Summary¹

In the history of the United States, science, technology, engineering and mathematics—the disciplines that are linked together in the acronym STEM—have been uniquely powerful engines of growth and development. Harnessing the power of these disciplines and shaping and deploying them in ways that promote the wellbeing of all citizens and long-term stewardship of the natural world requires a well-informed population of critical thinkers that understands the STEM disciplines. To ensure that the United States continues to be a global leader in STEM literacy, innovation, and employment, it is essential that all Americans have lifelong access to high-quality STEM education.

Over several decades, the federal government has allocated resources to the improvement of STEM education. This has led to the development of a rich variety of educational innovations (i.e., programs, practices, models, and technologies), all in the service of supporting teaching and learning within the STEM disciplines. Although a number of these innovations have had the potential to impact learners on a broad scale, that potential often remains unrealized, and there are still ongoing questions of how to address the expansive goals of Pre-K–12 STEM education across the different grade bands, inclusive of all learners.

The Board on Science Education of the National Academies of Sciences, Engineering, and Medicine in response to a mandate within the CHIPS and Science Act of 2022 with support from the National Science Foundation

___________________

¹ This summary does not include references. Citations for the information presented herein are provided in the main text.

convened an expert committee to examine the interconnected factors at local, regional, and national levels that foster or hinder the widespread implementation of promising, evidence-based, Pre-K–12 STEM education innovations, identify gaps in the research, and provide guidance on how to address barriers to implementation.² The 15-member expert committee had extensive expertise across the STEM disciplines (science, engineering, mathematics, computer science, and data science) in various settings (rural and urban) and with different roles and spheres of influence (local, regional, and national).

The committee explored the available evidence on what it takes to successfully develop, implement, scale, and sustain Pre-K–12 STEM education innovations. In particular, the committee focused on the barriers to widespread implementation and looked for examples of innovations that found solutions to implementation challenges. Significantly, the committee examined the education system at various levels to better understand how the structure of the system itself can facilitate or hinder the scalability of promising, evidence-based, Pre-K–12 STEM education innovations.

Overall, the committee found that investments in innovations in STEM education have resulted in numerous promising programs. However, these programs vary in their success in reaching large numbers of students across different educational contexts. Perhaps more importantly, it is not clear that assembling an array of discrete innovative programs will result in the kind of robust, coherent, large-scale, systemic change that is likely needed to create the kinds of major improvements in student outcomes that many policy leaders seek.

PRE-K–12 STEM EDUCATION LANDSCAPE AND POLICY CONTEXT

Understanding how innovations in STEM education can take hold and result in improved outcomes for large numbers of learners requires an understanding of the larger educational landscape in the United States. The formal K–12 education system in the United States is organized across federal, state, and local levels. The federal government and states influence education through regulations and through financial support for education programs that often comes with particular guidance or restrictions for receipt and use of the funding. States are constitutionally responsible for public elementary and secondary education, and, as a result, most policymaking and governance of schools happen at this level. School districts, in turn, implement policy set by both the federal government and the state,

___________________

² The full statement of task appears in Box 1-1 in Chapter 1.

and also make numerous decisions about policy and practice that impact what happens in schools and classrooms.

Often policy and decision makers at the various levels of the system (including teachers) may not share the same priorities and goals with the result that there is misalignment in policies and priorities across federal, state, district, and school levels. These misalignments create obstacles to implementing coherent educational programs and are especially challenging when it comes to integrating innovative programs into existing structures and local contexts. In particular, the decentralized system means that it is difficult to propagate large-scale improvement across the country, and it presents a challenge for how the federal government can incentivize large-scale, sustained, and well-resourced improvement efforts.

Influence of Accountability

Currently, K–12 education is shaped by the past 25 years of accountability-based improvement efforts that began at the federal level with the passage of the No Child Left Behind (NCLB) Act in 2001. This legislation marked a number of critical changes in the federal education policy landscape and significantly increased the role of states in holding schools responsible for the academic progress of all students. State accountability systems serve to provide transparent data for instructional improvement, make visible learning gaps across equity groups, and encourage innovations in assessment practices. However, the high-stakes assessments have often narrowed the curriculum with an intense focus on English Language Arts and mathematics and emphasis on teaching approaches that result in high performance on standardized tests. This narrow focus can create barriers to implementation of some innovations in STEM education.

Preschool and Out-of-School Time

The challenges of disconnection across levels of the formal education system are compounded for preschool and for learning in out-of-school programs. Both of these sectors function almost independently from K–12 formal education. Each state designs its own preschool system through authorizing legislation and funding, and determines eligibility, quality standards, and monitoring. Because of this, the governance is highly variable and fragmented. Young children are served through a variety of programs (e.g., federal programs like Head Start, state-funded preschool, and various community agencies), which vary in their alignment to K–12 standards and curriculum.

Out-of-school time learning spaces comprise a vast range of environments and situations, including youth development programs, museums, libraries, zoos, botanical gardens, science centers, and community centers.

Out-of-school spaces not only provide opportunities for learners to engage in Pre-K–12 STEM education innovations but also are important spaces for innovation development. These settings can support experimentation as well as the development of resources (e.g., professional development, curriculum) that can translate into formal Pre-K–12 educational settings. Yet, these settings are often entirely disconnected from the formal education system.

SCALING AND SUSTAINING PROMISING PRE-K–12 STEM EDUCATION INNOVATIONS

Within this complex, disconnected system, the challenge becomes how an innovation that is successful in a single location or for a particulate group of students can be expanded to reach more students in more places. This is often thought of as a question of “scale.”

What Is “Scale”?

Many discussions of scale focus solely on increasing numbers of participants (spread), and common measures of scale often focus on number of beneficiaries, presence of materials, and time spent using the materials. However, scale can also mean depth of implementation (i.e., extent to which the innovation is intended to create or entails substantial shifts in the core of educational practice), sustainability (i.e., innovation endures over time in the original and new contexts when the initial circumstances run their course), and ownership (i.e., extent to which knowledge of and authority over the innovation is deepened and expanded over time). These additional ways of understanding scale are important because spread alone does not support educators or researchers in knowing whether an innovation is resulting in the desired improvement and for whom, whether the innovation is sustained as enactors change, and why or why not.

In fact, Pre-K–12 STEM education innovations can be designed for different purposes. They can be designed to have deep impact with a more local focus, designed explicitly for large-scale impact with less attention to depth, or designed to be some combination of both. Targeted innovations can be just as impactful as those designed for broad reach.

The more ambitious an innovation is—that is, the more it requires a substantial change to “business as usual” in order to be implemented—the more difficult it may be to sustain or to scale. If an innovation requires only superficial or minor changes to current practice, its adoption and assimilation by practitioners might be more easily achieved. However, the changes may not result in meaningful and long-lasting improvements. In fact, there is a tension between adoption of innovations that spread easily but may have limited impact and those that are harder to implement

yet show robust evidence of impact in the settings where they have been developed.

Features of Innovations That Facilitate Scaling

There are some key characteristics of innovations that appear to facilitate scaling. First, innovations with a strong core program (that is clearly stated) with room for adaptation to different contexts and learners are more likely to scale. Second, an innovation is more likely to scale when the goals and practices of the program align with the goals, priorities, and existing practices of the adopting organization or individual (e.g., district, school, educator, or out-of-school setting). Third, professional learning³ or other activities that build the capacities of individuals or organizations to implement the innovation are key. Finally, partnerships can be valuable for scaling in a variety of ways. If an innovation is developed by an entity outside of the K–12 system, such as a university or a nonprofit, partnering with districts, schools, and teachers as the innovation is designed and enacted can provide developers with in-depth insights into how contextual factors influence implementation and outcomes. These insights can guide subsequent improvements. In moving forward with implementation, external partners such as philanthropy or local business and industry can be valuable sources of resources, funding, and support.

ENABLING AND CONSTRAINING FACTORS

Many promising and innovative projects are not sustained in a significant way beyond their original instantiation. This is true for innovations that originate as research studies and those that emerge as educators on the ground work to expand and improve their practice. In some cases, the promising innovation does not gain traction beyond where it is developed because there is no strategy for sharing the innovation more widely. However, some innovations may also be developed in ways that do not take

___________________

³ As noted below, the committee recognizes that, as evidence-based practices have evolved with regard to the most effective ways to enable teachers to expand their knowledge and refine their practice in support of students’ learning, there has been a shift away from short-term professional development workshops or passive lectures and toward sustained, interactive learning experiences that include opportunities for educators to practice, customize, and reflect on how they can apply new learning in their schools and classrooms. The term “professional learning” is increasingly being used to refer to the latter approach. This report generally uses the term professional learning in line with current best practices, except when describing the work of others, in which case their own terms and descriptors are retained. Because various local, state, and national education agencies; funders; and other organizations may use one or the other, the committee’s recommendations include both terms.

into consideration variation across educational contexts—essentially there is a disconnect between the context where the innovation is developed and the contexts where it will need to be implemented. Collaboration across multiple sites and iterative cycles of design across time can be a model for addressing this kind of disconnect. In fact, innovations that have shown promise for scaling and sustaining are often developed with input from practitioners about the needs of educators and students and with explicit attention to the varying contexts in which the innovation could be implemented.

Research Incentives and Constraints

The incentives and constraints faced by researchers in obtaining funding and establishing the efficacy and effectiveness of new innovations can pose obstacles to developing flexible innovations that are more easily scalable. Research designs and methodologies that clearly show impact may push against designing flexible innovations. In addition, some researchers may lack the expertise to flesh out the kinds of supports that are needed in the initial design and/or evaluation to support more widespread implementation and allow adaptation to a broad set of contexts and student populations.

Once a program has scaled broadly, it is challenging to monitor and evaluate. When a program is being implemented and adapted in many different contexts, it is difficult to draw conclusions across sites and learners. Additionally, the cost of evaluating something at this scale in a deep way can be prohibitive.

The Realities of Educational Contexts

The complexity of educational environments creates additional challenges that can negatively affect successful scaling and sustainability efforts. If a program is designed to align with specific policies or to meet specific goals in a particular context, there can be a large threat to scaling if these goals vary across contexts. There is also a threat to sustainability when policies or priorities change and are not in line with the direction of the program.

Also, scaling and sustainability are aided by the buy-in and capacity building of individuals and organizations. If individuals championing the program at an organization or those trained to implement it leave, it can hinder continuity and efforts to scale the program. For programs that are sustained over time, it can be a challenge to keep materials up to date with advances in STEM fields or to update delivery technologies, as these would require additional development and testing.

Capacity of Educators and Education Leaders

The preparation and development of preservice and in-service STEM educators and education leaders does not routinely include opportunities to learn how to identify, evaluate, and implement innovations and adapt them for the needs of different students. This makes high-quality professional learning supports for those charged with enacting innovations essential. However, professional learning for the initial cohort of implementers is not sufficient for supporting sustainable implementation, and the sustainability of an innovation is easily threatened by turnover in the enactors. Thus, whether the innovation entails a minor or substantial change to “business as usual,” it is critical to build in structured opportunities to “onboard” new enactors.

Alignment with Existing STEM Initiatives and Priorities

The alignment of a particular STEM innovation with existing educational improvement efforts in a state, district, or school can shape whether and how an innovation is implemented. If an innovation aligns well with existing priorities, it may be easier to implement. If it contradicts existing priorities, it may meet resistance and fail. By aligning with—or building upon—other improvement strategies and reducing the perceived burden of educational reform, education leaders and decision makers can better leverage innovative STEM programs to advance educational improvement. In fact, one of the major challenges to improving STEM education through individual innovations is how to ensure that different programs are coherent and are leading to lasting change to the system.

THE FEDERAL ROLE IN SUPPORTING AND ADVANCING INNOVATIONS IN STEM EDUCATION

Although the federal role in STEM education is limited, some previous federal initiatives did make significant progress. For example, the systemic initiatives funded by National Science Foundation in the 1990s built significant capacity in states and districts for supporting high-quality STEM education. Many of these efforts, particularly those with robust (and coordinated) plans, showed promise leading to improvement; however, they were often not sustained when federal funding was eliminated.

Currently, federal funding agencies that support development of innovations in STEM education prioritize sequential studies of scaling innovations (i.e., pilot studies, efficacy studies, effectiveness studies, scale-up studies), whereby the innovation is implemented in tightly prescribed ways, in increasingly heterogeneous sites and/or populations, and in service of

specified outcomes. However, federal funding is not widely available to investigate the sustainability of innovations.

RECOMMENDATIONS

Based on the findings and conclusions summarized above, the committee developed a set of recommendations for federal, state, and local actors. These recommendations focus on building the capacity of educators and the education system for implementing innovations, enhancing the research infrastructure for developing innovations that are scalable and sustainable, developing methods to support systemic and continuous improvement, and understanding how to monitor progress.

Building Capacity and Monitoring Progress

The first four recommendations point to roles that federal agencies can play in building capacity within states to support implementation efforts. To build capacity, there needs to be significant investment in the professional development of the various enactors who are the main implementors of STEM education innovations (e.g., teachers, administrators) and designers (including researchers) as well as significant investment in building partnerships within the system to support organizational capacity. Building connected systems of teachers and administrators, curriculum specialists and developers, technology specialists, community learners and partners, and researchers, and creating new roles and spaces for them to work and learn together, is foundational for supporting organizational capacity. Moreover, previous systemic efforts showed greater promise when given the opportunity to build strong plans that enable systems-level change that can be sustained, many of which require time to take shape and then be implemented.

Recommendation 1: The U.S. Department of Education should allocate funding for teacher professional learning and development in all STEM disciplines, to include science, technology, engineering, mathematics, computer science, and other emerging STEM-focused subjects (e.g., data science). As part of the funding allocation, states will need to provide a plan for the use of funding for professional learning and development that is based on established best practice (e.g., curriculum-embedded, sustained over time), metrics to achieve the goals (e.g., measures of quality teacher professional learning and development), and data that show evidence for achieving those goals. The funding should be renewable up to ten years, and if states have not shown improvement by year four, they must revise their plan.

Recommendation 2: The National Science Foundation should develop a new generation of Pre-K–12 STEM education systemic initiatives with the goal of building infrastructure, capacity, and expertise to harvest promising evidence-based innovations, prepare them for wider implementation in new settings, and fund backbone organizations to organize the resources and support systems needed to carry out the implementations in schools. Funding should have a long enough time horizon (e.g., renewable up to ten years) for new structures and relationships to be iteratively refined and to take root in ways that could be sustained.

Recommendation 3: In alignment with the authorizing language of the CHIPS and Science Act of 2022 (P.L. 117–167, Sec. 10395), the National Science Foundation’s Directorate for Technology, Innovation, and Partnerships should partner with the Directorate for STEM Education (EDU) to fully leverage the expertise of the EDU Federal Advisory committee and ensure the inclusion of program officers with expertise in Pre-K–12 STEM education in the evaluation of proposals related to supporting multidisciplinary research centers for scaling promising Pre-K–12 STEM education innovations.

Recommendation 4: In an effort to facilitate coordination across federal agencies that implement, scale, and sustain Pre-K–12 STEM education innovations, the National Science and Technology Council’s Committee on STEM Education (CoSTEM) should identify key metrics for scaling and sustaining innovations and identify an appropriate schedule for reporting them to the public, exploring where the data should be reported (e.g., science.gov, National Center for Education Statistics, National Center for Science and Engineering Statistics) to keep the data evergreen. CoSTEM should use these data to inform future iterations of the strategic plan and coordinate consistent investments across the federal agencies.

Building a Research Infrastructure for Scalable and Sustainable Innovations

The committee also points to the need for additional research for better understanding the interrelated factors associated with sustaining Pre-K–12 STEM education innovations. The next three recommendations point to how federal actors can further develop the research infrastructure, helping researchers to fill the gaps in the current research base.

Recommendation 5: The U.S. Department of Education and the National Science Foundation should create a new funding category that allows for the study of sustainability of STEM education innovations,

which could allow for a deeper understanding of the dimensions of scaling. This should include developing system-level measures of STEM education innovations that attend to the interrelated actors, structures, and interactions that shape how innovations take hold in multiple, diverse contexts for different learners with an eye toward equity.

Recommendation 6: The U.S. Department of Education and the National Science Foundation should encourage practice-initiated partnerships and planning grants that will connect teams of researchers, designers, and practitioners with expertise and experience in different aspects of innovation development and implementation, either within a project or across successive related projects. These projects should include considerations of implementation and adaptation for different learners across multiple, diverse contexts early in the development process.

Recommendation 7: The U.S. Department of Education and the National Science Foundation should continue to encourage research and development in early STEM education: specifically, efforts that tackle integrating professional learning and development opportunities with curricula that address the many domains of learning that educators are expected to promote in early childhood, planning grants to support practice-initiated partnerships, and focus on coherence and alignment across preschool and elementary grades.

State and Local Actors: Systemic Change and Continuous Improvement

The next five recommendations are directed toward state and local actors in the Pre-K–12 education system. The recommendations focus on including strategies for making processes for continuous improvement the norm and creating partnerships to connect and align all the relevant roles and forms of expertise.

Recommendation 8: School and district leaders should adopt and/or evaluate a networked continuous improvement framework, emphasizing iterative assessment and refinement of strategies to meet the evolving educational landscape. This involves a cycle of planning, implementing, evaluating, and adjusting, with engagement of pertinent individuals to ensure ongoing relevancy. Data-driven analysis and feedback mechanisms should allow for real-time monitoring and responsive adaptation. Embracing this approach fosters a culture of innovation, learning, dexterity, responsiveness, and resilience within the schools and across the district.

Recommendation 9: To understand the implementation and scaling of Pre-K–12 STEM education innovations, state and district partners should develop data systems that capture information about opportunities to learn, including time for instruction, allocation of resources and funding, access to and enrollment in Pre-K–12 STEM education innovations, and qualifications of teachers and characteristics of teachers. These data should be disaggregated to examine trends by subgroups of students and by school characteristics.

Recommendation 10: To ensure continuous improvement in Pre-K–12 STEM education innovations, school and district leaders should engage with professional learning and development providers to offer curriculum-embedded, ongoing opportunities within and across years that includes specific emphasis on new teachers to ensure that their learning is commensurate with those who participated in opportunities in years prior.

Recommendation 11: Local school and district leaders should initiate and sustain partnership agreements across all levels of the STEM education learning ecosystem (e.g., teachers, teacher educators, education researchers, designers of Pre-K–12 STEM education innovations, families, etc.) in order to combine STEM education expertise and local knowledge to attend to specific problems of practice and advance sustained development and implementation of promising, evidence-based, Pre-K–12 STEM education innovations.

Regional Actors and Impact

The final set of recommendations acknowledge some of the broader range of actors that can have regional impact as they support the implementation, scaling, and sustaining of Pre-K–12 STEM education innovations.

Recommendation 12: Leaders of local and regional Pre-K–12 systems should work to strengthen learning opportunities in STEM education among key actors in the STEM education learning ecosystem (e.g., teachers, school/district leaders, school board leaders, teacher educators, professional development providers, universities and colleges, museums, nonprofits, families, etc.) with an emphasis on building relational connections among communities and sharing knowledge.

Recommendation 13: In their support of Pre-K–12 STEM education innovations, the federal government, philanthropic organizations, and business and industry should provide support to projects that include

designers of curricula partnering with education leaders, teachers, families/communities, and researchers to co-design resources that are evidence-based, meaningful, accessible, and able to be feasibly implemented to support STEM teaching and learning. Attention should be given to features that can lead to meaningful STEM learning, while also considering components needed to ensure resources can be sustained and adapted for use by others.