•  
  •  
 

ORCID

Danielle McNamara, https://orcid.org/0000-0001-5869-1420

Abstract

The increasing scale of digital learning environments has transformed the data available for educational measurement, learning analytics, and the learning sciences. However, learning at scale is often treated as a matter of sample size or novel analytic methods rather than as the outcome of institutional infrastructure that governs how learning data are generated, curated, transformed, and accessed. This paper offers an infrastructure-centered description of learning at scale, focusing on learning management systems (LMSs), platform-integrated tools, and data pipelines that produce research-ready educational data. As an illustrative exemplar, we draw on the design and operation of the Arizona State University Learning@Scale (L@S) digital learning network platform to describe (a) sources of data commonly available at scale in higher education settings, (b) processes by which those data are curated and made accessible, and (c) constraints imposed by privacy, governance, and equity considerations. Particular attention is given to language-based data—such as discussion boards and written assignments—which represent a uniquely rich but underutilized source of evidence for educational measurement. We argue that understanding learning at scale requires attention not only to analytic methods, but also to the infrastructural conditions that shape construct representation, subgroup visibility, and validity.

DOI

https://doi.org/10.59863/ZVLT7884

Download

Share

COinS