Rearchitecting Data for Researchers: A Collaborative Model for Enabling Institutional Learning Analytics in Higher Education

Abstract

This article presents the case of the Learning Analytics Architecture (LARC) dataset, a collaborative effort at the University of Michigan to develop a common and extensible tool using administrative data and designed primarily for learning analytics researchers to investigate enrolled students' academic careers, demographics, and related teaching and learning outcomes. The institutional context prior to the creation of the dataset and the rationale, design, development, and maintenance involved in creating LARC are all detailed. Also discussed are the procedures for access, documentation, and ensuring the continued usability and relevance of the dataset for a growing learning analytics and data science research community. The authors conclude the case description with recommendations for institutions seeking to replicate this effort. Notes for Practice• Administrative data in higher education is not typically ready "out of the box" for learning analytics researchers. How can institutions leverage their own resources and knowledge of their data to reduce time-intensive and repetitive data-cleaning efforts to make the most of their own data?• This article presents the case study of the LARC dataset, as well as the decision points and process that the committee members undertook to build and maintain the data. This kind of effort remains a major hurdle for many higher education institutions.• A framework is provided for other institutions interested in building a similar dataset for learning analytics research. Six areas of consideration are discussed: (1) facilitating partnerships across departments, roles, and levels;(2) level setting and arriving at a common level of understanding among team members; (3) obtaining buy-in from relevant stakeholders and data stewards; (4) designing for the needs of the specific institutional context; (5) utilizing national and international standards whenever possible; and (6) understanding the institutional landscape of learning analytics and designing datasets accordingly.

Practice Makes Better: Quantifying Grade Benefits of Study

“…The University of Michigan's definition of underrepresented racial/ethnic minority (URM) is tied to nationality and requires some exposition. As defined in the Learning Analytics Data Architecture (LARC) 15 Data Dictionary, students are considered international (ITL) if they are neither U.S. citizens nor U.S. permanent residents. The international student population is roughly two-thirds Asian, with the plurality of students from China (46%) and another 5% from the Republic of Korea.…”