THE EFFECTS OF COURTROOM CAMERAS ON VERBAL BEHAVIOR:
AN ANALYSIS OF SIMULATED TRIAL WITNESS TESTIMONY
IN COURTROOMS USING TELEVISION CAMERAS
DONALD LEWIS SHORES, JR.
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Donald Lewis Shores, Jr.
Many individuals contributed to the research recorded
in this volume. First and foremost, recognition should
be accorded my wife, Kathy, and my two daughters, April
and Robyn, for waiting patiently and constantly encourag-
ing me in the completion of this study.
For almost 30 years my parents have encouraged me,
believed in me, and supported me in more ways than they
know. Without their help and training this volume would
One person who has probably thought and done more than
anyone to aid me in clearly expressing ideas has been Dr.
Donald E. Williams. As chairman of my supervisory com-
mittee, Dr. Williams spent countless hours reading, editing,
discussing, and encouraging in the development of this re-
search. To him I am deeply grateful and indebted.
To each member of the supervisory committee, Dr. Thomas
B. Abbott, Dr. Anthony J. Clark, Dr. Harry H. Griggs, and
Dr. Andrew J. Rosalsky, I extend my gratitude for shuffling
schedules, offering advice, and reading manuscripts in an
effort to make this endeavor as beneficial yet painless as
The following persons are but a few of the many who
cooperated to make this study possible: Circuit Judge
Theron Yawn and Court Executive Assistant Ben North
approved the use of courtrooms for the experimental phase
of the project; John Thorne and Barbara Belue from the
University of Florida Institute for Food and Agriculture
Sciences provided the edited videotape of the simulated
robbery; Joe Davis devoted three full days without
remuneration to serve as the attorney in the courtroom;
Katy Paschall and Don and Debby Nickens volunteered numerous
hours as research assistants; Don and Fran Cruse provided
computer programming guidance above and beyond the call of
duty; and five men, Milton Craven, Fred Roma, Don Hall,
George Sternfels, and Nick Rupus, volunteered their time
and abilities to role-play the presiding judge during the
experiment. To all of the above I offer my thanks for
jobs well done.
I dedicate this volume to two people who inspired me
from my earliest remembrances to work hard, do my best,
and to love and trust God, my grandparents, Carl and Cecil
TABLE OF CONTENTS
ACKNOWLEDGMENTS . . . . .. . .
LIST OF TABLES . . . . . . . .
ABSTRACT . . . . . . . . ....
I INTRODUCTION. . . . . . .
Background of the
Current Controversy . . .
Review of Relevant Literature.
Witness Testimony . . .
Effects of Videotaping on
Speech Students. . . .
and Verbal Behavior . .
Rationale for This Study . .
Research Hypotheses. . . .
Notes. . . . . . .
II METHODOLOGY . . . . . .
The Laboratory Versus Field
Setting . . . . . . .
Experimental Controls. . . .
Subjects Used in This
Study . . . . . . .
Procedure. . . . . . .
Analysis of Results. . . . .
Notes. . . . . . . .
TABLE OF CONTENTS (Continued)
III RESULTS . . . . . . . .
Self-Report Findings . . . .
Dependent Measures . . . . .
Hypotheses Test Results. . . .
Summary. . . . . . . .
IV CONCLUSIONS . . . . . . .
Limitations of the Study . . .
Implications for Future Research
APPENDIX A--THE PERSONAL REPORT OF COMMUNI-
CATION APPREHENSION . . .
APPENDIX B--INFORMED CONSENT . . . .
APPENDIX C--RESEARCH QUESTIONNAIRE. . . .
APPENDIX D--ATTORNEY'S QUESTIONS USED IN
OBTAINING TESTIMONY. . . .
APPENDIX E--SELECTED TRANSCRIPTS . . .
APPENDIX F--POST-EXPERIMENTAL QUESTIONNAIRE
APPENDIX G--DIAGRAM OF COURTROOM USED
IN EXPERIMENT . . . . .
APPENDIX H--RESULTS OF THE SURVEY OF
SUBJECT'S ATTITUDES . . .
APPENDIX I--COMPUTER PRINT-OUT OF STATIS-
TICALLY SIGNIFICANT RESULTS.
REFERENCES . . . . . . . . .
BIOGRAPHICAL SKETCH . . . . . . . .
LIST OF TABLES
1 Camera and No-Camera Group Mean Scores
for the Independent Variables Communi-
cation Apprehension (CA) and Vocabu-
lary Proficiency (VOCAB). . . .
2 Subjects Reporting Distractions in
the Courtroom . . . . . .
3 Group Mean Scores of Dependent
Variables . . . . . . .
4 The Effect of the Television Camera
on Verbal Behavior. . . . . .
5 The Effect of Vocabulary Proficiency
on Verbal Behavior. . . . . .
6 Effect of Communication Apprehension
on Verbal Behavior. . . . . .
7 Interaction Effect of the Camera
Situation and Communication Appre-
hension on Verbal Behavior. . . .
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
THE EFFECTS OF COURTROOM CAMERAS ON VERBAL BEHAVIOR:
AN ANALYSIS OF SIMULATED TRIAL WITNESS TESTIMONY
IN COURTROOMS USING TELEVISION CAMERAS
Donald Lewis Shores, Jr.
Chairman: Donald E. Williams, Ph.D.
Major Department: Speech
This study investigated the effects of the presence of
a television camera on the content of trial witness testi-
mony. The primary reason for conducting this study was to
begin providing a sound factual base for decision-making
concerning the psychological effects of cameras in the
courtroom. Specifically, this inquiry investigated possible
effects of the television camera's presence on the ability
of witnesses to present cogent testimony.
Testimony was obtained from 58 college-aged subjects
during a simulated trial situation using an actual court-
room. The testimony was analyzed to determine the type-
token ratio, mean word length, average word frequency of
the first 100 words of each subject's testimony, the total
adjusted length of the testimony, and the ratio of trivial
words used to the total number of words used by each sub-
ject. Using three independent variables (the camera situa-
tion, communication apprehension scores, and vocabulary
scores), an analysis of covariance revealed no statis-
tically significant differences among the dependent varia-
bles between subjects testifying with the television camera
in the courtroom and those testifying without the camera
However, significant (p < .05) main effects were found
between-vocabulary and average word frequency (p < .04),
and between communication apprehension and total adjusted
testimony (p < .001). A significant interaction effect
(p<.04) existed between the camera situation and communica-
tion apprehension for the average word frequency of sub-
A self-report instrument completed by subjects after
testifying in the courtroom indicated no-camera subjects
perceived more distractions (10 of 26) in the courtroom
while testifying than camera-condition subjects (9 of 32).
Four individuals mentioned the television camera as a
source of distraction.
This research and analysis led to the following con-
1. The television camera alone has no significant
effect on the lexical diversity of witness
2. An interaction of the camera situation with
an individual's personal level of communica-
tion apprehension significantly affects that
person's pattern of repeating words while
testifying in the courtroom.
3. A person's normal level of communication appre-
hension significantly affects his/her length
of testimony. Individuals possessing higher
levels of communication apprehension presented
longer testimony than individuals with lower
communication apprehension scores.
After more than 60 years, the legal debate continues
over the use of cameras in the courtroom. As this delibera-
tion continues, social scientists recognize that television
viewing occupies a significant part of the lives of our
country's citizens. A recent national poll (Roper, 1979)
indicated people in the United States rely on television
for most of their news and information. In the same vein,
United States Supreme Court Chief Justice Warren Burger in
the majority opinion in Richmond Newspapers, Inc. v.
Virginia (1980) recognized the decline in firsthand ob-
servation of courtroom trials and an increasing reliance
of citizens on the media to acquire information about legal
proceedings. Justice Burger stated that this reliance on
media in one sense "validates the media claim of function-
ing as surrogates for the public" (pp. 5012-5013).
Nevertheless, a strong argument articulated by the
Supreme Court in Estes v. Texas (1965) provided the basis
for prohibiting cameras in many state courts because of the
undeveloped technology of broadcasting and the public's
lack of familiarity with broadcasting. Later in a report
to the Florida Supreme Court, Circuit Judge Paul Baker
(1977) indicated he perceived a change in the public's view
The day has indeed arrived when television
is commonplace in our daily lives. Tech-
nical advances over the last ten years
have made equipment less obtrusive than
during Estes and the majority of the media
personnel have come to view their role as
the conscience of the community with
greater responsibility. (p. 14)
In the 1980's, a second argument appears to constitute
the primary cause for delay in implementing courtroom
camera coverage--a fear of the possible psychological
effects of the camera on trial participants. The abstract-
ness of this argument makes rebuttal difficult. In
Chandler v. Florida (1980) attorney Joel Hirschhorn em-
phasized to the court that "common sense" and "human
nature" require the striking down of the Florida rule allow-
ing cameras in the courtroom (Associated Press, Gainesville
Sun, November 13, 1980, p. LA). Hirschhorn's "common sense"
plea, in addition to his lack of supporting evidence, under-
scores the need for research to ascertain what the psycho-
logical effects of cameras in the courtroom actually are, if
such effects indeed exist.
The conflicts between our country's lawyers and judges
and members of the news media as they interpret issues
related to the Constitutional concepts of fair trial and
free press comprise a classic clash involving these two
elements of our Federal Constitution's Bill of Rights.
The first amendment's guarantee of press freedom and the
sixth amendment's assurance of the defendant's right to a
fair, public trial appear to collide in the controversy
concerning the use of cameras (both still and television)
in the courtroom (Fatzer, 1979, p. 232). The assumption
that the mere presence of cameras "could deny the defendant
the constitutional right to a fair trial" (Graham, 1978,
p. 546) lies at the center of the debate regarding the use
of cameras in the courtroom.
Since the 1930's, members of the legal community in
discussing the courtroom behavior of trial participants,
have emphasized the possibility of constraining psycho-
logical effects being caused by the presence of still
and/or television cameras. The Dean of the Harvard Law
School, Erwin Griswold (1962), suggested that even if the
noise and disturbance elements of courtroom cameras could
be completely controlled "there would still remain the
fact of broadcasting and televising, and the inevitable
psychological impact which arises from that fact" (p. 617).
U.S. Supreme Court Justice William Douglas (1960)
echoed Griswold's concern as he also alluded to the factor
of "staged performance" as a possible detriment caused by
the presence of courtroom cameras:
It is the fact of photography, the fact
that the intrusion is present, the fact
that all the principals to the trial--
judge, witness, lawyer, jury--are "on
stage" which is inescapably distracting
from the task at hand. It is the fact
that these participants are made actors
which is dangerous and disturbing.
Other members of the legal profession have expressed
concern that cameras recording judicial proceedings might
impair the ability of the witness to respond.. U.S. Supreme
Court Justice Tom Clark in the majority opinion in Estes v.
Texas (1965) wrote that the "impact upon a witness of the
knowledge that he is being viewed by a vast audience is
simply incalculable" (Gillmor, 1974, p. 457). The possible
outcomes mentioned by Clark could be collectively cate-
gorized as distortion. Agreeing with this classification,
Griswold (1962) feared the presence of cameras would cause
a measurable difference in participant output and "an
inevitable interference with the administration of justice"
While various claims could be conjectured, methodical
investigation is needed to determine what effects cameras
have on witnesses. Approaching the courtroom scene as a
communication situation, we find the circumstances promote
a continuing interaction among the participants in the situa-
tion. The witness comprises one vital element in the trial
process (Hirschhorn, 1979, p. 30; Pryor, strawn, Buchanan,
and Meeske, 1979, p. 22; Tongue & Lintott, 1980, p. 792);
the lawyers, the judge, the jury members, and other court
officials are the other parties filling specific roles. It
is the witness, however, who must enter the unfamiliar
environment of the courtroom, swear to tell the truth, and
then respond to inquiries from competing advocates in the
controversy. As many jurists have stated, cameras could
possibly change or even distort the continuing interactions
in the courtroom. The problem lies in the discovery and
the objective measurement of this possible distortion, if
it does exist.
This dissertation reports the findings of a study
designed to quantify the effects of a courtroom television
camera on the ability of the witness to present cogent
testimony. By the observation of witness testimony in a
controlled, experimental environment and the analysis of
the testimony for relative complexity, the study begins
the process of quantifying the effects of a television
camera on the witness's ability verbally to express ideas.
The complexity of the communication process makes
particular demands of research conducted in this particu-
lar communication situation. Viewing the trial as a com-
munication situation, we observe countless variables inter-
acting. With the scope of this inquiry limited to the
witness's communication behavior, this study initiated in-
vestigation of a single component or participant in the
trial process. The factors affecting even this one com-
ponent could comprise a long list. For the purposes of
this study three specific factors received attention: The
effects of the presence of a television camera in the court-
room on the witness's ability to present testimony; on
the witness's anxiety about communicating; and on the
witness's ability to use language (vocabulary).
Historical Background of the
The Bruno Hauptmann trial in 1935 prompted the
American Bar Association seriously to consider the poten-
tially obtrusive nature of courtroom media coverage (White,
1979, p. 4). The arrest and trial of Hauptmann for the
kidnapping of the Charles Lindbergh baby provided the news
media sufficient incentive to enter the courtroom and
report directly to the public. The sensational nature of
the trial encouraged intense competition among members of
the media to provide the most complete, realistic coverage.
The large number of media personnel present for the trial
(approximately 700 news reporters and over 100 photogra-
phers) presented the court with the problem of controlling
media persons' movements during the trial to prevent
obtrusive interference in the trial proceedings (White,
1979, p. 4).
Subsequently, the American Bar Association explored
the possible repercussions of the presence of the media in
the courtroom and in 1937 adopted Canon 35 as one of its
Canons of Judicial Ethics (Aspen, 1978, p. 82). Canon 35
forbade the taking of photographs in the courtroom or the
broadcasting of trial proceedings. The ABA edict stated
that such reporting was "calculated to detract from the
essential dignity of the proceedings, degrade the court
and create misconceptions with respect thereto in the mind
of the public" (Gillmor, 1974, p. 452). In 1952, the ABA
amended Canon 35 to include television coverage as well
(Wolf, 1978, p. 419).
Pressure from media representatives for a relaxing of
Canon 35 resulted in a 1955 Texas experiment in which live
proceedings of a murder trial were televised to the public
(White, 1979, p. 5). Following the experiment, the Texas
Bar Association declined to endorse Canon 35. The next
year, the Colorado Supreme Court began a series of hearings
examining the use of cameras and their effects in the
courtroom. Thus in 1956, Colorado replaced Canon 35 with
a rule allowing camera coverage of trials (Wolf, 1978,
Events in Texas and Colorado began to erode support
for Canon 35, or at least raise a reasonable doubt about
its desirability. The decade of the 1950's closed with
Colorado, Oklahoma, and Texas allowing cameras to cover
judicial proceedings at the discretion of the judges and
the media, and at least twelve other states reporting use
of cameras in isolated cases or as common practice
(White, 1979, p. 6).
The 1965 review of the Billie Sol Estes case by the
U.S. Supreme Court slowed the progress of courtroom camera
coverage. The Court reviewed Estes' 1962 swindling con-
viction and ruled that televising a state criminal trial
was a denial of due process (Wolf, 1978, pp. 420-421).
The courtroom scene during the Estes trial was graphically
described to the Supreme Court: Cables and wires "snaked
across the courtroom floor," microphones "beamed at the
jury box and counsel table," yielding the conclusion "that
the activities of the television crews and news photogra-
phers led to considerable disruption" (Gillmor, 1974,
In its Estes decision the Supreme Court outlined at
least four ways televised trials could cause unfairness.
In the majority opinion, Justice Tom Clark enumerated these
areas of concern:
(1) The potential impact of tele-
vision on the jurors is perhaps of the
greatest significance. . The con-
scious or unconscious effect that this
/-trial notoriety_7 may have on the
juror's judgment cannot be evaluated. .
/-I_7t is practically impossible to
assess the effect of television on jury
attentiveness, . and new trials
plainly would be jeopardized in that
potential jurors will often have seen
and heard the original trial when it was
(2) The quality of testimony in
criminal trials will often be impaired.
The impact upon a witness of the knowl-
edge that he is being viewed by a vast
audience is simply incalculable.
(3) A major aspect of the problem
is the additional responsibilities the
presence of television places on the
(4) Finally, we cannot ignore the
impact of courtroom television on the
defendant. Its presence is a form of
mental--if not physical--harassment,
resembling a police line-up or the
third degree. . A defendant on trial
for a specific crime is entitled to his
day in court, not in a stadium, or a
city or nationwide arena. (Gillmore,
1974, pp. 456-458)
Although the high court essentially banned camera
trial coverage in reversing the Estes decision, Justice
John Harlan in his concurring opinion spoke optimistically
about the future development of camera technology and the
public's acclimation to the presence of television cameras
which could prompt a rehearing of the issue.
Finally, we should not be deterred from
making the constitutional judgment which
this case /Estes 7 demands by the pros-
pect that the day may come when tele-
vision will have become so commonplace
an affair in the daily life of the average
person as to dissipate all reasonable
likelihood that its use in our court-
rooms may disparage the judicial process.
(Gillmor, 1974, p. 460)
In his Estes majority opinion, Justice Clark also in-
dicated cameras might find a place in the courtrooms of
It is said that the ever-advancing tech-
niques of public communication and the
adjustment of the public to its presence
may bring about a change in the effect
of telecasting upon the fairness of
criminal trials. But we are not deal-
ing here with future development in
electronics. Our judgment cannot be
rested on the hypothesis of tomorrow
but must take facts as they are presented
today. (Gillmor, 1974, p. 458)
The 1970's witnessed the acknowledgment of the poten-
tial value of the electronic media in courtroom trial
situations even though restrictions remained intact. In
August, 1972, the American Bar Association replaced Canon
35 with Canon 3A(7) of the new Code of Judicial Conduct
(Wolf, 1978, p. 421). The new canon continued the ban on
photographing and broadcasting trial proceedings but
recognized the positive uses of audio and videotaping for
presentation of evidence, maintaining court records, and
educating lawyers and judges (Kornblum, 1973; Miller and
Fontes, 1979; and Miller, Bender, Florence and Nicholson,
1974). Proponents of camera courtroom coverage have tried
almost yearly further to relax the restrictions of Canon
3A(7). Even so, as late as February, 1979, the ABA House
of Delegates rejected any change in the canon (White,
1979, p. 7).
By early 1981, the following thirteen states, by
judicial decision, allowed still and television camera
coverage of courtroom proceedings on a permanent basis:
Alabama, Alaska, Colorado, Florida, Georgia, Montana,
New Hampshire, New Jersey, North Dakota, Tennessee, Texas,
Washington, and Wisconsin (Carter, 1981, pp. 52-86). In
addition, sixteen states now allow courtroom cameras on
an experimental basis: Arkansas, Arizona, California,
Idaho, Iowa, Louisiana, Maryland, Massachusetts, Minnesota,
Nevada, New Mexico, New York, Ohio, Oklahoma, Pennsylvania,
and West Virginia (Carter, 1981, pp. 87-123, 128).
At the present time, these eight states are either
studying or considering the possibilities of allowing
cameras in the courtroom: Kentucky, Michigan, Missouri,
Nebraska, Oregon, Rhode Island, Utah, and Virginia (White,
1979, pp. 27-32). Two states have recently declined to
change existing rules forbidding cameras in the courtroom.
The Illinois Supreme Court denied a petition by the state's
News Broadcasters Association to allow courtroom broad-
casting in May, 1978 (White, 1979, p. 27). North Carolina
journalists had their petitions for relaxed photography
restrictions turned down in 1961 and 1969 (White, 1978,
The courts forbid photography in the courtrooms of
Delaware, Hawaii, South Carolina, Maine, and Wyoming.
Experts in those states foresee no change in the policy
regarding use of cameras in the courtroom in the near
future. The only state with no rule forbidding courtroom
cameras, Mississippi, has no courtroom cameras by orders
of various state judges (White, 1979, p. 33).
Currently, the prevailing guideline for controlling
cameras in the courtroom is Canon 3A(7) of the American
Bar Association's Code of Judicial Conduct. The canon
A judge should prohibit broadcast-
ing, telecasting, recording, or taking
photographs in the courtroom and areas
immediately adjacent thereto during
sessions of court or recesses between
sessions, except that the judge may
(a) the use of electronic and
photographic means for the presenta-
tion of evidence, for the perpetuation
of a record, or for other purposes of
(b) the broadcasting, televising,
recording, or photographing of investi-
tive, ceremonial, or naturalization
(c) the photographic or electronic
recording and reproduction of appro-
priate court proceedings under the
(i) the means of recording
will not distract participants
or impair the dignity of the
(ii) the parties have con-
sented and the consent to being
depicted or recorded has been
obtained from each witness
appearing in the recording
(iii) the production
will not be exhibited until
after the proceeding has been
concluded and all direct ap-
peals have been exhausted;
(iv) the reproduction will
be exhibited only for instruc-
tional purposes in educational
institutions. (Wolf, 1978,
The responsibility of whether to allow cameras in
the courtroom has been shifted back to the individual
states by the U.S. Supreme Court's ruling in Chandler v.
Florida (1981). In that case two Miami, Florida, police-
men contended that television coverage of their trial
denied them the right to a fair trial. They argued that
the mere presence of cameras had an adverse effect on trial
participants thus preventing a fair trial.
The Supreme Court ruled that the appellants had pre-
sented insufficient evidence to.support their claim of
the inherent negative effect of television trial coverage.
Chief Justice Burger wrote in the majority opinion:
No one has been able to present em-
pirical-data sufficient to establish
that the mere presence of the broad-
cast media inherently has an adverse
effect on that 5"the trial]7 process.
(Chandler v. Florida, 1981, p. 4146)
Justice White, who dissented in Estes supported an
overturning of the Estes decision and voiced his doubts
about the inherent prejudicial nature of televising
I am unwilling to assume or conclude
without more proof than has been mar-
shalled to date that televising criminal
trials is inherently prejudicial even
when carried out under properly con-
trolled conditions. (Chandler v. Florida,
1981, pp. 4148-4149)
Reacting to the Chandler decision, Washington legal
reporter Lyle Denniston (1981) commented that "Chandler
thus may stand, at this point, only for the idea that
states may allow television to come into court, but have
no constitutional duty to permit cameras" (p. 21). The
Chandler decision allows the states to decide if cameras
should be present in their courtrooms and it allows indi-
vidual judges wide discretion in limiting the camera's
access to trials (Denniston, 1981, p. 20). Although this
decision (Chandler) may be a step forward for the news
media, the Supreme Court refused to allow an absolute right
of access to media cameras. Thus, in 60 years of debate
the courts have approved a tenuous privilege to the media
allowing cameras to photograph and televise judicially
approved and designated courtroom trial situations.
Review of Relevant Literature
This study's focus on witness testimony calls for a
review of the literature dealing specifically with the
effects of television cameras on trial witnesses. Because
of the public nature of courtroom proceedings, at least
two other bodies of literature relate to this study: The
effects of videotaping on the oral performance and atti-
tudes of speech students, and the effects of communication
apprehension on verbal behavior.
Since little research has been completed assessing
the effects of cameras on the performance of witnesses,
it seems logical to investigate related environments where
the effects of cameras can be studied. With the increased
educational use of videotaping, researchers have studied
the effects of videotaping on students. One such group
was beginning speech students at the collegiate level in
classes featuring oral presentations before audiences.
The review of the research on videotaping that follows
reveals experimental results relative to the study of
witnesses' verbal behavior in the courtroom.
As a second consideration, an often mentioned effect
of the camera's presence on witnesses in the courtroom is
nervousness or stage fright. An examination of the
literature dealing with the phenomenon of communication
apprehension discloses discussion of verbal behaviors
generally attributed to individuals in apprehension arous-
ing situations. The courtroom, with its robed judge and
carefully controlled aura of decorum and solemnity, could
be considered an apprehension arousing situation. Thus,
to make a specific observation, some witnesses' verbal
behaviors may be the result of apprehension due rather to
the courtroom setting than to the presence of a camera.
First, therefore, research pertaining to witness testimony
will be reviewed; this will be followed by a review of re-
search on the two related topics cited previously.
As mentioned earlier in this study (p. 6), television
camera coverage of witness testimony became a specific con-
cern of the legal profession in 1952, when the American Bar
Association amended Canon 35 of the Code of Judicial Conduct
to exclude television cameras from the courtroom and added
a phrase indicating the electronic media's potential to
"distract the witness in giving his testimony" (Graham,
1978, p. 546). In a speech to the Colorado Law School,
Supreme Court Justice Douglas (1960) enumerated some of
the areas of concern related to witness testimony:
Will not a witness' recollection of the
facts in favor of the defendant be
colored by the publicity and public
sentiment? Again, even if a witness'
recollection is unaffected, will not
the fact that his testimony is broad-
cast make him reluctant to fully state
the facts in favor of the defendant, in
the face of adverse public sentiment?
(pp. 7, 8)
Justice Clark in Estes specifically mentioned witness
testimony (see page 9 of this study) as one of four areas
of interest regarding televised trials (Gillmor, 1974,
pp. 456-458). Judge Baker in his report on camera coverage
of the Florida v. Zamora (1977) trial shared Justice Clark's
concern, but labeled that concern "misplaced" when directed
solely toward the broadcast media. Baker (1977) suggested
that sensational trials are usually reported "in greater
detail on the printed page" than in radio or television
broadcasts (p. 15).
In a survey of Florida attorneys during that state's
year-long experiment with cameras in the courtroom,
Buchanan, Pryor, Meeske, & Strawn (1979) listed some of
the attorneys' concerns. These included fears that a wit-
ness might alter testimony "concerning matters be believes
very unpopular," or that a witness might hesitate to give
testimony which "tends to contradict a fact testified by a
greater number of witnesses" (p. 35).
Early experiments. Court officials in Waco, Texas,
conducted what may have been the first courtroom test of
cameras in 1955. Following the camera-recorded murder
trial, Judge D. W. Bartlett stated that he had observed.
no grandstanding by either reporters or witnesses. Bartlett
noted that cameras distracted no more than a court re-
porter (Blashfield, 1962, p. 432). Apparently, officials
conducted no formal survey to ascertain the effect of the
cameras on participants.
Beginning in 1956, Colorado allowed cameras in state
courts. Although no controlled experiments or surveys
were conducted, State Supreme Court Justice Frank Hall
(1962) reported six years later that the court could detect
no witnesses being distracted in presenting testimony and
that "no witness has been heard to complain" (p. 1121).
The circuit court system in Washtenaw County, Michigan,
in 1961, installed a closed-circuit television system in
one of the county's courtrooms and broadcast the pro-
ceedings to students at the University of Michigan Law
School. The circuit judge involved in the experiment
reported nothing about the process had "been observed
to be distracting to a witness or any other participant
in a court proceeding" (Blashfield, 1962, p. 429).
A December, 1974, Washington state experiment allowed
the installation of cameras in one courtroom and explained
to all trial participants the nature of the experiment.
The proceedings were taped for later closed circuit view-
ing. Following the trial, an informal survey indicated
witnesses had not been disturbed, distracted, or interfered
with in giving their testimony. Most of the witnesses
reported not even noticing the use of still cameras. They
also acknowledged not being "conscious of the operation
of the TV equipment while waiting to testify or while
testifying" (Blonk, 1974, p. 18).
An attorney involved in that Washington experiment
"felt the witnesses might have reacted differently if
they had not been advised by an attorney that the TV was
not live" (Blonk, 1974, p. 18). Thus the size of the
receiving audience as perceived by the witness may affect
responses to the cameras.
Recent experiments. A Nevada district court allowed
a Las Vegas television station to tape an attempted murder
case in 1976 (State v. Solorzano, 92 Nev. 144). The
television station set up two cameras in the courtroom
with all other broadcasting equipment placed in an
adjoining room. Following the trial, television station
personnel interviewed participants and reviewed the taped
record of the trial and discovered no irregular behavior:
There is no indication from the Solorzano
record of any of the detrimental behavior
modification--strutting, playing to the
camera, embarrassment, witness intimida-
tion, politicking, or general foolishness
--which the Estes court had so firmly
asserted would befall televised trials.
(Goldman, 1978, p. 2040)
The defense attorney in the Solorzano case told
interviewers after the trial that in his opinion the video-
taping "had no visible adverse effect on the jurors, judge,
witnesses, attorneys, or most important, Xavier Solorzano"
(Goldman, 1978, pp. 2262-2263). The prosecutor in the case
reiterated the defense attorney's remarks and praised the
positive effects of videotaping:
The psychological effect upon jurors,
witnesses, and attorneys was positive.
I did not detect anything which would
suggest that the presence of the cameras
was an unduly inhibiting influence or
that it produced a cavalier--showboating
type--response in anyone. On the con-
trary, the ever-present cameras were
conducive to better preparation by
attorneys, increased attentiveness by
jurors, and improved articulation by
witnesses. (Goldman, 1978, p. 2264)
During the internationally-televised Florida murder
trial of Ronney Zamora in 1977, Judge Baker (1977) reported
no problems regarding witness testimony, especially in
the areas of concern mentioned by Justice Clark in Estes
(p. 15). Baker described the press as having "policed
themselves beautifully! I have no complaints with the press
whatsoever" (Knopf, 1977, p. 23). After admitting to a
reporter his initial misgivings about camera coverage of
the trial, Baker said "The press has gotten a lot more
sophisticated since then J-Hauptmann7. I never made a
request that was argued with, or not complied with"
(Knopf, 1977, p. 23). A similar response resulted from
a one-trial experiment in Ohio in 1978, in which the
judge in the case reported no perceptible "posing or act-
ing for the camera. . However, one witness acknowledged
that the possibility of being on television made him more
nervous" (Wolf, 1978, p. 427).
Other than post-trial interviews with trial partici-
pants, the first widely-publicized survey research
appeared after the one-year Florida experiment (July 4,
1977 through June 30, 1978). Researchers from the Univer-
sity of Central Florida administered questionnaires to
participants immediately following five trials in 1977-78.
The researchers obtained responses from only nine wit-
nesses. Of the witnesses who chose to respond, two out
of nine perceived the television cameras had inhibited
testimony. None of the witnesses perceived adverse
effects from still cameras (Pryor, Strawn, Buchanan, &
Meeske, 1979, p. 22).
In response to questions concerning fairness of
camera coverage to witnesses, Pryor's survey found just
less than 50% (59 of 119) of the respondents felt camera
coverage was fair to witnesses. The witnesses themselves
split evenly in their responses to the question of fair-
ness: Three reported TV coverage fair to witnesses,
three reported coverage unfair, and three were uncertain
of fairness (Pryor et al., 1979, p. 23).
A related study in 1978, regarding the Florida experi-
ment conducted by the same investigators surveyed 247 of
the state's attorneys. Of the attorneys responding to the
survey, 39% (96 individuals) had experienced camera trial
coverage. Over one-half the attorneys (125) felt tele-
vision cameras had no adverse effect on witnesses, while
more than two-thirds (164) of the respondents said still
cameras had no adverse effect on testimony (Buchanan et al.,
1979, p. 35).
Although a majority of the attorneys in Buchanan's
(1979) survey observed no adverse effects on witnesses,
those attorneys with over twenty years' experience "were
almost unanimous in their agreement" that cameras ad-
versely affected witnesses (p. 35). One prosecuting
attorney said "Some witnesses were reluctant to put
themselves in a bad light. Several were afraid for their
own safety or that of their families" (Buchanan et al.,
1979, p. 35).
Buchanan et al. (1979) concluded that although most
attorneys responding to the survey did not view cameras as
"procedurally disruptive," many did express "concern about
juror distraction and the effects of live camera coverage
on witnesses' testimony" (p. 36). However, attorneys who
had experienced cameras in the courtroom were less negative
toward camera use than were attorneys without courtroom
The Supreme Court of Florida administered a survey
in July, 1978, to assess the attitudes of non-judicial
participants in the one-year experiment. Forty-four
percent of the witnesses (654 of 1,500) contacted re-
sponded to the survey (A Sample Survey of Attitudes, 1978,
Appendix A). All respondents reported being "slightly"
or "moderately" aware of the cameras' presence in the
courtroom. Almost 40% of the witnesses perceived the
presence of the cameras made them feel "slightly more
responsible for their actions" (L-Florida_7 Code of
Judicial Conduct, 1979, p. 156). More than 80% of the
witnesses felt slight distraction or no distraction by
the cameras' presence. The cameras' presence slightly
to moderately affected almost 40% of the witnesses' per-
ception of the trial's importance. Court personnel (84%)
and attorneys (85%) perceived the presence of the cameras
to affect the flamboyancyy" of witnesses not at all or
slightly. They (66% of the court personnel and 57% of
the attorneys) did feel that witnesses were slightly
inhibited by the presence of the cameras ("Florida_7
Code of Judicial Conduct, 1979, p. 156).
The Wisconsin Supreme Court appointed an eleven-
member committee to monitor and evaluate the use of audio
and visual equipment in the courtroom during a one-year
experiment beginning April 1, 1978. The evaluation com-
mittee used trained court observers to report on trial
proceedings using cameras and to interview trial partici-
The Wisconsin court observers reported no cases of
objectionable distraction by media representatives.
Through interviews the observers discovered some judges
and attorneys who felt witnesses may have experienced
more nervousness due to cameras in the courtroom but the
majority felt the quality of the testimony was undamaged
(Report of the /-Wisconsin_7 Supreme -Court Committee,
1970, pp. 11, 19, 30, 41, 54, 57). At least three inter-
view respondents observed that initial witness appre-
hension subsided or vanished during the process of testify-
ing (pp. 20, 39, 47). Two trial participants expressed
opinions about the cameras' positive influence on wit-
nesses (pp. 16, 21).
The California Supreme Court began a one-year camera-
in-the-courtroom experiment from June 1, 1980, to July 1,
1981, and enlisted a Sacramento law firm to assess the
effects of the camera coverage on the performance of
courtroom participants. The Vice-President of that law
firm, Charles Doolittle, outlined the areas of concern
regarding witnesses in a letter to this researcher:
(1) There is the question of wit-
nesses coming forward or not coming
forward because of camera presence.
That is, an individual may be "scared
off" because of camera coverage, or a
witness may surface because they L-sic.
became aware of the trial through
(2) The question of presentational
ability is being researched by our pro-
ject. Are witnesses nervous in front
of cameras to the extent that their
testimony is impaired? Is there a
problem with telling the truth or the
whole truth because of camera presence?
(3) Are witnesses reluctant to
participate again in court processes
because of their experience in front
of the cameras? (Doolittle, 1980,
Although incomplete, preliminary findings of the
California study indicate a minimal level of witness
distraction due to camera presence. However, witnesses
have generally voiced three types of concerns:
(1) Apprehension about subsequent
editing of their testimony by the broad-
(2) A feeling that some harm might
come to them because of their broadcast
testimony, either reputational, pro-
fessional, financial, or physical; and
(3) An undefined feeling of distrac-
tion and apprehension about the presence
of the camera, that is, a fear of its
unknown effects. (Doolittle, 1980,
Hoyt (1977) has conducted the only controlled ex-
periment purporting to examine the effects of the camera
on a person's ability to answer questions. Using college
students and an informational film, Hoyt attempted to
quantify the effects of an obtrusive camera, a hidden
camera, and no camera on the length, detail, and correctness
of subject responses to factual questions. The study
revealed that subjects in the obtrusive camera situation
talked longer when responding, waited less time before
speaking, and used more words in their answers than sub-
jects in either of the other conditions (hidden camera or
no camera). Findings also indicated that the testimony
of subjects in the obtrusive camera situation contained
significantly more correct information than subjects in the
other two conditions. Hoyt's findings appear to corroborate
a 1967 court opinion concerning the openness of the court-
The public notice of judicial proceeding
tends to improve the quality of testi-
mony, first by producing in a witness'
mind a disinclination to falsify, and
second, by securing the presence of
other non-parties who may be able to
furnish testimony to contradict falsi-
fiers. (Loewen, 1978, p. 512)
Although a controlled experiment, Hoyt's study failed
to provide control for important variables affecting the
subjects' responses. Certain subjects may have possessed
larger vocabularies than others and may have experienced
varying degrees of apprehension; either or both of these
factors could have affected their responses. Another
factor that might have contaminated the results concerned
the population of student subjects. All of the subjects
were enrolled in a course pertaining to the study of mass
media and society. Mass media students could have been
more familiar with and more positively disposed toward the
presence of the camera. Hoyt's study also did not place
subjects in a courtroom setting.
In summary, the research findings in the area of
camera effects on witness testimony consist primarily of
self-report survey studies. Many of the responses are
personal estimations and speculations about contexts in
which no variables were experimentally controlled. Studies
reveal some self-reported instances of witness distraction
or inhibition attributed to the presence of cameras in the
courtroom. Considered wholly, the survey results tend to
disprove, or at least provide scant support for, the con-
cerns enumerated in the Estes opinion.
The results of methodical surveys to date include a
relatively small number of witness responses compared to
the responses from judges, attorneys, jury members, and
other court personnel. Pryor's Florida survey secured
responses from only nine witnesses. Although the Florida
Supreme Court survey obtained responses from 654 witnesses,
the witness response rate of only 44% was the lowest
return rate of any participant category. So even though
the results of these surveys may appear favorable toward
camera coverage, in terms of its effects on witnesses,
a relatively small population provides the basis for
The only available experimental results imply
productive, desirable, positive effects when persons
answer questions in the presence of cameras. Subjects
responding in the camera condition appeared to provide
more correct, detailed, and accurate information in answer-
ing questions than subjects in the no-camera condition.
Effects of Videotaping
on Speech Students
Although experimental and survey findings have pro-
vided limited data on the effects of cameras in the court-
room, research with cameras in educational settings provides
information about the effects of cameras on students making
oral presentations in the classroom. The public testimony
of a courtroom witness differs from the oral presentation
of a speech student, yet certain similarities exist: Both
student and witness are verbalizing in a "public" setting;
both may be unfamiliar with the presence of cameras; and
both may experience additional fear or anxiety in the public
situation either with or without cameras present. Because
of these similarities, a brief review of the effects of
videotape use in the classroom may help provide a basis
for understanding certain aspects of the courtroom effects
of cameras on witnesses.
The use of videotape recorders (VTR) and cameras in
the classroom began in the late 1950's and early 1960's.
Research into the effectiveness of VTR in classroom situa-
tions began in earnest in the late 1960's. The research
of the early 1970's found that the use of cameras and
recording equipment in the classroom did not create any
negative reactions from student speakers. Bush, Bittner,
and Brooks (1972) found no significant differences in
levels of anxiety, exhibitionism, or reticence among
students exposed to use of the VTR and students whose
instructors did not use VTR as a teaching tool to record
and playback student presentations. Bush et al. hypo-
thesized that an "anxiety threshold" may exist where the
presence of the classroom audience of peers creates an
inherently anxious situation for the speaker (p. 129).
Therefore, the addition of the camera may add little or
no anxiety stimuli to this environment.
Other research dealing with the effects of VTR use
in the classroom revealed that students presenting more
formal, organized speeches reflected a more rigid delivery
than students not subjected to the VTR (Deiker, Crane, and
Brown, 1971, p. 141), and students speaking in the VTR
condition experienced significantly fewer nonfluencies
than non-VTR students (Deihl, Breen, and Larson, 1970,
An early study by Welke (1968) involved student
speakers appearing in three audience conditions: A visible
audience, both a visible audience and an invisible
audience (through use of closed-circuit television), and
an invisible audience only. He found that student speakers
experienced significantly less "audience-induced anxiety
over time" when speaking to either the visible audience
only, or the audience comprised of both visible members
and closed-circuit viewers than when speaking only to the
invisible audience (p. 18). The least anxiety arousing
situation approximates the audience conditions of a court-
room with television cameras. Welke also discovered that
individual speakers responded differently to the various
conditions. Some speakers reported less anxiety when
speaking to the visible audience, while others preferred
the invisible audience.
At least three studies yielded positive attitudinal
results related to VTR use. Two different studies found
that students exposed to VTR use developed significantly
more positive attitudes toward its use than students not
exposed to classroom use of the VTR (Goldhaber & Kline,
1972; McCroskey & Lashbrook, 1970). In a slightly
different finding, Bradley (1970) concluded that students
who used the VTR daily in the classroom developed a more
positive attitude toward the "intellectual atmosphere"
and "content evaluation" of the course than did students
in the no-VTR situation (p. 166).
Research related to use of cameras in educational
settings suggests the presence of a camera may promote a
more formal, conscious effort by participants to present
themselves in the best possible manner within an audience
setting considered "naturally" anxiety arousing. Perhaps
the courtroom witness may experience the "anxiety thres-
hold" mentioned by Bush et al. If so, the witness's aware-
ness of the camera in the courtroom may not increase an
already existing level of anxiety.
The finding by Welke that students reported a
reduction of anxiety over time may prove inconsequential
in this study of witnesses since a majority of witnesses
testify either only briefly or infrequently. However,
results indicating an increasingly positive attitude
of students toward VTR use over time tends to corroborate
some courtroom surveys indicating lawyers more accustomed
to camera coverage possess more positive attitudes toward
its use (Buchanan et al., 1979, p. 36).
and Verbal Behavior
Just as increased exposure to VTR equipment may pro-
mote more positive attitudes toward its use in the class-
room, a witness's positive adjustment to the courtroom and
cameras in the courtroom may aid that witness in feeling
less apprehensive or frightened in a trial situation. In
regular trial proceedings, courtroom observers frequently
mention the perceived existence of apprehension among
witnesses. Participants in trials frequently suggest
that witnesses may experience "detrimental psychological
reactions" (Blashfield, 1962, p. 434), nervousness and
tension (Gillmor, 1974, p. 459; Douglas, 1960, p. 5),
stage fright (Griswold, 1962, p. 616; Buchanan et al.,
1979, p. 35), or heightened self-consciousness (Code of
Judicial Conduct, 1979, p. 156). These observations suggest
an important variable in a study of the camera's effect on
McCroskey (1977) defined communication apprehension
(CA) as "an individual's level of fear or anxiety
associated with either real or anticipated communication
with another person or persons" (p. 78). This concept
appears to describe one type of reaction some or most
witnesses experience in the courtroom (with or without
A primary concern of this research was the effect of
CA on a witness's ability to testify, and whether this
effect was intensified by the presence of a camera.
Earlier researchers have discovered differences in the
verbal behavior of low and high apprehensive individuals.
Lerea (1956) discovered that low apprehensives pro-
duced a larger quantity of verbal output, a higher type-
token ratio, and fewer nonfluencies and speaking errors
(p. 233). Powers (1977) found high apprehensives talked
less and used more rhetorical interrogatives (non-Ah
interjections such as "you know?") than low apprehensives.
Freimuth (1976) found a significant relationship
between CA and an individual's perceived "communication
effectiveness" (p. 297). Behaviors resulting from the
interaction of CA and communication effectiveness included
high apprehensives reporting more.silence in their speech,
low ratings in ability to use language, and overall in-
effectiveness in being understood by an auditor.
Jordan and Powers (1978) investigated verbal behavior
in relation to CA and social context. These researchers
found that social distance (context) had no significant
effect on verbal behavior; however, the level of appre-
hensiveness clearly accounted for differences in verbal
behavior (p. 298). Jordan and Powers also discovered that
high apprehensives responded more methodically to stress
by altering their verbal behavior to appear more socially
acceptable. Low apprehensives may be more confident in
their verbal ability and thus feel less need to adapt to
the situation (p. 299). High apprehensives may sometimes
adjust their verbal behavior in some stressful situations
to the degree that their verbalization resembles that of
A courtroom appearance without the presence of
cameras may represent an unusually stressful situation
for some witnesses. However, the stress felt by witnesses
when television cameras are present in the courtroom could
affect their testimony in a number of ways: The witness may
say less, experience longer periods of silence, use fewer
different words to frame responses, or experience a
diminished capacity to use language; on the other hand,
the apprehensive witness may adjust verbal behavior to
approximate that of a low apprehensive witness.
Rationale for This Study
The controversy surrounding the presence of cameras
in the courtroom continues to plague legal and media pro-
fessionals. After more than 60 years of debate little factual
evidence exists to support either side of the question as
to what effects the camera has, if any, on the communication
behavior of witnesses (Fatzer, 1979, pp. 230, 239). Attor-
ney Joel Hirschhorn claimed at an American Judicature
Society convention there were "no objective scientific
studies on the effects of the presence of cameras in
courts" (Chance, 1980, p. 9). To date, each side of the
controversy bases arguments "largely on theories and
assumptions rather than facts" (Wolf, 1978, p. 427).
The possible psychological effects of cameras on
trial participants remain a major concern with jurists.
The effects of cameras on witnesses will never be known
until systematic research efforts probe the question.
Although many legal professionals voice uncertainty about
allowing news cameras unlimited access to courtrooms, most
recognize the need for continued research. Justice Burger
in the majority opinion of the Chandler v. Florida (1981)
case declared "Further experimentation is necessary to
evaluate the potential psychological prejudice associated
with broadcast coverage of trials" (p. 4146).
The absence of research may be the result of jurists'
thinking that cameras must prove their positive value and
demonstrate an absence of influences on the proceedings
before being allowed in the courtroom where researchers may
conduct in-depth experimentation (Block, 1978, p. 455;
Gerbner, 1980, p. 426). However, experimental programs in
various states indicate research in the courtroom and
judicial due process can co-exist (Baker, 1977; Blonk, 1974;
Goldman, 1978; Hall, 1962; Loewen, 1978; Netteburg, 1980;
Pryor et al., 1979).
Research into the effects of cameras on witness
testimony is sparse (Barker, Note 1; Gerbner, Note 2; Hoyt,
Note 3; Miller, Note 4; and van Peski, Note 5). Former
Kansas Supreme Court Chief Justice Harold Fatzer (1979)
held that the "actual impact of the camera's presence . .
on . witnesses remains speculative and unknown" (p. 239).
The only study tangetially related to the camera's effects
on witnesses is Hoyt's (1977), in which college students
studying mass communication generally gave more detailed,
more correct responses to questions in the presence of
The need for research specifically dealing with wit-
nesses relates to the "critical" (Hirschhorn, 1979, p. 56)
role of the trial witness. Cases are adjudicated on the
basis of evidence supplied by witnesses. The witness's
ability to recount events and frame responses in the most
appropriate language is crucial to the due process aspect
of our judicial system. At the same time, the perceived
need for the media to disperse information in an informative,
realistic manner to the public helps to create the current
controversy about the presence of cameras in courtrooms.
The public's need and right to know have been upheld
in recent U.S. Supreme Court decisions (Richmond News-
papers, Inc. v. Virginia, 1980; Gannett Co., Inc. v. De
Pasquale, 1979). By conducting carefully-controlled
experimental research, the media, jurists, and the general
public can begin to perceive what effects cameras have
on trial participants.
Based on the preceding review of related research,
this study sought to test the following hypotheses:
1. The presence of a television camera in the
courtroom will significantly affect the
verbal behavior of witnesses so that the
content of testimony from such subjects
will be judged more lexically diverse than
that of subjects testifying in a courtroom
with no camera present.
2. The acquired verbal ability (vocabulary) of
the subject will have more influence on the
lexical diversity of the subject's testimony
than either communication apprehension or
the presence of the television camera.
3. The subject's reported measure of communica-
tion apprehension (PRCA) will have more im-
pact on the lexical diversity of that sub-
ject's testimony than the presence of the
4. An interaction effect between communication
apprehension and acquired verbal ability will
have a significant influence on the lexical
diversity of the subject's testimony.
1. Barker, L. Professor of communication, Auburn
University. Personal communication, October 17,
2. Gerbner, G. Dean of the Annenberg School of Com-
munication, University of Pennsylvania. Personal
communication, May 8, 1981.
3. Hoyt, J. L. Personal communication, October 23, 1980.
4. Miller, G. R. Professor of Communication, Michigan
State University. Personal communication, October 13,
5. van Peski, D. Staff assistant, American Bar Associa-
tion, Chicago, Illinois. Personal communication,
October 21, 1980.
The Laboratory Versus.Field Setting
The social science researcher faces an almost un-
answerable dilemma in balancing the need for maximum
realism in research with the need for optimal experimental
control. Some experts concerned with the effects of tele-
vision cameras on witnesses' verbal behavior contend that
realism and control cannot co-exist in the same experi-
mental design (Gerbner, Note 1; Hirschhorn, Note 2).
Miller and Fontes (1979) addressed the question of
realism versus control and concluded that a balance be-
tween the two should be attained:
Awareness of the degree of experimental
control built into a study permits better
assessment of the extent to which the
stimulus presentation deviates from the
phenomenon being investigated. Both
generalizability and validity of the find-
ings will be affected by the amount of con-
trol imposed. Stated differently, too much
control may produce findings that are not
really generalizable to the phenomenon being
studied even though they may be treated as
highly probable knowledge claims. Conversely,
lack of experimental control may culminate
in propositions where the consequences are
attributed to the wrong antecedents or
"causes." (p. 42)
Bermant, McGuire, McKinley, and Salo (1974) criticized
most psycholegal research conducted to date for yielding
results of limited generalizability. Bermant et al.
described two properties to assess and classify research
findings: Structural verisimilitude and functional
verisimilitude. Structural verisimilitude was achieved
when research designs replicated the object or system being
studied in physical, proportional similarity. An example
would be using minutely scaled model trains or wind
tunnels to study causes of train derailments or the effects
of stress on airplane design. Functional verisimilitude was
present to the degree a study provided approximation of the
"conduct under specified conditions to the conduct of the
object or system being modeled" (p. 226).
Bermant et al. (1974) used the computer terms, "input,"
"throughout," and "output," to operationalize the concept
of verisimilitude in research. Structural verisimilitude
is concerned with input as well as demographic and psycho-
logical variables relative to the witness such as age, sex,
race, income, education, occupation, self-esteem and dogma-
tism, etc.; while functional verisimilitude refers to
throughput, or the operation of the model in relation to
an actual case, and output, or the end product of the model
compared to an actual situation (p. 227). Bermant asked
two questions in determining structural verisimilitude
of jury studies.
1. How realistic are the setting and circum-
stances wherein participants are asked to
behave as jurors?
2. How similar are the participants in the re-
search to-persons likely to serve on real
juries? (p. 227)
To determine functional verisimilitude one question was
1. How well does this model mimic the behavior
of actual juries behaving under conditions
of similar input? (p. 227)
Studying trial witness testimony using a laboratory
approach would emphasize functional verisimilitude, or the
processing and end product, i.e., the similarity of the
procedure used in the model to the actual procedure in a
real trial, and the final result (the witness's testimony)
of the model compared to the end product in a real trial.
The conduct or procedure used to elicit the testimony
would definitely affect the final form of that testimony.
This study concentrated on examining possible variables
affecting the trial witness's output or end product,
specifically the verbal testimony.
By using a laboratory approach such variables as the
type of trial (criminal, civil, etc.), the attorney-
questioning behaviors,.the distractions of the courtroom
audience, and the witness's knowledge of and relationship
to the crime could be controlled and kept constant with a
large number of witnesses. This study controlled the
variable of trial type by having all subject-witnesses
testify about the same crime. By using a videotape of
a simulated crime, each subject witnessed the same event
from a vantage point common to all participating witnesses.
Thus, the subjects' relationships to the crime and
the perpetrators and victims of the crime were held con-
stant. By using one attorney as questioner, each subject
experienced the same or similar questioning techniques.
The attorney in this study used a set of predetermined
questions for each subject. The slight variations in
the questions posed by the attorney were meant to improve
the interaction aspect and the realism of the situation.
Each subject testified before the same number of
persons in the courtroom. In order to limit possible varia-
tions of behavior by a large number of people in the
courtroom, the subjects testified with only five persons
present. Each of these confederates was instructed to
behave in the same manner during the questioning of each
By controlling a large number of variables, and by
using an actual courtroom with realistic court personnel,
this research improved the structural verisimilitude of
an earlier experiment (Hoyt, 1977). Hoyt did not use a
court setting, but only a "large room" where subjects were
seated at a table in the center of the room (p. 490). The
subjects were shown an informational, two-minute, color
film describing the operation of the Berlin Post Office.
In contrast, the study reported in this dissertation
used an actual courtroom as the setting; the subjects
viewed a videotape of a realistic convenience store rob-
bery; and the subjects in this study represented a variety
of backgrounds and interests.
Even though the in-court proceedings in this study
did not constitute an actual trial, the physical surround-
ings were purposely intended to reflect the solemnity and
decorum of an actual trial situation. The research design
used in this study attempted to combine the advantages of
a laboratory approach to control a large number of variables
and the field setting to add realism to the simulated trial
Subjects Used in This Study
This project used 58 college student volunteers who
were enrolled in an introductory speech communication
course at the University of Florida during the Spring
Quarter, 1981. Subjects received a modest bonus credit
toward their term evaluation in this speech course by
participating in the project. Although the subjects' age
range was a relatively narrow eight years (18-26), the
subjects varied in their backgrounds and interests as
evidenced by the 27 different academic areas of concentra-
tion represented in the group. Subjects reported academic
majors in agriculture, business, criminal justice and law,
education, engineering, health professions, journalism,
and various areas of liberal arts and sciences. There
were 39 women and 19 men who participated in the study.
Of those 58 subjects, 32 testified with the camera in the
courtroom while the remaining 26 had no camera in the
courtroom during their testimony. No subject had ever
testified as a witness in a trial proceeding and few (six)
had ever been in an actual courtroom.
Since it might be assumed that a majority of trial
witnesses are not college students between the ages of 18
and 26, the structural verisimilitude of this study was
limited in this respect. However, after discounting expert
trial witnesses such as law enforcement officers, doctors,
psychiatrists, etc., it might also be assumed that this
study's population of novice witnesses does reflect the
limited familiarity with court proceedings of the average
Two weeks before the experimental phase of the study,
all students in the sections of the introductory speech
course were administered the long-form Personal Report of
Communication Apprehension (PRCA) instrument (see Appendix A)
in individual class sections comprised of between 18 and
35 students. Before the administration of the PRCA, the
subjects were told they were being asked to complete a sur-
vey designed to discover how people perceive their personal
communication. Subjects were informed their scores would
be kept confidential and would have no bearing on their
The PRCA is widely accepted in the field of study
of human communication as a valid, reliable measure of
communication apprehension. After eight years of use the
PRCA internal reliability estimates have ranged from .92
to .96, while reliability judged through the test-retest
method was .82 (McCroskey, 1978, p. 201). The validity
of the PRCA has been upheld in comparison to other CA
measuring instruments. Daly (1978), in comparing eight
scales used to measure anxiety in its various forms, found
the PRCA to be one of the three most reliable instruments
among the eight measures (p. 216). The other seven scales
included Watson and Friend's Social Anxiety and Distress
Scale; Paivio's Exhibitionism instrument; Gilkinson's Per-
sonal Report of Confidence as a Speaker Scale; and three
scales developed by Beatty, Kruger, and Springhorn to measure
prior apprehension avoidance, and tension (Daly, 1978, pp.
205, 206). Daly also found the PRCA to be the most "encom-
passing" instrument evaluated within this group (p. 216).
One week before the experimental phase of the study,
potential subjects were told when and where to report in
order to participate in the study. Students wishing to
participate were required to read and sign an "informed
consent" statement indicating they understood the purpose
of the project, the experimental procedures to be used,
and were willing to participate (see Appendix B). The
only element of the study the students were not informed
of concerned the use of a television camera as an integral
factor in the experiment.
The experimenter presented the explanation of the
procedures to be used in the project to the large audience
of students attending the lecture meetings of the intro-
ductory communication course attended by all members of
the course's smaller discussion sections. During this
presentation, the prospective subjects were informed that
the study sought to learn more about the oral communication
patterns of courtroom witnesses. The students were also
informed that their involvement in the study would take
a maximum of one hour.
The experimental phase of the .study was conducted for
three consecutive days, April 15, 16, 17, 1981, from 8 A.M.
until 5 P.M. Subjects were required to report to the
Alachua County Court Building in Gainesville, Florida;
the building houses both county and district court adminis-
trative offices and several courtrooms. The court building
was constructed in 1978, thus the rooms and furnishings
are of modern design. This project used two small court-
rooms and adjoining jury deliberation rooms.
Subjects met the experimenter either individually or
in groups of two to four, in one of the two courtrooms.
If the PRCA or the informed consent form had not been filled
out, the subjects completed either or both before proceeding.
If both forms had been completed, the subject or subjects
(not more than four at once) were taken by a research
assistant to a small jury deliberation room adjoining the
After seating the subjects in the jury room, the
research assistant (not the experimenter) informed them
they were about to view a videotape of a crime. Subjects
were not told if this crime was actual or simulated. The
subjects were asked to pay close attention to the videotape
because they would be asked to testify about the recorded
events. In a real situation, witnesses would not be fore-
warned to watch carefully the details of a crime about to
be committed. Yet, in a real situation, witnesses would be
questioned as soon as feasible after the crime and later
requestioned by police investigators and attorneys; the
events witnessed would be incrementally imprinted on the
witnesses' minds. Since time constraints prevented a
similar succession of events in this experiment, the sub-
jects were forewarned to observe the crime carefully; in
this way, memory acuity might be encouraged, somewhat as
it is in real life.
In order to enhance uniformity among the individual
groups of subjects, the research assistant read the
directions from an instruction sheet to each group. After
receiving these instructions, the subjects viewed the four-
minute videotape of a replicated convenience store robbery
(Note 3). Briefly, the color videotape began by showing
the exterior front view of a small convenience store. Two
men drove up to the front of the store in a small, foreign-
made car. One man went into the store and purchased a small
item while apparently observing the physical arrangement
of the store's interior. After making the purchase, the
man returned to the car and conversed with the second man
in the car about the advisability of robbing the store.
Shortly, the two men entered the store, waited a few mo-
ments, then forced at gunpoint two store employees to open
the cash register. After one man emptied the register, he
directed the two women employees through the store into
what appeared to have been a stockroom at the rear of the
store, while the second man exited the store and quickly
started the motor of the car. The first man ran back
through the store and got into the waiting car. As the car
sped away, the two employees escaped from the back room
and walked to the check-out counter where one woman, at
the very end of the tape, telephoned the police, reporting
Following the viewing of the videotape, the re-
search assistant administered a questionnaire to each
subject containing three sections: An introductory section
seeking demographic information (name, age, sex, major in
school, place of birth, and student classification in the
university); section two containing a 50-item standardized
vocabulary test; and the third section containing a 16-item
attitude questionnaire to assess the subjects' opinions of
the judicial system and the news media (see Appendix C).
This initial questionnaire was administered after the
subjects viewed the videotape in order to more closely
approach a real situation. Normally a crime witness would
experience a period of time between perceiving the crime and
testifying. The questionnaire not only provided necessary
data for the study but also served as a mental task to in-
volve the subjects' minds for an interval of time with
something other than the content of the videotape.
The Vocabulary Test for High School Students and
College Freshmen compiled by Arthur Traxler and published
by Bobbs-Merrill Company (Note 4) was used in this project
because of its ease of administration, short administration
time (15 minutes), reliability and validity. Reliability
scores for the Traxler test range from .90 to .92. When
compared to other tests widely used in academic placement
of students, such as the ACT, ACE, and the American
Council Psychological Examination, correlations ranged
from .83 to .88 (Traxler, 1964, p. 6). In 1963, a com-
parison of scores from the Traxler vocabulary test with
the SAT scores of freshmen college students resulted in a
.65 correlation. In this study, subjects' vocabulary scores
obtained from the Traxler test were compared to their SAT
verbal scores; a .76 correlation resulted. These results
indicate a strong relationship between the Traxler test
and established verbal indices. The strong positive correla-
tion between the SAT verbal scores of subjects in this study
with the vocabulary scores obtained from the Traxler test
indicate the test is an adequate measure of verbal aptitude.
The second portion of the initial questionnaire was
a 16-item attitude measure constructed by this experimenter
to probe such areas as the subject's attitudes toward the
fairness and honesty of both our judicial system and the
news media. The attitude items also sought to assess
the subject's understanding of the operation of our judicial
system and news media. Using a Likert-type scale, the
subjects indicated their agreement or disagreement with
the nine statements pertaining to the judicial system and
the seven statements relating to the news media. Although
not used as independent variables in this study, the re-
sponses provided insight into subjects' perceived relation-
ships with the court system and the media.
Following the pretest, each subject waited outside the
second courtroom until a student role-playing the bailiff
escorted the subject into the courtroom to the witness
stand. The bailiff then administered the oath to each sub-
ject. The subject sat down and the bailiff moved to a seat
in the jury box on the witness's immediate left. A retired
senior citizen role-playing the court judge directed the
attorney to question the witness.
A third-year law student questioned each witness using
a predetermined list of 15 questions (see Appendix D). The
questions asked of each subject were purposely open-ended
to allow the subject to express original thought rather
than any more presumed, predetermined replies. Since the
purpose of this study was to assess effects of the camera
on the content of testimony, the subject was given the
opportunity to respond in a natural, free-form manner as
no attempt was made to provide directive structure for the
subject's thought pattern, thus making the camera the
primary independent variable rather than the questions or
the responses. A second consideration in formulating the
15 questions was the need to obtain an adequate quantity
of testimony for analysis.
The same questions were asked of each subject with
only minor variations to reflect individuality of the
participants' responses. The variation in questioning
included repeating questions when asked, asking follow-up
questions related to subjects' responses, and offering
subjects the opportunity to elaborate on their answers.
Although the reliability of these variations may be ques-
tioned, the desire to achieve a realistic situation pre-
vailed in the decision to simulate an actual direct examina-
tion by an attorney rather than an experimenter simply
asking a set of stock questions with minimal interaction be-
tween the subject and the questioner.
The attorney maintained the same physical distance
(approximately eight feet) and general demeanor as he
questioned each subject (see Appendix E for selected
transcripts). After responding to the questions, each
subject was dismissed by the judge and escorted from the
courtroom by the bailiff, who in turn escorted the next
subject to the witness stand.
After testifying, each subject reentered the first
courtroom and completed a short questionnaire (see Appendix
F). This questionnaire contained two parts: The first
section sought to ascertain the subject's awareness of
distractions perceived in the courtroom while testifying;
the second section contained 14 statements adapted from
the PRCA which applied specifically to the situation
featuring courtroom testimony; and a final question
asked the subjects to describe their understanding of.the
Earlier survey research assessed camera distraction
by simply asking: To what extent did the camera's pre-
sence in the courtroom distract you in giving testimony?
Although yielding information, this approach could possibly
cause respondents to view the camera as a distraction. In
this study subjects were asked an open-ended question:
"Did anything or anyone distract you during your testimony
in the courtroom?" The subject replied either "yes" or
"no" and then was asked to write a description of any dis-
tracting element, if such was perceived. This approach did
not suggest possible distractions, but allowed the subjects
individually to recognize disconcerting elements within the
The second portion of the final questionnaire con-
tained 11 statements adapted from the PRCA, and three
reworded versions of the open-ended questions in part one
of the questionnaire. Subjects responded to all the
questions of this questionnaire using a five-point Likert-
type scale. Although not a primary objective of this
study and not used as a variable in this study, the ex-
perimenter sought to assess the relationship between PRCA
scores and scores obtained from an instrument related to
the delivery of courtroom testimony as a specific communica-
tion situation. The three remaining questions of part two
served as a reliability check of questions contained in part
one. In an effort to provide control for any biased or
irresponsible statement which might appear, the final item
of this instrument assessed the subject's knowledge of the
After completing the final questionnaire, the experi-
menter debriefed each subject individually. The debriefing,
consisted of general questions repeated for each subject,
such as: "What was your general reaction to testifying in
a courtroom; did the scene in the videotape appear realis-
tic; and did the courtroom scene seem real?" The experi-
menter scanned the completed final questionnaire and asked
for clarification of any ambiguous responses. Clarifica-
tions offered by the subjects were written on the ques-
tionnaire by the subject. After this initial questioning
period, the experimenter informed each subject that the
primary purpose of the study focused on the effect of the
television camera on the content of witness testimony.
By design, subjects were randomly assigned to one of
two experimental conditions before testifying in the court-
room: The condition with a portable videotape camera pre-
sent in the courtroom and the condition with no camera
being used in the courtroom. Every five subjects experi-
enced a different camera condition, i.e., the first five
subjects testified with the camera in the courtroom, the
second group of five subjects testified in the no-camera
condition, etc. A camera operator noticeably adjusted the
camera, which was mounted on a tripod during the testimony
of each witness, and started and stopped the video recorder
before and after each subject testified (see Appendix G
for a floor diagram of the courtroom). During each sub-
ject's testimony only five other persons were in the
courtroom: The judge, the attorney, the bailiff, the
camera operator (who served as an audience member in the
no-camera condition), and a person who tape-recorded all
the testimony for analysis.
The recording of all testimony was accomplished using
a portable audio cassette tape recorder. The person re-
cording was visible to the subjects as they entered the
courtroom, but subjects could see neither this person nor
the tape recorder while testifying (see Appendix G). The
first items of the final questionnaire (Did anything dis-
tract you while giving testimony in the courtroom?) was
used to assess any effects of the tape recorder in the
courtroom as well as the presence of the camera.
Analysis of Results
The testimony elicited during the courtroom question-
ing was transcribed from the audiotape and twice checked
for accuracy by the experimenter who compared the transcript
with the contents of the audiotape. The only change made
in the transcripts from the original audiotape was to change
vocalizations such as "uh huh," "huh uh," or "em huh" to
"yes," "no," and "yes," respectively. This change was
made only where the vocalization clearly represented a
positive or negative response to a question.
An integrated set of Fortran IV programs known as
FAMULUS (Ashley & Cruse, 1980) was used to analyze the
testimony. The FAMULUS family of computer programs has
been used since its development in 1967, to organize, sort,
and enumerate various types of numerical and lexical data
(Ashley & Cruse, 1980, p. 4) (Note 5). The specific
FAMULUS program "count" yielded the total number of words
used by each subject, the total number of different words
used, the frequency of each word used, the mean word length
of all words used by each subject, and the frequency with
which each subject repeated words while testifying.
The total number of words used and the number of dif-
ferent individual words used was needed to compute the type-
token ratio (TTR) of each subject. The TTR is computed by
dividing the number of individual words (types) used by the
total number of words (tokens) in a spoken sample (Carpen-
ter, 1969, p. 162; Johnson, 1944, p. 1). The most re-
liable TTR must be computed using the same-sized samples
of not less than 100 words, since the normal person repeats
a word every 10 to 15 words spoken (Miller, G. A., 1963,
p. 89). Therefore, generally, the larger the language
sample, the smaller the number of different types used in
proportion to the total number of words.
The mean word length (MWL) for each subject was used
as an indicator of vocabulary (Johnson, 1944; Carpenter,
1969). The average word frequency (AWF) reflected how
many words a subject spoke before repeating a previously
used word. For example, an AWF of 9.50 would indicate the
subject spoke an average of nine and one-half words before
repeating a word already used during the testimony.
By searching the list of words and their frequencies
used by each subject, the total numberof articles (a, an,
the) and vocalized pauses (eh, un, um, and).was calculated.
This total (named trivial words) was subtracted from the
total number of words used by the subject to obtain the
total adjusted testimony length (TAT) for each witness.
In addition, the division of the total number of trivial
words by the total number of words used (plus trivial words)
yielded a trivial word ratio (TWR) for each subject.
The use of TAT and TWR seemed more valid to this
researcher as dependent variables than simply considering
the total number of words used by each subject. An in-
dividual might testify using a large total number of words,
yet if 25% or 33% of those words were simply vocalized
pauses or articles the lexical quality and complexity of
the testimony would be judged of lower quality than an
individual using the same total number of words but fewer
trivial words. To date, TAT and TWR have not been used
in the analysis of testimony by other researchers.
The first 100 words of each subject's testimony re-
ceived an additional analysis to determine the TTR.
Earlier studies have indicated that the first one or two
minutes, or about 100 words, of a person's verbal output
typifies that person's verbal behavior (Fairbanks, 1924,
p. 24; Jordan & Powers, 1978, p. 294). By studying word
frequencies, researchers may hypothesize about individual
and group language norms (Bradac, Konsky, & Davies, 1976,
p. 71; Johnson, 1944, p. 3).
For analysis of the research hypotheses, the sub-
jects were grouped according to their relationship to the
three independent variables: Communication apprehension
scores (PRCA), vocabulary scores, and camera condition
(either testifying with the camera or with no camera in
the courtroom). The following statistical null hypotheses
H1: The presence of the television camera in the
courtroom has no significant effect on the
subject's TTR, MWL, AWF, TAT, or TWR of
testimony when adjusting for verbal ability
H2: The verbal ability of the subject has no
significant effect on the TTR, MWL, AWF,
TAT, or TWR of testimony when adjusting
for the presence of the camera and com-
H3: The subject's measured level of communica-
tion apprehension has no significant effect
on the TTR, MWL, AWF, TAT, or TWR of testi-
mony when adjusting for verbal ability and
the presence of the camera.
An analysis of covariance was used to test the
relative impact of each independent variable upon the
dependent variables of this study. Due to unequal cell
sizes and the existence of ordinal and interval values
for two of the independent variables (communication
apprehension and vocabulary), an analysis of covariance
was chosen to evaluate the first three hypotheses. As
opposed to the analysis of variance, analysis of covariance
considers the relative values of the ordinally and
intervally-scaled independent variables of this study,
thus more clearly indicating the effect of the nominal
variable (camera condition) upon the witness's verbal
output (Agresti & Agresti, 1979, pp. 458, 459).
H4: The variables--camera condition, verbal
ability, and communication apprehension--
exhibit no interaction effects when sub-
jects testify in the courtroom.
Both an analysis of variance for unbalanced data and
an analysis of covariance was used to test for interaction
among the three variables--camera condition, verbal
ability, and communication apprehension. The ANOVA for
unbalanced data controls for the unequal cell sizes and
indicates the possibility of differing slopes of the
regression lines, thus indicating that two variables may
interact to produce the measured difference in witness
response (SAS, 1979, p. 242).
The courtroom television camera alone may or may not
produce a significant change in the verbal output of the
witness. The process of analysis presented in this chapter
seeks to assess either positive or negative effects of
the television camera on a trial witness while considering
and controlling for the possible effects of the witness's
normal level of communication apprehension and the ability
of the witness clearly to verbalize while testifying.
1. Gerbner, G. Dean of the Annenberg School of Com-
munication, University of Pennsylvania, Personal
communication, May 8, 1981.
2. Hirschhorn, J. Attorney for the defendant to
Chandler v. Florida case heard before the U.S.
Supreme Court. Personal communication, April 24,
3. The videotape used in this research was produced by
and used with the permission of the Editorial
Office of the Institute of Food and Agricultural
Sciences, University of Florida, Gainesville.
4. The Vocabulary Test for High School Students and
College Freshmen by Arthur Traxler was used in this
research with the express, written permission of
5. Computing was done utilizing the facilities of the
Northeast Regional Data Center of the State Univer-
sity System of Florida located on the campus of the
University of Florida in Gainesville.
This chapter presents the tabulated results of the
communication apprehension test, the pre-experimental
questionnaire, and the post-experimental questionnaire.
Also reported are the mean scores of the dependent measures
(type-token ratio--TTR, mean word length--MWL, average
word frequency--AWF, total adjusted testimony--TAT, and
trivial word ratio--TWR). Finally, statistical findings
in response to the study's four hypotheses are reviewed.
Each subject completed the PRCA instrument approxi-
mately two weeks before participating in the courtroom
phase of the experiment. The highest PRCA score obtained
from the 58 subjects was 104 (high apprehensive), while
the lowest score was 44. In developing the PRCA,
McCroskey (1978) used 25 questions with the highest
possible score being 125 and the lowest possible score
25. McCroskey found the mean score to vary between 73
and 75 in studies using over 12,000 college students and
4,000 adults (p. 201). The mean score for the subjects
in this study was 73.5, with 23 subjects scoring above the
mean and 35 scoring at the mean or below. The 32 subjects
randomly placed in the camera experimental situation had
a mean PRCA score of 71.16, while the 26 subjects in the
no-camera-condition averaged 76.39, thus no statistically
significant difference existed (.80 >p > .50) between the
two groups in relation to their normal communication appre-
hension levels. The mean PRCA score obtained in this
study served as a dividing point between high and low
The Traxler vocabulary test comprised the first section
of the pre-experimental questionnaire administered immedi-
ately after the subjects viewed the videotape of the simu-
lated crime. In the development of the vocabulary test,
the scores for the 50th percentile ranged between 35 and
36 for college freshmen taking the test (Traxler, 1964,
p. 9). The mean score on the vocabulary test for all sub-
jects in this study was 33.21, with a range of 45 (high)
to 14 (low). Subjects placed in the camera-condition (where
the camera was present during testimony) averaged 34.84
on the test while subjects in the no-camera-condition
(where no camera was in the courtroom during testimony)
averaged 31.19 (see Table 1). Thus, no statistically Z
Table 1. Camera and No-Camera Group Mean Scores for the
Independent Variables Communication Apprehen-
sion (CA) and Vocabulary Proficiency (VOCAB).
Camera CA VOCAB CA VOCAB
Situation Level Level Scores Scores
High High 94.2 38.5
Low 86.0 23.5
Low High 61.1 39.9
Low 54.5 26.3
High High* (100.0) (45.0)
Low 93.4 28.1
Low High* 61.2 37.2
Low 66.8 25.0
*Only one subject in this cell.
significant difference (.20> p > .10) in vocabulary scores
existed-between the camera group and the no-camera group.
On the basis of the results of the PRCA scores and
the vocabulary scores, subjects were divided into eight
groups. There were two camera situations (with and without
the camera present), two levels of CA (high and low in
each camera situation), and two levels of vocabulary pro-
ficiency (a high and low vocabulary group within each
Section Two of the pre-experimental questionnaire
contained a 16-item attitude survey. The first nine ques-
tions attempted to assess the subjects' attitudes toward
our judicial system, and the remaining seven questions
assessed the subject's attitudes toward the newsmedia.
The items contained in this survey were constructed by the
experimenter. The Kendall tau b test of association was
used to assess the relationship of positively and nega-
tively phrased forms of the statements. All positive and
negative pairs of the statements were found to be asso-
ciated with probability of at least p < .01.. This section
of the questionnaire was included primarily to provide the
experimenter information on the subjects' general attitudes
toward the two entities (the judicial system and the news
nedia) germane to this study. However, even though the
results of the survey were not used to assist in inter-
preting a dependent variable in this study, the results
may help explain some of the other findings (see Appendix
H for survey results).
Immediately after testifying in the courtroom, sub-
jects completed a questionnaire containing four open-
ended questions and 14 Likert-type statements adapted
from the PRCA. The four open-ended questions asked the
subjects if they had perceived any distractions while
testifying, the nature of any distractions observed, the
effect of nervousness on their ability to perceive dis-
tractions, if such were perceived, and their perception
of the study's purpose. The experimenter reviewed each
subject's responses to these four questions during the
debriefing session in an effort to clarify the meaning of
unclear written responses.
In reply to the question asking subjects if they had
observed any distractions, 19 subjects indicated that some-
thing had distracted them while testifying. Nine of the 32
subjects in the camera-condition noted distractions, while
10 of the subjects in the no-camera-condition reported
Three of the nine distraction-reporting subjects in
the camera-condition indicated the camera was the primary
distractor while one subject named both the camera and the
person operating it as the distracting factor. The person
in the audience who operated the camera was identified
as the distracting factor by two additional camera-con-
dition subjects. The person who sat in the audience
portion of the courtroom during the no-camera-condition
was mentioned as the source of distraction by five sub-
jects in that situation.
Other factors mentioned by subjects as distractions
and their frequency of mention included: The tape-recorder
(2), the unfamiliar atmosphere of the courtroom (3), some-
one moving outside the courtroom (1), the bailiff clearing
his throat (1), extraneous noise outside the courtroom
(1), the presence of the microphone (1), an unexplained
sense of nervousness (1), and the "nice-looking" attorney
(1), Some subjects mentioned more than one source of dis-
traction, thus explaining the larger number of distracting
factors than persons distracted.
Two of the three subjects who named the camera speci-
fically as a distracting agent indicated the presence of
the camera hindered their ability to testify. All three
subjects disturbed by the camera felt the distraction
affected the content of their answers.
Five of the nine subjects mentioning distractions in
the camera-condition were classified "high apprehensive"
with PRCA scores ranging from 83 to 103. Four of the nine
subjects were in the high vocabulary group.
In the no-camera-condition, two of the 10 subjects
distracted were high apprehensive and low vocabulary.
The majority of the subjects (8) distracted in the no-
camera-condition were low apprehensive. Five of those
eight were in the high vocabulary group (see Table 2).
Table 2. Subjects Reporting Distractions in the Court-
Camera CA VOCAB Subjects
Situation Level Level Distracted
High High 4
Low High 4
High High 0
Low High 5
In summary, 28% (9 of 12) of the camera-condition
subjects reported distractions) in the courtroom while
testifying. In the no-camera condition 39% (10 of 26) of
the subjects reported distractionss. Of the nine dis-
tracted in the camera-condition, only three reported the
camera as the sole source of distraction.
Section Two of the post-experimental questionnaire
compared 10 selected statements from the PRCA with 10
adapted forms of the statements to reflect the specific
situation of the courtroom. The items selected from the
long-form PRCA for this specific application are similar
in nature to the short-form PRCA (McCroskey, 1978, p. 201).
The mean composite score for all items adapted to the
specific courtroom situation was 2.64 (with 1.0 indicating
low apprehension), while the composite mean for the similar
items on the PRCA was 3.08. However, only five of the 10
pairs yielded statistically significant (p < .05) levels
of association (using the Kendall tau b test of associa-
The final question of the post-experimental question-
naire asked subjects their perceptions of the study's pur-
pose. In an effort to verify the meaning of responses to
this question, the experimenter specifically asked each
subject to explain their responses in detail. A total of
six subjects stated the purpose of the study correctly on
the questionnaire. Five of the six had experienced the
camera-condition while testifying; three of these five
subjects indicated they were distracted while testifying.
Two of the six subjects indicated they determined the
purpose of the study only after completing the preceding
parts of the final questionnaire.
The five dependent measures examined in this study
included the type-token ratio (TTR), the mean word length
(MWL), the average word frequency (AWF), the total ad-
justed testimony (TAT), and the trivial word ratio (TWR).
The mean TTR for the entire subject sample was 0.592.
The mean TTR for the subjects in the camera-condition was
0.5919, while the mean for no-camera-condition subjects
was 0.5915. Thus, the verbal output as measured by the
TTR revealed no statistically significant difference be-
tween the groups of subjects in the two camera situations
(t = .305, df = 57, .80 > p > .50). Subjects reporting
low levels of CA had higher TTR scores than high appre-
hensives (0.600 to 0.579), but the difference was insigni-
ficant (t = .395, df 57, .80 > p > .50). The difference
in TTR scores among high and low vocabulary subjects
(0.603 to 0.577) was also insignificant (t = .495, df = 57,
.80 >p > .50).
The testimony of the 58 subjects in this study had an
average MWL of 4.32. No statistically significant dif-
ferences existed in the scores of the various independent
variable subgroups. No-camera subjects had a slightly
higher MWL than camera witnesses (4.33 to 4.31). Sub-
jects reporting low levels of CA had a higher MWL than
higher apprehensive subjects (4.34 to 4.29). Individuals
with higher vocabulary scores also had a slightly higher
MWL than those subjects scoring lower on the vocabulary
test (4.33 to 4.31).
The mean AWF for the entire subject sample was 9.54 with
no significant differences among subgroups of witnesses. Sub-
jects repeating less words, thus recording a higher AWF were
camera-condition subjects (9.64), low apprehensives (9.84),
and high vocabulary subjects (10.11). These averages
compare to 9.43 for no-camera subjects, 9.10 for high
apprehensives, and 8.85 for low vocabulary individuals.
The adjusted testimony length of subjects varied from
subgroup to subgroup, but again the differences were statisti-
cally insignificant. The mean TAT for all subjects was
553.4 words. Subgroups yielding the longest testimony were
the no-camera subjects (558.5 to 549.2 for camera subjects)'
low apprehensives (638.1 to 424.4 for high apprehensives),
and subjects.with high vocabulary scores (625.9 to 464.0
for low vocabulary individuals).
The ratio of trivial words to the total number of
words in subjects' testimony also exhibited no significant
differences from group to group. The overall TWR was 0.188.
Unlike the other dependent variables reported, a lower
TWR score would indicate more diverse, complex testimony.
The no-camera group had a slightly lower TWR than the
camera group (0.186 to 0.189). Virtually no difference
existed between high and low apprehensives (0.1877 to
0.1881). High vocabulary subjects had a lower mean TWR
(0.183) than low vocabulary subjects (0.194) (see Table 3).
Table 3. Group Mean Scores of Dependent Variables.
Camera CA VOCAB
Situation Level Level TTR MWL AWF TAT TWR
High High .588 4.27 9.97 467.5 .179
Low .585 4.33 9.07 432.0 .198
Low High .598 4.32 9.97 622.8 .189
Low .585 4.35 8.67 553.3 .195
High High* (.570) (4.40) (9.50) (526.0) (.176)
Low .570 4.27 8.56 383.9 .189
Low High .628 4.40 10.51 748.2 .176
Low .577 4.33 9.24 570.2 .199
*Only one subject in this cell.
One subject group registered the best mean scores
on all five dependent variables. The low-apprehensive,
high-verbal, no-camera group yielded a mean TTR score of
0.628, a MWL of 4.40, an AWF of 10.51, a TAT of 748.2,
and a TWR of 0.176. The lowest TTR was produced by the
no-camera, low-apprehensive, low-vocabulary group (0.570).
The lowest MWL (4.27) was registered in both the camera
and no-camera high-apprehensive groups. The camera-
condition, low-apprehensive, low-vocabulary group produced
the lowest AWF (8.67). The lowest TAT (383.9) was pro-
duced by the no-camera, high-apprehensive, low-vocabulary
group. The camera-condition, high-apprehensive, low-
vocabulary group yielded the highest TWR (0.198).
No significant differences were recorded in the
scores of male and female subjects. Males had slightly
higher scores on TTR (0.613 to 0.481), MWL (4.42 to 4.18),
AWF (10.22 to 9.22), and TAT (622.1 to 419.9). However,
females used a slightly smaller ratio of trivial words
than males (0.183 to 0.188).
By examining the mean scores, little difference can
be observed among the various levels of the subjects as
grouped according to the independent variables. However,
generally, the higher (indicating more verbally complex)
scores in relation to the three dependent variables were
produced by the groups scoring higher on the vocabulary
Hypotheses Test Results
The first hypothesis tested concerned the effect of
the camera on the dependent variables:
H1: The presence of the television camera in
the courtroom has no significant effect
(p < .05) on the subject's TTR, MWL, AWF,
TAT, or TWR contained in testimony when
adjusting for the effects of CA and vocabulary
Using an analysis of covariance test to evaluate
the first hypothesis, the addition of the television
camera to the courtroom situation accounts for a statis-
tically insignificant amount of the variance in the de-
pendent variable TTR (p < .13); similarly with the MWL,
the camera accounts for little variance in the scores
(p < .98). The camera also appears to have little effect
on AWF scores (p < .64), TAT scores (p < .41), or TWR
scores (p < .61). Therefore, when considered alone, the
presence of the camera appears to have a statistically
insignificant effect on verbal behavior. Thus, the first
hypothesis can not be rejected at the .05 level (see
Table 4. The Effect of the Television Camera on Verbal
Variable F (1, 57) P <
Type-token ratio 2.33 0.133
Mean word length 0.00 0.976
Average word frequency 0.23 0.637
Total adjusted testimony 0.70 0.406
Trivial word ratio 0.26 0.611
The second hypothesis tested concerned the effect of
the subject's verbal ability (as measured by the Traxler
vocabulary test) on the dependent variables.
H2: The vocabulary proficiency of the subject
has no significant effect (p < .05) on the
TTR, MWL, AWF, TAT, or TWR within a sub-
ject's testimony when considering the
possible effects of the presence of the
camera and the level of the subject's CA.
The measured effect of a subject's vocabulary pro-
ficiency on the resulting TTR of the testimony was insigni-
ficant (p < .20) when the presence of the camera and the
subject's level of CA were considered. When examining
the effect of vocabulary proficiency on MWL, the effect
was also insignificant (p < .49). Vocabulary proficiency
had an insignificant effect on TAT (p < .08) and TWR
(p < .99). However, the subject's verbal ability appears
to significantly affect the AWF (p < .04) (see Appendix I).
Considering these findings, the second hypothesis can be
rejected at the .05 level only for the dependent variable
AWF (see Table 5).
Table 5. The Effect of Vocabulary Proficiency on Verbal
Variable F (1, 57) P<
Type-token ratio 1.68 0.200
Mean word length 0.48 0.493
Average word frequency 4.38 0.041*
Total adjusted testimony 3.11 0.083
Trivial word ratio 0.00 0.990
*Significant at the .05 level.
The third hypothesis tested the effect of CA on the
verbal behavior of witnesses.
H3: The subject's measured level of CA has no
significant effect (p < .05) on the TTR,
MWL, AWR, TAT, or TWR of the subject's
testimony when regulating possible effects
of the camera's presence and the subject's
When adjusting for the possible effects of verbal
ability and the presence of the camera, the subject's CA
accounts for an insignificant amount of the variance in
the TTR (p < .09). CA also has an insignificant effect on
MWL (p < .58), AWF (p < .30), and THR (p < .32). CA does
appear to have a significant effect on TAT (p < .001)
(see Appendix I). Because of these findings, the third
hypothesis can be rejected only for TAT (see Table 6).
Table 6. Effect of Communication Apprehension on Verbal
Variable F (1, 57) p <
Type-token ratio 2.95 0.092
Mean word length 0.31 0.581
Average word frequency 1.08 0.303
Total adjusted testimony 12.42 0.001*
Trivial word ratio 1.00 0.322
*Significant at the .05 level.
The fourth hypothesis sought to test for possible
interaction effects among independent variables.
H4; The variables (camera situation, verbal
ability, and CA) exhibit no interaction
effects when subjects testify in the
An analysis of variance for unbalanced data (designs
yielding uneven cell sizes) revealed no significant inter-
actions among the independent variables. In fact, a least
square means comparison indicated only two groups of the
eight to be significantly different at the .05 level.
Both groups were in the no-camera condition. The no-camera,
low-apprehensive, high-vocabulary group had significantly
different mean TTR scores (p < .02) than the no-camera,
high-apprehensive, low-vocabulary group.
Using an analysis of covariance to assess possible
interaction effects revealed a possible trend toward inter-
action when considering TTR scores. An interaction be-
tween the camera situation and CA yielded a probability
of p < .16. Similarly, the AWF scores were affected by
the interaction of the camera situation and CA (p < .04)
(see Appendix I). No other interactions among variables.
approached significance. Thus, the fourth hypothesis may
be rejected at the .05 level only for an interaction effect
between the camera situation and CA in relation to the
dependent variable AWF (see Table 7).
Table 7. Interaction Effect of the Camera Situation and
Communication Apprehension on Verbal Behavior.
Variable F (1, 57) P <
Type-token ratio 2.07 0.156
Mean word length 0.01 0.904
Average word frequency 4.24 0.044*
Total adjusted testimony 0.35 0.554
Trivial word ratio 0.00 0.962
*Significant at the .05 level.
By dividing the subjects into two groups (high and low)
in relation to each of the two variables, CA and vocabulary
proficiency, the researcher's ability to detect significant
differences between groups decreases as the distinction
between groups decreases. However, as evidenced by the
mean scores reported for the five dependent variables, no
statistically significant differences existed among the
mean scores of the eight groups of subjects.
The presence of the camera, when considered as a
single factor, had little or no effect on the verbal be-
havior of witnesses as far as possible effects can be
measured statistically. However, the camera situation
and CA appeared to interact to create a significant effect
on the AWF of subjects.
In this experimental situation, CA, as an isolated
factor, had a statistically significant effect on the
witnesses' length of testimony--the higher the subject's
CA score the shorter the testimony. When considering TTR
scores, CA approached a significant effect. As might be
expected by the very nature of AWF, vocabulary proficiency
had a statistically significant effect on this dependent
The question of the television camera's effects
on courtroom trial participants has not been answered.
with any real degree of certainty for almost 30 years,
since the American Bar Association included television in
its prohibition of camera trial coverage in 1952. Through
this study, an effort was made to continue.the process
of researching the effects of the television camera on
courtroom communication, and specifically, its effects
on the content, or lexical nature and diversity, of wit-
The first conclusion to be drawn from this study is
that the presence of the television camera, in this
experimental situation, had no perceivable effect on
the overall lexical nature or diversity of witnesses.
The research hypothesis suggested the presence of the
camera would enhance the verbal behavior of witnesses.
This study found no evidence to support that hypothesis
directly. The mean TTR and AWF scores were slightly higher
for camera than no-camera subjects (TTR 0.5919 to 0.5915,
AWF 9.64 to 9.43), but they were not statistically signifi-
cant. No-camera subjects used slightly longer words than
camera subjects (MWL 4.33 to 4.31), but the difference be-
tween the groups did not reflect statistical significance.
It should be noted that the higher mean vocabulary scores
for the camera group may explain the differences found
among these dependent measures.
The courtroom camera may affect the testimony of a
particular segment of the population. The high-appre-
hensive, high-vocabulary camera group reported three of
the four instances of actual camera distraction. Although
the dependent measures for this group were not significantly
different from the study's population as a whole, the
perceived distraction of the camera may have affected
this group's ability to testify in some manner undetected
by the dependent measures. Hoyt (1977) discovered the
presence of the camera enhanced the quality of subjects'
responses, but the findings of this study neither conform
nor refute his results.
The statistically significant finding that an inter-
action exists between the presence of the camera and the
individual's level of communication apprehension tends
to support the conclusion that highly apprehensive,
articulate individuals may experience distraction caused
by their perception of the camera's presence, yet this
perceived distraction may not be reflected in the lexical
nature of their testimony. As suggested by Jordan and
Powers (1978), high apprehensives may be more aware of
the stressful situation and attempt to adjust their be-
havior to approximate that of low apprehensives. The high
apprehensives in the camera-condition may have been more
cognizant of various distracting factors in the environ-
Eight of the 19 subjects reporting distractions during
their periods of testifying indicated the distractions
either hindered their testifying or caused them to alter
the content of their testimony. This agrees somewhat with
earlier responses to item 25 of the PRCA, "I would enjoy
presenting a speech on a local television show." The
mean score of all subjects on this item was 3.56, yet in
the post-experimental questionnaire, when asked if they
would answer questions in the same manner if their court
appearance was being televised to the Gainesville com-
munity, the mean score of all subjects was a more positive
2.33. This may indicate that having experienced the
courtroom situation, witnesses react to a more concrete
stimulus more positively than to the abstract possibility
.of appearing on television. This difference in mean
scores may also be the expression of relief after ex-
periencing a stressful situation.
The second research hypothesis suggested that vocabu-
lary proficiency would have a greater effect on the lexical
nature and diversity within the testimony than either com-
munication apprehension or the presence of the television
camera. The findings of this study only partially support
A stepwise regression procedure (SAS, 1979, p. 391)
used to measure the relative effect of each independent
variable revealed vocabulary proficiency the best pre-
dictor variable for MWL and AWF. Vocabulary was the second
most important predictor for TTR and TAT. Although the
stepwise procedure indicated vocabulary as an important
variable in predicting the dependent variable scores, the
vocabulary proficiency accounted for only between 1.5% to
9% of the total variance recorded in this study's dependent
measures. Thus, vocabulary proficiency had more effect on
two dependent variables than either camera situation or CA,
but accounted for a relatively small amount of the differ-
ence in subjects' responses. A conclusion that may be
drawn from this data is that the three independent varia-
bles leave much of the variance in TTR, MWL, AWF, TAT,
and TWR scores unaccounted for, thus suggesting that some
variable(s) not reported in this study may explain better
the differences in subject response. A second conclusion
might be that each level of the independent variables
should be more precise.
The third research hypothesis postulated that the
CA level accounted for more variation in witnesses' lexical
nature and diversity than the presence of the television
camera. As with the second hypothesis, the data resulting
from this experiment only partially supports this assumption.
The stepwise regression procedure indicated CA was
the best predictor variable (of the variables included in
this study) for TTR and TAT scores. Although statistically
insignificant for TTR, CA had a significant effect
(p <.001) on TAT scores. In both TTR and TAT the effect
of CA was negative, i.e., the higher the CA score, the
lower the TTR and TAT scores. Thus the conclusion that
witnesses with lower CA levels would provide the more
complex, complete testimony than more highly apprehensive