Re: New Member from Hong Kong
Fri, 15 Aug 1997 12:39:57 +0000

Dear Robert,

It seems to me that your task as researcher is to ensure that your
elements are sufficiently representative of the domain of interest to
enable you to elicit a sample of relevant constructs. I reckon from
my position of relative ignorance you are well on the way to doing

We expect that elements will be similar and *different* and your
participants seem to be having problems coming up with dimensions of
difference - maybe? This is only a problem if they think you want
them to tell you how they are similar alone.

In introducing the task it may help to suggest that you are looking
for issues relevant to your research question, just like when we ask
participants to contrast elements in terms of 'personality features',
'your relationship with them', their views on ecology' etc...

> Date: Fri, 15 Aug 1997 18:38:52 +0800
> To:
> From: (Robert Wright)
> Subject: New Member from Hong Kong
> Reply-to:

> Dear PCP Group Members,

> My Repertory Grid Question:
> ---------------------------------------
> Presently, after one pilot test after another (3 in total), I still haven't
> been able to "WORD MY ELEMENTS" in such a way so that they appear to be
> comparing APPLES WITH APPLES; rather they apear to be apples with organes -
> as they presently are.
> For example, in evaluating the effectiveness of an appraisal system
> (performance management system), the main elements within my domain would be:
> E1: Performance Appraisal (PA) training
> E2: PA annual interview
> E3: PA communication (including notes and guidelines, brochures)
> E4: PA progress reveiws (done informally throughout the year)
> E5: PA standards & critieria used to appraise performance
> E6: Link to recognition of good performance
> E7: Self-appraisal
> E8: PA form design
> E9: PA objective / work goal setting
> (these 9 elements fundamentally represent the typical employee performance
> appraisal system)
> But my concern is, during my Pilot test Repertory Grid Interviews, the
> respondents had great difficulty in comparing say E2, E5 & E8 (E2=PA annual
> interview; E5=stds & criteria; E8=PA form).
> Onc interviewee objected with great frustration, stating that I was forcing
> him to compare between a piece of paper with what I actually do in the
> interview!!!
> The more I thought about this, the more I realised that my elements may not
> be apples and apples!
> I feel confident that the 9 elements do make up the framework - or the four
> corners - of an appraisaly system. And to appropriately evaluate the
> effectiveness of any appraisal system, I needed to include these elements.
> But I can see what some of the respondents are saying. IT IS DIFFICULT TO
> So my QUESTION IS: Then how can I evaluate a "system" of performance
> (remember that Kelly emphasised that elements must be either ALL people, or
> ALL events, or ALL situations, or ALL objects ...

> I suspect that my problem is with the "wording of my elements".
> Can any one provide me with some insights into what I am doing wrong or not
> doing right?
> Many thanks for taking the time to read about my concern.
> Sincerely yours,
> Robert

malcolm c. cross
department of psychology
city university
northampton square
london ec1v ohb

telephone +44 (0)171 477 8531
facsimile +44 (0)171 477 8590