Re: help please/rising to the concrete

From: Mike Cole (mcole@weber.ucsd.edu)
Date: Tue Jul 08 2003 - 10:19:41 PDT


You raise a very interesting point, Bill, and one which I wrestle with a
lot.

First, what fascinated me when we were attempting to code things such as
"needs help" (in *The Construction Zone* or in our analysis of the kids
behavior in the cooking club, only part of which ever made it into print)
is the way that coders working together could come up perfectly reasonable
categories (such as needs help) or coding by speech act categories (in
the study that took kids to the supermarket and brought them back to the
preschool classroom to talk about their trip) and be able to convince themselves
that they were consistent and "objective" (nice term to think about in CHAT
discussion "context"!). Then I would have them (where them was sometimes
me and someone else) either code with another person or by themselves and
find that initially there was good reliability with the initial partner, but
that over time, and as they encountered new transcripts, they drifted apart.

I interpreted the relevant section of Victor's note to be referring to such
a process.

While conducting this work our collective of researchers included David Roth
who had worked with Aaron Cicourel and Bud Mehan, both of whom did micro
ethnography and at least in Bud's case, were working in an ethnomethodology
tradition precisely to get at the issue of member's point of view which, so
far as I can tell, was never a part of Bronfenbrenner's practice, despite
his theoretical/methodological position.

I believe that while one has to maintain scepticism in all such effort, they
can prove useful depending upon the pattern of results. So, for example,
when coding speech acts in classroom and supermarket, we obtained an interesting result. The mean length of utterance was significantly longer in the super
market than the classroom. This was the sort of evidence that Labov was then
using to beat on Jensen and cultural deprivation theorists. But when we
broke down MLU by speech act, we found that what really shifted was the kind
of speech act and the MLU within speech act was more or less invariant
across settings. This leads to a very different conclusion, e.g. that there
is an identifiably different structure to classroom discourse than the
discourse setting of a trip to the supermarket (in this example) and that
this form of discourse makes demands on little kids that they have to learn...
and were not being taught... in the classroom. There is stuff about this
in cultural psychology and the original refs are available.

In our current work the students who write fieldnotes about their participation
in the 5thD are actually co-constructing the interactions they record in their
notes. I believe this makes them priveleged observers and there are times
when the results are very plausibly ecologically valid.

But then I want to go beyond individual cases and find out, for example,
whether I can document a process of inculturation into the 5thD. That means
reading lots of fieldnotes and categorizing them with respect to theoretically
relevant behaviors that are indices, presumably, of enculturation.

Does this sound time consuming?
And does it pose serious difficulties when one has 15 years of fieldnotes
written by 45 generations of undergrads, site coordinators, visitors, etc?
And does it solve the problem you pose?

I believe you have hit on a methodological issue of central importance
which has been lurking in all of our discussions about subjects, agents,
objects, etc.

My solution is to engage in a variety of ways of representing what happens
in an idioculture and hope that they cohere. Sometimes they do.
mike



This archive was generated by hypermail 2b29 : Fri Aug 01 2003 - 01:00:08 PDT