[Sigia-l] Re: Replacement for EZSort

[Bill Bell]

"Trent Mankelow" <trent.mankelow at optimalusability.com> wrote, in part:

> I'm keen to get people's feedback about an idea for a software tool
> that we want to write to replace EZSort. We have just completed a
> reasonably large card sorting exercise (125 cards, 84 participants in
> 29 groups) and analysed the results using EZSort. We ...

1. When assessing the needs of a collection of users that represent a population with limited 
skills, is it reasonable to expect them to be able to use a computerised tool to express 
themselves effectively? Even software developers tend to use physical 3x5 cards when they 
partipate in CRC exercises to investigate system designs.

2. Building a good card sorter is, in my view, about the same as building a good interface for 
manipulating outlines.

(a) I have used, or attempted to use, a wide variety of PIMs, most of which depend in one 
way or another on the outline as a central metaphor for representing what we might 
otherwise keep in our heads. Although programmers find it very natural to manipulate trees 
the usability of PIMs that they create varies widely and I suspect that people vary widely in 
their preferences for them. I prefer the outliner that I built myself for DOS (note: this is a 
feeble joke, even though it's true).

(b) Why re-invent the wheel? How about looking for an outliner that seems particularly well 
suited to the task of card sorting?

3. To return, in a way, to my first point: how about making it easier for those conducting 
studies to create and manipulate decks of cards? 

Print one shuffled stack of 3x5 cards for each study participant. Identify each card in a stack 
with it own unique bar code. After the user has sorted his/her cards into piles, ask her/him to 
label each pile with a fresh 3x5 card.

Spread the cards and read them with a wand. Key the user's labels into the scanned lists.

IOW, put the additional cognitive effort where it should be, in the responsibility of the 
individual performing the study.

Bill