The Digital Media and Learning Initiative
Data Visualization for K-12 Learning
This convening brought groups together in Princeton and in Palo Alto at the Carnegie Foundation for the Advancement of Teaching to discuss the question of how to make interactive data meaningful—collecting, integrating, analyzing, and presenting—over long time periods. Both groups had experts in data collection and interactive learning environments, building data visualization tools, school-based and informal learning, and digital assessment. The varied perspectives allowed for richer conversation around how to address the primary question.
Participants agreed that data visualization for K-12 learning holds unique challenges. There are many stakeholders in this arena for whom data is increasingly important, not just at the school level for No Child Left Behind, but for parents who want to know how their child is doing more specifically than a single letter grade for half a term; for teachers who want faster feedback on when students are struggling with a new concept; for students who need to be able to assess their own progress and judge areas where they can do more work or where they excel.
Since NCLB became law, ever more data is coming from standardized tests. Additionally, new forms of data are being produced from innovative learning tools, for example in schools with SMALLab (Situated Multimedia Art Learning Lab) or in after school programs like Global Kids. The growing multitude of data also takes a multitude of forms, as representation techniques advance with new technologies. These changes present unique opportunities and challenges: There is ever more data about ever more varied topics, which goes far beyond simply having an end of year test score on a state-required test.
As data becomes increasingly complex and varied, strategies to make that data meaningful must similarly grow more sophisticated, while keeping the results clear and intelligible. For example, if a teacher sees the output of a large blue circle for a given project, that has to mean something useful: that a child did well, or that a child needs more help. This is a challenge that designers must keep in mind when making tests or building tools for young people, as eventually data gains more power when connected to other data.
Participants discussed how in some ways, one good analogy for successful data visualization in learning in learning is a well-designed game: the player and anyone watching the play know immediately where the player is in the game, how she is doing, what resources she has, and where she goes next. There is transparency, quick feedback, clear scoring, and evident objectives; most importantly, all of these are easily and immediately interpretable.
Producing data is an important first step. But it is far from enough. In schools across the country, parents can see teachers and administrators wrestle with these challenges in the simple question: how should a child’s report card look? Is a letter grade enough? Should there be handwritten comments? Should tropes brought into the classroom to help students learn be reflected in what a parent sees? How often should information on the student go home?
After discussing challenges that each person was facing in his or her own work, participants agreed that they wanted to learn more. Coming out of these discussions, some participants wanted to begin working on a resource that will show examples of data visualization, both stronger and weaker. Ideally, this will evolve into a resource that people in the education field can share and contribute to in all phases, from data sets to analysis to visualizations.