Diversity in learner usage patterns

Affordances and limitations of learning analytics for computer-assisted language learning: a case study of the VITAL project

“…Among the reasons cited to explain this situation is the fact that the data were not very accessible and the collection, processing and interpretation of data in non-CMC learning environments was very time-consuming or required a lot of technical resources (Fischer, 2007(Fischer, , 2012. Supporting instructors and course designers to collect and process behavioural data arising from CALL contexts will require the development of more intuitive and easy-touse tools as well as a wider variety of relevant functions.…”

“…When the computer is used as a tool or tutor for learning, artefacts created by learners interacting with each other can be analysed to inform second language acquisition. For example, computer logs of human-computer interactions reflecting a learner's actions in a dedicated learning program include a lot of diverse data about the use of tools and resources (Fischer, 2012). According to Fischer (2007), this kind of "empirical evidence in support of the value of pedagogical features built into CALL programs […] is fundamental to the advancement of knowledge in the field of CALL" (p. 414).…”

“…In task-based language teaching, for example, researchers have examined the difference between the task-asworkplan (the lesson plan outlining how the instructor intended the students to complete a series of tasks), versus the task-as-process (the way students actually completed the tasks) (Samuda & Bygate, 2008). To ascertain to what extent students use desirable strategies while learning, researchers have found new strategies to observe students' actual learning behaviour (Chapelle, 1989;Hegelheimer & Chapelle, 2000), particularly as the use of student questionnaires alone was not a reliable source of information (Fischer, 2012). Computerdocumented data indicating the processes through which learners complete a task were found to deliver the most reliable basis for empirical evidence for hypotheses from Second Language Acquisition (SLA), such as the noticing hypothesis or most effective types of learner feedback (Hegelheimer & Chapelle, 2000).…”

“…While the use of Learning Analytics for CALL is relatively new, learner tracking data, whether script-based or collected by means of screen recording software, has been used for several decades to answer specific research questions in the field of second language learning (Fischer, 2007(Fischer, , 2012Hwu, 2013). A study by Debski (2003) which analysed 91 CALL research articles published between 1980-2000 in seven English language journals found that the use of computer-based tracking methods evolved from 0% in 1984-1988 to 25 % in 1990-1994 and 15.94 % in 1996-2000; the other main methods were questionnaire/survey (36.23 %), document examination (34.78 %), interview, observation, think aloud protocols, and language tests.…”

“…When considering the collection and analysis of learner tracking data, it is important to distinguish between human-human interactions via the computer (e.g., Computer-Mediated 5 Communication (CMC)) and Human-Computer Interactions (HCI) (e.g., mouse clicks, keystrokes, and all entries made by a learner) (Chun, 2013;Fischer, 2012). When the computer is used as a tool or tutor for learning, artefacts created by learners interacting with each other can be analysed to inform second language acquisition.…”