ORIGIN OF THE LEARNING DESIGN
From observing Andrew Vincent and John Shepherd when they were introducing the Middle Eastern Politics role-play at The University of Melbourne in the early 90’s.

NUMBER OF TIMES LEARNING DESIGN HAS BEEN IMPLEMENTED
Three so far.

MODIFICATIONS SINCE FIRST USE
Modification is ongoing based on feedback after each implementation. Modifications made include:

• Better character descriptions that are more clearly linked to the problem to be solved.
• Need more moderators, at least one for each camp, to keep the participants focused on the problem rather than on surface socialization and clearer instructions/training for those moderators.
• More pre-prepared kick start episodes for the moderators to introduce as the role-play progresses.
• Guidelines for the teacher about assessment and its role in keeping focus.
• More clarity on the process of moving from discussion to consensus to framing a resolution.
• Need for clearer student learning outcomes both in regard to content and in regard to generic skills.
• The convict database which is a great resource for those participating as convicts but more background is needed for those participating as officers.

To provide flexibility in use with differing class sizes, a new smaller version has been created reducing the number of characters to 12 for classes of 12 students or multiples of 12.

Next time it is run across classes that are meeting at different times we will make changes to the organization of the classes: one class per role-play world, one moderator per class in addition to the overall moderator, each class with minor variations to test reaction to the variations.

A demonstration version and a version for school level students is underway.

RESEARCH CONDUCTED ON THE DESIGN IMPLEMENTATION
The evaluation methods of the design include email logs, video interviews with students, open-ended email survey, designer's ad hoc journal and external peer review via this AUTC Learning Designs Project.

The outcome of the first experience could be summarized as: many lurkers and a very small number of active and highly creative participants. This meant that the role-play fizzled out towards the end because the active participants were not receiving enough feedback from the other characters who they were meant to be collaborating with. As this was a voluntary staff development exercise to merely expose university teachers to the concept of role-play as a learning design this didn’t really matter as it served to teach us all a valuable lesson: the need for active participation by all characters if the plot is to progress towards a resolution.

The second experience was more fruitful as the four face-to-face sessions provided the discipline and structure to keep the momentum going. However many of the participants were frustrated at sometimes having to wait a whole week for a reply due to the limitations we caused by spreading the characters across 4 classes. In hindsight this was a mistake but at the time we had envisaged that students would be so motivated that they would be using the role-play at many times other than just the set class time. In addition many students were merely "play acting" and having fun and only skimmed the surface of what was required in order to reach a resolution. Fun can be conducive to learning but firm moderation may be needed to move the learning to deeper levels. Both these problems can in part be attributed to the fact that the role-play was not tied to any assessment. Unfortunately it is human nature that the weight of assessment gives weight to the worth of the exercise.

Comments from the external peer review conducted by this project include:

Feedback from other participants [in the role-play] can highlight limits in students’ knowledge of interactive skills, research methods, and/or First Fleet content. A wealth of links and resources for extending learning are provided. However learners are free to minimise challenge, as there is no assessment. In the sample class files, the evidence suggests that few students went beyond minimal use of the database or minimal engagement in problem solving. The requirement to interact with others could motivate students to discover information to sustain a case, and to evaluate input from others. However sample interactions among real students do not support this assumption.

The product structure could be strengthened pedagogically by building in: more opportunities for self-assessment and feedback; more opportunities for practice in role-play and discussion; various forms of assessment integrated into the activities; a clearer definition of user groups and relevant instructions that match each group; guidance and links to learning resources about [moderation for moderators and about group work and problem solving for students]. Also some interface design aspects could be improved to ease navigation, especially between the role-play and the database.

Results of the evaluation of the third implementation are not yet available. Future research will focus on reusability issues, namely: What do teachers need in order to take the First Fleet and use it themselves without help from the original designers?

THOUGHTS ABOUT EFFECTIVENESS
Quotes from students, transcribed from the interviews, included many which we needed for improving the interface and debugging the software; for example the need for a synchronous chat function in addition to the asynchronous functions, which Fablusi provides but we had not thought to add to this implementation of the role-play. Feedback also included the following:

About learning outcomes related to content aspects:

• Interesting to find out about First Fleet as I only vaguely knew [about it] before.
• Learnt much about history – I had already done First Fleet at primary school and high school and now university [sigh] but yes I can [now] understand it from their perspective.
• We didn’t have this kind of opportunity at high school.
• Great fun, especially for younger people 14-15 years – my brother is currently doing a project on Roman History at school and this type of thing would be good for him to use.
• Already covered [First Fleet] at university so could do one that is more political and current e.g. terrorism.

About learning outcomes related to affective aspects:

• Doing a profile at the beginning is good because it makes you establish your character and who you are and then how to approach the rest of the problems.
• My character is a "bad guy" but it’s because he’s interested in his own survival – yes I have empathy for him.

About organizational aspects:

Probably we needed another week because not enough people were using it in their own time rather than just in this class lab.

UNEXPECTED LEARNING OUTCOMES
Several students reported that they now have a better understanding of discussion forums and email which surprised us as our impression of most students is that they are very comfortable already with online technologies. In future evaluations a gender breakdown of these comments might be useful and a comparison with faculties other than the Arts Faculty.

HOW LEARNER ENGAGEMENT IS SUPPORTED
A learner is immersed in a role and immersed in a real world problem to solve and all other learners around them are similarly immersed. They need to engage with the content in order to play their role. They are free to portray the role with a degree of wit, fun, competitiveness and humour which assists engagement. That the real world tasks can also be the assessment tasks supports engagement and there is a final reflective task which engages them in looking not only at the historical insights but also at their communication and negotiation skills as well as empathetic emotions.

ACKNOWLEDGEMENT OF LEARNING CONTEXT
Immersion in this early colonial negotiation provides first hand insight into the development of democratic processes as well as environmental sustainability and racial issues. Although the First Fleet context seems white Anglo-saxon there are in fact roles that raise conflicts that are broader: male vs female, officer vs marines vs convicts, class structure, Jews vs Protestants, black vs white plus Aboriginal roles could be added if the moderator is suitably skilled. The skills of negotiation and communication should be transferable. The learning design itself does not explicitly state the above however it would need to be a key feature of the debriefing.

HOW THE LEARNING DESIGN CHALLENGES LEARNERS
Feedback from other team members, from other students in the class (in their roles) and from the moderator ensures that self-questioning is almost automatic in this learning design. However assessment would still be a key factor in encouraging students to push themselves.

OPPORTUNITIES FOR PRACTICE
There is no role-play if students are not articulating and participating. A role-play is only successful if all students are active. The key to successful role-playing however is a moderator that assists the students to go beyond mere play-acting to more sustained deep thinking about the problem and the role. The moderator cannot be just monitoring from the sidelines but must be intensively involved in advising roles and in seeding scenarios to kick-start discussion if it starts to fizzle.