Dear Mike,

Interesting! Here's some comments, in haste.

> I was wondering whether I should use multi-agency working as the theme
(eliciting constructs in terms of multi-agency working) using people in the team
as elements.

Do you mean named individuals? This could be inhibiting if you
intend to feed inormation back to the team. Unless of course you're
comfortable with personal development / group faciltitation
techniques. An alternative to named individuals would be roles.

>I could then perhaps use concepts from activity theory (e.g.
tools/means; rules; community, etc) as elements against which I
could elicit constructs also in terms of multi-agency working.

That's a second grid. Abstract nouns make for tricky elicitation. Try
it ourself: 'Tools, rules, and community: in what way are two of these
the same and one different?' No. In any case, why are you doing
this? To see how your team construes activity theory? If so, I would
offer them as elements 6 or 7 different theories relevant to their jobs,
and include activity theory among them. Use an Ideal element ('the
most useful theory to help me in my job'). Compare ratings on the
other theories, esp acitivty theory, against the ideal. Focus
discussion on the constructs that contribute most to the difference
between a theory and the idal theory.

I would use effective multi-agency working ,,,,,,,,,,,,,,,,,,, ineffective
> multi-agency working as the donated bi-polar construct.

Fine

I was wondering too whether I should ladder the elicited constructs in each case

Always ladder downwards suficiently to ensure appropriately
specific constructs that avoid cliches and motherhoods.. Laddering
upwards is a different task, a different grid activity, to ascertain
people's personal values if that's what you want to do. Make that a
seperate activity from the basic grid elicitation.

> I then want to co-construct a group grid for individuals to
> use......

What do you mean? Constructs are personal and may not
necessarily be shared between people. Trying to do so would be a
good team-building exercise but the outcome would not have much
use as a standard instrument. If the latter is what you seek, use a
conventional psychometric advice with norms etc. Horses for
courses.

I would then suggest an intervention for improving
> multi-agency working and ask the participants to re-do their grids
> sometime after the intervention.

Sounds very valuable

> Do you have any views on what I've suggested?

As above.
>
> Finally what software programme would yo suggest for the factor analysis of
the grids?

Why on earth factor-analyse?
What is your analysis purpose?

Having said that, a good, free, platform-independent and competely
secure programme is WEBGRID. It offers cluster analysis, principal
components analysis, sociogrids (don't like that myself), and other
goodies. Access details in my book or via my website.

Byee!

Devi



Devi Jankowicz PhD, ABPsS, C.Psychol.
Professor of Constructivist Managerial Psychology
University of Luton
Graduate Business School
University of Luton
Hitchin Road
Luton LU2 8LE
(0044)-(0)-1582-734111 ext 5050
devi.jankowicz@...