Devi -



Addendum! Other reasons I had for eliciting the constructs in a group
setting and not using the 1-5 ranking include:



Anonymity (more honest constructs with some degree of confidentiality rather
than pressure of one-on-one interview)



And rating bias (e.g. self is rated a 1 on smart but college-bound is rated
a 5, for more intelligent); I had read that it might be easier to rate
oneself as less than or more than in relation to others rather than to
provide a concrete numerical rating of one's intelligence (or other
attributes). Also, I was concerned with getting only middle-of the road
rankings, although that is useable data too!



Maggie







_____

From: devi.jankowicz@... [mailto:devi.jankowicz@...]
Sent: Monday, January 31, 2005 8:42 AM
To: RepGrid@yahoogroups.com
Subject: Re: [RepGrid] New member seeking design feedback on PCT project
rsearching socio-economic status



Dear Maggie,

Some specific reactions to what sounds like a most interesting
topic.
a) I'd suggest you elicit the grids individually rather than in a group
setting. In the latter case, the constructs you obtain are likely to be
shallow, cliched, and insufficiently operationalised. It's worth taking
the extra time to get detailed constructs.
b) Can you say why it is important to you to use just 3 elements?
You are likely to get very few, sufficiently detailed, constructs that
way. Think of a grid as a map of the terrain. The elements are your
triangulation points in mapping that terrain, and there should be
sufficient to cover the ground adequately. 'Adequately' is a relative
word, but at least 6 and up to 12 or so seems about right. The
point ois to be able to provide your interviewee with different
combinations of three at a time, in order to prompt the elcitiation of
_different_ constructs. I should aim at anything between 8 and 15
per interviewee.
c) Then there are the varieties of 'self'. In any study of change, for
example, you might consider a 'self as I am now' and 'the self I
would like to be'.
d) You need to specify how you intend to analyse the consructs.
Presumably some sort of content analysis? Reliability will be an
issue here. Now, if you were to use a 5-point rating scale, rather
than Kelly's original approach of Xs and blanks (which is, by the
way, just a rating scale with 2 points!) there are some quite
powerful qualitative approaches available to you, which utilise some of the
information in the ratings in a simple, uncomplicated way (I know some
people
aren't comfortable with lost of statistics).

If you're looking for a good guide, can I suggest my own book?
Jankowicz A.D. _The Easy Guide to Repertory Grids_. Chichester: Wiley 2004.
0-470-85404-9. Some details are at
http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0470854049.html

Good luck with the project,

Kind regards,
Devi

> Greetings fellow PCPers,
>
> I am working on my dissertation and plan to elicit individual bipolar
> constructs in a group setting. The elements would be the self and two
> other elements in the role "college student" for a grid total of three
> elements. I plan to use the difference method and the single triad
> (the total number in the grid is three) to ask students to identify
> how two are alike (to generate the similarity construct) and how the
> third is different (to generate the opposite pole of similarity
> construct) Then I plan to ask them to indicate with an x how the self
> is construed on the similarity/difference continuum.
>
> I then plan to qualitatively analyze the individually elicited
> constructs to generate aggregated/super-ordinate constructs that apply
> to the group as a whole. I have questions about this approach and
> would love to engage with someone to discuss:
>
> 1) Dyadic vs. triadic implications (use of only one triad, is this
> valid?) 2) Point out a better administration/elicitation script 3) The
> feasibility (and validity) of creating aggregated constructs from a
> qualitative analysis of individually elicited constructs.
>
> The paragraph below is an abstract of my research interest and
> approach. Please respond if you are interested in working with me on
> this; I am also seeking a visiting scholar for my committee! Best
> Regards, Maggie
>
> ABSTRACT
> The relationship of lower socio-economic status (SES) - defined as
> enrollment in a free or reduced lunch program, parent(s) occupation,
> and parent(s) education level - and social class - defined as
> attending a public secondary school with a high percentage of students
> on free or reduced lunch, a low percentage of graduates continuing on
> to college, and a high drop-out rate - and the decision to participate
> in higher education is investigated. The research hypothesis is that
> lower SES students hold specific perceptions of persons
> self-identified as college material and that these perceptions - or
> constructs - differ from the students' personal self-concept. Personal
> construct theory and the repertory grid (Kelly) is used to elicit and
> collect individual constructs. Qualitative analysis of individual
> constructs identifies super-ordinate themes applicable to the students
> as a LSES group. Administration of an appropriate survey as indicated
> from the construct analysis provides corroboration of the initial data
> collection method.
>
>
>
>
>
>
>
>
> Maggie C. Case, MPA, Ph.D. candidate
> Adjunct Faculty
> College of Graduate Business & Management
> College of Business
> University of Phoenix
> Jacksonville, FL
> 904 - 347 - 1365
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
>
>
>
> ------------------------ Yahoo! Groups Sponsor
> --------------------~--> Has someone you know been affected by illness
> or disease? Network for Good is THE place to support health awareness
> efforts! http://us.click.yahoo.com/RzSHvD/UOnJAA/79vVAA/SJYolB/TM
> --------------------------------------------------------------------~-
> >
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>





_____

Yahoo! Groups Links

* To visit your group on the web, go to:
http://groups.yahoo.com/group/RepGrid/

* To unsubscribe from this group, send an email to:
RepGrid-unsubscribe@yahoogroups.com
<mailto:RepGrid-unsubscribe@yahoogroups.com?subject=Unsubscribe>

* Your use of Yahoo! Groups is subject to the Yahoo!
<http://docs.yahoo.com/info/terms/> Terms of Service.



[Non-text portions of this message have been removed]