Artificial Intelligence, Data Literacy, and Preservice Mathematics Teacher Training
Heather Gallivan, University of Northern Iowa
Eric Weber, Iowa State University
The rise of Artificial Intelligence. The advent of ChatGPT – an interactive artificial intelligence (AI) platform – has started a national and global conversation of what AI does for us and what it can (and cannot yet) do. These and other AI entities have the potential to touch upon every aspect of the human experience. Since AIs are built upon data science and machine learning methodologies, data literacy among the populace at large is as crucial to society as ever. While all disciplines will play a role in developing the data literacy of K-12 students–hence, all disciplines will have contributions to deliver–we believe that mathematics teachers at the primary and secondary levels are best positioned to implement the charge of informing our students of the issues, challenges, and possibilities presented by data literacy, data science methodologies, and AI in general. In particular, the potential for fundamental transformation of society that AI poses calls for mathematics teacher educators to train mathematics teachers in the relevant data literacy and data science content. This position paper will accomplish the following intertwining objectives: 1) define data science and data literacy; 2) review the current state of data science and literacy education and mathematics teacher training within the State of Iowa and at the national level; 3) our own contributions to mathematics teacher preparation for data science; 4) support our claim that in-service and pre-service mathematics teachers will be at the forefront of nationwide efforts to increase data literacy across all sectors of society as full scale deployment of AI becomes actualized.
What is Data Science? As the evolution of data-driven methodologies accelerates, the scope of our efforts to further data science education in Iowa remains in flux–even the terminology itself evolves rapidly. Despite this constant churning, let us start with the terminology. Within academia–higher education especially–the dominant term in recent years has been “data science”. Broadly, data science is “the science of learning from data” (Engel, 2017, p. 44). However, there is no consensus on how to define data science or how it overlaps and/or differs from other terms like artificial intelligence (NASEM, 2018, 2023; Rosenburg & Jones, 2023). For our purposes here, we shall refer to data science and artificial intelligence interchangeably–not because they are, but rather because we believe that ChatGPT and similar AI platforms will ultimately render the “data science” phrase obsolete. Data literacy on the other hand, is more well-defined in the field of mathematics and statistics education. Data literacy involves not only being able to analyze, interpret, and evaluate data and statistics (i.e. statistical literacy), but to also be a critical consumer of data (Gould, 2017); recognizing “what we and others can do with data, what data can do to us, and what kind of world we can create with data” (Louie, 2023, p. 1).
We emphasize that regardless of how we refer to the content or the discipline, the content itself is a re-coupling of the academic disciplines of mathematics, statistics, and computer science. This places mathematics teachers at the forefront of delivering data science/literacy content in the nation’s K-12 schools. Let’s now consider where data science education currently stands at the K-12 level.
Data Science at the K-12 Levels. Just as the terms we use for data science are ever changing, the field of data science in K-12 education is “still developing and open to being shaped” (Rosenburg & Jones, 2022, p. 1). However, policy documents and reports from national organizations have made statements regarding what data science and data literacy concepts K-12 students need to know. The Pre-K–12 Guidelines for Assessment and Instruction in Statistics Education II: A Framework for Statistics and Data Science Education (GAISE II) report (2020) acknowledges that “the demands for statistical literacy have never been greater” (p. 1). This report provides a framework for how statistical and data literacy should be developed from the early grades through high school. The report highlights new skills for students, including a focus on the entire statistical investigative process, multivariate thinking across all grade levels, and incorporating technological tools to aid in data analysis. The Iowa Core Standards for Mathematics also have standards that reference data analysis and statistics from Kindergarten (Classify objects into given categories; count the numbers of objects in each category and sort the categories by count; K.MD.B.3.) through high school (e.g. summarize, represent, and interpret data; S-ID.A, S-ID.B, S-ID.C).
Given this emphasis on developing data and statistical literacy and analysis skills for K-12 students, many states nationally have begun to offer programs and coursework in data science. Currently, there are 14 states which have data science programs of varying depth and size; the majority of which are being taught by mathematics teachers (Drozda, Johnstone, & Van Horne, 2022). All 50 states have standards that reference data, but only 5 states have standards that are data science specific. For example, California released a Mathematics Framework in July 2023 that has an entire chapter devoted to data science across all grade bands from pre-kindergarten to high school (California Department of Education, 2023). There are several data science and statistics curricula that have been developed for high school students in recent years. These include CourseKata, YouCubed, Introduction to Data Science, and Bootstrap: Data Science (web addresses available in the resources below). All of these web-based resources are freely available for use by teachers. In Iowa, data science has also gained in popularity. Many high schools in Iowa teach coursework in statistics, which has overlapping concepts with data science (e.g. analyzing multivariate data in high school); and at least one high school in Iowa is currently offering coursework in data science utilizing the Skew the Script curriculum materials. Further, the state of Iowa has officially adopted course descriptions for data literacy and data science (you can search for data science and data literacy course descriptions at the link under Resources below), anticipating the desire for high schools to start officially offering such coursework for credit.
With this increased national and state-wide interest in and need to offer K-12 coursework in data science, there will be a need for teachers to teach these courses. But where do we start? For example, do we need to define state-level standards for data science? Offer professional development for in-service teachers to learn to teach data science? We contend that a good place to start is in teacher preparation, especially at the preservice mathematics teacher level.
Mathematics Teacher Preparation in Data Science. Since the goal of K-12 education in data science is to develop students’ ability to develop data literacy and data analysis skills (GAISE II, 2020), teacher preparation needs to focus on current and future teachers developing these skills as well. Teachers need to have experiences where they work with multivariate data sets, ask statistical questions, etc. with data sets that are meaningful in order to be able to give those experiences to their students. In other words, teachers need experience engaging with data themselves before creating these opportunities for students. The vast majority of K-12 data science courses offered in the United States are taught by mathematics teachers (Drozda, Johnstone, & Van Horne, 2023) and thus, are in the best position to continue to move data science education forward. We feel that future mathematics teachers are an important group to target for teacher preparation.
Our long-term vision is for all future teachers, regardless of discipline, to experience substantial data literacy and data science content in teacher preparation courses. In STEM fields at the secondary level, “substantial” may involve several courses dedicated to data science. For example, the AMTE Standards document (2017) recommends middle school mathematics teachers having two courses in statistics and data science and high school mathematics teachers to have three to be sufficiently prepared to teach statistics and data science content. However, the current challenge is two-fold: 1) fully developed data science curricula for preservice mathematics teachers do not exist; 2) we do not have space in current teacher preparation programs to introduce new courses to deliver data science content. Both of these challenges are exacerbated by the shifting specifications of the data science discipline and the associated licensing requirements for teachers.
Despite these challenges, we feel it is necessary to start, even if in small increments. We argue that the mathematics community has a significant advantage in the form of a fully developed infrastructure–curriculum, educational standards, pipelines from preservice to in-service teaching opportunities–over statistics and computer science (where mathematics teachers often teach coursework in these areas as well). Thus, we believe that future mathematics teachers are in the best position currently to be trained in teaching data science. In response to challenge 1), we have created a 6-week data science module to develop preservice secondary mathematics teachers’ data science content knowledge and mathematical knowledge for teaching data science. We have implemented this module with preservice teachers in the state of Iowa during one of their required content courses for the teaching major and licensure. Thus far, we have shown positive results in developing preservice secondary mathematics teachers’ knowledge of a few data science concepts (e.g. data classification and model fitting). The purpose of this module is to begin the conversation on what content knowledge the field believes is important for preservice mathematics teachers to know and how we can best prepare them to teach data science concepts in the future.
In response to challenge 2), we are beginning to explore how we can deliver content within the existing teacher preparation programs. First, our intent is to design and develop short 4-8 week drop-in data science modules that could be embedded within other teacher education coursework. Further, every course at the post-secondary level has a list of learning objectives that are meant to advance the students’ understanding of the overall program’s objectives. We contend that within mathematics teaching programs, we can design and deliver data science content that still meets the course learning objectives while also conveying relevant data literacy concepts. For example, within a teaching methods course, our learning objectives are often pedagogical in nature (i.e. how to teach mathematics content). To engage students in meeting those pedagogical goals, data science concepts can be the content in which preservice teachers engage in those pedagogical goals (i.e. writing a lesson plan over a data science concept). Additionally, mathematics teacher education programs often contain content courses that have learning objectives to develop preservice teachers’ mathematical knowledge for teaching (Ball, Thames, Phelps, 2008); namely, the specialized content knowledge required to teach mathematics. Data science could also be a topic covered to support preservice teachers in developing this knowledge. Based on our pilot project, course content can be changed moderately to meet both objectives in single courses, and we believe that this approach can be successful in multiple courses.
To advance our mission of preparing teachers to teach data science in K-12, we intend to further develop our modules as the field of data science and AI evolves to meet the needs of current and future teacher learning. We also intend to expand our modules to a full-length course in data science and create smaller modules that can be used to introduce and develop data science content in other relevant mathematics teacher education coursework. To bring our vision to fruition, we will eventually expand our modules to meet the needs of teachers other than those of mathematics–computer science, general science, elementary, and other content area teachers also will likely have opportunities to teach coursework in data science at the K-12 level and it is important to prepare them as well. Finally, we would like to expand our efforts in the future by providing professional development opportunities or graduate level coursework for in-service mathematics teachers. To reiterate, the conversation needs to start somewhere, and we feel we are in a position as mathematics teacher educators to begin that conversation through the development of curriculum materials for preservice teachers.
“All-Hands” Approach to Data Literacy. Because AI has such great potential to transform every aspect of society, data science and data literacy instruction at the primary and secondary levels will require a similar transformation. We don’t know what the future holds for data scientists in terms of the problems and goals they will have, which means the goals for data science education will have to adapt and change as the field progresses (Rosenburg & Jones, 2023). This will necessitate an “All Hands” approach to accomplish such a sizable task–all disciplines will be affected eventually, and thus teachers across all disciplines will need to be prepared for those changes. We are beginning the task of adapting preservice mathematics teacher curriculum as the “tip of the spear” effort to accommodate the coming changes due to AI, whatever the scale of those changes may be. We conclude with a call for partners: we would be delighted to partner with in-service mathematics teachers who desire to join the effort to better prepare our students for the future in which AI is a prevalent reality.
This article is based upon work supported by the Iowa Space Grant Consortium under NASA Award No. 80NSSC20M0107.
References
Association of Mathematics Teacher Educators. (2017). Standards for preparing teachers of mathematics. Association of Mathematics Teacher Educators. amte.net/standards
Ball, D. L., Thames, M. H., & Phelps, G. (2008). Content knowledge for teaching: What makes it special? Journal of Teacher Education, 59(5), 389-407. https://doi.org/10.1177/0022487108324554
Bargagliotti, A., Franklin, C., Arnold, P., Gould, R., Johnson, S., Perez, L., & Spangler, D. A. (2020). Pre-K–12 guidelines for assessment and instruction in statistics education II (GAISE II): A guideline for precollege statistics and data science education. National Council of Teachers of Mathematics. https://www.amstat.org/asa/files/pdfs/GAISE/GAISEIIPreK-12_Full.pdf
California Department of Education. (2023). Mathematics framework for California Public Schools: Kindergarten through grade twelve (Mathematics Framework)
Drozda, Z., Johnstone, D., Van Horne, B. (2023). Previewing the national landscape of K-12 data science implementation (National Academy of Sciences, Engineering, and MedicineFoundations of Data Science for Students in Grades K-12: A Workshop). https://www.nationalacademies.org/event/09-13-2022/docs/D16254F310D01BBDA873920E4EFB8151F2D8334181AA
Engel, J. (2017). Statistical literacy for active citizenship: A call for data science education. Statistics Education Research Journal, 16(1), 44-49.
Gould, R. (2017). Data literacy is statistical literacy. Statistics Education Research Journal, 16(1), 22-25.
Louie, J. (2023).Critical data literacy: Creating a more just world with data (National Academy of Sciences, Engineering, and Medicine Foundations of Data Science for Students in Grades K-12: A Workshop). 3https://www.nationalacademies.org/event/09-13-2022/docs/D16254F310D01BBDA873920E4EFB8151F2D8334181AA
National Academies of Sciences, Engineering, and Medicine [NASEM]. (2018), Data science for undergraduates: Opportunities and options. National Academies Press.
National Academies of Sciences, Engineering, and Medicine [NASEM]. (2013), Foundations of data science for students in grades K-12. National Academies Press.
Rosenburg, J. M. & Jones, R. S. (2023). A secret agent? K-12 data science learning through the lens of agency (National Academy of Sciences, Engineering, and Medicine Foundations of Data Science for Students in Grades K-12: A Workshop). https://www.nationalacademies.org/event/09-13-2022/docs/DD667E469D0EC5DD91A7D85BC839A9852491A3CF9F15
Resources
Bootstrap: Data Science: https://www.bootstrapworld.org/materials/data-science/
CourseKata: https://coursekata.org/
Introduction to Data Science: https://centerx.gseis.ucla.edu/idsucla/
Iowa Course Descriptions Search: https://nces.ed.gov/scedfinder/Home/Search
Skew the Script: https://skewthescript.org/ap-stats-curriculum
YouCubed: https://www.youcubed.org/