Teacher Education Clearinghouse for AI and Data Science
The next presidential administration should develop a teacher education and resource center that includes vetted, free, self-guided professional learning modules, resources to support data-based classroom activities, and instructional guides pertaining to different learning disciplines. This would provide critical support to teachers to better understand and implement data science education and use of AI tools in their classroom. Initial resource topics would be:
- An Introduction to AI, Data Literacy, and Data Science
- AI & Data Science Pedagogy
- AI and Data Science for Curriculum Development & Improvement
- Using AI Tools for Differentiation, Assessment & Feedback
- Data Science for Ethical AI Use
In addition, this resource center would develop and host free, pre-recorded, virtual training sessions to support educators and district professionals to better understand these resources and practices so they can bring them back to their contexts. This work would improve teacher practice and cut administrative burdens. A teacher education resource would lessen the digital divide and ensure that our educators are prepared to support their students in understanding how to use AI tools so that each and every student can be college and career ready and competitive at the global level. This resource center would be developed using a process similar to the What Works Clearinghouse, such that it is not endorsing a particular system or curriculum, but is providing a quality rating, based on the evidence provided.
Challenge and Opportunity
AI is an incredible technology that has the power to revolutionize many areas, especially how educators teach and prepare the next generation to be competitive in higher education and the workforce. A recent RAND study showed leaders in education indicating promise in adapting instructional content to fit the level of their students and for generating instructional materials and lesson plans. While this technology holds a wealth of promise, the field has developed so rapidly that people across the workforce do not understand how best to take advantage of AI-based technologies. One of the most crucial areas for this is in education. AI-enabled tools have the potential to improve instruction, curriculum development, and assessment, but most educators have not received adequate training to feel confident using them in their pedagogy. In a Spring 2024 pilot study (Beiting-Parrish & Melville, in preparation), initial results indicated that 64.3% of educators surveyed had not had any professional development or training in how to use AI tools. In addition, more than 70% of educators surveyed felt they did not know how to pick AI tools that are safe for use in the classroom, and that they were not able to detect biased tools. Additionally, the RAND study indicated only 18% of educators reported using AI tools for classroom purposes. Within those 18%, approximately half of those educators used AI because they had been specifically recommended or directly provided a tool for classroom use. This suggests that educators need to be given substantial support in choosing and deploying tools for classroom use. Providing guidance and resources to support vetting tools for safe, ethical, appropriate, and effective instruction is one of the cornerstone missions of the Department of Education. This education should not rest on the shoulders of individual educators who are known to have varying levels of technical and curricular knowledge, especially for veteran teachers who have been teaching for more than a decade.
If the teachers themselves do not have enough professional development or expertise to select and teach new technology, they cannot be expected to thoroughly prepare their students to understand emerging technologies, such as AI, nor the underpinning concepts necessary to understand these technologies, most notably data science and statistics. As such, students’ futures are being put at risk from a lack of emphasis in data literacy that is apparent across the nation. Recent results from the National Assessment of Education Progress (NAEP), assessment scores show a shocking decline in student performance in data literacy, probability, and statistics skills – outpacing declines in other content areas. In 2019, the NAEP High School Transcript Study (HSTS) revealed that only 17% of students completed a course in statistics and probability, and less than 10% of high school students completed AP Statistics. Furthermore, the HSTS study showed that less than 1% of students completed a dedicated course in modern data science or applied data analytics in high school. Students are graduating with record-low proficiency in data, statistics, and probability, and graduating without learning modern data science techniques. While students’ data and digital literacy are failing, there is a proliferation of AI content online; they are failing to build the necessary critical thinking skills and a discerning eye to determine what is real versus what has been AI-generated, and they aren’t prepared to enter the workforce in sectors that are booming. The future the nation’s students will inherit is one in which experience with AI tools and Big Data will be expected to be competitive in the workforce.
Whether students aren’t getting the content because it isn’t given its due priority, or because teachers aren’t comfortable teaching the content, AI and Big Data are here, and our educators don’t have the tools to help students get ready for a world in the midst of a data revolution. Veteran educators and preservice education programs alike may not have an understanding of the essential concepts in statistics, data literacy, or data science that allow them to feel comfortable teaching about and using AI tools in their classes. Additionally, many of the standard assessment and practice tools are not fit for use any longer in a world where every student can generate an A-quality paper in three seconds with proper prompting. The rise of AI-generated content has created a new frontier in information literacy; students need to know to question the output of publically available LLM-based tools, such as Chat-GPT, as well as to be more critical of what they see online, given the rise of AI-generated deep fakes, and educators need to understand how to either incorporate these tools into their classrooms or teach about them effectively. Whether educators are ready or not, the existing Digital Divide has the potential to widen, depending on whether or not they know how to help students understand how to use AI safely and effectively and have the access to resources and training to do so.
The United States finds itself at a crossroads in the global data boom. Demand in the economic marketplace, and threat to national security by way of artificial intelligence and mal-, mis-, and disinformation, have educators facing an urgent problem in need of an immediate solution. In August of 1958, 66 years ago, Congress passed the National Defense Education Act (NDEA), emphasizing teaching and learning in science and mathematics. Specifically in response to the launch of Sputnik, the law supplied massive funding to, “insure trained manpower of sufficient quality and quantity to meet the national defense needs of the United States.” The U.S. Department of Education, in partnership with the White House Office of Science and Technology Policy, must make bold moves now to create such a solution, as Congress did once before.
Plan of Action
In the years since the Space Race, one problem with STEM education persists: K-12 classrooms still teach students largely the same content; for example, the progression of high school mathematics including algebra, geometry, and trigonometry is largely unchanged. We are no longer in a race to space – we’re now needing to race against data. Data security, artificial intelligence, machine learning, and other mechanisms of our new information economy are all connected to national security, yet we do not have educators with the capacity to properly equip today’s students with the skills to combat current challenges on a global scale. Without a resource center to house the urgent professional development and classroom activities America’s educators are calling for, progress and leadership in spaces where AI and Big Data are being used will continue to dwindle, and our national security will continue to be at risk. It’s beyond time for a new take on the NDEA that emphasizes more modern topics in the teaching and learning of mathematics and science, by way of data science, data literacy, and artificial intelligence.
Previously, the Department of Education has created resource repositories to support the dissemination of information to the larger educational praxis and research community. One such example is the What Work Clearinghouse, a federally vetted library of resources on educational products and empirical research that can support the larger field. The WWC was created to help cut through the noise of many different educational product claims to ensure that only high-quality tools and research were being shared. A similar process is happening now with AI and Data Science Resources; there are a lot of resources online, but many of these are of dubious quality or are even spreading erroneous information.
To combat this, we suggest the creation of something similar to the WWC, with a focus on vetted materials for educator and student learning around AI and Data Science. We propose the creation of the Teacher Education Clearinghouse (TEC) underneath the Institute of Education Sciences, in partnership with the Office of Education Technology. Currently, WWC costs approximately $2,500,000 to run, so we anticipate a similar budget for the TEC website. The resource vetting process would begin with a Request for Information from the larger field that would encourage educators and administrators to submit high quality materials. These materials would be vetted using an evaluation framework that looks for high quality resources and materials.
For example, the RFI might request example materials or lesson goals for the following subjects:
- An Introduction to AI, Data Literacy, and Data Science
- Introduction to AI & Data Science Literacy & Vocabulary
- Foundational AI Principles
- Cross-Functional Data Literacy and Data Science
- LLMs and How to Use Them
- Critical Thinking and Safety Around AI Tools
- AI & Data Science Pedagogy
- AI and Data Science for Curriculum Development & Improvement
- Using AI Tools for Differentiation, Assessment & Feedback
- Data Science for Safe and Ethical AI Use
- Characteristics of Potentially Biased Algorithms and Their Shortcomings
A framework for evaluating how useful these contributions might be for the Teacher Education Clearinghouse would consider the following principles:
- Accuracy and relevance to subject matter
- Availability of existing resources vs. creation of new resources
- Ease of instructor use
- Likely classroom efficacy
- Safety, responsible use, and fairness of proposed tool/application/lesson
Additionally, this would also include a series of quick start guide books that would be broken down by topic and include a set of resources around foundational topics such as, “Introduction to AI” and “Foundational Data Science Vocabulary”.
When complete, this process would result in a national resource library, which would house a free series of asynchronous professional learning opportunities and classroom materials, activities, and datasets. This work could be promoted through the larger DoE as well as through the Regional Educational Laboratory program and state level stakeholders. The professional learning would consist of prerecorded virtual trainings and related materials (ex: slide decks, videos, interactive components of lessons, etc.). The materials would include educator-facing materials to support their professional development in Big Data and AI alongside student-facing lessons on AI Literacy that teachers could use to support their students. All materials would be publicly available for download on an ED-owned website. This will allow educators from any district, and any level of experience, to access materials that will improve their understanding and pedagogy. This especially benefits educators from less resourced environments because they can still access the training they need to adequately support their students, regardless of local capacity for potentially expensive training and resource acquisition. Now is the time to create such a resource center because there currently isn’t a set of vetted and reliable resources that are available and accessible to the larger educator community and teachers desperately need these resources to support themselves and their students in using these tools thoughtfully and safely. The successful development of this resource center would result in increased educator understanding of AI and data science such that the standing of U.S. students increases on such international measurements as the International Computer and Information Literacy Study (ICILS), as well as increased participation in STEAM fields that rely on these skills.
Conclusion
The field of education is at a turning point; the rise of advancements in AI and Big Data necessitate increased focus on these areas in the K-12 classroom; however, most educators do not have the preparation needed to adequately teach these topics to fully prepare their students. For the United States to continue to be a competitive global power in technology and innovation, we need a workforce that understands how to use, apply, and develop new innovations using AI and Data Science. This proposal for a library of high quality, open-source, vetted materials would support democratization of professional development for all educators and their students.
For the United States to continue to be a competitive global power in technology and innovation, we need a workforce that understands how to use, apply, and develop new innovations using AI and Data Science.
Students, families and communities want and need more STEM learning experiences to realize the American Dream, and yet they cannot access them. Prioritizing STEM education must be an urgent priority for the federal government and the Department of Education.
The Department of Education must provide guidance for education decision-makers to evaluate AI solutions during procurement, to support EdTech developers to mitigate bias in their applications, and to develop new fairness methods.
The incoming presidential administration of 2025 should champion a policy position calling for strengthening of the connection between K-12 schools and community workplaces.