Most early-grade teachers rely on basic assessment strategies centered on observational tools provided by the local education department. While these observational tools offer a broad overview of classroom learning, they fall short in measuring specific skills and tracking individual children’s progress – thus limiting their overall effectiveness.
At Trackosaurus, we’re addressing this gap by developing self-directed assessment games that provide deeper insights into each child’s developmental journey. Collaborating with leading education experts, we design our games to align with cultural contexts and emphasize a strengths-based approach to skill measurement.
No single assessment method – whether fully observational or fully direct – works for every skill or every teacher. Some skills are best assessed through observation, while others require direct measurement. Teachers also have different assessment preferences, which can evolve over time. For these reasons, we believe a hybrid approach to formative assessment works best.
However, designing effective hybrid tools comes with challenges. The Trackosaurus UX design team is tackling these challenges by exploring ways to seamlessly integrate multiple assessment methods in a single tool without overwhelming teachers. Our goal is to create a tool that balances usability and accuracy, ensuring that different modalities of assessment complement rather than contradict one another.
A key focus is data presentation – how to provide clear, actionable insights when, for example, observational and direct assessments of the same child yield different results. We’re developing intuitive ways to reconcile these types of differences in our hybrid tool, helping teachers make informed instructional decisions without added complexity.
The Trackosaurus team is investigating how to use Voice AI to assess children’s foundational skills in low-resource environments. For example, we’re currently helping to develop the following models and processing systems to unpack children’s early literacy skills using their verbal responses during gameplay:
Each model requires specialized input data and fine-tuning strategies, which differ between low-resource and medium-resource languages. Once the models are built, we also need to employ clever app development techniques to ensure our Voice AI-powered assessment games work well across a range of mobile devices and connectivity conditions.
