Against the backdrop of increasing global humanitarian migration to highly

This study investigated what log files can reveal about learner behaviour of low-and non-literate adults learning to read for the first time in Finnish as a second language. The participants' reading development was supported by practising in an online training environment. Log files, automatically created usercomputer interaction records, were chosen as empirical evidence as their analysis enables in-depth post-activity exploration of student behaviour. The quantitative analysis resulted in user profiles containing information on learner engagement, performance and productivity. Overall, the results demonstrate that individual learning performance, process, and progress can be studied and reflected on holistically by investigating the individual's digital learning footprints, their log files. Log files are an accurate and precise, yet currently underemployed research tool. More easy-to-use tools for non-experts are in demand, as current Data Mining (DM) tools are designed for computer scientists and need to be developed further to become accessible and applicable by practitioners and educational researchers.

No abstract