Communications of the ACM

Viewpoints

The Long Road to Computer Science Education Reform

By Cameron Wilson, Peter Harsha
Communications of the ACM, September 2009, Vol. 52 No. 9, Pages 33-35
10.1145/1562164.1562178
Comments

Whenever computer scientists discuss the human capital issues that plague the computing fields—the significant lack of diversity among computer science graduates (especially at the master's and Ph.D. levels),^a the underrepresentation of women in computing, and the general decline in interest among U.S. students in the science and engineering fields—inevitably, the lamentable state of computer science education in the U.S.'s primary and secondary schools figures prominently.

While there are significant issues within higher education, there is a growing realization that we must address challenges at the beginning of the education pipeline. Excellent computer science classes are being taught in U.S. schools today, but looking across the country they are the exception to the rule. In general, we find too few students have the opportunity to take engaging and rigorous computer science courses in high school; there is little diversity among those that do. Too few opportunities exist for professional development for teachers. Too little innovation has happened in creating an engaging and rigorous curriculum for students. There is general agreement that this is a national failing—and one that we can ill afford—as computing is a central part of society, and key enabler of innovation and economic growth.

If "fixing" computer science education in kindergarten through grade 12 (K–12) is so clearly necessary, why has there not been more progress in the U.S.? In an age when the ability to think computationally already is, or certainly will be, a prerequisite for success in so many endeavors, why do we still struggle to reform K–12 computer science and make it more relevant?

In large part, it is because reform of the K–12 education system at any level or in any subject is notoriously difficult. Control over education is decentralized. States and school districts play varying leadership roles in determining what students must learn. Federal policy and bureaucracy, driven largely through strings attached to federal funds, layer on top of state and local responsibilities. Add to this mix numerous outside organizations from teachers' advocates to parent groups demanding a range of reforms and the result is an immensely complex web of policies, institutions, and players shaping the U.S. education system. From the outside looking in, any large-scale reform effort seems doomed for marginalization within this behemoth.

To further complicate the calculus, pressure is increasing, both nationally and internationally, for countries to take immediate steps to strengthen science and mathematics education to foster future innovation in high-technology fields.^b President Barack Obama made this national push plain in a major speech to the May 2009 annual meeting of the prestigious National Academies in Washington D.C. stating, "since we know that the progress and prosperity of future generations will depend on what we do now to educate the next generation, today I'm announcing a renewed commitment to education in mathematics and science."^c This statement moved discussions of improving science, technology, engineering and mathematics education (also known as STEM) to the forefront of national debates about education reform.

With the stage thus set for STEM education reform, two important questions come to mind: First, is this going to be the classic case of the irresistible force (the need for coordinated STEM reform) meeting the immovable object (state and local control of education)? Second, and most relevant to the computing community, how does computing fit into overall K–12 STEM education reform? Before we can begin to answer these questions it is useful to understand the basic workings of the education system in the U.S.

---

Any reform discussion starts with two important state-based concepts: standards and assessments.

---

Any reform discussion starts with two important state-based concepts: standards and assessments. These are the backbone of the education landscape. Each state sets learning standards for students in the state's K–12 school system. For example, one part of the state of Virginia's sixth-grade mathematics standard is that students should be able to "...identify, represent, order, and compare integers."^d Then, in most states, it is up to the school districts to establish curriculum implementing these standards. State and school districts assess whether the concepts are learned through testing. And yes, just because something is in the standards does not mean students will be exposed to it, and just because they are tested does not mean they know it. The point is to understand the education policy framework because it looms large in efforts to reform education.

Graduation requirements are equally important. Most states set or provide guidance on the credits students must accumulate to graduate from high school (also known as secondary school). These requirements fall into two general categories: a set of "core" courses that students must take to graduate; and electives that constitute everything not in the core. For example, the state of Texas requires that students have four years of high school mathematics credits in order to graduate. California aligns its graduation requirements with the higher education system by mandating a set of courses that are the minimum admission requirements to state-funded schools.

While states and/or local governments generally make graduation and curriculum decisions, these decisions are influenced by national goals and accountability requirements of certain federal laws. The most notable federal law is the controversial No Child Left Behind Act. Its accountability provisions require that states test in reading, mathematics, and science at certain grade levels. If individual schools do not meet state-based benchmarks for student achievement in reading or mathematics then those schools can face federal sanctions.

Because standards, assessments, curriculum, and graduation requirements are state and/or local decisions, advocates for national STEM reform face a difficult challenge of making their case to policymakers in each state, or even to the 14,000 school boards across the nation. Local governance of education has rendered any discussions of "national standards" political nonstarters. However, as President Obama's speech signified, the political landscape for reform has begun to change. Two years ago, the National Science Board identified the crises in STEM education^e stating that the system was failing America's youth and curriculum needed reform and coordination both within and among the states. Soon after the president's speech, powerful state education groups announced that 46 states would work toward harmonizing (not nationalizing) some standards.^f The administration is also putting money on the line with $5 billion targeted for overall education reform efforts, some of which will likely go to STEM education. All of this points to a solution that is driven from the bottom, but with top-down pressure and incentives being applied.

However, there are some important caveats to this progress—particularly for computer science education. First, harmonizing standards between states is politically difficult because standards affect what is assessed and therefore tested. Scores on those tests can affect federal funding and parents' perceptions of schools. Second, of the $5 billion in new federal education funding, it is not clear how much of it will go to STEM education reform. Third, it appears that states will focus initially on harmonizing math and reading standards, then turn to reviewing science standards. Most states do not have a specific set of computer science standards, but even if they exist, the state's focus on the "common core" likely will not include existing computer science standards. We need to delve even deeper into the education system to understand this.

Standards, graduation credits, and No Child Left Behind drive students and administrators toward emphasizing—both in terms of what students take and resources dedicated to developing them—the "core" courses of the curriculum. It is a gross oversimplification of an incredibly complex system to say that students across the nation are taking a similar set of core courses. The key issue for computer science education is, as a general rule, computing is absent from the "core." Much of what is called computing education by states at the K–12 level, particularly high school, is placed within the technology curriculum both in the states standards and the schools. However, the curriculum of so-called computing classes within this category largely focuses on the use of technology (keyboarding, or learning word processing/spreadsheets) instead of core computing concepts. Further, technology classes are generally elective credits for students on par with health or shop class.

This categorization puts efforts to get rigorous computing courses into the college-bound academic curriculum at a significant disadvantage. What is considered technology in school is typically not an academic subject area for the college-bound student; rather classes to help bolster vocational education for those about to enter the work force. Students pursing college often do not have the time for elective credits, particularly those focused on a vocation.

Despite these daunting obstacles, there are exciting efforts already under way led by different parts of the community to address this national failing:

---

Reform of the K–12 education system at any level or subject is notoriously difficult.

---

• Two years ago, ACM created an Education Policy Committee^g to focus on public policy issues in science and math education relevant to computing and computer science. Since its inception, the committee has identified many of the issues outlined in this column. Two key parts of its agenda are: First, educate policymakers and national groups on the importance of the field of computing, the value of teaching computer science, and that computer science curricula focus is on conceptual knowledge. Second, ensure that rigorous computer science classes count toward a student's core graduation requirements in math or science areas.
• Realizing that reform begins with the states, the Computer Science Teachers Association (CSTA) has formed a cohort of master teachers in many states to serve as a network for sharing information and communicating issues to state and local education leaders.
• The single point of national leverage for computer science reform is the current Advanced Placement (AP) computer science course. The AP system is run by the College Board, which is a national non-profit organization that helps set the curriculum and writes and administers the tests. Every student taking an AP course is, or should be, exposed to the same curriculum, and every student takes the same test to assess the knowledge. The National Science Foundation is currently funding a project to create a new rigorous and engaging Advanced Placement Computer Science (AP CS) course that will attract more college-bound students.
• Some states are moving toward allowing the AP CS course to count toward a credit in students' core courses requirements. Texas now allows AP CS to count as a mathematics credit as part of the student's four-year mathematics requirement. North Carolina and Ohio also are moving toward allowing AP CS to count as a mathematics credit.
• Georgia has launched its Georgia Computes! program. As part of this effort Georgia overhauled its state standards for computing and now counts AP CS as a science credit.
• The National Center for Women and Information Technology (NCWIT) has formed the K–12 alliance, bringing together leaders from more than 20 organizations with a potential reach of approximately half the girls in the U.S. to provide resources and advocate for reform.
• As noted in a previous issue of Communications,^h recognizing the need for a rigorous and engaging computing course taken before AP, Joanna Goode and Jane Margolis are working with a team to create an exciting new curriculum based on the ACM and CSTA model curriculum for K–12 education.ⁱ This course is being scaled up in the Los Angeles school district, and they are working with the state to build it into their higher-education requirements.

These are all promising signs, but any education reform, whether it is STEM broadly or computing specifically, will have to be measured over years, if not decades. Progress can be made, but the states will have to lead, with federal policymakers offering support with resources and a heightened sense of national urgency. There is a significant risk that computing will be left on the outside looking in. The community needs to come together and use the existing efforts mentioned in this column as a starting point for reform. We need to work together to prove why rigorous and engaging computing education should be included in the K–12 landscape. In short, we need to ensure computer science education is part of the national dialogue of what students need to learn in the high-technology and highly competitive global economy.

Back to Top

Back to Top

Back to Top

Back to top

The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.