An interesting article in First Monday, Why first-year college students select online research resources as their favorite by James P. Purdy
523 students at “a mid-sized Midwestern university” responded to a questionairre about their preferences for doing online research; the questionairre also asked for free-entry answer as to why they preferred what they preferred.
Google and Google Scholar were by far the most preferred sources for online research, although library online databases were third (ahead of “Books”, public records, face-to-face interaction, online forums, etc).
Most interesting is why they said they preferred what they preferred: Ease of use, ease of use, ease of use.
However, “the second most popular reason students identified a research resource as their favorite was quality. Students indicated that they favored a resource that returns trustworthy, credible, and scholarly sources…. That students simultaneously valued scholarliness and favored Google and Google Scholar may signal a misperception of the scholarliness of Google’s results and an overreliance on Google Scholar’s judgment of scholarliness.”
Also note, “The third most frequent reason students cited as guiding their choice of favorite research resource was connectivity. In other words, they valued a resource’s digital affordance to connect directly to a source text.”
I think these findings are consistent with the analysis and strategy outlined in my position paper on improving article search: If the library wants to keep a hand in providing a research service we have some control over, we need to focus on providing a service that is easy to use above all else. Of course, we also care about the quality of the results students get — even if they don’t. Not only may students not be adept at judging the quality of results, but they may not even care about relevance of results: “These results suggest that students favor resources that return any results, particularly scholarly results, above resources that return relevant results.”
That students valued quality over relevance indicates students may define research as meeting particular task criteria, rather than generating knowledge. For example, they may see good research as referencing five scholarly sources rather than conversing with topically relevant sources.
If these students are representative, this is perhaps a cautionary note with regard to my study (which you’ll hear about later) to compare quality of search results from different tools by asking them to rate results — they simply don’t care. On the other hand, if we want to provide users with better results (whether they care or not), the first step is getting them to use the tools — and this suggests the quality bar may be fairly low in getting users to use the tools, so long as the tool is easy to use.
And the importance of providing a tool that links users directly to fulltext, or perhaps access/delivery options.
This study reveals that students preferred research resources they find straightforward and uncomplicated. They likewise preferred resources that return scholarly results. They also favored resources that link them directly to sought texts rather than provide only bibliographic information. Students indicated they are motivated less by how quickly a resource returns results or if it offers a variety of results. They are also motivated less by whether a resource provides topically relevant sources.
I am feeling confident about the position laid out in the position paper: To meet the desires of our users, we need to provide tools that are as simple as possible to use, and well-integrated with library delivery options. If we can’t do that, a certain portion of our users won’t even use our tools, making the quality of their results or sophistication of their search functions irrelevant.