News

The third and final blog in our series on qualitative data analysis is where we gather up our notes, pictures, photos and recordings to immerse ourselves fully in the data, interpret meaning and draw our learnings and conclusions.

Part 1 took us beyond coding to see a much broader landscape of what’s possible in qualitative data analysis, whilst Part 2 shared ways to gather better quality data by analysing as you go and developing observation and fieldwork skills.

Here we look at methods to find a way through the data, practising and refining skills in pattern-spotting, clustering and mapping. We keep our human participants front and centre with behavioural analysis, personas and creative ways to bring others into the process to enrich and sense-check our work.

Finding your way: patterns & clusters

When you’re not sure where to start, maps are really handy. Even if you think you know where you’re going, understanding the territory you’re in and the different ways you can get to your destination are crucial.

Maps can be extremely versatile and help you to explore the big picture and relationships across time, place, sentiment, concepts, words and visuals. Maps you can use in qualitative research analysis include:

  • Time maps (e.g. student journeys, where you or the participant maps experiences spanning a period of time)
  • Place maps (e.g. where interactions took place, where someone visited, virtual visits to websites/ apps)
  • Sentiment/emotion maps (e.g. highs and lows of experiences – often useful together with a ‘time’ map to explore experience across days, months or years)
  • Concept maps (e.g. clusters of post-it notes, diagrams and mind maps)
  • Word/verbal maps (e.g. text analysis, word clouds)
  • Conceptual matrix (e.g. plotting two axes and mapping behaviours, patterns and behavioural traits to see where clusters and gaps emerge)

You’ve almost certainly been involved in this technique if you’ve ever generated ideas on post-its, then mapped them into similar clusters and themes, like the picture below:

But don’t stop now! Once patterns start to emerge, you can challenge what you’re seeing, test assumptions against your fieldwork notes and re-organise your maps as many times as you need.

Plotting your route: mapping & models

One way to start testing your emerging theories is to map them onto a model that already exists – perhaps something from a previous related project or literature from your earlier reading. What matches or contradicts? Where are the dense clusters and where are the gaps?

Here’s an example of a really detailed student journey map from a project at UTS. How could you use something like this in your own research on student experience?

Meeting the locals: personas & profiles

If you’re analysing observed human behaviours as part of your qualitative research, one really useful tool is to plot patterns across a spectrum – ideally across quite a few, given the complexity of human behaviour. In one project, we created 20+ behavioural spectra in the first stages of analysis, before whittling them down to 5-6 of the most relevant for our learnings.

Working across a continuum like this pulls your analysis away from ‘either-or’ binary thinking, and is helpful when collaborating with other researchers and comparing notes from interviews you’ve each conducted. When you’ve identified the relevant ‘opposites’, e.g. students who ‘want to be here’ versus those who ‘have to be here’, you can discuss the different people you interviewed, and where their behaviours and actions might be ‘mapped’ (example below):

As with mapping activities, you’re also looking for clusters and gaps – is everyone at one end, spread across the spectrum, or at the extremes? Why might this be?

In larger projects where you have enough participants (at least 15-20, as a rough guide), you may also be able to identify a number of personas, or ‘pen portraits’ illustrating a typical profile you have observed. These should draw on the behaviours you explored and can help those not involved in the project to engage with the individuals you discovered in your analysis.

Involve community with collaboration and playback

When you’re nearly finished with the analysis process, you can also take your findings back to interested stakeholders, colleagues and even those who participated in your research. By sharing your theories and findings, you can invite another layer of input and analysis and discover perspectives you and your fellow researchers may have missed. This can include:

  • Sharing emerging findings in a workshop; asking what resonates, what contradicts and tests people’s assumptions;
  • Sharing back with student participants and inviting their reflections;
  • Creating different formats to share findings (blogs, video, interactive tasks and activities)

Whilst it’s tempting to wrap up the findings, sign off your report and move onto the next project, some of the best qualitative projects I’ve worked on continue to evolve and are used in follow-up projects and initiatives that build on the findings. Stay open to challenges and the ‘long tail’ of the analysis process; it’s extremely rewarding, and makes all the effort for you and participants worthwhile!

Further resources and reading

As Nick Hopwood mentioned in Part 1 of this series, ResHub is trialling a monthly drop-in Qualitative Data Analysis Clinic for anyone working with qualitative data. Drop by and become part of a community sharing interest in qualitative data! You can register for upcoming clinics via the links below:

Tuesday 27th Jul (12:30 – 1:30pm)
Tuesday 24th Aug (12:30 – 1:30pm)
Tuesday 14th Sep (12:30 – 1:30pm)

There’s a great, short book by Design Researcher Erika Hall called Just Enough Research that shares a practical and real-world lens on research. If qualitative research is just another item on your to-do list, her tips and advice on using ‘just enough’ are great.

Sue Bell, an Australian market research professional, offers an introduction to qualitative analysis in this short webinar presentation.

Feature image by Charles Deluvio on Unsplash

Join the discussion