What’s in 4,000 stories? What about 100?

Two contrasting weeks working on very different narrative projects – both fascinating in their own right. And both doing SenseMaker™ analysis under time pressure, but with datasets that couldn’t have been more different.

  1. At a conference with friends in the US, we gathered people’s feedback on a particular organisational issue. Over two days, we collected just over 100 stories – about 8,000 words. Two of which were swearwords – I love it that people drop the pretence when they give feedback like this. Closing the survey at 8pm on the Thursday night, I had a few hours to analyse and feedback to the plenary session on the final day.

    The analysis itself was actually straightforward and fascinating – the difficult bit was deciding what had to be left out in the final version – I had only 20 minutes to present. And then generating the slides from the material. The results were well-received – and as always happens with narrative research, there were plenty of “oh! Really?” moments, as well as humorous ones
      

  2. The second project has been a much larger affair. Over 4,000 stories collected over the past couple of months. Initial analysis of topline trends and outliers is easy – and equally fascinating. Again, it’s possible to get headlines and interesting materials quickly from this much material but then you get into the nuances and hypotheses that arise from the first cut of results.
    A different difficulty arises here – there are so many interesting routes to take through the analysis:
    1. What happens as level of education goes up? Age?
    2. If I combine ideas about the future with ideas about poverty how do the results look? And then level of education? And age?
    3. How do all of the above change depending on a) time of day b) day of the week c) where we collected

The answer to both is, of course, to know what’s interesting to the project owners and put degrees of focus in those areas after the initial top-level analysis. And to look for evidence around some of the key assumptions – are they right or misjudged? And to indulge in a small amount of exploration according to my own instincts – after all this time working in this field I’ve got a good (but not infallible) sense of spotting oddities and anomalies that need to be highlighted.