This week we are asked to map our previous barriers to implementation of learning analytics to the ROMA model.
This is a huge task, as the 7-step ROMA models throws up issues that can occur throughout the process, which can require years of investment and development.
“In spite of significant investment over a nine‐year timeframe, Course Signals has not yet been deployed across the entire university.”
For this reason, I’m going to focus on Step 1: Map political context. Or rather: the change that needs to be managed:
Discursive change: changing how information is communicated and shared?
Procedural change: changing how something is done: how decisions are made, how learners are supported
Content change: changing written policy with regard to evidence‐based support of learners
Attitudinal change: changing how key stakeholders perceive the project
Behavioural change: making sustainable changes in the way student success is achieved or supported
Mapping this to the idea of communities, we can see parallels between my own questions of what issues arise from different stakeholder communities and where changes need to be implemented.
Are enforced policies required to change behaviours or can training and education on the possibilities of learning analytics result in creating the desired and necessary behaviours for effective use of LA? As Ferguson observes:
“Educators need to be able to evaluate any implementation of analytics tools in order to use them effectively. Learners need to be convinced that analytics are reliable and will improve their learning without unduly intruding into their privacy. Support staff need to be trained to maintain the infrastructure and to add data to the system. Library staff need to be able to use the analytics to shape their practice and resources. University administrators need to be convinced that the implemented analytics provide a sound return on investment and demonstrably improve teaching and learning quality. IT staff need to put workflows into place so that raw data are collated, prepared for use, and made readily available to end users. ” (pg. 126)
The core issue of learning analytics seems to be its scale: this is not a small project that can be quickly implemented or ‘tried out’ on a whim. LA requires serious investment, time and resources, and commitment; on a system that until now there is little evidence makes a significant difference to the learning experience. As a topic I find it fascinating, but big data only starts to become valuable when they can be monetised. In an increasingly commercial higher education market – when and how will investments be realised, and what are the ethical repercussions of this? That’s a topic for next week and the next section of this module.