Identifying components of learning analytics in education and providing a conceptual framework for optimizing learning

Research output: Contribution to journalArticleAcademicpeer-review


Background and Objectives: Learning analytics is a new and promising field of study in education that seeks to capture, analyze and report data about learners and their learning environment for the purpose of optimizing learning and its environment. Data and analytics are the two main keywords for learning analytics in which data is the feeder of analytics to provide evidence-based insights about teaching and learning. Although learning analytics has been of interest to many scholars during the last decade, little research has been done to provide a comprehensive framework of learning analytics. This study is aimed at identifying learning analytics components in education to provide a conceptual framework for optimizing learning. Methods: This is a qualitative study in which the design of the study is content analysis. The thematic analysis which is the research method suggested by Braun and Clarke was used in the following six steps: (1) familiarizing with collected data, (2) generating the initial codes, (3) searching for the themes and components, (4) reviewing the potential themes and components, (5) defining and naming the themes and components, and (6) reporting the results. In this study, 14 experts in the field of learning analytics were interviewed. Purposeful sampling method was used to select the participants. Moreover, the strategy for selecting these experts was based on the relationship between their theoretical and research activities. The reason why 14 experts were interviewed is the theoretical saturation which means data collection process continues until no new data is collected. That is to say that the theoretical saturation method was used to determine the sample size. To collect data, unstructured interview was performed. Data analysis was performed in three stages including open coding (line by line coding), axial coding (combining codes and developing a category of more general concepts) and selective coding the process of choosing one category to be the core category, and relating all other categories to this category). MAXQDA software version 2018 was used to run data analysis. The validity of the findings was assessed by the content validity index (CVI) and the reliability of the findings was determined based on Cohen’s kappa coefficient. Findings: The results showed that learning analytics is comprised of seven main components, including environment (background, culture, communication), objectives (optimization, learning, recognition, awareness of the process of learning, feedback, self-regulation, personalization, motivation, supervision, and assessment), stakeholders (learners, teachers, learning designers, administrators, and parents), data )meta-data, meaningful data, academic background data, academic data, performance data, interaction data, and psychological data), levels of analytics (descriptive, diagnostic, predictive, and prescriptive),], process (collection, analysis, report, and interpretation), and technique (analysis of social networks, clustering, categorization, prediction, regression, decision tree, factor analysis, discovery of association rules, discovering sequential patterns, and descriptive analysis], which altogether provide the conceptual framework of the learning analytics components in teaching for optimization of learning. Conclusion: Based on the findings of the study, the users of learning analytics in education can be recommended to consider these seven components when they are using them to optimize learning. Based on the findings, recommendations for future research and practical activities are made.
Original languagePersian (Iran, Islamic Republic of)
Pages (from-to)937-948
Number of pages12
JournalJournal of Technology Education
Issue number4
Publication statusPublished - Oct 2020
Externally publishedYes

Cite this