The afternoon session at LASI gave us a chance to look at Gooru a “search engine for learning”. It an interesting tool that aims to foster a blended learning platform that allows teachers to pull full curricular activities, or small snippets of activities from both existing repositories (e.g., Khan Academy, National Geographic). Teachers can implement quizzes, interactives, web pages and other rich media content, and their own designs can be uploaded to the repository for other instructors to use or remix. Underneath all of this is a rich (HUGE) amount of tracked user data to give teachers insight into the state of their class (from individual students to whole class information). So much data can be both good and bad though.
Prasad Ram (the project lead), highlighted some of the challenges in this space for practitioners: What’s the granularity that the teacher is interested in? Is it one student? The whole class? What kinds of reports are actually useful to the teacher? How does this data allow teachers to personalize the learning for students?
These questions are not easy ones to answer and in many ways lie at the heart of effective Learning Analytics implementation. One approach for making this data relevant is through effective visualizations – but as many people know good visualizations are hard (and bad visualizations are worse than none at all!). Gooru is still trying to think of all the ways that these might be used and so far they are making headway, although I don’t envy their task. I think this is something that is going to take a lot of work and they will have to tread lightly to not overwhelm the teachers – giving them everything may result in them using nothing (see orchestrational load).
I applaud their really interesting approach to large-scale implementation of such an ambitious platform, but I have to admit that it concerns me a bit about their overall model of approach. It seems that most of their curricular designs fall into the lecture/drill/quiz model (also popular with Khan Academy, but I actually think Gooru does it better) – which puts the learning a bit too much on rails for my tastes. Also the work seems to be very much siloed to individual students (rather than collaborative work) and goes against some of my ideas on the needs to support authentic STEM practices required in today’s “Knowledge Society”. If I were to push I’d like to see how Gooru could use its vast collection of knowledge resources to support students in collaborative inquiry curricula.*
* Talking to Prasad at the end of the talk he mentioned that there are social features built into Gooru but that these elements (for moderation, abuse filtering) haven’t been fully built out yet. I’d be interested into see further how these are done and the kinds of interactions they are supporting, as I see this what would make it really transformational