39 "A\‘. «final... LIBRAi-i Y Michigan State .' University This is to certify that the thesis entitled THE EFFECTS OF CAMERA SHOT AND WITNESS TYPE ON JURORS' RESPONSES TO A VIDEOTAPED DEPOSITION presented by Edmund Paul Kaminski has been accepted towards fulfillment of the requirements for Ph . D . degree in Communication Major professor Au ust 8, 1978 Date g 0-7639 THE EFFECTS OF CAMERA SHOT AND WITNESS TYPE ON JURORS' RESPONSES TO A VIDEOTAPED DEPOSITION BY Edmund Paul Kaminski A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1978 Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Doctor of Philosophy degree. 4;} 7/ Guidance Committee: £ZAQQZ X fizz/(ZR f, Chairman ABSTRACT THE EFFECTS OF CAMERA SHOT AND WITNESS TYPE ON JURORS' RESPONSES TO A VIDEOTAPED DEPOSITION BY Edmund Paul Kaminski The advent of videotape in the legal system has aroused many concerns for jurists. One concern deals with the type of camera shot employed when videotaping testimony. Currently, there are few rules governing taping and presentation of testimony. Moreover, there is a paucity of research that has examined the effects of different camera shots on viewers' per- ceptions. The purpose of this thesis was to examine the effects of three different camera shots (close-up; medium; long) and two different wit- ness presentational styles (strong; weak) upon jurors' responses. Spe- cifically, the dependent measures included jurors' perceptions of the witness': (l) composure, (2) credibility, (3) authority, and (u) char- acter. In addition, measures of the following were also obtained: (1) information retention, (2) juror interest, (3) verdict, and (u) award. With the assistance of legal experts, a trial deposition was se— lected. The deposition was from a defendant who was accused of negli- gence in an industrial accident. The deposition, which was approximately 30 minutes in length, consisted of cross-examination by the plaintiff's attorney. It did not contain direct examination by the defendant's attorney. Professional actors played the roles of the witness and the defense attorney. An actual attorney played the role of the plaintiff's attorney. Edmund Paul Kaminski The type of witness manipulation consisted of the same actor playing two different roles: (l) a strong witness and (2) a weak witness. The presentational style of the strong witness was characterized as fluent, assertive, and attentive. The presentational style of the weak witness was characterized as uncertain, hesitant, fumbling, and inattentive. The actor was trained to emit verbal and nonverbal cues derived from previous research which were found to elicit impressions of these presentational styles. The deposition was videotaped in color. Three cameras were used simultaneously in order to achieve the three levels of camera shot. Moreover, the deposition was taped twice -- once for each witness type. One hundred and sixty-two undergraduate students role-played jurors for this study. Subjects read a trial synopsis, which was used in order to place the deposition in context. The videotaped deposition was then shown. Following the deposition, subjects completed a questionnaire. The results revealed significant main effects for the presentational style of the witness. Specifically, the strong witness was perceived to be significantly more composed, qualified and dynamic than the weak wit- ness. Moreover, subjects exposed to the strong witness retained signif- icantly more information and expressed greater interest than subjects exposed to the weak witness. No significant main effects were obtained for camera shot. However, a few significant camera shot x witness type interactions were obtained. The strong witness was perceived significantly more authoritative in the close-up shot than in the long shot. Perceptions of the weak wit- ness' authority did not differ significantly across camera shots. The weak witness was perceived significantly more composed in the long shot Edmund Paul Kaminski than in either the medium or close-up shots. No significant differences in the perceptions of the strong witness' composure were obtained across camera shots. Finally, subjects exposed to the weak witness retained significantly more information when presented with a long shot than with either a medium or close-up shot. Mean information retention scores did not significantly differ across camera shots for subjects exposed to the strong witness. While significant interactions were obtained, the largest of these effects accounted for only five percent of the variance. From a practi- cal standpoint, five percent is not very much. Therefore, in the final analysis, given a strong or weak witness, it probably does not matter whether a close-up, medium, or long shot is used to videotape testimony. Due to some limitations with the present study, this recommendation is offered cautiously. ACKNOWLEDGMENTS I would like to express my appreciation and gratitude to the members of my committee, Dr. Erwin P. Bettinghaus, Dr. Vincent R. Farace, Dr. Norman E. Fontes, Dr. Eugene Jacobson, and Dr. Gerald R. Miller, for their support, guidance, and patience. To Gerald R. Miller, I extend my deep appreciation and friendship for his understanding, guidance, and friendship. He has been a constant source of intellectual inspiration and emotional support. Truly a man of patience, he has treated me with fairness and kindness. To Norman E. Fontes, I extend my friendship and thanks fer his help, patience, understanding, and Mexican temper. Hopefully, he'll remain in academia long enough to read this. To Michael Sunnafrank, I extend my thanks and friendship for putting up with my insanities during the final year and for his support. Over a period of four years, there are so many people who have helped either by being supportive or by adding humor to my existence by simply being themselves. Due to space limitations, I cannot explain in detail the accomplishments of every individual. So, in keeping with non- tradition, I'll mention them below and hope that they'll know what they did. Thanks is extended to: The Montana Miracle Machine, The Animal, The Condor, The Enforcer, The Mole, The Toke, The Greek, The Magnet, Gar Face, The Kid, Johnny Research, Cosmo, Miz Laura, Slick, M2, M3, Mr. Affect, Beech Street Fats, The Califbrnia Fox, Jungle Man, Andy ii Pandy, Mr. Interpersonal, The Buzzed Czars, Ma Bell, Merkins' Marauders, Squeaky, Ms. Hope, Chuckles, Bozo the Clown, Dr. of Love, Paul Revere, The Goonies, Captain Video, The Women's Lib Group, Claw Woman, Leepin' Leena, Lemonade, Organizational Man, Pandora, and Captain Communication. I would also like to extend my thanks to the continuing saga of The Toke vs. The Goonies, for adding humor and excitement. I understand the movie will be released as soon as the credits are worked out. I'd also like to thank Miz Laura for helping me find my sanity and helping me lose it at the same time. Special thanks goes to Judy Mead for typing this thesis and being very patient. I would like to thank my family for their faith and support. Finally, I would like to express my gratitude for having a loving wife and son who have been especially understanding and supportive. Thank y'all. iii TABLE OF CONTENTS Page LIST OF TABLES .................................................... vi LIST OF FIGURES ................................................... viii Chapter I INTRODUCTION ....... . .................... . ............... 1 Relevant Research .............................. . ........ 7 Hypotheses....... ....................................... 16 II METHODS AND PROCEDURES ............... . .................. 26 Taping the Witness ................. ....... .............. 27 Trial Synopsis.. ....... . ...................... . ......... 28 SllbjectSOOOOOOOOOOOOOOOO ..... O ..... .... ........... .00... 28 Demographic Characteristics.............. ............... 29 Power Estimation....... ............................. .... 31 Measurement Techniques .................................. 32 III RESULTS. 0 OOOOOOOOOOOOOOOOOOOOOOOOO O OOOOOOOOOOOOOOO 0 .0. O O ‘41 The Effects of Witness Type...................... ....... 41 The Effects of Camera Shot for the Strong Witness ....... Q7 The Effects of Camera Shot for the Weak Witness ......... 53 IV DISCUSSION .............................. .. ..... ......... 59 The Effects of Witness Type ......... . ................. .. 59 The Effects of Camera Shot ........... . ........ . ......... 63 sumarYOOOOOOOOOOOOOOOOOO...COOO ..... 0...... ...... 0.0... 67 Limitations....... ................. . ............ ........ 68 Future Research... ................. ........... .......... 7n Recommendations. .......................... . ............. 76 APPENDICES APPENDIX A INTRODUCTION TO JURORS ................ . .............. ... 77 iv TABLE or CONTENTS (cont'd.) Pagg_ APPENDIX B TRIAL SUMMARY ......................................... 78 c QUESTIONNAIRE ON JURY SIZE ............ . ............... so LIST or REFERENCES ................................................ 102 Table 10 ll 12 13 LIST OF TABLES Frequencies of Subjects' Demographic Characteristics ....... Means and Analysis of Variance Summary of Subjects' Age.... Factor Loadings of Composure Items ......................... Factor Loadings of Credibility Items...... ................. Factor Loadings of Authority and Character Items ........... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Composure. ........ ......... ...... .......... ....... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Qualification..... ...... .......... ..... . .......... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Dynamism.00...........OOOOOOOOOI... ...... O... ..... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on Retained Information ....... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of Juror InterestOOOOOOOOOOOO...00.000000000000000.................. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' safetYOOOOOOOOOOO ........ 0.0.0.0.... ...... O ..... .0 Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' AuthoritYOOOOOO.....0.........OOOCOOOOOOOOOOO ..... Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Character.‘0....0.00000.........OOOOOIOOOO... ..... vi Pa_g_e_ 30 31 31+ 34 36 1+2 an #5 #6 H8 1+9 50 51 Table la 15 16 17 LIST or TABLES (cont'd.) Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on Jurors' Award to the PlaintiffOOOOOO ........ ......OOOOOOOIOOOOO ...... O 0000000000 Frequencies of Verdict for the Strong and Weak Witnesses... Frequencies of Verdict for the Three Camera Shots of the Strong WitneSSOIOOO OOOOOOOOOOOOOO ......OOOOOOOOO. 0000000000 Frequencies of Verdict for the Three Camera Shots of the Weak Witness .................... .. ..... . ................... vii 52 52 53 55 Figure LIST OF FIGURES Graphic Representation of Witness' Composure as a Function of Camera Shot and Witness Type........ ...... .. Graphic Representation of Information Retention as a Function of Camera Shot and Witness Type.... ............ viii 57 58 Chapter I INTRODUCTION The advent of videotape in the legal system has aroused many concerns for jurists. One concern deals with the technical aspects of the use of videotape (Bermant, McGuire 8 Chappell, 1975). From a technical stand- point, the concern encompasses the quality of the equipment, the skills of the technicians, the production techniques applied, and the editing of the videotape (see Doret, 1974). Currently, there are few rules governing taping and presentation of testimony. For example, Ohio's Superintendence Rule 15 stipules that standard one-half inch videotape equipment constitutes the standard for filming and playback of testimony and other evidence. However, the ruling allows for deviations from the standard as long as compatible equipment is supplied or the original tape is converted such that it is compatible with the standard. The only other requirement is that there must be a minimum of one monitor having at least a 1n inch screen for playback to the jury. Clearly, Ohio's Superintendence Rule 15 allows the litigants a good deal of freedom in deciding how and where to videotape. The ruling sup— plies no limitations on lighting, panning, zooming, camera angles, special effects, backdrops, etc. The lack of specificity concerning the use of videotape in the legal setting has prompted researchers to express con- cern that the "techniques of film and television art will soon become 2 applied to videotaped depositions and testimony" (Bermant, 33.2}:9 1975, 8). Conceivably, the use of various production techniques could system- atically affect the information that is presented in a trial. Thus, it is important to understand how these techniques affect jurors' perceptions of trial participants as well as the information provided by them. Unfortunately, very little research has examined the effects of pro- duction techniques on viewers. Perhaps film has been considered an artis- tic medium lending itself to evaluation by aesthetic criteria. While aes- thetic criteria may be acceptable for evaluating many types of films (e.g., dramas, comedy, suspense), they are inadequate for evaluating the effects the techniques may have on jurors. Thus, research is needed to determine what systematic effects, if any, various production techniques have on jurors' responses to videotaped testimony. A basic question raised by many leading jurists concerns the type of camera shot that should be employed when videotaping a witness. Doret (197u) has addressed most of the central issues involved in this question. Many alternatives are available when taping a witness and the use of any particular camera shot has advantages and disadvantages. For example, Doret (1974) states that a shot providing a panoramic view of the court- room offers the jury: ...a viewing experience similar to that of watching a movie of a stage play. The advantage of this method is that it deviates least, in terms of the visual field offered the juror, from the traditional trial, and offers the juror the widest possible universe of sensory data to fbrmulate his impressions upon. The disadvantage of this method is the inability of the panorama to capture in detail the nuances of the demeanor of the witness (233-23”). The problem of the lack of visual detail associated with the panoram- ic shot could be alleviated by using a close-up (head 8 shoulders) or 3 medium (head 8 torso) shot of the testifying witness. However, these shots also have disadvantages. First, the amount of sensory data available to the juror would be greatly reduced. Whether this reduction has any sys- tematic affect on the jurors' decision-making process is unknown. Addi- tional disadvantages are dependent upon how the shot is executed in the context of the entire trial. If the camera is positioned for a close-up or medium shot and remains stationary, then jurors may detach the witness from the courtroom environment, or whatever setting in which the taping occurred. Further, the jurors would not be able to see the behavior of off-camera participants. The other trial participants' reactions to a witness' testimony may constitute important information to the jurors. Currently, three types of camera shots are predominately used in the legal system when videotaping witness testimony: (l) close—up shot: tight focus on the head and shoulders of the witness (2) medium shot: focus from the head to just above the waist of the witness (3) long shot: full focus of the witness from head to foot In addition, a fourth shot (a very long shot) is often used at the begin- ning of a taping session in order to allow the jurors to see all the par- ticipants. It is not presently known what effect these camera shots have on jurors' responses to a videotaped witness. Thus, the purpose of the present study is to determine whether the three different camera shots (i.e., close-up; medium; long) used to videotape witness testimony have any systematic effect upon jurors' impressions of a witness. A logical beginning fer determining the effects of camera shots on jurors' responses would entail an examination of television production texts. Most television and film production texts include a discussion u of camera techniques. However, some of these are limited to a "how to" discussion. That is, they discuss camera techniques from a technical per- spective describing different lenses, cameras, dollies, etc. (cf., Fulton, 1960; Quick 8 LaBau, 1972; Scott, 1975). At best, these texts include descriptions of how images change as a function of camera distance, lens selection, f—stops, etc., but offer nothing in the way of how these tech- niques affect the reactions of viewers. In contrast, other texts go beyond the basic "how to" discussion, and provide discussions of how viewers may react to various camera techniques (Eisenstein, 1960; Bretz, 1962; Millerson, l96u; Zettl, 1966; Lewis, 1968; Davis, 1960; Madsen, 1973). While the effects of many different camera techniques are discussed in these texts, only camera shots are of central concern. According to television production texts, the long, medium and close- up shots have distinct functions. Millerson (196a) summarizes the utility of each of these shots. The long shot serves to personalize the individual(s) being filmed. ‘In contrast to longer shots (e.g., the very long shot), movement becomes more recognizable. Moreover, facial expressions and gestures become more dominant. Thus, in the long shot, emphasis is placed more on the actor(s) rather than the setting. The medium shot serves to direct the attention to one or two indi- viduals and provide ample cues. Facial expressions and gestures become more prominent In the medium than in the long shot. The purpose of the close-up shot is to concentrate the viewer's in- terest. It forces the viewer to notice detail that might otherwise be S overlooked. The close-up shot is usually used for dramatic emphasis of detail (Madsen, 1973). In sum, the three camera shots differ with respect to the number of sensory cues made available to the viewer as well as the saliency of the cues presented. Of the three shots considered, the close—up provides the fewest sensory cues and the greatest saliency of the cues presented. In contrast, the long shot provides the greatest number of sensory cues and the lowest saliency of the cues presented. The medium shot falls in be- tween the close-up and long shots. That is, the medium shot provides more sensory cues than the close-up shot, but fewer than the long shot. The saliency of the cues presented by the medium shot is less than the close- up, but greater than the long shot. Considering the videotaped witness, these effects can significantly alter jurors' perceptions of that witness. The literature on person perception adds credence to this last point (e.g., Bruner 8 Taguri, 195u; Bruner, 1957; Hastorf, Schneider 8 Polefka, 1970; Shaver, 1975, 1977). Early research in person perception has shown that subjects alter their impressions of a hypothetical stimulus person by simply rearranging a list of descriptive adjectives (Asch, 19u6; Luchins, 1957; Anderson 8 Hubert, 1963). Additionally, researchers contend that some traits are more central than others, and that their inclusion in a list of adjectives can significantly alter subjects' impressions of both a hypothetical stimulus person (Asch, 1946) and a live stimulus person (Kelley, 1950). Thus, when people are privy to information about a per- son prior to meeting them, the order and type of adjectives used in des- cribing that person can affect initial impressions. This is directly analogous to what transpires in most trials, par- ticularly for the defendant. Usually, jurors receive information about 6 the defendant before he/she takes the stand. The order and type of infor- mation presented will probably influence the jurors' initial impressions. Therefore, jurors will form an impression of the defendant prior to view- ing his/her behavior during testimony, and may look for behaviors that support their initial impression. This is called the prior entry effect (Jones 8 Gerard, 1967). The presentation of the defendant to the jurors is significant. Conceivably, the defendant could be presented such that the behaviors shown may confirm or contradict the jurors' initial percep— tion. Therefore, an understanding of how people form impressions from a person's behavior is important. Bruner (1957) contends that impression formation is an act of cate- gorization of attributes of a stimulus person. That is, people learn through experience what elements of the stimulus person are related and these are combined into meaningful categories. Some attributes (stimulus elements) are more relevant than others. Bruner calls these criterial attributes, which serve to define the boundaries of the category. As one might expect, when perceivers define categories of behavior, they begin to respond to the stimulus person in terms of those categories. Thus, the stimulus person becomes a role occupant in a category rather than a unique individual. This increases the likelihood that the per- ceiver will make errors in judgments (Bruner, 1957; Goffman, 1963; Miller 8 Steinberg, 1975). Although perceptual accuracy has concerned many researchers in the area of person perception (see Cline, 196%), it is not a major concern here. Rarely, if ever, will jurors know if their perceptions are accu- rate. More important is the fact that they think they are accurate, and 7 respond according to their perceptions. Thus, it is important to under— stand how the different camera shots may alter jurors' perceptions of the witness. Earlier it was argued that camera shots differ both in the number of sensory cues presented to the viewer and the saliency of those cues. The reduction of sensory cues may reduce the number of criterial attributes available to the jurors, which could result in a different categorization of the witness. Differences in the saliency of the cues presented may also affect jurors' perceptions of the witness. These differences may alter the sa- liency of certain attributes. For example, the close-up and medium shots may cause jurors to concentrate on attributes of the witness that may have been overlooked, or perceived as less relevant if a long shot had been used. The alteration of relevant attributes could also lead to different impressions of the witness. Recently, some researchers have undertaken the task of empirically testing the effects of different camera shots on viewers' responses. An examination of the relevant research concerning these effects follows. Relevant Research Williams (1965) examined the effect of varying camera shots on viewers' expressed interest level toward a televised lecture. The results of his study indicated that expressed interest level did not significantly differ as a result of using a variety of close—up and long shots compared to utilizing a static medium shot. However, when examining shot differences using a film screen, Williams (1968) found that viewers' expressed inter- est level significantly decreased when a long shot was employed. 8 Wurtzel 8 Dominick (1971-72) examined the effects of acting style and camera shot on viewers' evaluations of television drama. An 11 min- ute emotional scene was performed by three professional actors utilizing two different acting styles: (1) film acting and (2) stage acting. Stage acting is different from film acting in that gestures and expressions are more elaborate and pronounced when an actor is performing on a stage. The scene was filmed four times in order to obtain the levels of the two in- dependent variables of acting style (film acting and stage acting) and type of shot (close-up and medium). The dependent variable was an evaluation measure comprised of ten bi-polar adjectives. Six of these scales were derived from an evaluative measure developed by Osgood, Succi 8 Tannenbaum (1957). The other four scales consisted of "adjectives most used in instructor evaluations of the media performance class at Queens College," which correlated highly with the other six scales. The main effects for acting style and camera shot were not signifi- cant, but a significant interaction was observed. Specifically, viewers evaluated the scene more favorably when the actors were film acting and a close-up shot was used as opposed to a medium shot. Viewers in the medium shot condition evaluated the scene more favorably than did viewers in the close-up shot condition when stage acting was employed. McCain 8 Repensky (1972) examined the effect of camera shot on interpersonal attraction. The stimuli subjects were two comedy perform- ers, Edmonds and Curly. The comedians worked together as a team and per- formed two comedy routines. These routines were taped using three camera shots simultaneously: (1) close—up shot, (2) medium shot, and (3) long shot. The measure of interpersonal attraction was comprised of three 9 dimensions: (1) physical attraction, (2) social attraction, and (3) task attraction. The authors derived these three dimensions utilizing orthog- onal factor analysis and thus treated each dimension independently in the analyses. The results of this study indicated that camera shot does affect in— terpersonal attraction. However, the effects differed for each performer. Analysis of the physical attractiveness data yielded a significant main effect for Comedian and a significant interaction of Comedian and Camera Shot. Edmonds was perceived as being more physically attractive in the close—up shot than in the medium or long shot. Curly was perceived as being more physically attractive than Edmonds in the medium and long shots. Neither comedian was perceived significantly more attractive in the close- up shot. No significant differences were obtained from the analysis of the social attractiveness data. A significant interaction was obtained from the analysis of the task attractiveness data. Curly was perceived as most task attractive in the close-up condition while Edmonds was per— ceived as least task attractive in the close-up condition. The two co- medians did not differ in task attractiveness in any other condition. Clearly, some characteristic(s) of the two comedians interacted with camera shot in this study. However, it is difficult to determine what source characteristic(s) were operating. The authors offer a number of plausible explanations which include, the roles of the comedians (i.e., straight—man vs. funny-man), their physical appearance, and the quality of their performances. In an attempt to uncover the influence of two particular source characteristics, McCain 8 Divers (1973) examined the effects of body type, sex of the source and camera shot on interpersonal attraction and 10 source credibility. Three males and three females were selected as stim— ulus subjects and were classified into three categories of body type based on the work of Sheldon (195”). The three body types were: (1) endomorph (fat or plump), (2) mesomorph (muscular or athletic), and (3) ectomorph (thin or skinny). Three camera shots were employed: (1) close—up shot, (2) medium shot, and (3) long shot. Interpersonal attraction was com- prised of three independent dimensions: (1) physical attraction, (2) social attraction, and (3) task attraction. Source credibility was com- prised of five independent dimensions: (1) competence, (2) sociability, (3) dynamism, (4) composure, and (5) character. The six stimulus subjects delivered a "three minute neutral speech," which was taped utilizing three cameras simultaneously in order to obtain the three levels of camera shot. The results of the study yielded significant main effects for body type and sex of the source. No main effect for camera shot was found, but a number of significant interactions were obtained. However, once again the interpretation of these results is clouded by the lack of control in the study. For example, the results suggested that the sex of the speaker had a strong impact on the results of the study. This result was inter- preted cautiously by the authors, for as they noted: Since only one person of each sex represented each body type, the differences are really personal at- tribute differences of single individuals. Facial expression, fluency of presentation and other non- verbal variations between the males and females may well provide better explanations for differences be- tween them than their gender differences (McCain 8 Divers, 1973, 9-10). Thus while it would appear that body type, sex of the source and camera shot have an effect on interpersonal attraction and source credibility, 11 this result must be interpreted cautiously given the lack of control of potentially relevant source characteristics. Summarizing the research discussed thus far, it would appear that different camera shots have an impact on viewers' perceptions of a source's attractiveness and credibility. However, the precise nature of these ef- fects is still unknown. In virtually all of the studies discussed earlier, relevant source characteristics were not controlled. It is not known what effect these source characteristics may have had in the results reported in these studies. An additional point is worth mentioning here. In the studies reported earlier, the role of the sources differed. That is, sources served as oo- medians (McCain 8 Repensky, 1972), lecturers (Williams, 1965; 1968), ac- tors (WurtzeltSDominick, 1971-72) and neutral speakers (McCain 8 Divers, 1973). As the role of the source changes, the role of the receiver and the purpose of the message changes. Conceivably, as the relationship be- tween the source and the viewer changes, different cues emitted by the source become salient for the viewer. That is, the viewer may attend to different cues if he/she knows that he/she is being entertained, than if he/she knows he/she is being informed or persuaded. This point attains particular relevance when considering the role of the juror. Presumedly, the juror is aware that he/she will receive conflicting testimony through- out the course of a trial. The jurors' task is to weigh the evidence pre- sented, assess the veracity of the information, the credibility of the witnesses, etc. In essence, the juror acts as a judge of various wit- nesses' character and testimony. Clearly, the role relationship between juror and witness is very different than the role relationship between comedian and viewer. The cues emitted by the source that are salient in 12 one situation, may not be as salient in the other. The results of the studies reported earlier may not be applicable to the legal setting. Hence, the need for research on the effects of different camera shots on jurors' perceptions of a witness still exists. Recently, a study was conducted which examined the effects of differ- ent camera shots on jurors' perceptions of a witness (Halbert, 1978). Three independent variables were manipulated: (1) camera shot (close-up; medium; long; extreme long), (2) attractiveness of the witness (attrac— tive; unattractive), and (3) sex of the witness. The dependent variables measured included: (1) identification with the witness, (2) information retention, (3) perceived credibility, (u) interpersonal attraction, and (5) viewer interest. Four sources were used in this study: an attractive male and female and an unattractive male and female. The attractiveness levels of the sources were determined by a pretest. The study employed an actual trial deposition of an automobile acci- dent. Due to a limitation of available cameras, each source presented the testimony twice in order to obtain the four levels of camera shot. The results of this study yielded a number of significant three-way interactions. However, due to some problems with the manipulations these results are somewhat suspect. Hence, the findings will not be discussed. Instead, attention will be devoted to the areas of concern surrounding this study. One problem with this study surrounds the attractiveness manipulation. The mean attractiveness ratings of the sources taken from the experimental groups indicated that the two males were not perceived significantly dif- ferent from each other. The two females were perceived significantly 13 different in attractiveness. The weakness of the attractiveness manipu- lation for the male sources creates obvious problems in interpreting any of the results that involve this variable. The procedures used when taping the stimuli persons may also have had an effect on the results. Originally, five camera shots were used. The fifth shot, a "very long shot" was dropped from the design. As was mentioned earlier, due to a shortage of cameras, the witnesses presented the testimony twice in order to achieve the levels of camera shot. This was achieved by taping three of the shots during one presentation, and the remaining two during the second. Unfortunately, no consistent pat- tern of shots was used when taping the witnesses. Thus, in the final de- sign, different shots were from different presentations, which varied across witnesses. For example, for the attractive female, the close—up, medium, and long shots were taken during one presentation, while the ex— treme long shot was taken from another. For the attractive male, the close—up and long shot were from one presentation, while the medium and extreme long shot were from another. Therefore, some of the differences obtained may be partially attributable to differences in presentation rather than differences in camera shot. The final area of concern surrounds the lack of control of potentially relevant source characteristics. The study employed only one individual of each sex for the levels of attractiveness. Thus, the differences ob- tained may be attributable to individual differences among the sources. Coupling this problem with the previous problem, it becomes clear that the results may have been affected by differences in presentations within sources as well as between sources. 11+ Clearly, most of the research on camera shots conducted to date has lacked control for potentially relevant source characteristics. One study that appears to have avoided this problem was discussed earlier (Wurtzel 8 Dominick, 1971-72). Recall that the findings from this study indicated that acting style interacts with camera shot. Assuming that acting style is analogous to the presentational style of a source, then presentational style may serve as a useful construct to investigate in relation to camera shot. Still, given the many idiosyncratic behaviors of communication sources, the problem becomes one of determining a useful categorization of presentation style, particularly for the trial participant. Confronted with a similar dilemma, Miller 8 Siebert (1975) offered a useful, albeit rough, distinction of two different types of witnesses: (1) strong witness and (2) weak witness.* A "strong witness" was con- ceived to be "...assertive, attentive, and unhesitant...", while a "weak witness" was seen as "...uncertain, fumbling, and inattentive..." (p. 18). The study was designed to investigate the effects of type of witness (strong and weak) and mode of presentation (color and monochromatic) on jurors' responses. Utilizing an actual trial deposition of an industrial accident, a professional actor was taped in color using a medium shot. The shot included the witness and two attorneys who were seated at a small table. The testimony was presented twice in order to obtain the *The original study employed a third type of witness called the "modal witness." In the modal condition, the witness read the testimony, simula- ting the case where the witness' testimony is read into the record. In a replication of this study (Miller 8 Siebert, 1975) the modal condition was dropped for two major reasons: (1) it was unrealistic and (2) sub- jects had difficulty determining whether they were supposed to rate the actual witness or the reader. The results of the replication supported the results of the original study for the effects of witness type. There- fore, the general results will be considered without regard for the modal condition. 15 levels of witness type. The tapes were then shown to role-playing jurors. The dependent measures included: witness credibility, witness authorita- tiveness, witness character, and juror information retention. The results of the study indicated that the strong witness was per- ceived significantly more credible and authoritative than the weak witness. The strong witness was perceived as having better character than the weak witness, but this finding was not significant. Witness type had no sig- nificant effect upon jurors' retention of trial-related information. The present study offers a replication and extension of the study reported by Miller 8 Siebert (1975) with some important differences. The latter study employed a static medium shot, which encompassed the witness. The present study will employ three different camera shots (i.e., close-up, medium, and long) which will also encompass only the witness.* In addition, the present study will add a trial synopsis which will precede the videotaped witness in order to obtain a measure of ver- dict. Finally, the information retention measure will be refined and a measure of witness composure will be added. In sum, the present study will assess the effects of camera shot and witness type on jurors‘ responses to a videotaped witness. The dependent measures will include: (1) witness composure, (2) witness credibility, (3) witness authority, (u) witness character, (5) information retention, (6) interest, (7) verdict, and (8) award. *While the three shots of interest will only encompass the witness, a longer shot, which includes the two attorneys, will be used at the be- ginning and the end of the deposition in order to provide the jurors with a sense of location. 16 Hypotheses This section will delineate the specific hypotheses of this study and the rationale behind their derivation. In order to achieve clarity in the derivation of specific hypotheses, the main effects for witness type will be discussed first. These hypotheses will then be modified in order to take into account the effects of the different camera Shots. The first variable to be considered is the perceived composure of the witness. Composure is being used here as a measure of the general presen- tational style of the witness. The specific items that comprise this mea- sure are derived from the research on verbal and nonverbal behavior. For example, one characteristic of the strong witness is a fluent speech pat- tern, while the speech pattern of the weak witness is nonfluent. Research on speech fluency indicates that a nonfluent speech pattern is indicative of high anxiety; i.e., low composure (Dibner, 1956; Krause 8 Pilisuk, 1961; Pope 8 Siegman, 1962; Zimbardo, Mahl 8 Barnard, 1963; Kasl 8 Mahl, £1965; Cook, 1969). Therefore, one of the items of the composure measure its a measure of perceived anxiety of the witness. In essence, the com- pxbsure measure serves as a check on the manipulation of witness presenta- tixonal style. Thus, the following hypothesis is posited: H : A strong witness will be perceived to be more composed than a weak witness. The next variable to be considered is the perceived credibility of the witness. Jurors are often placed in the position of having to make decxisions in the face of conflicting testimony. That is, the plaintiff's Case presents one impression of the events that transpired, while the defendant's case presents a different image. The jurors must then make 'thEE decision of whom to believe. While many factors may impinge upon 17 their decisions, most jurists would agree that the credibility of the witness has a large impact on the jurors' decision-making process. Credibility* has been defined as a receiver's attitude toward a source at a given point in time (cf., Auer, 1969; McCroskey, 1972). In addition, credibility has been conceived as a multidimensional construct (Lemert, 1963; McCroskey, 1966, 1972; Auer, 1969; Berlo, Lemert 8 Mertz, 1969-70). Still, researchers disagree on the number of relevant dimen- sions as well as the labels for those dimensions (see McCroskey, 1966). Berlo, Lemert 8 Mertz (1969—70) report three dimensions that comprise credibility: (1) safety, (2) qualification, and (3) dynamism. The first two dimensions comport with dimensions found by other researchers. How- ever, questions concerning the validity of the third dimension as a separ- ate dimension of credibility have been raised. McCroskey (1972) reports findings from a series of studies that found a dynamic source to be "... consistently more competent, and usually more trustworthy, than a passive source" (p. 66). Moreover, McCroskey (1972) reports that items of the dynamism scale were often represented on the competence factor. While the controversy still remains concerning whether or not dynamism is a separate dimension of credibility, it is not of central concern here. Rather, it is important to note the importance of dynamism to the perceived credibility of the source. The writings of Aristotle stressed the need for good delivery in ac- quiring credibility (Cooper, 1932). More recently, researchers have in- vestigated the effects of delivery on credibility (Winthrop, 1956; Miller 8 *The literature also contains the terms ethos and prestige for the same construct. For purposes of clarity, the present paper will consis— tently use the term credibility. 18 Hewgill, 1964; Sereno 8 Hawkins, 1967; McCroskey 8 Mehrley, 1969; McCroskey, 1972). Using the credibility scales developed by Berlo 8 Lemert (1961), Miller 8 Hewgill (1964) and Sereno 8 Hawkins (1967) found that sources who had a fluent presentational style were perceived significantly more dynamic and competent than sources who had a nonfluent presentational style. Moreover, Miller 8 Hewgill (1964) found that a fluent speaker was rated significantly more trustworthy than a nonfluent speaker, but this result did not emerge as clear as the results fOr the factors of competence and dynamism. Sereno 8 Hawkins (1967) report findings in the same direction for the ratings of the source's trustworthiness, but the differences were not significant. McCroskey 8 Mehrley (1969) employed the scales of authority and character which were developed by McCroskey (1966). These two scales are considered to be related to the dimensions of competence and trustworthi— ness. In addition, Berlo 8 Lemert's (1961) dynamism scale was employed. The authors report that a source who used a fluent presentational style was rated significantly higher on all three dimensions than a source who used a nonfluent presentational style. Given the two types of witness presentational style to be employed in this study, the relationship between witness type and credibility can now be hypothesized. The strong witness, whose presentational style is characterized as fluent, assertive, and attentive (good delivery) would be perceived as being more credible than the weak witness, whose presen- tational style is characterized as uncertain, hesitant, fumbling and in- attentive (poor delivery). Given the conceptualization of credibility l9 offered by Berlo, Lemert 8 Mertz (1969-70), the following three hypotheses have been generated to test the effects of witness type on credibility: H2: A strong witness will be perceived to be safer than a weak witness. H3: A strong witness will be perceived to be more qualified than a weak witness. H : A strong witness will be perceived to be more dynamic than a weak witness. The variables of witness authority and character will be considered conjointly. As was noted above, the measures for these two variables were developed by McCroskey (1966) and are considered to be related to dimen- sions of credibility derived by other researchers. Given this relation— ship, one would expect the presentational style of a source to have the same effect upon these variables as it does upon credibility. Some sup- port is offered for this contention from the results of the study by McCroskey 8 Mehrley (1969). Recall that in that study, the authors found that a source who had a fluent presentational style was perceived to be more authoritative and as having better character than a source who had a nonfluent presentational style. This finding lead to the following hypotheses: H5: A strong witness will be perceived to be more authorita- tive than a weak witness. H : A strong witness will be perceived as having better character than a weak witness. The next variable to be discussed in relation to witness type is information retention. When describing the behaviors of the strong and weak witness, it was stated that the weak witness would be nonfluent, fumbling, and inattentive. The nonfluency in speech may be disruptive to the extent that jurors would find it difficult to follow the testimony. 20 Moreover, the nonverbal behaviors of the weak witness may be distracting to the jurors, causing them to attend more to those behaviors than to the information presented verbally. Research on distraction and message recall has produced inconsistent results. Some researchers have reported findings that indicate distrac- tion decreases recall of message content (Vohs, 1964; Gardner, 1966; Haaland 8 Vankatesan, 1968), while others have found distraction to in- crease recall (Silverman 8 Regula, 1968). One explanation for these inconsistent findings is offered by Baron, Baron, 8 Miller (1973). The authors contend that the effects of distrac- tion on message recall are dependent upon whether or not the distraction can be ignored. If the distraction is mild, then an attempt will be made to block the distracting stimulus by attending more to the message. How- ever, if the distraction is severe, then more attention will be paid to the distracting stimulus. Assuming the distracting behaviors of the weak witness (e.g., the verbal nonfluencies and fUmbling) are disruptive, the following hypoth- esis is advanced: H7: Jurors exposed to a weak witness will retain less trial- related information than jurors exposed to a strong witness. The presentational style of the witness is expected to have an effect upon jurors' expressed interest in the proceedings. Since the writings of Aristotle (Cooper, 1932), the importance of the presentational style of a source and its effect upon receiver interest has been discussed by stu- dents of public speaking. While many factors will ultimately affect re- ceivers' interest, one general relationship is clear. The better the 21 delivery style of the source, the greater the likelihood of evoking interest in the receiver. Earlier it was argued that the strong witness' presentational style characterizes good delivery, while the weak witness' presentational style characterizes poor delivery. Given the general relationship between pre- sentational style and interest, the following hypothesis is posited: H : Jurors exposed to a strong witness will express greater interest in the proceedings than jurors exposed to a weak witness. The relationship between witness type and verdict will now be con— sidered. Earlier it was argued that jurors are often faced with the di— lemma of making a decision from conflicting testimony. Further, it was argued that one variable that plays a major role in their decision-making process is the perceived credibility of the witnesses. If this is true, then a significant correlation between credibility and verdict should be found. Recently, Kaminski (1977) reported evidence that supported this relationship. That is, as the perceived credibility of the plaintiff increases, the likelihood of a verdict in favor of the plaintiff increases. The same relationship was found for the defendant. Given this relation— ship between witness type and credibility, the following hypothesis was derived: H : Jurors exposed to a strong witness will be more likely to offer a verdict congruent with the witness' testimony than jurors exposed to a weak witness. Considering all of the hypotheses posited thus far, a general pattern emerges. They all indicate that a strong witness will be perceived more favorably than a weak witness. This general relationship may have an effect upon the award offered to the plaintiff for those jurors who find in favor of the plaintiff. 22 While an award may be considered restitution for the plaintiff, it may also be viewed as punishment for the defendant. Therefore, when de- termining how much money to award a plaintiff, jurors may not only con- sider reasonable recompense, they may also consider the degree to which they are punishing the defendant. Conceivably, jurors will be less likely to punish a defendant they view favorably as opposed to a defendant they view unfavorably. Thus, jurors who look favorably upon a defendant may award the plaintiff less than jurors who look unfavorably upon the defen- dant. The following hypothesis has been posited to test this reasoning: H10: For those jurors who find for the plaintiff, jurors exposed to a strong Witness Wlll recommend lower awards than jurors exposed to a weak witness. It should be noted that this hypothesis assumes that the witness is either the defendant, or a witness for the defense. If the witness was the plain- tiff, or a witness for the plaintiff, the effect would be reversed. Thus far, the hypotheses have focused on the relationship between witness type and the dependent variables without regard for the effects of camera shot. Those effects will now be discussed and the hypotheses will be modified to take these effects into account. In order to discuss the effects of camera shot, it is necessary to recall the differences between the three different shots. Earlier, it was argued that the shots differ in the number of sensory cues made avail- able to the viewer as well as the salience of the cues presented. Specif- ically, the number of sensory cues is greatest in the long shot and mini- mized in the close-up shot, with the medium shot falling in between the two. Moreover, the cues are most salient in the close-up shot; less salient in the medium; and least salient in the long shot. Given the differences in the saliency of the cues presented, one would expect the 23 characteristics of the witness to become most prominent in the close-up shot and least prominent in the long shot. Additionally, the character— istics of the witness in the medium shot would be less prominent than in the close-up, but more prominent than in the long shot. Therefore, the effects of witness type upon the dependent variables in Hypotheses 1-10 are expected to change with respect to camera shot. That is, the effects should be most pronounced in the close-up shot and least pronounced in the long shot. The effects in the medium shot should be less pronounced than in the close-up, but more pronounced than in the long shot. The following 20 hypotheses have been posited to test the effects of camera shot upon the variables of composure; safety; qualification; dyna- mism; authority; character; information retention; interest; verdict; and award. The first ten are derived fer the strong witness, while the last ten are derived for the weak witness. In order to reduce unnecessary con- fusion, a greater than sign (>) is used in these hypotheses. The effect upon the dependent variable discussed in each hypothesis is expected to be more pronounced in the shot that precedes the greater than (>) sign. Stated differently, the shots are expected to be significantly different from each other, with the ratings of the dependent variable being greater in those shots that are on the left of the greater than (>) sign. The hypotheses are as follows: For the strong_witness: H11: composure: close—up Shot > medium shot > long shot H12: safety: close-up shot > medium shot > long shot H13: qualification: close-up shot > medium shot > long shot H1”: dynamism: close—up shot > medium shot > long shot 30‘ authority: character: information retention: interest: verdict (congruent with the witness' testimony): award: the weak witness: composure: safety: qualification: dynamism: authority: character: information retention: interest: verdict (congruent with the witness' testimony): award: 24 close-up shot close-up shot shot close-up close-up shot medium medium medium medium close-up shot > long shot > long Shot > long shot > long shot > long long shot > long shot > shot > long long shot > long shot > shot > medium medium medium medium medium medium medium medium medium medium medium shot shot shot shot shot shot shot shot shot shot shot > long shot > long shot > long shot > long shot > long > close—up > close—up > close-up > close-up > close-up > close-up > close-up > close-up > close-up > close-up close-up shot > medium shot > long shot shot shot shot shot shot shot shot shot shot shot shot shot shot shot shot 25 Hypotheses 20 and 30 deserve further explanation. The direction of the effect of camera shot on award is dependent upon whether the witness' testimony is in favor of the plaintiff or the defendant. These hypotheses are posited for a witness whose testimony is in favor of the defendant. If the witness' testimony was in favor of the plaintiff, the effects would be reversed. It should be noted that the hypotheses for the effects of camera shot for the strong witness are reversed for the weak witness. Therefore, the last 20 hypotheses combined produce a general interaction hypothesis: H31: Jurors' perceptions of a strong witness will be more favorable in the closer shots, while jurors' percep- tions of a weak witness will be more favorable in the longer shots. Chapter II METHODS AND PROCEDURES With the aid of legal experts, a transcript of an actual deposition was selected. The deposition was of a defendant who was accused of neg- ligence resulting in an industrial accident. The deposition, which was approximately 30 minutes in length, consisted of cross-examination by the plaintiff's attorney. It did not contain direct examination by the de- fendant's attorney. Professional actors played the roles of the witness and the defense attorney. An actual attorney played the role of the plaintiff's attorney. The type of witness manipulation consisted of the same actor playing two different roles: (1) a strong witness and (2) a weak witness. The presentational style of the strong witness was characterized as fluent, assertive, and attentive. The presentational style of the weak witness was characterized as uncertain, hesitant, fumbling, and inattentive. The actor was trained to emit verbal and nonverbal cues derived from previous research which were found to elicit impressions of the presenta- tional styles of interest. For example, for the strong witneSs, the actor was instructed to speak normally, fluently and with confidence; hold his head erect; maintain eye contact with the questioning attorney and lean slightly toward him. In addition, he was instructed to relax and not fidget, tap his feet or place his arms akimbo (elbows out with hands placed on hips). For the weak witness, the actor was instructed to speak 26 27 softly and nonfluently; insert pauses, "um's" and "uh's" in his sentences; maintain low eye contact with the questioning attorney and lean slightly away from him. In addition, he was instructed to tense his muscles slightly; sigh occasionally; fidget and tap his fingers and feet. These behaviors have been found to be indicative of the presentational styles of interest (cf., Dibner, 1956; Kraus 8 Pilisuk, 1961; Reece 8 Whitman, 1952; Pope 8 Siegman, 1962; Zimbardo, Mahl 8 Barnard, 1963; Kasl 8 Mahl, 1965; Dittman 8 Llewellyn, 1968; Cook, 1969; Mehrabian, 1969, 1971; Harrison, 1974). Tapingfthe Witness The deposition was videotaped in a television studio at Michigan State University. The participants were seated at a rectangular table in front of a plain backdrop. The witness was seated at the middle of the table, with the attorneys at the ends. The deposition was videotaped in color utilizing three cameras simultaneously in order to achieve the three levels of camera shot (close-up; medium; and long). Each shot contained only the witness. Neither attorney appeared in the shot. A fourth camera shot (a very long shot), containing the witness and both attorneys, was used to offer the viewer a sense of location. This very long shot was edited onto the beginning and end of the videotape. The angle used for all camera shots was 90° to the vertiCal plane. This angle was selected because past research has indicated that deviations from this angle has biasing effects on viewers (Tiemens, 1970; McCain 8 Wakshlag, 1974; McCain, Chilberg 8 Wakshlag, 1977). The deposition was videotaped twice—-once for each witness type. The testimony was identical in both presentations. Only the delivery style differed. 28 Trial Synopsis In order to place the deposition in context, a brief trial summary was written with the assistance of legal experts. In addition, an injury for the plaintiff was contrived, as the deposition did not mention the in- jury sustained by the plaintiff. It was decided not to make the injury either too serious or too trivial, as this might have affected the sub- jects' impressions of the incident, which could have carried over and altered their impressions of the witness. Therefore, a moderately serious injury was considered to be most desirable. In order to determine what a "moderately serious" injury would be, a list of 25 injuries which the plaintiff could have sustained were gen- erated. This list was offered to 98 undergraduate students enrolled in communication courses at Michigan State University. The subjects were asked to rate each injury on a seven-point scale, ranging from "not seri- ous" to "very serious." While the results indicated that a number of injuries were perceived as moderately serious, only one (a broken leg with no additional compli- cations) was perceived moderately serious and was minimally skewed in either direction of the scale. Therefore, the injury ascribed to the plaintiff was a broken leg with no additional complications. Subjects The subjects used in this study consisted of 197 undergraduate students enrolled in communication courses at Michigan State University. The study was conducted on three separate evenings in one week, two conditions each evening. The conditions were randomly assigned to each of the three even- ings. Subjects volunteered to participate during one of the evenings. 29 Upon arriving, the subjects were randomly divided into two groups, one group for each condition. The subjects were told they were participating in a study on jury size and that they would be assigned to a jury after the presentation of the testimony. The subjects were given the trial synopsis which was read aloud by the experimenter. The synopsis was then collected and the video- taped deposition was presented. At the conclusion of the deposition, the subjects completed the questionnaire. The subjects were then taken to another room and informed that they would not be deliberating. At this time, the subjects were partially debriefed. After all the data were collected, the subjects were fully debriefed. Thirty-five subjects were randomly deleted in order to obtain an equal number of subjects for each condition. Thus, the total number of subjects used in this study was 162 (27 subjects for each condition). Demographic Characteristics Certain demographic characteristics were obtained from the subjects in order to check for homogeneity of sampling across conditions. Specifi- cally, measures of the subjects' sex, age, education, marital status, and ethnic affiliation were obtained. In addition, subjects were asked whether they served as a juror before and whether they had been a party to a suit before. The frequencies of the above measures, with the exception of age, are reported in Table l. A chi-square test was used for each of these measures to assess differences across conditions. The results indicated that the subjects did not differ in sex (x = .50, df = l, p > .05), educa- tion (x = .90, df = 3, p > .05), marital status (x = 3.35, df = l, p > .05), Ethnic affiliation (x = 2.99, df = 2, p > .05), whether they served as a 30 Table 1. Frequencies of Subjects' Demographic Characteristics. Sex Male Female (Missing Data) Education Freshman Sophomore Junior Senior (Missing Data) Marital Status Single Married (Missing Data) Ethnic Affiliation Caucasian Black Other (Missing Data) Ever a Juror Before? Yes No (Missing Data) Ever a Party to a Suit Before? Yes No (Missing Data) Strong Witness Close-up Medium Long l4 l3 (0) 27 (0) 23 (3) 27 (0) 25 (O) 17 10 (0) 10 (O) 27 (0) 19 (6) 27 (0) 21+ (0) 12 15 (O) 27 (0) 19 (4) 26 (0) 2n (0) Weak Witness Close-up Medium Long 11 15 (l) 23 (l) 16 (6) 26 (l) 21 (1) ll 16 (0) 26 (0) 2O (3) 26 (0) 25 (O) 15 12 (0) (2) 26 (0) 25 (0) 31 juror before (x = .00, df = 1, p > .05), nor whether they had been a party to a suit befOre (x = .00, df = l, p > .05). The mean age of the subjects and analysis of variance summary are reported in Table 2. The results indicated that the subjects did not differ in age across conditions. Given the above results, the sample was considered to be homogeneous across conditions. Table 2. Means and Analysis of Variance Summary of Subjects' Age. xl Strong Witness Weak Witness Close-up 19.78 19.12 19.45 Medium 20.15 20.00 20.08 Long 19.44 19.96 19.70 E 19.79 19.69 i = 19.79 2 Source Sum of Squares df Mean Square F n Camera Shot 10.49 2 5.25 2.27 ---- Witness Type .35 1 .35 <1 ---- Camera Shot X Witness Type 9.38 2 4.69 2.03 ---- Within Groups 358.35 155 2.31 Total 378.57 160 Power Estimation A common problem in social science research stems from the interpre- tation of insignificant results. Often, the researcher is left asking whether a Type II error has been committed; i.e., accepting the null 32 hypothesis when it should be rejected. The probability of committing a Type II error can be minimized by increasing the power of an experiment. Power is defined as the probability of rejecting the null hypothesis when an alternative hypothesis is true (see Keppell, 1973). The power asso- ciated with a given experiment can be affected by a number of factors, one of which is the size of the sample employed. Given the concern with committing a Type II error, an estimation of the power associated with the present experiment was calculated. The for- mula used was derived by Cohen (1969). In calculating the power, an esti- mation of the size of the effect (n2) was made. The estimation of effect size is arbitrary, but Cohen (1969) suggests that an effect at n2 = .06 is reasonable for most social science data. Therefore, the effect size was set at a minimum of n2 = .06. With a sample size of 27 subjects per cell and an estimated effect size set at n2 = .06, the resultant power of this experiment was .70. This means that if an effect is present at n2 = .06, the probability of rejecting the null hypothesis is .70. For smaller effects (n2 > .06) power decreases, while for larger effects (n2 > .06) power increases. Measurement Techniques This study was designed to determine the effects of camera shot and type of witness upon jurors' perceptions of the witness' composure, credi- bility, authority, and character; the amount of information retained by jurors; the amount of interest expressed by jurors; the verdicts rendered by jurors; and the award offered to the plaintiff for those jurors who find in favor of the plaintiff. Jurors' perceptions of the witness' composure was measured utilizing ten seven-point Likert-type scales. The adjectives used for this scale 33 were: friendly-unfriendly; confident-unconfident; relaxed-tense; atten- tive-inattentive; assertive-nonassertive; poised-nervous; calm—anxious; comfortable-uncomfortable; unhesitant-hesitant; outgoing-reserved. The ten items were factor analyzed which yielded a single factor so- lution. One item (friendly—unfriendly) failed to load adequately and was deleted from the scale (see Table 3). The remaining nine items yielded an alpha coefficient of .90. Witness credibility was measured utilizing the scales developed by Berlo, Lemert 8 Mertz (1969-70). The measure consisted of 15 seven-point Likert-type scales, which theoretically comprise three dimensions of credi- bility: (1) safety, (2) qualification, and (3) dynamism. The subjects were presented all 15 scales and were asked to place a check in the space beside the answer which best described their opinion of the witness. For example: very qualified ______qualified somewhat qualified somewhat unqualified unqualified very unqualified The 15 scales were factor analyzed using the multiple-group method (see Nunnally, 1967). The results of this analysis indicated that the credibility measure was comprised of the three dimensions of safety, qualification, and dynamism (see Table 4). The alpha coefficients for these dimensions were .72, .72, and .85, respectively. Jurors' perceptions of the witness' authority and character were measured utilizing scales developed by McCroskey (1966). Twenty-two items 34 Table 3. Factor Loadings of Composure Items. Factor 1 Item (Composure) 1. Friendly - Unfriendly .05 2. Confident - Unconfident .70* 3. Relaxed - Tense .84* 4. Attentive - Inattentive .58* 5. Assertive - Nonassertive .56* 6. Poised - Nervous .88* 7. Calm - Anxious .80* 8. Comfortable - Uncomfortable .86* 9. Unhesitant - Hesitant .66* 10. Outgoing - Reserved .51* *Indicates item was retained for the factor. Table 4. Factor Loadings of Credibility Items. Factor 1 Factor 2 Factor 3 Item (Qualification) (Safety) (Dynamism) 1. Safe - Unsafe .15 .54* .04 2. Just - Unjust .08 .68* .08 3. Kind - Cruel -.11 .68* .01 4. Friendly - Unfriendly .01 .65* .08 5. Honest - Dishonest .20 .56* .02 6. Trained - Untrained .69* .10 .15 7. Experienced - Inexperienced .74* .07 -.02 8. Skilled - Unskilled .61* .02 .06 9. Qualified - Unqualified .55* -.Ol .06 10. Informed - Uninformed .52* .20 .11 11. Aggressive - Meek .03 -.08 .80* 12. Emphatic - Hesitant .05 .12 .71* 13. Bold - Timid .ll -.13 .82* 14. Active - Passive .07 .20 .74* 15. Energetic - Tired .19 .20 .55* *Indicates item was retained for that factor. 35 were designed to measure authority and 20 items were designed to measure character. The measure consisted of statements about the witness to which the subjects rated their amount of agreement with the statement. For example: I have confidence in this witness: ______ Strongly Agree Agree Undecided Disagree Strongly Disagree The 42 items were factor analyzed to a two factor solution using varimax rotation with communalities in the diagonal. The results indi— cated that a number of items failed to load adequately, given a criterion of loading no less than .50 on one factor and no more than .20 on the other. After eliminating those items which failed to meet this criterion, the authority scale was comprised of seven items and the character scale was comprised of ten items (see Table 5). The resulting alpha coeffi- ents for the authority and character scales were .85 and .90, respectively. The amount of information retained by jurors was measured in the following manner. Sixty-four multiple choice questions were constructed concerning the testimony presented during the deposition. The items were pretested using a sample of undergraduate students enrolled in communica— tion courses at Michigan State University (N=58). The subjects were di- vided into two equal sized groups. One group was shown the medium shot of the strong witness. The other group was shown the medium shot of the weak witness. After viewing the videotape, they completed a questionnaire which included the 64 infbrmation retention items. The items were 36 Table 5. Factor Loadings of Authority and Character Items. Factor 1 Factor 2 Item (Character) (Authority) 1. I respect this witness' opinion on the topic. .55 .46 2. This witness is pp£_of very high intelligence. .03 .46 3. This witness is a reliable source of information on the topic. -.06 .04 4. I have confidence in this witness. .61 .37 5. This witness lacks information on the subject. .26 .49 6. This witness has high status in our society. .12 .25 7. I would consider this witness to be an expert on the topic. .14 .66* 8. This witness' opinion on the topic is of little value. .26 .27 9. I believe this witness is quite intelligent. .OO .61* 10. This witness is an unreliable source of information on the topic. .41 .52 11. I have little confidence in this witness. .62 .46 12. This witness is well informed on this subject. .19 .69* 13. This witness has low status in our society. .00 .24 14. I would not consider this witness to be an exEEFt on this topic. .07 .71* 15. This witness is an authority on the topic. .05 .70* 16. This witness has had very little experience with this subject. .12 .36 17. This witness has considerable knowledge of the factors involved with this subject. .20 .61* 18. Few people are as qualified to testify on this topic as this witness. .22 .36 19. This witness is not an authority on this subject. .19 .68* 20. This witness has very little knowledge of the factors involved with this subject. .29 .50 21. This witness has had substantial experience with this subject. .18 .38 22. Many people are much more qualified to speak on this topic than this witness. .12 .31 37 Table 5 (cont'd.). Item 23. I deplore this witness' background. 24. This witness is basically honest. 25. I would consider it desirable to be like this witness. 26. This witness is pg£_an honorable person. 27. This witness is pp£_concerned with my well-being. 28. This witness is a reputable person. 29. I trust this witness to tell the truth about the topic. 30. This witness is a scoundrel. 31. I would prefer to have nothing at all to do with this witness. 32. Under most circumstances I would be likely to believe what this witness says about the topic. 33. I admire this witness' background. 34. This witness is basically dishonest. 35. The reputation of this witness is low. 36. I believe that this witness is concerned with my well-being. 37. This witness is an honorable person. 38. I would pp:_prefer to be like this witness. 39. I do pp£_trust this witness to tell the truth on this topic. 40. Under most circumstances, I would pp:_ be likely to believe what this witness says about the topic. 41. I would like to have this witness as a personal friend. 42. The character of this witness is good. Factor 1 (Character) .34 .73* .20 .69* .49 .32 .70* .61* .59 .46 .59* Factor 2 (Authority) -.03 .ll .35 .OO .13 .18 .20 -.O2 .16 032 .27 -001 .17 .13 .17 .32 .2O .31 .17 .17 *Indicates items was retained for that factor. dichotomously coded as 38 being either right or wrong. These data were then subjected to an item analysis and those items that demonstrated low relia- bilities were eliminated. Forty-six items were retained which yielded an alpha coefficient of .89. Juror interest was measured utilizing three, seven-point scales. Subjects were asked to While respond to the following items: watching this witness, I was: very interested interested somewhat interested undecided somewhat uninterested uninterested very uninterested watching this witness, my mind wandered: all of the time most of the time quite often some of the time occasionally rarely n ever I found the testimony presented by the witness: very easy to follow easy to follow somewhat easy to follow 39 undecided somewhat difficult to follow difficult to follow ______ very difficult to follow These three items yielded an alpha coefficient of .70. Jurors' verdicts were measured by having the subjects indicate whether they decided the defendant was guilty of negligence or not guilty of neg- ligence. The measure was obtain immediately after the presentation of the defendant's testimony. Award was measured by having subjects write the amount of money they decided the defendant should pay the plaintiff. The subjects could award a minimum of $0.00 and a maximum of $3,000.00. These figures were derived with the assistance of legal experts, given the nature of the case and the extent of the injury suffered by the plaintiff. An additional point about the measure of award deserves attention. After consulting with a Judge and two attorneys, it became apparent that in this case, a plaintiff could sue for damages that would fall into three categories: (1) medical expenses, (2) lost wages, and (3) pain and suf- fering. The costs for the first two categories were fixed, while the latter was variable. Moreover, if the jurors found the defendant guilty of negligence, the first two categories would usually be paid automati- cally. That is, if found guilty, the defendant would usually have to pay for medical expenses and lost wages. After contacting a local hospital and a labor negotiator, the following amounts were established for these two categories: $300.00 for medical expenses and $1,920.00 for lost wages. Thus, for this case, if found guilty, the defendant would have to pay the plaintiff $2,220.00 to cover medical expenses and lost wages. The subjects 40 had to decide how much more the defendant would have to pay to cover pain and suffering. As was mentioned earlier, this amount was set with a min- imum of $0.00 and a maximum of $3,000.00. Chapter III RESULTS This section presents the results of the data analyses for the 25 hypotheses posited earlier. The results will be discussed in four sec- tions. The first three sections consider: (1) the effects of witness type, (2) the effects of camera shot for the strong witness, and (3) the effects of camera shot for the weak witness. For these sections, the significant results will be presented first, followed by the nonsignifi— cant results. The fourth section will discuss the acceptability of the general interaction hypothesis posited in Hypothesis 31. Two-way analysis of variance was used to test the effects of camera shot and witness type upon each of the eight dependent variables. The Newman-Keuls procedure was employed for comparisons among cell means. For all statistical tests, the .05 level of significance was required. The Effects of Witness Type The presentational style of the witness was found to have a signifi- cant effect upon subjects' ratings of the composure of the witness. The means and analysis of variance summary are reported in Table 6. An in- spection of the mean composure ratings indicated that the strong witness was perceived more composed than the weak witness in all three camera shots. The results of the Newman-Keuls test indicated that all three differences were significant. Moreover, the strength of the relationship 2 was found to be large (n = .50). 41 Table 6 . Means and Analysis of Variance Summary of the Effects of Cam- era Shot and Witness Type on the Assessment of the Witness' Composure. Strong Witness Weak Witness 3;- Close-up 37.33a 28.52b 32.93 Medium 37.74a 22.78C 30.26 Long 39.62a 24.37C 32.00 §' 38.23 25.22 §'= 31.73 Source Sum of Squares df Mean Square F n2 Camera Shot 197.98 2 98.99 2.46 —--- Witness Type 6857.51 1 6857.51 170.10** .50 Camera Shot X Witness Type 357.38 2 178.69 4.43* .03 Within Groups 6289.19 156 40.32 Total 13702.06 161 Note: Means containing a common letter do not differ significantly. *p<.05. **p<.001. 43 Perceptions of the witness' qualification was found to be signifi- cantly affected by the presentational style of the witness. The mean qualification ratings are reported in Table 7. An inspection of the means indicated that the strong witness was perceived more qualified than the weak witness in all three camera shots. However, the results of the Newman-Keuls test indicated that only the ratings in the close—up shot were significantly different. A measure of the strength of the re- lationship indicated the effect obtained was small (n2 = .03). The perceived dynamism of the witness was also found to be signifi— cantly affected by witness type. The mean dynamism ratings are reported in Table 8. An analysis of the means indicated that the strong witness was perceived significantly more dynamic than the weak witness in all three camera shots. In addition, the strength of the relationship was found to be large (n2 = .40). The presentational style of the witness was found to have a signif- icant effect upon the amount of information retained by the subjects. The mean retention scores are reported in Table 9. An inspection of the means indicated that subjects retained more information when exposed to the strong witness in the close-up and medium shots. However, in the long shot, subjects retained more information when exposed to the weak witness. The results of the Newman-Keuls test indicated that only the differences found in the close-up and medium shots were significant. Overall, subjects retained more information when exposed to the strong witness. Clearly, the retention scores in the close-up and medium shots are the major contributors to this effect. It should be noted that the . 2 strength of the relationship was found to be small (n = .02). 44 Table 7. Means and Analysis of Variance Summary of the Effects of Cam- era Shot and Witness Type on the Assessment of the Witness' Qualification. Strong Witness . Weak Witness E Close—up 24.37a 22.04b 23.20 Medium 23.44ab 22.85ab 23.15 Long 23.81 22.92 23.37 ab ab E 23.88 22.60 3? = 23.40 2 Source Sum of Squares df Mean Square F n Camera Shot 1.44 2 .72 <1 ---- Witness Type 65.49 1 65.49 4.59* .03 Camera Shot X Witness Type 23.42 2 11.71 <1 ---- Within Groups 2227.26 156 14.28 Total 2317.61 161 Note: Means containing a common letter do not differ significantly. *p<.05. 45 Table 8. Means and Analysis of Variance Summary of the Effects of Cam- era Shot and Witness Type on the Assessment of the Witness' Dynamism. Strong Witness Weak Witness £- Close-up 19.48a 14.89b 17.19 Medium 19.88a 14.15b 17.02 Long 20.11a 13.59b 16.85 §' 19.83 14.21 §'= 17.02 Source Sum of Squares df Mean Square F n2 Camera Shot 3.00 2 1.50 <1 ---- Witness Type 1277.93 1 1277.93 105.44** .40 Camera Shot X Witness Type 25.35 2 12.68 1.05 ---— Within Groups 1890.67 156 12.12 Total 3196.95 161 Note: Means containing a common letter do not differ significantly. **p<.OOl. Table 9. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on Retained Information. xl Strong Witness Weak Witness Close-up 36.89ac 33.26b 35.07 Medium 37.74ac 33.00b 35.37 Long 34.70abc 36.85C 35.78 3? 36.44 34.37 3'6 = 35.41 Source Sum of Squares df Mean Square F n2 Camera Shot 13.48 2 6.74 <1 -——- Witness Type 174.22 1 174.22 3.99* .02 Camera Shot X Witness Type 369.33 2 184.67 4.23* .05 Within Groups 6798.07 156 43.58 Total 7355.10 161 Note: Means containing a common letter do not differ significantly. *p<.05. 47 The amount of interest expressed by subjects was also found to be significantly affected by the presentational style of the witness. The mean interest scores are reported in Table 10. An inspection of the means indicated that subjects expressed greater interest when exposed to a strong witness than a weak witness in all three camera shots. The re- sults of the Newman-Keuls test indicated that only the interest ratings in the close-up and medium shots were significantly different. A measure of the strength of the relationship indicated a moderate effect (n2 = .10). The means and analysis of variance summary for the dependent variables of safety, authority, character and award are reported in Tables 11, 12, 13, and 14 respectively. The results of the analyses indicated that the presentational style of the witness had no significant effect upon these variables. The frequencies of verdict for the strong and weak witnesses are reported in Table 15. A chi-square test was utilized to assess the re- lationship between witness type and the verdicts reported by the subjects. The results indicated that the relationship between these variables was not significant (x2 = .91, df = 1, p > .05). The Effects of Camera Shot for the Strong Witness Perceptions of the strong witness' authority was found to be signif- icantly affected by the camera shots. The mean authority ratings are re- ported in Table 12. An inspection of the means indicated that the strong witness was perceived most authoritative in the close—up shot and least authoritative in the long shot. Moreover, the authority ratings in the medium shot were higher than in the long shot, but lower than in the close-up. The results of the Newman-Keuls test indicated that only the Table 10. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of Juror Interest. Strong Witness Weak Witness 2' Close-up 14.56a 11.92b 13.24 Medium 15.07a 11.81b 13.44 Long 14.07a 13.33ab 13.70 I 14.57 12.36 36: 13.46 Source Sum of Squares df Mean Square F n2 Camera Shot 5.81 2 2.91 <1 -—-- Witness Type 197.78 1 197.78 18.13** .10 Camera Shot X Witness Type 46.38 2 23.19 2.13 ———- Within Groups 1702.30 156 10.91 Total 1952.27 161 Note: Means containing a common letter do not differ significantly. **p<.001. 49 Table 11. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Safety. x| Strong Witness Weak Witness Close-up 21.22 21.52 21.37 Medium 21.96 20.89 21.43 Long 21.26 21.37 21.31 E 21.48 21.26 ‘x' = 21.37 2 Source Sum of Squares df Mean Square F n Camera Shot .33 2 .17 <1 ---- Witness Type 2.00 1 2.00 <1 ---- Camera Shot X Witness Type 14.93 2 7.47 <1 ---- Within Groups 2408.52 156 15.44 Total 2425.78 161 50 Table 12. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Authority. Strong Witness Weak Witness 3? Close-up 19.92a 17.63aC 18.78 Medium 18.66ab 16.03C 17.35 Long 16.77bC 18.15ac 17.27 I 18.46 17.27 I = 17.86 2 Source Sum of Squares df Mean Square F n Camera Shot 67.94 2 33.97 1.68 ---- Witness Type 56.89 1 56.89 2.82 ---— Camera Shot X Witness Type 133.00 2 66.50 3.29* .04 Within Groups 3151.19 156 20.20 Total 3409.02 161 Note: Means containing a common letter do not differ significantly. *p<.05. 51 Table 13. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on the Assessment of the Witness' Character. x| Strong Witness Weak Witness Close-up 35.26 33.93 34.59 Medium 34.74 34.56 34.65 Long 33.15 34.44 33.80 3? 34.38 34.31 36 = 34.35 2 Source Sum of Squares df Mean Square F n Camera Shot 24.53 2 12.27 <1 ---- Witness Type .22 1 .22 <1 ---— Camera Shot X Witness Type 46.93 2 23.47 <1 ---- Within Groups 5514.96 156 35.35 Total 5586.64 161 52 Table 14. Means and Analysis of Variance Summary of the Effects of Camera Shot and Witness Type on Jurors' Award to the Plaintiff. Strong Witness Weak Witness R- Close-up 842.86 810.00 826.43 Medium 708.33 1257.00 982.67 Long 727.27 931.82 829.55 2' 759.49 999.61 §'= 879.55 2 Source Sum of Squares df Mean Square F n Camera Shot 251212.31 2 125606.15 <1 ---- Witness Type 918159.59 l 918159.59 1.28 ---- Camera Shot X Witness Type 960261.36 2 480130.68 <1 ---- Within Groups 44770044.20 62 722097.49 Total 46893747.06 67 Table 15. Frequencies of Verdict for the Strong and Weak Witnesses. Strong Witness Weak Witness Innocent 44 50 Guilty 37 31 53 mean authority ratings in the close-up and long shots differed signifi- cantly. The mean authority ratings in the medium shot did not differ significantly from the mean authority ratings in the close-up or long shots. The means and analysis of variance summary for the dependent vari- ables of composure, safety, qualification, dynamism, character, informa- tion retention, interest, and award are reported in Tables 6, 7, 8, 9, 10, 11, 13, and 14 respectively. The results of the analyses indicated that camera shot had no significant effect upon these variables for the strong witness. The frequencies of verdict for the three camera shots of the strong witness are reported in Table 16. A chi-square test was used to assess the relationship between camera shot and the verdicts reported by sub- jects exposed to the strong witness. The results indicated the rela- tionship was not significant (x2 = .70, df = 2, p > .05). Table 16. Frequencies of Verdict for the Three Camera Shots of the Strong Witness. Close-up Medium Long Innocent 13 15 16 Guilty 14 12 11 The Effects of Camera Shot for the Weak Witness The results fer the effects of camera shot for the weak witness yielded two significant effects. Camera shot was found to have a sig- nificant effect upon: (1) subjects' perceptions of the weak witness' composure and (2) the amount of information retained by the subjects. 54 The mean composure ratings for the weak witness are reported in Table 6. An inspection of the means indicated the weak witness was per- ceived most composed in the close—up shot and least composed in the me- dium shot. The results of the Newman-Keuls test indicated that the weak witness was perceived significantly more composed in the close—up than in either the medium or long shots. The composure ratings in the medium and long shots were not fbund to differ significantly. The mean information retention scores are reported in Table 9. An inspection of the means indicated that subjects retained the greatest amount of information in the long shot, while subjects exposed to the medium shot of the weak witness retained the least amount of information. The results of the Newman-Keuls test indicated that the infbrmation re- tention scores were significantly greater in the long shot than in either the medium or close-up shots. The information retention scores in the medium and close-up shots were not found to be significantly different. The means and analysis of variance summary for the dependent vari- ables of safety, qualification, dynamism, authority, character, interest, and award are reported in Tables 7, 8, 10, ll, 12, 13, and 14 respec- tively. The results of the analyses indicated that camera shot had no significant effect upon these variables for the weak witness. The frequencies of verdict for the three camera shots of the weak witness are reported in Table 17. A chi-square test was utilized to test the relationship between camera shot and the verdicts reported by subjects exposed to the weak witness. The results indicated that the relationship was not significant (x2 = .10, df = 2, p > .05). 55 Table 17. Frequencies of Verdict for the Three Camera Shots of the Weak Witness. Close—up Medium Long Innocent 17 17 16 Guilty 10 10 11 Summarizing the results of the data analyses discussed thus far, it was found that the presentational style of the witness had a significant effect upon the variables of composure, qualification, dynamism, informa- tion retention, and interest. Camera shot had a significant effect upon subjects' perceptions of the strong witness' authority and the weak wit— ness' composure. Moreover, camera shot had a significant effect upon the amount of information retained by subjects exposed to the weak witness. No other significant effects were obtained for either camera shot or wit- ness type. Thus, the data supported Hypotheses 1, 3, 4, 7, and 8; and partially supported Hypotheses 15 and 27. Hypotheses 2, 5, 6, 9, 10, 11, 12, 13, l4, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 28, 29, and 30 were not supported by the data and therefore were rejected. Given the lack of support for 18 of the 20 hypotheses posited for the effects of camera shot, Hypothesis 31 (the general interaction hy- pothesis) was rejected. Still, it should be noted that the results yielded three significant interaction effects. Specifically, camera shot was found to interact with witness type producing significant ef- fects upon the variables of composure, authority, and interest. The nature of these interactions have already been discussed in terms of their simple main effects. However, the effects of camera shot and 56 witness type upon the variables of composure and information retention warrant further inspection since witness type alone was found to have a significant effect upon these variables. That is, additional analysis was necessary in order to determine if the main effect for witness type could be interpreted given that significant interactions were also obtained. Keppell (1973) contends that when both significant main effects and interactions are obtained, the ability to interpret the main effects is dependent upon whether the interaction is ordinate or disordinate. An ordinate interaction occurs when the relative ranking of the levels of one factor do not change at the different levels of the other factor. A disordinate interaction occurs when the relative ranking of the levels of one factor changes at the different levels of the other factor. The mean composure ratings for the strong and weak witness are plot- ted as a function of camera shot in Figure 1. An inspection of Figure 1 indicated that the strong witness was consistently perceived more com- posed than the weak witness at all three levels of camera shot. There- fore, the interaction was found to be ordinate, allowing the main effect for witness type to be interpreted. The mean information retention scores for subjects exposed to the strong and weak witness are plotted as a fUnction of camera shot in Fig- ure 2. An inspection of Figure 2 indicated that subjects retained more information when exposed to a close-up or medium shot of the strong wit— ness than subjects who were exposed to a weak witness in these same shots. However, for the long shot, subjects exposed to the weak witness retained more information than subjects exposed to the strong witness. Therefore, the interaction is clearly disordinate given that a consistent pattern did not emerge. Thus, no general conclusion regarding the effect of witness type upon information retention can be made. Mean Composure Ratings 57 40— 39- 38' STRONG WITNESS 37- 36- 35- 34— 33- 32- 31- 30- 29‘ WEAK WITNESS 28- 27- 26- 25- 24- 23- 22- 21- 20- l l [—7 CLOSE-UP MEDIUM LONG SHOT SHOT SHOT Figure 1. Graphic Representation of Witness' Composure as a Function of Camera Shot and Witness Type. Mean Retention Scores 58 40— 39- 38- 37' STRONG 36_ WITNESS 35- 34- WEAK 33- WITNESS 32- 31- 30- 29- 28- 27— 26— 25- 24- 23- 22- 20- l 7| ’1 CLOSE-UP MEDIUM LONG SHOT SHOT SHOT Figure 2. Graphic Representation of Information Retention as a Function of Camera Shot and Witness Type. Chapter IV DISCUSSION The results of this study revealed a number of significant effects for the independent variables of witness type and camera shot. This sec- tion will discuss those effects and provide plausible explanations for their occurrence. The effects of witness type will be discussed first, followed by a discussion of the effects of camera shot for each witness type. A discussion of the limitations of the study as well as recommen- dations to the legal community will also be provided. The Effects of Witness Type The presentational style of the witness was found to have a signifi- cant effect upon the subjects. Specifically, subjects perceived the strong witness to be significantly more composed, qualified, and dynamic than the weak witness. Moreover, subjects exposed to a strong witness retained significantly more information and expressed greater interest than subjects exposed to a weak witness. As was mentioned earlier, the composure measure was primarily em- ployed as a check on the manipulation of witness type. The items used in this measure were derived from the literature as indicants of the presentational style of the strong and weak witnesses. The results in- dicated that the strong witness was perceived significantly more composed than the weak witness at all three levels of camera shot. Therefore, the results support the findings obtained by previous researchers in that 59 60 subjects did perceive the two witnesses as hypothesized. Perhaps more important to the present study, the results indicated the manipulation was successful. It should be noted that the results also revealed a significant cam- era shot by witness type interaction fer the composure variable. However, further analysis indicated that this interaction was ordinate, allowing interpretation of the main effect for witness type. The interaction ef- fect will be discussed later in this chapter. The perceived qualification of the witness was also significantly affected by witness type. The results indicated that the strong witness was perceived more qualified than the weak witness. This finding repli- cates the results obtained by previous researchers (Miller 8 Hewgill, 1964; Sereno 8 Hawkins, 1967). However, the results also indicated that the strong witness was perceived significantly more qualified only in the close-up shot. Hence, the difference in the close-up supplied the great- est contribution to the overall effect. A plausible explanation for this result is that the close-up shot tended to emphasize the characteristics of the witness and focused the subjects' attention to those characteris- tics. Some support is lent to this contention since the strong witness received his highest ratings in the close-up shot, while the weak witness received his lowest ratings in the close-up shot. The findings revealed that subjects perceived the strong witness to be significantly more dynamic than the weak witness. An examination of the means indicated that this pattern was consistent across all three camera shots. One possible explanation for this finding stems from the presentational fluency of the two witnesses. The strong witness was flu- ent and nonhesitant while the weak witness was nonfluent and hesitant. 61 The nonfluencies of the weak witness may account for his lower dynamism ratings. This interpretation is supported by the results of previous research (Miller 8 Hewgill, 1964; Sereno 8 Hawkins, 1967; McCroskey 8 Mehrley, 1969). The amount of infbrmation retained by the subjects was also found to be affected by witness type. Overall, it appeared that subjects exposed to the strong witness retained significantly more information than sub- jects exposed to the weak witness. However, the results also revealed a significant camera shot by witness type interaction. Further inspection indicated that this interaction was disordinate, making it difficult to interpret any general effect for witness type. Therefore, the specific effects for the interaction will be discussed in lieu of the general main effect. An inspection of the mean information retention scores revealed that subjects retained more information when exposed to the strong witness in the close-up and medium shots. However, in the long shot, subjects ex- posed to the weak witness retained more information than subjects exposed to the strong witness. Subsequent analysis revealed that only the differ— ences in the close-up and medium shots were significant. A plausible explanation for this finding involves the nonverbal be- havior of the weak witness. As was mentioned earlier, the behaviors mani— fested by the weak witness included fidgeting, tapping his fingers and feet, tensing his muscles, occasional sighs, etc. These behaviors may have been distracting to the subjects. Assuming this is true, the medium and close-up shots may have emphasized those cues, increasing the amount of distraction experienced by the subjects. However, the long shot may not have emphasized those cues to the same extent as the medium and 62 close-up shots. Thus, the nonverbal cues may have been only mildly dis- tracting in the long shot. Research on the effects of distraction upon information retention has produced some seemingly inconsistent results. Distraction has been found to increase recall of message content (Silverman 8 Regula, 1968) and to decrease recall of message content (Vohs, 1964; Gardner, 1966; Haaland 8 Vankatesan, 1968). In a review of the literature, Baron, Baron and Miller (1973) suggest that one factor that may account for the dis- crepancy in results is whether the distraction could be ignored. They argue that if a distracting stimulus is sufficiently severe, subjects may attend more to the distracting stimulus than to the content of the message. However, if the distraction is mild, subjects may attempt to block the distracting stimulus by attending more to the content of the message. Assuming that the cues emitted by the weak witness were much more distracting in the close-up and medium shots than in the long shot, then subjects exposed to the weak witness in the close-up and medium shots may have been attending to the distracting behaviors rather than to the con- tent of the message. The subjects exposed to the long shot of the weak witness may have attended more to the information presented in order to block the distracting behaviors. This would explain the lower informa- tion retention scores obtained in the close-up and medium shots, as well as the increase in retention in the long shot of the weak witness. The last variable significantly affected by witness type was the amount of interest expressed by the subjects. The results revealed that subjects consistently expressed greater interest when exposed to the strong witness than the weak witness. However, significant differences 63 were obtained only for the close-up and medium shots. A plausible expla— nation for this finding also involves the nonverbal behavior of the weak witness. As was argued above, the cues emitted by the weak witness may have been distracting to the subjects. Moreover, it was argued that this distraction may have made it difficult for subjects to attend to the tes- timony presented by the weak witness, particularly in the close-up and medium shots. The amount of difficulty associated with attending to the testimony may account for the lower interest ratings reported by subjects exposed to the weak witness. Still, the lack of significant differences for subjects' expressed interest in the long shot deserves attention. Earlier it was argued that the nonverbal cues emitted by the weak witness in the long shot may have been only mildly distracting, while the cues presented in the medium and close-up shots may have been much more distracting. Hence, subjects might have been better able to follow the testimony presented by the weak wit- ness in the long shot than in the close—up or medium shots. The reduction in difficulty of following the weak witness' testimony in the long shot may account for the increase in subjects' expressed interest for that shot. The Effects of Camera Shot The results indicated that camera shot had a significant effect upon subjects' perceptions of the strong witness' authority and the weak wit- ness' composure. In addition, camera shot had a significant effect upon the amount of information retained by subjects exposed to the weak witness. The results for the effect of camera shot upon perceptions of the strong witness' authority indicated that the strong witness was perceived to be significantly more authoritative in the close-up shot than in the long shot. One explanation for this result is that the close-up shot 64 emphasized the characteristics of the strong witness more than the long shot, resulting in higher authority ratings in the close-up. Given this reasoning, one would expect the close-up shot to emphasize the character— istics of the weak witness more than the long shot. Thus, the authority ratings of the weak witness should be higher in the long shot than in the close—up shot. The mean authority ratings for the weak witness were higher in the long shot than in either the close-up or medium shots, but the dif- ferences were not significant. It is not immediately clear why the effect failed to hold for the weak witness. One possible explanation is that subjects interpreted the nonverbal cues emitted by the weak witness differently. Some support for this contention is offered by subjects' responses to an open-ended ques- tion regarding what they liked least about the trial. Some subjects ex— posed to the weak witness derogated the questioning attorney for making the witness nervous. Others commented on the evasiveness of the witness. From an attribution viewpoint, it seems reasonable to assume that percep- tions of the weak witness will differ between subjects who attributed the witness' behavior to be a result of the attorney's questioning behavior, as opposed to those subjects who attributed the witness' behavior to his own evasiveness (see Jones 8 Davis, 1965; Kelley, 1967; Shaver, 1975). Thus, there may have existed some ambiguity associated with the cues pre- sented by the weak witness, which resulted in subjects making different attributions for the weak witness' behavior. If a given Shot increased the ambiguity of the cues presented by the weak witness, then an indication of this effect should be revealed in the variances for each shot. That is, if subjects were interpreting the cues similarly, then the variance should be relatively small within each 65 condition, and consistent across the conditions. If the subjects were interpreting the cues differently, and the shots had no effect on this difference, then the variances should be large within each condition, but consistent across the conditions. However, if the shots had different ef- fects upon the ambiguity of the cues presented, then the variances should be inconsistent across conditions. An inspection of the variances for the weak witness for each camera shot revealed an interesting pattern. The variance in the medium shot was . . 2 _ much smaller than the var1ances 1n the close-up and long shots (s close—up- 2 2 25.01; s medium=ll'72; s =25.90). A test for homogeneIty of var1ance long was computed using Hartley's test (see Winer, 1971). The results indi— cated that the differences in variance was not significant, but approached significance (F = 2.42; p < .07). The critical F value for this test was 2.67. Thus, there is some support for the contention that subjects were interpreting the cues emitted by the weak witness differently in the close-up and long shots. This remains speculative since it is not known precisely what cues are being used nor exactly how they were being interpreted. It should be noted that the variances for the Strong witness were =22.38 =23.72). How- . 2 2 2 also different (8 close-up=l3‘”69 S 3 5 long medium ever, the test for homogeneity of variance indicated the variances were not significantly different (F = 1.76; p < .19). In sum, differences in perceptions of the strong witness' authority may be the result of the close-up shot emphasizing the cues emitted by the strong witness. The lack of significance obtained for the weak wit- ness may have been the result of differences in subjects' interpretations of the cues emitted by the witness. It should be noted that the strength 66 of the effect on perceptions of the witness' authority was found to be small (n2 = .04). Camera shot was found to have a significant effect upon subjects' perceptions of the weak witness' composure. The results indicated that the weak witness was perceived more composed in the close-up shot than in either the medium or long shots. The composure ratings in the medium and long shots did not differ significantly. A plausible explanation for this pattern of effects involves the behavioral cues used by the weak witness to indicate low composure. The behaviors of the weak witness included nonfluent speech; low eye contact; leaning away from the questioning attorney; fidgeting; tapping of fingers and feet; a closed body position, etc. These behaviors, as well as the measure of composure, were derived from the literature on verbal and non— verbal behavior (cf., Reece 8 Whitman, 1962; Mehrabian, 1969, 1971; Harrison, 1974). An inspection of these behaviors indicates that a majority of the cues indicative of low composure emanate from the body and not the face. Of course, the paralinguistic cues also indicate com- posure, but these were consistent across all three shots. The body cues were not consistent, for they were not seen in the close-up shot. As- suming the body cues add information revealing the degree of composure, then it makes sense that the weak witness would appear more composed in the close-up shot where these cues were absent. Additional support is lent to this contention when the long shot is examined. In the long shot, the body cues are available, but they are not as prominent as in the medi- um shot. Moreover, facial affect cues are also not as prominent as in the close-up shot. Given the above explanation, one would expect to find the composure ratings fer the weak witness to be higher in the long shot than 67 in the medium shot, but lower than in the close-up shot. The data revealed this pattern of relationships. Still, it should be noted that the overall effect was weak (n2 = .03). The final relationship to be discussed concerns the effect of camera shot on the amount of information retained by subjects exposed to the weak witness. The results indicated that subjects retained significantly more information in the long shot than in either the medium or close-up shots. The amount of information retained in the medium and close—up shots did not differ significantly. The pattern of effects obtained can be explained by the nonverbal behavior of the weak witness. Earlier it was argued that the nonverbal behaviors of the weak witness may have been distracting to the subjects. It was fUrther argued that the distraction may have been more pronounced in the close-up and medium shots than in the long shot. Therefore, sub- jects may have attended more to the cues of the weak witness in the close- up and medium shots and not to the information presented. In the long shot, the distracting cues may have been only mildly distracting. Thus, subjects could have attended more to the information presented in order to block the distracting cues. This would explain why subjects retained more infermation in the long shot fer the weak witness. However, once again it should be noted that while the effect was significant, the strength of the relationship was weak (n2 = .05). Summary The results of this study revealed a number of significant effects fer witness type, but few effects for camera shot. In fact, camera shot alone had no significant effect upon any of the dependent variables 68 examined. Camera shot did interact with witness type producing signifi— cant effects upon the subjects' perceptions of the strong witness; author- ity and the weak witness' composure. Moreover, camera shot affected the amount of information retained by subjects exposed to the weak witness. Clearly, the number of significant effects obtained for camera shot is far less than the number predicted. The majority of the hypotheses for the effects of camera shot were rejected by the data. This might lead one to conclude that the conventional wisdom of television and film producers concerning the effects of different camera shots is in error. However, such a conclusion is not warranted by the data. Conceivably, the lack of significant differences between the shots may be a function of the way in which they were employed. Once established, all shots remained static with no other shots introduced. The effects may very well have been dif- ferent if different shots were employed contiguously. Thus, the close-up shot may consistently provide dramatic emphasis only when preceded by a longer shot. Moreover, the emphasis may increase if the preceding shot is a long shot rather than a medium shot. The present study cannot address these issues. It compares the dif- ferent shots with regard to the image conveyed. Therefore, it is important to note that the findings generalize only to the static nature of the shots themselves, and not to the numerous combination of shots that could be employed. Limitations Given the desire to generalize the findings of the present study to courtroom settings involving actual jurors, the present study, like pre- vious studies that have employed role-playing jurors as well as trial syn- opses, possesses a number of limitations. The limitations associated with 69 the procedures and sampling in legal research have been discussed else— where and the relevant points will only be summarized here (see Miller, Fontes, Boster 8 Sunnafrank, 1977). The first limitation concerns the use of students as role-playing jurors as opposed to using actual jurors. Previous research has suggested that student jurors are demographically different than actual jurors. Miller, e£_§1, (1975) conducted a series of studies examining the effects of different modes of presentation on jurors' information retention. In one study, student role-playing jurors were used. In another, actual jurors were used. The authors found the information retention scores to be practically at ceiling in all conditions when student jurors were em- ployed. However, when the same study was conducted using actual jurors, the information retention scores were much lower, and significantly dif- ferent from the scores obtained from the student jurors. The demographic data obtained from these two samples indicated differences in educational level, with the educational level of actual jurors falling far short of college students. Miller §£_213 (1977) reason that college students may simply be smarter than typical jury panels. Student jurors also tend to differ from actual jurors in their dis- tribution of verdicts. Recent research has found that student jurors tend to manifest a liberal bias in trials; i.e., their verdicts favor the de- fendant significantly more frequently than actual jurors (Simon 8 Mahan, 1971; Miller, Bender, Florence 8 Nicholson, 1974; Miller, g£_§13, 1975). Miller, gt_al. (1977) offer two possible explanations for these findings. First, student jurors may have more liberal social and judicial attitudes because they are younger, or may have a "more sophisticated, complex def- inition of reasonable doubt" due to their extensive educational background 70 (p. 5). Whichever explanation is preferred, it still appears that student jurors differ from actual jurors in their final decision. An additional area of concern surrounds the differences in expecta- tions between student jurors and actual jurors. In most of the research conducted by the Michigan State Legal Communication Research Team involv- ing actual jurors, elaborate cover stories were used to explain the un— usual number of jurors employed in each study (cf., Miller 8 Siebert, 1975; Miller 8 Fontes, 1977). Moreover, in some studies, videotape cam- eras were visible in the courtroom and jurors usually completed a ques- tionnaire prior to their deliberation. One might have expected that jurors would become suspicious of these procedures, as they deviate from typical trial procedures. However, as Miller, g£_a13 (1977) note "...not a single juror expressed doubt about the trial's validity. Instead, they all indicated that they thought the trial was authentic and that their verdicts would be binding" (p. 6—7). In contrast, student jurors are often suspicious of the procedures employed even when no deception is involved (Miller, g£_a13, 1977). The suspicious nature of the college student is most probably due to the num- ber of studies they have participated in which have involved deception. Miller, pt_al, (1977) contend that since "...deception is a common pro- cedural commodity, they seem determined to discover the latest subter- fuge" (p. 7). Thus, student jurors may differ from actual jurors in that they are suspicious at the outset, which may ultimately affect their be- haviors and responses. Apart from the limitations involved in using a student sample, two additional limitations arise as a result of the procedures employed. First, a complete trial was not used. Rather, subjects were presented 71 with a trial syn0psis and then viewed a deposition from the defendant. Obviously, the procedure employed differs markedly from a complete trial. For one thing, the amount of information presented is less than the amount available in a complete trial. This reduction can have a number of seri— ous consequences. First, the manipulation may become quite transparent in a truncated version of a trial. Second, the elimination or oversimplica- tion of many variables may cause the independent variable(s) to achieve exaggerated importance. Similarly, a truncated version may eliminate variables that could heighten or reduce the effects of the independent variable(s). Recently, Sunnafrank, Miller 8 Fontes (1978) conducted a study com- paring the effects of a full videotaped trial, a videotaped synopsis, and a written synopsis on the decisions of role—playing jurors. The authors found that subjects differed significantly in their verdicts. Specifi- cally, subjects exposed to the full videotaped trial or the videotaped synopsis were more likely to find the defendant guilty and recommend more severe sentences than were subjects who read the written synopsis. The authors conclude that the mode of presentation in trial simulations exerts a significant impact on juror behavior (Sunnafrank, Miller 8 Fontes, 1978, 7-8). The procedures employed in the present study do not easily fit any previous presentational mode. A written synopsis was employed, but a videotaped deposition was added. Moreover, the testimony of only one participant (the defendant) was presented. Thus, it is difficult to assess the effect this procedure had on the subjects. However, it is noteworthy that the procedures clearly differ from what would occur in an actual trial. 72 The last procedural limitation concerns the measures of verdict and award. These measures were obtained using the individual as the unit of analysis. This obviously differs from what occurs in the courts. Clearly, all group process variables were eliminated by not having the subjects form juries for deliberation. Conceivably, verdicts and awards obtained from juries after deliberation may not coincide with mean individual awards or individual frequencies of verdict. Some support for this contention is offered by Miller, e£_§l, (1977). The authors report findings that indi- cate that mean individual awards were substantially larger than group awards. Hence, although there is limited data, there appears to be some merit to the contention that group process variables affect pre-delibera— tion verdicts and awards. It should be noted that the above limitations arise since the present study deviates from the "ideal" experiment. That is, all of the limita- tions could have been eliminated by conducting a full trial in a courtroom with actual jurors. Of course, such an endeavor is costly and necessary resources are not always available. Such was the case for the present study. The necessary resources were simply not available. Despite the limitations presented above, the present study has advan- tages over previous research conducted on the effects of camera shot. As was mentioned in Chapter I, most of the research conducted to date has failed to control for differences in potentially relevant source charac- teristics. The present study avoided that problem by using the same source and altering the presentational style of that source. Thus, the present study offers a better picture of the differences between three types of camera shots. 73 Of course, one of the objectives of the present study was to assess the impact of different camera shots in order to assist the legal commu- nity in making policy decisions concerning the use of camera shots when taping depositions. Given the limitations presented above, this objec- tive may be difficult to realize. The sample and procedures employed may seriously hamper the generalizability of the findings. However, as is the case with any research endeavor, certain trade—offs had to be made. A discussion of these trade-offs will now be offered. The present study attempted to uncover the effects of different cam- era shots for the method that is most commonly used in the legal setting: the deposition. This decision immediately raises a number of problems. First, it is unrealistic to present a deposition apart from an actual trial. However, including the deposition in an actual trial poses prob- lems. Generally, there arises the problem of controlling extraneous vari- ables which may alter the effects of camera shot (e.g., the behaviors of other witnesses, additional testimony, etc.). Of course, one could argue that these additional variables would be present anyway, and could be con- trolled across conditions. While there may be some merit to this argument, it too poses problems. For example, how should the trial (apart from the deposition) be presented: live or on videotape? The problems associated with showing the trial live are clear. For the present study, the trial would have had to have been presented six times. Attempting to control for variation in actors' presentational style across six conditions would have been exceedingly difficult. More- over, the monetary costs associated with the experiment would have been quite large. 74 The trial could have been videotaped, which would have eliminated the problem of variation in performance across conditions and might have reduced the cost as well. However, the question of which shot to use when taping the other participants arises. If all three were employed, the complexities of the design would have increased drastically. More- over, the answer to the basic question of differences between camera shots may have easily become obscured. Finally, regardless of whether a full trial is presented live or on videotape, the problem of controlling the presentational style of the other trial participants arises. Their style could have affected sub- jects' evaluations of the witness of interest apart from the effects of camera shot. Attempting to control the presentational style of the other trial participants would have increased the complexities of the study even more. Clearly, alternatives to the procedure employed create problems of experimental control and demand resources beyond what was available. Given available resources and a desire to assess the differences in cam- era shot controlling for extraneous variables, the procedures and sample utilized were employed. While the present study may be limited in its generalizability to the legal setting, it provides a useful beginning for future research. Some suggestions for future research will now be offered. Future Research One obvious direction fer fUture research on the effects of different camera shots would entail designing experiments that more closely approxi- mate the trial setting. This would greatly improve the generalizability of the findings and allow more confident recommendations to the legal community. 75 Given the findings of the present study, additional suggestions can be made. First, many of the explanations offered for the effects observed concern the nonverbal behavior of the weak witness. It was argued that the weak witness' nonverbal behavior may have been distracting to the subjects. Unfortunately, the present study did not measure distraction. Therefore, future research should employ a measure of distraction which would offer an assessment of the validity of these explanations. In addition, while some of the differences obtained may be due to the nonverbal behavior of the witnesses, it is not known which behaviors account for most of the variance. For example, the paralinguistic cues may account for more variance than any other cues for some variables or vice versa. This could be tested in a number of ways. The design of the study would have to be expanded to include two additional conditions which would contain only the audio portion of the testimony for the strong and weak witness. This would allow an assessment of the effects of the paralinguistic cues alone. The other conditions could be presented once with the audio portion included and once with it removed. This would allow comparisons of different cues alone as well as combinations of some of the cues. Finally, the present study examined three different camera shots independently. Future research may consider examining the processual nature of camera shots. That is, one could examine differences in cut- ting from a long shot to a close-up, a medium shot to a close-up, and a long shot to a medium shot. Of course, such a study contains problems of deciding when to change shots as well as generalizability problems. Yet, if these could be worked through, the results would be most useful in assessing the impact of shot changes. 76 Recommendations Given the pattern of results, the decision to use one camera shot over another is dependent upon the type of witness to be videotaped. If a strong witness is videotaped, and there is a concern fer the perceived authority of the witness, then a close-up shot should be employed. If authority is not a concern, then it does not really matter which shot is used. If a weak witness is being videotaped, and there is a concern for the perceived composure of the witness, then a close-up shot is recom- mended. However, if there is a greater concern for the amount of infor- mation retained by the jurors, then a long shot is recommended. Still, a final note concerning the selection of camera shots is worth mentioning. In all cases, the effects of the interaction of cam- era Shot and witness type were small. At best, five percent of the vari- ance could be explained by these interactions. From a practical stand- point, five percent is not very much. Therefore, in the final analysis, given a strong or weak witness, it probably does not matter whether a close-up, medium, or long shot is used. Perhaps it might be best to select a camera shot for standard usage when videotaping testimony; e.g., mandate that all videotaped testimony will be taken using a medi- um shot of the witness. Considering the limitations discussed earlier, the recommendations offered above are offered cautiously. Additional research that more closely approximates the realities of the legal setting is needed before recommendations may be offered with greater confidence. APPENDICES APPENDIX A INTRODUCTION TO JURORS 77 APPENDIX A INTRODUCTION TO JURORS MICHIGAN STATE UNIVERSITY College of Communication Arts East Lansing - Michigan - 48824 Department of Communication Dear Juror: You are about to hear a civil case involving an industrial accident. While we would like to show you the actual trial proceedings, we are un- able to because they were not videotaped. However, part of the defen- dant's testimony was taken during a deposition, which was videotaped and presented to the jury during the trial proceedings. Since we are unable to present the entire trial, a written summary of the important aspects of the trial is being provided. After reading this summary, you will have an opportunity to view the defendant's videotaped deposition. The plaintiff in this trial, John Hickson, is suing the defendant, Robert Montague, for costs resulting from an injury Mr. Hickson suffered. You will be asked to decide whether the defendant was negligent and caused the accident. If you determine that the defendant was negligent, you will also be asked to decide how much money the defendant will pay the plaintiff. Mr. Hickson is asking for $5,220.00 in damages: $300.00 for medical costs; $1,920.00 for lost wages; and $3,000.00 for pain and suffering. APPENDIX B TRIAL SUMMARY 78 APPENDIX B TRIAL SUMMARY JOHN HICKSON vs. ROBERT MONTAGUE The defendant in this case, Robert Montague, is charged with negli- gence. Specifically, the defendant is charged with negligence on a construction site which resulted in an injury for the plaintiff, John Hickson. Mr. Hickson is a laborer for the Field Construction Company. His duties include unloading materials delivered to the construction site and general cleanup. On February 9, 1976, Mr. Hickson was unloading a truckload of pro- pane tanks delivered to the construction site by the defendant, Robert Montague. The tanks were being lifted directly off the truck by an over- head crane. In addition, empty tanks were being lowered directly on the truck by the same crane. In the beginning, Mr. Montague was helping Mr. Hickson hookup and unload the tanks. During the loading and unloading of the tanks, Mr. Montague got off and on the truck a number of times to go over to a fire barrel to warm up, due to the fact that the temperature was below freez- ing. During one of these trips, Mr. Montague changed the position of the tailgate of the truck in order to use it as a step. Originally, the tailgate was even with the bed of the truck, and he lowered it such that it was halfway between the bed of the truck and the ground. This placed the tailgate approximately two feet below the bed of the truck. 79 During one of the lifts, Mr. Hickson was watching the tanks being lifted by the crane, and he started walking backwards on the truck. He continued looking up and walking backwards, and fell off the bed of the truck onto the tailgate. As a result of the fall, Mr. Hickson broke his right leg. He suffered a simple fracture of the tibia (the large bone between the ankle and the knee). At the time of the accident, Mr. Montague was standing by the fire barrel. Mr. Hickson contends that Mr. Montague did not tell him that he changed the position of the tailgate, and that he still expected the tailgate to be even with the bed of the truck. Mr. Hickson further contends that if the tailgate had not been changed, the accident would not have occurred. Mr. Hickson had his leg in a cast for four weeks, and after the cast was removed, the leg was bandaged for two additional weeks. He spent a total of six weeks on crutches, and was unable to work during that period of time. After six weeks, the leg had healed completely, and there were no additional complications. At that time, Mr. Hickson returned to work. APPENDIX C QUESTIONNAIRE ON JURY SIZE 80 THE NATIONAL SCIENCE FOUNDATION Questionnaire on Jury Size Department of Communication Michigan State University East Lansing. Michigan 81 Dear Juror: Now that you have read the trial summary and heard the testimony of the defendant, we would like you to respond to a number of questions concern— ing the testimony you have just heard. First, we would like to find out what your verdict is in this case. You must decide whether or not the defendant is guilty of negligence. If you find the defendant NOT GUILTY, then he will not have to pay the plaintiff any money. However, if you find the defendant GUILTY, then he must pay the plaintiff $2,220.00 to cover medical bills and lost wages. In addi- tion, you must decide how much money the defendant must pay the plaintiff for pain and suffering. The plaintiff is asking fer $3,000.00 fer pain and suffering. You may award the plaintiff any amount you choose, from a minimum of $0.00, to a maximum of $3,000.00. I find the defendant: not guilty of negligence guilty of negligence If you marked "not guilty," please turn to the next page. If you marked "guilty," how much money should be awarded to the plaintiff? (minimum amount: $0.00/maximum amount: $3,000.00) (write in amount) 82 THE FOLLOWING QUESTIONS CONCERN SOME OF THE INFORMATION THAT THE WITNESS PROVIDED DURING THE COURSE OF THE DEPOSITION YOU HAVE JUST HEARD. PLEASE ANSWER EACH QUESTION BY PLACING A CHECKMARK (V5 ON THE LINE BESIDE THE RE- SPONSE WHICH YOU FEEL IS MOST CORRECT. PLEASE WORK CAREFULLY AND ANSWER EACH QUESTION. YOU MAY HAVE ALL OF THE TIME YOU NEED TO COMPLETE THE QUESTIONNAIRE. IF YOU HAVE ANY QUESTIONS WHILE FILLING THE QUESTIONNAIRE OUT, PLEASE RAISE YOUR HAND AND WE WILL BE HAPPY TO ASSIST YOU. 1. What is the name of the witness whose deposition was being taken? a Donald 0. Montague b. John Hickson c. Charles R. Barnett d. Robert Montague e. Robert Stein 2. Where does the witness live? a. Lexington Road, Wacousta b Upton Road, Chicago c. Upton Road, Wacousta d. Upton Road, Grand Rapids e. the witness did not say where he lived 3. What type of discharge was the witness given when he returned to civilian life? a. medical discharge b. dishonorable discharge c. court martialed d. honorable discharge e. general discharge 4. How much education had the witness completed prior to entering the a. 8 years of education (i.e., through the 8th grade) b. one year of high school c. high school d. one year of college e. two years of college 5. How long has the witness been married? a. 2 years b. 3 years c. 4 years d. 6 years e. 7 years 10. ll. 83 Where is the witness currently employed? a Propane FUel Distributors, Inc. b Liquipane Fuel Services, Inc. c. Liquipane Gas Services, Inc. d. Liquipane Gas Distributing Company e. Propane Gas Distributing Company How long has the witness been employed by that company? a. 1 year in October b 2 years in October c. 4 years in October d 5 years in October e 6 years in October How heavy are the full propane tanks? a seventy pounds b one hundred pounds c. one hundred fifty pounds d one hundred seventy pounds e two hundred pounds What was the length and width of the bed of the vehicle? a. 12 feet long and 8 feet wide b. 15 feet long and 6 feet wide c. 17 feet long and 7 feet wide d. it did not have a bed e the witness was a dispatcher and did not know the dimen- sions of the vehicle bed How were the cylinders positioned when being transported? a. one level of cylinders in an upright position b. two levels of cylinders in an upside down position 0. three levels of cylinders placed sideways in the vehicle d. four levels of cylinders stacked vertically in the vehicle e. five levels of cylinders placed angularly in the vehicle a. 1973 b. 1974 c. 1975 d 1976 12. 13. 14. 15. l6. 17. 84 During what month did the mishap occur? a. c. d. January February April June October did the witness drive the vehicle in question? every day several times a week once a week once a month he was a dispatcher and did not drive the vehicle in question days a week did the witness work in the winter time? two days a week three days a week four days a week five days a week six days a week Who was responsible for unloading the cylinders when they were to be placed on the ground? c. d. e. When they were to be placed on the roof? the driver (on the ground)/the rigger (on the roof) the witness was a dispatcher and did not know whose responsibility it was to unload the cylinders the driver (on the ground)/the oiler (on the roof) the lead man (on the ground)/the oiler (on the roof) the lead man (on the ground)/the rigger (on the roof) the propane tanks used for? fuel to operate equipment temporary heat for the workers temporary heat for the building temporary heat for the foreman's trailer the propane had many uses At what address did the mishap occur? 28th and Division 28th and Lexington 29th and Lexington 40th and Division the witness did not recall the address 18. 19. 20. 21. 22. 85 What type of activity was being conducted at the location where the mishap occurred? construction of a high—rise office building surveying of a building site construction of a single—story office building excavation of a land-fill site construction of a high—rise apartment building On how many previous occasions had the witness made deliveries to this site? a. b. c. d. e. How many a. b. c. d. e. What was two or three times four or five times nine or ten times seven or eight times the witness was a dispatcher and made no deliveries to the site cylinders of the product make up one vehicle load? at least 10 per delivery at least 20 per delivery at least 30 per delivery at least 40 per delivery at least 50 per delivery the nature of the cable that the witness talked about? the witness said it was a choker which was a plastic device that was attached to the sleeve at the bottom of the cylinder the witness said it was a choker which was a wire cable device that was attached to the collar on top of the cylinders the witness did not talk about any cable the witness said it was a choker which was a wire cable device that was attached to the collar on the bottom of the cylinders the witness said it was a choker which was a wire cable device that attaches to the sleeve on the bottom of the cylinders cylinders were unloaded at once using the above mentioned 01-pr none--the cylinders were unloaded by hand 23. 24. 25. 26. 27. 28. 86 How many cylinders had been unloaded before the mishap occurred? is the 12 to 16 14 to 18 16 to 2O 17 to 21 19 to 23 bed of the vehicle from the ground? 1 1/2 2 1/2 4 1/2 feet 5 1/2 feet the vehicle in question did not have a bed on it feet feet Initially, how many laborers were helping the witness load and unload the tanks? When the one two three four no one was helping the witness witness reached the site of the mishap, what was the name and position of the person who told him where to locate his vehicle so it could be unloaded? Red: a labor foreman Red: a job superintendent Jack: an oiler Jack: a forklift operator Jack: a dispatcher Where on the vehicle are the levers that control the tail-gate located? back of the vehicle a pushbutton located inside they are on both sides of the there are no levers; there is the cab of the vehicle they are on the driver's side they are on the driver's side they are on both sides on the on the back of the vehicle on the front of the vehicle front of the vehicle Which lever was used to raise the tail-gate on the day of the mishap? the lever on the passenger's side on the front of the vehicle the lever on the driver's side on the back of the vehicle the lever on the driver's side on the front of the vehicle no lever was used; rather, a pushbutton was used inside the cab of the vehicle the lever on the passenger's side on the back of the vehicle 2g. 30. 31. 32. 33. 34. 87 How long did the witness remain on the vehicle while assisting with the unloading of the first set of cylinders? 5 to 10 minutes 11 to 15 minutes 16 to 20 minutes 21 to 35 minutes the witness was a dispatcher and did not assist with the unloading of any cylinders When the witness stepped off of the vehicle how was the tail-gate positioned? Where was a. b. c. d. e. a few inches from the ground the witness did not get off or on the vehicle midway between the ground and the bed of the vehicle folded away even with the bed of the vehicle the tail-gate positioned after it was changed? even with the bed of the vehicle 3/4 of the way between the bed of the vehicle and the ground half-way between the ground and the bed of the vehicle a few inches from the ground it was folded away the witness when the accident occurred? on the vehicle in the labor foreman's office in his own office on the ground beside the vehicle at the fire barrel How far was the fire barrel from the vehicle in question? a. c. d. e. 3 to 4 feet 6 to 7 feet 8 to 9 feet 10 to 11 feet 12 to 13 feet How far was it from the full cylinders to the end of the bed of the vehicle? a. b. c. d. e. 1 foot 4 feet 10 feet 13 feet 16 feet 35. 36. 37. 38. 39. 40. 88 Ho re the cylinders moved around on the bed of the vehicle? a. they were rolled b. they were moved by the forklift c. they were moved by the crane d. they were pulled e. they were lifted Where did the worker who was involved in the mishap fall? a onto the tail-gate b. onto the bed of the vehicle c. onto the ground d on the floor of the labor foreman's office e he did not fall--he suffered a severe cut When the worker fell off of the bed of the vehicle, how many cylin- a none b one C. two d three e the worker did not fall off of the truck How many persons were with the witness at the time of the mishap? a. none b one o. two d. five e. six. a. the tail-gate is left folded underneath the truck b. the tail-gate is placed down on the ground 0. the tail—gate is placed halfway between the ground and the bed of the truck ' d. the tail-gate is placed even with the bed of the truck e. there is no standard procedure that is followed How many vehicles like the one involved in the mishap does the company that employs the witness operate at the present time? a. none b. one c. ten d. five e. fifteen 41. 42. 1+3. 44. 45. 46. the weather like 89 on the day of the mishap? it was cold and it was cold it was cold the witness day it was warm calm; it was snowing and windy; it had snowed the previous night and windy; it was snowing did not recall what the weather was like that and calm; it looked like it was going to snow the witness do after the worker had the mishap? neither helped nor talked to him the witness did not help him but did talk to him no mishap occurred the witness helped him but did not talk to him the witness was not aware of any mishap occurring because he was in his office Who finished unloading the vehicles after the mishap occurred? the oiler two crane operators no mishap occurred a crane operator the forklift operator What is the name and position of the witness' supervisor? a. b. c. d. e. Steve Montague: Steve Nugent: Art Richardson: Steve Richardson: Mike Richardson: a comptroller a comptroller a dispatcher a dispatcher a dispatcher 0n the day of the mishap, how many loads of cylinders had the witness delivered prior to the mishap occurring? a. b. c. d. e. What were none one two three four the witness' normal working hours? he started between 6:00 and 8:00 AM and finished between 4:30 and 5:00 PM he started between 5:00 and 7:00 AM and finished between 3:30 and 4:00 PM he started between 3:00 and 5:00 AM and finished between 1:30 and 2:00 PM he started between 8:00 and 10:00 AM and finished between 6:30 and 7:00 PM he started between 10:00 AM and 12:00 noon and finished between 8:30 and 9:00 PM 90 Now we would like to ask you some questions about your opinion of this witness. Please place a check mark (J) in the space beside the answer which best describes your opinion of this witness. For example, if you strongly agree with the statement: "This witness is a likeable person" then you should place your check in the space beside "Strongly Agree," for example: Strongly agree Agree Undecided Disagree Strongly disagree If you are undecided about a statement, if you have pg_opinion about a statement, or if your opinion about a statement is neutral; then place your check mark in the space beside "undecided." 47. I respect this witness' opinion on the topic. Strongly agree Agree Undecided Disagree Strongly disagree 48. This witness is not of very high intelligence. Strongly agree Agree Undecided Disagree Strongly disagree 49. This witness is a reliable source of information on the topic. Strongly disagree Disagree Undecided Agree Strongly agree 50. I have confidence in this witness. Strongly agree Agree Undecided Disagree Strongly disagree 51. 52. 53. 54. 55. 56. 91 This witness lacks information on the subject. Strongly disagree Disagree Undecided Agree Strongly agree This witness has high status in our society. Strongly agree Agree Undecided Disagree Strongly disagree I would consider this witness to be an expert on the tOpic. Strongly disagree Disagree Undecided Agree Strongly agree This witness' opinion on the topic is of little value. Strongly agree Agree Undecided Disagree Strongly disagree I believe that this witness is quite intelligent. Strongly agree Agree Undecided Disagree Strongly disagree This witness is an unreliable source of information on the topic. Strongly disagree Disagree Undecided Agree Strongly agree 57. 58. 59. 60. 61. 62. 92 I have little confidence in this witness. Strongly agree Agree Undecided Disagree Strongly disagree This witness is well informed on this subject. Strongly agree Agree Undecided Disagree Strongly disagree This witness has low status in our society. Strongly agree Agree Undecided Disagree Strongly disagree I would not consider this witness to be an expert on this topic. Strongly agree Agree Undecided Disagree Strongly disagree This witness is an authority on the topic. Strongly agree Agree Undecided Disagree Strongly disagree This witness has had very little experience with this subject. Strongly disagree Disagree Undecided Agree Strongly agree 63. 64. 65. 66. 67. 68. 93 This witness has considerable knowledge of the factors involved with this subject. Strongly agree Agree Undecided Disagree Strongly disagree Few people are as qualified to testify on this tOpic as this witness. Strongly agree Agree Undecided Disagree Strongly disagree This witness is not an authority on this subject. Strongly disagree Disagree Undecided Agree Strongly agree This witness has very little knowledge of the factors involved with this subject. Strongly agree Agree Undecided Disagree Strongly disagree This witness has had substantial experience with this subject. Strongly agree Agree Undecided Disagree Strongly disagree Many people are much more qualified to speak on this tOpic than this witness. Strongly agree Agree Undecided Disagree Strongly disagree 94 69. I deplore this witness' background. Strongly disagree Disagree Undecided Agree Strongly agree 70. This witness is basically honest. Strongly disagree Disagree Undecided Agree Strongly agree 71. I would consider it desirable to be like this witness. Strongly agree Agree Undecided Disagree Strongly disagree 72. This witness is not an honorable person. Strongly agree Agree Undecided Disagree Strongly disagree 73. This witness is not concerned with my well-being. Strongly agree Agree Undecided Disagree Strongly disagree 74. This witness is a reputable person. Strongly disagree Disagree Undecided Agree Strongly agree 75. 76. 77. 78. 79. 80. 95 I trust this witness to tell the truth about the topic. Strongly agree Agree Undecided Disagree Strongly disagree Strongly agree Agree Undecided Disagree Strongly disagree This witness is a scoundrel. I would prefer to have nothing at all to do with this witness. Strongly agree Agree Undecided Disagree Strongly disagree Under most circumstances I would be likely to believe what this witness says about the tOpic. Strongly agree Agree Undecided Disagree Strongly disagree I admire this witness' background. Strongly agree Agree Undecided Disagree Strongly disagree This witness is basically dishonest. Strongly agree Agree Undecided Disagree Strongly disagree 81. 82. 83. 84. 85. 86. 96 The reputation of this witness is low. Strongly agree Agree Undecided Disagree Strongly disagree I believe that this witness is concerned with my well—being. Strongly agree Agree Undecided Disagree Strongly disagree This witness is an honorable person. Strongly disagree Disagree Undecided Agree Strongly agree I would not prefer to be like this witness. Strongly agree Agree Undecided Disagree Strongly disagree I do not trust this witness to tell the truth on this topic. Strongly agree Agree Undecided Disagree Strongly disagree Under most circumstances I would not be likely to believe what this witness says about the tOpic. Strongly disagree Disagree Undecided Agree Strongly agree 97 87. I would like to have this witness as a personal friend. Strongly agree Agree Undecided Disagree Strongly disagree 88. The character of this witness is good. Strongly disagree Disagree Undecided Agree Strongly agree Now we would like to ask you a few more questions about your Opinion of this witness. Please place a check mark (V) in the space beside the answer which best describes your opinion of this witness. For example, if you thought that this witness was somewhat strong, then you should place your If you are statement, your check 89. check in the space beside "somewhat strong," for example: This witness was. . . Very strong Strong Somewhat strong Undecided Somewhat weak Weak Very weak undecided about a statement, if you have no opinion about a or if your opinion about a statement is neutral; then place mark in the space beside "undecided. " This Witness Was. . . . very safe 90. very just safe just somewhat safe somewhat just undecided undecided somewhat unsafe somewhat unjust unsafe unjust very unsafe very unjust 91. 93. 95. 97. 99. 101. This Witness Was. very kind kind somewhat kind undecided somewhat cruel cruel very cruel very dishonest dishonest somewhat dishonest undecided somewhat honest honest very honest very experienced experienced somewhat experienced undecided somewhat inexperienced inexperienced very inexperienced very unqualified unqualified somewhat unqualified undecided somewhat qualified qualified very qualified very aggressive aggressive somewhat aggressive undecided somewhat meek meek very meek very bold bold somewhat bold undecided somewhat timid timid very timid 98 . (Continued) 92. very friendly friendly somewhat friendly undecided somewhat unfriendly unfriendly very unfriendly 94. very untrained untrained somewhat untrained undecided somewhat trained trained very trained 96. very skilled skilled somewhat skilled undecided somewhat unskilled unskilled very unskilled 98. very informed informed somewhat informed undecided somewhat uninformed uninformed very uninformed 100. very emphatic emphatic somewhat emphatic undecided somewhat hesitant hesitant very hesitant 102. very active active somewhat active undecided somewhat passive passive very passive 103. 105. 107 O 109. 111. 113. 99 This Witness Was. very energetic energetic somewhat energetic undecided somewhat tired tired very tired very tense tense somewhat tense undecided somewhat relaxed relaxed very relaxed very assertive assertive somewhat assertive undecided somewhat nonassertive nonassertive very nonassertive very calm calm somewhat calm undecided somewhat anxious anxious very anxious very uncomfortable uncomfortable somewhat uncomfortable undecided somewhat comfortable comfortable very comfortable very reserved reserved somewhat reserved undecided somewhat outgoing outgoing very outgoing . . (Continued) 104. 106. 108. 110. 112. very confident confident somewhat confident undecided somewhat unconfident unconfident very unconfident very attentive attentive somewhat attentive undecided somewhat inattentive inattentive very inattentive very nervous nervous somewhat nervous undecided somewhat poised poised very poised very truthful truthful somewhat truthful undecided somewhat untruthful untruthful very untruthful very hesitant hesitant somewhat hesitant undecided somewhat unhesitant unhesitant very unhesitant 100 FOR THE FOLLOWING QUESTIONS, PLEASE PLACE A CHECK MARK (I) IN THE SPACE BESIDE THE ANSWER THAT BEST DESCRIBES YOUR OPINION. 114. 115. 116. 117. While watching this witness, I was: very interested interested somewhat interested undecided somewhat uninterested uninterested very uninterested While watching this witness, my mind wandered: all of the time most of the time quite often some of the time occasionally rarely never I found the testimony presented by the witness: very easy to follow easy to fellow somewhat easy to follow undecided somewhat difficult to follow difficult to follow very difficult to follow What did you like most about this trial? 101 118. What did you like least about this trial? Now we would like to know a few things about you. . . 119. Sex: Female Male 120. Age: 121. What is your year in school? Fr. Soph. Jr. Sr. 122. Marital status: Single Married 123. What do you consider your ethnic affiliation to be? 124. Have you any defects in your hearing? YES NO If yes, please explain: 125. Have you any defects in your vision? YES NO If yes, please explain: 126. Have you ever served as a juror? Yes No 127. Have you ever been a party to any suit, either civil or criminal? Yes No If yes, please explain what the nature of each suit was and how you were involved: THANK YOU VERY MUCH FOR YOUR COOPERATION LIST OF REFERENCES LIST OF REFERENCES Anderson, N. 8 Hubert, S. Effect of concomitant verbal recall on order effects in personality impression fermation. Journal of Verbal Learning and Verbal Behavior, 1963, 2) 379-391. Asch, S. Forming impressions of personality. Journal of Abnormal and Social Psychology, 1946, 41, 258-290. Auer, J. The rhetoric of our times. New York: Appleton-Century—Crofts, 1969. Baron, R., Baron, P. 8 Miller, N. The relation between distraction and persuasion. P§ychological Bulletin, 1973, 89, 310-323. Berlo, D. 8 Lemert, J. A factor analytic study of the dimensions of source credibility. Paper presented at the annual Convention of the S.A.A., New York, 1961. Berlo, D., Lemert, J. 8 Mertz, R. Dimensions for evaluating the accept- ability of message sources. Public Opinion Quarterly, 1969-70, 39, 563-576. Bermant, G., McGuire, M. 8 Chappell, D. Videotaped trial presentations: Advantages and disadvantages. Battelle Memorial Institute, 1975. (Stencil) Bretz, R. Techniques of televisionpproduction. New York: McGraw—Hill, 1962. Bruner, J. Going beyond the information given. In J. Bruner, E. Brunswik, L. Festinger, F. Heider, K. Muenzinger, C. Osgood, 8 D. Rapaport (Eds. ) Contemporary approaches to cognition. Cambridge, Mass.: Harvard University Press, 1957. Bruner, J. 8 Taguiri, R. The perception of people. In G. Lindzey (Ed.) Handbook of social psychology. Vol. 2. Reading, Mass.: Addison- Wesley, 1954, 634-654. Cline, V. Interpersonal perception. In B. A. Maher (Ed.) Progress in experimental research. Vol. 1. New York: Academic Press, 1964, 221-284. Cohen, J. Statistical power analysis for the behavioral sciences. New York: Academic Press, 1969. 102 103 Cook, M. Anxiety, speech disturbances, and speech rate. British Journal of Social and Clinical Psychology, 1969, g) 13-21. Cooper, L. The rhetoric of Aristotle. New York: Appleton-Century- Crofts, 1932. Davis, D. The gpsmmar of television production. London: Barrie, 1960. Dibner, A. Cue-counting: A measure of anxiety in interviews. Journal of Consulting Psychology, 1956, 293 475-478. Dittman, A. T., 8 Llewellyn, L. G. Relationship between vocalizations and head nods as listener responses. Journal of Personalityyand Social Psychology, 1968, 2, 79-84. Doret, D. Trial by videotape--Can justice be seen to be done? Temple Law Quarterly, 1974, 41) 228-268. Eisenstein, A. The film sense. New York: Harcourt, Brace 8 World, 1960. Fulton, A. Motion pictures. Norman: University of Oklahoma Press, 1960. Gardner, D. The effect of divided attention on attitude change induced by a marketing communication. In R. M. Haas (Ed.) Science, tech- nology, and marketing. Chicago: American Marketing Association, 1966, 532-540. Goffman, E. Stigma: Notes on the management of spoiled identity. Englewood Cliffs, N. J.: Prentice-Hall, 1963. Haaland, G. 8 Vankatesan, M. Resistance to persuasive communications: An examination of the distraction hypothesis. Journal of Person- ality and Social Psychology, 1968, 2, 167-170. Halbert, R. The effects of image size on jurors' perceptions of a videotaped deposition: A medium shot in the dark. Unpublished master's thesis, Department of Communication, Michigan State University, 1978. Harrison, R. P. Bsyond words: An introduction to nonverbal communica- ‘tion. Englewood Cliffs, N. J.: Prentice-Hall, 1974. Hastorf, A., Schneider, D. 8 Polefka, J. Person percsption. Reading, Mass.: Addison-Wesley, 1970. Jones, E. 8 Davis, K. From acts to dispositions: The attribution process in person perception. In L. Berkowitz (Ed.), Advances ip_experimental social psychology. Vol. 2. New York: Academic Press, 1965, 220—266. Jones, E. 8 Gerard, H. Foundations of socialppsychology. New York: Wiley, 1967. 104 Kaminski, E. The effect of differing editing techniques on juror re- sponses. Unpublished master's thesis. Department of Communication, Michigan State University, 1977. Kasl, S. 8 Mahl, G. The relationship of disturbances and hesitations in spontaneous speech to anxiety. Journal of Personality and Social Psychology, 1965, A, 425-433. Kelley, H. The warm-cold variable in first impressions of persons. Journal of Personality, 1950, 18, 431—439. Kelley, H. Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation, 1967, Vol. 15. Lincoln, Nebraska: University of Nebraska Press, 1967, 192-240. Keppell, G. Design and analysis: A researcher's handbook. Englewood Cliffs, N. J.: Prentice-Hall, 1973. Krause, M. 8 Pilisuk, M. Anxiety in verbal behavior: A validation study. Journal of Consulting Psychology, 1961, 2E, 414-419. Lemert, J. Dimensions of source credibility. Unpublished manuscript, Department of Communication, Michigan State University, 1963. Lewis, C. The TV director/interpreter. New York: Hastings House, 1968. Luchins, A. Primacy-recency in impression formation. In C. I. Hovland (Ed.) The order of presentation ingpersuasion. New Haven, Conn.: Yale University Press, 1957, 33-61. Madsen, R. The impact of film. New York: McMillan, 1973. McCain, T. 8 Repensky, G. The effect of camera shot on interpersonal attraction for comedy performers. Paper presented to the 58th annual Convention of the Speech Communication Association, Chicago, 1972. McCain, T. 8 Divers, L. The effect of body type and camera shot on interpersonal attraction and source credibility. Paper presented to the 59th annual Convention of the Speech Communication Associa- tion, New York, 1973. McCain, T. 8 Wakshlag, J. The effect of camera angle and image size on source credibility and interpersonal attraction. Paper presented to the International Communication Association Convention, New Orleans, 1974. McCain, T., Chilberg, J. 8 Wakshlag, J. The effect of camera angle on source credibility and attraction. Journal of Broadcastipg, 1977, 21, 35-1‘6 0 McCroskey, J. Scales for the measurement of ethos. Speech Monogpsphs, 1966, 82: 65-72. 105 McCroskey, J. An introduction to rhetorical communication. Englewood Cliffs, N. J.: Prentice-Hall, 1972. McCroskey, J. 8 Mehrley, R. S. The effects of disorganization and non- fluency on attitude change and source credibility. Speech Monographs, 1969, 33, 13-21. Mehrabian, A. Methods and designs: Some referents and measures of non- verbal behavior. Behavioral Research Methods and Instrumentation, 1969, _1_, 203-207. Mehrabian, A. Silent messages. Belmont, Cal.: Wadsworth, 1971. Miller, G. R., 8 Fontes, N. The effects of videotaped court materials on juror responses. Preliminary Report. Unpublished manuscript, Department of Communication, Michigan State University. NSF-RANN #APR75-15815, 1977. Miller, G. R. 8 Hewgill, M. A. The effect of variations in nonfluency on audience ratings of source credibility. The Quarterly Journal of Speech, 196M, 593 36-uu. Miller, G. R. 8 Siebert, F. Effects of videotaped testimony on informa- tion processing and decision-making in jury trials. Final Report. Unpublished monograph, Department of Communication, Michigan State University. NSF-RANN #GI-38398, 1975. Miller, G. R. 8 Steinberg, M. Between people: A new analysis of inter- personal communication. Chicago: S.R.A., 1975. Miller, G. R., Bender, D., Florence, B., 8 Nicholson, H. Real versus reel: What's the verdict? Journal of Communication, 197a, 253 99-111. Miller, G. R., Fontes, N., Boster, F., 8 Sunnafrank, M. Methodological issues in jury research: What can simulations tell us? Paper pre- sented at the Annual Convention of the American Psychological Asso- ciation, San Francisco, 1977. Miller, G. R., Bender, D., Boster, F., Florence, B., Fontes, N., Hocking, J., 8 Nicholson, H. The effects of videotaped testimony in jury trials. Brigham YoungUniversity Law Review, 1975, 123 331-373. Millerson, G. The technique of television production. New York: Hastings House, 196a. Osgood, C., Suci, G. 8 Tannenbaum, P. The measurement of meaning. Urbana: University of Illinois Press, 1957. Pope, B. 8 Siegman, A. The effect of therapist verbal activity level and specificity on patient productivity and speech disturbance in the initial interview. Journal of Consulting Psychology, 1962, 26, #89. 106 Quick, J. 8 LaBau, T. Handbook of film production. New York: McMillan, 1972. Scott, J. Film: The medium and the maker. New York: Holt, Rinehart 8 Winston, 1975. Sereno, K. 8 Hawkins, G. The effects of variations in speakers' non- fluency upon audience ratings of attitude toward the speech topic and speakers' credibility. Speech Monggraphs, 1967, 323 58-64. Shaver, K. An introduction to attribution processes. Cambridge, Mass.: Winthrop, 1975. Shaver, K. Principles of socialppsychology. Cambridge, Mass.: Winthrop, 1977. Sheldon, W. Atlas of man: Aguide for somatypipg the adult male at all ages. New York: Harper and Row, 1954. Silverman, I. 8 Regula, C. Evaluation apprehension, demand characteris— tics, and the effects of distraction on persuasibility. Journal of Social Psychology, 1968, 253 273-281. Simon, R.,8 Mahan, L. Quantifying burdens of proof: A view from the bench, the jury, and the classroom. Law and Society Review, 1971, _5_, 319-330. Sunnafrank, M., Miller, G. R., 8 Fontes, N. The "whole truth" or part of it? A procedural note on trial simulations in juror research. Unpublished manuscript, Department of Communication, Michigan State University, 1978. Tiemens, R. Some relationships of camera angle to communicator credi- bility. Journal of Broadcasting, 1970, $33 n83-u90. Vohs, J. An empirical approach to the concept of attention. Speech Monographs, 196”, 81) 355-360. Williams, R. 0n the value of varying TV shots. Journal of Broadcasting, 1965, 9) 33-u3. Williams, R. Film shots and expressed interest levels. Speech Monographs, 1968, 35, 166-169. Winer, B. Statistical principles in experimental design. New York: McGraw-Hill, 1971. Winthrop, H. Effect of personal qualities on one—way communication. Psychological Reports II, 1956, 323-32u. Wurtzel, A. 8 Dominick, J. Evaluation of television drama: Interaction of acting styles and shot selection. Journal of Broadcasting, 1971-72, 16, 103-110. 107 Zettl, H. Television production handbook. Belmont, Cal.: Wadsworth, 1966. Zimbardo, P., Mahl, G. 8 Barnard, J. The measurement of speech disturb- ance in anxious children. Journal of Speech and Hearipngisorders, 1963, _2_3_, 362-370.