Women in Computer Science

Women in Computer Science
Final Report - June 1, 1999

Summer Adams / Naisa Ali / Julia Medina / Libby Nix / Bill Rivera

Introduction
We were interested in taking up this study because we noticed that there was a drastic reduction in the number of women entering fields of Computer Science. Comparing those who attended the AP classes during their junior and senior years of High School, which prepare them for majoring in Computing, to the number of women who were actually majoring in Computing at the Institutions we were familiar with. We were not happy with this reduction of about 50% to less than 20% and we wanted to see how widespread it was and what might be its cause. After reviewing the MIT study (http://www-eecs.mit.edu/women-stu.html) we learned that the low percentages at the College of Computing at Georgia Tech were also the norm at MIT and other institut ions of similar caliber (presumably top ten or twenty graduate institutions are inferred by this statement). The MIT study had concluded that one of the main reasons was the perception that women had of themselves with regard to their preparedness for a grueling freshman level programming course along with some negative perceptions about the field of Computing in general. However, their study (and they were the first to admit it) led to several further questions and did not seem to provide a complete ex planation for the reduction of the number of women in the field of Computer Science. We decided to follow their survey technique but attempted to get at other possible root causes such as lack of encouragement from peers, parents, teachers, and counselor s both in High School and in College (see Methods).

Research
A recent survey done by the Gallup Organization in conjunction with CNN, USA Today, and the National Science Foundation included 744 children in grades 7 through 12. "Telephone Interviews were conducted from March 20-27, 1997 from Gallup in terviewing centers throughout the country. The focus of the survey was on students' familiarity with and use of modern technology with special attention given to use of computers and the Internet." (Gallup 1997) Some things that they found were that "Am erican teenagers are very interested in and reliant on modern technology." The paper gives a lot of information about findings regarding students usage of computers and other household technologies. The somewhat surprising facts to us came with the find ings on comparisons of boys and girls. The similarities between boys and girls were staggering. "Boys and girls express nearly identical levels of confidence in using computer, with average scores (on a zero to 10 point scale) of 6.8 and 6.7 respectivel y. ... One-third of both boys and girls feel their computer/technology education is on track..."(Gallup 1997).

Unfortunately, these numbers signifying gender equality do not represent the trends seen in boys and girls entering fields in computer science. According to Cynthia Lanius, author of an article entitled Getting Girls Interested in Computer Science, th ese results are encouraging because "increasing girls' computer use may be necessary to increase girls' interest in computer science". However, she also feels that this is not enough. "Computer science (which really should be called computing science) i s the study of computation; computers are merely the tool performing the computation." (Lanius) Her basic conjecture is that saying that girls are using computers or even are comfortable with computers just as much as boys means nothing to how it will eff ect the computer science field. The fact is the number of girls entering computer science is dropping and we would like to figure out why.

A data brief written by Susan Hill from the National Science Foundation says that there was a decrease in the number of bachelor's degree awards in mathematics and computer science between the years of 1985 and 1995. It also says, "the percentage decl ine in computer science was much larger among women (51 percent) then among men (28 percent) from 1985 to 1995." (Hill 1997) Hill also notes that only 29 percent of the computer science bachelor's degree recipients were women, as opposed to the 47 percen t in mathematics and 42 percent in chemistry.

This all leads to a very interesting article by Tracy Camp called The Incredible Shrinking Pipeline. The pipeline that Camp is referring to "represents the ratio of women involved in computer science from high school to graduate school." (Camp 1997) She finds that "while the percentage of bachelor's degrees awarded in CS to women decreased, corresponding percentages of other science and engineering disciplines increased." (Camp 1997) One of Camp's reasons for such great concern in this matter is the critical labor shortage in the computer science field and "Although women are more than half the population, they are a significantly underrepresented percentage of the population earning CS degrees." (Camp 1997) She also found that the number of women g raduating in computer science rose to 37.1 percent in 1984 and in 1994 had dropped to 28.4 percent. The thing she feels makes this even more shocking is the number of women in similar disciplines as CS has increased. In fact, "CS is the only science and engineering discipline where the percentage of bachelor's degrees awarded to women decreased." (Camp 1997) All of this is showing to us that some discovery needs to be made as to why these facts are true. We intend to show that this is greatly being ef fected by the influences women are exposed to in high school.

An article written by Ilana DeBare talks about the issues of female students in computer science high school courses. She takes a look at several different high schools that have computer science courses or labs and discovered in most of them there we re significantly fewer girls than boys. "...In a wide variety of schools and classrooms, the story is the same: Girls are doing great with computer applications. But they're not learning the skills that would someday allow them to design those applicati ons." Why is this is a question she poses, one of the answers she comes up with is girls' perception of what a computer scientist is. "The dearth of girls in computer classes has much more to do with stereotypes of what are acceptable "male" and female" activities -- and the broader problem of girls shying away from math and science in general." Another answer was the image of the computer scientist as a "nerd". She also notes that for those girls that do go into computer science find that they are al one in their field. It may make them feel they don't belong there. DeBare feels that we must push women harder into these fields because they won't be able to get past the stereotypes to get there on their own.

Dr. Anita Borg, President of the Institute for Women and Technology writes an article that also explores why there are so few women in computer science. She feels the decrease in the number of women going into these fields is mainly because of the inf ormation age hitting for two reasons. First how it has effected schools, "Girls and boys got through most of secondary school without experiencing unequal participation with computers. Today there is both an imbalance in the availability of software tha t appeals to girls and few attempts to address boys' tendencies to "hog" the computer."(Borg) The second reason is "culturally pervasive stereotypes of the computer professional did not exist. There are two particularly damaging aspects to the stereotyp e: the image of the nerd or hacker, and the pervasive belief that 60-80 hour workweeks are required." (Borg) She also gives a little personal example of her cousin being exposed to one of her professor's view on what computer science is all about. This professor showed a video entitled "Triumph of the Nerds". "It is all about the founders of Apple and Microsoft and the "guys in Silicon Valley. Technical women are almost entirely absent..." (Borg) This article completely supports our conjecture that wom en are not entering computer science majors from high school because of the male "geek" stereotype that is put on people in this field.

Ellen Spertus gives a report that explores the reasons, as stated in the introduction of the report, "why females consistently stay away from computer science, why people of both sexes seem to expect less from women, and why a woman is considered unfeminine if she is an engineer"(Spertus 1991). The author found it strange that other females did not have the inclination that she had to pursue technological careers. She did not take the conventional route and just explore the neurological an d psychological differences, but she also found great evidence of culture biases which could lead to the disparity of females in Technological, specifically Computer science fields.

The first section of her report analyzed the societal pressures, which result in the lack of women in the technical fields. She cites stereotyping as a one of the most influential societal pressures over women. The separation of feminine and < I>masculine occupations, have pushed women into their certain roles. She infers that from childhood, society teaches women that technical fields are unfeminine through various avenues. Media portrayals of women, the difference in male and female toy s and the treatment of boys and girls, all help provide this classification of what is feminine and what is not. She outlines several instances in which males and females are treated differently due to societal biases. She concludes that society class ifies "Assertiveness, confidence, and high achievement "(Spertus 1991) as masculine (unfeminine) qualities, while homeliness and submissiveness are considered proper feminine attributes.

The report also takes a look at how the environment of the technical workplaces has an effect on women entering those fields. She showed examples of Masculine workplace environment, which makes female employees, feel uncomfortable. She defines these such environments as being a result of rampant inappropriate sexual discussions, and behaviors, lack of opportunities for female advancement, differences in the communication styles of the sexes, lack of female mentors, and difference in the priorit ies of males and females. She faults this environment for the lack of desire of many females to maintain and/or pursue technical careers. She concludes that women tend to feel uncomfortable in these environments and as a result "feel out of place", and have difficulty seeing themselves in these working environments.

Another section of this report discusses the gender biases that exist in our daily speech. She analyzes these biases because she feels that "Biases in language are important because they show both the biases people hold and how they are communica ted."(Spertus 1991). She analyses the different connotations that feminine words infer as oppose to their masculine counterparts. An example given, is the use of prince vs. princess. When you refer to someone as a prince, they are seen as speci al, strong, gallant…and other positive attributes. As where you call someone a princess, they are seen as spoiled pampered, conceded…and other negative attributes. She also discussed the assumptions made, as to sex, when it comes to technical professio nals. In most instances, when sex is undefined, masculine words are used as default. She hypothesizes that upon the entry of more females in the technological fields these stereotypes and assumptions would all but disappear. She concludes that in order to overcome the gender biased, "A girl must have a strong vision to see herself in a career considered masculine, and a woman must have strong character not to be affected by all the masculine assumptions about her"(Spertus 1991).

The original paper that sparked our interest in this topic is a web site found at MIT summarizes the findings of the Final Report (submitted on Feb. 17, 1995) by the MIT Electrical Engineering and Computer Science (EECS) Women Undergraduate Enrollment Committee. The report observed that EECS has the lowest ratio of women to men of any major at MIT with more than a few students. It found that women are half as likely as men to major in EECS despite the fact that men and women major at the same rate in t he School of Engineering as a whole. Moreover women applying to MIT seem to express the same interest in EECS as they do in other fields. The report found that "women come to MIT less prepared to major in EECE than their peers while both men and women pe rceive EECS to be very competitive." They found that the same ratio existed at other schools as well.

However our conjecture is that this report does not give a complete description of the situation. It does not explain why High School women reported as having an even balance with men in AP courses leading to possible interest in CS may not choose to p ursue a career in Computer Science. It also does not explain the difference between the numbers of women showing interest and the numbers of women who graduate with an EE or CS degree.

Our questions address the possibility that some women are deterred by their peers and/or counselors in High School from majoring in CS. We also think that the stereotypes held by high school students' effect their college major decisions.

Methods
Our basic method was to administer a questionnaire to high school juniors and seniors and students in a freshman level programming class. After gathering completed questionnaires, we attempted to apply normative statistics in order to summa rize this data. We wrote 10 multiple choice questions which respondents answered on a scale of 1-5 (where 1 means "strongly disagree" and 5 means "strongly agree"). We also included two qualitative response questions. Survey question s were distributed either by hand or by teachers from the high school's who participated in the study. We are especially grateful to Jason Dorough of Woodward High School in South Atlanta, Dr.Gale Marshall of Alexander High School in Western Douglas Count y, and administrator Mary Isele at the College of Computing for supporting and administering our surveys. We entered the data into spreadsheets, used spreadsheet formulas to compute the mean and standard deviation, and used the chart wizard to display th e scores on scaled histograms and pie graphs. The main problem we encountered was time. Being spring quarter and close to graduation, many seniors had early exams and graduation practice. These factors forced us to reduce the number of students being s urveyed as well as rely on the results of input by juniors. The size of our sample achieved sufficient information including females and males from metro-Atlanta and suburban high schools.

Data and Statistics

Total Females: Suburban Females: Metro Females:

N = 29 N = 16 N = 13

Total Males: (all males were classified as suburban)

N = 12

The following questions were rated on a scale consisting of the following choices and the values the team assigned each choice: Strongly Disagree = 1, Disagree = 2, No Opinion = 3, Agree = 4, Strongly Agree = 5. *Note: The order of the questions below is not the same as the order of the questions on the survey. The order was altered to better group answers to similar questions.

1. My parents have encouraged me to pursue a career in a technical field.