Communications of the ACM

Contributed articles

Crossroads For Canadian CS Enrollment

By Jacob Slonim, Sam Scully, Michael McAllister
Communications of the ACM, October 2008, Vol. 51 No. 10, Pages 66-70
10.1145/1400181.1400199
Comments

The decline in computer science enrollment in practically all Canadian universities has resulted in a shortage of technical graduates that in turn jeopardizes the ability of the Canadian information and communication technology (ICT) sector to remain vibrant, innovative, and competitive in the global economy. The sector's anticipated wave of retirements will exacerbate the situation in the next five to 10 years. With upward of 30% of Canada's GDP directly or indirectly involving ICT, university CS educators must rejuvenate the discipline and rethink the country's long-range strategy to teaching CS to ensure continual renewal of the ICT work force. While some preliminary signs indicate that enrollment levels in 2007–2008 are stabilizing, the public, private, and education sectors must coordinate their approaches to ICT recruiting, training, and development to achieve the required renewal.

We collected CS enrollment data (1999–2007) for most Canadian universities to determine the quantitative extent of the enrollment decrease and personally interviewed many CS department chairs and senior university administrators to learn their views on the causes of the decline. Each university meeting we conducted began with a report on the enrollment decrease for that university, a discussion structured around three basic questions:

• What is the reason for the decline in your university's CS enrollment?;
• How well prepared are the students entering CS in your university?; and
• What actions, if any, is your university taking to mitigate the decrease in enrollment?

We followed with a discussion regarding the personal views of the interview participants about the decline in enrollment.

We interviewed universities until all regions in Canada were represented and the interviews became repetitive as to the information they were able to disclose.

We also interviewed five large industry partners—Business Objects, CIO of Quebec, IBM, Research In Motion, and SG1—for their views on the state of the Canadian ICT sector; the interviews were secured through a list we obtained from Industry Canada as the sponsor of the study. We asked the industry partners whether they are seeing any shortage in the ICT labor market that might be attributed to the enrollment decline and whether they had plans to outsource work as a result. An open discussion again followed the questions.

More important than explanations of past enrollment decreases is the design and articulation of initiatives that mightreverse the trend. If the decline is cyclic (see Figure 1), as some of our colleagues in a number of North American universities suggest, then taking immediate steps to reengage students in CS can shorten the duration of the current trough, hasten recovery, and lessen the depth and impact of future troughs. If the decline is not cyclic, as we believe it is, then action (such as inclusion of the societal impact of computing, promotion of the discipline, and reexamination of CS curricula in light of the current economic and social environment) is necessary to rejuvenate CS enrollment throughout North American universities.

Back to Top

Enrollment Status

While we focus on data from Canadian universities, enrollment data from the Computing Research Association's Taulbee reports suggest common CS enrollment trends in both Canada and the U.S.^1,6,7 We studied enrollment figures available from Statistics Canada⁴ that were normalized to report full-time equivalent students; part-time students were counted as a fraction of an enrollment of a full-time student. We augmented the data sets with (non-normalized) enrollment figures from 32 major Canadian universities (1999–2007) provided by the 32 affiliated CS departments (see Figure 2).

We were surprised by the range of enrollment declines across the various regions of Canada. The largest was in Atlantic Canada, with a drop to 37% of its 2001–2002 peak. British Columbia was the exception, at just over 97% of its 2001–2002 peak in 2006–2007. The remaining regions were in between: Ontario and Quebec were between 50% and 60% of their peaks, and the three prairie provinces—Alberta, Manitoba, and Saskatchewan—were near 65% of their peak. These declines are resulting in much smaller graduating classes through at least 2011, thereby adding pressure on hiring in the ICT sector.

Universities in British Columbia attribute part of their success maintaining CS enrollment levels to several actions: First, the University of British Columbia introduced a new multidisciplinary program within its Faculty of Science that added almost 100 students to CS. Previously, in the 1990s, the same university also began to emphasize recruitment of women students through the Supporting Women in Information Technology program (www.cs.ubc.ca/labs/swift/) and Alternate Routes to Computing program (www.arc.cs.ubc.ca/) that attracted individuals with prior learning to CS. These programs have given the university the highest percentage of women pursuing IT degrees in Canada, ranging from 23% to 26% of 850–1,000 students from 2000 to 2006. Second, the province of British Columbia restructured its university landscape, producing a one time influx of 200 students from a previously uncounted source in the Statistics Canada data.

Many institutions have compensated for declining undergraduate enrollment by reducing admission standards and/or increasing the size of their master's and Ph.D. programs (see Figure 3). However, the master's programs are also now beginning to see decreased enrollment as graduating undergraduate classes shrink. Canadian undergraduate graduates declined from 4,900 in 2003–2004 to 3,300 in 2006–2007 and have continued to decline.

The reported explanations for this decrease are not new. The CRA has had a group of CS chairs and deans examining the issue since 2000. The people we interviewed echoed the reasons cited in reports from Australia² and Western Europe.⁵ The interviews identified common themes across Canada, along with the occasional distinguishing reason specific to a region. We categorize the commonly cited reasons as perception, preparation, and curriculum design and report here the ideas raised in more than 50% of our university interviews.

A negative view of CS, typically false, arose in a number of contexts:

• Unhealthy ICT sector in the face of outsourcing;
• Fewer opportunities and lower salaries in the ICT sector;
• Narrow view of the types of problems addressed in the discipline;
• Lack of recognition of the breadth of CS application areas;
• Requirement of, and emphasis on, the discipline's mathematical sophistication;
• Stereotypes of the discipline's students; and
• Earlier CS frontiers people now take for granted.

The relative ease of use and ubiquity of general-purpose computing also mask the challenges that have yet to be solved in CS. As computers have become commonplace, students are increasingly less motivated to study them as a discipline.

Negative perceptions of the discipline and the focus on machines rather than on people have been particularly detrimental in attracting women to the field, so student recruitment is effectively being drawn from significantly less than the full high-school population.

First-year students are often cited as ill-prepared for the mathematically and science-based approaches common in computing in the majority of CS university programs. Explanations for this lack of preparation include a lack of opportunity to be introduced to the discipline in high school; a shortage of teachers with an adequate understanding of CS; a dearth of computing infrastructure in Canada's high schools; and overwhelmed guidance counselors unable to keep up with the ever-changing discipline or offer comprehensive guidance about IT careers and options.

CS programs in North American universities often exhibit low undergraduate retention rates from the first to the second year. Whatever the reason—incorrect perception of CS, lack of preparation, or loss of interest due to the curriculum—the low retention rate fosters the perception that the discipline is difficult and requires even higher student recruitment numbers.

Finally, CS curricula in North American universities have not adapted to the incoming students or to changes in the industry; CS curricula go stale quickly, and specific topics can be seen by potential students as irrelevant. First-year courses often focus on teaching a computer language to address toy problems like temperature conversion, elevator operations, and simple report generation. The curricula emphasize mathematical accomplishment and typically develop from first principles. Also, Canadian universities often lack mandatory exposure to technology across all their academic programs, leaving fewer students exposed to computing as a discipline on its own or as a first-class contributor to or enabler of multidisciplinary work. We elaborate on the data and the causes of declining Canadian CS enrollment in a separate report.³

Back to Top

Looking Ahead

Canadian universities have reacted to declining CS enrollment through recruitment programs, a new area for CS, along with innovative educational programs. These efforts include outreach, variations on first-year courses, joint programs with other disciplines, special adaptation to teaching and learning environments, and more focus on recruiting and retaining students. However, they have not been sufficient. Needed still is a concerted and coordinated Canadian national action plan among the public, private, and education sectors.

An October 2007 IBM Centre for Advanced Studies conference workshop (www-927.ibm.com/ibm/cas/cascon_main/) provided a brainstorming opportunity for 100 industry partners (organized into 10 groups) from throughout Canada, federal and provincial government leaders, CS professors and department chairs, and high school educators. Just as there is no single explanation for declining enrollment, there is no single remedy. The workshop identified the responsibilities all participants needed to accept to restore CS enrollment.

Canada's 10 provinces and three territories have constitutional responsibility for education at all levels, while the federal government focuses on the national health of industry sectors. Thus, federal and provincial responses to boost CS enrollment must differ. The brainstorming workshop (morning session) produced a "top 10" list of directions that might help mitigate the decline:

• Change the high-school curriculum and pedagogy;
• Change the university curriculum and pedagogy;
• Use and promote co-op internships and role models more broadly;
• Improve students' perceptions of CS as a discipline;
• Develop a national advertising and awareness campaign;
• Engage a national society to lobby Canadian federal and provincial governments and function as a repository of teaching resources;
• Win the commitment of the provincial ministers of education to address the specific challenges in CS education at a joint meeting of ministers;
• Change university admission policies to recognize CS as a formal part of the academic high school curriculum;
• Recruit and retain more students in CS; and
• Leverage the interest of high school and university students already in CS to develop new and engaging programs.

---

All must focus on changing the public's perception of the industry and its employment opportunities and adapt the CS curricula to remain engaging and relevant for all Canadian students.

---

The afternoon session developed action plans for each item in each sector:

Private sector. The private sector must promote the demand, availability, and diversity of employment opportunities. It also must set the standards for salaries that, in part, fueled the earlier rush to ICT by some students. These factors make it the most logical and effective sector to spearhead promotion of ICT to the public. The private sector must also help fund the activities that are needed to promote ICT, requiring financial, in-kind, and time commitments from industry partners. The private sector also includes the role models most able to demonstrate success in the ICT sector to which students and their parents can relate and aspire.

The curricular advice from industry partners must also emphasize the interest of the ICT sector as a whole rather than as specific to individual companies or specializations. These partners must help introduce new models of education that leverage and value the hands-on experience available through centers like the IBM Centre for Advanced Studies (www-927.ibm.com/ibm/cas/) and the Business Objects (an SAP company) Advanced Academic Research Centre (labs.businessobjects. com/arc/).

Public sector. As with the private sector, the public sector must also support the activities that promote ICT, especially at the federal level, whether through financial support, in-kind contributions, or information. The federal government, through the Natural Sciences and Engineering Research Council of Canada (www.nserc.gc.ca/index.htm) and other research agencies, must also increase its funding for CS research to create and sustain dynamic projects that will attract more graduate students to the discipline.

The provincial ministries of education must provide a consistent and relevant curriculum to primary and secondary schools, recognize CS as a teachable academic subject in all 10 provinces, and promote and support the continued development of CS teachers in an area that is always changing.

The federal and provincial governments must also help communicate and promote the number and diversity of employment opportunities in the ICT sector, providing incentives for students to study the discipline as a cornerstone of the national and international knowledge economy.

Universities. Universities must adapt their curricula to engage and retain new students and provide multidisciplinary programs that are attractive to students and industry alike. Moreover, they must create an environment and teaching style that is more appealing to women, introducing new notions of creativity into their programs, and improve retention rates through better engagement of students.

The very structure of university CS programs must also change. Foundation CS courses in the first and second years must capture the breadth of CS; meanwhile, specializations (in both university programs and industry) rely on only a subset of the core, branding the unreferenced material irrelevant. Also, many first-year CS programs assume that students lack a worthwhile computing background on which to build. Universities must rethink what they expect from their core courses and simplify advanced placement in their programs.

Universities are positioned to provide role models (including students, teachers, and practitioners) at the leading edge of technology. They must be excited about the discipline and find ways to convey to their audience the exciting directions in which it is heading. They must also be able to connect their audience with how work in the ICT sector affects the social and economic aspects of daily life.

Universities must also adapt their own regulations to recognize the importance of CS in the competitive global economy by, say, including a beginning course in all university programs, transfer credits or advanced placement into CS programs based on high-school CS courses, and basic CS concepts in all degree requirements, even the humanities. They must therefore revise their silo-based organization to facilitate and encourage multidisciplinary education and programs.

High schools. The high school partners across the Canadian educational system preparing students to consider ICT as a possible career path could now take several new directions: First, they could offer CS courses that introduce students to computing concepts beyond the applications in the students' own studies (some already do this) but must provide consistency and continual updates. Delivery of the material, as well as the resources for teaching it, must engage the students. The high schools, and likely the primary schools, must also find ways to keep women connected with all the sciences as they reach grade 12.

Back to Top

Conclusion

The products and services of the ICT sector represent a cornerstone of the developing global knowledge society. ICT is one factor increasing the productivity of the Canadian work force; ICT must continue to strengthen this productivity if Canadian businesses are to remain competitive. Sustained success in the ICT sector, essential to Canada's national interest and prosperity, requires that all participants—public, private, and educational—continually adapt to changes in technology and its applications. This requires a Canadian national work force that is dedicated, informed, sophisticated, and agile.

For the sake of the overall Canadian economy, it is imperative that the country's public, private, and educational sectors all play a role in increasing the numbers of students studying CS in high school and university in order to sustain the required ICT work force. All must focus on changing the public's perception of the industry and its employment opportunities and adapt the CS curricula to remain engaging and relevant for all Canadian students.

All must invest time and money in solving the current crisis of falling Canadian CS enrollment. Failure to do less than what we propose here would harm future generations. That investment must include stable, long-term financial support for CS education initiatives that recognize the time commitment required of the leaders in all sectors.

Back to Top

Acknowledgments

We would like to thank the people who helped us prepare this article:

• The chairs of Canadian university CS departments for their participation, openness, and feedback in discussing and quantifying the issues facing the Canadian ICT sector;
• The industry partners who met with us for our on-site interviews concerning their perspective on the direction of the discipline;
• Paul Swinwood of the Information and Communications Technology Council (www.ictc-ctic.ca/) for helping sponsor our work;
• Business Objects, an SAP company, IBM, and Research In Motion for providing additional industrial contacts; and
• The Natural Sciences and Engineering Research Council of Canada (www.nserc.gc.ca/index.htm).

Back to Top

Back to Top

Back to Top

Back to Top

Back to top

The Digital Library is published by the Association for Computing Machinery. Copyright © 2008 ACM, Inc.

---

No entries found