When your students are researching, they are constantly looking for resources and explanations to help them understand the subject area or to make references in their work.
You may have noticed that search engines don’t always place the most useful stuff at the top. Actually, the most useful resources are often those that have been shared by the students themselves, which they normally just post on Facebook or send their fellow coursemates a link.
At the School of Engineering in the University of Liverpool, we were really interested in developing this method of sharing resources amongst our students, since it is something that is currently not systematically captured. We wanted to find a way to associate resources with particular study modules and so, with support from Jisc, we developed a community in which students share their impressions of which resources are useful and in what sense they’re useful.
For our engineering students, existing search engines don’t always return the most relevant results at the top, for a number of reasons:
- Many engineering terms are also used in common speech, though with a different or less well-defined meaning, e.g. ‘strength’, ‘beam’ or ‘section’
- Some engineering terms are shared with other academic disciplines; medicine is a prime example. Terms such as ‘stress’ and ‘fatigue’ mean different things to the engineering and medical/healthcare professions. Engineers often have to scroll through several pages of medical related results before they find things relevant to their field of study
- Engineering-related resources with pedagogic value can often be lost amongst those with a more commercial slant.
These issues are transferable to any discipline , so taking them on board, our aim was to develop an enhanced visual search engine with student involvement every step of the way. We adopted a crowdsourcing approach, to ask them what problems they faced in finding and sharing resources, and what would make this quicker and easier. To get us started they even spent the summer going through the internet and identifying resources that would be relevant to their particular modules. As they were searching, they also marked the ones which were not relevant to engineering modules, to save time in the future for their fellow students. For the project, we specifically targeted resources relevant to the users’ studies.
We’ve christened our new service ‘Kritikos’, from the Greek for ‘critical’ or ‘judicious’. To help the students speed up the process of identifying relevant online resources, we adopted three main approaches:
1. Use Google’s Custom Search Engine
We have created a Google Custom Search Engine to help filter out those results most relevant to the subject area. We chose Google over Bing or Yahoo as we found that their filters were easier to use, both for actual query terms and for the different media formats, and the results are returned very quickly. There are limitations however; the Google Custom Search Engine API is not free, and results are returned only ten at a time. We developed a main search page which allows students to find resources by different media types, with the results from our Google Custom Search being returned in real time.
2. Present the results as images, not text
The results are then displayed as thumbnail images. Clicking on a thumbnail opens up a detail page and an embedded version of the resource is displayed. Having thumbnail images and embedded resources saves time as it means that students can see at a glance if is of use to them, rather than having to wait to download a file only to find that it wasn’t what they were looking for.
3. Take advantage of student feedback
The crowdsourcing method hasn’t only been used in putting the resource together, we’ve continued it within the service by setting up an activity stream of other students’ interactions with the resource which is displayed alongside the media. Instead of storing this information in our own database, we decided to make use of an exciting new US-based initiative called Learning Registry. Students logged into the site can add their own interactions, noting if they found the resource useful and suggest which modules it’s relevant to. From here, they can also conveniently view all the resources that others have found useful to that specific module. Others can then agree or disagree with the judgements of their peers to help build a useful picture of the relevance of the resource. By doing this, it means that students are helping each other with their research, and ultimately makes the search for quality resources quicker and more accurate.
We are currently trialling Kritikos with our own students, many of whom have recently been completing projects, but who are now turning their attention (hopefully!) to revision. We are also talking to academics in other departments. This is exciting for them because although we’ve initially developed Kritikos for engineering, it’s completely transferable to different disciplines and to different organisational levels, e.g. institution, department, individual module/course. We also hope to trial Kritikos for a 3rd Year class at the Cambridge University Engineering Department between October and December 2013 and are also discussing trials with other universities overseas.
If you’d like to find out more on the Kritikos approach or how crowdsourcing might work in your institution, please contact Kritikos academic director Dr Tim Bullough (firstname.lastname@example.org) or me (email@example.com).