A student came to our reference desk last semester asking for help finding books on a narrow civil rights topic. The librarian couldn’t find any relevant books either and recommended that we assess our civil rights collection. A couple of us reviewed our civil rights monograph holdings from the past decade against a list of titles provided by ProQuest.  

The verdict? Our collection is quite robust overall. We could not identify any monographs focused on the students research topic (but there are journal articles out there).

My biggest takeaway? Our discovery layer can be a big pain to use.

Instead of using the catalog, I spent hours searching for books using the patron-facing discovery layer. I already use it regularly but never so intensively. It wasn’t a great experience. I made that choice on a whim and it turned out to be very educational for me. Now I am more confident than ever that students are missing good, relevant books when they search. Here are the most significant problems I experienced during this project:

  • Book reviews clutter and confuse the results. Including book reviews adds minimal additional value and wastes researcher time in having to filter them out or scan through them for the actual book result. I was frustrated with them. It’s uncommon someone will want reviews instead of the book.
  • Results that link out to a book index are useless. I don’t want to see whether EBSCO has a book in its system. I only want to see if we have it.
  • JSTOR results are inaccurate. I would often get a result showing we have the book through JSTOR. Often we did. But several times I clicked on a JSTOR result only to be taken to a page on JSTOR telling me we don’t have the book. 

In short, there were lots of false positives (JSTOR, EBSCO, and book reviews masquerading as book results). I did not experience many false negatives. However,  false negatives happen when I mistyped or the spell-checker changed a word to something I don’t want. Catalogs and discovery layers aren’t as good at finding close matches as Google is. If I were a student, I would be super frustrated.

What to do about this? Librarians can:

  • Provide feedback on the discovery layer. Settings can and should be tweaked. Feeds and search can be improved. 
  • Consider making the catalog the default search setting. That’s what I prefer if you have a separate discovery system for law students and faculty use.
  • Curate lists. Curated lists can take a ton of time but remain a valuable tool for researchers, particularly for topics that arise frequently in paper classes. (And then we have to bring those lists to our students’ attention, but that’s a different challenge…)

In conclusion, federated searching can be overrated. There’s still something to be said for having the option to go directly to a tool with a narrow focus (like CRS reports, or articles from a specific newspaper) to search for what you need, and there is something to be said for forcing students to intentionally select the content type they are searching for before they search. The same is true for book catalogs. More isn’t more—sometimes it just causes confusion and wastes time.