Throughout my action research project, I followed a cycle of planning, action and reflection (as per Bradbury, 2015) to influence an aspect of teaching and learning which was the focus of my project. In this process, I used the following methods to collect qualitative data:
- Participant observations (throughout 5 lectures, 3 student-led workshops, and the student presentations);
- Evaluation questionnaire (to collect feedback from students and refugee collaborators at the end of the project).
In this project, I adopted an ethnographic approach, and conducted participant observations to holistically investigate several aspects of the intervention for an extended period of time. As per Mattelmäki (2006), this was an exploratory, rather than evaluative, method which took place in natural settings (i.e. the classrooms), and was open to refinements throughout the process as new learning shaped future observations. I considered whether I should hire a research assistant or collaborator to help me with the data collection, and to allow participants to be more open and honest in sharing their feedback. However, this was not possible given the limited resources available (I didn’t have budget to pay someone else, and I didn’t want to ask people to volunteer and do the work for free) but also with the view of avoiding having external people join the masterclasses as observers, as it could feel intrusive for the participants (especially for vulnerable people such as the refugee collaborators).
I had initially considered asking participants to fill in two questionnaires, one at the beginning and one at the end of the project, but finally – as discussed also in the group tutorials – I decided to use only one questionnaire at the end of the project. This choice was due to the time limitations of the project, and the complexity of handling large amounts of data, but also to avoid overwhelming participants with filling in too many forms at the start of the research (when their first impression of the project is built), considering they already had to fill in the Ethics Forms. In the design of the questionnaire, I included a mix of multiple choice questions (to be quick and easy to answer, to break the ice) and semi-structured questions to allow participants to openly share their insights on their change in skills and knowledge, their main take-away from their project, their feedback on what worked well and what could be improved. I was very keen for the questionnaire to be no longer than 2 pages, to make it easier and not overwhelming for people to complete, especially considering that refugees are often asked to fill in many forms. Moreover, I decided for the questionnaires to be anonymous to allow participants to more honestly share their feedback.
I started drafting the questionnaires during the PgCert workshops and iterated them thanks to feedback from my peers. I created one questionnaire for students and one for refugees, with slightly different wording, to gather insights on how the project contributed (or not) to fostering a safe space for interactions and changing their perspectives of their collaborators. After getting feedback from my tutor and peers at the group tutorial, I printed the final version of the questionnaires (see files below), and used them at the final session of the Collaborative Challenge on 6th December. I handed in the questionnaires in the middle of the session, to give people more time to fill them in without rushing, and to avoid the risk that they would leave the room without returning the questionnaires to me. In the end, I collected data from 26 students and 6 refugee collaborators, which I think was substantial, considering the scope of the ARP.
Drawing on Morgan (1988), I had initially considered also conducting a focus group with 3 students and 3 refugee collaborators to gain more in-depth data and compensate the limitations of a written questionnaire. I had also considered whether the focus group should be conducted online or in person; while in person would likely be more engaging, I decided that the online delivery would contribute to overcoming financial barriers for refugees to travel to the venue of the focus group; moreover, delivering the session online, I could automatically transcribe it, saving me time for the data analysis process. I had planned this session to be delivered the week after the students’ project presentations, allowing myself one week of time for reflection and planning before the next cycle of data collection. I am aware that a focus group would have opened up more interesting issues, but I had already collected enough data through the above-mentioned methods, and this extra step was not deemed needed considering also the time limitations of this ARP, and to avoid over-asking of participants, who had already shared with me lots of insights throughout the ARP and my on-going larger research project.
I then thematically analysed the data (as per Kara, 2015), following a manual, iterative process that encompassed data reduction, display and conclusion drawing. Drawing from Miles and Huberman (1994), the process of data reduction entailed selecting, synthesising, and transforming data. I produced data displays in the form of tables to represent coded data (with category names attributed to meaningful segments of the transcriptions) and facilitate the recognition and comparison of themes, identified as patterns cutting across the data. Finally, I drew conclusions from the data by identifying themes and sub-themes, and outlining relationships between them.