Problem A major motion picture studio wanted to know if there was a market for a new movie service, one where patrons we're encourged to use their mobile phones to share the experience over social media. I initially designed a focus group research study with my graduate students, but this failed miserably.
Our protocol was to begin each focus group with the same statement: “Imagine you've gone to the movie theater to see a film with friends, and when you arrive you see two choices for the film you are interested in. The first choice is a regular movie showing. The second is a showing of the same film, but everyone will be allowed, or even encouraged to use their mobile phone to share the film experience over social media,”
We never go to finish the orientation message. The focus groups errupted in vehement objection. We couldn't gather data.
Searching for a Solution After the second failed attempt at a focus group, we looked for a new procedure by conducting a literature review on methods to validate personas. In the process the reearch team found an article by 2 researchers at Microsoft who said that personas as a research method were non-scientific because they were non-falsifiable. The slide below—from my presentation at the Service Design Network Conference in Cardiff quotes their argument (with some paraphrasing)...
Getting Scientific The claim that personas were an invalid research construct led me to a mathematically sound but little used methodcalled Q-Method. Invented by the physicist and psychologist (he held doctorates in both fields) William Stephenson, Q-methodology is a scientific and falsifiable methid for studying subjectivity. With a little tweaking, we were able to adapt the method to the problem of validating personas for the movie service.
For the mathematically-inclined, Q-sorts allow individuals to array statements of opinion in a very personally releveant manner. These arrays are then recorded and transformed using factor analysis with data set transformations such as Eigen-vector rotation.
The results were very interesting. We did find a persona very interested in the Social+Movie experience, and we were able to use Q-Methodology to flesh out three different personas, two of which have distinct preferences and behaviors for socializing around media and films.
The results were striking enugh that I was able to publish the method in Touchpoints, the international journal of service design, as well as present the method as a valuable means to generate valid, testable personas for service design.
