Re: Social Constructionism

Devi (
Thu, 4 Jun 98 13:33:10 +0100

Bob Parks writes:

>Do you know of any attempt to understand language as a socially conditioned
>variable in PCP analyses? This has been an important conceptual difficulty
>for me in implementing a PCP analysis. Its never clear whether (or to what
>extent) PCP methodology leads to a socially conditioned defnition of a
>term, or an individual's own experientially rooted conceptualization. If we
>could use a socio-grid analysis to first clarify the shared dimensions of
>the terms being investigated, and then proceed to unravel a person's unique
>experiential associations, the methodology would move an important step
>toward integration. I'd be delighted if someone has taken up this direction
>of analysis. But if not, can someone take this reasoning a step further?

In the context of performance appraisal, Borman (1983) has used pcp
techniques to explore the fairly common (but to a personnel practitioner,
disturbing) finding (Schweder 1980) that there is more in common between
a word and its work-related synonyms as shared amongst a working group
than there is between the word and the job behaviour it is being used to
describe. Loosely, that we have a clearer, shared understanding of what
we look for in judging a person's job performance, than of how to measure
what we're looking for; more precisely, that there is a greater
correlation between behavioural rating-scale scores and ratings of the
conceptual similarity of the behaviour being rated, than there is between
ether of these two factors and the actual behavious as observed.

As I recall it, Borman's technique involves
a) developing a set of simple adjective-traits which he assumes everyone
understands in the same way (I know, I know, but you've got to start
b) using these to rate a given construct
c) defining shared meaning in different constructs in terms of the
correlation between the ratings involved.

I'm hazy about the detail as it's a long time since I looked at the
paper, but you might find it relevant to your query.

In return, do you or anyone else know of any work in which an attempt has
been made to explicitly identify the _web_ of shared meanings involved in
related words by this, or any similar technique: ie., where the
assumption that the adjective-traits aren't identically understood by all
respondents? (Or is that what a repertory grid is? I get tangled up
between signifiers as internal held, publicly available representations;
the meanings they stand for; and the external signifieds to which both
refer. So, for example, I can't make up my mind if Saussure, who asserts
the arbitrariness of the signifier-signified relationship but draws
attention to the importance of contrast between different signifiers in
understanding meaning, has or hasn't anything to offer Kellian
constructivists- who make inferences about meanings by looking at the
relationship between constructs. When a construct is expressed, as in a
rep grid- not the same thing as the internal construing which it
represents, remember!- is it in fact a signifier in Saussure's sense?)



Borman W.C. (1983) "Implications of personality theory and research for
the rating of work performance in organizations" in Landy F., Zedeck S.,
& Cleveland J. (eds.) _Performance Measureent and Theory_ Hillsdale NJ:
Lawrence Erlbaum.
Schweder R.A. (1980) "Factors and fictions in person perception: a reply
to Lamiell, Foss and Cavenee" _Journal of Personality_ 48, 455-484.