debate. Two other studies (Cox, 1974; Cross & Matlon, 1978) reported that judges utilize five different paradigms. Those two studies agreed on: (1) chooser of policy system, (2) evaluator of argument skills; (3) tabula rasa; and (4) evaluator of stock issues. One study (Cox, 1974) suggested the fifth paradigm to be a critic of rhetorical proposition while the other study (Cross & Matlon, 1978) suggested that it was a hypothesis tester. Other studies (Hample, 1979; Crable, 1976) have suggested a legal model. (Also see footnote 3). Dunne, Mack, and Pruett (1971) conducted an analysis of judging styles in debate. Judges were classified as either logical (issues) or

PAGE 57

50 proficiency (skills) judges. Only about half the judges sampled agreed with the logical-proficiency dichotomy. Of those judges who did agree, a majority favored the logical judging paradigm. One possible reason for the large number of judges that rejected the logical-proficiency paradigm was that five different judging paradigms had emerged (Cox, 1974) where the logical paradigm overlapped with four different paradigms and the proficiency paradigm became the evaluator of argument skills. Cox (1974) conducted an analysis of judging philosophies in academic debate and reported that judges viewed themselves primarily as reflecting one of five judging philosophies: (1) chooser of policy systems with 42.9% of the judges reporting that they saw themselves as primarily using this philosophy, (2) evaluator of issues with 32.1%, (3) evaluator of argument skills with 15.5%, (4) critic of rhetorical proposition with 3.6%, and (5) tabula rasa with 5.9 % (p.62). Cox also found that many judges expressed willingness for the debate to include theoretical issues, including the judging paradigm to be used in the debate round. Similar to Cox's study, Cross and Matlon (1978) conducted an analysis of judging philosophies in academic debate and concluded that five philosophies prevail, including: (1) The hypothesis testing model; (2) The chooser of policy systems; (3) The evaluator of argument skills; (4) The Tabula Rasa judge; and (5) The stock-issues judge. The stock-issues is much the same as the evaluator of issues in Cox's study. It is also noteworthy that Cox's critic of rhetorical proposition was replaced by the hypothesis tester. Cross and Matlon reached two important conclusions. First, they found that" ... the majority of judges in the academic debate community view debates with extraordinary consensus regardless of their stated judging philosophies" (p. 123). This finding is in contrast to the results of the Burgoon and Montgomery study. The second finding was that judges appear to accept new theory arguments if the theory is forcibly and convincingly defended by the team that originated the argument (p. 123). Four different studies (Mccroskey & Camp, 1964 ;Williams & Webb, 1964; Williams, Webb, & Clark, 1966; and Williams, Clark, & Wood, 1966) investigated the criteria employed by the debate judge in making a decision. Mccroskey and Camp (1964) examined stock issues, judging criteria, and the debate decisions. The results included the finding that the "need" issue or the importance of the problem was paramount. A second finding was that the judge usually determines the central issue during the constructive speechs. The successful debaters were those who came to an agreement on the stock or central issues. Mccroskey and Camp further noted that debaters were unable to render an objective decision about the outcome of the rounds they had participated in. This final finding supports the forty percent disagreement that King and Clevenger (1960) reported.

PAGE 58

51 Bauer and Colburn (1966) wrote that judges differ in their decisions. Four reasons for the judges' differing were offered: (1) the contest was a close debate; (2) the judges used different evaluation criteria; (3) the judge was less competent due to a lack of knowledge; and (4), an unethical decision was made (pg. 23-24). One measure of determining the rationale for the judge's decision is through the judge's ballot. The ballot serves to provide direct feedback to the debaters and their coach. Ellis and Minter (1967) wrote that 11 one can never directly test the validity of debate decisions, since there are no absolute criteria against which to judge them, but the validity of decisions can be indirectly investigated by seeing how consistent the decisions are from judge to judge" (p.53). One of their findings regarding consistentcy was that 27% of the judges might be inconsistent because of travel fatigue (p.56). Lewis and Larsen (1981) conducted a study of forensic judges in individual events that examined interrater agreement and found that additional training significantly increased inter-judge agreement, and also found that 11 the present study indicates that practical results can be achieved by applying empirical research methods to the judging of forensics activities" (p.16). Of course, interjudge agreement may be affected by bias. Three studies (Mccroskey & Camp, 1966; Brooks, 1971; and Hill, 1973) investigated judging bias. Investigations regarding bias has centered around either a bias regarding the topic or a geographic bias. Mccroskey and Camp (1966) asked 95 debate judges of their bias regarding the topic. They found no bias by judges regarding the topic. They reported an earlier study by Scott found no bias and they concluded that 11 Finally, Scott's conclusion that judges' bias on the topic has no effect on their decisions was supported" (p. ) Bro6ks (1971) noted that ''an integral part of learning is evaluation and feedback" (p.197), and in debate the ballot is the feedback. Brooks attempted to find geographical bias in judges' decisions. The results supported some bias due to geographical distance and concluded that more research was needed in investigating the psychological and social distance. Part of the social and psychological distance between the debaters and the judge is that the judge usually h~s a holistic view of the debate round while the debaters only see their viewpoints. It is not uncommon for debate partners to have misperceptions about each other's stands in a debate round. Hill's dissertation (1973b) found no bias towards either the topic or the geographical proximity of the judge's and debater's schools. No study has supported any bias towards the topic by debate judges. The existing evidence on any geographical bias is less clear. One study (Brooks, 1971) suggests that some bias exists while a second study (Hill, 1973) found no support for a geographic bias. One reason for investigations into potential judging bias is because debaters rarely agree with the judge's decision.

PAGE 59

52 Three studies (King & Clevenger, 1960; Klopf, Evans & DeLozier, 1965; Hill, 1973) conducted comparisons of the debater and the judge's assessment. Not surprisingly, King and Clevenger (1960) found that debaters disagreed with the judges over the result of a given round almost 40 percent of the time. Klopf, Evans and De Lozier (1965) found that laymen, students, and other faculty could judge debate if their sole function was to make the win-loss decision. Debaters were found to not agree with the judges. One study, however, which did find a correlation between the judges' evaluations and the debaters' evaluations was Hill's (1973). Hill found that debaters had similar assessments to those assessments of the judges. Hill noted that his results were different from other research and suggested the use of classroom debaters might have caused the different result. Hill also speculated" . that involvement in tournament debating tends to destroy the individual's ability to accurately .. assess his own performance"(p.377). Three studies (Markgraf, 1966; Pearce, 1969; and Berthold, 1970) examined the debaters' ability to adapt to judges. Markgraf (1966) wrote that debaters feel confusion because of the absence of uniform standards of judging. Another problem was" ... the great discrepancy between professed ideals and actual debate judging" (p.38). Markgraf argued that the solution was for the judges to adapt to the debaters and not the other way around. While he did not give his perspective a name, Markgraf was advocating a tabula rasa judging paradigm. One possible method to reduce the disagreement between the debaters and judges would be for the judges to state their viewpoints. In support of this method, Verderber (1968) found that when judges state their criteria and preferences before the debate, the majority of debaters would adapt to the judges. Debaters also can learn the judge's viewpoints over time. Pearce (1969) reported that repeated exposure to a judge made" ... debaters more receptive to communication"(p.77) because the debaters could learn the judge's criteria. Debaters improved because they learned the judge's preferences and paradigm. Berthold (1970) noted that" ... most debate coaches and debaters would agree that the single most important item in any debate tournament is the judge's ballot. The ballot not only reveals which team won a given debate, but it also presumably indicates in a number of ways the participants' debate skills" (p.30). Berthold analyzed the C-form ballots that Burgoon (1975) also utilized. Berthold found that judges wrote more in the first round of a tournament but found that this was not related to the judges' overall performance. lO No all female teams in the sample.

PAGE 60

CHAPTER TWO METHODOLOGY Research Design The study investigated any effect of the use of emotive words by four speakers involved in NDT championship debate rounds. Operationally, the relationships between three dependent variables and five independent variables will be investigated. The three dependent variables will be the overall use of emotive words, and the uses of positive and negative emotive words. The five independent variables are: result of the debate round (winner or loser); side of the debate team (affirmative or negative); position of the debate speaker in rebuttal speeches (first rebuttal, second rebuttal); the year of the debate final (1967 to 1985); and gender make-up of the team (mixed or single-gender). Variables The dependent variable in this study was emotive words. Those words were selected from two original instruments, the Gottschalk-Gleser content analysis scale and the Osgood, Suci, and Tannenbaum semantic differential scale. Three dependent variables were identified in terms of the emotive 53

PAGE 61

54 word list: positive emotive words, negative emotive words, and the overall use of emotive words. Positive emotive words, negative emotive words, and the overall use of emotive words were further divided into the following categories: 1. The emotive words used by the first negative speaker in the first negative rebuttal speech, including the use of positive words, negative words, and total emotive words. 2. The emotive words used by the second negative speaker in the second negative rebuttal speech, including the use of positive words, negative words and total emotive words. 3. The emotive words used by the first affirmative speaker in the first affirmative rebuttal speech, including the positive words, negative words, and total emotive words. 4. The emotive words used by the second affirmative speaker in the second affirmative rebuttal speech, including the positive words, negative words, and total emotive words. The five independent variables will be: 1. Result of the debate round (winner or loser) as measured by having a majority of judges vote for the team. 2. Side of the debate team (affirmative or negative).

PAGE 62

3. Position of the debate speaker in rebuttal speeches (first rebuttal, second rebuttal). 55 4. Year of the debate final (1967 until 1985). 5. Gender make-up of the team (mixed or single-gender teams). Procedures Transcripts for Analysis The transcripts of the final rounds of the National Debate Tournament were published annually by the American Forensic Association in JAFA from 1967 until 1985. This study will analyze the transcripts of all of the rebuttal speeches from those debates. The transcripts of the rounds were made from audio tape recordings of the rounds. Rives and Boaz (1977) explained that II the debate was edited from a tape recording. Except for the correction of obviously unintended errors ... this is as close to a verbatim transcript as was possible to obtain from the recording" (p.10). Thirty-eight debate teams representing twenty different schools participated in the final round of the NOT between 1967 and 1985. Several schools made multiple appearances in the final round, including six by Harvard and four each by Dartmouth and Northwestern. A total of seventy-six debaters were involved in these debates but because five debaters appeared in two different final rounds, there was a total of seventy-one different debaters involved. Negative teams won eleven of the nineteen debates. Of the one hundred and nine

PAGE 63

56 judges, fifty-seven voted for the negative teams and fifty one voted for the affirmative teams. Of the topics, five were international in scope and fourteen were national topics. The sides, teams, topics and results of the final rounds from 1967 to 1985 are displayed in note 1. Subjects The study utilized the existing published record of nineteen years of the transcripts of the final round of the National Debate Tournament. A total of seventy-one debaters took part in these debates. Four women and sixty-six men debated in the final rounds between 1966 and 1985. A total of one hundred and nine judges were used to judge the nineteen debates. Materials The study utilizes three pieces of hardware: the optical scanner; the computer that will utilize the wordscan and other content analysis software, a Macintosh PC; and the computer that will conduct the multiple analysis of regression upon the results of the data, a mainframe VAX. The study utilizes the wordscan and browser software that was discussed above and the study will also utilize SAS, a statistical package available from the Center for Instructional and Research Computing Activities (CIRCA} at the University of Florida. The wordscan program examined the "text only" debate transcripts for emotive words. Those emotive words were taken from Gottschalk and Gleser (1969)

PAGE 64

57 content analysis scales (the hostility scale) and the semantic differential scales of Osgood, Suci and Tannenbaum (1957). Both scales contain bipolar terms that fall into positive and negative categories. Items from both scales were combined and additional words were derived from the original by the use of a thesaurus and grammatical derivations. Identifying Emotive Words The transcripts were processed by an optical scanner (Model: Apple Scanner 300dpi) that will convert the transcripts into a word processed form (MS Word 4.0). The scanned document was manually checked against the original to ensure its accuracy. The resulting document was divided into smaller documents: one document for each of the rebuttal speeches and two or three documents for each of the constructive speeches. This division was necessary for ease of analysis and to accommodate limitations of the Wordscan software. All documents were placed into a "text only" format that have all non-word elements such as punctuation removed. The resulting document(s) were examined by a word processing program called Wordscan that was developed by Durkee and distributed through the public domain by EduComp Services (Baez, 1989). The program compared the "text only" document to a separate dictionary and provided a display of any dictionary ideas that were contained in the document and the number of uses. In this study the dictionary is a list of emotive words that were derived from two other instruments, the

PAGE 65

58 hostility scale (Gottschalk, Winget & Gleser, 1969) and the semantic differential (Osgood, Suci, & Tannenbaum, 1957). The hostility scale was designed to measure attitudes and the semantic differential was designed to measure emotions. This study investigated the use of emotive words in academic NDT debate. From Gottschalk's hostility scale and Osgood's semantic differential scale, an aid has emerged to examine emotive words (See note 2). Another public domain word processor, Browser-developed by Zimmerman (Baez, 1989) allowed any word from a document to be displayed in context. Together, Wordscan and Browser examined the processed transcripts to determine the existence, types, amount and context of both negative and positive emotive words. The examination of context indicated whether or not the word use was anomalous. For example, the word was not used in a quote or the word was not used as a filled pause, one example is the word well. Content analysis scales such as the Gottschalk-Gleser content analysis method can be utilized through the use of a computer. Gottschalk, Hausmann, and Brown (1979) wrote that the use of the computer can save time, increase the uniformity of the analysis and aid in efficiency and rapidity. Viney (1986) stated that computerization is both feasible and cost-beneficial. Viney reported on three different computer-based scoring systems based on the Gottschalk-Gleser scales. Viney also underscored an ethical value to the use of content analysis:

PAGE 66

Insofar as content analysis is a technique for analyzing data and not for collecting it, it raises no ethical issues. Yet, because it can be applied to data collected in ethically appropriate ways, it has value. It provides a means of making sense of and rendering useful/ the information from a simple question posed to research participants such as 'I am interested in X. I'd like you to tell me about your experience of X.' (pp. 7576) 59 The participant's statement that Viney described could be examined for use of emotive language. The Wordscan software produced lists of the positive and negative words that the debaters utilize. Statistical Model The best statistical model for conducting the analysis is the use of the multiple regression model. Lewis-Beck (1980) explained that" ... multiple regression has great range and its mastery will enable the researcher to analyze virtually any set of quantitative data" (p. 47). This model allows formulation of the relationship between the many independent variables and the dependent variable(s). Agresti and Agresti (1979) also explained the value of the multiple regression model; it enables one" to develop a better predictor of a dependent variable than can be obtained by using only one independent variable. Also, these models allow one to analyze partial relationships between two variables, controlling for other variables" (p.322). Since the study utilized a number of different variables, the possibility that partial relationships existed between the variables might explain the existing inconclusive results in the literature.

PAGE 67

60 The formula that would be utilized for the multiple regression was: Y = Ln (P/1-P} =Bo+ B1X1 + B2X2 + B3X3 + B4X4 + B 5 x 5 + E The variables deal with the result (X 1 ), side (X 2 ), rebuttal position (X 3 ), year of the debate final (X 4 ), and gender make-up of the team (X 5 ) (See note 3).

PAGE 68

61 Notes 1 The participating debate teams, result, strength of the judges' decisions and the topic/resolution are listed: 1967 Dartmouth (negative) defeated Wayne State (affirmative) on a six to one judges' decision. TopicResolved: That the United States should substantially reduce its foreign policy commitments. 1968 Wichita State (affirmative) defeated Butler (negative) on a six to one judges' decision. TopicResolved: Should the federal government guarantee a minimum annual cash income to all citizens? 1969 Harvard (affirmative) defeated Houston (negative) on a five to two judges' decision. TopicResolved: Should executive control of United States foreign policy be significantly curtailed? 1970 Kansas (negative) defeated Canisius (affirmative) on a four to three judges' decision. TopicResolved: Should the federal government grant annually a specific percentage of its income tax revenue to the states? 1971 U. California at Los Angeles (affirmative) defeated Oberlin (negative) on a four to three judges' decision. TopicResolved: Should the federal government adopt a program of compulsory wage and price controls? 1972 u. California at Santa Barbara (affirmative) defeated Southern California (negative) on a four to three judges' decision. TopicResolved: Should greater controls be imposed on the gathering and utilization of information about U.S. citizens by governmental agencies? 1973 Northwestern (negative) defeated Georgetown (affirmative) on a four to three judges' decision. TopicResolved: Should the federal government provide a program of comprehensive medical care for all United States citizens? 1974 Harvard (affirmative) defeated Augustana of Illinois (negative) on a three to two judges' decision. TopicResolved: Should the federal government control the supply and utilization of energy in the United States? 1975 Baylor (negative) defeated Redlands (affirmative) on a three to two judges' decision. TopicResolved: Should the power of the presidency be significantly curtailed?

PAGE 69

---------------~ 1976 Kansas (negative) defeated Georgetown (affirmative) on a four to one judges' decision. 62 TopicResolved: Should the federal government adopt a comprehensive program to control land use in the United States? 1977 Georgetown (affirmative) defeated Southern California (negative) on a four to one judges' decision. TopicResolved: Should the federal government significantly strengthen the guarantee of consumer product safety required of manufacturers? 1978 Northwestern (negative) defeated Southern California (affirmative) on a three to two judges' decision. TopicResolved: Should United States law enforcement agencies have significantly greater freedom in the investigation and/or prosecution of felony crime? 1979 Harvard (negative) defeated Northwestern (affirmative) on a three to two judges' decision. TopicResolved: That the federal government should implement a program which guarantees employment opportunities for all United States citizens in the Labor force. 1980 Northwestern (affirmative) defeated Harvard (negative) on a five to zero judges' decision. TopicResolved: That the federal government should significantly strengthen the regulation of mass media communication in the United States. 1981 Pittsburgh (negative) defeated Dartmouth (affirmative) on a five to zero judges' decision. TopicResolved: That the United states should significantly increase its foreign military commitments. 1982 Louisville (negative) defeated Redlands (affirmative) on a three to two judges' decision. TopicResolved: That the Federal Government should significantly curtail the powers of labor unions in the United States. 1983 Kansas (affirmative) defeated Dartmouth (negative) on a four to one judges' decision. TopicResolved: Should United States military intervention into internal affairs of any foreign nation or nations in the Western Hemisphere be prohibited? 1984 Dartmouth (negative) defeated Louisville (affirmative) on a four to one judges' decision. TopicResolved: Should any and all injury resulting from the disposal of hazardous waste in the United States be the legal responsibility of the producer of that waste?

PAGE 70

63 1985 Harvard (negative) defeated Iowa (affirmative) on a five to zero judges' decision. TopicResolved: Should the United States government significantly increase exploration and/or development of space beyond the Earth's mesosphere? 2 Norman Markel combined the hostility and semantic differential scales into an emotive word list. 3 The value, P, is the percentage of emotive words in each rebuttal speech. The actual value choices for each independent variable is displayed below: where x 1 = 1 if team won the round = o if team lost the round where x 2 where x 3 where x 4 where x 5 = 1 if affirmative = 0 if negative = 1 for first speaker of that team. = 2 for second speaker of that team. = 1 for 1967 = 2 for 1968 = 3 for 1969 = 4 for 1970 = 5 for 1971 = 6 for 1972 = 7 for 1973 = 8 for 1974 = 9 for 1975 =10 for 1976 =11 for 1977 =12 for 1978 =13 for 1979 =14 for 1980 =15 for 1981 =16 for 1982 =17 for 1983 =18 for 1984 =19 for 1985 = 1 if single gender team = 0 if mixed gender team

PAGE 71

CHAPTER THREE RESULTS Results of the study, comparison of the emotive word use in the rebuttal speeches, results of the linear regression, findings of the research questions, and a summary of results are included. The Study The study was designed to examine the use of emotive words by academic debaters in the National Debate Tournament's championship round for each year from 1967 to 1985. The use of emotive words varied between debaters. The first four tables (3-1, 3-2, 3-3, 3-4) present the numbers of emotive words used by each debater. Table 3-1 displays the use of positive, negative, and overall use of emotive words by the rebuttal speakers in the NDT from 1967 to 1969. Table 3-2 displays the use of positive, negative, and overall use of emotive words by the rebuttal speakers in the NDT from 1970 to 1974. Table 3-3 displays the use of positive, negative, and overall use of emotive words by the rebuttal speakers in the NDT from 1975 to 1979 and table 3-4 displays the use of positive, negative, and overall use of emotive words by the rebuttal speakers in the NDT from 1980 to 1985. These tables display the number of emotive words spoken by each debater. 64

PAGE 72

65 Table 3-1: Use of Emotive Words in the NOT Rebuttal Speakers from 1967 to 1969: Year/Speaker Negative Positive Overall 1967 lNR 36 58 94 lAR 48 71 119 2NR 12 9 21 2AR 9 81 90 1968 lNR 68 59 127 lAR 51 81 132 2NR 67 78 145 2AR 46 76 122 1969 lNR 65 68 133 lAR 81 68 147 2NR 70 55 125 2AR 110 86 196

PAGE 73

66 Table 3-2: Use of Emotive Words in the NOT Rebuttal Speakers from 1970 to 1974: Year/Speaker Negative Positive Overall 1970 lNR 52 60 112 lAR 74 73 147 2NR 80 61 141 2AR 76 95 171 1971 lNR 78 62 140 lAR 71 55 126 2NR 90 114 204 2AR 105 78 183 1972 lNR 83 44 127 lAR 59 41 100 2NR 93 47 140 2AR 95 64 159 1973 lNR 69 44 113 lAR 69 87 156 2NR 98 82 180 2AR 65 75 140 1974 lNR 94 80 174 lAR 53 42 95 2NR 87 55 142 2AR 91 55 146

PAGE 74

67 Table 3-3: Use of Emotive Words in the NOT Rebuttal Speakers from 1975 to 1979: Year/Speaker Negative Positive Overall 1975 lNR 107 47 154 lAR 79 92 171 2NR 80 66 146 2AR 70 93 100 1976 lNR 103 65 168 lAR 70 62 132 2NR 106 85 191 2AR 104 61 165 1977 lNR 91 72 163 lAR 83 137 220 2NR 139 61 200 2AR 82 69 151 1978 lNR 81 35 116 lAR 21 79 100 2NR 98 65 163 2AR 88 88 176 1979 lNR 99 49 148 lAR 95 83 178 2NR 92 68 160 2AR 90 55 145

PAGE 75

68 Table 3-4: Use of Emotive Words in the NDT Rebuttal Speakers from 1980 to 1985: Year/Speaker Negative Positive Overall 1980 lNR 79 86 165 lAR 105 122 227 2NR 82 98 180 2AR 89 109 198 1981 lNR 92 61 153 lAR 84 79 163 2NR 114 85 199 2AR 73 63 136 1982 lNR 96 103 199 lAR 107 120 227 2NR 103 111 214 2AR 73 91 164 1983 lNR 94 73 167 lAR 129 75 204 2NR 82 72 154 2AR 79 95 174 1984 lNR 98 97 195 lAR 88 140 228 2NR 86 100 186 2AR 106 75 181 1985 lNR 105 74 179 lAR 182 108 290 2NR 99 98 197 2AR 128 103 231 ---------------------------------------------------------

PAGE 76

69 The debaters varied in their choices of emotive words as well as the amount of emotive language that they used. Tables 3-5 through 3-80 display the most frequently used emotive language for each speaker. For each speaker, the five most often used choices are presented, assuming that each of those words were used more than two times. If the speaker used identical numbers of some choices, then all those choices are also displayed. The number of word choices varies from zero (when a speaker never used an emotive word choice more than two times) to ten choices (when a speaker had six ties among those words used three times). Certain emotive words were used more than other words. This was clear for the negative words (tables 3-5 through 380). The debaters used a greater number of different positive words than negative words. The debaters were generally consistent in their use of the negative words. The word "not" was used more than any other negative word. "Not" was the most used negative word in all nineteen years of the National Debate Tournament studied, for both the second negative and second affirmative speaker. It was the most used word for the first negative and first affirmative speakers for eighteen of the nineteen debates. Both times that the word "not" was not the most used word, it was the second most used word. This happened in 1985 for the first negative rebuttal and in 1983 for the first affirmative rebuttal. In both debates, "argument" was used more than any other negative emotive word.

PAGE 77

Table 3-5: Most Often Used Emotive Words for First Negative Rebuttal 1967 Positive words Negative words united maintain certainly strong 8 3 3 3 not 11 Table 3-6: Most Often Used Emotive Words for First Affirmative Rebuttal 1967 Positive words Negative words gentleman united know free fact 14 8 5 3 3 not breaking no 24 3 3 70

PAGE 78

Table 3-7: Most Often Used Emotive Words for Second Negative Rebuttal 1967 Positive words Negative words Table 3-8: not 4 Most Often Used Emotive Words for Second Affirmative Rebuttal 1967 positive words negative words not 9 argument 3 71

PAGE 79

Table 3-9: Most Often Used Emotive Words for First Negative Rebuttal -1968 Positive words negative words kind 5 not 24 present 4 no 9 solution 4 poor 5 advantage 3 problems 3 social 3 without 3 solving 3 deny 3 welfare 3 Table 3-10: Most Often Used Emotive Words for First Affirmative Rebuttal 1968 Positive words negative words gentleman 8 not 20 advantage 6 poor 7 new 4 no 5 assistance 3 never 3 care 3 important 3 security 3 significant 3 social 3 72

PAGE 80

Table 3-11: Most Often Used Emotive Words for Second Negative Rebuttal 1968 Positive words present gentleman assistance social able better kind solving welfare 10 7 4 4 3 3 3 3 3 negative words not poor problem 31 16 5 Table 3-12: Most Often Used Emotive Words for Second Affirmative Rebuttal -1968 Positive words gentleman advantage conclude security significant social 7 3 3 3 3 3 Negative words not argument poor problem bit 19 7 6 3 3 73

PAGE 81

Table 3-13: Most Often Used Emotive Words for First Negative Rebuttal 1969 Positive words gentleman intelligence value true judgment united 11 7 6 5 4 4 Negative words not war without threat 18 4 4 4 Table 3-14: Most Often Used Emotive Words for First Affirmative Rebuttal 1969 Positive words Negative words gentleman 10 not 35 intelligence 7 harms 9 value 6 no 4 true 5 war 4 united 4 74

PAGE 82

Table 3-15: Most Often Used Emotive Words for Second Negative Rebuttal 1969 Positive words Negative words intelligence 10 not 23 gentleman 9 never 8 important 3 deny 4 like 3 no 3 Table 3-16: Most Often Used Emotive Words for Second Affirmative Rebuttal 1969 Positive words gentleman respect intelligence united judgment 15 12 11 5 4 Negative words not harm harms no 33 9 9 7 75

PAGE 83

Table 3-17: Most Often Used Emotive Words for First Negative Rebuttal 1970 Positive words Negative words health 9 not 18 care 6 poor 10 aid 4 deny 4 advantage 3 problem 3 increasing 3 critical 3 cut Table 3-18: Most Often Used Emotive Words for First Affirmative Rebuttal 1970 Positive words Negative words health 13 not 16 gentleman 10 poor 10 care 7 no 6 advantage 3 never 6 fine 3 76

PAGE 84

Table 3-19: Most Often Used Emotive Words for Second Negative Rebuttal 1970 Positive words Negative words care 9 not 28 health 7 poor 10 solve 6 no 10 even 4 problem 8 like 4 cut 8 right 4 Table 3-20: Most Often Used Emotive Words for Second Affirmative Rebuttal 1970 Positive words Negative words advance 14 not 18 gentleman 8 no 19 solve 9 cut 8 health 6 problem 8 care 5 please 5 77

PAGE 85

Table 3-21: Most often Used Emotive Words for First Negative Rebuttal 1971 Positive words Negative words care 7 not 22 fact 3 no 8 power 3 harms 4 advantage 3 denies 3 hurts 3 problem 3 Table 3-22: Most Often Used Emotive Words for First Affirmative Rebuttal 1971 Positive words Negative words gentleman 6 not 28 health 5 deny 5 care 3 poor 5 just 3 lose 3 well 3 78

PAGE 86

Table 3-23: Most Often Used Emotive Words for Second Negative Rebuttal 1971 Positive words Negative words solve 12 not 37 whole 8 problem 17 care 5 problems 7 health 4 poor 5 present 4 Table 3-24: Most Often Used Emotive Words for Second Affirmative Rebuttal 1971 Positive words Negative words advantage 9 not 33 help 5 deny 14 increase 5 no 13 even 4 critically 6 right 4 79

PAGE 87

Table 3-25: Most Often Used Emotive Words for First Negative Rebuttal 1972 Positive words Negative words well 5 not 24 might 4 no 14 win 3 argument 3 ok 3 war 3 strong 3 Table 3-26: Most Often Used Emotive Words for First Affirmative Rebuttal 1972 Positive words Negative words united 8 not 23 guard 3 argument 8 right 3 lose 4 security 3 no 4 greater 3 80

PAGE 88

Table 3-27: Most Often Used Emotive Words for Second Negative Rebuttal 1972 Positive words Negative words united 7 not 40 know 3 no 9 join 3 argue 5 gentlemen 3 death 5 argument 3 die 3 suffering 3 Table 3-28: Most Often Used Emotive Words for Second Affirmative Rebuttal 1972 Positive words Negative words well 5 not 23 win 5 argument 10 right 4 no 6 guard 4 war 6 like 3 never 6 constructive 3 winning 3 81

PAGE 89

Table 3-29: Most Often Used Emotive Words for First Negative Rebuttal 1973 Positive words Negative words well 5 not 26 might 4 no 10 strong 3 drug 3 win 3 drugs 3 ok 3 nothing 3 poor 3 Table 3-30: Most Often Used Emotive Words for First Affirmative Rebuttal 1973 Positive words Negative words health 13 not 20 care 11 no 8 increase 5 blood 6 please 4 problem 6 well 4 82

PAGE 90

Table 3-31: Most Often Used Emotive Words for Second Negative Rebuttal 1973 Positive words Negative words health 6 not 31 increase 5 no 21 advantage 5 threat 3 advantages 3 blood 3 clearly 3 deny 3 free 3 destroy 3 persists 3 never 3 true 3 Table 3-32: Most Often Used Emotive Words for Second Affirmative Rebuttal 1973 Positive words Negative words advantage 9 no 18 care 8 not 16 present 8 strike 3 good 4 bad 3 save 4 drug 3 absolutely 4 83

PAGE 91

Table 3-33: Most Often Used Emotive Words for First Negative Rebuttal 1974 Positive words Negative words protection 8 not 31 clean 6 pollution 15 know 6 no 10 win 6 harms 6 advantage 4 argument 4 increase 4 Table 3-34: Most often Used Emotive Words for First Affirmative Rebuttal 1974 Positive words gentleman protection 6 3 Negative words not no problem pollution 18 7 4 3 84

PAGE 92

Table 3-35: Most Often Used Emotive Words for Second Negative Rebuttal 1974 Positive words Negative words advantage 7 no 21 clean 5 pollution 16 win 3 argument 7 well 3 harm 6 Table 3-36: Most Often Used Emotive Words for Second Affirmative Rebuttal 1974 Positive words increase advantage protection please 5 4 3 3 Negative words not harm pollution no against 32 12 12 11 3 85

PAGE 93

Table 3-37: Most Often Used Emotive Words for First Negative Rebuttal 1975 Positive words Negative words special 6 not 32 please 4 no 17 know 4 harm 11 just 3 deny 7 Table 3-38: Most Often Used Emotive Words for First Affirmative Rebuttal 1975 Positive words Negative words justice 8 not 25 right 5 no 13 advantage 5 never 5 better 4 argument 5 new 4 86

PAGE 94

Table 3-39: Most Often Used Emotive Words for Second Negative Rebuttal 1975 Positive words justice please power true good 9 5 5 4 4 Negative words not no 36 8 Table 3-40: Most Often Used Emotive Words for Second Affirmative Rebuttal 1975 Positive words Negative words justice 8 not 14 advantage 6 no 13 present 5 argument 6 better 5 never 6 value 5 87

PAGE 95

Table 3-41: Most Often Used Emotive Words for First Negative Rebuttal 1976 Positive words Negative words new 7 not 25 absolutely 6 no 19 aid 5 problem 7 allow 4 destroy 4 just 4 argument 4 united 4 Table 3-42: Most Often Used Emotive Words for First Affirmative Rebuttal 1976 Positive words Negative words increase 8 not 23 adequate 3 argue 11 reasonable 3 no 6 respect 3 arguments 3 increased 3 failure 3 88

PAGE 96

Table 3-43: Most Often Used Emotive Words for Second Negative Rebuttal 1976 Positive words Negative words fact 11 not 28 increase 6 no 24 absolutely 5 argument 10 present 4 pollution 3 great 3 negative 3 wrong 3 Table 3-44: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1976 Positive words win constructive please even solve trust 8 5 4 3 3 3 Negative words not argument arguments no 37 11 10 8 89

PAGE 97

Table 3-45: Most Often Used Emotive Words for First Negative Rebuttal 1977 Positive words Negative words present 10 not 36 safety 7 no 13 please 7 argument 8 certainly 4 deny 5 resolution 4 negative 5 Table 3-46: Most Often Used Emotive Words for First Affirmative Rebuttal 1977 Positive words Negative words present 12 not 31 safety 10 no 10 respect 3 argument 8 increase 3 negative 6 new 3 90

PAGE 98

Table 3-47: Most Often Used Emotive Words for Second Negative Rebuttal 1977 Positive words Negative words even 5 not 36 just 5 no 16 safety 5 argument 16 new 4 argue 7 win 4 arguments 6 Table 3-48: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1977 Positive words Negative words well 6 not 19 present 5 no 13 please 4 argument 8 right 4 wrong 7 just 3 safety 3 91

PAGE 99

Table 3-49: Most Often Used Emotive Words for First Negative Rebuttal 1978 Positive words justice know 3 3 Negative words not no 40 8 argue 7 pollution 3 Table 3-50: Most Often Used Emotive Words for First Affirmative Rebuttal 1978 Positive words even 3 Negative words increase heart solve solves 6 3 3 3 92

PAGE 100

Table 3-51: Most Often Used Emotive Words for Second Negative Rebuttal 1978 Positive words Negative words increase 8 not 47 fact 6 no 10 significant 3 pollution 7 well 4 cut 7 increased 3 Table 3-52: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1978 Positive words right advantage necessary soft increase please reasonable significant solve 5 4 4 4 3 3 3 3 3 Negative words not no negative cuts pollution 28 16 4 4 4 93

PAGE 101

Table 3-53: Most Often Used Emotive Words for First Negative Rebuttal 1979 Positive words voluntary know important 3 3 3 Negative words not no harm negative base illness 30 30 6 5 3 3 Table 3-54: Most Often Used Emotive Words for First Affirmative Rebuttal 1979 Positive words Negative words opportunity 5 not 20 absolutely 4 no 20 adopt 3 argument 5 arguments 4 strike 4 wrong 4 94

PAGE 102

Table 3-55: Most Often Used Emotive Words for Second Negative Rebuttal 1979 Positive words fact please 3 5 Negative words not no arguments argument argue 21 12 8 6 6 Table 3-56: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1979 Positive words Negative words regard 7 not 23 even 4 no 11 like 3 argument 4 amendment 3 illness 4 never 4 nothing 4 95

PAGE 103

Table 3-57: Most Often Used Emotive Words for First Negative Rebuttal 1980 positive words Negative words better 10 not 35 advantage 8 no 19 good 5 arguments 5 free 4 problem 4 well 4 Table 3-58: Most Often Used Emotive Words for First Affirmative Rebuttal 1980 Positive words Negative words free 9 not 32 better 7 no 20 resolution 7 argument 16 please 5 argument 10 new 4 arguing 3 reasonable 4 bad 3 united 4 without 3 96

PAGE 104

Table 3-59: Most Often Used Emotive Words for Second Negative Rebuttal 1980 Positive words Negative words resolution 7 not 30 advantage 7 no 16 better 5 negative 4 free 5 argue 3 well 5 argument 3 arguments 3 Table 3-60: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1980 Positive words Negative words advantage 12 not 32 resolution 12 argument 17 free 8 argument 6 united 6 lying 6 necessary 4 no 6 97

PAGE 105

Table 3-61: Most Often Used Emotive Words for First Negative Rebuttal 1981 Positive words Negative words just 6 not 19 constructive 5 no 19 even 3 strike 7 important 3 argument 5 like 3 please 3 Table 3-62: Most Often Used Emotive Words for First Affirmative Rebuttal 1981 Positive words Negative words constructive 5 not 36 union 5 no 14 true 4 strike 6 assistance 4 problems 3 increased 3 harm 3 like 3 solve 3 win 3 98

PAGE 106

Table 3-63: Most Often Used Emotive Words for Second Negative Rebuttal 1981 Positive words Negative words good 7 not 30 new 7 no 30 aider 4 accidents 5 advantage 3 strike 6 even 3 greater 3 might 3 Table 3-64: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1981 Positive words Negative words fact 5 not 21 please 5 no 11 increased 5 strike 6 increase 3 argument 3 constructive 3 harm 3 war 3 99

PAGE 107

Table 3-65: Most Often Used Emotive Words for First Negative Rebuttal 1982 Positive words Negative words unions 9 not 33 just 8 argument 11 union 6 negative 8 increase 5 no 7 absolutely 4 desirable 4 Table 3-66: Most Often Used Emotive Words for First Affirmative Rebuttal 1982 Positive words Negative words unions 19 not 35 union 7 argument 20 advantage 7 no 7 constructive 6 arguments 5 just 5 bad 4 strikes 4 100

PAGE 108

Table 3-67: Most Often Used Emotive Words for Second Negative Rebuttal 1982 Positive words Negative words unions 20 not 33 please 7 negative 10 advantage 7 no 9 good 5 argument 8 constructive 4 arguments 6 power 4 Table 3-68: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1982 Positive words Negative words unions 17 not 23 advantages 7 no 6 right 6 strikes 5 pleasure 5 never 4 able 4 eliminate 4 101

PAGE 109

Table 3-69: Most Often Used Emotive Words for First Negative Rebuttal 1983 positive words Negative words united 10 not 38 increase 5 no 15 please 4 against 6 soft 4 abandon 4 solve 4 Table 3-70: Most Often Used Emotive Words for First Affirmative Rebuttal 1983 positive words Negative words advantage 6 argument 38 united 6 not 29 together 5 no 14 constructive 4 war 6 amendment 3 worse 3 increase 3 terrible 3 just 3 arguments 3 soft 3 well 3 win 3 102

PAGE 110

Table 3-71: Most Often Used Emotive Words for Second Negative Rebuttal 1983 Positive words Negative words united 8 not 19 advantage 7 no 19 win 6 war 7 whole 4 against 4 obviously 3 arguments 4 well 3 Table 3-72: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1983 Positive words Negative words constructive 8 not 18 true 6 war 16 advantage 5 none 7 good 5 bomb 6 union 4 threat 5 even 4 greatest 4 might 4 103

PAGE 111

Table 3-73: Most Often Used Emotive Words for First Negative Rebuttal 1984 Positive words Negative words right 11 not 27 increase 7 no 14 interest 5 argues 14 know 5 argue 8 new 5 argument 6 Table 3-74: Most Often Used Emotive Words for First Affirmative Rebuttal 1984 Positive words Negative words know 9 not 27 new 8 no 26 just 7 waste 9 better 7 arguments 4 right 6 increase 6 like 6 104

PAGE 112

Table 3-75: Most Often Used Emotive Words for Second Negative Rebuttal 1984 Positive words Negative words know 10 not 33 new 14 no 20 important 6 argument 18 increase 5 bad 6 interest 5 never 4 Table 3-76: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1984 Positive words Negative words new 10 not 33 increase 9 no 20 resolution 6 argument 18 open 6 bad 6 confidence 5 never 4 105

PAGE 113

Table 3-77: Most Often Used Emotive Words for First Negative Rebuttal 1985 Positive words Negative words important 7 argument 24 good 5 not 19 like 4 no 11 knowledge 3 bad 4 true 3 wars 4 Table 3-78: Most Often Used Emotive Words for First Affirmative Rebuttal 1985 Positive words Negative words true 24 not 69 just 7 argument 36 increase 6 no 30 good 6 bad 8 even 5 argues 5 106

PAGE 114

Table 3-79: Most Often Used Emotive Words for Second Negative Rebuttal 1985 Positive words Negative words good 8 not 27 increase 7 no 15 just 7 argument 11 knowledge 6 never 9 like 5 bad 7 Table 3-80: Most Often Used Emotive Words for Second Affirmative Rebuttal Speech 1985 Positive words Negative words increase 18 not 53 true 12 no 33 increased 8 bad 11 solver 6 never 7 able 5 107

PAGE 115

108 Rouse and Thomas (1987) postulated that word economy was important in making an effective rebuttal speech and they found that there was a "steady decline in lAR elaboration and rhetorical amplification over the years (p.108)." Their finding suggests that the debaters would demonstrate changes in their use of emotive words. An examination of tables 3-5 through 3-80 demonstrated a number of findings. The debaters were consistent in their use of negative emotive words. In addition to the use of "not" as the most widely used negative word, the second most widely used word was "no." "No" was used the second most in 12 of the 19 first negative rebuttals. "No" was used the second most in five of the 19 first affirmative rebuttals. "No" was also used the second most in ten of the 19 second negative rebuttals and in nine of 19 rebuttals by the second affirmative speakers. Other often used negative words included "argument" and "against." Not all negative words were used consistently. There were differences in the use of some negative words. The word "strike" was among the most often used in six affirmative rebuttals and the affirmative team lost four of those debates (1973, 1979, 1981, and 1982) while the word "strike" was in the most often used group for the negative only in the 1981 rebuttals. The word "war" was among the most often used group for the affirmative rebuttals in two winning years (1969, and 1981) and one losing year (1983). The negative debaters most often used the word "war" in

PAGE 116

109 three different years (1969, 1972, and 1983) and lost in all of them. The words "harm" or "harms" were among the most often used group in five first negative rebuttal speeches, but only in one second negative rebuttal. Of the five debates, the negative won two (1975 and 1979) and lost three (1969, 1971, and 1975). The first affirmative debaters most often used "harm" or "harms" in two years (1969, and 1981) while the second affirmative rebuttal speech also most often used those words in two years (1969, and 1974). Of the three different debates that included a large use of the words "harm" or "harms" by the affirmative, only one was a loss (1981). The use of the positive emotive words was less predictable than the use of the negative words. Among the more often used words were "gentleman," "just," "increase," "health," "know," "win," "new," "care," "advantage," "right,"' "well," "important, 11 "like, 11 "united," "present, 11 "constructive," "good," and "please." The use of these words varied between the four different rebuttal speeches. Some of these words were more often used by the winning debaters. The word "gentleman" was often used by the debaters until the middle 1970's. First affirmative rebuttalists consistently used "gentleman" from 1967 until 1971 and again in 1974. The affirmative won in 1968, 1969, 1971, and in 1974. In contrast, the first negative rebuttal speaker only used "gentleman" in 1969. The second negative rebuttal used

PAGE 117

110 "gentleman" in 1968, 1969, and 1972. All were losses for the negative. The second affirmative rebuttal used "gentleman" in 1968, 1969, and 1970. The word "just" was among the most often used words by the first negative rebuttalist in 1975, 1976, 1981 and 1982all negative wins. The first affirmative rebuttalists used "just" in 1971, 1982, 1983, 1984, and 1985. 1971 and 1983 were affirmative wins. The second negative rebuttal speaker used "just" in 1977, a loss, and 1985, a win. The second affirmative rebuttal only used "just" in 1977, a win. The word "increase" was among the most often used words in four second negative rebuttals (1976, 1978, 1984, and 1985) that the negative won. The first negative rebuttal speaker included "increase" or "increasing" in two victories (1970 and 1984) and two defeats (1974 and 1983). "Increase" was included in the most used group by the first affirmative rebuttal in two wins (1977, and 1983) and five losses (1973, 1976, 1981, 1984, and 1985). The second affirmative speaker used "increase" in a similar fashion with two wins (1971 and 1974) and four losses (1978, 1981, 1984, and 1985). The word "health" was more often used in the first affirmative and second negative rebuttal speeches than by the first negative and second affirmative speakers. Both the first affirmative and second negative rebuttal speakers used "health" in 1970, 1971 and 1973. The other two rebuttals only used "health" in 1970. 1970 and 1973 were negative wins.

PAGE 118

111 The word "know" was among the most often used words by the first negative rebuttalists in 1974, 1975, 1978, 1979, and 1984. All of those debates were negative wins except 1974. The first affirmative rebuttal speaker used "know" in 1967 and 1984; both debates were affirmative losses. The second negative rebuttal speaker used "know" in 1972 and 1984. 1972 was a negative defeat. The second affirmative rebuttal speaker did not include the word "know" among the most often used words. The word "win" was among the most often used words by the first negative rebuttal speaker in 1972, 1973, and 1974. Only 1973 was a negative victory. The first affirmative rebuttalist used "win" in 1981, a loss, and 1983, a win. The second negative rebuttalist used "win" in 1974, 1977, and 1983. All three debates were losses for the negative. The word "win" was used by the second affirmative speaker in 1972, a win, and 1976, a loss. The word "new" was among the most often used words by the first negative rebuttalist in 1976 and 1984; both debates were won by the negative. The first affirmative rebuttalist used "new" in 1968, 1975, 1977, 1980, and 1984. 1968, 1977, and 1980 were affirmative wins. The second negative rebuttalist used "new" in 1977, 1981, and 1984. The two later debates were wins. The second affirmative rebuttalist only used "new" in 1984 and lost the debate. The word "care" was among the most often used words in 1970 and 1971 by the first negative rebuttalist, the first affirmative rebuttalist, and the second negative rebuttal

PAGE 119

---112 speakers. In 1970, the negative won and the affirmative won in 1971. The second affirmative rebuttalist only used "care" in 1970. The word "advantage" was among the most often used words by the first negative rebuttalist in 1968, 1970, 1974, and 1980. Only 1970 produced a negative victory. The first affirmative rebuttalist used "advantage" in 1968, 1970, 1975, 1982, and 1983. The affirmative rebuttalists won in 1968 and 1983. The second negative rebuttalist used "advantage" in 1973, 1974, 1981, 1982, and 1983. The negative won in 1973, 1981, and 1982. The second affirmative rebuttal speaker used "advantage'' in 1968, 1971, 1973, 1974, 1975, 1978, 1980, 1982, and 1983. Of the nine years, the affirmative won five years 1968, 1971, 1974, 1980, and 1983. The word "right" was among the most often used words by the first negative rebuttalist in 1972 and 1973. The negative won in 1973. The first affirmative rebuttalist used the word "right" in 1972, 1975, and 1984. Only 1972 was a victory. The second negative rebuttalist only used "right" in 1970, a win. The second affirmative rebuttalist used "right" in 1971, 1972, 1977, 1978, and 1982. The affirmative won the first three debates and lost the last two. The word "well" was among the most often used words by the first negative rebuttal in 1972, 1973, and 1980. The negative only won in 1973. The first affirmative rebuttal used "well" in 1971, 1973, and 1983. 1971 and 1983 were

PAGE 120

113 wins. The second negative used "well" in 1974, 1978, 1980, and 1983. Only 1978 was a negative win. The second affirmative rebuttal used "well" in 1972 and 1977. Both debates were affirmative victories. The word "important" was among the most often used words by the first negative rebuttalists in 1979, 1981, and 1985. All three debates were negative wins. The first affirmative rebuttalist only used "important" in a win in 1968. The second negative rebuttal used "important'' in 1969, a loss, and 1984, a win. The second affirmative rebuttalist did not use the word "important" as one of the most often used words. The word "like" was among the most often used words by the first negative rebuttalists in 1981 and 1985. The negative won both debates. The first affirmative rebuttalist used "like" in 1981 and 1984. Both debates were losses. The second negative rebuttal speakers used "well" in 1969, 1970, and 1985. Both 1970 and 1985 were negative wins. The second affirmative rebuttalists used "like" in 1972, a win, and in 1979, a loss. The word "united" was among the most often used words by the first negative rebuttalists in 1967, 1969, 1976, and 1983. The negative won the debates in 1967 and 1976. The first affirmative rebuttalists used "united" in 1967, 1969, 1972, 1980, and 1983. All the debates except for 1967 were wins. The second negative used "united'' in 1972, and 1983. Both were negative losses. The second affirmative rebuttalist used "united" in 1969 and 1980. Both debates were affirmative wins.

PAGE 121

114 The word "present" was among the most often used words by the first negative rebuttalists in 1968 and 1977. The negative lost both debates. The first affirmative rebuttal speaker only used "present" in 1977, a win. The second negative rebuttal speakers used "present" in 1968 and 1971. Both debates were negative losses. The second affirmative rebuttalists used "present" in 1973, 1975, and 1977. The affirmative won in only 1977. The word "constructive" was among the most often used words by the first negative rebuttalist in 1981, a win. The first affirmative rebuttalists used "constructive" in 1981, 1982, and 1983. Only in 1983 did the affirmative win. The second negative rebuttalist only used "constructive" in 1982, a win. The second affirmative rebuttalist used "constructive" in 1972, 1976, 1981, and 1983. The affirmative lost in both 1976 and in 1981. The word "good" was among the most often used words by the first negative rebuttalists in 1980, a loss, and 1985, a win. The first affirmative rebuttalist only used "good" in 1985, a loss. The second negative rebuttal speakers used "good" in 1975, 1981, 1982, and 1985, all wins for the negative. The second affirmative rebuttal speakers used "good" in 1973, a loss, and 1983, a win. The word "please" was among the most often used words by the first negative rebuttalist in only 1975, a win. The first affirmative rebuttalist used "please" in 1973, a loss, and 1980, a win. The second negative rebuttalists used "please" in 1975, 1979, and 1982, all wins for the negative.

PAGE 122

115 The second affirmative rebuttalists used "please" in 1970, 1974, 1976, 1977, and 1981. Only two of the five debates (1974 & 1977) were affirmative victories. Overall, there was a much less consistent use of positive words than negative words by both the affirmative and negative teams. Those positive emotive words that were used more often were used in different numbers by the rebuttal speakers and with different results. Linear Regression Analysis Linear regression analyses were run to determine separate prediction equations for the negative emotive language, the positive emotive language, and the overall emotive language as generated from the six independent variables. The initial analysis was conducted by applying the general linear models procedure (GLM) of the Statistical Analysis System (SAS). The GLM was run for negative, positive, and the total use of emotive words. The results for the negative emotive words are in table 3-81. The results for the positive emotive words are in table 3-82 and the results for the total emotive words are in table 3-83. The GLM for the negative emotive words failed to find a significant effect for round result, side of topic, and speaker position. The GLM did discover significant effects for the year and gender variables. The GLM for the positive emotive words did not determine a significant effect for any of the independent variables. The GLM for

PAGE 123

116 the total emotive word usage demonstrated a significant effect for both the year and gender variables.

PAGE 124

Table 3-81: General Linear Models Procedure: Yl Negative Emotive Words f. value 3.45 R-sguare 0.212887 Source Result Side Position Year Gender 0.0043 f. Value 0.38 1.06 0.16 12.72 4.63 Pr> f. 0.5418 0.3075 0.6927 0.0007 0.0350 Table 3-82: General Linear Models Procedure: Y2 Positive Emotive Words f. value 1.13 R-sguare 0.116708 Source Result Side Position Year Gender 0.3435 f. Value 1. 71 2.36 0.67 1.02 1.94 0.1967 0.1301 0.4150 0.4589 0.1691 117

PAGE 125

118 Table 3-83: General Linear Models Procedure: Y3 Total Emotive Words f. value Pr 2:. f. 3.97 0.0032 R-sguare 0.220802 Source f. Value Pr 2:. f. Result 0.01 0.9064 Side 0.05 0.8154 Position 0.04 0.8427 Year 15.34 0.0002 Gender 4.39 0.0398

PAGE 126

119 Regression Analysis Multiple regression analyses were run to determine a prediction equation for the three dependent variables and the five independent variables. Five different linear regression analyses were run for each of the three dependent variables. The analyses included a forward selection, a stepwise procedure, a backward elimination, a maximum R 2 improvement, and a minimum R 2 improvement. The forward selection technique starts with no variables in the model. It then 11 adds the variable that has the largest F statistic to the model" (SAS, 1985, p.764). Following this step, the forward selection technique finds the next largest F statistic, given the largest value already included, and adds that variable to the model. It then adds the third largest F statistic, given the two strongest values, and adds that variable to the model. The forward selection 11 .. continues either until all variables have been added, or until the largest remaining partial F-value is too small to bother with" (Younger, 1979, p.397). The results of the forward selection are contained in tables 3-84 through 3-86. The forward selection technique for the use of negative emotive language provided evidence that supported research questions four and five. That is, both the year and gender variables exhibited significant relationships with the dependent variable. For both the positive emotive words and the total emotive words, only the year variable was significantly related.

PAGE 127

120 The stepwise procedure" ... is a modification of the forward selection technique and differs in that variables already in the model do not necessarily stay there" (SAS, 1985, p.764). The stepwise procedure adds variables one at a time, but previously entered variables can be dropped out if their F statistic becomes nonsignificant. Any deleted variable can be reincluded at a later step. The stepwise procedure" ... terminates either when no new variables can be entered or when the one to be entered was the one dropped at the previous step" (Younger, 1979, p.403). The results are in tables 3-87 through 3-89. The stepwise procedure for the use of negative emotive language provided evidence that supported research questions five and six. Both the year and gender variables exhibited significant relationships with the dependent variable. For both the positive emotive words and the total emotive words, only the year variable was significantly related. The stepwise procedure for the positive emotive words included debate side, but this failed to reach significance.

PAGE 128

121 Table 3-84: Forward Selection Procedure Summary: Yl Negative Emotive Words Step 1 2 3 Variable Year Gender Side f. Statistic 12.3477 5.6351 0.5178 Prob>F 0.0008 0.0202 0.4741 Model R**2 0.1430 0.2044 0.2101 Table 3-85: Forward Selection Procedure Summary: Y2 Positive Emotive Words Step Variable .E Statistic Prob>F Model R**2 1 Year 4.0720 0.0472 0.0522 2 Side 3.0616 0.0844 0.0903 3 Result 1.1328 0.2907 0.1044 4 Position 0.6929 0.4080 0.1131 Table 3-86: Forward Selection Procedure Summary: Y3 Total Emotive Words Step Variable f. Statistic Prob>F Model R**2 1 Year 15.2379 0.0002 0.1708 2 Gender 3.6067 0.0615 0.2098 3 Result 0.7097 0.4023 0.2175

PAGE 129

122 Table 3-87: Stepwise Procedure Summary: Yl Negative Emotive Words Step Variable E. Statistic Prob>F Model R**2 1 Year 12.3477 0.0008 0.1430 2 Gender 5.6351 0.0202 0.2044 Table 3-88: Stepwise Procedure Summary: Y2 Positive Emotive Words 1 2 Table Step 1 2 Variable E. Statistic Year 4.0720 Side 3.0616 Prob>F 0.0472 0.0844 3-89: Stepwise Procedure Summary: Total Emotive Words Variable E. Statistic Prob>F Year 15.2379 0.0002 Gender 3.6067 0.0615 Model R**2 0.0522 0.0903 Y3 Model R**2 0.1708 0.2098

PAGE 130

123 The backward elimination technique begins with all the variables in the model. The technique then deletes variables from the model, one at a time according to which variable has the smallest non-significant F statistic. The technique ends when all the remaining variables have F statistics that are significant. The results are found in tables 3-90 through 3-92. The backward selection technique for the use of negative emotive language provided evidence that supported research questions five and six. Both the year and gender variables exhibit significant relationships with the dependent variable. For both the positive emotive words and the total emotive words, only the year variable was significantly related. The debate side variable approached significance for the positive words. The maximum R 2 improvement technique is different from the forward, stepwise, and backward techniques. It does not suggest a single model. Rather the technique ... does not settle on a single model. Instead, it tries to find the best one-variable model, the best two-variable model, and so forth, although it is not guaranteed to find the model with the largest R 2 for each size'' (SAS, 1985, p.765). The results are contained in tables 3-93 through 3-110. The maximum R 2 improvement technique for the use of negative emotive language provided evidence that supported research questions five and six. Both the year and gender variables exhibit significant relationships with the dependent variable. In every one of the six ''best" variable model, the year variable was significant. In all but the best one

PAGE 131

124 variable model, the gender variable was significant. For both the positive emotive words and the total emotive words, only the year variable was significantly related. For both dependent variables, the year variable was consistently included in all best models and was consistently significant. No other variable was significant. The next included variables were consistent for both dependent variables. The debate side variable was the next included variable in the best two through six variable models for the positive emotive words. For the total emotive words, the gender variable was the next included variable in the best models. The minimum R 2 improvement technique is similar to the maximum R 2 technique. Both techniques often produce the same best model, but the minimum R 2 considers more models. The results are presented in tables 3-111 through 3-126. The results were identical to the maximum R 2 improvement technique results. Consistently the year variable was significant for all three dependent variables. The gender variable was significant for the negative emotive words.

PAGE 132

125 Table 3-90: Backward Selection Procedure Summary: Yl Negative Emotive Words Step 1 Variable Result Position Side E Statistic 0.0910 0.1595 0.5178 Prob>F 0.7639 0.6908 0.4741 Model R**2 0.2119 0.2101 0.2044 2 3 1 2 3 1 2 3 Table 3-91: Backward Selection Procedure Summary: Y2 Positive Emotive Words Variable Gender Position Result E Statistic 0.2896 0.6929 1.1328 Prob>F 0.5922 0.4080 0.2907 Model R**2 0.1131 0.1044 0.0903 Table 3-92: Backward Selection Procedure Summary: Y3 Total Emotive Words Variable Position Side Result E Statistic 0.0397 0.2595 0.7097 Prob>F 0.8427 0.6120 0.4023 Model R**2 0.2204 0. 2175 0.2098

PAGE 133

126 Table 3-93: Maximum R 2 Improvement Procedure: Yl Negative Emotive Words: Best One Variable Model Variable Year E. Statistic 12.35 Prob>F R-sguare 0.0008 0.1429 Table 3-94: Maximum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Two Variable Model Variable Year Gender E. Statistic 16.63 5.64 Prob>F R-sguare 0.0001 0.0202 0.2044 Table 3-95: Maximum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Three Variable Model Variable Side Year Gender E. Statistic 0.52 16.25 4.85 Prob>F R-sguare 0.4741 0.0001 0.0309 0.2101

PAGE 134

127 Table 3-96: Maximum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Four Variable Model Variable Side Position Year Gender E. Statistic 0.51 0.16 16.06 4.79 Prob>F R-sguare 0.4767 0.6908 0.0001 0.0319 0.2118 Table 3-97: Maximum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Five Variable Model Variable Result Side Position Year Gender E. Statistic 0.09 0.54 0.16 15.94 4.63 Prob>F R-sguare 0.7639 0.4628 0.6927 0.0002 0.0350 0.2129

PAGE 135

128 Table 3-98: Maximum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best One Variable Model Variable Year E. Statistic 4.07 Prob>F R-sguare 0.0472 0.0522 Table 3-99: Maximum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Two Variable Model Variable Side Year E. Statistic 3.06 4.19 Prob>F R-sguare 0.0844 0.0444 0.0903 Table 3-100: Maximum R 2 Improvement Procedure: Y2 Negative Emotive Words: Best Three Variable Model Variable Result Side Year E. statistic 1.13 2.44 4.19 Prob>F R-sguare 0.2907 0.1228 0.0442 0.1044 J

PAGE 136

129 Table 3-101: Maximum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Four Variable Model Variable Result Side Position Year .f Statistic 1.13 2.43 0.69 4.18 Prob>F R-sguare 0.2918 0.1237 0.4080 0.0447 0.1131 Table 3-102: Maximum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Five Variable Model Variable Result Side Position Year Gender .f Statistic 1.38 2.58 0.69 4.42 0.29 Prob>F 0.2436 0.1125 0.4104 0.0391 0.5922 R-sguare 0.1167

PAGE 137

130 Table 3-103: Maximum R 2 Improvement Procedure: Y3 Total Emotive Words: Best One Variable Model Variable Year .f statistic 15.24 Prob>F R-sguare 0.0002 0.1708 Table 3-104: Maximum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Two Variable Model Variable Year Gender .f Statistic 18.55 3.61 Prob>F 0.0001 0.0615 R-sguare 0.2098 Table 3-105: Maximum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Three Variable Model Variable Result Year Gender .f Statistic 0.71 18.97 4.29 Prob>F 0.4023 0.0001 0.0420 R-sguare 0.2175

PAGE 138

-------------~----131 Table 3-106: Maximum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Four Variable Model Variable Result Side Year Gender .E Statistic 0.61 0.26 18.90 4.45 Prob>F 0.4376 0.6120 0.0001 0.0385 R-sguare 0.2204 Table 3-107: Maximum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Five Variable Model Variable Result Side Position Year Gender .E Statistic 0.60 0.26 0.04 18.65 4.39 Prob>F 0.4407 0.6144 0.8427 0.0001 0.0398 R-sguare 0.2208

PAGE 139

132 Table 3-108 Minimum R 2 Improvement Procedure: Yl Negative Emotive Words: Best One Variable Model Variable Year E Statistic 12.35 Prob>F 0.0008 R-sguare 0.1429 Table 3-109: Minimum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Two Variable Model Variable Year Gender E Statistic 16.63 5.64 Prob>F 0.0001 0.0202 R-sguare 0.2044 Table 3-110: Minimum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Three Variable Model Variable E Statistic Prob>F R-sguare 0.2101 Side 0.52 0.4741 Year 16.25 0.0001 Gender 4.85 0.0309

PAGE 140

133 Table 3-111: Minimum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Four Variable Model Variable .E Statistic Prob>F R-sguare 0.2119 Side 0.51 0.4767 Position 0.16 0.6908 Year 16.06 0.0001 Gender 4.79 0.0319 Table 3-112: Minimum R 2 Improvement Procedure: Yl Negative Emotive Words: Best Five Variable Model Variable .E Statistic Prob>F R-sguare 0.2129 Result 0.09 0.7639 Side 0.54 0.4628 Position 0.16 0.6927 Year 15.94 0.0002 Gender 4.63 0.0350

PAGE 141

134 Table 3-113: Minimum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best One Variable Model Variable Year ,E Statistic 4.07 Prob>F R-sguare 0.0472 0.0522 Table 3-114: Minimum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Two Variable Model Variable Side Year ,E Statistic 3.06 4.19 Prob>F R-sguare 0.0844 0.0444 0.0903 Table 3-115: Minimum R 2 Improvement Procedure: Y2 Negative Emotive Words: Best Three Variable Model Variable .E Statistic Prob>F R-sguare 0.1044 Result 1.13 0.2907 Side 2.44 0.1228 Year 4.19 0.0442

PAGE 142

135 Table 3-116: Minimum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Four Variable Model Variable .E Statistic Prob>F R-sguare 0.1131 Result 1.13 0.2918 Side 2.43 0.1237 Position 0.69 0.4080 Year 4.18 0.0447 Table 3-117: Minimum R 2 Improvement Procedure: Y2 Positive Emotive Words: Best Five Variable Model Variable .E Statistic Prob>F R-sguare 0.1167 Result 1. 38 0.2436 Side 2.58 0.1125 Position 0.69 0.4104 Year 4.42 0.0391 Gender 0.29 0.5922

PAGE 143

136 Table 3-118: Minimum R 2 Improvement Procedure: Y3 Total Emotive Words: Best One Variable Model Variable Year E. Statistic 15.24 Prob>F 0.0002 R-square 0.1708 Table 3-119: Minimum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Two Variable Model Variable Year Gender E. Statistic 18.55 3.61 Prob>F 0.0001 0.0615 R-square 0.2098 Table 3-120: Minimum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Three Variable Model Variable E. Statistic Prob>F R-square 0.2175 Result 0.71 0.4023 Year 18.97 0.0001 Gender 4.26 0.0420

PAGE 144

137 Table 3-121: Minimum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Four Variable Model Variable Statistic Prob>F R-sguare 0.2203 Result 0.61 0.4376 Side 0.26 0.6120 Year 18.90 0.0001 Gender 4.45 0.0385 Table 3-122: Minimum R 2 Improvement Procedure: Y3 Total Emotive Words: Best Five Variable Model Variable Statistic Prob>F R-square 0.2208 Result 0.60 0.4407 Side 0.26 0.6144 Position 0.04 0.8427 Year 18.65 0.0001 Gender 4.39 0.0398

PAGE 145

Findings of the Research Questions Research Question One 138 Do winning and losing debate teams differ in their use of emotive words during rebuttal speeches in a debate round? The study failed to support any difference. Research Question Two Do the affirmative and the negative debaters differ in their use of emotive words in rebuttal speeches? The study failed to support any difference. Research Question Three Do the first and second rebuttals for each debate team differ in their use of emotive language? to support any difference. Research Question Four The study failed Does the quantity of the emotive words used by debaters vary over a period of years? The study found a significant difference in the use of emotive words over time. Debaters were found to use more emotive words over time. This finding was consistent for all three dependent variables and all statistical tests except for the GLM test of positive emotive words. Research Question Five Does the quantity of emotive words used differ between mixed-gender debate teams and single-gender debate teams? The study found some support for a difference. There was a significant difference in the use of the negative emotive

PAGE 146

------------139 words. Mixed-gender debate teams used more emotive words. This result was also consistent for the statistical tests. The GLM demonstrated a difference for the total use of emotive words. A Summary of Results This chapter described results of this investigation. The study examined only the rebuttal speeches in nineteen years of the final round of the National Debate Tournament. The research investigated the relationship of five independent variables result, debate side, rebuttal position, year, and gender to three dependent variables use of negative emotive words, use of positive emotive words, and total use of emotive words. Linear regressions were run using a general linear models procedure on all three dependent variables. The procedure indicated that year and gender were significant for the use of negative emotive words. Year was significant for both the total use of emotive words and for the use of negative words. The GLM failed to demonstrate any significant relationship between the independent variables and the use of positive emotive words variable. Following the general linear models procedure, five different linear regression analyses were run for each of the three dependent variables. The analyses included a forward selection, a stepwise procedure, a backward elimination, a maximum R 2 improvement, and a minimum R 2 improvement. The results were consistent with the results of the general linear models procedure for the negative and

PAGE 147

140 total emotive words. The results were not consistent with the use of the positive emotive words. While the GLM failed to demonstrate any significance, the other measures demonstrated that the year variable was significant. The debates were also examined for the use of individual emotive words. There was a great deal of consistency to the use of the negative emotive words. The most often used negative words are "not" and "no". The use of the positive emotive words is less consistent than the use of the negative words. Among the more often used words are "gentleman," "just," "increase," "health," "know," "win," "new," "care," "advantage," "right," "well," "important," "like," "united," "present," "constructive," "good," and "please." The use of these words varied between the four different rebuttal speeches.

PAGE 148

Overview of the Study CHAPTER FOUR DISCUSSION The present study examined the relationship between emotive words employed in debate and several contextual factors of interest to debaters and debate coaches. Emotive words were examined in terms of three dependent variables: 1) positive words, 2) negative words, and 3) positive and negative words (i.e., total emotive words). Contextual debate factors were examined in terms of five independent variables: 1) result of the debate, 2) side of the resolution, 3) speaker position, 4) year of the debate, and 5) gender composition of the debate teams. Transcripts of nineteen years of the final round of the National Debate Tournament provided the raw data for this research. The emotive words in the transcripts were located and counted by a computer aided content analysis technique. Once identified, the relationship of the dependent variable to the independent variable was examined by means of SAS statistical procedures. The statistical procedures were generally consistent in the results. One test, the GLM, had results slightly different from the other tests. The year of the debate variable was significant for all three dependent variables. The gender independent variable was found to be significant for the use of the negative emotive language dependent variable and with the total use of emotive language according to the GLM. 141

PAGE 149

142 The results of this study extends prior research in three ways. First, the importance of debate rebuttals as a focus of research was affirmed. The results are consistent with prior research (Rouse & Thomas, 1987) which demonstrated the value of research regarding the first affirmative rebuttal. The present study found no differences between any of the rebuttals and also demonstrated the value of examining all four rebuttals. No previous study, to our knowledge, has attempted to examine all four rebuttal speeches in a series of academic debates. Thus, this study fills a gap in the debate literature that had previously existed. Second, the value of using regression techniques to study a chronological series of debates in order to examine all the debates for selected variables and applying statistical tests is new. Rouse and Thomas {1987} were the first to examine a series of debate speeches. But there were no tests of significance. Third, the use of emotive language in debate was established. While some previous research examined language use in debate (Giffin, 1959; Williams & Webb, 1964; Williams, Webb, & Clark, 1966; Vasilius & Destephen, 1979; and Rouse & Thomas, 1987}, analysis of emotive language in academic debate has established as an increasingly significant factor in academic debate because this study found that the use of emotive language increased at a statistically significant level.

PAGE 150

143 Interpretation of the Results This research proposed to answer five questions relating to emotive language and debate. Each question is discussed below. 1. Do winning and losing debate teams differ in their use of emotive words during rebuttal speeches in a debate round? The results of the statistical tests failed to demonstrate any significant relationship. One possible explanation for the failure to find any significance is that the debaters in these rounds were the best in the United States and that the debaters used the emotive language to the same approximate degree. Although the research failed to find any significant difference between the dependent variables and the results of the debate, the following are a series of observations based upon differential frequency of individual emotive words. The results indicated that debaters were much more consistent in their use of the negative words than they were in the use of the positive words. For both the negative side team and affirmative side team, "not" was the most widely used negative emotive word. The second most widely used word was "no." Other often used negative words included "argument", and "against." Certain negative words were more likely to be used by the losing debate team, regardless of whether of which side of the resolution they argued. The word "strike" was among the most often used in six affirmative rebuttals and the affirmative side team lost all four of those debates. The

PAGE 151

144 negative side debaters most often used the word "war" in three debates and lost all of them. The use of positive words was less predictable than the use of the negative words. Among the more often used positive words were "gentleman," "just," "increase," "health," "know," "win," "new, 11 "care," "advantage," "right," "well," "important," "like," "united," "present," "constructive," "good," and "please." The use of these words varied between the four different rebuttal speeches. Some of these words were more often used by the winning team. Some positive words were used more often by affirmative side teams that won, than by negative side teams that won. Those words include "gentleman," and "united." The word "gentleman" was often used by both debate teams until the middle 1970's. The affirmative speakers usually won when they used "gentleman" and the negative side lost. The affirmative side won four of the six debates while the negative side lost all three debates. The word "united" was among the most often used words by the negative side rebuttals in two wins and four losses. The affirmative side won four and lost one debate. Other positive words were more often used by winning negative side teams than affirmative side teams. Those words include "please," "just," "know," "good," "increase," "health," and "constructive." The word "please" was among the most often used words by the first and second negative side rebuttals in three debates that were all won by the

PAGE 152

145 negative side. The affirmative side were less consistent with "please." They only won three of seven debates in which one of the affirmative side speakers included "please" among the most often used emotive words. The word "just" was among the most often used words by the first negative rebuttalists in four debates that were all negative side wins. In contrast, the second negative rebuttalists won only one of two debates. The affirmative side rebuttals were less successful in their use of "just". They won three of six debates. The word "know" was among the most often used words by the negative side rebuttals in four wins and one loss. The affirmative side rebuttal speakers used "know" in two losses. The word "good" was among the most often used words by the negative side rebuttals in four wins and one loss. The affirmative side used "good" in one win and two losses. The word "increase" was among the most often used words in negative side rebuttals that resulted in four wins and two losses. The first affirmative side speaker won two debates and lost five and the second affirmative side speaker won two and lost four. The word "health" was more often used by the first affirmative side and second negative side rebuttal speeches than by the first negative side and second affirmative side speakers. Two of the three debates were negative side wins. The word "constructive" was among the most often used words by the negative side rebuttals in two wins. The affirmative side used "constructive" in three wins and four losses.

PAGE 153

146 A few positive words were used by both successful affirmative side and negative side teams. Among the words were "well," "important," and "new." The word "well" was among the most often used words by the negative side rebuttals in three losses and one win. The affirmative side used "well" in four wins and one loss. The word "important" was among the most often used words by the negative side rebuttals in four wins and one loss. The affirmative side only used "important" in one debate, a win. The word "new" was among the most often used words by the negative side rebuttals in three wins and one loss. The affirmative side used "new" in three wins and two losses. Some emotive words were roughly evenly distributed between wins and losses by both sides. Among those words were "care," "advantage," and "right." Each side used the word "care" in two wins and two losses. The word "advantage" was among the most often used words by the first negative side rebuttal in three losses and one win. The second negative side rebuttal used "advantage" in two losses and three wins. The affirmative side rebuttals used "advantage" in nine different debates. They won five and lost four of the debates. The word "right" was among the most often used words by the negative side rebuttals in two wins and one loss. The affirmative side won three debates and lost three. Two words were found to be evenly distributed between wins and losses for the negative side, but more often to be found in affirmative side losses. Those words are "like"

PAGE 154

147 and "increase." The word "like" was among the most often used words by the negative side rebuttals in two wins and two losses. The affirmative side used "like" in three losses and one win. Two other words were almost evenly distributed between affirmative side wins and losses. But they were more often found in negative side losses. The words are "win" and "present." The word "win" was found to be used in two affirmative side wins and two losses. The negative side only won one and lost four debates. The word "present" was most often used by the affirmative side in one win and two losses. "Present" was among the most often used words by the negative side rebuttals in three debates that were all losses. Overall, there was a much more consistent use of negative emotive words by both the affirmative side and negative side debaters. "No" and "not" were the most often used by both sides. Two negative words were more often used by losing debaters "strike" for the affirmative side and "war" for the negative side. The use of the positive words was less consistent than the negative words. Some positive words were more likely to be used by a winning side. The words "well," "important," and "new" were found to be used by both the affirmative side and negative side teams that won. The words "united" and "gentlemen" were found more often in the word choices of the affirmative side teams that won. More words were found to be used by winning negative side teams. Those words included "please,'' "just," "know,"

PAGE 155

----148 "good," "increase," "health," and "constructive." Some positive emotive words were evenly found between both sides and results. These words include "care," "advantage," and "right." Other words were evenly found in both affirmative wins and losses, but were more likely to be found in negative side losses. These words were "win" and "present." Other words were found evenly in both negative side wins and losses, but more likely in affirmative side losses. The words include "like" and "increase." 2. Do the affirmative and the negative debaters differ in their use of emotive words in rebuttal speeches? The results of the statistical tests failed to demonstrate any significant relationship. The debate side variable was included in the stepwise procedure model for the positive emotive language dependent variable, but was not significant. The variable also was the second chosen variable in the maximum R 2 improvement procedure for the best two through six models for the positive emotive words, but was also not significant. The above discussion of the first research question of the differences in the use of individual emotive words suggested that different debaters were found to use different emotive words and that there were differences in the use of emotive words, but the differences were not statistically significant. 3. Do the first and second rebuttals for each debate differ in their use of emotive words? The results of the statistical tests failed to find any significant relationship.

PAGE 156

149 4. Does the quantity of emotive words used by debaters vary over a period of years? The results of almost all of the SAS statistical procedures clearly indicated that the year variable was significant. The general linear model demonstrated that the year variable was significant for the use of negative emotive words and the total usage of emotive words. All the other statistical procedures, including, the forward selection, the stepwise procedure, the backward elimination, the maximum R 2 improvement, and the minimum R 2 improvement techniques all found that the year variable was significantly related to all three dependent variables. The fact that each of the procedures found "year" to be significant, amplifies the finding. Carney (1972) explained that" .. the most convincing demonstration of the validity of an interpretation comes when findings from one analysis after another all produce the same kind of data" (p. 261). Younger (1979) discussed the forward selection, stepwise procedure, backward elimination, maximum R 2 improvement, and minimum R 2 improvement techniques and concluded that" .. only in extremely clear-cut problems would we expect all five techniques described here to produce the same subset of variables for the reduced model" (p.416). The finding that the use of emotive words was related to the year variable is related to previous findings. Rouse and Thomas (1987) examined the first affirmative side rebuttal of 13 years of the NOT and found that there was a consistent decline in elaboration and

PAGE 157

150 rhetorical amplication. They pointed out that successful debaters had" .. a gift of economical language" (p.101). Because of the strict time limits in debate and the limited length of the rebuttal speeches, Rouse and Thomas (1987) found the debaters to be utilizing more economical words over a 13 year period. They reported that debaters were found to have reduced excess wordage. In other words, the present research provides an empirically based explanation of the Rouse and Thomas results. The present study found the debaters to be utilizing greater use of emotive words over a period of nineteen years (1967-1985) and Rouse and Thomas found that debaters used more economical words over a period of thirteen years (1972-1984). The two findings are similar and are most likely related. Because the debaters were found to be cutting unnessary words and that they also were found to be using greater numbers of emotive words, they were not reducing the amount of emotive words used. Tacitly at least, debaters valued emotive words. The relationship between the use of the emotive words and the year variable is the most obvious finding of this study. This relationship held for all three dependent variables. 5. Does the quantity of emotive words used differ between mixed-gender debate teams and single-gender debate teams? The gender variable was found to be significantly related to the use of negative words. The results of all the SAS statistical procedures clearly indicated that the gender variable was significant. All six different

PAGE 158

151 statistical techniques concluded that the gender variable was significantly related to the use of negative emotive words. The gender variable was significant for both the positive and total words when the GLM procedure was utilized but not for the other procedures. Mixed-gender teams used more emotive words than the single-gender teams. All four of the mixed-gender debate teams lost their rounds. All the other debate teams were all-male in composition. The fact that no mixed-gender team won in the nineteen years studied supported Hill's (1973b) finding that found that all-male teams did better than mixed-gender or all female teams. The lack of mixed-gender success was in contrast to the studies (Hayes & McAdoo, 1972; Rosen, Dean & Willis, 1978) that reported that mixed-gender teams did better than single-gender teams. One possible explanation for this finding is that debate has been dominated by men and that judges might be biased against women. The fact that only four teams out of 38 included women would suggest that either few women were involved in the NDT or that it was difficult for women to reach the final round. One contra-indication to sexism is that three of the four teams including women occurred in the last three years of the debates studied. This study did support the earlier finding that mixed-gender debate teams differ from single gender teams. The mixed-gender variable was found to be related to the use of positive words. One explanation for the use of the emotive words by the mixed-gender teams is that those teams might have used emotive words in an attempt

PAGE 159

152 to overcome the judge's bias. Another explanation is that women use more emotional words as a response to their loss of power in our society (Lutz, 1990). Implications: Pragmatic and Academic Implications are discussed in two areas: implications for the practice of academic debate and implications for future research. For coaches and practitioners of debate, this study has provided evidence that the use of emotive words increased in the final round of the NDT from 1967 until 1985. While the relationship of emotive words and the debate round result was not established, the results indicate that both coaches and debaters should raise their consciousness regarding debater's language generally and their selection of emotive words in particular. The information obtained from this investigation regarding differences in individual emotive words have an implication for both the debater and debate coach. The use of certain emotive words should be considered carefully. Both sides should try to use positive words such as "well," "important," and "new." The negative side team should try to use words such as "please," "just," "know," "good," "increase," "health," and "constructive." The negative side team should try to avoid using words such as "war," "win," and "present." The affirmative side debater should try to use more positive words like "gentlemen," and "united". The affirmative side debaters should avoid words such as "strike". This brief summary of

PAGE 160

153 the results of this study is a practical guide for debate coaches and practioners. In terms of further research on emotive words in debate, the results of this study have indicated the value of pursuing the relationship between the use of emotive words with both year and gender variables. Further research should be conducted with investigating the use of emotive words in non-policy debate or in debates where the teams differ greatly in debate ability. Because the debaters in this study were relatively evenly matched (National champion and runner-up) in their abilities, additional research might compare dissimilar debate teams. If the teams are of different qualities, differences in the use of emotive language might be clearer. Research comparing the use of emotive words of NOT and CEDA debaters could demonstrate an empirical difference between the two different debate types. In regards to implications for research, it is, perhaps, the most important aspect of this study is that academic debate is placed in an empirical context. That is, this research is a significant step in closing the breach between the academic debate community and the quantitative researchers in the speech communication field. Limitations of the Study One limitation of this study was the nature of the data set. The debates that were studied were only the championship rounds of the National Debate Tournament. These were policy debates. NOT is a specialized type of

PAGE 161

154 debate and the results and conclusions are best generalized to NDT policy debates. Another limitation is that this study used the emotive words drawn from the content analysis scale and from the semantic differential scale. Other possible emotive words would not have been included. The lack of significance with the result variable might be explained by this limitation.

PAGE 162

REFERENCES Achen, c. H. (1982). Interpreting and using regression. Newbury Park, CA: SAGE Publications. Agresti, A., & Agresti, B.F. (1979). Statistical methods for the social sciences. San Francisco: Dellen Publishing. Allen, M. (1987). An extension of the subjective probability model. Paper presented at the Central States Speech Association Convention. st. Louis, MO. Allen, M., & Kellermann, K. (1988). Using the subjective probability model to evaluate academic debate arguments. Argumentation and Advocacy: Journal of the American Forensic Association, 25,93-107. Allen, R.R. (1963). The effects of interpersonal and concept compatibiltiy of the encoding behavior and achievement of debate terms. Central States Speech Journal, 14, 23-26. Allport, G.W., & Odbert, H.S. (1936). Trait-names: A psycho lexical study. Psychological Monographs,47, v-211. Anderson, K.E. (1966). Quantitative research in debate. Journal of the American Forensic Association, 1, 112115. Anderson, K.E. (1974). Quantitative research in debate. Journal of the American Forensic Association, 10, 147155. Arens, R., & Meadow, A. (1956). Psycholinguistics and the confession dilemma. Columbia Law Review, 56, 19-46. Armstrong, D., & Van Schooneveld, C.H. (Ed.s) (1977). Roman Jakobson: Echoes of his scholarship. Lisse, Belgium: The Peter de Ridder Press. Baccus, J. (1937). Debates judge each other. Quarterly Journal of Speech, ll, 74-80. Baez, A. (1989). The validity of a negative attitude vocabulary as an indicator of attitude. Unpublished manuscript, University of Florida. 155

PAGE 163

---------------------156 Barker, L.L. (1965). A comparative analysis of debater-judge ratings. Journal of the American Forensic Association, 2., 17-20. Bateson, G. (1955). A theory of play and fantasy. Psychiatric Research Reports, 2., 39-51. Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine. Batten, B.W., & Insko, C.A. (1986). Detection of speaker's attitude from successive pro and con advocacy. Representative Research in Social Psychology.16,28-37. Bauer, O.F., & Colburn, C.W. (1966). The maverick judge. Journal of the American Forensic Association, i, 22-25. Beard, R.S. (1969). Legal cross-examination and academic debate. Journal of the American Forensic Association, .., 61-66. Beeman, w. o. (1988). Affectivity in persian language use. Culture. Medicine and Psychiatry, 12, 9-30. Benson, J.A. (1975). An investigation of the relationship between speaking position and rank assignment in forensic competition. Journal of the American Forensic Association, ~, 183-188. Benson, J.A., & Friedley, S.A. (1982). An empirical analysis of evaluation criteria for persuasive speaking. Journal of the American Forensic Association, 19, 1-13. Berger, C.R., & Bradac, J.J. (1982). Language and social knowledge:Uncertainty in interpersonal relations. Baltimore: Edward Arnold. Berger, P.L., & Luckmann, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. Garden City, NY: Doubleday. Bergmann, B.L. (1988). The psychosocial profile of the health promoting adult. (Doctoral dissertation, University of Maryland, 1988). Dissertation Abstracts International, 49, 2545. Berry, W.D., & Feldman, s. (1985). Multiple regression in practice. Newbury Park, CA: SAGE Publications. Berthold, C.A. (1970). A descriptive study of selected characteristics of debate judge's ballots. Journal of the American Forensic Association, 2, 30-35.

PAGE 164

157 Bies, R.J. (1987). Beyond "voice": The influence of decision-maker justification and sincerity on procedural fairness judgments. Representative Research in Social Psychology, 17, 3-14. Binedell, N.A. (1989). Corporate characteristics and sociopolitical responsiveness: An empirical analysis in South Africa. (Doctoral dissertation, University of Washington, 1989). Dissertation Abstracts International, 51, 915. Boaz, J.K. (Ed.) (1978). 1978 national debate tournament final debate: Should United States law enforcement agencies have significantly greater freedom in the investigation and/or prosecution of felony crime? Journal of the American Forensic Association, 15, 22-61. Boaz, J.K. (Ed.) (1979). 1979 national debate tournament final round: That the federal government should implement a program which guarantees employment opportunities for all United States citizens in the labor force. Journal of the American Forensic Association, 16, 29-67. Boaz, J.K. (Ed.) (1980). 1980 national debate tournament final round: That the federal government should significantly strengthen the regulation of mass media communication in the United States. Journal of the American Forensic Association, 17, 18-58. Boaz, J.K. (Ed.) (1981). 1981 national debate tournament final round: That the United States should significantly increase its foreign military commitments. Journal of the American Forensic Association,18, 17-59. Boaz, J.K. (Ed.) (1982). 1982 national debate tournament final round: That the federal government should significantly curtail the powers of labor unions in the United States. Journal of the American Forensic Association, 19, 14-58. Boaz, J.K. (Ed.) (1983). 1983 national debate tournament final round: Should United States military intervention into internal affairs of any foreign nation or nations in the Western Hemisphere be prohibited? Journal of the American Forensic Association, 20, 23-61.

PAGE 165

158 Boaz, J.K. (Ed.) (1984). 1984 national debate tournament final round: Should any and all injury resulting from the disposal of hazardous waste in the United States be the legal responsibility of the producer of that waste? Journal of the American Forensic Association, 21, 16-59. Boaz, J.K. (Ed.) (1985). 1985 national debate tournament final round: Should the United States government significantly increase exploration and/or development of space beyond the earth's mesosphere? Journal of the American Forensic Association, 22, 26-62. Bolton, R. (1979). People skills: How to assert yourself. listen to others. and resolve conflicts. Englewood Cliffs, NJ: Prentice-Hall. Brainpower Inc. (1986). Statview 512+. Calabasas, CA: Brainpower Inc. Brooks, W.D. (1971). Judging bias in intercollegiate debate. Journal of the American Forensic Association, 1., 197-200. Budd, R.W., Thorp, R.K., & Donohew, L. (1967). Content analysis of communications. New York: Macmillan. Burgoon, J.K. (1975). Evaluation criteria as predictors of debate success. Journal of the American Forensic Association, 12, 1-4. Burgoon, J., & Montgomery, C. (1976). Dimensions of credibility for the ideal debater. Journal of the American Forensic Association, ll, 171-178. Burgoon, J.K., Wilkinson,M., & Partridge,R. (1979). The relative effectiveness of praise and derogation as persuasion strategies. Journal of the American Forensic Association, 16, 10-20. Bush, L. (1973). Individual differences multidimensional scaling of adjectives denoting feelings. Journal of Personality and Social Psychology, 25, 50-57. Carey, T.F. (1972). Content analysis:A technique for systematic inference from communications. Winnipeg, Canada: University of Manitoba Press. Chronicle of Higher Education (1991). The Chronicle of Higher Education Almanac, 38(1). Cliff, N. (1969). Adverbs as multipliers. In J.G. Snider, & C.E. Osgood, (Eds.) Semantic differential technigue:A sourcebook. Chicago: Aldine Publishing Company, 143-160.

PAGE 166

159 Clore, G.L., Ontony, A., & Foss, M.A. (1987). The psychological foundations of the affective lexicon. Journal of Personality and Social Psychology. 53, 751766. Coburn, C.W. (1972). Strategies for educational debate. Boston: Holbrook Press. Colbert, K.R. (1987). The effects of CEDA and NOT debate training on critical thinking ability. Journal of the American Forensic Association, 23, 194-201. Colbert, K. & Biggers, T. (1985). Why should we support debate? Journal of the American Forensic Association, 21 ,237-240. Conley, J.M. (1979). Language in the courtroom. Trial:The National Legal Magazine, 15(9), 32-36. Cox, J.R. (1974). A study of judging philosophies of the participants of the national debate tournament. Journal of the American Forensic Association, 11, 6171. Cox, M. (1976). Argumentation and debate: A classified bibliography (Review of Argumentation and debate: A classified bibliographyl. Journal of the American Forensic Association, il, 57. Crable, R.E. (1976). Models of argumentation and judicial judgment. Journal of the American Forensic Association, 12, 113-120. Cronen, V.E. (1970). Forensics and behavioral science: A response to Walwik and McGlone. Journal of the American Forensic Association, 2, 92-96. Cross, J.D., & Matlon, R.J. (1978). An analysis of judging philosophies in academic debate. Journal of the American Forensic Association, 15, 110-123. Dahl, H., & Stengel, B. (1978). A classification of emotion words: A modification and partial test of de Rivera's decision theory of emotions. Psychoanalysis and contemporary thought, 1, 269-312. Deatherage, S. (1985). First judge critique. In J.K. Boaz (Ed.) 1985 national debate tournament final round: Should the United States government significantly increase exploration and/or development of space beyond the earth's mesosphere? Journal of the American Forensic Association, 22, 55-56. Dresser, W.R. (1964). The use of evidence in ten championship debates. Journal of the American Forensic Association, 1, 101-106.

PAGE 167

160 Dresser, W.R. (1966). The impact of evidence on decision making. Journal of the American Forensic Association, J., 43-47. Dunne, D.P., Mack, H.L., & Pruett, R. (1971). Empirical evidence on the "logical" "proficiency" dichotomy in debate judging. Journal of the American Forensic Association, 2, 201-207. Eco, u. (1977). The influence of Roman Jakobson on the development of semiotics. In D. Armstrong & C.H. Van Schooneveld (Ed.s), =R=o=m=a=n-=--=J~a=k=o=b=s~o=n=:"-------'E=c=h=o~e=s::.-....;o=f=--=h=i=s scholarship (pp. 39-58). Lisse, Belgium: The Peter de Ridder Press. Ehninger, D. (1958). The Debate about debating. Quarterly Journal of Speech, 44, 128-136. Ell, J.F. (1988). Faculty perceptions of financial constraint in universities. (Doctoral dissertation, University of Alberta, 1988). Dissertation Abstracts International, 49, 2872. Ellis, D.S., & Minter, R. (1967). How good are debate judges? Journal of the American Forensic Association, .i, 53-56. El tinge, E.M. (1988). Linguistic content analysis of the Holt, Rinehart and Winston series of high school biology textbooks: A longitudinal study focusing on the use of inquiry. (Doctoral dissertation, Iowa State University, 1988). Dissertation Abstracts International, 49, 2606. Eman, V., & Lukehart, J. (1976). Information use in academic debate: An information theory perspective. Journal of the American Forensic Association,12, 178183. Emmert, P., & Barker, L.L. (Eds.) (1989). Measurement of communication behavior. New York: Longman. Emmons, N.J. (1988). Theory and practice in conflict: A content analysis of basic considerations affecting experimental design in social science research. (Doctoral dissertation, Georgia State University, 1988). Dissertation Abstracts International, 50, 309. Farrell, T.B. (1979). Habermas on argumentation theory: Some emerging topics. Journal of the American Forensic Association, 16, 77-82.

PAGE 168

161 Fay, P., & Middleton, W.C. (1942). Judgment of introversion from the transcribed voice. Quarterly Journal of Speech, 28, 226-228. Fiske, J. (1982). Introduction to communication studies. New York: Methuen. Frana, A.W. (1989). Characteristics of effective argumentation. Journal of the American Forensic Association, 25, 200-202. Freeley, A.J. (1976). Argumentation and debate: Reasoned decision-making. Belmont, California: Wadsworth Publishing Company. Freund, R.J., & Littel, R.C. (1986). SAS system for regression, 1986 edition. Cary, NC: SAS Institute, Inc. Freund, R.J., Littel, R.C., & Spector, P.C. (1986). SAS system for linear models, 1986 edition. Cary, NC: SAS Institute, Inc. Friedman, R.P. (1972). Reflections of an incompetent judge. Journal of the American Forensic Association,~' 123126. Fryar, M., Thomas, D.A., & Goodnight, L. (1989). Basic debate. Lincolnwood, Illinois: National Textbook Company. Gerber, G., Holsti, O.R., Krippendorff, K., Paisley, W.J., & Stone, P.J. (Eds.) (1969). The analysis of communication content:Developments in scientific theories and computer techniques. New York: John Wiley & Sons. Gibson, J.W., & Felkins, P.K. (1974). A Nixon lexicon. Western Speech,~, 190-198. Gibson, J.W., Gruner, C.R., Kibler, R.J. & Kelly, F.J. (1966). A quantitative examination of differences and similarities in written and spoken messages. Speech Monographs, ll, 444-451. Giffin, K. (1959). A study of the criteria employed by tournament debate judges. Speech Monographs, 26, 6971. Gift, T., Cole, R., & Wynne, L. (1979). An interpersonal measure of hostility based on speech context. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 87-94). New York: Springer-Verlag.

PAGE 169

162 Goffman, E. (1959). Presentation of self in everyday life. Garden City, NY: Doubleday Anchor Books. Gottschalk, L.A. (Ed.) (1961). Comparative psycholinguistic analysis of two psychotherapeutic interviews. New York: International Universities Press. Gottschalk, L.A. (1986). Research using the Gottschalk Gleser content analysis scales in english since 1969. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 29-46). New York: Springer-Verlag. Gottschalk, L.A., & Gleser, G.C. (1969). The measurement of psychological states through the content analysis of verbal behavior. Berkeley, CA: University of California Press. Gottschalk, L.A., Eckard, M.J. & Feldman, D.J. (1979). Further validation studies of a cognitive-intellectual impairment scale applicable to verbal samples. In L.A. Gottschalk (Ed.), The content analysis of verbal behavior: Further studies. New York: SP Medical & Scientific Books. Gottschalk, L.A., Hausmann, c., & Brown, J.S. (1979). A computerized scoring system for use with content analysis scales. In L.A. Gottschalk (Ed.), The content analysis of verbal behavior: Further studies (pp. 237250). New York: SP Medical & Scientific Books. Gottschalk, L.A., & Hoigaard, J. (1986). A depression scale applicable to verbal samples. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 105-122). New York: Springer-Verlag. Gottschalk, L.A., Winget, C.N., & Gleser, G.C. (1969). Manual of instructions for using the Gottschalk-Gleser content analysis scales: Anxiety, hostility, and social alienation personal disorganization. Berkeley: University of California Press. Gow, J.E. (1967). Tournament debating: A time for changes. Journal of the American Forensic Association,!, 107111. Gregg, R.B. (1967). The rhetoric of evidence. Western Speech, ll, 180-189. Hample, D. (1979). Motives in law: An adaptation of legal realism. Journal of the American Forensic Association, 15, 156-168.

PAGE 170

163 Hanson, C.T. (1987). Seeding as a tournament practice: Scheduling elimination rounds. Journal of the American Forensic Association, 23, 206-209. Harris, T.E., & Smith, R.M. (1973). A systems analysis of the current debate controversy. Journal of the American Forensic Association, 2, 355-360. Haslett, B.J. (1987). Communication:Strategic action in context. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Hayes, M.T., & McAdoo, J. (1972). Debate performance: Differences between male and female rankings. Journal of the American Forensic Association,~' 127-131. Heath, S.B. (1979). The context of professional languages: An historical overview. In Alatis & Tucker (Eds), Language in public life (pp. 102-135). Washington, DC: Georgetown University Press. Hill, S.R. (1973a). A study of participant evaluations in debate. Journal of the American Forensic Association, 2, 371-377. Hill, S.R. (1973b). A study of the effect of non-ability variables on the outcome of intercollegiate debates. (Doctoral dissertation, University of Florida, 1973) Dissertation Abstracts International, 34, 7366A. Hill, S.R. (1975). Prestige as a factor influencing the validity of decisions in intercollegiate debate. Southern Speech Journal,40, 351-364. Hollihan, T., Riley, P., & Baske, K. (1985). The art of storytelling: An argument for the narrative perspective in academic debate. In J. Cox, M. Sillars, & G. Walker (Eds), Argument and social practice: Proceedings of the fourth SCA/AFA conference on argumentation (pp. 807-826). Annandale, VA: Speech Communication Association. Howe, J. (1981). CEDA's objectives: Lest we forget. In D. Brownlee (Ed.), The philosophy and practice of CEDA. New York: Cross examination Debate association. Hufford, R. (1965). The logician, the historian, and rhetorical criticism. Journal of the American Forensic Association,,14-16. Hufford, R. (1965b). Toward improved tournament judging. Journal of the American Forensic Association,, 120-126.

PAGE 171

164 Hurrat, K.S. (1988). Some factors affecting the coverage of foreign events by 107 american newspapers. (Doctoral dissertation, University of Texas at Austin, 1988). Dissertation Abstracts International, 50, 284. Irvine, J. T. (1990). Registering affect: Heteroglossia in the linguistic expression of emotion. In C. A. Lutz and L. Abu-Lughod (Eds.) Language and the politics of emotion. (pp. 126-161). New York: Cambridge University Press. Iversen, G.R., & Norpoth, H. (1987). Analysis of variance. Newbury Park, CA: SAGE Publications. Jakobson, R. (1972). Verbal communication. Scientific American, 227, 72-80. Jakobson, R. (1960). ciosing statement: Linguistics and poetics. In T.A. Sebeok (Ed.) Style in language (pp. 350-377). Cambridge, MA: Massachusetts Institute of Technology Press. Jennings, J.C. {1987). Fifth grade students' perceptions of selected effective teacher characteristics. (Doctoral dissertation, University of Virginia, 1987). Dissertation Abstracts International, 49, 2545. Kellerman, K. (1980). The concept of evidence: A critical review. Journal of the American Forensic Association, 16, 159-172. Kemp, R. (1975). National debate tournament cross examination, hurrah? Journal of the American Forensic Association, li, 12. Kerlinger, F.N. (1964). Foundations of behavioral research. New York: Holt, Rinehart and Winston. King, T.R., & Clevenger, T. (1960). A comparison of debate results obtained by participant and critic judging. Southern Speech Journal, 25, 223-232. King, T.R., & Phifer, G. (1968). The college debater as seen by himself and by his peers. Journal of the American Forensic Association, 2, 48-52. Klopf, D.W. (1964). Practices in intercollegiate tournaments. Journal of the American Forensic Association, 1, 48-52. Klopf, D., Evans, D., & De Lozier, M.L. (1965). Comparative studies of students, laymen, and faculty members as judges of speech contests. Speech Teacher, 14, 314318.

PAGE 172

165 Klopf, D.W., & Mccroskey, J.C. (1964). Debating both sides ethical? Controversy Pau! Central States Speech Journal, 15, 36-39. Klopf, D., & Mccroskey, J. (1964b). Ethical practices in debate. Journal of the American Forensic Association, i, 13-16. Klopf, D., & Rives, s. (1965). Characteristics of high school and college forensics directors. Journal of the American Forensic Association, i, 33-37. Kneupper, C.H. (1979). Paradigms and problems: Alternative constructivist/interactionist implications for argumentation theory. Journal of the American Forensic Association, 15, 220-227. Kobayshi, V.N. (1988). The self-reflexive mind: The life's work of Gregory Bateson. Qualitative Studies in Education, i, 347-359. Koch, U. (1986). Studies proving the validity of the Gottschalk-Gleser content analysis scales in gerrnan speaking countries. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 47-54). New York: Springer-Verlag. Kully, R.D. (1972). Forensics and the speech communication discipline: Analysis of an estrangement. Journal of the American Forensics Association,~, 192-199. Laase, L.T. (1942). An evaluation of the quality rating system in measuring debate achievement. Quarterly Journal of Speech, 28, 424-430. Larson, C.E., & Giffin, K. (1964). Ethical considerations in the attitudes and practices of college debaters. Journal of the American Forensic Association, i, 86-90. Lebovits, A.H., & Holland, J.C. (1986). Use of the Gottschalk-Gleser verbal content analysis scales with medically ill patients. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 133-148). New York: Springer-Verlag. Lewis, J.J., & Larsen, J.K. (1981). Inter-rater judge agreement in forensic competition. Journal of the American Forensic Association, 18, 9-16. Lewis-Beck, M.S. (1980). Applied regression: An introduction. Newbury Park, CA: SAGE Publications.

PAGE 173

166 Lichtman, A.J., Rohrer, D.M., & Corsi, J. (1987). The debate resolution. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 126129. Littlefield, R.S. (1987). An analysis of tabulation procedures used to produce contestants for elimination rounds at national individual events tournaments. Journal of the American Forensic Association, n, 202205. Littlejohn, S.W. (1983). Theories of human communication. Second Edition. Belmont, CA: Wadsworth Publishing. Lutz, C.A., & Abu-Lughod, L. (Eds) (1990). Language and the politics of emotion. New York: Cambridge University Press. Mann, M.P. (1989). Critical events in faculty careers. (Doctoral dissertation, Indiana University, 1989) Dissertation Abstracts International, 51, 756. Markgraf, B. {1966). Debate judging and debater adaptation: A reply. Journal of the American Forensic Association, ~, 37-39. Matlon, R.J., & Keele, L.M. (1984). A survey of participants in the national debate tournament, 19471980. Journal of the American Forensic Association, 20, 194-205. Markel, N. (1989). Personal communication. Markel, N. (1990). The emotive function of soeech. Gainesville, FL:University Copy Center. Mayer, M.E., & Meldrum, V.D. (1987). The effect of various time limits on the quality of rebuttals. Journal of the American Forensic Association, n, 158-165. McBath, J.H., & Aurbach, J. (1967). Origins of the national debate resolution, Journal of the American Forensic Association,~, 96-103. Mccawley, J.D. (1977). Jakobsonian ideas in generative grammar. In D. Armstrong & C.H. Van Schooneveld (Ed.s), Roman Jakobson: Echoes of his scholarship (pp. 269284). Lisse, Belgium: The Peter de Ridder Press. Mccroskey, J.C., & Camp, L.R. (1964). A study of stock issues, judging criteria, and decisions in debate. Southern Speech Journal, 30,158-168.

PAGE 174

167 Mccroskey, J.C., & Camp, L.R. judge's bias in debate. Forensic Association,~, (1966). Judging criteria and Journal of the American 59-62. McGlone, E.L. (1969). Toward improved quantitative research in forensics. Journal of the American Forensic Association,, 59-54. Muller, H.J. (1943). Science and criticism. New Haven: Yale University Press. Murphy, J.J. (1964). Two medieval textbooks in debate. Journal of the American Forensic Association, i, 1-6. Murrish, W.H. (1964). Training the debater in persuasion. Journal of the American Forensic Association, i, 7-12. Newman, S.S. (1939). Personal symbolism in language patterns. Psychiatry,~, 177-182. Newman, S.S., & Mather, V.G. (1938). Analysis of spoken language of patients with affective disorders. American Journal of Psychiatry,94, 913-942. Newman, R.P., & Sanders, K.R. (1965). A study in the integrity of evidence. Journal of the American Forensic Association,~' 7-13. Nimicks, M.J. (1986). The indian independence movement under the leadership of Mahatma Gandhi and the U.S.A. civil rights movement under the leadership of Dr. Martin Luther King, Jr.: A comparison of two social movements to asess the utlity of nonviolence as a rhetorical strategy. (Doctoral dissertation, University of Florida, 1985). Dissertation Abstracts International, 48, 1866A. O'Barr, W.M. & Conley, J.M. (1976). When a juror watches a lawyer. Barrister, ~(3), 8-11,33. Ogden, C.K., & Richards, I.A. (1923). The meaning of meaning. London: Kegan, Paul, Trench, Truber. Osgood, C., Suci, G., & Tannenbaum, P.H. (1957). The measurement of meaning. Urbana: University of Illinois Press. Ovid (1955). Metamorphoses. (R.Humphries, Trans.). Bloomington: Indiana Indiana University Press. Parkinson, M.G. (1981), Verbal behavior and courtroom success. Communication Education, 30, 22-32. Parkinson, M.G., & Parkinson, L.M. (1979). Speech tactics for successful trials. Trial:The National Legal Magazine, 15(9), 36.

PAGE 175

168 Parkinson, M.G., Geisler, D., & Pelias, M.H. (1983). The effects of verbal skills on trial success. Journal of the American Forensic Association, 20, 16-22. Patterson, J.W., & Zarefsky, D. (1983). Contemporary debate. Boston: Houghton Mifflin Company. Pearce, W.B. (1969). Communicating the reasons for decision by the ballot. Journal of the American Forensic Association,, 73-77. Petelle, J.L. (1967). A comparison of team quality ratings against strong versus weak opposition. Journal of the American Forensic Association,~, 104-106. Petrie, H.G. (1969). Does logic have any relevance to argumentation? Journal of the American Forensic Association,, 55-60. Pomorska, K. (1977). Roman Jakobson and the new poetics. In D. Armstrong & C.H. Van Schooneveld (Eds.), Roman Jakobson: Echoes of his scholarship (pp. 363-378). Lisse, Belgium: The Peter de Ridder Press. Raghavan, A. (1989). Effect of seating arrangement and gender on speaking style. (Doctoral dissertation, University of Florida, 1989). Dissertation Abstracts International, 51, 681A. Reeves, M., & Osborn, L.R. (1965). Judges of high school debate tournaments: Sources, criteria, and orientation. Speech Teacher, 14, 59-62. Rickert, W.E. {1978). Debate poiesis. Journal of the American Forensic Association, 14,141-143. Rieke, R.D., & Smith, D.H. (1968). The dilemma of ethics and advocacy in the use of evidence. Western Speech,32, 223-234. Rives, S.G. (Ed.) (1967). 1967 national debate tournament final debate.Journal of the American Forensic Association,~, 118-139. Rives, S.G. (Ed.) (1968). 1968 national debate tournament final round: Should the federal government guarantee a minimum annual cash income to all citizens? Journal of the American Forensic Association,~, 115-141. Rives, S.G. (Ed.) (1969). 1969 national debate tournament final round: Should executive control of United States foreign policy be significantly curtailed? Journal of the American Forensic Association,, 139-163.

PAGE 176

169 Rives, S.G. (Ed.) (1970). 1970 national debate tournament final debate: Should the federal government grant annually a specific percentage of its income tax revenue to the states? Journal of the American Forensic Association, 2, 159-187. Rives, S.G. (Ed.) (1971). 1971 national debate tournament final debate: Should the federal government adopt a program of compulsory wage and price contrls? Jounral of the American Forensic Assocation, ~, 1-28. Rives, S.G. (Ed.) (1972). 1972 national debate tournament final debate: Should greater controls be imposed on the gathering and utilization of information about U.S. citizens by governmental agencies? Journal of the American Forensic Association,~, 235-260. Rives, S.G. (Ed.) (1973). 1973 national debate tournament final debate: Should the federal government provide a program of comprehensive medical care for all United states citizens? Journal of the American Forensic Association, 10, 16-45. Rives, S.G. (Ed.) (1974). 1974 national debate tournament final debate: Should the federal government control the supply and utilization of energy in the United States? Journal of the American Forensic Association, 11, 16-45. Rives, S.G. (Ed.) (1975). 1975 national debate tournament final round: That the power of the presidency should be significantly curtailed. Journal of the American Forensic Association, 12, 5-41. Rives, S.G. (Ed.) (1976). 1976 national debate tournament final debate: Should the federal government adopt a comprehensive program to control land use in the United States? Journal of the American Forensic Association, ll, 1-50. Rives, S.G., & Boaz, J.K. (Eds.) (1977). 1977 national debate tournament final debate: Should the federal government significantly strengthen the guarantee of consumer product safety required of manufacturers? Journal of the American Forensic Association, 14, 10-52. Rohrer, D.M. (1987). Debate as a liberal art. In D.A. Thomas, & J. Hart, (Eds) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 7-14. Rosen, N., Dean, L.M., & Willis, F.N. (1978). The outcome of debate in relation to gender, side, and position. Journal of the American Forensic Association, 15,17-21.

PAGE 177

170 Rouse, R.D., & Thomas, D.A. (1987) The first affirmative rebuttal: A content analysis of 13 NOT transcripts. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 101-111. Rowland, R.C. (1986). The relationship between realism and debatability in policy advocacy. Journal of the American Forensic Association, 22, 125-134. Rubin, J. (1973). Sociolinguistics. In J.J. Honigmann (Ed.) Hanbook of social and cultural anthropology. Chicago: Rand McNally, 479-508. Sapir, E. (1921). Language. New York: Harcourt Brace. Sapir, E. (1927). Speech as a personality trait. American Journal of Sociology .J.,Z, 892-905. SAS Institute Inc. (1985). SAS user's guide: Statistics. version 5 edition. Cary, NC: SAS Institute, Inc. Sayer, J.E. (1974). Debater's perception of nonverbal stimuli. Western Speech,38, 2-6. Schlueter, D.W., Barge, J.K., & Blankenship, D. (1990). A comparative analysis of influence strategies used by upper and lower-level male and female managers. Western Journal of Speech Communicauion, 54, 42-65. Schofer,G., Koch, U., & Balck, F. (1979). Possible applications of the Gottschalk-Gleser content analysis of speech in psychotherapy research. In L.A. Gottschalk (Ed.), The content analysis of verbal behavior: Further studies. New York: SP Medical & Scientific Books. Semlak, W.D., & Shields, D.C. (1977). The effect of debate training on students participating in the bicentennial youth debates. Journal of the American Forensic Association, 13, 192-196. Snider, J.G., & Osgood, C.E. (Eds) (1969). Semantic differental technigue:A sourcebook. Chicago: Aldine Publishing Company. Smith, C.C. (1937). Sportsmanship in debating. Quarterly Journal of Speech, D., 83-86. Smith, C.G. (1989). The federal service impasses panel: A decision model. (Doctoral dissertation, The Ohio State University, 1989). Dissertation Abstracts International, 49, 4004.

PAGE 178

171 Solarz, A.K. (1969). Perceived activity in semantic atlas words as indicated by a tapping response. In J.G. Snider, & C.E. Osgood, (Eds.) Semantic differential technigue:A sourcebook. Chicago: Aldine Publishing Company, 266-272. Southworth, W. (1984). Fourth judge critique. In J.K. Boaz (Ed.) 1984 national debate tournament final round: Should any and all injury resulting from the disposal of hazardous waste in the United States be the legal responsibility of the producer of that waste? Journal of the American Forensic Association, 21, 53-56. Sproule, J.M. (1974). Constructing, implementing and evaluating objectives for contest debating: A critique of critiques on debate. Journal of the American Forensic Association, 11, 8-15. Staats, A.W., & Staats, C.K. (1969). Meaning and m, correlated but separate. In J.G. Snider, & C.E. Osgood, (Eds.) Semantic differential technigue:A sourcebook. Chicago: Aldine Publishing Company, 197-205. Stewart, J.R., & Merchant, J.J. (1969). Perceived differences between debaters and non-debaters. Journal of the American Forensic Association, Q, 67-72. Sullivan, P.A. (1989). The 1984 vice-presidential debate: A case study of female and male framing in political campaigns. Communication Quarterly, 37, 329-343. Tajeu, K.S. (1989). Four-H EFNEP volunteer involvement: A case study of a cooperative extension effort to involve short-term, Four-H EFNEP helpers in on-going, independent, Four-H volunteer leadership roles. (Doctoral dissertation, Cornell University, 1989). Dissertation Abstracts International, 50, 2755. Tanaka, Y., Oyama, T., & Osgood, C.E. (1969). A cross cultural and cross-concept study of the generality of semantic space. In J.G. Snider, & C.E. Osgood, (Eds.)Semantic differential technigue:A sourcebook. Chicago: Aldine Publishing Company, 289-302. Thomas, D.A. (Ed.) (1979). Rebuttal strategy and tactics. In D.A. Thomas, Advanced debate. Skokie, Illinois: National Textbook Company, 288-301. Thomas, D.A. (1987a). Research in debate from the past five years: Trends, omissions, and recommendations. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 15-28.

PAGE 179

172 Thomas, D.A. (1987b). Rebuttal strategies. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 90-100. Thomas, D.A. (1987c). Judge adaptation. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 116-124. Thomas, D.A., & Hart, J. (Eds.) (1987). Advanced debate: Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company. Thompson, W.N. (1967). Quantitative research in public address and communication. New York: Random House. Uliana, R.L. (1979). Measurement of black children's affective states and the effect of interviewer's race on affective states as measured through language behavior. In L.A. Gottschalk (Ed.), The content analysis of verbal behavior: Further studies. New York: SP Medical & Scientific Books. Van Eemeren, F.H., & Grootendorst, R. (1983). Unexpressed premises: Part ii. Journal of the American Forensic Association, 19, 215-225. Vasilius, J.M., & Destephen, D. (1979). An investigation of the relationship between debate tournament success and rate, evidence, and jargon. Journal of the American Forensic Association, 15, 197-204. Verderber, R. (1968). Judge's criteria and debater adaptation: Empirical evidence. Journal of the American Forensic Association,~' 28-30. Viney, L.L. (1986). Assessment of psychological states through content analysis of verbal communications. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 55-86). New York: Springer-Verlag. Viney, L.L., Westbrook, M.T., & Preston, C. (1986). Some sources of alienation for drug addicts. In L.A. Gottschalk, F. Lolas, & L.L. Viney (Ed.s), Content analysis of verbal behavior: Significance in clinical medicine and psychiatry (pp. 189-196). New York: Springer-Verlag. Von Moltke, H. (1964). Decision debating: A judge's point of view. Journal of the American Forensic Association, i, 98-100.

PAGE 180

173 Walizer, M.H., & Wienir, P.L. (1978). Research methods and analysis: Searching for relationships. New York: Harper & Row. Walwik, T.J. (1969). Research in forensics: An overview. Journal of the American Forensic Association, Q, 43-48. Watzlawick, P., Beavin, J.H., & Jackson, D.D. (1967). Pragmatics of human communication: A study of interactional patterns. pathologies, and paradoxes. New York: W.W. Norton & Company. Weber, R.P. (1990). Basic content analysis. Newbury Park,CA: SAGE Publications. Weinreich, U. (1969). Travels through semantic space. In J.G. Snider, & C.E. Osgood, (Eds.) Semantic differential technique: A sourcebook. Chicago: Aldine Publishing Company, 116-129. Weiss, C.E. (1987). Variables that influence speech intelligibility most. In D.A. Thomas, & J. Hart, (Eds.) Advanced debate:Readings in theory practice & teaching. Lincolnwood, Illinois: National Textbook Company, 101-111. Welden, T.A. (1966). A congruity prediction of attitude change from debate win-loss record. Journal of the American Forensic Association,~, 48-52. Wickenden, T.H. (1988). The collective communication of social choice messages. (Doctoral dissertation, University of Pennsylvania, 1988). Dissertation Abstracts International, 49, 2433. Wiethoff, W.E. (1980). A classical rhetoric for 'powerful' argumentation. Journal of the American Forensic Association, 17, 1-10. Wildt, A.R., & Ahtola, O.T. (1978). Analysis of variance. Newbury Park, CA: SAGE Publications. Williams, F., Clark, R.A., & Wood, B.S. (1966). Studies in the dimensionality of debate evaluation. Journal of the American Forensic Association,~' 96-103. Williams, F., & Webb. S.A. (1964). Factors in debate evaluation: A pilot study. Central States Speech Journal, 15, 126-128. Williams, F., Webb. S.A., & Clark, R.A. (1966). Dimensions of evaluation in high school debate. Central States Speech Journal, 17, 15-21.

PAGE 181

174 Wilson, J.L. (1979). Competitiveness of intercollegiate debaters: A multivariate analysis. Journal of the American Forensic Association, 15, 148-155. Wilson, K.L., & Brondfield, J. (1967). The Big Ten. Englewood Cliffs, NJ: Prentice-Hall. Wood, R.V., & Goodnight, L. (1989). Strategic debate. Lincolnwood, IL: National Textbook Company. Younger, M.S. (1979). Handbook for linear regression. North Scituate, MA: Doxbury Press.

PAGE 182

BIOGRAPHICAL SKETCH Gerald R. Kish was born on April 4, 1957, in Hattisburg, Mississippi. He attended the University of Florida as an undergraduate. While there, he was a member of the speech and debate team, and he placed fifth in extemporaneous speaking at the 1979 DSR-TKA National championships. He graduated in 1980 with his baccalaureate in psychology. After experiencing private industry, he was accepted into the master's program in communication at the University of South Florida in 1983. He was also awarded a graduate fellowship and he graduated in the summer of 1984. He next enrolled in the doctorate program at the University of Florida in the fall of 1984. While a graduate assistant, he coached the speech and debate team. Two of his debaters won the DSR-TKA National Championship in 1986. In 1987, he was hired to be the forensic laboratory coordinator by the Department of Psychiatry at the University of Florida. Also in 1987, he married Ms. Lynn Herrington. In 1989, he accepted a position as a primary therapist at the North Florida Evaluation and Treatment Center. Mr. Kish expects to graduate with his Doctor of Philosophy in the Department of Communication Processes and Disorders in December of 1991. 175

PAGE 183

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. Norman N. Markel, Chair Professor of Communication Processes and Disorders 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 disser on for the degree of Doctor of Philosophy. 0 Processes and 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 dissertatio for the degree of Doctor of Philosophy. Anthony J. C rk Associate Professor of Communication Processes and Disorders 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. Professor of Law 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. Professor of Statistics

PAGE 184

This dissertation was submitted to the Graduate Faculty of the Department of Communication Processes and Disorders in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May 1992 Dean, Graduate School

PAGE 185

UNI V ER S IT Y O F FL O RIDA II I II IIIIII Ill Ill l llll lllll II IIIIII IIII II IIIII I I I IIII IIIIII Il l I I 3 1262 08553 4880