Action research: experimenting with roleplay

By Scott Donald

It’s time to put my money where my mouth is. This post will describe some action research I recently carried out and will function as a template for anyone else who is interested interested in collaborating. It should be clear that my main aim is to make it as accessible as possible. I’m not the first person to try and present research in this way, but there is (naturally) a lot of very scientific, jargon-loaded material out there that puts many teachers off reading any research. There are plenty of places to go if that type of thing tickles your fancy, but here, the emphasis is on making things easy and practical.

The problem

Many teachers struggle with students who don’t contribute enough in class, especially in terms of speaking activities in ELT classrooms. This may be because the students are too embarrassed to speak up in class. This could be to do with personality, but can also be related to cultural factors, e.g. Japanese learners (Swan and Smith 2001). Another reason for students not communicating could be described (perhaps harshly) as a lack of imagination, but may, in fact, have more to do with a lack of motivation.

Motivation is a complex area affected by many factors; things like personality, external influences, or age. In a discussion I had recently with colleagues, we talked about the lack of motivation (imagination) in adolescent classes. Here, instead of a Japanese factory worker, afraid to make a mistake and lose face in front of factory owner, we can imagine a slouching, smirking teenager. Let’s call him ‘Pedro’ for now. Pedro is also worried about losing face, but he’s smirking because he’s trying to carefully word his response to the teacher, so as to walk the fine line between ‘cool’ and pissing off the teacher so much that he ends up with extra homework. Couple that with the Spanish vergüenza (‘shame’, previously mentioned in my Post-talk) and we have a recipe for long, painful silences; or (perhaps worse) Pedro might resort to the dreaded one-word answers.

‘How was your weekend, Pedro?’

‘Fine.’

‘Did you do anything nice?’

‘Study.’

Isn’t it amazing how Pedro saps your enthusiasm with so few syllables. Of course, not all young learners are like this. I’ve been quite lucky this year, and have had some really good young learner classes with high motivation. There is no Pedro.

However, I teach two separate classes of eleven-year-old students, and within them, there are a couple of shy students who don’t contribute as much as the others. Again, this could be an issue of either shyness, or motivation. I have a boy who, rather than speak, gives me a shrug and a smile every time I nominate him in open class, and a girl whose parents have told my boss that she really enjoys classes, but who I don’t think I’ve seen smile once in my class. These students are often sandwiched between some very loud and excitable young humans. We have Sara, who, even though she knows my name is Scott, will insist on calling me ´Scotland´. We have another trio of female troublemakers whose ringleader loves nothing more than drawing pictures of overweight naked people and then trying to incite a giggle-riot. The boys are more or less kept at bay by this dominance of girl power, but I still have to intercept scribbled notes of blossoming romances going between them and the girls.

The hypothesis

Let’s put these ‘characters’ to one side for now, and focus on the quieter students. One way of dealing with students who are reluctant to contribute is through the use of roleplay in class.

Roleplay is said to:

• be fun (and therefore encourage participation)
• help shyer students who may otherwise be embarrassed talking about themselves
• offer a wide range of language opportunities

These benefits are promoted on popular teaching websites in scientific journals, and in popular books, such as ‘Role Play’ by Gillian Porter Ladousse.

The experiment

The nice thing about having two classes at the same level doing the same course is the ability to set up something of a control group, i.e. one that doesn’t receive any special treatment. I checked the coursebook for some upcoming grammar, and decided to use ‘be going to’ for the experiment. I often plan these classes with my colleague, James Sampson, who also a class at this level. (He likes to claim he does most of the planning, but this is simply not true.) So, James found an exercise on ‘be going to’ from Teach-This.com, which we/he adapted to better suit our classes.

For the experiment, I designated my Monday group as the non-roleplay group (NRG), and the set-up was as follows:

• T hands out worksheet and elicits follow up questions in Open Class, e.g. ‘Are you going to call someone after class? Who are you going to call?’
• T explains that the Ss are going to walk around the class, asking all the questions, until they have spoken to everyone. They also must write the name of the person they have spoken to, and their answer. They must also answer ‘yes’ to the first questions, even if it’s not true, and then invent an answer.
• Ss mingle and T adopts the role of facilitator, monitoring the students, not interfering unless students require help.
• Once most Ss have asked all their questions, T conducts Open Class Feedback, and asks students about their answers.

As you can see, this is a fairly straightforward activity. For my Tuesday group (RG), I did something very similar, except I allowed the students to choose a role beforehand. Instead of answering the questions as themselves, they would instead do it as a celebrity, e.g. Rihanna, Nadal, Justien Bieber, etc.

I thought about assigning celebrities to the students, but:

• I have to appreciate that I am no longer cool or relevant (if I ever was) and that I might have struggled to name 12 celebrities that the kids would definitely know.
• I thought it would be more motivating for the students choose their own roles.

The final element was that James would also do the activity with his own class (NRG2), to give me a little bit more data.

The results

Number of students: NRG (9) NRG2 (8) RG(8).

Length of the activity: The RG took about ten minutes longer than NRG and NRG2. This was due to the extra time it took to set up the activity while students chose their roles. It turned into something of a competition to see who could choose the best celebrity.

Task completion and problems: Most students completed majority of the task in all three groups. There were a few pieces of vocabulary that needed clarifying and very little emergent language in any of the groups.

Student engagement: High in all three groups, but James and I agreed that the activity could have been shorter, perhaps with some scaffolding language included. The answers the students produced in the RG were slightly better than in the NRGs, e.g. students seemed more willing to answer the question,

‘Who are you going to get married to in the future?’

They also seemed to giggle less at the question. There was some engagement with the character in the RG group. ‘Emma Watson’ said she was going to watch Harry Potter at the cinema. But there was also evidence to the contrary, e.g. ‘Adele’ saying she’d go to the cinema with ‘Emma Watson’ (which seemed to be more to do with the fact that students were sisters.) This lack of substantial engagement with their roles was apparent in Open Class Feedback too where I asked questions such as, ‘Do you think Adele and Emma Watson would like to go to the cinema together/what would they watch?’, and I was repeatedly met with shrugged shoulders.

The Conclusion

I found the process of doing the action research itself really interesting. For example, I found it challenging to remain objective and not influence one of the groups over the other. I think it’s very tempting to apply the things you’ve learned from a Monday class to your Tuesday class, and I was trying to resist this for the sake of objectivity. Also, having the hypothesis in mind (that roleplay can help motivation), I had to keep reminding myself to give fair opportunities to both sides and not be disappointed when there didn’t seem to be a great deal of difference between the RG and NRGs.

Regarding this specific experiment, the fact that it produced very little evidence to support the hypothesis is interesting. It was clearly a very small sample, and a number of factors could have affected the results. So it would be foolish and unscientific to try and claim that roleplay therefore doesn’t have a positive effect on motivation. However, that’s not to say that there aren’t several useful conclusions that I can draw from the action research:

• While many attest to its benefits, Roleplay isn’t a panacea and won’t work as a motivational tool for all classes, students, or situations. This may sound obvious, but there is something satisfying for me in having tested it myself.
• With some set-ups, roleplay can even have a detrimental effect on the student. The shy student who tries not to smile in my class was unable to think of a character. In fact, she had the whole class waiting while she sat, obviously embarrassed, trying to think. It was a long time before she thought of one and, although she did the activity without any problems, I think her embarrassment during the set-up far outweighed any benefit from the main activity.
• Something that proved very effective in the lesson was James’s suggestion of getting students to say ‘yes’ to all the questions, i.e. that students should say ‘yes, I am going to the cinema this weekend’, even if they are not, and forcing them to imagine what they are going to watch. This approach seemed to increase the language produced and the students had no issues telling lies (surprise, surprise.) It also opened up the possibility of an extra task/question in Open Class Feedback. After the students had told me someone’s answer, we would decide whether we thought it was true or false. This idea, of allowing students to invent their answers, could be seen as somewhere between roleplay and non-roleplay — allowing students to use their imagination and personal experience at the same time.

Looking ahead

Two cheers for democracy

It was all very well-intentioned giving the students the chance to choose their own roles, but I won’t be doing it again in a hurry. It was time-consuming and embarrassed the shy student. There may be ways round this, but I think the most efficient would be the authoritarian approach,

‘You’re Justin Bieber, no arguments.’

To avoid them trying to argue with me, I’ll put the roles on cards and deal them face down — leave it up to fates. I’ll need to swot up on my Spanish celebrities though. Thank goodness Justin Bieber reinvented himself. I can at least continue to use him in class for a little longer.

Imagine all the students, lying in harmony

While I clearly haven’t been put off the idea of roleplay, I will think twice before doing it again. Especially considering the success of the in-between option of getting students to lie/imagine their answers. I have repeated this with a similar activity since and it also worked well.

A rewarding experience

I can definitely say that I found the process enjoyable. I would be interested in following up on either of the two areas outlined above, or a different area altogether. Doing it again, I think I would choose a problem that is more directly related to my own teaching, rather than my colleagues’ teaching. Another option for action research is also to focus on a change you want to make to your classroom, rather than a problem (Curtis, 2014). And another option is something I mentioned in my Pre-talk called exploratory practice. Here, instead of starting with a problem, it’s simply starting with an area that you would like to know more about.

I would be even more encouraged if others decided to try any of the above and share it on this blog. Please contact me via the blog’s Facebook page if you’re interested. It doesn’t need to follow the template I’ve used, problem, hypothesis, experiment, results, conclusion, looking ahead; it’s merely there as a suggestion, (which I’d also be happy to receive any feedback on.) Action research doesn’t even need to be particularly academic (Baker, 2014) ← you don’t even need to include these pesky little references! It’s about trying something in your classroom, and sharing it with others. So, don’t be afraid. As my experiment suggests, there is nothing to be gained from being shy :)

Bibliography

Baker, F., (2014) What is Action Research? Webinar (members only, available at: http://www.cambridgeenglish.org/cambridge-english-teacher/)

Curtis, A., (2014) Action Research for Language Teachers. Webinar (members only, available at: http://www.cambridgeenglish.org/cambridge-english-teacher/) (short version available at: https://www.youtube.com/watch?v=x0KWFJ-rQwU)

Kuśnierek, A., (2015) Developing students’ speaking skills through role-play 7 (2015)73–111. Available at: worldscientificnews.com

Ladousse, G.P., (1987) Roleplay, OUP

Phillips, D.K. and Carr, K (2010) Becoming a Teacher Through Action Research, 2nd ed. Routledge / Taylor & Francis Group, LLC.

Swan, M. and Smith, B., (2001) Learner English, 2nd ed. CUP

Wallace., (1998) Action Research for Language Teachers. CUP

(2004) Role-play. Available at: https://www.teachingenglish.org.uk/article/role-play

(2014) Are you going to…? Available at: https://www.teach-this.com/images/resources/are-you-going-to.pdf