Building on the success of previous workshops at LAK, this year's Workshop on Methodology in Learning Analytics is focussed on understanding the methodological lay-of-the-land in learning analytics - to examine not only what we should be doing, but what we are actually doing. This hands-on workshop will develop participants’ understanding of the breadth and interplay of analytical methods used in the field, from Grounded Theory to deep convolutional neural networks. The workshop will be of interest to anyone in the learning analytics community, novice or expert, quantitative or qualitative, who wants to better understand what methodology means to learning analytics.
We will create a private repository of all LAK and JLA papers prior to 2019.
The objective is to use a mixed methods approach to defining appropriate categories/dimensions as a framework for a taxonomy of analytical approaches used in learning analytics.
Working first as individuals but in group tables, we will prepare brief annotations for each paper identifying: individual methods used, degree of emphasis on methodology, and how methods were used together. We will use a process of constant comparison to arrive at a coding rubric and then use that rubric to quantify the data for the collection of papers. From this, we will generate a table of overall distributions, illustrate trends over time, and identify gap areas.
Having arrived at a process for coding and having coded a significant number of papers in a database, some of the workshop participants will switch to work on creating an interactive visualization of the findings as an R Shiny application.
The interactive R Shiny visualization tool will update as new data are added to the database.
The remainder of the papers that remain unsorted will provide a natural basis for continuing the work of the community. We will use our Slack channel to continue the conversations while asynchronously working through the paper load. A final synthesis paper may be a long-term product authored by the Methodology in Learning Analytics Bloc.
Contact info: lakmethlab (at gmail)