Re: +ve and -ve values of constructs

bob green (
Wed, 29 Sep 1999 07:09:42 +1000


You have already had some useful suggestions. Regarding your comment:

> This is not an isolated case in my recent experience, but I have
>not seen anything about it in the literature. Has anyone encountered this
>one before?

In practice it is most likely fairly common, I recall some mental health
staff struggling with whether drug use was positive or negative. The issue
most likely was that it wasn't construed as inherentlyly bad, though for
some clients it was disastrous. Then there there was the issue of the their
personal vs professional contruing on the subject. Context and the
instructions given are important in this regard.

Some time ago I asked about the midpoint on this list, a search of the
archives may provide some useful posts, certainly there were suggestions re
splitting constructs e.g., 'works indoors -v- works outdoors' construct,
making both positive poles and eliciting a negative opposite for each.

You might also find the following additional Yorke references of interest.

Yorke D (1978). Repertory Grids in Educational Research: Some
Methodological Considerations. British Educational Research Journal. 4(2),

Yorke D (1985). Administration, Analysis and Assumption: Some Aspects of
Validity, in Beail N (ed). Personal Constructs: Applications in Clinical
and Educational Settings. Groom Helm, London.

Yorke M (1988). Kelly's Eye: An Alternative View of PCT, in Fransella F
and Thomas L (eds). Experimenting with Personal Construct Psychology.
Routledge and Kegan Paul, London.

I can understand the methodological logic of seeking a preferred-most valued
pole (Mackay wrote on this), however if respondents find it difficult to
describe a construct in these terms, you could also look at

(a) the construct may not be useful to discriminate between elements, i.e.,
is it a significant discrimination? Kelly's discussion of concrete
constructs may be helpful. The suggestion to ladder would seem good advice.

(c) it may be possible to ask in some instances for 'the' more positive
pole, however where a client thinks doing this is conflictual with their
view of the world, forcing such choice raises questions of the validity of
any results which are obtained.

> For the sake of the ratings, I encouraged him to pick one over the
>other. He was very resistant, saying "I can pick one over the other, but
>it will NOT represent how I really feel about the elements I rate with
>that construct."


Bob Green