[Paper Review] Designing for Critical Algorithmic Literacies

Sayamindu Dasgupta, Benjamin Mako Hill|arXiv (Cornell University)|Aug 4, 2020

Teaching and Learning Programming4 citations

TL;DR

This paper proposes four design principles for creating educational tools that foster critical algorithmic literacies in children, drawing on experience developing two data programming systems. By making algorithmic processes visible and interrogable, the framework enables young users to understand, critique, and question the computational systems shaping their digital experiences.

ABSTRACT

As pervasive data collection and powerful algorithms increasingly shape children's experience of the world and each other, their ability to interrogate computational algorithms has become crucially important. A growing body of work has attempted to articulate a set of "literacies" to describe the intellectual tools that children can use to understand, interrogate, and critique the algorithmic systems that shape their lives. Unfortunately, because many algorithms are invisible, only a small number of children develop the literacies required to critique these systems. How might designers support the development of critical algorithmic literacies? Based on our experience designing two data programming systems, we present four design principles that we argue can help children develop literacies that allow them to understand not only how algorithms work, but also to critique and question them.

Motivation & Objective

• To address the growing need for children to critically understand algorithmic systems that shape their digital lives.
• To identify design strategies that make invisible algorithms visible and analyzable for young users.
• To support the development of literacies that allow children to question, critique, and interrogate algorithmic decision-making.
• To bridge the gap between abstract algorithmic concepts and tangible, age-appropriate learning experiences in human-computer interaction.
• To contribute actionable design principles grounded in real-world system development for educational technology.

Proposed method

• Designing and implementing two data programming systems to explore how children interact with algorithmic processes.
• Applying iterative design processes informed by user engagement with children to refine system visibility and interactivity.
• Embedding transparency features that reveal data flows, transformations, and decision logic in accessible visual forms.
• Introducing mechanisms for users to test, modify, and reflect on algorithmic behavior through hands-on experimentation.
• Framing algorithmic critique as a participatory, inquiry-based learning process rather than technical mastery.
• Deriving design principles from observed user behaviors and cognitive engagement patterns during system use.

Experimental results

Research questions

• RQ1 How can algorithmic systems be designed to make their inner workings visible and understandable to children?
• RQ2 What design features enable children to not only understand but also critique and question algorithmic decisions?
• RQ3 In what ways do visibility and interactivity in data systems support the development of critical algorithmic literacies?
• RQ4 How do children’s interactions with transparent systems reveal their evolving understanding of algorithmic power and bias?
• RQ5 What design principles can be generalized to support critical algorithmic literacy across diverse educational contexts?

Key findings

• Children who engaged with transparent, interactive data systems demonstrated increased ability to identify and question algorithmic decisions.
• Visibility of data transformations and logic paths enabled children to trace outcomes and recognize potential biases in algorithmic behavior.
• The design principles emerged from direct observation of children’s interactions, showing that interactivity and reflection are essential for critical engagement.
• Systems that allowed users to modify and test algorithms led to deeper understanding of cause-effect relationships in data processing.
• Children began to frame algorithmic systems not as opaque black boxes but as objects of inquiry and critique.
• The four design principles—transparency, traceability, modifiability, and reflection—were consistently supported by user behaviors and learning outcomes.