Folks who follow this blog will know that I try to strike a balance between topical, practitioner-oriented pieces and more academic articles such as scientific papers & other peer-reviewed content. I’m not always successful, but firmly believe that the most valuable use of this channel is to provide practical insights backed up by a solid theoretical basis wherever possible. Of course, it’s not an easy line to tread… get the balance wrong and you can end up with worthy but dull academic pieces that extend only marginally beyond the bounds of a narrow research community. Or conversely, anecdotal experiences that have little chance of delivering insights that generalise to other contexts and individuals.
But occasionally we get to deal with a theme that genuinely does cross boundaries – in other words, one that has a clear foundation in scientific research, but with immediate implications for design practitioners. I’m talking in particular about the search modes work – a theme that I’m finding has increasing value both as a way of describing information-seeking behaviour and for defining the types of design interventions and strategies needed to support that behaviour. At the moment, I’m working with Joe Lamantia and Stephann Makri to validate the initial framework, extending it to new domains and using on our own insights and practical experience to test its validity.
But we’ve reached a point where we’ve gone just about as far as we can with own insights and analysis. The coding process we’ve adopted is commonly used in the social sciences, but after a while the data starts to look increasingly like a Rorschach ink blot: we no longer perceive the ‘objective reality’ (to the extent that it exists) and instead see patterns and sequences that we expect to see (subconsciously or otherwise).
So, we need your help. Would you be interested in analysing our data? It entails reviewing our framework, then looking at a set of micro-scenarios and telling us what behaviours you see. I’m told the enterprise search scenarios take about 30 minutes to code, the site search ones perhaps 60 minutes. But if you have less time that we’ll of course be interested in any sort of constructive feedback – on both the process or its output.
We’d love to have your input on this. Feel free to contact me either here or offline!
Yes of course, very happy to help. I’ll try and get various members of our team to give feedback.
Thanks Charlie – glad to hear it! I’ll drop you a note offline…
I would be glad to support you in this analysis. Do you particular approach in mind or are you open ? On my side I’ve applied markovian models on such data which gave good understanding of the patterns.
Glad to hear it @gerdupont. Yes we do have a particular approach in mind, based on this model (which used a ‘grounded theory’ type approach): http://isquared.wordpress.com/2012/09/18/a-model-of-consumer-search-behaviour/
But of course we’re open to other possibilities. If you dop me a note with your email address I can send you further details. Thanks!