Dear Mike and all,
Mike, good to hear from you.
In my MDes, I was very interested in statistics. I took two courses with
the Ed Psych department at Cincinnati (I was fortunate to have Mike as my
master’s thesis advisor, thanks Mike!). My master’s thesis included
statistics, and I even created an infographic about how to decide basic
statistical methods.
Right now, I am less sure that statistical methods are the best for design
research. When a medical researcher, for example, wants to test if a
medical intervention is effective for a population, they go ahead and apply
the intervention to a sample, and then using statistics they are able to
infer if the outcomes are true or not for the population. This makes sense.
Imagine an HIV intervention, a medical research can select a sample in New
York. If the research outcome is that the intervention is effective,
physicians can apply the intervention to any human in the world with good
confidence.
In design research, there are some reasonable scenarios to apply statistics
in design research. Companies like Google and Uber use a good deal of
statistics. This makes sense for making design decisions of people that are
already users or for usability studies that focus on some cognitive issues
(e.g. info processing), which are very similar among all individuals in the
population.
We are tempted to think that some design products are interventions;
however, some of these interventions are created in a specific situation
that will not apply in other contexts. Imagine for example one designs a
quick service to grab a cheap sandwich for lunch in New York (think of
Subway). This design won’t work the same in every country. Service
designers cannot apply this “design product” the same everywhere. In
Colombia, Subway is not as quick and not as cheap. They try to make it very
similar, but it is not exactly the same.
For other design solutions that include less standard variables such as
preference, cognitive biases, behavior, or culture, I think statistics may
be less useful. I recently tried to do a feasibility study with a
randomized clinical trial to test an interaction design prototype, which
was aimed to help people make better behavioral health decisions. We
recruited people at a health care center. It did not go as expected because
there was high dropout for the second data collection. We had a public
health researcher with us, and we identified that there was a relevant
difference with standard public health studies: we did not have patients
but users. Patients usually adhere to at least clinic visits, users don’t.
And even if users behaved as patients, not everyone has the same preference
for a product or service. Not everyone likes to get lunch at Subway; not
everyone likes to have a Facebook account. Everyone does have the same (or
almost the same) body/anatomy so that an HIV intervention can be
generalizable to the whole world population. But not everyone has the same
preferences, taste, culture; these are critical in design.
My thinking now is that other methods such as case study research may give
us better evidence for explanatory studies in our field.
Best wishes,
Mauricio
--
G. Mauricio Mejía, MDes, PhD
Associate professor in design
PhD in Design and Creation program director
Universidad de Caldas, Colombia
https://co.linkedin.com/in/gmmejia
--
--
La información aquí contenida es para uso exclusivo de la persona o entidad
de destino. Está estrictamente prohibida su utilización, copia, descarga,
distribución, modificación y/o reproducción total o parcial, sin el permiso
expreso de la Universidad de Caldas, pues su contenido puede ser de
carácter confidencial y/o contener material privilegiado. Si usted recibió
esta información por error, por favor contacte al remitente o al correo
electrónico
[log in to unmask]
, borre el material y por ningún motivo haga público su contenido.
La Universidad de Caldas no es responsable por la información contenida en
esta comunicación, el directo responsable es quien la firma o el autor de
la misma."
-----------------------------------------------------------------
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------
|