Bruce Weaver <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>...
> Neil wrote:
> > ... Q-Sort Analysis / Q-Methodology, (a type of Factor
> > Analysis), does someone know what the "Q" actually stands for? 
> > 
...
> 
> I initially thought Neil was talking about "Q-mode factor 
> analysis", as did some other respondents.  
> ------------------------------------------------------
> 
> But I don't think this is what Neil was talking about.  The 
> following is from David Clark-Carter's book (reference given 
> below), in a section called "Techniques to Measure Meaning" 
> (p. 97):
> 
> "Q-methodology is an approach to research which was devised 
> by Stephenson (1953). It requires participants or judges to 
> rate statements or other elements on a given dimension or on 
> a given basis.  One technique which Q-methodology employs is 
> getting participants to perform Q-sorts. 

Thanks, Bruce. That clarifies some of my memories. I will not proceed
to further muddle the question.

When I was an undergraduate long ago, just basically running jobs and
writing code for statistical work, I was involved in some research in
which Q-methodology (Q-sorts) were used to collect data from
respondents.

However, the researcher then used Q-mode factor analysis as a way to
cluster the respondents (NOT the items). And, as Bruce discussed,
negative loadings were problemmatic and generally these people were
put into their own cluster -- they liked what some other group
disliked.

As Bruce clearly points out, Q-sorts and Q-mode factor analysis are
two different things. But there were at least some researchers who
were combining them. The projects I worked on were contract consulting
work, so I can't give an academic cite.  I don't remember what "Q"
stood for in "Q-sort" either.
.
.
================================================================
This list will soon be replaced by the new list EDSTAT-L at Penn
State.  Please subscribe to the new list using the web interface
at http://lists.psu.edu/archives/edstat-l.html.
================================================================

Reply via email to