This article is published in the August 2020 issue.

CUE.NEXT: Envisioning the future of computing in undergraduate education



Computing and Computer Science have become relevant to undergraduate education in all disciplines. Academic institutions are challenged to meet the demand of the growing and increasingly diverse student body seeking to learn more about computing, computer science, and the role of computation in their own disciplines. Courses and curricula aimed at teaching CS majors generally do not meet the needs of this growing student audience.

Three NSF-funded workshops were held between November 2019 and January 2020 to initiate a national dialog on envisioning the future of computing in undergraduate education (CUE).[1] Applications were solicited from teams of 2 to 5 faculty, educators, or administrators.  Each team including at least one member from a computing-centric department (including CS, CE, ECE, IS) and at least one member from a non-computing centric discipline. The three workshops drew a total of 201 participants forming 50 teams from 66 institutions, with slightly more than half the participants (106) from non-computing departments. The majority of the participants came from Ph.D. granting institutions; non-Ph.D. or -M.S. granting institutions included ten 4-year liberal arts colleges, four HBCU’s, one community college, and one Hispanic serving institution. This article summarizes perspectives, challenges, and potential strategies around a variety of themes that emerged during the discussions. For the full workshop report see

1. Curriculum Design

1.1.   Approaches to CUE Curriculum Design

The most often-heard participant comments about creating and implementing successful CUE curricula are itemized below.

Non-computing participants observe:

  • Have a low barrier approach to contextualized computing, with a focus on the impact on humanity and society.
  • Have shorter pathways and shorter prerequisite chains.
  • Have ongoing communication between CS and other domains to help CS faculty understand the computational needs of non-computing disciplines.
  • Recognize CUE-like efforts that are already happening in some non-computing departments (media, arts, biology, etc).
  • Recognize that computing is one family of problem-solving techniques among many, and that discipline-specific knowledge and skills are needed.
  • Recognize that CS students in non-CS courses often fail to engage with the other discipline. Just as non-CS students can be turned off by coding, CS students can be turned off by writing, or artistic expression.
  • Stop using CS1 as a weed-out course; stop making computing/programming look scary.
  • The need for strong programming skills is contested in some disciplines. Most people in non-CS disciplines may not program much.

Computing participants observe:

  • Students from other disciplines want to enroll in CS courses for which they don’t have the prerequisites. This particularly affects courses in machine learning and AI.
  • The programming languages most appropriate for other disciplines change over time, and it is time-consuming to adjust curriculum to track this.
  • Most people in non-CS disciplines may not program much, but knowledge of programming is essential if quality computational tools are to be created.
  • Data-centric computing might offer an alternative path into high-level computing courses with the right mixture of topics.
  • Student workload has to be reasonable. Many CS courses require much more time than other courses. Is that always necessary to achieve our learning objectives? Do we want to project a weed-out image?
  • Other departments often don’t think they need intellectual engagement with CS. They just want service courses for their students.

Two approaches to curriculum development emerged from the discussions. In the CS-centered approach, computing faculty work with representatives from other disciplines to identify the most important computer science principles and practices for inclusion in courses intended for students in those other disciplines but offered by computing-centric departments. In the discipline-centered approach, faculty from a discipline outside of computing work with computing faculty to develop curricula and courses addressing computational knowledge and skills for that discipline.

A strength of the CS-centered approach is that it can readily draw on the knowledge and experience of computing faculty around computing pedagogy, as well as an understanding of important aspects of computer science that even highly knowledgeable users of computing technology may lack, such as software engineering. The challenge of this approach is ensuring that the curriculum adequately connects with the needs of learners in other disciplines. “CS + X” approaches generally are CS-centered, in that they typically use existing CS courses. But to fully meet the objectives of CUE, these courses need to be tailored to fit the needs of students in other disciplines.

A discipline-centered approach ensures that the curriculum will meet the needs of learners in that discipline, and that computing concepts are presented in a context that will be meaningful to those learners. But work will be needed to make sure that key aspects of computing, like software engineering, are adequately addressed.

1.2 Critical Issues to be Addressed

Non-computing participants observe:

  • Context matters! Abstract examples don’t motivate.
  • CS courses are technicalities-first, rather than problem-first.
  • Identify core learning goals for computing. Develop a common core. Once articulated, faculty can adopt them and assessment can measure against them.
  • Be aware of the challenges non-CS/STEM students face:
    • Fear (content, competence)
    • Anxiety (failure, time to graduation)
    • Preparation (lack of foundational vocabulary, comprehension, and skills)
    • Stereotypes (“not cool”, who has access, lack of role models)
  • Faculty incentives and shared resources are the key challenges limiting scalability and sustainability.

Computing participants observe:

  • CS courses have high enrollment and designing courses for non-majors has low priority at the current time.
  • Staffing courses for CS majors is currently a challenge. Unclear how departments could manage an additional load.
  • Non-majors in CS courses may be perceived by CS instructors as weakening the rigor of the courses targeted at majors.
  • Managing students with diverse backgrounds in a single class is hard and requires a redesign of the course.
  • Students taking computing courses in non-CS departments are often not prepared for further CS courses.
  • Faculty incentives and shared resources are the key challenges limiting scalability and sustainability.

The recommendations made include:

Develop incentives for qualified faculty to get involved and actively contribute to CUE.

Computing departments are struggling to meet the demand from their own majors. Faculty in other disciplines have their own priorities. Few institutions can hire new faculty specifically for CUE. Innovation in incentives for existing faculty will be needed.

Develop and disseminate sharable resources.

While some faculty want content modules they can adopt, others instead want models of pedagogy that they can adapt for their courses.

Enable paths into more advanced computing topics.

Learners need to be able to increase their mastery of computational problem solving, while retaining their disciplinary identities. Early courses must provide enough computational knowledge and skill for students to engage with more advanced topics in computing that matter for their discipline.

2. Diversity and Inclusion

In many settings, sadly, computing has developed a culture, or is perceived as having a culture, of arrogance and exclusivity, in which the uninitiated are made to feel inferior and/or unwelcome. Diversity and inclusion need to be addressed wherever computing is taught, as there is always the possibility that students may display condescending attitudes, and for other students to feel left behind. Some participants felt that faculty, too, may see knowledge of computing as a kind of medicine that needs to be administered to those unfortunates who don’t have it, and may not recognize the value of approaches in other disciplines. Fixed mindset attitudes, as opposed to growth mindset attitudes, still exist on campus and in our society. When these attitudes are experienced in the classroom, learners may feel that they “don’t have what it takes”.

Participants identified a number of key elements that CUE efforts should incorporate.

  • “Technology first, application second” computing courses can limit engagement. Many students are more interested, certainly initially, in what computing can do than in how it works. Courses that begin with a focus on applications can have wider appeal.
  • Use a broad concept of audience(s). Inclusion in CS should support women, students with disabilities, LGBTQ students, first-generation college students, and students from rural areas, as well URM students and students from diverse ethnic backgrounds. Assume that all CS courses include students with all of these backgrounds when designing them. Faculty should understand the background, interests, and circumstances of students they hope to attract and retain.
  • Create courses or workshops that strengthen students’ background, aimed at underrepresented groups, either pre-college, or students already in college. Such bridge programs must actively link computing to students’ goals and interests. Just offering a computing course for a non-computing department that has a diverse student body will not by itself broaden participation in computing. Even in these contexts additional effort that will likely be required to reach the diverse student constituency we need to reach.
  • Care should be taken in creating a welcoming environment in classes from day one. Computing departments should recognize that many students feel they do not belong.
  • CS courses are known for highly competitive, timed tests. These create barriers for students who have less background, and possibly broader interests, than classmates. Mastery- or competency-based assessments reward learning at the pace and path that individual students need, providing multiple opportunities to learn difficult material, with less stress on competition.

3. The Need for Innovation

Participants from computing and non-computing disciplines alike emphasized that greater innovation is needed, especially for institutions struggling with limited resources. Key ideas viewed as having potential for innovation include the following.

Online tutorial material on virtually any topic in computing is freely available online. Integration of online materials in traditional learning approaches needs to be better understood.  Peer mentoring is already successfully used within CS at larger institutions. It would be especially exciting to create approaches that allow students in non-computing disciplines to learn about computing, without relying entirely on faculty to invest scarce time and effort. But such approaches need to be managed carefully to ensure that students who have less background, or might be assumed by peers to have less background, aren’t marginalized. Projects, if chosen by students, can increase intrinsic motivation, so that students are measuring their progress against concrete goals rather than relying on grades on assignments or exams for feedback.

Diversity-aware hackathons can create substantial learning opportunities on a no-credit basis.  Study abroad concepts and practices could be generalized to support a year in a different department on the same campus. Finally, “virtual” departments could be created to facilitate faculty cooperation across discipline lines.

Since the CUE.NEXT workshops were held, we have all experienced the abrupt transformation of higher education forced by the COVID pandemic. While we certainly hope that much of this disruption will prove temporary, we also believe that some instructors will seize the opportunity to experiment with new ways to teach, to assess learning, and to connect with students. This in turn may open up new possibilities for innovative approaches to CUE courses and curricula.

In the coming year, we plan to organize virtual follow-up workshops focused on further exploring CUE. Please check for information about upcoming workshops and the full CUE report.

[1] This work is supported by the NSF under grant number CNS-1944777. Any opinions, findings, and conclusions or recommendations expressed are those of the authors and do not necessarily reflect NSF’s views.