What do people look for when they search online for learning resources?

I’m speaking at the JISC CETIS Repositories and the Open Web meeting on Monday 19th April, even if it is my birthday. I’m not expecting a card from the audience, but cake would be acceptable.

Speakers have been asked to prepare a position paper. This isn’t quite a position paper, but it is an outline of what I’ll be speaking about. It’s not that I’m lazy, well not much, but I’m not taking any particular position on an issue at this meeting (makes a change some of you might say). Instead I’ll be reporting back on the resource discovery survey run as part of our open education resources project. I thought some of the findings might be useful as an evidence base from which to frame a discussion about repositories and sharing learning resources. The final report isn’t written yet, and I’ve not done a detailed analysis of the responses (it’s a rich data set – 155 responses) but there’s a lot I can say from the summary data so far.

I’ve titled my contribution ‘What do people look for when they search online for learning resources?’. As part of the OER project I ran a survey that asked questions about people’s search strategies, where they they look for learning resources, how they evaluate what they find, what’s useful, and what’s less so. My aim was not to formally evaluate metadata, but instead to get some insight into people’s behaviour while searching for learning resources. It’s just a different side of the same coin as far as I’m concerned.

A couple of focus groups were run to unpick some issues in more detail, one with academic staff and another with students. Perhaps not surprisingly, students used a richer collection of search sites and services than their teachers. My view is they’re more motivated to. They also seemed better skilled at searching, or at least using the advanced search facility, if one was available. Must be all that library skills training they get.

No prizes for guessing that Google is the most used search service by a considerable margin, especially when combined with Google images and scholar search. And yes Wikipedia was very popular, including with staff, although it was only the student group that told us they start with Wikipedia to get an overview, but find what they thought in their opinion were ‘more authoritative’ references to cite in their assignments. Sadly Jorum was one of the least used services. More than one focus group participant asked “what’s Jorum?”.

What do people think about provenance of learning resources, licensing, openness, reviews & ratings, to download or to link in place, and many other important issues? Well that’s what I’ll be talking about in my presentation. But for those not able to come along on Monday, our list of recommendations below more than hint at many of these issues, and if I end up doing PPT slides I’ll upload those next week too.

What are the recommendations we came up with? Well, nothing astoundingly new, but these are evidence-based, which is what we wanted, and still an important set of considerations to reflect on when sharing learning resources. So, in no particular order:

• Make your learning resources easy to find. Clearly written descriptions of resources are always valued, especially if they contain the kinds of keywords you might use yourself if you were looking for similar resources. The more technical metadata fields are seldom used to find resources.
• Clearly license your learning resources. Consider Creative Commons, as users find this clear licensing framework easy to use, and it takes the guesswork – and admin – out of permissions.
• Make sure your metadata is searchable by Google. Federate your metadata to other repositories, or at the very least allow searching by third party search engines. In terms of ease of finding learning resources, many people prefer to use a small number of trusted search engines, most notably of course Google. So if you must create another repository, at least open it up to external searching.
• Make sure your resource is downloadable. This is a common practice, so resources that require server/client interaction might be less favoured than those that are self-contained and downloadable. If you want people to use your resources, then make it easy for them to do so.
• Make sure open really does mean open. The majority of users prefer not to have to register with a site or service to access your learning resource.
• Don’t worry about peer review. User comments, reviews and star ratings are not as important as some might think they are. Think carefully before investing in this functionality if you are creating a new repository. Formal peer review processes can be expensive to implement and our evidence suggests it’s not that valuable.

The full OER project report will be published by the HE Academy/JISC some time over the summer, and a separate paper on the resource discovery survey is in preparation. This work will likely be a proposal I’ll submit for Open Ed in Barcelona too, so hopefully see you there. In the meantime I look forward to sharing more of the data with those attending the Repositories and the Open Web in London next week.