Date: Wed, 18 Jun 1997 19:02:42 -0400
From: Thomas F Brown <tombrown@jhunix.hcf.jhu.edu>
Subject: Re: ratings
To: socgrad@csf.colorado.edu
Message-id: <97Jun18.190244edt.1485-6@jhunix.hcf.jhu.edu>
Content-transfer-encoding: 7BIT

>Thank you for an excellent post! You've answered a great deal. It's great
>to have some data besides just anecdotal evidence. 

Better data already existed in Baldi's article. She answered the question
fairly definitively: hiring in academia is a function of prestige and
professional accomplishment. Is this really in dispute? Baldi found
that nothing else was even statistically significant in her sample.

>One danger with
>anecdotal evidence is that people are sometimes more likely to recall or
>bring up "dramatic cases" rather than typical ones.

 I think that the first thing you should do is look at the outliers and
ask why they differ. This is not a "danger", it should be standard practice
in my opinion. That's exactly why I brought up race and sex to begin with,
to try to explain an outlier in which prestige and accomplishment did not
seem to have been very important.

I also mentioned personal network connections as a possible factor, and
I should have mentioned physical attractiveness, speaking ability,
interpersonal skills, and how the applicant's specialty fits in with
departmental goals.

I would guess that all of these factors could explain some of the residual
variance left over once we account for prestige and accomplishment. The
thing is that race and sex are easy to measure, while the others all require
a lot more work to measure. I think this is one reason why race and sex
get more attention. Also, many sociologists are prone to overemphasize the
importance of sex and race as a result of their ideological perspective. 

>        Without knowing the data on the demographics of supply, of course, its
>hard to make definitive conclusions but it looks at this point that your
>data tenatively lean more towards my second of the four scenarios.  That
>is, hiring at lower levels does not appear to significantly favor or
>disfavor any group (i.e. it's a lousy job market no matter who you are),
>but that the further up the ladder one goes the more white male priviledge
>becomes an advantage that favors white men at the expense of others.  Of
>course we need more studies like yours that dare to attempt to answer such
>questions before we can really say this with any confidence.  I'll throw it
>out as the closest of the four for now.  

Rememeber that you are speculating about leftover variance that is not
even statistically significant. This means that "anecdotal", case by
case analysis is probably the only way to find out what's going on
in any given situation.

> I think there is good reason to deduce from your findings, your suggestion 
>that most minorities (i.e. those channelled into studying race/ethnicity)
>and most whites (i.e. whites who are not specializing in race/ethnicity) do
>not have much overlap in competing for the same positions.

There's no reason to try this unless you have data about the demographics
of supply.

  It's perhaps
>the case that a good number of minorities are channeled into "racial/ethnic
>studies" (a specific "intellectual neighborhood" set aside for them) where
>they are segregated away from the center of the discipline (generic theory
>about human 'social actors').  

Who is channeling? How does the process work? I have never encountered
anything like this, but it would be interesting to see some examples.

>I think it's unfortunate if, in fact, job
>committees are assuming 1) that minorities should NECESSARILY study
>race/ethnicity and 2) that applicants for race/ethnicity positions will
>necessarily be minorities.  

Well, it wouldn't be hiring committees that decide who should study what,
since they only come into the picture once the course of study has been 
chosen. As for #2, the demand for minorities teaching race/ethnicity
can come from undergrads themselves rather than from the hiring committee.
I have seen students protest along these lines at several schools and
heard about similar actions at many more.

>I'm particulary interested in the implications of this finding:
>
>*Minorities tend to be hired in the area of race/ethnicity.  Only women
>were hired in the area of gender, and women are underrepresented among
>hires for the area of theory.
>
>The implicit messages that this structuring seems to send are that: 1)
>Minorities are best qualified to study race, since minorities are more
>"raced" than whites.  2) Women are best qualified to study gender, since
>women are more "gendered" than men.  3) (white) men are best qualified to
>study general (generic) social theory since they are "generic" social
>actors, while women and minorities are "specialized" (gendered and raced)
>types of social actors. 
>
>Whether its self selection or hiring practices is unclear, but I find it
>"problematic" if certain groups are seen as automatically the experts in a
>certain specialty area.  Even what appears to benefit minorities and women
>on the surface may, in the end work against them. It seems to me that the
>discipline to some extent funnels minorities into studying and theorizing
>about "minorities" and women into theorizing about "women" (though this is
>less true than the first) but white men are not funneled into talking about
>the group-specific experiences of "white men."  Instead, white men tend to
>serve as a "false generic" (as Minnich calls it) for what a "social actor"
>is and tend to wind up theorizing more about "agents," "social actors,"
>etc.  This higher level generalizing is generally "more prestigious"
>because it is considered broader in scope.  Just a thought: how many
>ethnographies consisting of exclusively African-American subjects refer to
>their subjects as "social actors" and makes generalizations about humanity
>rather than "African-American culture?"  How many ethnographies with white
>subjects imply their subjects are representative of "white America" or
>generalize specifically about "white culture" rather than about general
>social actors situated in a certain structural setting?  
>
>Sorry to add a new tangent, but just couldn't resist commenting on that
>finding.

It is a very interesting tangent.