By jPablo Caballero | June 27, 2018
Introduction to the xAPI Cohort, by Torrance Learning, and to one of the projects in the Spring 2018 “semester”: xAPI and Instructor-Led Training. With an interactive video!
(It is recommended to read the article first to get the context, but if you want you may jump directly to the interactive video)
Twice a year, in the Spring and in the Fall, Torrance Learning organizes the xAPI Cohort. It is a learning initiative in which participants freely form groups of interest around various xAPI-related topics. Each group works on a project about their topic of interest. The Cohort lasts about twelve weeks. There is a weekly one-hour online meeting, in which the groups share their progress, and there is also a presentation about some area related to xAPI. Access is totally free, and participants can get involved as much or as little as they want. It is perfectly ok to participate just as an observer.
The participants’ degree of knowledge about xAPI varies greatly. Many are new or almost new to xAPI, and they learn through the weekly presentations and through interaction with other participants. There’s a Slack area for the Cohort, with various channels for the teams and for other topics of interest.
In the Spring 2018 I joined the group ILT-xAPI, whose interest was to take advantage of xAPI to express and record in a known and homogeneous way data about instructor-led sessions. The idea was to use some system that would allow a group of participants to share one or more presentations, that would record real-time data about how users were using the system, and that would finally send all the data as xAPI statements to an LRS (for later analysis). Thinking of the system in such a generic way, we reasoned that we could collect data not only about the behavior of the students (how they follow, react and answer questions during the presentation), but also about how the instructor uses the system. The general objective would be to analyze all the data at a later time to obtain information about the participants’ responses, but also to try to infer more general patterns, such as attention and/or participation levels, understanding, timings (use, response), pace set by the instructor, amounts of interactions that the instructor uses in his/her sessions, etc.
Since such a system did not exist, I developed a prototype with enough functionalities to materialize those initial ideas, at least partially. The system that was developed is Open Source, and it’s available at github. It is just a limited prototype, but it implements enough enough functionality to test out our initial ideas. Even with just the few features it implements, it is possible to collect and generate a fair amount of data from an ILT session.
This system is just an instrument to facilitate data collection. As in any setting in which you want to reach conclusions from data, it is necessary to plan which questions you want answered, what data would provide answers to those questions, and how to obtain that data. In our case, the data is gathered with the prototype system, but to make meaningful use of it, it would be necessary to define/design the presentations to be used in the ILT session, the interactions/questions/quizzes to use, etc.
Since it is just a prototype with a narrow objective and there was very little time to develop it, its interface and general usage is quite rudimentary. The system is usable and serves its purpose, but one needs a bit of technical knowledge to be able to use it.
Our group of interest was started initially by one person (Roland Barrera]) who was interested in gathering data from instructor-led sessions. A few of us who were already enrolled in the xAPI Cohort and were also interested in this topic by some reason or another, joined the group. In any case, the project was initiated in a small setting by a small group. What I would like to know is whether this type of system and this type of data would be useful to a wider range of professionals. That is why I would like to ask, particularly to those who deliver, manage, or carry out any professional activity related to instructor-led training, if they think that a system like the one described -but much more evolved and usable- would be useful.
Two of the main advantages that xAPI brings are homogeneity the expression of data about learning experiences, and centralized storage. These two characteristics may facilitate the cross-analysis of data that came from a wide variety of systems: eLearning modules, LMSs, other systems (ERP, APIs) etc. What we wanted, with our project in the xAPI Cohort, was to provide one more source of data, a source that can be particularly rich: what happens in an instructor-led session in the classroom.
Please watch this interactive video that briefly demonstrates and explains the system that was developed.