Observers' perceptions of an interpersonal dispute as a function of mode of presentation.

Communication researchers have employed a number of modes for presenting stimuli to subjects including written accounts, audio and videotapes, and live performances. The use of these different modes could be potentially problematic if the mode of presentation influences variables measured in

the research. This study investigated whether the modality employed to present stimuli to subjects (written transcript vs. videotape) influences perceptions of a conflict situation. The results revealed that whether observers learned about the dispute by transcript or viewing it on videotape had no effect on perceptions of argumentative or verbally aggressive conduct, the estimation of verbal aggression, or ease of assessing the dispute. Attitudes toward dispute behaviors and dispute resolution were influenced by presentation mode, however. These findings have implications for the role of mode of presentation in communication research.

**********

* A great deal of communication research in recent years has involved subjects assessing various communication situations. The mode for presenting the situations has involved written accounts, video recordings, photographs, audio recordings, participants observing live performances, participants enacting situations which they later assess, and participants being asked to recall particular situations from their past experiences. Written accounts, video recordings, participation ratings, and participation recall have been used much more than others. This could become problematic, if mode of presentation was found to influence some of the variables typically measured in communication research.

This presents a need for research on the mode of communication situation presentation. Some research has examined this issue. Wilke (1934) and Carver (1935) found vocal messages more persuasive than written messages. However, later studies produced mixed results for persuasion (e.g., McGinnies, 1965; Tannenbaugh & Kerrick, 1954; Worchel, Andeoli & Eason, 1975). Chaiken and Eagly (1976) found easily comprehended messages were more persuasive when presented on videotape. In a study by Wilson (1974), when broadcast media were compared to print media, there was greater recall for print (or written) media than for broadcast media. In a later study, Chaiken (1980) found that when participants were involved (interested) in the issue/topic being presented, the information was processed differently than when they were not as involved in the issue. Specifically, when participants were involved in the issue, information was processed systematically rather than heuristically. In systematic processing, recipients form opinion judgments by focusing primarily on persuasive argumentation; that is, the message content. In heuristic processing, recipients focus on, among other things, the identity or image of the communicator. Chaiken found that modality played a part in how messages were processed, that is, participants processed heuristically when exposed to videotaped as opposed to written messages.

LaPlante (1971) discovered media that allowed for nonverbal behavior to be observed were associated with more favorable evaluations. Tusing and Dillard (2000) found paralinguistics (i.e., volume and rate) influenced perceived dominance. In their study, paralinguistic or vocal cues were presented via videotapes of actors delivering short persuasive messages. Similarly, Mehrabian and Ferris (1967) and Argyle and Dean (1965) suggest that affiliative motivation and psychological closeness increases with the number of favorable nonverbal messages. In the Argyle and Dean study, nonverbal messages were presented by creating dyads composed of a participant and a confederate and measuring eye contact at various distances. In the Mehrabian and Ferris study, attitudes toward vocal communication were assessed as speakers communicated "like," "neutrality," and "dislike" through tone of voice.

The present study was primarily prompted by a recent study by Infante, Rancer, and Jordan (1996) who tested the hypothesis that observers would be less likely to perceive verbally aggressive behavior, and more likely to perceive argumentative behavior when they read transcripts of a conversation marked by conflict presented in an affirming versus nonaffirming communicator style. Although the hypothesis was supported, an anomaly in the results was that participants "overestimated greatly the occurrence of verbal aggression in the conversation" (p. 327). The results of this study, however, were not as conclusive as they could be for a major methodological reason. As in most communication research, only one mode for presenting the communication situation to participants was used. The Infante et al. (1996) study used scripts as the method of presenting the interpersonal dispute which contained affirming or nonaffirming messages. One explanation for the overwhelming tendency for participants to overestimate the number of verbally aggressive messages was that participants may not have received enough information through the transcripts to make accurate assessments. For instance, seeing the nonaffirming nonverbal behaviors on a videotape may have provided a basis for deciding that a particular verbal message was not as threatening to the receiver's self-concept as it might appear in a transcript of the interaction. If this is the case, there would appear to be a need for research comparing specific methods of observation. Although the nonverbal messages were described in the transcripts, actual presentation using both visual and vocal cues as in a video, might affect one's perception of the message. Indeed, there is evidence that in some cases nonverbal messages are relied upon more than verbal messages for social meaning (Andersen, 1999; Mehrabian & Wiener, 1967; Philpot, 1983). In the case of contradictory messages (e.g., when the verbal conveys one thing and the nonverbal message another), adults tend to assume that the nonverbal message is more accurate (Burgoon, 1985).

The purpose of the present study was to determine if mode of presentation alters the perceptions of communication in a conflict situation. Specifically, this work utilized the conflict situation studied by Infante et al. (1996). The use of transcripts was only one of several possible methods for presenting a dispute to observers for evaluation. Therefore, research may be needed on other methods of depicting or presenting communication behavior. Perhaps hearing and seeing the messages instead of only reading about them in a transcript provides observers with a basis for more accurate estimation of the amount of verbally aggressive messages delivered in a conflict situation. While arguments could be provided with regard to more information being available in certain modes (i.e., those that contain both verbal and nonverbal information for instance), a lack of research on this specific topic lead to the following research question:

RQ: Will video and transcript presentations of affirming and nonaffirming styles in an interpersonal dispute differ in terms of:

(a) the estimation of verbally aggressive messages?

(b) the perception of argumentativeness and verbal aggressiveness of the participants?

(c) the difficulty to assess the communication that took place in the conversation?

(d) the favorability ratings of the dispute behaviors as seen by observers?

(e) the observers' attitudes toward how the dispute is resolved?

Although the research question was posed mainly for practical reasons, to determine how modality functions in studying interpersonal conflict, some theoretical implications, while not the primary focus and motivation for the study, were recognized. If the research question is answered favoring the video condition with more perceptual accuracy and ease, and more favorable perceptions, this might suggest support for a "maximal sensory input" position in studying conflict. A multichannel message mode may provide the most accurate and useful information for observers of a conflict situation. Less than optimal information may represent a handicap from an observer's perspective. Perhaps perceptual accuracy and favorableness depend heavily upon an abundance of information from a variety of external inputs (i.e., verbal and nonverbal information).

On the other hand, if the data for the research question should favor the transcript method of presenting interpersonal conflict to research participants for assessment, this might suggest a "potency of language" interpretation. A verbal only presentation may serve to activate perceivers' imagination for enhancing the understanding process. Allowing the observers to construct some of the details of the situation in their minds may enhance perceptions more than by adding another channel of input for the presentation. More of a minimal channel message mode may allow language to function in a manner that is sometimes observed when, for example, a person concludes, "the movie was not as good as the book." Scholars of English Literature have long recognized this potential in verbal language. Providing people with abundant information may impede perceptions and discourage imaginative cognitive activities that stimulate perceptual accuracy and favorableness.

Technological advances in recording and editing have allowed researchers to utilize video more often as a method of presenting stimuli to participants. If mode of presentation provides respondents with more information with which to assess communication behavior (e.g., interpersonal conflict) then this added information might affect perceptions and reactions to that behavior. This speculation may have implications for the internal validity of previous research which utilized written descriptions for the evaluation and interpretation of communication behavior. Differences between transcripts and video presentations would also have implications for future research, especially if one mode made it more likely or less likely to detect a particular type of behavior.

Perceptions of argumentative and verbally aggressive conduct were included because according to the research cited earlier these appear to be important in interpersonal conflict. The other three factors were included in order to obtain possible insights into having subjects in a study use transcripts versus video recordings in making judgments about interpersonal conflict. The first concerned assessment ease. When subjects were asked to serve as observers of an interpersonal dispute and to assess the behavior of participants, did they find one presentation mode more difficult than the other in making those assessments? If perception of difficulty is related to perceptions of the dispute, this might be a consideration in future research. For instance, asking subjects to do the more difficult task might be justified if it results in fewer errors of perception. The second consideration was how favorably subjects perceived the behaviors in the dispute. If transcripts or videos inhibited or magnified how constructive, friendly, pleasant, and aggressive the conflict is perceived, then presentation mode should be considered when asking people to judge a dispute. For this same reason, attitude toward how the dispute was resolved was examined. Does a solution adopted by disputants seem to observers as good when reading about it as compared to seeing and hearing the disputants adopt it?

METHOD

Participants

The participants in this study were 200 students--71 males and 129 females enrolled in undergraduate communication courses at a large southern state university. The participants ranged in age from 18 to 50 with a mean age of 21.38. The students took part in the study in a prearranged place and time. Participation was voluntary, although all students received extra credit for participation as determined by their course instructors.

Procedures

The script of an interpersonal dispute used in this study was adapted from the one created by Infante et al. (1996). In that study, affirming and nonaffirming behaviors were identified empirically in a pilot study and used to create the affirming and nonaffirming scripts. The transcripts focused on an interpersonal dispute. The issue of family visitation in the apartment of two male college roommates was discussed. Roommate conflict was chosen due to its relevancy to most college students. "Specifically, one roommate informs the other that his family will be visiting for the homecoming weekend, then a discussion ensues regarding rules that should be implemented for such visits. The conversation was developed such that the conflict arose, climaxed, then ended in de-escalation/resolution" (Infante et al., 1996, p. 323).

Although not a variable with relevance to the current study, two versions of the transcript were developed, one that contained affirming nonverbal behaviors and one that contained nonaffirming nonverbal behaviors. These behaviors were identified by means of stage directions. All statements in the transcripts pertained to the conflict issue except eleven statements by the first person in the dyad and seven statements by the second person that attacked the other's self-concept. One person used three blame messages, one competence attack, three character attacks, one personality attack, one command, and two global rejections, for a total of eleven verbally aggressive messages. The other person in the dyad used one blame message, one character attack, two personality attacks, two competence attacks, and one command, for a total of seven verbally aggressive messages. The transcripts were four to four and one-half pages long.

Videotapes

Two videotapes were created which followed the two scripts exactly. Two Caucasian male actors were hired for this procedure. Actors for the study were 19 and 20 year old students with acting experience, majoring in Performing Arts. Both actors had semi-professional or professional acting experience. They were each paid and signed an agreement not to discuss the nature of the study with anyone.

Each actor was provided copies of the affirming and nonaffirming script one month before rehearsals began. The importance of identical verbal performance of the affirming and nonaffirming scripts was emphasized and carefully scrutinized during rehearsals directed by the researcher coordinating this portion of the study. The concept of affirming and nonaffirming communicator style was explained to the actors to aid them in delivering the messages appropriately.

Three one-hour videotaped rehearsals produced identical verbal presentation of affirming and nonaffirming behaviors. The actors were also informed of specific set (stage) layout and appropriate attire for taping. During the rehearsals, whenever specific verbal messages and/or nonverbal direction cues were missed, the director stopped the actors to make corrections. The actors viewed the videotapes as a reference for additional rehearsals they conducted on their own time when the researcher/director was not present. Scripts were memorized by the actors. The videotapes for the actual study were shot in the studio of a local television station. The stage was set up as the living room of an apartment in which the two actor/roommates lived. Specific directions regarding the type of set and placement of props were given to the stage manager two months prior to taping. These procedures produced two 3 to 3.5 minute videos for the affirming and nonaffirming conversations respectively which were used in the study.

Once the video conditions were produced from the written script, student participants were solicited from various communication classes. The instructors for classes were informed of the opportunity for their students to participate in a communication conversation study. Interested instructors were given information sheets to distribute to their students. The information sheet indicated that the study involved watching or reading a communication conversation, and then completing a questionnaire based on the conversation they read or viewed. Sessions were scheduled in blocks of 30-minutes.

Participants were assigned randomly to either view the video or read the transcripts. The design of the experiment was a 2 x 2 x 2 factorial with affirming/nonaffirming messages and transcript/video medium as manipulated variables, the participant's sex was an attribute variable. The data reported in this article represents the first part of a larger study which, in addition to the male actors in the present study, will include videotapes of female actors also portraying the affirming and nonaffirming conditions. The larger study currently in process will focus on the impact of gender in conflict situations.

Dependent Measures Questionnaire

The questionnaire contained instructions and general information about the nature of the study. The transcript of the conversation or the videotape followed. Participants were told to read the transcript or watch the videotape carefully so that they would be able to respond to the questionnaire, which followed. Once they had completed reading the transcript it was collected and the questionnaires were distributed. This procedure assured that the participants did not go back and reread the script once they began completing the questionnaire.

Demographics, argumentative and verbally aggressive conduct. The questionnaire consisted of several parts. Participants were asked to indicate demographics (sex, age, and major). Then, six items were used to assess perceptions of "argumentative conduct" (e.g., "To what extent were the participants rational in their communication behavior" (not at all rational-very rational). The measure was the same as the one used by Infante, Myers, and Buerkel (1994) and Infante et al. (1996). The "argumentative conduct" score was obtained by summing the six items. Coefficient alpha for this measure was .73. To assess participants' perception of "verbally aggressive conduct," six items developed by Infante et al. (1994) were used (e.g., "To what extent did the participants attack one another's self-concepts" (not at all-a great extent). Coefficient alpha was .87 for these six items.

Next, participants were asked to estimate the number of verbally aggressive message behaviors they recalled each roommate engaged in during the conversation. The six categories of verbal aggression used in the script were listed: blame, competence attacks, character attacks, personality attacks, command, and global rejection or disconfirmation. A total of 18 verbally aggressive messages of these six types were exchanged in the conversations, thus, the estimation of verbal aggression was computed by subtracting 18 from the total number of verbally aggressive messages identified by the participant (e.g., for character attack, "How many times did each roommate attack each other's character, honesty, trustworthiness, morality?").

Favorability of the dispute behaviors. Participants indicated how favorably they viewed the dispute behaviors by responding to eleven items which were created to tap perceptions of the aggressiveness of dyads (e.g., aggressive-nonaggressive), constructiveness of conversation (e.g., constructive-destructive), friendliness (e.g., friendly-unfriendly) and pleasantness of conflict management (e.g., pleasant-unpleasant), perceptions of how the relationship of the dyad might be affected by the dispute (e.g., "Will the relationship be more or less satisfying to the roommates?" (more satisfying-less satisfying), as well as the participants' own desire to participate in the conversation (e.g., "To what extent would you have wanted to participate in this exchange?" (a great deal-not at all). Responses toward these items constituted the measure of "favorability toward dispute behaviors." Coefficient alpha obtained for this measure was .80.

Solution assessment and assessment ease. Participants also assessed the solution in the dispute. This "attitude toward the solution" measure was obtained by creating four semantic differential-type scales (good-bad, positive-negative, right-wrong, valuable-worthless). Coefficient alpha for this measure of was .90. Assessment ease was measured by having participants respond to three questions developed for this study, using 7-point semantic differential-type scales. The scales included: "How easy was it for you to evaluate the communication behavior of the participants in this conflict?" (easy-difficult), "How confident are you in the accuracy of your perceptions of the participants' communication behaviors?" (confident-not confident), and "How easy was it for you to answer the questions in this survey based on your recall of the conversation?" (easy-difficult). These three questions constituted the measure of "assessment ease." Coefficient alpha for this measure was .78.

RESULTS

The focus of this study was the research question: Will video and transcript presentations of affirming and nonaffirming styles in an interpersonal dispute differ in terms of: the estimation of verbally aggressive messages, how argumentative and verbally aggressive conduct by the participants is perceived by observers, how easy or difficult is it for observers to assess communication, how favorably the dispute is seen by observers, and observers' attitudes toward how the dispute is resolved?

In order to explore this question, a 2 x 2 x 2 multivariate analysis of variance (MANOVA) was computed with three independent variables: affirming or nonaffirming nonverbal behaviors in the dispute, transcripts or videotapes as the channel for presentation of the dispute to observers, and male or female subjects. The latter variable was included to determine whether the results of the study needed to be qualified according to gender since research suggests that males and females differ in terms of aggressive communication trait predispositions (Infante, 1985, Infante, 1989; Infante & Rancer, 1996; Rancer & Dierks-Stewart, 1985; Schultz & Anderson, 1982), beliefs about aggressive communication (Rancer & Baukus, 1987), and how sex-based stereotypes influence perceptions of aggressive communication (Nicotera & Rancer, 1994). The dependent variables were: verbal aggression estimation (the number of times the subject said six different types of verbally aggressive messages were used, minus 18, which was the actual number of times the messages appeared in the dispute), argumentative conduct (sum of six items), verbally aggressive conduct (sum of six items), ease in assessing communication (sum of three items), how favorable dispute behaviors were perceived (sum of 11 items), and attitude toward the resolution of the dispute (sum of four items). The results produced a significant main effect for the transcript-video condition, Wilks' lambda of .89, F(6,187) = 3.86, p < .001, [R.sup.2] = .11. The means for the main effect are presented in Table 1.

Three-way univariate ANOVAs were then computed for each dependent variable to determine which contributed to the MANOVA main effect for modality. The results indicated that attitude toward the solution adopted to resolve the dispute accounted for most of the variability of modality, F(1,192) = 21.58, [eta.sup.2] = .10, p < .001. The solution was evaluated more positively in the transcript versus video condition. Favorability of the dispute behaviors was significant for modality as well, F(1,192) = 4.78, [eta.sup.2] = .02, p < .05. The dispute behaviors were perceived more favorably in the transcript as compared to video condition. The results for the other four dependent variables were not significant. Therefore, whether observers of an interpersonal dispute learned about it by reading or viewing a videotape of the dispute had no effect on the estimation of verbally aggressive messages, on the perception of argumentative or verbally aggressive conduct, or on how easy or difficult it was to assess communication in the dispute situation. Thus, while judgments about communication were not influenced by the mode of presentation, attitudes toward the dispute behaviors and the solution were affected by reading about, versus seeing and hearing, the dispute.

No significant main effect for gender was found in this study, and the participants' gender did not interact with the other two independent variables.

DISCUSSION

Overview and Summary

The results found in this study indicate that although modality did not affect perceptions of argumentative and verbally aggressive conduct, estimation of the number of verbally aggressive messages, or how easy or difficult it was to assess communication in the dispute situation, modality did affect attitudes toward the resolution of the conflict and the favorability of the dispute behaviors. The transcript mode of presentation elicited the more positive attitudes.

The methodological focus of this study is important since a great deal of communication research in the past two decades has involved presenting communication situations to subjects in a variety of ways and asking them to assess those situations. While live presentations, along with videotape and audiotape have been used as presentational modes, written depictions probably have been far more common. This being the case, a very basic issue involves what has been gained and lost by a reliance upon the written presentational mode.

Of course a major advantage of written stimuli is practicality. Often this is the most economical and sometimes the only feasible method of studying communication situations especially those which depict very unusual and perhaps illegal communicative behaviors. The results of the present study, which focused on aggressive forms of communication, provide some justification for electing the more economical and feasible method of conducting a study. At least in the case of aggressive communication, the communicative behaviors of interest were perceived by subjects to about the same degree when a videotape was viewed as when a transcript was read. Thus, in this sense, a nonverbal behavior that was described in writing functioned about the same as conveying that behavior through a videotaped image.

On the other hand, the results suggest that the two modes may differ in a fundamental way that should be considered in light of methodological economy and feasibility. The solution to the conflict in the dispute situation was viewed more positively in the transcript condition, along with how favorable the dispute behaviors of participants were perceived. This would seem to mean the attitudinal dimension of dispute perception was affected. A possibility is, at least in conflict situations, reading about a dispute enhances positive attitudes, or perhaps does not encourage as much negative affect to develop. It may be that more channels of input for a communication situation does not increase the chance of favorability developing. Relying more on the receiver's imagination, as is required by scripts, may be a more likely way for such attitudes to develop. A possible interpretation, therefore, of the vast body of prior communication research which has utilized written depictions of communication is that attitudes reported may have been more favorable than what may have been obtained if a different, perhaps more costly, methodology had been employed. On the other hand, studies which employed methodology using videotapes might have reported less favorable attitudes than what would have been obtained if written depictions of communication had been utilized.

Limitations and Future Research Directions

While this study moves us forward in considering the influence of mode of communication in attitude formation, some questions remain to be explored. First, only one conflict episode was presented to the participants. Will these same findings emerge if the conflict presented was between other adversaries such as parent-child, superior-subordinate, or between spouses? Secondly, only male dyads were used in the written transcripts and on the videotape. Will the reactions to the conflict be seen similarly if the dyads were composed of all females or mixed-sex dyads? Future research might also explore how perceptions of the conflict are affected by the race or religion of the participants in the dispute.

Conclusions

The results of the present study suggest that researchers should consider mode of presentation of communication when investigating observers' perceptions of interpersonal messages. It appears that some attitudes are influenced by how communication is conveyed to observers. The results are generally consistent with previous research which indicates that messages conveyed in certain modes account for differences in the persuasiveness, comprehension, and recall (see for example, LaPlante, 1971; Chaiken & Eagly, 1976). This study goes further by suggesting that modality affects perceptions of an interpersonal conflict, in particular, how positively or negatively the solution to a dispute is evaluated, as well as how favorably the dispute behaviors are viewed.

TABLE 1
Main Effect Means and Standard Deviations
for Transcript-Video Conditions

                                     Transcript
                                      Condition     Video Condition
        Dependent Variable            (n = 100)        (n = 100)

Verbal aggression estimation         7.99 (14.64)     9.08 (16.81)
Argumentative conduct               25.84 (6.21)     24.78 (6.06)
Verbally aggressive conduct         29.39 (7.54)     29.41 (7.59)
Assessment ease                     13.35 (3.87)     13.00 (4.14)
Favorability of dispute behaviors   46.31 (10.25)    42.73 (11.03)
Attitude toward solution            20.83 (4.80)     17.62 (5.04)

Note. Numbers in parentheses represent standard deviations for each
condition.

REFERENCES

Andersen, P. A. (1999). Nonverbal communication: Forms and functions. Mountain View, CA: Mayfield Publishing Co.

Argyle, M., & Dean, J. (1965). Eye-contact, distance and affiliation. Sociometry, 28, 289-304.

Burgoon, J. K. (1985). Nonverbal signals. In M. L. Knapp & G. R. Miller (Eds.), Handbook of interpersonal communication (pp. 344-390). Beverly Hills, CA: Sage.

Burns, K. L., & Beier, E. G. (1973). Significance of vocal and visual channels in the decoding of emotional meaning. Journal of Communication, 23, 118-130.

Carver, M. E. (1935). Listening versus reading. In H. Cantril & G. Allport, The Psychology of Radio. New York: Harper.

Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source message cues in persuasion. Journal of Personality and Social Psychology, 39, 752-766.

Chaiken, S., & Eagly, H. (1976). Communication modality as a determinant of message persuasiveness and message comprehensibility. Journal of Personality and Social Psychology, 34, 605-614.

Infante, D. A. (1985). Inducing women to be more argumentative: Source credibility effects. Journal of Applied Communication Research, 13, 33-44.

Infante, D. A. (1989). Response to high argumentatives: Message and sex differences. Southern Communication Journal, 54, 159-170.

Infante, D. A., Myers, S. A., & Buerkel, R. A. (1994). Argument and verbal aggression in constructive and destructive family and organizational disagreements. Western Journal of Communication, 58, 73-84.

Infante, D. A., & Rancer, A. S. (1996). Argumentativeness and verbal aggressiveness: A review of recent theory and research. In B. Burleson (Ed.), Communication yearbook 19 (pp. 319-351). Thousand Oaks, CA: Sage.

Infante, D. A., Rancer, A. S., &Jordan, F. F. (1996). Affirming and nonaffirming style, dyad sex, and the perception of argumentation and verbal aggression in an interpersonal dispute. Human Communication Research, 22, 315-334.

LaPlante, D. (1971). Communication friendliness, trust, and the prisoner's dilemma. Unpublished master's thesis thesis, University of Windsor, Canada.

McGinnies, E. (1965). A cross-cultural comparison of printed communications versus spoken communications in persuasion. Journal of Psychology, 60, 1-8.

Mehrabian, A., & Ferris, S. R. (1967). Inference of attitudes from nonverbal communication in two channels. Journal of Consulting Psychology, 31, 248-252.

Mehrabian, A., & Weiner, M. (1967). Decoding of inconsistent communications. Journal of Personality and Social Psychology, 6, 109-114.

Nicotera, A. M., & Rancer, A. S. (1994). The influence of sex on self-perceptions and social stereotyping of aggressive communication predispositions. Western Journal of Communication, 58, 283-307.

Philpot, J. S. (1983). The relative contribution of meaning of verbal and nonverbal channels of communication: A meta-analysis. Unpublished master's thesis, University of Nebraska, Lincoln.

Rancer, A. S., & Baukus, R. A. (1987). Discriminating males and females on belief structures about arguing. In L. B. Nadler, M. K. Nadler, & W. R. Todd-Mancillas (Eds.), Advances in gender and communication research (pp. 155-173). Lanham, MD: University Press of America.

Rancer, A. S., & Dierks-Stewart, K. J. (1985). The influence of sex and sex-role orientation on trait argumentativeness. Journal of Personality Assessment, 49, 69-70.

Schultz, B., & Anderson, J. (1982, May). Learning to negotiate: The role of argument. Paper presented at the annual meeting of the Eastern Communication Association, Hartford, CT.

Tannenbaum, P. H., & Kerrick, J. S. (1954). Effect of newscast item leads upon listener interpretation. Journalism Quarterly, 31, 33-37.

Tussing, K. J., & Dillard, J. P. (2000). The sounds of dominance: Vocal precursors of perceived dominance during interpersonal influence. Human Communication Research, 26, 148-171.

Wilke, W. H. (1934). An experimental comparison of the speech, the radio, and the printed page as propaganda devices. Archeological Psychology, 169.

Wilson, E. C. (1974). The effect of medium on loss of information. Journalism Quarterly, 51, 111-115.

Worchel, S., Andreoli, V., & Eason, J. (1975). Is the medium the message? A study of the effects of media, communicator, and message characteristics on attitude change. Journal of Applied Social Psychology, 5, 157-172.

Andrew S. Rancer (Ph.D., Kent State University, 1979) is Professor of Communication at the University of Akron. Felecia F. Jordan-Jackson (Ed.D., West Virginia University, 1989) is Associate Professor of Communication at Florida State University. Dominic A. Infante (Ph.D., Kent State University, 1971) is Professor Emeritus of Communication at Kent State University. An earlier version of this manuscript was presented at the annual meeting of the National Communication Association, November 2000, Seattle, WA.

"The authors wish to thank Florida State University for a grant awarded to the second author. We also wish to thank Erica N. Dawson and Kimberly L. Dawson for help with data collection, and Professor Beth Le Poire for editorial advice with earlier versions of this paper."