Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!orion.oac.uci.edu!ucivax!gateway
From: msw@unix.cis.pitt.EDU (Matt S Wartell)
Newsgroups: soc.feminism
Subject: Re: equal pay interesting statistic
Summary: correlation is not causation
Keywords: statistics causation correlation salary
Message-ID: <73709@unix.cis.pitt.edu>
Date: 30 Dec 90 21:39:53 GMT
References: <93640@aerospace.AERO.ORG> <1990Dec8.214629.14521@cbnewsk.att.com>
Organization: University of Pittsburgh
Lines: 24
Approved: tittle@ics.uci.edu
Nntp-Posting-Host: zola.ics.uci.edu

Jim Baranski cites a radio program:
   While women in general make ~$.69 for every ~$1.00 that men make, on the
   same scale, married men make ~$1.30, and single women make ~$1.10...

Susan Hallander responds:
   ...                        Could it be that the raise-dolers figure
   that single men don't need as much money as their married counterparts
   because they aren't supporting a family, and single women need more
   money then married women because they aren't being supported by
   anyone. I bet raise-dolers are generally married men.

Susan assumes that the rates of pay are _caused_by_ the gender and
marital status of the worker; the statistics above are correlational,
not causal.  An alternate hypothetical explanation could be that
married men seek training and positions that allow them to support a
family and single women do the same in order to support themselves.
Married women may opt for less stressful, lower paying jobs because
they have the second income to augment their own.

I am not asserting that there is no gender bias in salary, just
presenting a statistical caution.  (Remember 95.378% of all statistics
are made up).
--
matt wartell, university of pittsburgh             msw@unix.cis.pitt.edu