Design and validation of learning trajectory-based assessments for computational thinking in upper elementary grades

Brian D. Gane,Maya Israel,Noor Elagha,Wei Yan,Feiya Luo,James W. Pellegrino
DOI: https://doi.org/10.1080/08993408.2021.1874221
2021-02-26
Computer Science Education
Abstract:<span><b>Background &amp; Context:</b> We describe the rationale, design, and initial validation of computational thinking (CT) assessments to pair with curricular lessons that integrate fractions and CT.<b>Objective:</b> We used cognitive models of CT (learning trajectories; LTs) to design assessments and obtained evidence to support a validity argument<b>Method:</b> We used the LTs and evidence-centered design to develop assessments that 144 Grade 3 and Grade 4 elementary students completed following the integrated instruction. We analyzed data using multiple psychometric approaches.<b>Findings:</b> The design approach and data analysis suggest that the assessments are well-suited to evaluate students' CT knowledge, skills and abilities across multiple LTs.<b>Implications:</b> We show how to use LTs to design assessments that can yield valid inferences about students' CT competencies; these methods can be adopted and extended by others to create additional assessments that can advance computer science education. </span>
What problem does this paper attempt to address?