ABSTRACT
The data for learning analytics can come from many different sources and can be classified in various ways. The types of data can be vital for learning analytics but they are recorded and held primarily for other institutional purposes. One of the primary types is the demographic data the institution holds about students, such as their date of birth, sex, socio-economic status or ethnicity. The UK Information Commissioner's Office suggests a method of classifying personal data, based on its origins: provided data, observed data, derived data and inferred data. The learning management system (LMS) and the student information system (SIS) are likely to be the primary systems from which data for learning analytics is extracted in large-scale learning-analytics deployments. Commercial solutions generally target these systems as their main data sources. A significant problem for learning analytics is that learners may carry out many of their learning activities outside the monitored and recorded confines of the institution's LMS.
