Teaching Conflict Resolution: A Study of Two Interactive Methods

Background

Conversation is one of the basic means of oral interaction; therefore, being able to participate actively and appropriately in a conversation is a skill that many language learners would like to and need to acquire. Multidimensional models of communicative competence (e.g. Bachman, 1990;Celce-Murcia et al., 1996;Bachman and Palmer, 1996) have identified oral discourse competence as a distinct component of L2 speakers' communicative language ability. Among others, conversational competence can be characterized by the ability to perform openings, re-openings, closings and pre-closings, to establish and change topics, to hold and yield the floor, to backchannel, to interrupt and to collaborate, as well as to recognize and produce adjacency pairs (Celce-Murcia et al., 1996).

Purpose-oriented oral communication is an important part of most of communication skills courses. These courses aim at developing conversation skills related to a particular area. Since the conversation skills mentioned in the earlier paragraph radically change with reference to dimensions such as audience, communication goal, and context, purpose-oriented communication demands a special set of abilities.

Role-plays and simulations have long been used in the discovery-based approaches to teach purpose-oriented oral communication in a variety of disciplines (Klabbers, 2001). Simulations are now being conducted with the aid of computers. Both role-plays and Educational computer-based games (ECBGs) can effectively create a teaching context that is conducive for constructivist learning. ECBGs also reduce stress on the learners' imagination. Students can actually see a character and can hear a dialogue. In role-play situations the students need to use their imagination to recreate a scene. Karmos (1999) examined how the type of task candidates who have to perform in oral exams affects conversational structure and the candidates' opportunities to display their conversational competence. On the basis of the analysis of 30 non-scripted interviews and 30 guided role-play tasks, it was concluded that role-plays can be better measures of the candidates' abilities of managing conversations, that is, performing openings and closings, initiating and rejecting topics, as well as interrupting. In non-scripted interviews, due to the unequal distribution of power, candidates rarely have the chance to initiate a new topic and have no right to initiate the opening or the closing of the conversation. Consequently, these components of conversational competence cannot be tested in non-scripted interviews. In contrast, the findings suggest that the guided role-plays in the study display several characteristics of conversations: candidates have the right and opportunity to introduce, ratify and reject new topics in the course of the interaction. Depending on the script of the role-play task, they can also initiate the closing and opening of the conversation. These results indicate that guided role-plays, scripted with a focus on conversational skills, have the potential to compel candidates to exhibit their conversational competence. In this project we decided to consider Role Play with the use of an educational computer game because both of these methods broadly fall within the constructivist paradigm. There main obvious similarities and differences in these two methods are that both allow a free play of imagination; and, while role-plays normally need pair or group work, ECBGs can be used by a single player. Both can be used to encourage students to construct hypotheses, make decisions, and discover principles by themselves (Kearsley 1994). The instructor's task is to transform the learning content into a format appropriate to the learner's current state of understanding and organize it in such a manner that the student is able to build upon what they have already learned, and progression is facilitated. This requirement is very much applicable to the design of a Role Play activity and, equally, to the design of an educational game. The issue of the theory of instruction becomes relevant in this case. Bruner (1966( , as cited in Kearsley 1994 states that a theory of instruction should address the following aspects:

1. the most effective sequences in which to present material 2. the ways in which a body of knowledge can be structured so that it can be most readily grasped by the learner

In the light of these two aspects it becomes interesting to analyse the instructional design of educational games. Gaming software ranges from simple simulations to first-person adventure with stunning graphics, compelling storylines and state-of-the-art video. Researchers have, however, questioned the social consequences of gaming as an entertainment activity. It had been argued (Billen, 1993) that games affect cognitive functions, motivation, and may create a divide between the players and the real world.

Others have however found that the gaming industry has potential to motivate users by stimulating curiosity. (Thomas and Macredite, 1994). Quinn (1997) argues that educational games need to combine fun elements in a game with aspects of instructional design.

Your name is Soon Leng. You have been working for just over a year as a copywriter for a medium sized public relations company. Your principal task has been to write text for annual reports, which you hate. Recently you've also been asked to write text for web sites.

Your best experience over the year was writing a humorous speech for a CEO at an annual stockholders' meeting.

You believe you are a great speech writer and would love the opportunity to demonstrate your skills again.

Your worst experience was a press release you wrote that had the client confronted with a libel suit. It wasn't entirely your fault though; you relied on confidential information provided by the client which you assumed was accurate.

Overall you believe your performance has been above average. You also work longer hours than the other two writers, Tony Lau and Hui Fen.

Tony Lau has been with the company three years and gets the plum accounts and makes twice as much as you.

Hui Fen has been with the company only six months but makes more than you because of her previous awards as a newspaper columnist.

You are a bit of a loner and don't socialize with the others in your office. You have never asked your boss for anything and have not made any effort to build any sort of relationship is with him.

Context, research questions, and research method

Overview: In this study, the effectiveness of Role Play was compared against that of an educational computer-based game. These two methods were applied in the conflict resolution (CR) part of a business communication course in a Singaporean university. This is a technological university where most of the students study engineering. All engineering students have to complete three core courses in communication skills. Students have a high level of motivation towards some of the components of theses courses, one of them being conflict resolution. Like many other courses, this communication skill course follows a lecture-tutorial pattern. Strategies of resolving conflicts are discussed in the lecture while interactive exercises and activities are done during tutorials. All tutorials are conducted in a multimedia laboratory where each student has a multimedia PC to work on. Three groups of computer engineering students studying communication skills participated in the study.

The aim of the CR lessons was to teach the students to deal with a difficult person in the context of asking for a raise. All groups were initially given a lecture on CR methods. Then one group (RP group) was asked to work through a Role Play activity; another group (G group) with a computer game and the last group (RPG group) was asked to do both.

The study was designed to answer the following three questions: 1. Which method leads to more effective learning in terms of the achievement of the course objectives and Why? 2. Which method do the students seem to prefer? 3. Which particular aspects of the method seem to work better?

A detailed questionnaire was used as the measuring instrument. (See Appendix 2.) To avoid a halo effect, some questions asked participants to compare the new teaching method with the established lecture method. The RPG group was asked to compare the two methods with each other and with the lecture. In an attempt to make this study more exploratory rather than a straightforward Method-Comparison Study, we asked the participants to rank various features of the method(s) in order of the importance assigned by them. All responses were kept anonymous and confidential.

Participants: In all 63 participants took part in this study. They were divided according to the class they were attending forming three groups as follows: 18 (RP), 12 (G), and 33 (RPG). We achieved a 100% response rate. All participants are first year students of the Computer Engineering degree programme at a technological university. They were in the 17-24 age group. Their mean (self-rated) speaking ability was 2.32 (closer to 'Satisfactory') on a 5 point scale, while their mean (self-rated) ability to understand English was 2.84 (closer to 'Good'). They had minimal experience with Role-Play, and considerable experience with Computer games. ANOVA results show that none of the background information is significantly different (P<0.10) across the three experimental groups. This means that the impact on scores might be largely due to a change in the teaching method, rather than any of the background factors. This suggests that our random allocation of the participants to different teaching methods was successful. Description of RP activity: Students in the tutorial group were subdivided in two major groups. One group was given the role of the boss, while the other was given the role of the employee. The employee was supposed to ask for a raise, while the boss was supposed to act as a difficult person. The two major groups were further subdivided in teams. Each team had two students. One student would actually conduct the discourse (the boss and the employee), while the other student (coach) observed and assisted his/her team mate during the proceedings. This arrangement is summarized below. Both groups were given time to read and understand the handouts with their roles. The 'employees' needed the time to plan the CR strategies they would adopt. The tutor spent most of this time preparing the 'bosses'. This was because their role included keeping certain information from the 'employee' unless the 'employee' asked the right questions. The 'bosses' had to be instructed not to divulge this information too readily. The 'bosses' also needed time to rehearse their roles together with their coaches. (See Appendix 1 for more details.)

When two teams were engaged in the interaction and their proceedings, the other teams observed them, and tried to improve on their communication strategies when their turn came. Brief comments were made immediately after each team's performance. The class had a debriefing where the tutors and students offered their comments on the way the proceedings were conducted.

The Game: Students in the G and RPG groups used a computer-based game that had the same theme and storyline as the Role Play. The game was composed using simple flash animation. Dialogues were provided using both voice-over and text. An introduction to the scenario and characters, as well as instructions was provided at the beginning of the game itself. At the end of the tutorial, the class was debriefed and the tutors provided comments on student performance and their use of CR and communication strategies. With the RPG group, the role play was conducted first. Independent Variables: The experimental treatment in our study is the impact of teaching methods (role play, ECBG, combination of role play and ECBG). Separate descriptions were developed and pre-tested for each method. The lecture was delivered before the tutorials in a lecture theatre using slides and video. The Role Play was specially written for the experiment, and closely paralleled the scenario, the characters, and the storyline in the interactive game. This was intentionally achieved to avoid introduction of an unwanted variable. The scenario was that of an average employee asking a very exacting and difficult employer for a raise.

Dependent Variables:

The participants answered twenty close-ended and one open-ended question after they complete the learning activities. The first nine questions related to the background of the learners in terms of age, sex, year at university, nationality, pre-university education, English proficiency, learning from elective courses, prior experience with the teaching method(s) being trialed. The remaining questions asked the participants to rate the learning in the experimental teaching session, attribute the learning to a method, indicate their preferences for the method, and rank their preferences for the features of their learning method. Most questions were on a seven-point Likert scale with verbal anchors ranging from Very Much or Very High (1) to Not at all or Very Low (7). For the ranking of the methods, participants could rank their preferred method from 1 to 3.

Control Variables:

We controlled other variables that may affect the relationship between learning and the teaching method (dependent variable). We controlled these variables by holding them constant across our experimental conditions. These variables were: learning from the lecture, the tutor, the subject matter, students' previous subject knowledge, and the story line. We made sure that all participants attended the lecture and that all groups had the same lecturer and tutor. We also made sure that the story line of the game and the role play were similar. We used the same subject matter (content) for teaching. All participants were first year engineering students, and by running ANOVA for the 9 questions related to their background we made sure that none of the background information was significantly different across the three experimental groups (P<0.10).

Findings of the experiment

Research question 1-Method Comparison: We tried to measure the impact of each teaching method on five different aspects. They were:

• Knowledge of strategies • Ability to choose appropriate strategy (Selection)

• Ability to ask for raise (negotiate) (Application)

• Ability to deal with difficult people (Broad Application)

• Overall communication skills The participants were asked to rate their learning across these five aspects before the tutorial and after the tutorial. They were asked to rate their learning on a seven-point Likert scale where 1 means 'Very High' and 7 means 'Very Low'. The mean difference between the 'Before the tutorial' score and 'After the tutorial' score indicates measured learning for that particular method. Figure 1 shows the mean scores obtained by each subject group in the five aspects of learning. Since the difference in mean scores seems to be significant only in terms of acquisition of knowledge and ability in strategy selection, we need to closely look at the statistical analysis related to these two features. This part of the analysis is presented in Table 3. Table 2 indicate that in terms of knowledge acquisition, the RP method is significantly better than G. As expected, the other two comparisons (RPG with RP and RPG with G) are not statistically significant. This must be due to the dominance of RP in the RPG group. If G was more dominant than RP (in the RPG group), this dominance would have reflected in the statistical significance. However that does not seem to be the case. The data, thus, seem to support the thesis that the RP method is marginally better than G even though the effect is not statistically significant. Our confidence level in RP as an interactive method is stronger than what it was before the experiment was conducted. However, the support for RP is weak in terms of three learning features (Application, Broad Application, and overall development of communication skills) and the data are inconclusive in these three respects. Our analysis of confidence interval demonstrates that the effect of RP is considerable (See Table 3). Method Preference: We checked the manipulation of our experimental treatment by requesting subjects to state their preference for the teaching method as compared against the lecture method. We asked them to rank the preferred method where 1 indicates that they learnt most from this method and 3 indicates that they learnt the least from this method. The results of this study show that the Role Play method is generally preferred by the students who wish to transfer their skill to real-life situations. As seen from the figures in Table 5, the difference between learners' preference for lecture and for the game is not significant. (P>0.05). This shows that the learners are indifferent when it comes to compare the lecture method with the game. This indifference is interesting to note because common sense expects us to assume that game would be preferred to the lecture method. Another interesting fact is that the students definitely prefer RP method when they compare it with lecture and with the game. The comparisons are statistically significant (P<0.05) in these two cases. Table 5's chi square value of 32.323, with 3 degrees of freedom, clears the related critical value of 15.5, so we can affirm the claim that the participants do significantly differ in their self-reported preference for a certain teaching method.

Investigation into the Teaching Methods: After attempting a comparative analysis, let us try to investigate which features of each of these methods are more significant than the other. Let us first analyse features of the Role Play method. Ratings given by the subjects indicate that among the given five features of the Role Play method, Tutors' comments helped them to learn the most, while group discussion on the Role Play was the least important in this regard (Figure 3). Observing other teams playing the roles was the secondmost important factor, followed closely by enactment of the situation, and then by the description of the Role Play situation.

Descriptors

When it comes to the features of the Interactive Game, ratings given by the subjects indicate that Animation helped them to learn most, followed closely by the use of Humour. Storyline of the game was third-most important, followed by depiction of Characters in the game. Description of the game's scenario at the start of the game was the least important in this context ( Figure 4). A statistical analysis of these results show that the ratings of the descriptors are not significant in any way for RP. For the ratings of the game descriptors, they are only statistically significant when we compare them with those students who said they learned most from the game. In this case, the results of a two-way Anova show that Humour as a descriptor is significant (.035). When we compare the rating of the descriptors with those students who chose the game as the preferred method, the game's storyline is significant (.033) as a descriptor. Discussion

The manipulation of variables for this experiment suggests that since the participants were Computer Engineering students, a game with an interesting story line and adequate learning mechanism would be preferred for the following reasons:

• Opportunity to work at your own pace • Opportunity to repeat the learning exercise as often as required • Opportunity to revisit a desired specific part of the learning exercise • Freedom to access the Learning resources at any time.

• Economies of cost in terms of time and energy Interestingly, despite the above-mentioned advantages of the game method, students seem to prefer the Role Play method. Moreover, it was also seen that in terms of knowledge acquisition, students seem to learn significantly better using the Role Play method. These findings strongly highlight the role of face-to-face-communication in teaching purpose-oriented oral communication such as conflict resolution. However, we believe that this research has implications for both teachers and game developers. Findings related to the features of each method show that skills required for both RP and ECBG are: logic, memory, visualization, and problem solving. Established educational theories (Saljo,1979, Schank andCleary, 1995) argue that development of such skills is fundamental to all learning. While teachers need to focus on using these findings for the selection of an appropriate game, the game developers need to focus on providing a pedagogic base to the gaming activity. Fundamentally, the game developers need to work on the basis of their understanding of the relationship between educational needs and game elements. Such an understanding will help them develop educational games that include visualization and problem solving skills. Games that incorporate a blend of visualization and problem solving can provide sufficient motivation and engagement to the learners. In order to achieve this, the following implications of the present experiment can be made use of.

1.

It is important to build in conversation and interaction in the games. The game used in this study was entirely based on conversation and interaction between characters. Students however still seem to prefer the Role Play method. This preference shows that there is a scope for including greater person-to-person interaction in games.

2.

By their very nature, ECBG follow a programmed learning structure. Each question has a finite set of alternative answers from which learners choose a path and proceed accordingly. This structure is however highly unsuitable for the development of purpose-oriented oral communication. The developers need to explore other options for the development of these games.

3.

Feedback seems to be very important for the learners. Aspects such as animation, dialogue, story line can be significant to hold learners' attention. However, consolidation of learning seems to be the key for the success of a teaching method. Although interactive games do provide feedback in terms of what is right and what is not, and do make suggestions, developers need to improve the quality of the feedback and simulate the existence of a 'live' tutor.

4.

Students seem to benefit from watching how other learner perform and learn. It might be useful if performance of each learner is electronically stored, and made available to the other players. According to Malone (1981), three elements contribute to the fun in games: fantasy, curiosity, and challenge. The element of challenge can be further enhanced by letting the learners compare their performance with the performance of other players. Amory et al. (1999) have developed an interesting model on the premise that educational games contain different aspects, those that promote educational objectives (abstract) and those that allow for realization of these objectives (concrete). According to this model, a game space consists of different components that contain discrete interfaces. These interfaces are either abstract or concrete. The game space embodies all the components (each with their own interfaces) and interfaces that define the interactive learning environment. If the game developers use such a model, it can provide them with a systematic approach for their work. Also, it will allow them to easily identify the game type that is most appropriate to the particular pedagogic purpose.

Conclusion

Computer Engineering students appear to favour Role Play over games, and also seem to benefit more from Role Play in terms of knowledge acquisition. Students identified animation, humour, and storyline of the game as important features. In the Role Play method, tutors' comments helped to learn them the most, while observing other teams playing the roles was the second-most important factor. Development of learning tools based on a combination of the two methods could provide educators with a superior mechanism to enhance learning in virtual environments where knowledge is acquired and skills are effectively transferred to real-life situations. It appears that such development will work at least in the case of computer engineering students.

You believe in giving your employees a raise based solely on the quality of their work, not on how long they have been in your company.

You have no compassion and do not allow sympathy or pity to sway your decisions.

You are prepared to invest in training for your employees.

You do not like pets, you think it is unsanitary to have animals living in your house and you think it is stupid to spend money on them.

SECRET info do not disclose unless asked You will be giving a 'small' party with 300 of your closest friends next week.

You have not prepared a speech for this party and are willing to pay $1500 for a good one.

If Soon Leng asks for training you are willing to pay for it and then consider a raise in six months time if his work improves.

Role: Employee

Study the situation and background information carefully.

Is your gut instinct to avoid conflict?

If you are confronted, do you ignore the aggressor and try to escape? Do you put on a strong offense or do you engage the person in conversation?

What is the winning strategy?

In this drama you will speak to your boss about getting a raise.

Your task

You will rely on logic and your knowledge of human nature in order to successfully negotiate yourself out of a difficult spot.

In this simulation you will encounter a difficult person, your boss, with whom you will have to negotiate to get a raise.

Your challenge will be to navigate the minefield of this person's psyche and emerge with the best possible outcome.

Try to make the choices in your negotiation that you would normally make in this situation and then observe yourself as you play.

Immediate financial problems

Last month, you maxed out your credit card with the purchase of a new car, and your loan payments eat up all your discretionary income. You need a raise simply to maintain your current standard of living.

Two days ago your financial situation hit a crisis because you discovered that your beloved cat "Mr Whiskers" requires a six hundred dollar operation or it will die.

You need a 20% raise.

You also feel that you need some training in web development. You know of a sixmonth web development course starting soon.

Please rate your own learning below. Circle the appropriate number below.( 1= Very little achievement 5 = Excellent achievement)

Before this tutorial After this tutorial

Table 1 : Design of the study Role Play Game