Xlist stats

From: Eva Ekeblad (eva.ekeblad@ped.gu.se)
Date: Fri Jan 28 2000 - 06:34:28 PST


Hi all

Elizabeth's question last week, about the rate of contributors over
subscribers by gender made me go back to my files -- to my horror I had
discovered that the messages-contributors-subscribers table that I had sent
to Mike Jan 1st was an old one from November 97, picked up later at some
unfortunate turn of forgetting, and complemented with data from 98 and 99,
without noticing that the figures for 97 were prematurely closed and the
figures from 96 didn't include a batch missing from the online archives,
which I later got complemented with the help of first Judy D and then Rick
Smith at the LCHC. Embarrassing: shows how I don't "see" numbers when I
cannot work with the formulas in Excel and make graphs to look at. I'm
always amazed and impressed with people who can read, with ease, codes I
can only negotiate bit by bit: sheet music or numbers or...

This has taken some time, partly because I decided to go over my database
and get it all linked up -- 98 annd 99 were in a separate format from the
others -- and partly because I wanted to put graphs and tables on the Web.
So now you can look at the stuff, graphs and more figures-in-tables, at
http://www.ped.gu.se/ekeblad/xmca/xmca.html There is also some more
explanation about the limited availability of lists of subscribers there.

The following table is the one Elizabeth asked for: contributors over
"present" by gender. I chose to label it "present" as it includes cartain
participant categories not on the subscriber list for the year in case (and
for those years where I have no list):

         1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
иииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииии
Fcon/Fpr 0,64 0,60 0,44 0,45 0,38 0,35 0,51 0,81 0,76 0,55 0,49
Mcon/Mpr 0,61 0,70 0,56 0,57 0,46 0,37 0,57 0,85 0,83 0,53 0,52

What stands out here is mostly how the rate of contributors over
people-present in the activity system follows the changes in the
subscription procedures, that I have mentioned in an earlier posting.

         1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
иииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииииии
Fms/Fcon 3,67 3,29 3,83 4,83 4,51 3,93 5,14 6,32 9,96 7,97 11,18 10,49
Mms/Mcon 5,13 7,92 8,31 9,96 8,02 8,21 9,82 10,00 13,51 9,99 14,74 21,20

The figures for messages over contributors by gender show a gender gap,
with men posting, on an average, more than women -- curves running roughly
in parallel, following the same fluctuations in the mailflow. Now, this
should probably be seen together with the very stable "Zipf law" (1/f)
phenomenon of "a few posting a lot, a middling number posting a middling
amount, and a large number posting little" -- a small turnover, genderwise,
among the very frequent contributors will make quite a lot of difference in
the overall distribution.

You know, if anybody out there uses these data for any purpose, I'd be
happy to have some feedback. Perhaps I'm totally headless even to be
handing out raw data like this?

Eva

******************************************************************************
This work was supported by
The Swedish Council for Research in the Humanities and Social Sciences (HSFR).
Opinions are those of the author and are not necessarily those of the Council.



This archive was generated by hypermail 2b29 : Tue Feb 01 2000 - 01:03:25 PST