Wine and cheese and grids: market research
Sun, 14 Sep 97 22:19:12 +0100

Miguel Sottomayer wrote:

>I'm trying to apply PCP int the marketing research area. Specifically,
>I'm tryng to use repertory grids for the elicitation of product
>attributes and for the identification of personal constuct systems both
>in the context of product choice (wine - retail market).
>The original idea was just to use the rep. grid technique as way to
>elicitate attributes for being used afterwards for the specification of
>a multiattribute choice models. However, I would like also to attempt the
>use of individual grids themself as the basis for consumer segmentation.
>If have you experience in something similar or are simply interested in
>this sort of things please send me comments.


Did you know that there's a major centre for PCP at Reading, your own
university? Contact Prof. Maureen Pope or Dr. Pam Denicolo, the latter
being very closely involved in running the European Personal Construct
Association. They're both at the Bulmershe Campus in Earley, Reading.

You might find the following review article of mine useful:

Jankowicz, A.D. "Applications of personal construct psychology in
business practice" in Neimeyer G. & Neimeyer R. (eds.) _Advances in
Personal Construct Psychology_ Greenwich, Conn.: JAI Press, 1991.

And there's an excellent account of grid technique, and how to "fit" it
to a variety of occupational applications, in:
Stewart V. & Stewart A. _Business Applications of Repertory Grid_ London:
McGraw Hill 1982.

NB Doing individual grids is fine, for the topic you have in mind. But at
some stage, especially if you're wanting to run a taste panel or a new
product development idea generation group, (let alone using the grid as
the front end of the development of a consumer questionnaire) when you'll
want to aggregate the results of several grids. Try:

Honey P. "The Repertory Grid in action" _Industrial and Commercial
Training_ 1977, 11, 11, 452-459.

The only similar work to your own that I've done personally has been in
running a market research taste panel for the development of a new
cheese. It's written up briefly in an early chapter by Loman O'Byrne in:
O'Cinneide B. _The Case for Irish Enterprise_ Dublin: Enterprise
Publications 1986.
You'll probably need to get this through inter-library loan: tell your
library to try Trinity College Dublin library first of all. If that
doesn't work, give me a shout and I'll mail you a photocopy of the
chapter in question.

We used existing cheeses as elements, asking people to taste them in
triads, plus a supplied element, "my favourite cheese"; analysis was a
simple computation of % matching scores for elements, looking at the
match between each cheese and the supplied element over all the
constructs we elicited. Great fun!

(If you're wondering which cheese came favourite: it was Tallegio, an
Italian cheese, and the company went on to produce an Irish version of
that, rather than creating an entirely new cheese. As you can imagine,
with something as biological as cheese or, indeed, wine, it ain't that
simple to create a new product which combines all the construct
characteristics _across_ elements that match the supplied, "favourite"
element most closely!)

Sorry I didn't respond earlier, but I was on holiday until a few days ago
and it's amazing how the e-mails and newsgroups items build up while
you're away...

Kindest regards,

Devi Jankowicz