The truth about search seems to be more astonishing than anything you could imagine. Lin Lin, EBSCO Senior User Experience Researcher, talking at the UKeiG Annual Meeting last week, provided some startling revelations, drawing on EBSCO’s wide experience of observing search behaviour with students ranging from age seven to postgraduate - they should know, since they claim to have the largest user research team in the industry, So what is the reality of student search?
When a student is asked to use the Web to write a research paper, they typically start their research between 11pm and midnight, when they are sitting on a couch, not at a desk. The research starts with four steps:
- Then “serious research”.
Students show a perhaps surprising anxiety about the whole process. Their starting point is to something they find reassuring, something they call “our oxygen”: Google. After they have looked for the topic with Google, they usually turn to Wikipedia, which after all is often the first hit retrieved, and the students use three specific parts of the Wikipedia entry:
- The definition – the overview, at the start of the article, in lay-person’s language (this confirms they are looking at the correct topic).
- The table of contents for the Wikipedia article – this becomes the table of contents for their research essay.
- The external links and references at the end of the Wikipedia article. This is where they jump to get further information.
Equally surprising is how the “serious research” is actually carried out. Says Lin, “linear navigation is over” (which she explains as looking at one topic, then another topic, then another topic). Instead, users open multiple browser tabs and obtain multiple search results for the same topic. This technique is derived from online shopping. The tab browsing involves collecting pages that look to be relevant. These tabs are kept on their desktop, sometimes for several days, a practice described by Jakob Norman of Nielsen Norman as “page parking”, which he contrasts with a seemingly similar sophisticated search methodology called “parallel browsing”, and provides helpful tips for how to deal with it at that site).
The restricted range of this searching is the most worrying: if the Wikipedia entry is not particularly good, the student is unlikely to turn to anything better.
Another alarming discovery is the number of search-related terms that students don’t understand. Based on a survey of 208 US students, the following terms were not generally understood:
- Catalog (or catalogue)
So what are the lessons here? Well, it is worth thinking about the student sample. In the questions after the presentation, Lin Lin stated that these results apply to search behaviour at all levels, but the results above were obtained from US college and high-school students writing humanities research papers in an area about which they have little prior knowledge. Fortunately, perhaps, not all research starts from zero knowledge of the subject. Nonetheless, some principles for presenting search interfaces for publishers are clear: avoid technical terminology in any help pages. Recognise the limited search strategies of most users. Wherever possible, make it clear there are other sources of information than Wikipedia, and that Wikipedia’s list of references may not be a fully considered bibliography on the subject.