Date: Sun, 14 Dec 1997 12:41:19 -0800
To: Editors of the Math Forum Internet Newsletter
From: Ray Olszewski
Subject: Re: Newsletter: Math Forum Internet News No. 2.50 (Dec. 15)
Cc: Cynthia Lanius

The following report was interesting enough to prompt me to connect to the
URL right away.

At 10:56 PM 12/13/97 -0500, you wrote:

    Although the numbers of women in most areas of science and
    mathematics are increasing, the number of women choosing
    computer science as a career is actually decreasing.
    Cynthia Lanius offers a page of tips for "Getting Girls
    Interested in Computers," with links to sites on gender
    equity in technology:
  
    http://math.rice.edu/~lanius/club/girls.html

Although this is a fair paraphrase of what Ms. Lanius says on her Web page,
it is not a very good summary of the research she is actually citing. Her
page provides a link to the underlying research (URL
http://www.nsf.gov/sbe/srs/databrf/sdb97326.htm), and the results there are
reported quite differently. The actual numbers are in a gif-image table, but
the text describes them (accurately) as follows:

    However, awards in mathematics and computer science declined among both
    men and women during the entire period. The declines in mathematics among
    women were not as great as those among men. The percentage decline in
    computer science was much larger among women (51 percent) than among men
    (28 percent) from 1985 to 1995. 

Note three differences between your summary and the actual data:

1. The data are for BA degrees, not career choice. They are not the same,
and the computer industry in particular has a long tradition of employing
people with degrees in areas far from computer science.

2. There is a substantial decline in degrees awarded to men as well as to
women. Although the decline among women is greater, both numbers are large
enough to cause me to think first about what is causing the overall decline
and only second about the source of the difference between the sexes.

3. There is a parallel decline in mathematics degrees for both sexes (albeit
a smaller one).

Although the data are too sketchy to be sure, my first guess is that we are
seeing mainly a change in how people prepare for careers in computer
software development. As the field has grown (quite rapidly over the decade
spanned by the observations, I'd imagine), I believe that computer science
has diminished as a training path, at least at the undergraduate level, in
favor of other degrees loosely grouped under the heading "software
engineering." Women may or may not be participating as much as men, but the
number cited in the underlying research do not really help us answer that
question.

In looking at the overall data, I would be considerably more concerned by
the extraordinarily low number of women receiving physics degrees, and by
the declines in degrees for both sexes combined in physics, chemistry,
mathematics, and engineering. In fact, only biological sciences, psychology,
and "social sciences" show substantial growth. 

In making these comments, I don't want to imply an indifference to the
preparation of women to participate in the sciences. I share the concern of
many that women need both equal opportunity and equal preparation for
careers in science, mathematics, and computer-related fields. (Though I do
have some concern about steering women in this direction at the same time
that men's participation is dropping; given the traditionally greater career
orientation of men, I fear their behavior is a signal that these professions
have become increasingly unrewarding career choices. But that is another,
and bigger, issue.) 

Most of the concrete suggestions Ms Lanius offer on her Web page are good,
practical suggestions for increasing the involvement of girls with computers
in productive ways. But when advocates of better preparation use data
carelessly, they undermine their own efforts. I think this is such a case.

------------------------------------"Never tell me the odds!"---
Ray Olszewski                                        -- Han Solo
762 Garland Drive
Palo Alto, CA  94303-3603
650.321.3561 voice                               
650.322.1209 fax                 http://www.comarre.com/ray.html
----------------------------------------------------------------

Date: Sun, 14 Dec 1997 15:37:27 -0600 (CST)
To: Ray Olszewski, Editors of the Math Forum Internet Newsletter
From: Cynthia Lanius
Subject: Re: Newsletter: Math Forum Internet News No. 2.50 (Dec. 15)

I agree with the writer that there are many reasons for much grave concern
in the report that I referenced, but I quote below the opening statement
from NSF on their report,

        "A record number of science and engineering (S&E) bachelor's degrees
were awarded to women in 1995. Women accounted for 47 percent of S&E
bachelor's degrees awarded in that year. 

        The number of women graduating in S&E was 47 thousand higher in 1995
than in 1985 while the number of men graduating in S&E was similar in both
years (but with a decline in 1990). Over this period, the greatest growth in
S&E bachelor's degree awards to women occurred from 1990 to 1995 (table 1)."

I quote from Table 2 Bachelor's Degree Awards in S/E fields, by sex: 1985,
1990, and 1995, citing the 1990 and 1995 figures for women:

S/E Fields                     90        95
Astronomy                      43        52
Physics                       679       675
Chemistry                    3324      4233
Other Phys Sci.               273       332
Earth, Atmospheric Sci.       775      1524
Mathematics                  6811      6491
Computer Sci.                8374      7063
Engineering                  9973     10950
Agricul. Sci                 2992      5637
Biological Sci.             19402     29918
Psychology                  38619     52963  
Social Sci.                 48740     56093

So my statement that "the numbers of women in most areas of science and
mathematics are increasing," seems accurate from this table. I do not mean
to imply that the problem is solved. He is right that I should add to my
paper that the figures are for bachelor degrees awarded. He is also right
that other significant questions need to be addressed. I was especially
interested in addressing the drop in computer science. Since 1985, it
actually dropped from over 14 thousand to the 7,000 figure. 

I would certainly appreciate further correspondence if you still believe
that I've used the data carelessly.

Thanks,
Cynthia Lanius

Date: Sun, 14 Dec 1997 14:21:52 -0800
To: Cynthia Lanius
From: Ray Olszewski
Subject: Re: Newsletter: Math Forum Internet News No. 2.50 (Dec. 15)
Cc: Editors of the Math Forum Internet Newsletter

Thanks for copying me on your reply. As to further correspondence, I would
only note that you do not say anything about two of the three concerns I
raised, namely:

 -- with regard to computer science, you cite the drop in women's degrees but
    ignore the substantial (though smaller) 28% drop in men's CS degrees. I
    think this omission encourages a reader to draw an incorrect
    inference from the presentation, namely that the drop in CS degrees
    is specific to women.

 -- women (and men) also show a decline in mathematics. So while it is 
    true that women's numbers show growth in science, lumping together
    science (up) and mathematics (down) obscures the result for 
    mathematics taken alone. This is important in context because, at 
    least traditionally, Computer Science degrees are basically applied
    math, and the CS declines mirror similar, though smaller, declines in
    math. Mixing in science (particularly a definition of science
    that includes psychology and social science) obscures thise parallel 
    in the data.

Thanks again for your reply, especially for such a quick reply on a Sunday.

Date: Mon, 15 Dec 1997 04:49:38 -0600 (CST)
To: Ray Olszewski, Editors of the Math Forum Internet Newsletter
From: Cynthia Lanius
Subject: Re: Newsletter: Math Forum Internet News No. 2.50 (Dec. 15)

Hello,

I have re-written the opening of the report - adding the point that the
number of men's cs degrees also decreased. (Although I still believe that is
not the issue, especially since they decreased in so much smaller number.)
But I certainly want to paint an accurate picture.

Thank you,
Cynthia Lanius