“I believe that we owe it to our children to prepare them for the world that they will encounter—a world driven by data. Basic data fluency is a requirement not just for most good jobs, but also for navigating life more generally, whether it is in terms of financial literacy, making good choices about our own health, or knowing who and what to believe.” – Steve Leavitt, Co-Author, Freakonomics

Because data science and related fields in data will be among the fastest growing professions in the U.S. over the next several years, interest in educating people to participate in those fields has been proliferating. This has manifested most visibly and immediately at the university and post-graduate level as a rapidly increasing number of schools are offering soon-to-be professionals formal degrees in the subject so they can seize the opportunity to launch lucrative careers. But, as the public begins to apprehend the pervasive impact of data on our daily lives and the importance of fostering interest in and awareness of the fundamentals of data science and analytics, there is a growing desire to expand educational opportunities in the subject to the K-12 level.

Creating data literacy has become a priority in many businesses. Like any language, however, the earlier we can introduce it to students, the better chance they have to truly become literate in the subject. K-12 educators see the value of adding data science, visualization and analytics to curricula early, but few standards exist and the body of research on the best way to do that is—while growing—still relatively scarce.

Early Efforts

Educators were formally considering the best ways to educate young students in data science—though it was couched in more mathematical terminology—as far back as 2005 with the first Guidelines for Assessment and Instruction in Statistics Education (GAISE) for Pre-K-12 report. Researchers updated their work in 2020 with GAISE II.

GAISE I established a basic framework for educators at the elementary, middle and high school levels, but GAISE II noted its limitations.

“GAISE I primarily focused on traditional data types of quantitative and categorical variables and on study designs using small data sets of samples from a population,” authors of the updated report wrote. “Fifteen years later, data types have expanded beyond being classified as quantitative and categorical thus necessitating the acquisition of different and often state-of-the-art statistical skills. Today, for example, data include text posted on social media or highly structured (or unstructured) collections of pictures, sounds, or videos. Data are immense and readily available. Data are multidimensional. Data representations and visualizations are also often multidimensional and interactive displaying many variables simultaneously.”

GAISE II strove to incorporate new skills that would equip students to recognize and make sense of the far more complex sources and forms data is taking. IT stressed the importance of asking questions throughout the problem-solving process and sought to include multivariate and probabilistic thinking as well as recognizing the role technology plays.

Even at the time, however, experts expressed that in order to meet the vision set forth in GAISE II, the educational system as a whole needed significant transformation. A 2020 article in the Harvard Data Science Review (HDSR) detailed the need for additional work on K–12 school curriculum, K–12 teacher professional development, K–12 teacher preparation, statistics and data science education research, and policies. And those needs have only grown as the amount of data produced each year is measured in hundreds of zettabytes and AI has increased our capacity to process and analyze that data.

How should we think about data education moving forward?

Since GAISE II, researchers and scholars have continued to examine data science learning at the K-12 level, but diverse research agendas and learning environments relative to data science education has made implementing policy on a nationwide basis difficult.

A more recent HDSR article examined just this question. Depending on their perspectives and motivations, researchers studying how to improve data science education are producing results that aren’t cohesive.

“As data science education is emerging at the grades K–12 and higher education levels, the research and curriculum development are highly variable and often lack a shared focus, language, or set of values,” the authors wrote.

Part of the problem, they assert, is that data science itself is interdisciplinary, involving a range of competencies that traditionally have been taught and learned separately. But a review of all the available research on data science education found that most of the differences that lead to a lack of standards in the space can be attributed to variations in the priorities of the researchers—what the authors called agency.

Different kinds of agency—reflected in different research studies—emphasize different techniques and can lead to different curricula and methods employed by teachers. How, the authors ask, can data science educators discuss their work with such diverse ideas and practices? They made several recommendations:

• There isn’t one correct form of agency for K-12 teachers to emphasize; instead, teachers can consider what their goals are when teaching particular students specific data science capacities or ideas.
• Data science education researchers need to be more explicit about what we prioritize and what motivates our work.
• We must consider essential research on data science learning at the grades K–12 levels by scholars not only with backgrounds in computer science and statistics education, but also in the arts, humanities, social studies, and everyday life and experiences

Conclusion

While efforts have begun to establish overarching principles for data science education and data literacy in K–12 education in many states through GAISE II, experts agree that more large-scale collaboration between major stakeholders needs to take place. While universities are beginning to train undergraduate and graduate students to compete in the working world, experts say, they also need to create academic programs to prepare data science educators and education researchers. Additionally, state, district, and school leaders and curriculum specialists must create opportunities for current teachers to engage in professional learning focused on supporting their development of data science and statistics concepts and practices.