By the Writing Data Stories Research Group
WRITING DATA STORIES is a new project that seeks to reorganize how young people, especially linguistically and ethnoracially minoritized students, learn about and interact with data. The project will engage middle school students in exploring scientific datasets about earth and the environment using flexible online data visualization and analysis tools.
Typically, school data investigations use small datasets that students create themselves, or larger datasets that clearly illustrate simple relationships and are less connected to students’ lives. Our goal is instead for students to author “data stories” that reorganize everyday and scientific conventions to position students themselves and the complex issues they care about at the center of each data investigation. In the process, students will be supported in learning to elicit support for their cause, critically reflect on their experiences, craft scientific arguments, and manipulate or wrangle large datasets.
At the core of this reorganization is a “syncretic approach” in which students deeply study everyday and scientific ways of knowing or doing that are traditionally in tension. This approach was developed specifically to support learners from nondominant backgrounds, including students identified as Dual Language Learners.
We are interested in bringing together both experience and data. It is common for people to dismiss personal experiences as subjective and suggest that 
empirical data are objective truth. Of course, neither of these are correct – experience is empirical, and data are subject to measurement error, sampling error, bias, and omission.
What’s more, personal experience can expose problems with data, and data can help situate and contextualize personal experience. Consider the Flint, Mich., water crisis: Official data analyses contradicted resident reports for years with tragic consequences; citizen science efforts where residents collected their own data eventually exposed the city’s neglect.
Writing Data Stories similarly seeks to put students’ personal experiences and public scientific datasets into direct conversation. In this way, students’ everyday knowledge and practices have equal value to scientific knowledge and practices, inviting students more authentically into the practices of science as they learn them.
While data serve as a powerful form of evidence that is valued in science, data is certainly not the only way that people convince one another of something. For example, in our first iteration of the curriculum, we are exploring advertisements as an everyday genre. Advertisements use a variety of conventions – emotional appeal, aesthetics, humor, celebrity endorsement, and yes, statistics – to sell a product. They also raise important questions that we don’t often ask, but should, about data: What assumptions are made about a population? Who is represented, and who is missing? What counts as a norm, a difference, and an outlier? Why are certain attributes of a person (such as ethnicity, income, popularity, age) recorded, while others are not?
Scientific reports of data analyses are often data- and evidence-rich, but often do not address these questions. They also rarely contextualize their findings in ways that call readers to action, or help them understand the everyday and human impacts of issues – both things that advertisements do very well.
Writing Data Stories has just started, and will continue for three years. In this first year, we will work with a small group of partner teachers to develop curriculum, software, professional learning supports. In years two and three, we will expand to serve more classrooms and students as we revise and share our curriculum. Given the interdisciplinary nature of the project, we expect that it will inform research in science education, data science education, and literacy education.
Editor's note: Read the abstract for "Writing Data Stories: Integrating Computational Data Investigations into the Middle School Science Classroom." The collaborators on this project are: Michelle Wilkerson and Kris Gutierrez at UC Berkeley; William Finzer and Natalya St. Clair at the Concord Consortium; Hollylynne Lee at North Carolina State University; and Anthony Petrosino at the University of Texas at Austin.