Open Education Evidence Hub

Open Education Evidence Hub

> An [Evidence Hub]1 is designed to help your members add their insights where they’ll make the highest impact, and through the use of different visualizations/maps, helps answer questions such as the following: > > * Who in my region is working on this problem? > * Are there any partnerships between projects in these two areas, on this theme? > * What are the key challenges we’re facing? > * Who has potential solutions to these, and what’s the evidence that they work? > * What evidence-based claims can we make with confidence? > * What are the most controversial issues? > > That’s why in every Hub you’ll find the core building blocks for this kind of collective intelligence: Key Challenges, Issues, Potential Solutions, Research Claims, Evidence, Resources, Organizations, Projects and People. In addition, you define a set of core Themes specific to your interests, which help organise all the ideas, people, projects and organizations. One such evidence hub collected [evidence around Open Education.]2 Since then a different code base has been developed using wordpress, as used – for example – in the learning analytics [evidence hub]3.  There are lots of reasons for that shift, which I won’t go into here, but while I’m a big fan of wordpress and rather like the new EH design, I’ve also spent some time recently thinking (with Ruth Deakin Crick) about how we can improve on the current instantiation, and make an effective evidence hub to support [improvement science]4. * * * So far, my thinking is being framed like this: Evidence specification: 1. Goals – what are we aiming for? What problem/key challenges are we trying to solve? (Goals) 2. Drivers – how will we get there?  What will it look like if we’ve solved them? (Outcomes/challenges) 3. Indicators – how will we know we’re on the right path? How will we measure our success? (Indicators/outputs and measurement model) 4. Activities – what are the routes or changes in direction to get to our goal? What activities or interventions can we make to produce the desired change? (Activities, learning design or strategies) 5. Evidence: What research and practice evidence do we have to support those activities? (Internal and External Evidence) 1. Did you get to where you were aiming, describe what happened, what worked and what didn’t? What did your indicators show? (new evidence) 2. What practical evidence from you, or your colleagues, did you draw on? (internal evidence) 3. What research evidence did you draw on? (external evidence) User experience: Thinking about how users interact with this system: 1. we’d want them to be able to add/edit content (especially at levels 3-5, but probably not at levels 1-2) 2. query the content to explore solutions to a problem, the evidence base for particular activities, etc., 3. and perhaps to propose new innovations for the community to try out 4. Each of these should involve interaction with structured, but seamless information, and adding/editing information shouldn’t be too arduous (involve lots of mandatory fields, or involve creating lots of new items when they only want to enter one, etc.). Design specification: 1. Page types should be structured to ensure relevant data is entered by users – e.g. ‘evidence’ might need a URL to an article, and a brief summary of claims made. 2. It should be possible to tag ‘relationships’ between pages; these relationships might be symmetrical (i.e. x is related to y, y is related to x) or asymmetrical (x is a parent of y, y is a child of x) – 3. We want to be able to visualise the page tree 4. For any ‘thread’ through the evidence tree (from outcome to evidence base) it should be possible to have a (visual and non-visual) overview of the connections; e.g. for all ‘top level’ outcomes, I want to know which measures measure them, and what research and practice evidence there is for those measures.  Similarly for each measure, I want to know which outcome it relates to, and what it’s evidence base is . 5. For any object, I want to be able to view within level relations (as well as the across level relations in ‘2’), e.g. to view activities that might be conducted together, or are similar. 6. Users shouldn’t need to enter the WordPress dashboard to add data, there should be a nice front-end form To do this I think I need: 1. To create ‘categories’ for each post-type in the schema (outcome, output/indicator, activity, evidence) – it may be desirable in the future to break each type down, e.g. ‘evidence’ as a taxonomy might contain terms: peer-review, practitioner, etc.) (but if this is the case, it would be better to create the types as hierarchical taxonomies, and to be as organised as possible from the beginning!). If using taxonomies, [UI plugin]5 is useful. 1. QUESTION: How do we link the categories so that their relationships are clear (e.g. outputs map to outcomes), without making them hierarchical (i.e., outputs are not a subtype of outcomes). 2. Create [advanced custom fields]6, such that when users select a particular post-type in the schema (using categories or taxonomies) they are prompted to fill in particular fields. Including: 1. A title (for all types) 2. Some description (for all types) 3. Relationships – for all types, this might include multiple boxes such that users are prompted to connect the evidence both ‘up’ and ‘down’ the logic model.  They might also be asked what items on the same level of the hierarchy are related (alternatively a [related posts]7 plugin might work for this). 3. Implement this mod of the [advanced custom fields plugin to create bi-directional relationships]8 4. Create taxonomies for various key concerns (which won’t be used for advanced custom fields, but might be useful for filtering/searching), e.g. discipline, level (1st year, M-level, etc.), target-group (e.g. particular accessibility needs, etc.) – these can be hierarchical and grouped as taxonomies in a way they can’t as tags, plus if people select a lower level in the hierarchy the upper levels should also be marked. 5. Use [gravity forms]9 (which apparently integrates well with advanced custom fields) to setup a nice frontend for users. 6. Use post templates to ensure the advanced custom fields are displayed appropriately to users 7. Visualisation – checkout [category D3 tree]10, [bu-navigation]11, [page-list]12, (but none of these do what I want). Martin Hawksey pointed to [his visualisation of blog posts]13. Looking at plugins to allow [sql queries from the admin interface]14 might be interesting. 8. On Ollie’s wise advice, I’ve also added a custom post type ‘internals’ to keep our internal-admin stuff away from the evidence hub main content. Ideation in the evidence hub I’m also interested in whether we can add a layer beyond the ‘static’ evidence hub. So, where I’m suggesting ‘activities’ above, we’d split that into two categories ‘we know it works’ for activities with an established research and practice evidence base, and ‘innovation’ for new ideas. The idea with the ‘innovation’ bit would be to have a space to come up with, discuss, and possibly vote on new ideas, to create novel approaches, usher the collective intelligence to flesh them out, and prioritise these. We’re using the [DW Q&A plugin]15 elsewhere which provides a stackexchange style vote up/down interface. You can imagine implementing more structure (using advanced custom fields) as ideas progressed, to e.g. prompt for project planning responses (I like aspects of the Wikimedia Foundation’s [idea lab]16 and [grants-application]17 space for this). With a Q&A element built in, there’s also potential to ask questions, with answers that could be connected to items in the evidence hub (again, using advanced custom fields).