@todd >last week working through 1000+ data points collected So... Just because you have collected all this data, does not prove or show anything. Many people in the stock market collected 1,000,000's and 1,000,000's of data points, and their models where wrong, very wrong.
What matters is does your method predicts the behaviour of real people using the site you are designing. And can the method be tested? Nobody is arguing that you can't collect themes. What is been argued is that you can pick and mix themes to create a composite human? That then becomes fiction. We can say that most Scandinavians are taller than most American people. We can say more Americans go to university than Scandinavians. Additionally people that go university are clever than people that don't. As soon as we mix then up so that we create our composite persona we end up with the clever short American, and the dumb tall Scandinavian. Both are fiction. Some would argue that my example is a bad persona, and they only create good personas. The question is how do you tell what a bad persona is? Each theme is interesting but combined they are fiction. How do doctors identify illnesses? Not by personas! unless you go to a Witch doctor, but by a set of heuristics (as defined by Imre Lakatos, not *Nielsen*), and there are many other methods, and theories. The doctor builds evidence. Some of it is qualitative , some of it is quantitative. You can create quantitative findings from qualitative data, but it is very hard to go the other way. The maths gets very hard (you move from A vs B to multivariate). I said earlier > When writers merge real life characters together the work becomes > fiction. How do get around the challenge of theory? > The challenge of personas is that I know of no theory that backs them up. Find me one! Can you describe how you communicate the mappings one to one? How would you > communicate findings from 40+ interviews across 17 different customers from > 5 different countries > We talk about the themes, and give examples using real people. The same way anthropologists have been doing it since Malinowski. When a new theme develops it means that we do not have to alter a persona. James http://blog.feralabs.com 2009/3/10 Todd Zaki Warfel <li...@toddwarfel.com> > > On Mar 10, 2009, at 7:31 AM, James Page wrote: > > @todd > > The issue here is that personas are a generalisation of the user base. As > Christine > Boese a couple of months back on the list said: > > > Yes, and the problem with that is? > > Descriptive, rich, qualitative methods are by definition NOT generalizable. >> That would be the whole point. One can inductively triangulate data, amass >> evidence that reinforces emerging categories of data, develop heuristics, >> and even conduct parallel studies and discover points of intersection >> between similar qualitative or ethnographic-type studies. >> > >> So replicate to some extent, but generalize, never. > > > Bullocks. > > We just spent the better part of last week working through 1000+ data > points collected from qualitative research, synthesizing, looking for > patterns and finding 19 themes, each with subgroups/themes. Those themes > showed a definite pattern. From those patterns, or generalizations, were > identifiable across customers of various sizes and industries. > > The individual stories might not be able to be generalized, but there are > definite patterns that can be found and used to communicate what it's like > for your customers and to provide empathy and understanding. > > When writers merge real life characters together the work becomes > fiction. How do get around the challenge of theory? > > > Bullocks again. How do doctors identify illnesses? Not every single case of > lung cancer is identical. Are you going to claim that lung cancer is > theoretical? > > How do you communicate your research findings to your clients? >> > For qualitative data probably very similar to the way you communicate to > clients but the mapping is one to one, not many people summarised as one. > > > Can you describe how you communicate the mappings one to one? How would you > communicate findings from 40+ interviews across 17 different customers from > 5 different countries? > > The time saving is because you do not have to create a pseudo person > between the research, and the report. As more information is discovered it > is very easy to add to the knowledge base. > > > The time spent on creating a persona is saved in spades against the time > spent trying to communicate individual stories of 30-100 individuals and the > edge case arguments that come from not having the data. > > > Cheers! > > Todd Zaki Warfel > Principal Design Researcher > Messagefirst | Designing Information. Beautifully. > ---------------------------------- > *Contact Info* > Voice: (215) 825-7423 Email: t...@messagefirst.com > AIM: twar...@mac.com > Blog: http://toddwarfel.com <http://toddwarfel/> > Twitter: zakiwarfel > ---------------------------------- > In theory, theory and practice are the same. > In practice, they are not. > > > > > ________________________________________________________________ Welcome to the Interaction Design Association (IxDA)! To post to this list ....... disc...@ixda.org Unsubscribe ................ http://www.ixda.org/unsubscribe List Guidelines ............ http://www.ixda.org/guidelines List Help .................. http://www.ixda.org/help
