We are delighted that we now have around 20 members of our Talking Mats (TM) research group. Members come from a variety of countries including the United Kingdom, Denmark, Cyprus, Germany, Sweden, Australia and Japan! We are a mix of academics and practitioners, with many combining both roles. So far we have spent time getting to know one another via video sessions and thinking about how the group might work.
We have decided our initial focus will be thinking about ways of analysing the data that is generated from conversations that are supported by TMs. This idea was suggested by Nikita Hayden. Nikita is a PhD student at the University of Warwick exploring the outcomes of siblings of children and adults with learning (intellectual) and developmental disabilities. Part of her research has used TMs with children with severe learning disabilities and their siblings to further understand their sibling relationships.
The types of data generated have been rich, vast and varied, leading to an overhaul of Nikita’s initial plan to analyse her TM data. This has raised questions about how TMs are interpreted and analysed in a research context, and what scope there is for our group to explore and synthesise the analysis potential of TMs. This is a question that the TM team is often asked and so having some information on the different options would be useful.
TM discussions generate various types of data, including:
- The photograph of the mat (which symbols are placed under the various columns);
- The conversation generated during the discussion;
- The body language and facial expression of the ‘thinker’;
- The speed of placement of symbols;
- The symbols that are moved following feedback etc.
We would like to review existing publications that have used TMs as research data and think about possible methods of analysis. This may include consideration of both within and between group research analysis techniques. It may also involve exploring the potential of both traditionally qualitative and quantitative analysis techniques, such as thematic or conversation analysis, or by drawing on data from the symbol placements to provide pre-post evaluation data.
We hope to generate a list of guidelines about what you might need to take into account when considering how to analyse these data. A challenge when analysing TMs data, is how to handle the variation in the types of data collected between participants. For example, some participants may place a large number of symbols, whereas other participants may have placed relatively few. This raises questions about how we deal with ‘missing data’. In small samples, how can we conduct a pre-post evaluation where some symbols are missing for some participants? If some participants use a five-point scale, and some use a two-point scale, what numerical analysis potential is there, if any? How can we appropriately derive qualitative themes from across our sample if some of our participants were minimally verbal? What sorts of non-verbal cues have been analysed in research using TMs?
Please do share any ideas or questions you have with Jill Bradshaw, our Talking Mats Research Associate – J.Bradshaw@kent.ac.uk