The recommendations are divided into four categories: programs that could encourage women in technical fields, ways for women to build their self-image as scientists, ways for women to deal with biased behavior, and suggestions for men and women who wish to encourage women in the sciences.
Programs that could be implemented to encourage women are discussed by [Keith et al 1990,Leveson 1989,Sandler 1986], and interested readers are urged to read the full suggestions in these documents. Many of the following suggestions are from these sources. While the suggestions are geared to universities, most can be adapted to industry as well. Recommended actions are:
One way an EECS department head did this was by announcing during a graduate orientation session that sexual harassment was unacceptable and that any harassment victims or witnesses should alert someone in a position of authority. See [Rowe 1981,Rowe 1985] for advice on implementing effective programs to fight sexual harassment.
As described throughout this report, being a woman in computer science is unusual in that the majority of one's peers are male and many people, consciously or subconsciously, expect the men to perform better. These expectations are subtly communicated to both males and females. Fortunately, there are many ways that a woman can increase her self-confidence and her self-image as a computer scientist.
A female computer science student often find herself in a classes where she is one of the only women. Additionally, she is likely to have few, if any, female professors. Even in the absence of being treated differently, this is likely to affect a student's self-image and perceptions of women. In reality, females in coed classrooms are usually treated differently from the males:
[T]eachers praise boys more than girls, give boys more academic help and are more likely to accept boys' comments during classroom discussions... While girls sit patiently with their hands raised, boys literally grab teacher attention. They are eight times more likely than girls to call out answers ([Sadker et al 1985] in [Van Nostrand 1990, page 187,]).(See also Section
.) By taking classes at women's
colleges or in women's studies, which almost always have a
large majority of women, female students feel free to more actively
participate in the classroom and are taken more seriously by teachers
and other students. While advising technical females to consider
single-sex high schools and colleges may appear counter-intuitive,
particularly because these schools often have less adequate laboratory
facilities, studies have shown that females at single-sex schools
study more science and mathematics than those in coeducational schools
[Kelly 1982, page 499,], are more likely to continue in science
[Ferry et al 1982, page 27,], and are disproportionately successful
compared to other women [Gilbert et al 1983].
Another advantage of women's studies programs is their emphasis on women's achievements. One female computer science student had the following experience:
When I entered my first women's studies class, on literature written by women, I expected to encounter second-rate works that were only being taught when the ground rules eliminated male competition. Instead, the books were first-rate, and I wondered why I had never read them before. The class taught me that women's achievements had often been overlooked. (I had been pretty misogynic before that.)Because of men's numerical and figurative dominance in engineering classes, a student in a women's studies course can find herself listening to vocal female students for almost the first time since entering college.
Another benefit of women's studies programs is that they document the barriers women have faced. It is important for women to realize that the paucity of women in the field does not imply that women are inherently unable and that negative behavior they encounter may be due to their sex and not to any actual shortcomings on their parts.
Finding female role models and mentors can also be helpful to a woman's self-image. While male mentors are certainly useful, and many women are happy with male advisors, female role models serve the additional purpose of providing a living example of a successful female scientist. One study found that ``[f]emale graduate students who identified female professors as role models viewed themselves as more career oriented, confident, and instrumental than did female students identifying male role models'' [Gilbert et al 1983, page 597,], although, as the report notes, self-selection may also have been a factor [Gilbert et al 1983, page 605,]. As noted in [Leveson 1989, page 20,]:
The importance of mentoring and role models cannot be stressed too much. One female student ... wrote about her experience in a summer job working for a female manager who she described as: `one of the most respected people in the company. I had never expected that having a female role model would change the way I felt about myself, but it did'.Thus, a female student should consider finding a senior woman in her field to work for at school or during a summer job. While a person's sex should not be the primary factor in choosing a mentor, spending some time with a female role model can be psychologically beneficial. (See also [Simeone 1987, pages 104--109,].)
In part because female computer science students so outnumber female professors, one-to-one contact is not always possible. Many schools have get-togethers for women in computer science or engineering, ranging from informal annual meetings to regular meetings of the Society of Women Engineers. Women find these activities useful for receiving encouragement and advice, and they ``generate a lot of good feeling'' [Leveson 1989, page 23,].
Perhaps the most effective group currently existing for women in computer science is an electronic mail discussion group with over seven hundred members which provides women with a ``forum for discussion of both the problems and joys of women in our field and a medium for networking and mentoring'' [Frenkel 1990, page 36,]. Women share advice on female-specific topics such as the best time in one's career to have children, how to dress for conferences, and how to deal with sexism, as well as general-purpose networking. The list is particularly important to women in computer science because they are so dispersed and might not otherwise be able to interact with many other female computer scientists and engineers. Being part of a discussion group with hundreds of female computer scientists can change a woman's default image of a computer scientist from male to female (or neutral) --- quite a difference from a student's academic experience, where two thirds of the computer science departments have zero or one female professors [Gries et al 1991].
In cases of blatant sexual harassment, a woman can usually complain to some authority. In less clear-cut but nevertheless offensive situations, this option is often not open. While acting as a group and writing a report was effective for women at MIT [MIT 1983] and Carnegie-Mellon [CMU 1989], often women must react individually to behavior they find unacceptable.
As discussed in Section , women are often the victims of
subtle subconscious bias. For example, a woman may find that comments
are primarily directed to men in a group, unintentionally leaving her
out. If the woman does not feel comfortable directly confronting the
individuals in question, there are more diplomatic methods of calling
the behavior to the person's attention:
of this report.) After the
meeting, the woman took him aside and mentioned that she had noticed
he had directed his answers to [John] and asked, in a concerned way,
whether she had done anything to cause this behavior. When he replied
that she had not done anything wrong and began to apologize, she
dismissed his apologies by saying that she was just relieved she had
not given him any negative signals. He continued to apologize.
The indirectness of her approach allowed the man to save face while still communicating the point. It appears to have been effective, because, when she saw the man later in the day, he made eye contact and spoke directly to her.
Indirect methods are also often more effective than direct means for
dealing with offensive sexist and sexual comments. As mentioned in
Section , the only consequence of complaining is often
that one is dismissed as a `feminist'. In a recent column in The New
York Times, a woman described her experience as a college journalist:
Though I was militantly middle of the road in perspective, by the second time I mentioned sexism in print I was pegged. People I met seemed to treat me like a Marxist, a radical, a testosterone-fueled male-basher; others asked me why I suddenly hated men so much; and certain folks at home warned me that I was going to jeopardize my future by scaring off potential male suitors [Kamen 1990].Two techniques that are often more effective than anger (even when justified) are role reversal, which involves the substitution of other terms for female terms, and humor. Combinations of these two techniques are especially effective.
Treating men in the manner in which they treat women can be an effective response to poor treatment:
I'm much more often complimented for my dress, my hair, or my accent than for the content of what I say and do. So I just turn the compliments around and tell men how pretty their ties are, or how the cut of their jackets bring out their broad shoulders. They squirm under such scrutiny just as we do [Frenkel 1990, page 41,].
[A few years ago,] I wrote to Time Magazine to register my opinion about a `humor' piece titled `Women Are Getting Out of Hand', which included lines like `Women are getting too big for their britches'. I suggested that they would not be likely to publish articles like `Blacks Are Getting Out of Hand' or sentences like `Hispanics are getting too big for their britches'.
Humor is effective for several reasons. First, when men make offensive remarks, if someone protests, they retort: ``Can't you take a joke?'' When a woman uses humor to protest an offensive remark, she can offer the same response if challenged. Additionally, humor breaks tension and often allows one to point out that something is wrong without a direct confrontation.
One female computer professional told the following stories:
That turned out to be the perfect response --- I got my point across and it broke the tension with laughter. The moral of this story is, that if you can make your point with humor, the message goes down easier.
[W]hen I first starting teaching 13 years ago, I was married at the time and both my husband and I had teaching positions in the same department. My department head was sitting in my class (at the time we were in a joint math and CS department and my department head was retraining). [O]ne day, one of my students referred to me as Mrs. [Smith] in class and I was feeling pretty frustrated by that point since I had noticed that I kept getting called `Mrs.' while my husband was [called] `Dr.', so I looked at him and said `What do you call Karl?,' and he replied `Dr. [Smith]'. So I told him that I'd appreciate it if he would use the same title for me since I had the same degree. Well, he wasn't the type to quiet down quickly, so he retorted, `And what does Karl think of your feeling that way?,' to which I replied angrily, `It doesn't matter what he thinks; it was a helluva lot harder to get a Ph.D. than it was to get married.' Word apparently got out to the students after that because I wasn't called `Mrs.' for quite some time. (Of course, I was shaking by the time class ended wondering what my department head thought of that reaction. He never did say anything about it.)
Female computer scientists have combined the techniques, role reversal and humor, as in the following incidents:
A few years ago I was at a workshop held way out in the country.... There were about forty people there. I don't remember how many were women, but there couldn't have been many.At one point we were sitting around singing, and after a while someone started one of those make-up-a-verse of your own affairs.... It went around for a while, and someone offered,
I know a girl, her name is Jill.
dum-de-dum de dum-dum-dum-dum
She won't do it, but her sister will.
dum-de-dum de dum-dum-dum-dum
Well, I can't carry a tune in a bucket, but I co-opted the next verse:
I know a guy, his name is Bill.
dum-de-dum de dum-dum-dum-dum
He won't do it, but his brother will.
dum-de-dum de dum-dum-dum-dum
Everyone laughed, and I thought the matter was settled. But some people won't give up, and a few rounds later, the same fellow came around with:
I know a girl, her name is Sue.
dum-de-dum de dum-dum-dum-dum
She won't do it for me, but she'll do it for you.
dum-de-dum de dum-dum-dum-dum
Drat! Can't back down now ... I got as far as:
I know a guy, his name is Lou....
when the rest of my colleagues shouted down the offender.
A [campus] career recruitment poster, several years ago ... made quite a splash. The poster portrayed two men in shirts and ties picking up a printout off a line printer, dropping the fanfold paper all over the floor as they ogled at a woman walking by in a miniskirt. The caption: `We think about more than just work here at [Company Name].' Outraged women began tearing the posters off the walls, then (insert light bulb here) they instead made photocopies and put them up in great numbers all over campus. Meanwhile, the recruiters figured out something was wrong and tried desperately to get rid of them. A number of women went to the recruitment talk to disrupt it, and found that they were the only ones there. Several hack posters were created, for example one with two women in businesslike garb poring over a printout while a scantily clad beach-boy type walked by, with the caption, `We think about more than just sex here at [Company Name].' I believe the president of the company issued a formal apology afterward. This one is notable as much for the reaction of the student body as for the poster itself.
Most of the suggestions in this section involve avoiding behavior that was described elsewhere in the report as potentially offensive. Because the reader is assumed to have read the rest of the report, a minimum of justification appears in this section.
As described in Section , many people treat men and women
differently without realizing it. Once people understand that it is
possible for a well-meaning person to unintentionally discriminate,
they should train themselves to oversee or spot check their behavior
to make sure they do not behave in such a manner. Anecdotal evidence
suggests that, once people become aware of subconscious bias, they often
catch themselves behaving in such a way. This is true of women as
well as of men. Fortunately, once a person is aware of the problem,
the behavior is usually easy to change. Examples of behaviors to avoid are:
Additionally, one can help counteract another person's subconscious bias by either discreetly calling it to the person's attention or by directly offsetting it. For example, if a female colleague keeps getting interrupted or ignored, one can express interest in what she is saying without directly telling people to listen to her, which might be interpreted as patronizing.
A female graduate student provided me with some positive examples of how her advisor treated her. In addition to generally treating her with respect, by giving her challenging projects and the necessary resources to complete them, there were two specific things he did that especially impressed her:
While any well-intentioned person avoids behavior that they know to be offensive, some people do not realize that certain behaviors, discussed throughout this report, offend some women. Good intentions are no guarantee that women will not get the wrong message. For example, a woman might feel uncomfortable with a man who has revealing pictures of women on his walls or computer screen. Additionally, by not using language that could be interpreted as sexist, such as ``he'' for the generic computer architect and terms like ``manpower'', women feel more included. One positive example is the name on a sample application for the United States Department of Defense fellowships, included in the instruction booklet: ``Smith, Dana Robin.'' Both ``Dana'' and ``Robin'' can be either female or male names. Another positive example is the alternating usage of ``he'' and ``she'' as the default singular program in chapters of the influential Computer Architecture: A Quantitative Approach, by John Hennessy and David Patterson. There are many guides to ``nonsexist'' communication, such as [Persing 1978].
Someone who supervises a research or work group or organizes its events should try to choose activities at which all group members would feel comfortable. While it is not always possible to choose an activity that everyone relates to --- for example, some people are uninterested in any sport --- an effort should be made to include everyone in at least some of the social occasions. I have seen people repeatedly left out of social activities springing from the workplace because they are poor athletes or disabled, have unusual dietary requirements, or do not drink alcohol. Anyone who cannot socialize with the group will not feel as though they fully belong. (Conversely, anybody who does not want to socialize with the work group, for whatever reasons, should not be made to feel unwelcome as a worker.)
Some professors rarely encourage or praise their students, but such
encouragement, even if just a few spoken words, is greatly appreciated
and rarely forgotten [Widnall 1988, page 1743,]. For example,
suggesting that a student consider graduate school can make a
substantial difference [Leveson 1989, page 23,]. In one survey of female scientists, ``[t]he
encouragement of teachers --- along with that of fathers --- was the
influence most frequently quoted as steering [them] towards science''
[Ferry et al 1982, page 27,]. In practice, women often do not get encouragement from teachers and
guidance counselors ([Cooper Union 1989] in [Baum 1990, page
48,]). While encouragement should be directed to worthy males
and females, it has greatest effect with people who receive little
encouragement and who have low self-esteem, often women [Zappert et al 1984, page
8,]. (See, for example, page of this report.)
Additionally, it is important for parents to encourage their daughters
as well as their sons in technical areas. ``According to one computer
camp director, `Mothers bring their boys to the classes. Girls have
to beg to enroll''' [Hess et al 1985, page 201,]. Even if parents do
treat their children equally, it is impossible to make girls feel that
it's as normal for girls to use computers as it is for boys, with all
the biases in our society (Section ) and in children's
toys and computer games (Section ). Nevertheless, it is
clearly important to encourage one's daughters, and many female
computer scientists attribute their career choice and success to
parents who encouraged them.
Readers will notice that there are many situations for which I made no recommendations. For example, I had no advice for the woman who told me about dining with a group of men after a conference who then began telling jokes that were specifically degrading of women. I also have found myself with nothing to say when a brilliant female graduate student says she avoids a certain research area because of the specific men at the university involved in it. Unfortunately, there is a substantial class of behavior which is both offensive and non-actionable.