What People Eat: Youth Telling Data Stories through Critical Data Visualization Reading (Poster 6)

• In Event: Decolonizing Mathematics Education Through Storytelling: A Tool for Exploring Identity, Justice, and Community

Abstract

Traditional data science education, rooted in computer science and mathematics, can often exclude diverse participants, leading to a lack of representation (Jiang et al., 2022). To address this, one promising approach is to engage youth in critically reading and reimagining existing data visualizations. This method encourages students to analyze the formatting (what variables are represented), framing (how variables are related), and narrating (what stories can be told) of a visualization, then reformat, reframe, and re-narrate it (Rubel et al., 2021). Renarrating is particularly vital as it allows students to engage with data in a low-stakes manner (Zhao & Dyer, 2024). However, the process by which students construct their data stories is still underexplored.

We present our preliminary findings on youth data storytelling when critically reading a data visualization. We applied two of Radinsky's (2020) data narratives modes: encompassing data into existing narratives and narrating oneself into a data-represented world. Our research involves an analysis of six semi-structured interviews with middle school students (pseudonyms: Ronald, Roy, Ben, Clark, Ruth, and Faith) who participated in a 14-week data-art inquiry project at an afterschool youth community club in East Tennessee in Fall of 2023. During the interviews, students were given 15 minutes to explore a data visualization about calorie intake and diet structure, followed by eight data reimagining questions derived from Rubel et al.'s (2021) framework. We recorded and transcribed the interviews, then conducted a thematic analysis (Braun & Clarke, 2012) to identify patterns in their data storytelling.

Our analysis found two themes: participants narrating as part of the existing story and others narrating themselves into their own stories. Illustrating the first theme, Roy, Clark, and Faith constructed stories by presenting their data understanding into the given data visualization’s existing narrative. Roy observed that global diet structures vary significantly, with different percentages of calories coming from various food categories. Clark highlighted the stark contrast in calorie intake between countries, noting that Americans consume an average of 3,600 calories daily, primarily from sugar and fat, compared to the healthier 1,600-calorie diet in Somalia. Faith compared the calorie intake of the United States and India, concluding that the Indian diet is healthier due to lower calorie intake and a higher proportion of grains.

In contrast, Ronald, Ben, and Ruth narrated themselves into their data stories. Ronald linked the unhealthy diet structure in the U.S. to his experiences with fast-food prevalence. Ben discussed the decline in produce consumption since the 1990s, advocating for increased vegetable and fruit production for healthier diets. Similarly, Ruth called for more awareness of dietary habits by comparing American calorie intake with global patterns.

Ddata storytelling fosters a deeper understanding of the information presented in data visualizations. By allowing students to either build/extend narratives based on data or position themselves within these narratives, this approach offers a more inclusive and meaningful entry point into data science education compared to traditional methods focused solely on symbolic representation. It also opens up the potential of other ways students can narrate and develop connections in the context of data visualizations.

Authors

• Yilang Zhao, University at Buffalo - SUNY

• Lynn L. Hodge, University of Tennessee