A Decade in the University Pipeline

By Tracy Camp

Accompanying Tables

The number of women entering the computer science pipeline and earning B.A./B.S. degrees is decreasing, which is of extreme concern because fewer women entering the pipeline means that, over time, fewer women will go on to earn their Ph.D.s in computer science.

The 1996 CRA Taulbee Survey, which was published in the March CRN, included results on the gender of bachelor's and master's recipients for the Ph.D.-granting departments that responded to the survey (Table 4 in the survey). According to this table, only 16% of the recipients awarded bachelor's degrees during the 1995-96 academic year were women.

Although 16% is extremely low, it is not substantially different from previous surveys; in the results of the 1994 and 1995 surveys, the percentage of women awarded bachelor's degrees was 18%. Thus, 16% does not sound so bad.

Or does it?

Instead of only considering the gender of degrees awarded at departments responding to the survey, I will focus on how well women are doing across all departments awarding B.A./B.S., M.S. and Ph.D. degrees in computer science. In the three tables accompanying this article, the information on numbers of degrees awarded in computer science during the decade 1984-94 is from the Education Department's National Center for Education Statistics. (The number of degrees awarded during the 1994-95 academic year is not yet available.) The center classifies computer science departments within the computer and information sciences (CIS) category.

Table 1 shows the percentage of students within different fields of study in CIS for B.A./B.S., M.S. and Ph.D. degrees awarded. As illustrated, the bulk of the degrees awarded are in the general computer science category, with information science and systems a distant second. Ph.D. level looks good.

At the Ph.D. level, there is good news. Table 2 lists the number of Ph.D., M.S. and B.A./B.S. degrees awarded in CIS during 1984-94 and the percentage of female recipients.

The number of Ph.D. degrees awarded in CIS continues to increase. In the most recent academic year available (1993-94), the percentage of Ph.D. degrees awarded in CIS to women was at its highest level since 1988-89 (the year when the percentage was at its highest level). Also, the percentage of Ph.D. degrees awarded in CIS to women in 1993-94 is 50% higher than it was a decade ago (15.4% versus 10%). At the M.S. level, the news is mixed. The good news is that the number of M.S. degrees awarded in CIS continues to increase. The bad news is that the percentage of M.S. degrees awarded in CIS to women has steadily decreased since 1990. Furthermore, 1993-94 had the lowest percentage of women awarded M.S. degrees during the 1984-94 period.

At the B.A./B.S. level, there is only bad news. Table 2 shows that the number of B.A./B.S. degrees awarded in CIS continues to decrease, and this decrease is occurring at a faster pace for women. This decrease is especially upsetting when one considers the total population receiving B.A./B.S. degrees. Table 3 lists the total number of B.A./B.S. degrees awarded in all fields during 1984-94 and the percentage of the recipients that were women. As illustrated, the number of B.A./B.S. degrees awarded in all fields continues to increase, and the increase is at a faster pace for women than for men. Thus, even though more women are earning B.A./B.S. degrees, the percentage of B.A./B.S. degrees awarded to women in CIS continues to decrease. This is extremely bad news.

Pipeline shrinkage

The pipeline shrinkage problem for women in computer science is a known phenomenon and is illustrated in Table 2. For example, although 28.4% of the B.A./B.S. degrees awarded in CIS during the 1993-94 academic year went to women, only 25.8% and 15.4% of the M.S. and Ph.D. degrees awarded in CIS, respectively, went to women.

One way to increase the percentage of women at the high end of the university pipeline is to get more women to enter the low end of the pipe. Hence, it is critical that the decrease in the number of B.A./B.S. degrees awarded in CIS seen during the 1984-94 decade be reversed. There is positive evidence that this will occur.

The 1996 CRA Taulbee Survey indicates that the number of B.A./B.S. degrees awarded in CIS should increase; the number of new bachelor's students enrolled in computer science Ph.D.-granting departments increased 40% this year. (The increase is actually greater than 40% because more departments completed the 1995 survey). Furthermore, this 40% increase follows a 5% increase in 1995.

The number of B.A./B.S. degrees awarded in CIS at Ph.D.-granting departments should dramatically increase in the near future. In addition, one would expect to see a corresponding increase in the number of B.A./B.S. degrees awarded in CIS at non-Ph.D.-granting departments.

One question, however, remains. How will the future increase in the number of B.A./B.S. degrees awarded in CIS affect the percentage of women recipients? During 1984-94, the decrease in the number of B.A./B.S. degrees awarded was at a faster pace for women than for men.

Will the increase in the number of B.A./B.S. degrees awarded in CIS occur at a faster pace for women? Or will women continue to lag? The Taulbee Survey does not classify the new 40%-plus bachelor's students in CIS by gender so one can only speculate on what the outcome will be. Crediting the Web

The CRA Taulbee Survey proposes that the Web contributed to the increase in the number of new bachelor's students in CIS, and I agree. The Web has brought more attention to our field than anything in the past.

Furthermore, I believe that women like the Web as much as men do. The Web is a type of computer tool that is used for a purpose or goal, which many women like, and some studies have indicated that many Web surfers are women.

Thus, I expect that the increase in the number of B.A./B.S. degrees awarded in CIS will be at a faster pace for women than for men. To ensure that this expectation does materialize, the number of B.A./B.S. degrees awarded to women should be closely monitored.

If the percentage of women earning B.A./B.S. degrees in CIS continues to decrease, loud alarms should be sounded. A further decrease in the percentage of women earning B.A./B.S. degrees in CIS would be detrimental to the rest of the pipeline and to the community at large.

TracyCamp is a member of the ACM Committee on Women and is an assistant professor in the Computer Science Department at the University of Alabama. Her principal research interests are mobile computing and networking. Contact her at e-mail: camp@cs.ua.edu.