Open AccessSynergies of Learning Analytics and Learning Design: A Systematic Review of Student Outcomes
Author(s) -
Marion Blumenstein
Publication year - 2020
Publication title -
journal of learning analytics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.084
H-Index - 7
ISSN - 1929-7750
DOI - 10.18608/jla.2020.73.3
Subject(s) - learning analytics , personalization , computer science , blended learning , educational technology , higher education , educational data mining , psychology , knowledge management , data science , mathematics education , world wide web , political science , law
The field of learning analytics (LA) has seen a gradual shift from purely data-driven approaches to more holistic views of improving student learning outcomes through data-informed learning design (LD). Despite the growing potential of LA in higher education (HE), the benefits are not yet convincing to the practitioner, in particular aspects of aligning LA data with LD toward desired learning outcomes. This review presents a systematic evaluation of effect sizes reported in 38 key studies in pursuit of effective LA approaches to measuring student learning gain for the enhancement of HE pedagogy and delivery. Large positive effects on student outcomes were found in LDs that fostered socio-collaborative and independent learning skills. Recent trends in personalization of learner feedback identified a need for the integration of student-idiosyncratic factors to improve the student experience and academic outcomes. Finally, key findings are developed into a new three-level framework, the LA Learning Gain Design (LALGD) model, to align meaningful data capture with pedagogical intentions and their learning outcomes. Suitable for various settings — face to face, blended, or fully online — the model contributes to data-informed learning and teaching pedagogies in HE.