Gender Issues in Computer Science Education

Davies, A.R., Klawe, M., Ng, M., Nyhus, C., Sullivan, H.
Department of Computer Science,
University of British Columbia, Vancouver, BC, Canada

Introduction 

The low participation by women in both the information technology (IT) industry and in computer science courses in secondary and post secondary education is an important equity issue in science education. In addition to the increasingly intense need for more highly skilled people in the IT sector, women are missing out on many of today’s most attractive career opportunities. Equally importantly, the IT field is missing out on the broader range of perspectives and talents that would result from significantly increased participation by women.

While many people are at least somewhat aware of the low participation of women in IT careers, few realize that both the percentage and total number of bachelor's degrees awarded in Computer Science to women decreased almost every year over the last decade. This is in direct contrast to almost every other area of science and engineering where participation by women has significantly increased. Only 14.4% of employees in IT are female (Myers, 1999); and that there was only a 2% increase in the number of women in high-tech fields between 1991-1996 (Myers, 1999). The current percentage of Computer Science B.Sc. graduates from U.S. research-intensive universities (i.e. those offering a Ph.D. in Computer Science) is approximately 17%.

Although career opportunities in the IT industries are dramatically increasing, the ability of women to participate in this industry is dependent on their willingness to learn (at least some) computer science. Research indicates that most girls indicate a lack of interest and perceived ability in computer science and engineering in grade eight and earlier, and that this situation persists throughout secondary school. It is therefore very important to understand how to increase girls’ interest and achievement in computer science and their awareness of IT careers.

This paper aims to first understand why there is such a significant difference between girls and boys in choosing IT as their careers.  A literature review on the important factors influencing gender differences in IT career choice will be presented, as well as a summary of possible actions recommended by various research groups.  We will then introduce the SWIFT program, which is an overall program aiming to understand and tackle the issue of low participation of women in the IT field.  Under the SWIFT program are various projects that handle different aspects of the problem.  These include: (i) an overall career survey to explore the existing gender gap in their perception of science subjects;  (ii) the E-GEMS project aiming at developing gender-inclusive computer software to enhance female students’ interest in mathematics and educational computer games; (iii) the Virtual Family program focusing on increasing students’ programming knowledge by designing gender-inclusive software that provides an easy and enjoyable introduction to Java; (iv) a variety of outreach initiatives, such as the organization of a provincial conference to link practitioners and the design of a new introductory computer course emphasizing computer applications for university programs; and (v), the ARC program which offers retraining for adults who are interested in starting their careers in the IT industry.  We will then conclude this paper by highlighting how the above projects address the issue of low participation of female in the IT field, and the new directions our projects generate.

Literature Review: What is Known About Female Participation in the IT Field:

The gender imbalance in computer science is well documented (Camp, 1997; Klawe & Leveson, 1995; Pearl et al., 1990). Forecasting based on current enrollment statistics shows that into the year 2002, university computer science graduates will continue to be granted to men and women in percentages very similar to those seen today (Camp et al., 2000). Course taking patterns in high school indicate that the gender imbalance in computer science has already established itself prior to university. Based on a US Department of Education high school transcript study, the overall computer related course enrollment in high school had close to equal representation of male and female students in 1994 (US Department of Education, National Center for Education Statistics, 1998). However, a higher percentage of female students selected Clerical & Data Entry, whereas a higher percentage of male students selected Computer Applications. High school course taking patterns from British Columbia, however, have gender imbalances closer to those seen at the university level (GenTech, 2000). Table 1 shows the percentage of females (versus males) enrolled in BC high schools during the years 1994/5, 1995/6, and 1996/7.

Table 1: BC Secondary School Computer Course Enrolment Patterns

Course

Female : 94/5

Female: 95/6

Female: 96/7

Tech Education 8

42 (males 58)

42.1 (males 57.9)

42.3 (males 57.7)

Tech Education 10

13.1 (males 86.9)

16 (males 84)

16.8 (males 83.2)

Computer Education 11

35.5 (males 64.5)

35.1 (males 64.9)

34.8 (males 65.2)

Computer Education 12

25.6 (males 74.4)

23.2 (males 76.8)

21.5 (males 78.5)

The British Columbia statistics on course enrolment clearly shows that female students have already turned away from computer science in greater numbers relative to male students prior to entering university. Why so few female students choose computer science is less clear, though many reasons have been offered in the existing literature.

The various factors influencing female enrollment in IT related courses could be grouped under the following headings:

Socialization

Computers are perceived as belonging to the male domain of mathematics, science, electronics, and machinery (Inkpen et al., 1994). Computer use and expertise have been associated with masculinity, and therefore, gender socialization serves to act negatively on female students’ computer interactions. In one study it was found that boys aged 3-6 had already acquired a gender stereotyped view of computer users (Fletcher-Flinn & Suddendorf, 1996). Boys in general receive more support from teachers and parents, and are more likely to be the main computer user at home than girls are (Becker & Sterling, 1987).  Moving from the early years, the effects of socialization become more entrenched. It has been reported that masculine gender role traits are associated with a more positive computer attitude (Colley et al., 1994), and that there is a gender stereotype of science in general being masculine (Kahlee & Meece, 1994).

The Role of Computer Software

This gender stereotype is reflected in computer software, especially computer games.  Since society in general tends to stereotype females into weak dependent roles, the stereotypical woman in most computer games is weak, and if she is a main character, she is modeled to attract the male audience: slim, big-breasted and scantily clad.  Educational software has been found to have a male bias (Jenson, 1999); one study of mathematical software reported that the number of female characters declined, and the amount of violence increased, as the grade level the software was directed at increased (Chappell, 1996). Research in the area of educational computer games for the early grades has also suggested that violence, competition and the scarcity of females characters all impact on girls’ interest in using computer software (Jenson, 1999; Inkpen et al, 1994; Koch, 1995; Funk & Buchman, 1996). This finding coincides with E-GEMS research findings from their educational games (see section on E-GEMS research under Programs). Educational software programs, often modeled after video games, have been accused of serving as a gateway to technology for boys, but not for girls (Chappell, 1996).

This gender bias influences children’s interest in playing electronic games: boys play more than girls and prefer more realistic and violent games (Funk & Buchman, 1996).  Research has shown that, in the primary and middle school level, it is more socially acceptable for boys to play computer games, and those boys who play computer games are perceived as more popular than girls who do so (Funk & Buchman, 1996). One question being raised in research is that if girls see female figures in electronic games as problem solvers and skilled thinkers, would their views of their own ability change? (Inkpen et al., 1994)  There is a good possibility that the answer to this is yes. Looking into the field of social psychology and studies done on aggression it is noted that one possible reason that females are not as aggressive as males is that they are less likely to see females portrayed in the media and in real life as aggressive people (Aronson, 1998).

The Importance of Role Models

Researchers have reported that the main users of computers at home are male, and since there is evidence of same-sex modeling within families, boys are more likely than girls to get interested in computers (Hess & Miura as cited in Shashaani, 1994). At school, computer science is more frequently taught by male teachers. One 2-year study conducted in a secondary school reported that of the 13 computer science classes conducted during the course of the research, 12.5 were taught by men (Schofield, 1995). At the university level, women comprise only 16.4% of assistant professors, 11.7% of associate professors, and 7.6% of full professors (Kozen & Morris, as cited in Camp, 2000).

Coupled with the fact that there is an obvious shortage of female role models among science teachers, many educators, including female teachers themselves, are not aware of the danger of perpetuating the female stereotype. A 1993/4 survey in the US reported that the most time teacher educators spent on gender equity is two hours per semester, with one third spending one hour or less (Campbell & Sanders as cited in AAUW, 1999). The chilly climate for girls using computers is promoted by the beliefs, values, and practices of educators themselves (Becker & Sterling, 1987).  Teachers have been reported to play a role both in perpetuating gender socialization and impacting negatively on girls’ experience with computers (Shashaani, 1994; Volman, 1997; Sanders, 1995).

Access to Computers

The issue of differential access to computers experienced by female and male students has been well documented, and this difference becomes noticeable at a very young age. The pattern of gender imbalance (access, ownership, and usage) shown in elementary schools, high schools, and universities can be found among kindergarten and pre-school students (Fletcher-Flinn & Suddendorf, 1996). By grade four, the differences in computer usage and the attitudes towards computers and electronic games are becoming more noticeable. Between grades four and eight the hours spent on computers decrease for females and increase for males (Funk & Buchman, 1996; Buchman & Funk, 1996).  

In populations of secondary school students, more male than female students report having a home computer (Shashaani, 1994; Volman, 1997; Woodrow, 1994), and the same gender difference has been reported for university students (Shashaani, 1997). More male than female students report having access to home computers (Busch, 1995; Dugdale et al., 1998), having used computers at home (Colley et al., 1994), and having access to and using a home computer at least once a week (Comber, 1997). In school, although both boys and girls have access to school computers, boys tend to dominate the computer resources, leaving girls with less access to computers in the school setting (Schofield, 1995, Underwood, & Underwood 1999). One study reported that 83% of the girls responded that they used computers at school one hour or less per week, and 30% of the boys responded that they used computers at school five hours or more (Shashaani, 1994).   Girls in general have less access to computers in school and home, are underrepresented in elective programming courses, and over represented in word processing courses.  This is similar across age groups and countries (Fletcher-Flinn & Suddendorf, 1996).

School policy and classroom practice sometimes result in the alienation of female students from computers and computer science, hence denying them the opportunity to explore this field of study. When computer science is placed within the mathematics departments of schools, the image of computer science being a masculine course is exacerbated by its proximity to mathematics, which itself is seen as masculine (Kahlee & Meece, 1994). Although mathematics has seen the benefits of a concerted effort to attain gender equity (AAUW, 1999), the association of computing science and mathematics has added to male students feeling more at home in computer science than girls (Schofield, 1995).

Experience with Computers

As well as all the above factors, computer inexperience for girls both at home and at school has often been cited as an important factor in determining their attitudes and anxieties towards computers (Jenson, 1999). Due to the fact that boys have a greater tendency to dominate available computer resources, and parents and educators tend not to correct that fact, females have significantly lower experience levels than their male counterparts. In the classroom it has been observed that males tend to dominate the computers during free time, and females will only use the computers when given specific instructions allowing them to (Koch, 1995). Comparing to boys, girls spend less time per day playing video and computer games at home, own fewer games, and are less interested in and knowledgeable about the gaming industry (Klawe, 1998). 

There is a strong correlation between students who do not like and rarely play with computer games and those who judge their computer skills as weak and avoided spending time on the computer.  In the1995 study done by Koch et al. with a group of children in Ontario, they noticed that there were computer "in-groups" and "out-groups."  The computer "in-groups" were the group of children, predominately male, who controlled the computers during recess and free computer time, and even named the computers. The students on the margin of this computer in-group would not sacrifice their self-esteem to a computer game: these students were often girls (Koch, 1995).  Without the experience level of the boys, girls are often intimidated and feel unwelcome in the male-domain of computers in the elementary classroom.  Computer experience in the primary and middle school years is critical to ensure that gender equality in attitudes and computer use is reached because it is during these years that girls often turn away from science and technology (Koch, 1995).

Given the low level of exposure female students have, it is also noted that they have less programming experience as compared to their male counterparts.  The gender imbalance between those students with programming experience and those without is evident in the transcript and course enrollment data by gender (US Department of Education, National Center for Education Statistics, 1998; GenTech, 2000). The gender difference in this area was also reported in a study of teacher education students (Liu & Reed, 1992) and college students (Busch, 1995).

In studies that deal with gender difference in self-perception of computer experience, it is shown that males report having a greater range of computing experience than the range reported by females (Shashaani, 1994; Comber et al., 1997). Female students of single-sex schools also reported a greater diversity of computing experience relative to female students in co-educational schools (Jones & Clarke. 1995). The issue of breadth of experience is significant not only as an indicator of the students’ range of computer ability, but also as an indicator of attitude: diversity of computer experience has been identified as the major predictor of the score on computer attitude (Jones & Clarke, 1995). Testing of prior computer knowledge upon entry to a computer class resulted in lower scores for female secondary school students, and their estimate of their own competence was lower than the estimates their male classmates made of their own computer competence (Volman, 1997).

Attitude and Anxiety

The combination of socialization and lack of access and experience in computers have negative impact on female students’ attitude towards computers.  Computers are perceived as belonging to the male domain of mathematics, science, electronics, and machinery (Inkpen et al., 1994).  Computer use and expertise has been associated with masculinity, and therefore, gender socialization serves to act negatively on female students’ computer interactions.  Multiple findings have shown that gender differences in attitudes towards computer usage begin as young as 3 years old, and it was found that boys aged 3-6 had already acquired a gender stereotyped view of computer users (Fletcher-Flinn & Suddendorf, 1996; Kay, 1992; Koch, 1995; Becker & Sterling, 1987). There are fewer gender differences in computer attitudes among pre-school and primary school students than there are among middle, secondary, and university students (Kay, 1992). This difference in attitude is cumulative, however, and significantly contributes to the gender imbalance in computer course enrolment and interest in computers when female students reach the high school and university level (Becker & Sterling, 1987). Many studies across age and cultures have been conducted, and not one of them reported a case where the girls reported more positive attitudes towards computers than the boys did (Fletcher-Flinn & Suddendorf, 1996).

The attitudes of students towards computers have been assessed in a number of studies with both secondary school and university level students. Female students at the secondary level report their confidence level with computers lower (Shashaani, 1997) and their interest in computers lower (Shashaani, 1994; Volman, 1997), when compared with equivalent reports by male students. Secondary school male students were also more explicit about their lack of fear of computers (Volman, 1997).  One interesting gender difference observed by Koch (1994) is that when dealing with computers, girls blame their own inexperience and lack of ability if something goes wrong, hence decreasing their self-confidence level, while boys tend to blame the program or the computer system if something goes astray.  This is consistent with the findings in social psychology that girls feel it is their fault when something goes wrong, whereas boys tend to attribute fault to external causes such as bad luck (Aronson, 1998). In general, female students at the secondary and university level report higher levels of computer anxiety (Busch, 1995; Colley et al., 1994; Brosnan, 1998), lower levels of confidence with computers (Busch, 1995; Colley et al., 1994; Shashaani, 1997), and a lower level of enjoyment, or liking, of computers (Shashaani, 1997; Colley et al., 1994), than their male counterparts.

Ability and Perceived Ability

Directly related to the anxiety and less-than-positive attitude of female students for computers is their self-perceived ability to do well in computer-related fields.  During the early grades, there is a general consensus between both that girls in general are less proficient on the computers (Fletcher-Flinn & Suddendorf, 1996). These findings are reported through the secondary years as well, with most female students being less confident in their ability with computers than the male students. Although females may believe that there is no difference in general between males and females in computer skill and ability (Shashaani, 1994), when rating their own ability, female scores are lower relative to their male counterparts (Volman, 1997), a phenomenon known as the “we can but I can’t” paradox (Colley, as cited in Jenson, 1999). Female students scored lower in perceived self-efficacy in computer use, defined as the belief in one’s own ability to use computers successfully (Busch, 1995).

All the factors mentioned above influence female students’ likelihood to choose computer science as their subject of study and hence to opt for IT as their career. We now briefly turn to the literature of pedagogy describing women’s learning approaches.

Gender and Learning Approaches

The issue of socialization and its negative effects on the full participation of diverse learning communities is a subject with mounds of data within the education literature. While research to date does not point to a particular preference for learning based on gender, in the search to understand what discourages and encourages gender equity within computer science, a look at the views on learning held by feminist pedagogy, adult development and education reveals many useful clues (Taylor, 1995).

Within the literature of computer science, studies identify a preference in males to start the learning process with abstract concepts, where women tend to prefer concrete learning styles (Severiens, S. and Geert, T.D. 1997).  Men’s learning styles were found to be more congruent with traditional education than women (Philben et al. 1995). These findings support those of many who have noted that nearly all education systems were initially designed for the education of men, with a knowledge base predominantly based on a rationality that was socially constructed by white males.

As documented in the last twenty years, women’s ways of developing and learning display different attributes and emphases (Miller, 1984; Belenky et al.,1986; Gilligan et al., 1990;  Miller, 1984; Tisdale, 1993). These authors highlight the preference for connection, relationship, understanding and collaboration that women typically display in their preferred learning approaches. These findings mirror those indicating girls’ preference for electronic games that include elements of story line, characters,  relationships, collaborative learning and attention to physical, cognitive, and social space (Ching et al., 2000, Inkpen et.al, 1994).

Women and minorities were considered to be an under-served population by John Dewey (1938) and the progressive education movement.  This movement outlined directions for change from ‘traditional’ education to an approach that emphasized the idea “that there is an intimate and necessary relation between the process of actual experience and education.” (Dewey, 1938:20).  Adult education grew from the historical and philosophical frameworks of such thinkers as Dewey (1938), Freire (1970) and Knowles (1980) and their desire to serve a diverse learning community. Knowledge of cognitive developmental stages have helped educators understand the changes a learner encounters and provide educators and theorists with powerful tools for curriculum design (Kolb, 1984; McCarthy, 1987; Merriam, 1993; Weil and McGill, 1989).

Research shows that inclusion of reflection on life experience as an accompaniment to critical reflection and a discussion of the theoretical material contributes to both the voice and confidence of women ( Belenky et. al., 1986; Jordon et. al., 1991; Melemed and Devine, 1988). Within the current literature on gender equality in the learning environment are frequent references to differences in learning styles (Kolb, 1984).  Kolb’s Experiential Learning Model (Kolb and Lewis 1986) encourages an educational approach particularly friendly to women, one that integrates thinking, feeling, perceiving and behaving (Kolb, 1984; McCarthy, 1987; Merriam, 1993; Tisdale, 1993).  This valuing of each learner’s experience and encouragement to reflect on the impact and connection of learning to their lives is a particularly important contribution to the enhancement of self-esteem and personal competencies for women (Brown and Gilligan, 1992; Galbraith, 1994; Merriam, 1993; Miller, 1984; Steinem 1992; Tisdale, 1993). Ayersman (1996) found significant differences in computer anxiety amongst a post-treatment group of 86 undergraduates among learning style groups and urges further examination of this variable to gain understanding which would allow all learners to effectively and equitably succeed in computer–based learning.  Bunderson and Christensen (1995) exhort computer science courses and departments at all educational levels to become more friendly to women, using teaching methods inclusive of all learning styles and genders, class discussions, and positive responses to questions asked by all class members. These references to different learning approaches provide encouragement to pay attention to teaching methods, learning tasks and environments (Severiens, S.E. and Geert, T.M. 1994; Taylor, K and Marienau, C. 1995).  It is widely noted that understanding how women’s learning needs are different from those of men and ways the learning environment can be structured to empower women and foster their learning can help those promoting an equitable learning agenda (Belenky et al., 1986; Gilligan, Lyons and Hanmer, 1990; Saltonstall, 1990; Tisdale, 1993).

The above sections of the paper deal with identifying and understanding the contributing factors to low female participation in the IT field.  The rest of the paper will focus on recommendations and actual programs that are being launched to reduce the gender gap.   

Programs and Intervention

Research has been conducted on what kinds of interventions and programs can be developed to increase the number of female participants in computer science.  Studies that focus on identifying, describing, and explaining educational inequalities in relation to science and technology have provided us with an important baseline for intervention work.

To alleviate the low participation of females in computer-related fields and to address the above factors, researchers such as Koch et al. and action groups such as AAUW have proposed lists of actions to be taken.  In their paper, Koch et al recommend seven points that have to be addressed to reduce the gender gap.  They are exposure, educational reform, appreciating gender, choosing software, technology programs, all girl classrooms and role models (Koch, 1995).  The AAUW report makes seven recommendations on what needs to be addressed to overcome the under-representation of girls and women in computing (AAUW, 2000). They are:

·        Transform pink software by creating gender neutral software that challenges and appeals to a variety of students;

·        Look to girls and women to fill the IT job shortage: encourage girls into computing by using technology in a broad range of subjects to attract a more diverse group of students;

·        Prepare tech-savvy teachers: empowered teachers will empower students;

·        Educate girls to be designers, not just users, ensuring they have opportunities to fully explore the potential of technology;

·        Change the public face of computing so girls have a realistic image of computer professionals, and understand the importance of communication and team work in this field;

·        Create a family computer, placed where the entire family has access, and computer activities are associated within a social context, and not equated to isolation; and

·        Set a new standard for gender equity that seeks equal contributions to innovations in technology and equal mastery of the analytical and computing skills required to make these contributions.

The SWIFT Program           

The SWIFT (Supporting Women in InFormation Technology) program is a multi-facet, comprehensive program aiming to encourage more female participation in the IT sector and address some of the problems underlying the low participation rate.  The SWIFT project came into being as a way to house and refer to the various projects and activities initiated by Dr. Maria Klawe in her role as the NSERC-IBM Chair for Women in Science and Engineering in BC and the Yukon (for more information, please visit http://taz.cs.ubc.ca/swift).  SWIFT programs and initiatives aim to turn the tide on gender inequality in computer science, and are based on research findings that attempt to establish the causes behind the lack of female participation at all levels of computer science.

What follows are the description and major findings of the various projects that SWIFT oversees:

The SWIFT Survey

In the fall term of 1998, SWIFT conducted a survey of Vancouver high school students on their career choices. There were two reasons to conduct this survey. First, to establish where SWIFT initiatives should focus when trying to increase the number of female students choosing to study computer science, and second, to form a baseline with which to compare the future effects of SWIFT initiatives.

As mentioned in the section on literature review, there are various factors impacting on female students and affecting their decision to choose computer science as a subject of study; of these factors, the SWIFT survey focused on four areas. The first area was school subjects. Students were asked to rate different subjects, both on their perceived ability in the subject and their interest in the subject. Respondents rated their ability and interest on a scale of one to seven, with one indicating very little interest or ability, and seven representing high interest or ability. The second area was the influences on career choices; students were asked to rate the strengths of the various influences on their career choices, using the same seven point scale, with one indicating very little influence, and seven indicating very high interest. The third area focused on computers; students were asked about their access to, use of, and perceived ability with, computers. The last area asked about the students’ perceptions of a variety of professions; students were asked what skills and personality traits they thought were needed by someone pursuing different professions, with responses either indicating a trait or skill was needed or not. Results were tabulated based on the percentage of positive responses overall.

In total, 7, 411 surveys were filled out by students in grades 8, 10, and 12, of the Vancouver School District; this number represents 56% of the total number of students in those grades in the school district. There was little variation between grades, and therefore results are the combined results for all grades. Of the surveys filled out, 88% of the respondents indicated their gender, of those, 48% were male, and 52% were female.

School Subjects 

The first area was school subjects, and the results reflect the BC data on course taking with female students indicating that they were less interested and had less ability in computing science than did the male students. Overall, the average of scores for both interest and perceived ability were similar, for both male and female students. Figure 1 shows the average responses by gender to the question of interest in a subject, and figure 2 shows the average responses by gender to the question of perceived ability in a subject.

Figure 1. Interest in subject areas

Figure 2: Perceived ability in subject areas

Influences on Career Choices 

The second area the survey addressed was influences on career choices. Students were asked to rate the strengths of the various influences on their career choices. For both male and female respondents, personal interest and ability was rated as being the most influential factor when making a career decision. Based on reported gender differences in both interest and perceived ability with computers, and gender differences in computer science, decisions on career choices currently being made by women would follow from the pattern indicated by this survey. Figure 3 shows the average responses by gender. 

 

 

Figure 3:Influences on career directions

Computer Use 

The third area covered by the survey queried students on their use of computers. Students were asked to rate different computer activities according to how much time they spent on each activity. This question was asked for computer time at school, and computer time outside of school. Male students indicated they spent more time on computers both at school and outside of school. This measure, however, is only an indicator of the relative time spent at the various activities, as a 1-7 point scale was used, 1 indicating the least amount of time, and 7 indicating the most time; therefore direct comparisons between males and females are difficult to make. However, the responses are in-line with studies showing male students have greater access to computers, both in and out of school. The survey results also indicate male students spent more time programming and playing games relative to other computer activities compared to female students. The only areas female students report spending as much time as male students are assignments and e-mail.  This result is reflective of results found in the literature.  Figure 4 shows the average responses by gender to the question of computer use at school, and figure 5 shows the average responses to the question of computer use outside of school.

Figure 4: Computer use at school

Figure 5: Computer use outside of school

Students were asked to rate their computer skills in a variety of areas. The responses to this question are reflective of other studies that report gender differences in confidence and experience, and report male students having a stronger background in programming compared to females. The large difference seen in the response to Systems & Hardware proficiency could be the result of a number, or a combination of, different factors. Different studies have reported that male students are much more likely to overstate, and female students to understate, their general computer knowledge and accomplishments (Jenson, 1999; Schofield, 1995). Males are much more likely to use the ‘expert’ lingo of computers than are female students (Volman, 1997), and therefore may be more likely to view themselves as proficient in the concepts behind the terms they are familiar with. Males also have more experience, and spend more time on computers, and therefore have had more opportunity to increase their proficiency in this area. All the above factors may combine, resulting in the largest gender discrepancy being in the area of Systems & Hardware. Figure 6 shows the average responses by gender.

Figure 6: Perceived computer proficiency

Perception of Computer Professionals 

The fourth area the SWIFT survey covered was students’ perception of various professionals, including what skills and characteristics were required to be a successful computer professional. Both males and females chose computer skills most often, and females chose communication the least often as a skill requirement of a computer professional. This shows that female students would benefit from a better understanding of what the information technology industry looks for in a computer professional. Seeing communication as not being important to being a computer professional may also reflect the perception that the computer culture reflects male culture and therefore the skills associated with women would not be needed or important. Males and females responded differently to the requirements of business skills and creativity, with females rating business skills more often, and males rating creativity more often. Figure 7 shows the average response by gender to the skill requirement question.

Figure 7: Skill requirements for computer professionals

Males and females responded very similarly to the question of what characteristics were required to be successful as a computer professional. Female students generally responded positively more frequently to all characteristics, with the exception of: good memory, team player, and outgoing. Males were just slightly more likely to respond positively to the characteristic of having a good memory, but there is a much greater gap between the other two characteristics, team player and outgoing. This again indicates that female even more than male students are not receiving the information required to make an informed career choice.

Figure 8: Characteristics required for computer professionals

The results above clearly show that gender gap does exist in terms of their interest and perceived ability in handling computer science as a subject, and since these are the two main factors influencing students’ career choices, it is highly probable that more male students would choose to get into the IT field than females. Our findings also indicate that male students spend more time using the computer, have more experience in programming, and rate their computer skills as higher. With these findings as a guideline and yardstick, SWIFT has been running a number of different but inter-related programs to address these issues.

E-GEMS

Up until 1993 there had been little concrete research on the topic of children's interactions with computers and video games. It was partially due to this lack of research that the E-GEMS project was created in late 1992: to gain and provide insight into the interaction between grades 4-8 students and educational computer games. This age range was chosen because previous research indicated that this is when most children lose interest in math and science. The research in the E-GEMS group at UBC has focused on both commercially developed games and prototypes developed by teams of computer scientists, teachers, and undergraduate students working at E-GEMS during their cooperative work terms.   Throughout the research E-GEMS has conducted in the past seven years, particular attention has been paid to understanding how to ensure this type of software is effective for girls as well as boys (Klawe, 1999). 

Major Findings

            Gender differences-- Level of interest. The first major research study conducted at E-GEMS that took note of the gender imbalance in computer games took place in 1993 at Science World where observations were made of children's interaction with popular educational electronic games. During the study, girls were observed to be less interested in the electronic games compared to boys, but more interested in computer games than video games.  This interest level rises if the girls have a computer and/or a video game system at home. A higher percentage of girls were attracted to the computers compared to the percentage of boys, and far more girls than boys preferred the computer game design station. The research findings were that the girls preferred other activities, such as the design station, and only played the computer games if another group of girls were playing or there were free machines. This coincides with the findings that girls are more likely to play if there is ease of access to the computer games and there is a possibility of interacting with others while they play (Inkpen et al., 1995). 

            Gender differences -- What they want in computer games. Various studies, including those by the E-GEMS group, have found that girls have different desires and preferences for computer games compared to boys.  Girls, in general, feel that the important elements of computer games are story line, characters, worthwhile goals, social interactions, creative activities, and challenge.  Most boys want fast action, adventure, challenge, and violence (Klawe et al., 1996).  During the research at Science World, we observed that boys tended to dominate the fast action, high paced computer and video games while girls focused on story line and character development, educational games that they deemed worthwhile, and shied away from large groups playing the games.  Based on these findings, E-GEMS developed the Phoenix Quest game with the goal of creating a challenging adventure game incorporating mathematical concepts that emphasized story-line, interaction with the story's characters and creative activities, rather than fast-action and violence.  To emphasize the story line, interaction with characters, and creative activities, Phoenix Quest incorporates several language arts activities requiring the player to converse with the game characters and focus on the story line in order to gain access to the puzzles.

E-GEMS research using Phoenix Quest reveals that this approach to game design is effective in creating games that are attractive to females. On a session-exit questionnaire the girls voted significantly more often for "Human" aspects of Phoenix Quest: Helping Julie, Helping Darien, Getting to know the characters, Finding new chapters, and Following the story (Klawe et al., 1996). Although story line and social interactions has often been noted as major attractions for girls in computer games (Jenson, 1999; Koch, 1994; Chappell, 1996), there are still few examples of commercial computer games that emphasize these aspects.

            Gender differences -- Differences in play. In nearly every one of the E-GEMS field studies it was found that girls and boys show different preferences and levels of performance in activities and interaction styles.  Girls often spend more time exploring and communicating with partners, while boys progress faster through the activities, completing more levels and puzzles. For example, males completed more challenges and had higher scores in the activities compared to the females in a study using the computer game Builder (Graves & Klawe, 1997).  In a study using the computer game The Incredible Machine, it was found that girls on average finished fewer puzzles than the boys, whether they were working alone or as a pair (Inkpen et al., 1994).  With Phoenix Quest, it was found that boys were much more concerned about making rapid progress though the game than the girls, who tended to spend more time savoring each activity and exploring the game (Klawe, 1998).  Phoenix Quest rewards players who repeat puzzles to higher levels, even though the higher levels are not necessary to complete the game, another feature appealing to girls.

Despite progressing more quickly through game levels and puzzles in these studies, in the assessments of mathematical learning resulting from game play, boys have not shown greater achievement (Klawe, 1999).  Thus, even though there is a difference in the style of interaction girls and boys use when playing, with girls not racing through the games as quickly as boys, these differences do not result in lower learning gained by girls. It is important for computer games to allow the player to explore the game and repeat levels, so that girls who do not choose to play in the narrow goal-oriented way favored by boys can still benefit from the learning experience.

            Gender differences -- Importance of collaborative play. Preferences in computer games and styles of play are not the only gender differences that E-GEMS researchers have discovered. Another important discovery that coincides with findings outside of E-GEMS is that girls prefer collaborative play.  Females prefer computer games that allow them to communicate their ideas and knowledge with others (Graves & Klawe, 1997; Inkpen et al., 1994).  Girls prefer to share their ideas and help each other while playing a game.  In the study involving The Incredible Machine, it was shown that girls complete the puzzles faster and enjoy playing the game more when playing side by side compared to playing alone; they enjoy the game even more when playing together on the computer compared to sitting side by side (Inkpen et al., 1995).  Females preferred playing in partners and advanced further if they were given the chance to do so.

Strategies and Intervention

One E-GEMS researcher, Kori Inkpen, recommends four strategies to change the gender discrepancy: The first recommendation is ownership, that is, allowing girls to have equal voice in what is done with the computers, and have a girls-only and boys-only computer designation.  The second recommendation is creating a space for them, as girls are more likely to play if no one who might criticize them is around the computer.  In the study conducted at Science World, intervention was necessary in order to create a space for girls (Inkpen, 1994).  Different studies have been conducted showing that ensuring equal access to both boys and girls work in increasing young girls’ usage of computers (Inkpen, 1994; Koch, 1995; Fletcher-Flinn & Suddendorf, 1996; Becker & Sterling, 1987; Jenson, 1999; Klawe, 1998).  It is also shown that gender differences in attitudes and computer anxiety can also be reduced when equal access to computers is ensured (Jenson, 1999).  Single-sex schools may be a good example of this: studies suggest that girls in single-sex schools have more positive attitudes towards computers (Inkpen, 1994).  E-GEMS research has found that during classroom observations, boys were more aggressive in seeking or demanding access to computers and in their behavior while engaged in activities. However, minor interventions such as occasional girls only computer time or discussions resulted in enthusiastic and sustained engagement of girls with the prototype software (Klawe, 1999; Sedighian & Klawe, 1996; Upitis, 1995).  After minor interventions it was found that girls spoke up more and were more enthusiastic about playing the games.

The third recommendation is social interaction.  Research has shown that allowing girls to play within groups at the computer increases their time spent on the computer (Jenson, 1999; Klawe, 1998). This makes sense girls enjoyed working in groups more than boys, and computers have often been seen as a solo activity for young girls.  It was discovered that one of the most influential factors in increasing computer usage for girls was simply the opportunity to use computers in friendship groups (Klawe, 1998).  Inkpen (1994, 1995) and Sanders (1995) all found that girls fare better if they are given the option of interacting while they are on the computer. 

The last recommendation is a call to all software engineers. Several empirical studies have confirmed a strong role orientation in video games both in the arcade and at home.  This bias favors boys, and developing games that interest females will allow them more enjoyable playing time with things that interest them.  Game developers should be more aware of the female market and make games geared toward them (Koch, 1995).

Virtual Family

While the E-GEMS program tries to understand the gender differences in the use of computer games and to develop more appealing educational software for girls, Virtual Family is a program that attempts to develop materials and approaches to make learning programming more attractive to girls.  Efforts have also been made to organize workshops and to introduce the program to schools to make it more accessible to students and educators. 

Design Overview

One contributing factor to the low participation and high attrition rates of female students in computer science courses is the lack of programming experience female students have compared to male students prior to entering university (Bunderson & Christensen, 1995). When comparing the impact of different types of computer experience on computer anxiety, programming experience has also been related to lower levels of computer anxiety (Liu & Reed, 1992; Ayersman & Reed 1995/1996). A program to assist learning to program, designed in a way appealing to girls, has the advantage of increasing the awareness of computer science as a possible academic and career option, increasing the level of programming experience students have before entering university, and potentially decreasing computer anxiety. Virtual Family was developed by SWIFT to introduce an introductory programming activity that would appeal to female students.

Incorporating the important components of electronic games appealing to girls, namely story line, characters, and relationships, the Virtual Family program is based on a family and their interactions. The family includes Mom, Dad, Sis, and Junior, and a family mascot, a chicken named Flaxig. In addition, there are three family pets: a cat named cuddles; a dog named Rover; and a chameleon named Hue. These characters are line drawings, appearing on a changeable backdrop, that interact in animated action sequences. Each family member has their own menu where an item on the menu corresponds to an action sequence involving one or all of the family members, and none or all of the pets. The program was designed so that the action sequences could be easily modified by the addition of one line of code, or created by following the format of the action sequences already coded in the program. As students progress, they can add their own characters and pets and code action sequences involving their new characters interacting with each other and the core family members. Backdrops can be drawn or scanned in and added to the list of backdrops offered in the original program. A mood button on the start menu allows beginners to alter the initial mood and appearance of the characters. Moods can be used in the code, allowing students to alter character responses in action sequences based on the mood state they are currently in, which introduces the programming concept of conditionals and allows for more complex story development.

The modularity of object-oriented programming lends itself to software designed to be modified; the programming language chosen was Java. Each character was written as a separate class, and the core programming code was separated in other files, allowing beginners to start by both reading and modifying a small subset of the entire program, but allowing the possibility of a much more thorough exploration as the student became more experienced. Imbedding code to be modified within a larger program offers the student both a hands-on programming experience and the opportunity to benefit by reading the code of more experienced programmers. Offering program examples for students to read was a suggestion from the workshop, “In Search of Gender-Free Paradigms for Computer Science Education”, held at the National Educational Computing Conference (Frenkel, 1990). Virtual Family tutorials provide the guidance and expertise to carry out the program modifications.

Workshops and School Settings

Virtual Family workshops have been presented in a variety of settings and to a variety of audiences including elementary and high school classrooms, conferences aimed at female students in the secondary grades, and student teachers. Participants have come to the workshops from a wide variety of computing backgrounds ranging from students who have trouble controlling the mouse, to those with previous programming experience. Workshops have run between one to two hours. In this time, students add code to one of the java classes, compile the program, and run the new action sequence they have just coded. To accommodate as many backgrounds as possible, the tutorials offer lessons on basic computer operations, such as following links and cut and paste. Tutorials are currently being developed for the more experienced students and to assist those students who wish to continue developing their programming skills after taking a workshop. As well as the workshops, the Virtual Family program has been installed in school computer labs and classroom computers, allowing students to explore Virtual Family on their own.

Future Plans

Reaction to the Virtual Family workshop has been favorable, and future workshops in both elementary and high schools are planned. As well as continuing and refining the Virtual Family workshop, future efforts will be directed at using Virtual Family in an ongoing way to teach introductory programming skills and concepts. Although the Virtual Family program is currently installed on school computer labs and class computers, organized presentations have been limited to the workshop format and, due to time restrictions, the introduction to the programming aspect of the software has been brief. A review of the content and structure of the tutorials based on the observations of workshop participants and on individual one-on-one sessions between Virtual Family developers and students is currently underway. Future plans include developing and adjusting the tutorials to better meet the needs of a beginning student, with the aim of allowing Virtual Family to be used as a stand alone program not dependent on the structure of a workshop introduction. This necessitates the installation process of Virtual Family to be simplified, a project currently underway. Adding tutorials to both assist beginners and introduce more complex programming concepts will move Virtual Family in the direction of being a self-directed programming course that appeals to girls, introducing them to the world of programming and computer science in a way that is non-threatening, inclusive, and fun.

Outreach Activities

The following section reports on a variety of outreach activities with goals of changing attitudes towards and images of computer science, and of including the extended community in raising awareness of the issues and expanding the base for brainstorming about effective approaches. Over the past three decades, the science, mathematics and engineering communities have engaged in a wide variety of outreach and awareness activities for young women and their parents and teachers. Since there have been significant increases in the female participation rates in virtually all these fields over this period of time, it seems likely that similar approaches would also help with computer science.

Indeed, one might suspect that these efforts should already have addressed the gender imbalance in computer science along with the progress being made other areas of science and engineering. However, although computers are often used in outreach activities, the topic of computer science itself is rarely included. Those activities that do focus on computers such as computer camps tend to emphasize skills and the use of computer applications such as web page development. The image of an information technology career remains one of programming all day and night, and the image of the programmer programming remain one of a gifted loner engaged in a difficult, arcane obsession. Moreover, while parents, teachers and students are aware of the job opportunities associated with computer science, very few realize the low participation by women and the need to explicitly encourage girls to consider careers involving computer science. For these reasons many outreach activities now specifically target the image of computer scientists and programming. Some typical examples are described below.

Classroom Visits

Programs to support visits by volunteer scientists and engineers to K-12 classrooms, for example the Scientists and Innovators in the Schools (SIS) and Pathmakers programs (http://www.scienceworld.bc.ca http://www.carelton.ca/wise/pathmaker.htm), have been in place for decades. Such visits usually include a presentation by the volunteer about their work and how they got there, plus a hands-on activity engaging students in some aspect of their work. Evaluations of the SIS program indicate that interesting hands-on activities are an absolutely essential component of successful visits. Previously the lack (and disparity of models) of computers in schools made it difficult for computer scientists to engage students in hands-on activities related to computer science. Moreover, as computer science is almost never part of the required math or science curriculum, and because many teachers are neither familiar nor comfortable with computer science concepts, teachers are less likely to request computer science presentations compared to those in math and other areas of science or engineering.  Fortunately, the advent of relatively inexpensive hand-held (e.g. Palm) and laptop computers, together with portable computer projectors, and the wider availability of up-to-date computer labs and (supposedly) platform independent programming languages like Java, are making conducting hands-on computer science activities more feasible.

Some companies such as IBM recruit female volunteers from their employees and provide specific training and materials to help them make effective classroom visits to promote information technology careers. Similarly SWIFT hopes to develop a number of presentation and materials that could be used by computer science students and professionals in SIS and/or Pathmakers classroom visits. Already members of SWIFT are using Virtual Family and the E-GEMS computer games in their classroom visits to engage students in learning about programming, designing natural language dialogues, computer graphics, and the use of randomness in puzzle design. However more documentation needs to be developed in order to make these materials easily usable by other presenters.

IT Workshops for Girls

Workshops bringing together several hundred girls in grades 7-11 for a day or two are another popular format for science and engineering career outreach (see for example Ms Infinity, http://www.harbour.sfu.ca/scwist/).  Usually these include keynote presentations by highly enthusiastic and energetic women professionals about their work and careers, plus hands-on small group sessions in areas ranging from forensic entomology to building bridges out of spaghetti. Again computer science keynote and hands-on sessions have been rare in the past due to the low numbers of women in the field and the difficulty of offering good hands-on sessions. Even when computer science keynote speakers and session leaders are available, the organizers (often career counsellors) tend to view computer science as uninteresting and are less likely to include them in their program. Again, this has led to a number of workshops aimed entirely at information technology and computer science, as well as efforts to increase the presence of computer science material in existing workshop series.

SWIFT has been active in organizing IT workshops for female students.  Ms. Infinity is a one-day hands-on conference held in communities throughout British Columbia for young women in grades 9 and 10.  The day begins usually with a speech from a high profile female in science, and after that each student has a chance to attend 3 to 5 different hands on workshops.  Participation in these workshops is high, and the workshops have generated much interest in the province. 

Another program aiming at increasing female participation and interest in the IT field is IBM’s Women in Technology (WIT) workshops.  Half-day workshops have been held within the greater Vancouver region by female IBM staff and SWIFT volunteers.  To engage the female participants into thinking about computer science, each group of 10 girls is placed around a computer.  With the help of a facilitator, they develop their own group webpage.  There are also games aiming at testing participants’ knowledge of computer science and the IT field.  Feedback has been positive and plans are being made to extend the workshop to a full day one.

Post Secondary Programs and Courses 

Planning is currently underway at UBC for a new introductory level computer science course that will introduce students to fundamental computer science concepts through application in other disciplines. Students will learn about the principles of computer programming, how computer programs accomplish basic tasks such as organization and storage of data or pattern matching and how computer programs interact. Applications of these principles in Biology (such as constructing evolutionary trees or sequencing the genome), Psychology (such as modeling the brain, representing knowledge, or theories of language acquisition) and Fine Arts will be emphasized.  

ARC

The ARC (Alternate Routes to Computing) program is one of the main efforts made by SWIFT to make university computer science programs more accessible to, and more supportive of, women. The stimulus for ARC came from James Lau, Director of the IBM Canada Pacific Development Centre, who, in the fall of 1997, suggested the creation of an alternative computing program as a potential initiative for SWIFT. Mr. Lau was interested in increasing the number of IT workers who combined a significant level of knowledge and expertise in domains outside IT (e.g., biology, languages, and social work) with enough technical knowledge and expertise to contribute to software application development in those domains.

The information technology (IT) industry is predicted to continue to have a strong demand for highly trained people over the next decade. In the US the Bureau of Labor Statistics predicts that the demand for workers in IT will grow more rapidly than in any other field (Lazowska, 1999). Canada and the US are already experiencing a shortage of IT workers. However, despite many well-paid job opportunities, current participation in IT by women is low at all levels. The percentage of female undergraduate majors in computer science (CS) has dropped from 30-40% in the 1980's to 15-20% today (Kozen & Zweben 1998). CS is the only science and engineering field to experience a decline in participation by women over this period.

To address the skill shortage and low participation by women in IT, the University of British Columbia (UBC), Simon Fraser University (SFU), the IT industry, and the provincial government joined forces in December 1997 to develop a fast-track CS diploma program offering students academic and industry experience.  The goal of ARC is to provide a rapid path into IT careers for motivated individuals with a bachelor’s degree in any field, an excellent academic record, but little or no programming experience.

The steering committee that is responsible for developing the program comprises of faculty members of both universities who are responsible for undergraduate programs, administrators from co-op offices, student services, women students’ office, and female graduate students.  Representatives from the industry sector include Mr. Lau and a senior partner at Sierra Systems Consultants, Sonja Norman. The National Research Council and the BC Advanced Systems Institute joined the committee at a later stage.

In addition to designing the program to make it attractive to females, the committee also set out to raise funds for set up and the operation of the program.  Major sponsors include the BC provincial government and IBM.  Other companies like Sierra Systems Consultants not only support students during their co-op terms, but also make a $3,000 CDN cash contribution. 

The resulting ARC program is a two-year diploma program comprised of four four-month terms of university level courses (primarily in computer science) and an eight-month term of paid work experience in industry following the first two academic terms. Co-op programs alternate academic terms with paid work terms, making the program more affordable and offering valuable work experience for students.  The program was designed with the goal of attracting women, including mature female students who may have been out of the workforce for some time.  It is also based on standard undergraduate computer science courses so that students who wish to complete university degrees in computer science may do so.  A typical ARC course outline is shown below in Table 2: 

 

 

 

Academic Term

Course or Topic

Credit

Academic Term 1

 

Principles of Computer Programming

4

Introduction to Discrete Structures

3

Calculus I

3

Intro to object-oriented design and data abstraction

4

Academic Term 2

Computer Design

3

Discrete Structures part II

3

Computers and Society

3

Paid industry co-op term

0

Work Term 1

Paid industry co-op term

0

Work Term 2

Software Engineering I (recommended)

4

Academic Terms 3 & 4

Operating Systems I

3

Data Structures and Algorithms

3

Database Systems I

3

4 additional upper-level Computer Science courses

12

2 upper-level electives

6

 

Table 2: Typical ARC course outline

 

ARC 1998

A total of 205 complete applications were received by the May 15 deadline, with over 60% coming from women. The steering committee selected 78 applicants for interviews, based on the application packages: an application form, transcripts, a resume, and two reference letters. Although gender was not a determining factor in the selection process 66% of those interviewed were women. In addition to being interviewed, applicants took a standard IBM aptitude test to assess their ability to work with logic and symbolic patterns. The purpose of the interview was to assess each applicant’s communication skills, level of motivation and maturity, and ability to succeed in a program demanding high levels of time commitment and persistence. Forty students who had done well in the interview and the aptitude test were accepted into the program. Although gender was again not used as a factor in the selection process, over 70% of the accepted students were women.

The accepted students’ ages ranged from 22 to 51 with a mean age of 31. Their educational backgrounds included: anthropology, biology, business, chemistry, economics, engineering, English, geography, geology, genetics, nutrition, journalism, kinesiology, law, linguistics, mathematics, music, neuro-physiology, Spanish, paleontology, political science, and psychology.

Although 40 students had been accepted, two women withdrew early for personal reasons. After a one-day orientation session, the remaining 38 students began their classes in September with half attending UBC and half attending SFU. In addition to other support, feedback sessions were held during which faculty members and TAs listened to student issues and suggestions for improvements. The steering committee continued to meet approximately every two weeks to solve problems, complete the fund-raising, and plan the industry work term assignments. At the beginning of the second term, four student representatives joined the steering committee as full participants, with the exception of issues specific to individual students.

            Problems with ARC 1998. In this section some of the main difficulties encountered by students and organizers are listed.

·        Many students had difficulty with the math component of the program. A math refresher taken before beginning ARC would have been beneficial.

·        Students with no prior experience using computers initially had difficulties with basic computer operations like e-mail and printing. This engendered a lack of confidence and a feeling of being behind which persisted throughout the first two terms. Students felt a computer literacy course taken before starting ARC would have been beneficial.

·        Students felt they were not adequately informed of the time commitment needed for ARC. Stating it on the web page, in the interviews and in the orientation session was not enough.

·        The ARC web page, among other sources, had stated a minimum salary for the work term of $2500 CDN per month. The ARC sponsors who had committed to hire ARC students had agreed to this minimum; however there were 24 ARC sponsored spaces, compared to 29 ARC students. Since regular co-op employers set their own salary levels, it was difficult to ensure that all ARC students would achieve the stated minimum salary.

·        There have also been some problems in the co-op program. Some employers had unrealistic expectations of students’ knowledge about specific programming environments, while other work terms lacked challenge. The steering committee had to juggle shifting commitments on the part of sponsors and to find workable matches between companies and students.

            The positive outcome of ARC 1998. Despite the problems, virtually everyone involved in creating ARC feels the program is a great success. The 29 students remaining in the program have done well academically, achieving a grade-point average about 6% higher than the average achieved by students in the regular, already selective, CS programs. The ARC students are doing exceptionally well in their work terms and almost all are finding the work experience enjoyable, rewarding, and less difficult than the two previous academic terms. Employers are delighted with the students’ performance; they appreciate the students’ maturity, willingness to take on challenges, enthusiasm, and ability to learn. The ARC organizers find it most rewarding that the majority of students feel that ARC has changed their lives for the better. The ARC 1998 students will return to UBC and SFU for academic courses in January 2000. They will be outstanding mentors for the new group of ARC 2000 students behind them.

Two years in the making, and involving over 100 people, ARC has gone from a proposal to offer computer training to students from areas other than IT, to a CS diploma program training talented and enthusiastic people with little prior experience in computing. Although the program has been through some significant challenges, 29 people who never previously considered computing as a career are now headed to join the Canadian IT work force.

Moving on to ARC 2000

Due to the demand and success of ARC 1998, the ARC steering committee accepted another group of ARC students in January 2000. Several changes were made to the program to address the previously mentioned problems, including the following:

·        An application fee of $100 which helps to cover the administrative costs and ensures the seriousness of applicants.

·        The web page emphasizes the intense level of challenge and time commitment involved in the program.

·        All work term placements will be handled through the regular co-op process, with no promise of a minimum monthly salary. ARC students are required to take bridging workshops to help them succeed in the co-op process, and in transferring their skills to their new working environment.

·        More adequate funding is provided for staffing resources needed for the application, advising, and work term placement components.

·        All ARC 2000 students will attend classes at UBC for the first academic term and SFU for the second, making the program more efficient. The larger group will be potentially more cohesive overall, and hopefully minimize problems with group dynamics.

·        Applicants who do not have credit for four university math courses must complete a grade 12 math refresher within the two year period prior to their acceptance into the program.

·        Applicants must enter the program having a familiarity with basic computer operations.

ARC 2000 received 89 applications, 50% from women. Of the 62 students accepted, 56% are women. The applicants’ age range was 23 to 52 with an average age of 32 years. Although the number of students is less than the 1998 intake, their qualifications are significantly higher. The lower number of applicants may be due to the timing of the announcement and advertising, June to mid-August versus March to mid-May, and to the changes to the program: application fee, math and computer literacy prerequisites, emphasis on time commitments and challenge. These changes may also be responsible for the lower percentage of women applying.

Conclusion

Camp’s update on the status of the incredible shrinking pipeline (ISP) has good news and bad news.  The good news foresees a slight increase in the proportion of women earning CS degrees over what has occurred, and the bad, that women will continue to lag far behind their male colleagues. A sobering reflection is that if an equal number of women had joined their male counterparts in computing disciplines, there would not be the IT shortage problem stated in the recent ITAA study (Camp, Miller and Davies, 2000).

This current gender imbalance in technology and the role that technology will play in the future should be a concern for men and women, practitioners, policy makers and parents (Bunderson et al., 1995; Kirkpatrick and Cuban, 1998; Klawe and Leveson, 1995).  In this paper, we identified major causes such as gender-bias socialization, lack of access to and experience in computers, negative attitudes and low perceived ability that affect female students’ interest in pursuing IT as their career.  The various projects under the SWIFT program were launched not only to better understand gender differences in computer-related areas, but also to take concrete action to rectify problems that females are facing. The E-GEMS project attempts to alleviate the lack of interest and negative attitudes of girls towards computers in general and computer games in particular.  The Virtual Family project tries to help female students get beyond the rudimentary usage of computers to try their hands at programming.  This would lower their level of computer anxiety and increase their self-confidence and perceived ability in using computers. A new introductory level computer science course will serve to broaden interest and access to computer science. The ARC program aims at making IT as a career more attractive and accessible to women, and thereby trying to reverse the effects of gender-bias socialization and lack of experience of and access to computers. Outreach involvement such as classroom visits and IT workshops for girls constitute efforts to involve the larger shareholder community.

There are issues in program effectiveness and teacher preparation which will benefit from clarity around goals, objectives, criteria to measure outcomes, evaluations and incorporation of findings into the next delivery cycles.  There are clear indications for change in educational approaches for the teacher and the material/software emphasizing change in awareness, attitudes and interaction.  Industry partnering, as seen with the ARC program is essential.  In efforts to address issues of articulation, linking of secondary and post secondary education, again we see the need for efforts from the wider community. Involving girls in brainstorming, teaching and evaluating as partners in solution formulation would be essential. Involvement must be of sufficient interest and reward to all participants to make it sustaining.

Change of this magnitude depends on actively assisting groups to see the situation from many points of view. We are aware that our efforts alone are not sufficient to turn the tide, and that the solution lies in raising awareness in the larger community, including parents, schools, universities, IT employers, and the government.  

References

American Association of University Woman (AAUW) (2000). Tech-Savvy: Educating Girls in the New Computer Age. http:// www.aauw.org/2000/techsavvy.html. (April 27, 2000).

American Association of University Women (AAUW). (1999). Gender Gaps: Where Schools Still Fail Our Children. New York: Marlowe & Company.  

Aronson, E. (1999). The Social Animal. New York, Worth Publishers.  

Ayersman, D. J., Reed, W. M. (1995/1996). Effects of learning styles, programming, and gender on computer anxiety. Journal of Research on Computing Education. Vol. 28 Issue 2  pp. 148-161.  

Becker, H. J. and Sterling, C. W. (1987). Equity in school and computer use: National data and neglected considerations.  Journal of Educational Computing Research. Vol. 3 No. 3 pp. 289-311.  

Belenky, M.B., Clinchy, B.M., Goldberger, N.B. and Tarule, J.M. (1986). Women’s Ways of Knowing: The Development of Self, Voice and Mind. New York: Basic Books.  

Brosnan, M. J. (1998). The impact of psychological gender, gender-related perceptions, significant others, and the introducer of technology upon computer anxiety in students. Journal of Educational Computing Research. Vol. 18 No. 1 pp. 63-78.  

Brown, L.M. and Gilligan, C. (1992). Meeting at the Crossroads: Women’s Psychology and Girl’s Development. Cambridge, MA; Harvard University Press.  

Buchman, D.D. and Funk, J.B. (1996). Video and Computer games in the 90's: Children’s time commitment and game preference. Children Today. Vol. 24 No. 1 pp. 12-16. 

Bunderson, E. D. and Christensen, M. E. (1995). An analysis of retention problems for female students in university computer science programs. Journal of Research on Computing in Education. Vol. 28 No. 1 pp. 1-18. 

Busch, T. (1995). Gender differences in self-efficacy and attitudes toward computers. Journal of Educational Computing Research. Vol. 12 No. 2 pp. 147-158.  

Camp, T. (1997). The incredible shrinking pipeline. Communications of the ACM. October Vol. 40 No. 10 pp. 103-110.  

Camp, T., Miller, K., Davies, V. (2000). The incredible shrinking pipeline unlikely to reverse. The incredible shrinking pipeline unlikely to reverse. http://www.mines.edu/fs_home/tcamp/new-study/new-study.html. (April 7, 2000).  

Chappell, K. K. (1996). Mathematics computer software characteristics with possible gender-specific impact: A content analysis. Journal of Educational Computing Research. Vol. 15 No. 1 pp. 25-35. 

Ching, C.C., Kafai, Y. B. and Marshall, S. K. (2000). Spaces for change: gender and technology access in collaborative software design. Journal of Science Education and Technology. Vol. 9 No. 1 2000.  

Colley, A., Hill, F., Hill J., and Jones, A. (1995). Gender effects in the stereotyping of those with different kinds of computing experience. Journal of Educational Computing Research.  Vol. 12 No. 1 pp. 19-27.  

Colley, A. M., Gale M. T., and Harris, T. A. (1994). Effects of gender role identity and experience on computer attitude components.  Journal of Educational Computing Research. Vol. 10 No. 2 pp. 129-137.  

Comber, C., Colley, A., Hargreaves, D. J., Dorn, L. (1997). The effects of age, gender, and computer experience upon computer attitudes. Educational Research. Vol. 39 No. 2 pp. 123-133. 

D'Amico, M., Baron, L.J. and Sissons, M.A. (1995). Gender Differences in attributions about microcomputer learning in elementary School. Sex Roles. Vol. 33 Nos. 5/6 pp. 353-367. 

De Jean, J. and Upitis, R. (1995). Using CD-ROM Books and Paperbacks in a Grade 7/8 Poetry Unit. http://taz.cs.ubc.ca/egems/byAuthor.html

De Jean, J., Upitis, R., Koch, C. and Young, J. (1999). The Story of Phoenix Quest: How Girls Respond to a Prototype Language and Mathematics Computer Game. http://taz.cs.ubc.ca/egems/byAuthor.html

Dewey, J.  (1938).  Experience and Education.  New York: Collier Books. 

Dugdale, S., Dekoven, E., and Ju, M. (1998). Computer course enrollment, home computer access, and gender: Relationships to high school students’ success with computer spreadsheet use for problem solving in pre-algebra. Journal of Educational Computing Research. Vol. 18 No. 1 pp. 49-62.  

Fletcher-Flinn, C.M. and Suddendorf, T. (1996). Computer attitudes, gender and exploratory behavior: A developmental study. Journal of Educational Computing Research. Vol. 15 No. 4 pp. 369-392.  

Frenkel, A. (1990). Women and computing. Communications of the ACM. Nov. Vol. 33  No. 11 pp. 30-53.  

Friere, P.  (1974).  Pedagogy of the Oppressed. New York: Continuum.  

Funk, J. and Buchman, D.D. (1996). Children's perceptions of gender differences in social approval for playing electronic games. Sex Roles. Vol. 35 Nos. 3/4 pp. 219-231. 

Galbraith, M.C. (Ed.)  (1994). Facilitating Adult Learning: A Transactional Process. Malabar, FL:  Krieger. 

GenTech. BC Secondary School Enrolment Data by Grade, Course, and Sex: 1994/5, 95/6m and 96/7. http://www.educ.ubc.ca/faculty/bryson/gentech/data3yrs.html (April 5, 2000).  

Gilligan, C., Lyons, N.P., and Hanmer, T.J. (eds.)  (1990). Making Connections: The Relational Worlds of Adolescent Girls at Emma Willard School. Cambridge, MA: Harvard University Press. 

Graves, D. and Klawe, M. (1997). Supporting Learners in a Remote CSCL Environment: The Importance of Task and Communication. Proc. of CSCL '97, Toronto, Ontario. 

Inkpen, K., Booth, K. S., Klawe, M. and Upitis, R. (1995). Playing Together Beats Playing Apart, Especially for Girls. Proceedings of Computer Support for Collaborative Learning '95 (CSCL), Bloomington, Indiana. 

Inkpen, K., Upitis, R., Klawe, M., Lawry, J., Anderson, A., Ndunda, M., Sedighian, K., Leroux, S., Hsu, D. (1994). "We Have Never Forgetful Flowers in Our Garden": Girls' Responses to Electronic Games. Journal of Computers in Math and Science Teaching, Vol. 13 No. 4 pp. 383-403.  

Jensen, J. (1999). Girls ex machina: A school-based study of gender, culture and technology. Ph.D. Thesis, Simon Fraser University.  

Jones, T. & Clarke, V. A. (1995). Diversity as a determinant of attitudes: A possible explanation of the apparent advantage of single–sex settings. Journal of Educational Computing Research. Vol. 12 No. 1 pp. 51-64.  

Jordan, J.V., Kaplan, A.G., Miller, J.B., Stiver, I.P. and J.L. Surrey.  (1991). Women’s Growth in Connection: Writings from the Stone Center.  New York:  Guilford Press.  

Kahle, J. B. & Meece, J. (1994). Research on gender issues in the classroom in Handbook of Research on Science Teaching and Learning.  

Kay, R.H. (1992). Understanding gender differences in computer attitudes, aptitude, and use: an invitation to build theory. Journal of Research on Computing in Education. Vol. 25 No. 2 pp. 159-171. 

Klawe, M. M. (1999). Computer Games, Education and Interfaces: The E-GEMS Project. http://taz.cs.ubc.ca/egems/byAuthor.html

Klawe, M. M. (1999). Designing Game-Based Interactive Multimedia Mathematics Learning Activities. http://taz.cs.ubc.ca/egems/byAuthor.html

Klawe, M. M. (1998). When does the use of Computer Games and Other Interactive Multimedia Software Help Students Learn Mathematics? http://taz.cs.ubc.ca/egems/byAuthor.html. 

Klawe, M., Leveson, N. (1995). Women in computing: Where are we now? Communications of the ACM. January. Vol. 38 No 1 pp. 29-35. 

Klawe, M., Westrom, M., Davidson, K., Super, S. (1996). Phoenix Quest: lessons in developing an educational computer game for girls ... and boys. Presented at ICMTM96. http://taz.cs.ubc.ca/egems/papers/ICMTM96/ICMTM96.html

Knowles, M.S.  (1980). The Modern Practice of Adult Education – From Pedagogy to Andragogy.  Chicago:  Association Press. 

Koch, C. (1995). Is equal computer time fair for girls? A computer culture in a grade 7/8 classroom.  http://taz.cs.ubc.ca/egems/byAuthor.html.  

Koch, M. (1994). No girls allowed! Technos. Vol. 3 No. 3 pp. 14-19.  

Kolb, D.A.  (1984). Experiential Learning: Experience as the Source of Learning and Development.  New Jersey: Prentice Hall Inc.  

Kolb, D.A. and L.H. Lewis (1986). “Facilitating Experiential Learning: Observations and Reflections” in G.G. Darkenwald and A.B. Knox (eds.). New Directions for Continuing Education: Experiential and Simulation Techniques for Teaching Adults, no. 30.  San Francisco: Jossey-Bass.  

Liu, M. and Reed, W. M. (1992). Teacher education students and computers: Gender, major, prior computer experience, occurrence, and anxiety. Journal of Research on Computing in Education. Vol. 24 No. 4 pp. 457-468.  

McCarthy, B.  (1987). The 4Mat System: Teaching to Learning Styles With Right/Left Mode Techniques. Barrington, IL:  Excel Inc.  

Melemed, L. and I. Devine. (1988). “Women and Learning Style: An exploratory Study.” In P. Tancred-Sheriff (ed.) Feminist Research: Prospect and Retrospect. Montreal:  McGill-Queens University Press.  

Merriam, S.B. (1993).  “Taking Stock.” In S.B. Merriam (ed.). New Directions for Adult and Continuing Education.  No. 57 pp.105-110.  San Francisco: Jossey Bass.  

Miller, J.B.  (1984).  “The Development of Women’s Sense of Self”.  Work in Progress No. 12.  Wellesley, MA.: Stone Center Working Paper Series.  

Myers, J.  (July 1999). "Women in High Tech Fields in Science and Technology in British Columbia: Fact Sheet and Summary. Prepared for SCWIST / WISTTE Steering Committee." http://taz.cs.ubc.ca/wistte/exec_summary.pdf, March 2, 2000. 

Pearl, A., Pollack, M. E., Riskin, E., Thomas, B., Wolf, E., and Wu, A. (1990).  Becoming a computer scientist. Communications of the ACM. Vol. 33 No. 11 pp. 47-57.  

Philbin, M. et al. (1995). A survey of gender and learning styles. Sex Roles: A Journal of Research.  Vol.32  No.7-8  pp.485-94.  

Reinen I., J. and Plomp, T. (1996). Gender and New Technology. International Encyclopedia of Educational Technology 2nd Ed. 

Saltonstall, J.F.  (1989).  “Learning About Learning:  A Study of Women’s Ways of Learning and Being in a Formal Educational Environment.”  Ph.D. diss., Harvard University. 

Sanders, J. (1995). Girls and technology: Villain wanted. In S. V. Rosser. Teaching The Majority: Breaking the Gender Barrier in Science, Mathematics, and Engineering. New York: Teachers College Press.  

Schofield, J. W. (1995). Computers and Classroom Culture. Cambridge, USA: Cambridge University Press.  

Sedighian, K. and Klawe, M. (1996). Super Tangrams: A Child-Centered Approach to Designing a Computer Supported Mathematics Learning Environment. ICLS '96.  

Severiens, S. and Geert, T.D. (1997). Gender and Gender Identity Differences in Learning Styles.  Educational Psychology. Vol.17  No.1-2 pp.79-93.  

Shashaani, L. (1997). Gender differences in computer attitudes and use among college students. Journal of Educational Computing Research. Vol. 16 No. 1 pp. 37-51.  

Shashaani, L. (1994). Gender-differences in computer experience and its influence on computer attitudes. Journal of Educational Computing Research. Vol. 11 No.4 pp. 347-367.  

Steinem, G.  (1992).  Revolution From Within:  A Book of Self-Esteem. Toronto: Little, Brown and Company. 

Taylor, K. and Marienau C. (eds.) (1995). Learning Environments for Women’s Adult Development: Bridges Toward Change. New Directions for Adult and Continuing Education, No.65 pp. 5-12. 

Tisdale, E.J. (1993). Feminism and Adult Learning: Power, Pedagogy and Praxis. New Directions for Adult and Continuing Education, No. 57 pp. 91-104. San Francisco: Jossey-Bass.  

Underwood, J. and Underwood G. (1999). Task effects on co-operative and collaborative learning with computers. In K. Littleton & P. Light (Eds.). Learning with computers. New York: Routledge.  

Upitis, R. (1995). From Hackers to Luddites, Game Players to Game Creators: Profiles of Adolescent Students Using Technology. http://taz.cs.ubc.ca/egems/byAuthor.html. 

US Department of Education, National Center for Education Statistics (1998). The 1994 high school transcript study tabulation: Comparative data on credits earned and demographics for 1994, 1990, 1987, and 1982 high school graduates, Revised. NCES 98-532, by Stanley Legum, Nancy Caldwell, Bryan Davis, Jaqueline Haynes, Telford J. Hill, Stephen Litavecz, Lou Rizzo, Keith Rust, and Ngoan Vo. Project Officer, Steven Gorman. Washington, DC.  

Volman, M. (1997). Gender-related effects of computer and information literacy education. Journal of Curriculum Studies. Vol. 29 No. 3 pp. 329-349.  

Weil, S.W. and I. McGill. (eds.)(1989). Making Sense of Experiential Learning: Diversity in Theory and Practice. Milton Keynes: Open University Press.  

Woodrow, J. E. J. (1994). The development of computer-related attitudes of secondary students. Journal of Educational Computing Research. Vol. 11 No. 4 pp. 307-338.  

Yelland, N. (ed.)(1998). Gender in Early Childhood. New York: Routledge. 

Young, J. and Upitis, R. (1999). The Microworld of Phoenix Quest: Social and Cognitive Considerations.  http://taz.cs.ubc.ca/egems/byAuthor.html.