[Sigia-l] sub-groups in card sorts?

[Alexander Johannesen]

On 9/23/05, Eric Scheid <eric.scheid at ironclad.net.au> wrote:
> It's been suggested to me, by participants, that it might be useful to allow
> sub-groups in card sorts. I certainly can see the utility when it comes to
> pinning synonym terms together, but I'm wondering if the general use might
> be a self-defeating case? That is, since card sorting analysis is designed
> to generate a hierarchy as a result of examining the sortings by multiple
> participants, would inputting a hierarchy skew the results?

Some times there are greater things than IA and logic and reason
involved in creating these structures, most noteably culture, so yes,
sometimes there are perfectly good and valid reasons why there must be
some fixed/semi-fixed top-level categories in place which you can
refine further through card-sorting.

As to the card-sorting results; what exactly is a skewed result? There
are no perfect answer even if you do a full-blown card-sort on things.
Things we do through our IA tools are never perfect, only indications
of what might be good. Unless you're a die-hard philosopher with
concrete principles about this, I wouldn't worry too much; you can
well find more interesting and correct results through fixed top-level
categories than without, even if it goes against the grain of some
perfect idea.

The granularities we find through card-sorting is actually quite
interesting, where you have to balance the theoretical best solutions
with sub-optimal fixed ideas. But I'm sure DonnaM can fill us in on
this; I seem to recall her saying something about fixed anchors in
card-sorts and their interesting implications?


Alex
--
"Ultimately, all things are known because you want to believe you know."
                                                         - Frank Herbert
__ http://shelter.nu/ __________________________________________________