Re: Bestseller

Michael Cahill (MCBlueline@AOL.COM)
Sat, 5 Oct 1996 11:42:44 -0400

In a message dated 96-10-04 15:36:15 EDT, I wrote:

<< Aunger's melding of the more scientific comparative event history
method with the softer relexive analytical approach represents an effort to
overcome the science/art split within the ethnographic enterprise. It's
interesting to me that while the fusing is done so methodically (so
scientifically?), the final product can't seem to shake art's lead: "the
drawback is that the comparative event history approach examines data
suggested by the reflexive analytical results, which is a
less-than-perfect-melding, he feels. " Ironic, no? >>

I just read the original article. Actually, the reflexive analytical
approach is a formal method for removing ("cleaning") some of the bias
involved in data collection (e.g., "artiness" or "intuitional" selection in
the absence of knowledge of things like situational factors and
interviewer/informant characteristics and interactional features, all of
which affect informant reponses in identifiable ways). Wade's summary makes
this clear, but I missed the significance of it.

The problem seems to lie in how to combine (and recombine) models of the
"bias" (now seen as variability) -- e.g., the reflexive analytical approach
-- with models for the "bias" (causes of variability) -- e.g, the comparative
event history method. To wit:

"Further discussion of the...example on food avoidances in Zaire may help
clarify how the reflexive and comparative approaches could be used together
in ethnographic research. [In a prior study], I used the relexive analytical
approach to clean the data and to determine the relationship between
variables. At the same time, a related analysis using the reflexive
analytical approach showed that, _ceteris partibus_, better-schooled
individuals had greater knowledge of the food-avoidance domain than
unschooled individuals. The question therefore arises, is this knowedge
difference due to cognitive changes brought about by the training experience
of going to school, or does it represent the influence of some difference in
life history correlated with having gone to school?

To answer this question would require a deeper understanding of the
psychological and social causes behind differences in knowledge of culturally
sanctioned beliefs in Zairian society. Thus, case-based narrative techniques
could be used in a second step to examine the correlation between schooling
and increased knowledge. Because considerable life-historical information
was collected on each individual in the initial fieldwork, the same data set
used in the reflexive analysis could be used to assess the types of life
history that result in the learning of particular classes of avoidances.
This analysis might demonstrate that educated individuals exhibit an
increased knowledge of food avoidances from exposure to a wider range of such
beliefs when they leave their natal villages to attend schools in more
cosmopolitan areas. Further, in yet a third analytical step, individuals
might be classified by whether or not they went to school away from their
natal villages. This difference could be included in a relfexive-style
statistical model as an additional variable. Any new relationships hinted at
by this second reflexive analysis could then be examined by again turning to
a comparative event history analysis, *each step thus successively probing
deeper and deeper into the causal universe in which people live" [my
emphasis]. Aunger _CA_ 36(1):109, 1995.

Aunger seems to say that these combinations look easy enough, but that
there's really a disjuncture between the two methods that's hard to bridge --
"comparative event history does not *directly* make use of reflexively
cleaned data; rather, the relationships it examines are those *suggested* by
the reflexive analytical results [my emphases]. So, there may be room for
various kinds of error ("bias") after all. (And would the "arty" intuition
of the ethnographer play a role in ferreting it out?)

In a footnote, Aunger allows two other caveats: "First, ethnographic projects
have a great variety of goals. My proposal is restricted to ethnographic
research devoted primarily to the explanation or description of ethnographic
realities. [?] Ethnographies designed to capture the experiential nature of
fieldwork or life in the culture under study would have to rely on
alternative methods and presentation styles. Second, comparative-event
history methods are but one type of case-based formal procedures, restricted
to the analysis of even sequences; similar techniques are available for other
types of data." (p. 111)

Alternative presentation styles I can see [this article is a thicket of
words], but I see no reason why the methods and results could not profitably
be used by ethnographers interested in capturing the field experience and the
life of the culture. It seems to me that a lot of what the article
formalizes goes on in less formal ways in the research of most ethnographers.
I'm not sure that Aunger would take issue with this.

These methods might even fit into the research scaffolding of a bestseller :)

Mike Cahill