While enrollments in computing courses and degree programs vary among U.S. institutions of higher education, degree production has been surging on average since 2008, and many academic institutions are experiencing growth that has reached or exceeded the capacity of their departments. Furthermore, with the changing role of computing in society and the rapid evolution of computing fields and technologies, the committee sees both an urgent need and an opportunity to strategically evaluate the role of computing at academic institutions and to plan for a compelling future where student, departmental, institutional, and national needs can be met.
To this end, the committee makes the following recommendations:
RECOMMENDATION 1: The leaders of the institutions of higher education that have experienced rapid increases in computer science course enrollments should take deliberate actions to address this trend with a sense of urgency.
The increased interest and demand should be viewed as an opportunity for the institution to reassess the role of computing in general, now and in the future, for an inclusive student audience. With the institution’s mission and values as context, leaders should develop plans that address the needs of non–computer science (CS) students as well as CS majors—both traditional and those in emerging programs—and opportunities for interdisciplinary collaboration.
RECOMMENDATION 2: A range of actions should be considered as part of a comprehensive institutional strategy, from targeted controls on enrollments or resource additions to meet demand, to more
extensive institutional changes that extend beyond the computer science department.
Chapter 6 provides a template that can help institutions take stock before considering potential actions, several of which are also discussed in that chapter. While responses should and will vary by institution, there are some common issues and elements that the committee calls out here. The committee also notes that there is no crisp delineation between short-term tactics and long-term strategies, or between local actions and broader institutional initiatives. Indeed, what may seem to be a short-term, local “fix” can have long-term consequences that affect the entire institution.
Universities should develop ways to manage fluctuations in enrollments, without following enrollment booms and busts too closely, as they may be unpredictable or unrelated to long-term trends. They should settle on reasonable targets for long-run growth in computer science faculty and computationally trained faculty, aim for those targets (which will vary among institutions), and remain open to adjustments over time. There is no single approach—a silver bullet—for responding to enrollment growth that will be optimal at all institutions. Every tactic has benefits and costs. Leaders will need to select strategies and make trade-offs that are appropriate to their circumstances and to their institution’s mission and values, monitor the results over time, and update strategies as needed.
RECOMMENDATION 2.1: Institutions experiencing a computer science enrollment surge should seriously consider an increase in resources to address the rising workload on faculty and staff in computer science and related departments, and the limitations arising from inadequate facilities.
The key resources supporting and enabling a department’s teaching mission—faculty, teaching staff, teaching assistants, facilities, and support staff—are all dependent on a department’s budget. An institution’s capacity to respond to increased student demand for computing courses is limited by constraints on any one of these resources, as discussed in Chapter 6. At the same time, institutions need to recognize the challenge computer science departments face in hiring qualified tenure and tenure-track faculty and teaching faculty, and the vast opportunities for such individuals in the private sector. When faculty hiring is not possible, resources would make a difference to enable hiring of undergraduate or graduate students or staff to increase instructional capacity and help manage large course and advising loads. Another strategy could be to provide additional compensation to those faculty taking on teaching loads that are significantly larger than those typical of other faculty across the institution.
RECOMMENDATION 2.2: Some institutions may view the imposition of limits on enrollment in computer science and related courses as desirable or unavoidable. However, before imposing limits on
course or major enrollments, the consequences of doing so should be considered comprehensively, and the benefits and costs weighed for the entire university community.
Limiting course enrollments may cut off students from their true passion. It may also introduce stress or an environment of real or perceived competition for students who desire to enter a CS program, which could discourage participation among underrepresented groups. At the same time, institutions should not accept students with the promise of entering a major when the constraints on resources make their admission into the program unlikely. Any measures to limit major admission or declaration can result in reduced demand for more advanced courses and can affect the overall climate in such courses, and could lead to unexpected declines in overall degree production. As discussed in Chapter 5, specific strategies for limiting enrollments could also have the effect of disproportionately limiting the opportunities of underrepresented minorities and thus decrease diversity in the field.
Limiting students’ opportunities for taking CS courses may also affect the educational experiences of non-majors. Given the evidence of wide interest among non-CS majors, increasing requirements of CS courses for other majors, and increasing demand for computing skills across the workforce, institutions must understand and consider how limitations on CS enrollments will affect their entire student body, across all programs.
RECOMMENDATION 2.3: Institutional leadership should engage directly with computer science departments or programs to develop appropriate faculty hiring and faculty size targets, and develop strategies to improve faculty retention. Increasing the number and enhancing the role of academic-rank teaching faculty should be given serious consideration.
As institutions assess the role of computer science and the interests and needs of students, the reality of CS faculty hiring challenges and the opportunities available to new CS Ph.D.s need to be understood. As programs grow, the environment and workload for faculty needs to be taken into consideration. Institutions need to be proactive to help reduce the faculty retention challenges that departments face. Furthermore, the relatively small number of CS Ph.D.s pursuing an academic career, coupled with the broad opportunities for new Ph.D.s as well as CS faculty in the private sector, may impact faculty growth goals set by departments and institutions. While increases in today’s CS bachelor’s degree production will likely increase the number of Ph.D.s produced at some point in the future, this downstream growth of faculty potential will not provide a solution to the hiring problem in the near or mid-term. Furthermore, placing limits on CS major enrollments will likely prolong this challenge.
Larger departments have been increasingly hiring academic-rank teaching faculty to help deal with course enrollments and the increasing number of majors,
but even this can be challenging. Departments need to implement effective strategies to mentor, evaluate, and promote this new type of faculty, including tenure practices. This includes providing professional development as well as integrating teaching faculty into departmental faculty activities, defining the expected scholarship and leadership activities, and defining best practices for their success.
RECOMMENDATION 2.4: Larger institutions—in particular, research universities—should reevaluate the organizational placement of the computer science department and other departmental units with a computational mission.
In particular, institutional leadership should consider whether a College of Computing could help to improve organizational efficiency and benefit its students. The organizational placement of the CS department can have significant impact on the computing education for all students. The organizational structure also impacts approval and reporting channels, and can have significant impact on how quickly and effectively a department can respond to important changes. The appropriate structure should reduce duplication in courses and programs; coordinated programs can lead to better graduation rates.
RECOMMENDATION 2.5: Institutions should pursue innovative strategies for using technology to deliver high-quality instruction at scale to large numbers of students. Institutions should also pursue additional, creative strategies for meeting demand for quality computer science courses and skills development among the entire student body.
In this era where the demand for computer science is high, CS is a natural discipline for such experimentation and innovation. While this is certainly under way at some institutions (and is the basis for the massive open online course phenomenon—though this approach has shown mixed results), there are further opportunities for using novel technologies to deliver high-quality instruction at scale. Computing researchers should make use of advances in their own fields and work with education researchers to better understand how students learn and how to improve the efficacy and quality of automated instruction, labs, examinations, and grading in undergraduate computing education. Institutions should support such collaborations by providing needed startup resources. Additional strategies could include using qualified faculty outside CS to cover CS classes, bringing back retired faculty, and offering more summer classes covered by teaching faculty.
RECOMMENDATION 3: Institutions should take deliberate actions to support diversity in computer science and related programs. In particular:
RECOMMENDATION 3.1: Institutions should assess how computer science enrollment growth, and any actions or strategies for responding to it, affects the diversity of their student bodies, and deliberately align their actions and the culture of their programs with best practices for diversity and retention.
As discussed in Chapter 5, there is research from which institutions can draw when considering their strategies. Actions understood to have a negative effect on the participation of women and underrepresented minorities in CS or science, technology, engineering, and mathematics (STEM) should be avoided. Methods for managing growth while actively increasing diversity should be sought.
RECOMMENDATION 3.2: Institutions should leverage the increasing interest in computer science and related fields, among both non-majors and intended majors, to engage, recruit, and retain more women and underrepresented minorities into the field and help address the diversity problem proactively.
While the current enrollment growth is a challenge for departmental resources, it comes with the opportunity to take advantage of the broad excitement around CS in order to help shape the future of the field, and who will be included.
RECOMMENDATION 4: The National Science Foundation (NSF) can be especially helpful in advancing undergraduate computer science education in the context of increasing enrollments, for both majors and non-majors. The following actions should receive serious consideration:
RECOMMENDATION 4.1: Use NSF’s convening power to bring computer science faculty and institutional leaders together to identify best practices and innovation in computer science education in times of limited departmental resources. This should include assessment of the computer science skills and knowledge needed in non–computer science disciplines.
RECOMMENDATION 4.2: Support research on how best to use technology in teaching large classes. Such research should be multidisciplinary, spanning learning sciences, educational pedagogy for computer science, development and deployment of assessment instruments, and technology design.
RECOMMENDATION 4.3: Support research to advance the understanding of best practices for diversity in computing, including rigorous and longitudinal assessment of the efficacy of specific institutional practices, especially those taken or considered in times
of high enrollments. This research should be multidisciplinary, with experts in both micro- and macro-level social science research, statistics, computer science education, and diversity in STEM and computing.
RECOMMENDATION 4.4: Create an initiative to expand instructional resources in computer science, informed by an understanding of the constraints and dynamics of the supply and demand for computer science Ph.D.s. This might include research support and doctoral fellowships for domestic computer science undergraduates, and support for incorporating teaching into computer science doctoral programs and junior faculty research.
RECOMMENDATION 5: Computer science departments and the computing industry should develop new partnerships to help higher education meet workforce needs, continue to graduate well-prepared students, encourage industry to provide increased support for research funding, and allow a better exchange of Ph.D.-level researchers between academia and industry.
Academic institutions should pursue new partnerships with industry and to enable CS professionals in the private sector with academic experience to pursue an academic leave. This could open the door to teaching for individuals from industry, provide additional industry experience for CS courses, and provide industry an additional recruiting pipeline. Students would similarly benefit from the opportunity to network with professionals from the private sector.
RECOMMENDATION 6: Public institutions produce a significant fraction of each state’s workforce and the nation’s computer science undergraduate degrees. States should provide sufficient support to their public institutions to enable them to support fully their academic missions, including with respect to computer science education.
The widespread disinvestment by states in higher education has had a negative effect on all that these institutions do. Responding to high demand for CS education is extremely difficult when institutional budgets decline, as they have in most states over the last decade. Furthermore, strengthening CS programs can have positive benefits for regional economies.
RECOMMENDATION 7: To prepare students better for the expanding role of computing in academia, industry, and daily life underlying the increase in interest in computer science, government agencies and states should support local, state, and national programs for computing education for the purpose of increasing expo-
sure to computing, computational principles, information security, and data analytics throughout the K-12 pipeline.
Such initiatives should be informed by an understanding of which computer science principles and methods are most important in different academic disciplines and in daily life. They should also be linked to evidence about the CS and computing skills in demand in different industry sectors and occupational fields. These topics are important areas for future study.
RECOMMENDATION 8: Actions should be taken to facilitate an improved understanding of national undergraduate enrollment trends by improving the primary data available about them and facilitating the availability of that data in a timely fashion. In particular, the following actions should be considered:
RECOMMENDATION 8.1: Improved data sources about undergraduate enrollment should be pursued by federal and state governments in collaboration with academic institutions. To the extent possible, data should be made available in a time frame where the information can be useful for academic and government planning purposes.
This should include authoritative, longitudinal, and either comprehensive or representative data about course enrollments, declared majors, degree requirements, and the organizational unit responsible for administering degree programs. Data on enrollment trends should include details on the impact of pedagogical approaches and institutional environment on recruitment and retention of women and underrepresented minorities and the drivers of non-major enrollment in CS courses. Such data should be obtained at both the macro level, using rigorous sampling techniques to ensure representative data, and the micro level, by interviewing individual students and programs for a better understanding of individuals’ interests.
Such timely information would make it easier to track trends in higher education and predict degree production and workforce trends. Larger institutions and those with institutional research offices may have appropriate resources and infrastructure to support this objective; state higher education systems could play a role in coordinating such efforts to minimize the burden on institutions that lack such resources.
RECOMMENDATION 8.2: The taxonomies and classifications for undergraduate computing degrees and jobs should be reexamined and updated, so that those used in national statistics are more easily brought into alignment, and map more directly to the current organization of computer science and related fields in higher education.
In particular, the Integrated Postsecondary Education Data System (IPEDS) Classification of Instructional Program (CIP) “Computer Science” classifier and the Standard Occupational Classifications (SOCs) used by the Bureau of Labor Statistics (BLS) should be brought into better alignment. This could be done by
- Revising the CIPs and SOCs to that they map to each other more directly, or
- Creating a clearer crosswalk, perhaps identifying in the SOC background information how the SOC relates to CIPs, preferably while maintaining a continuity of data that continues to enable analysis of long-term trends.
Identifying the preferred level of CS education associated with various CS occupations would also be helpful.
In addition, the computing-related CIPs should be reorganized, renamed, and regrouped to reflect the current organization of these fields in higher education. For example:
- The existing “detailed” CIP of “computer science” should be reexamined and potentially redefined to include fields such as computer engineering and software engineering, and to make “information technology” a distinct or more clearly grouped category.
- Guidance given to institutions about how to classify CS and related programs could be clarified to enable accurate data reporting and analysis.
- Efforts could be made, perhaps through interviews with individual institutions, to map the classification of historical data to current CIPs to enable continuous analysis of temporal trends in the evolution of CS and related fields.
- Information could be collected about the organizational placement of these degrees within an academic unit to enable analysis of the emergence and prevalence of different institutional models.
This would make it much easier to correlate undergraduate degree production with labor market trends, and to assess emerging trends in the landscape of computing.
RECOMMENDATION 8.3: In the absence of comprehensive national statistics, the computer science community, in collaboration with education, social sciences, and statistics researchers, should continue to pursue or refine effective strategies for tracking enrollment, retention, and graduation rates and measuring student diversity.
In particular, these strategies should be based on surveys that are either comprehensive or based on representative and statistically rigorous samplings across
all institution types, with questionnaires crafted (and potentially administered) by (or with the guidance of) experts in survey design. The availability of such data in a timely fashion would be invaluable for informing academic institutions and federal agencies of current trends, and enabling these actors to make informed decisions in a reasonable time frame. Given the increasing requests for data made to institutions of higher education by state and federal agencies, accrediting groups, professional organizations, associations to which they belong, and organizations to which they subscribe, efforts to streamline such requests or use a common format would help to minimize the burden of such requests.
This page intentionally left blank.