If I were to ask you what you do for fun while you are at work, I am sure you wouldn’t tell me that you analyse your library’s OPAC transaction logs. Especially not those transaction logs where you peruse all those patron searches that don’t actually find anything! Well, I have been doing exactly that, and I have to admit I found it fascinating.
As you may already know, librarians have been analysing OPAC logs for many years. The first research paper I found in a very quick literature search was written by Thomas Peters in 1989, and reported on a study conducted at the University of Missouri-Kansas City. After reading through the study I was chortling to myself in glee because the author reported a search failure rate of 40.5% (Peters, 1989, p. 270). We at the Temasek Polytechnic (TP) Library only had a 26% search failure rate for the year 2010!
I then found a study conducted at another university in the US in 1998 where they did not actually report an overall failure percentage, but rather gave a post-test figure for the number of “correct syntax searches” that resulted in a failure of 31.35% (Blecic et al., 1998, p. 48). This paper also highlighted a study by Hunter in 1991 (Blecic et al., 1998, p. 40) at the North Carolina State University where the overall search failure rate was 54.2%. Wow, the TP Library was doing well, I thought!
When I found a more recent study from the College of New Jersey Library (Moulaison, 2008) where they were looking at a search failure rate of more than 30%, I was starting to feel really good about our OPAC. In fact, when I removed the figures for the failed ISBN searches (on the basis that these searches were conducted by book vendors and not by our patrons) I came up with the even more impressive figure of a 20% search failure rate for the TP library.
But then I started to delve a bit deeper. In the 1980s, the libraries that were fortunate enough to be online were using those funny old green and black screen terminals (usually situated right next to the even funnier card catalogue), and the default search was almost never a keyword search. I think some of us may remember as library staff having to learn how to enter a search correctly, and to think we expected our patrons to learn all this too! Today many libraries (including the TP Library) have a “Keyword Anywhere” OPAC default, which cuts down drastically on the number of search failures.
Moving on into the 1990’s. It seems that librarians during this time were starting to wake up to the problems inherent in OPAC, but many of us just lengthened the instructions on the introductory screen. When I think of Google’s simple search screen, I despair at how many of us tried to concentrate on teaching the patrons how to use OPAC, instead just making OPAC easier to use! Anyway, in terms of search failure figures, I think most of us had not yet started placing “Keyword Anywhere” searching first. I know that the TP Library OPAC in the late 90’s was emphasising Title searches.
So does this mean the TP Library OPAC is not as good as I originally thought? Certainly not! Although the day we start thinking there is nothing left for us to improve, is the day we should perhaps start looking for another job. Things are changing all the time, and at an incredible speed too. In fact, I was forcibly reminded of this the other day by a speaker at a recent presentation talking of Google “arcing” or nearing the downturn of the product lifecycle bell curve. I am not sure I can imagine a world without Google so I hope they re-invent themselves fast!
But I digress. I started this post by trying to entice you to look at your own library’s OPAC transaction logs because you can learn all kinds of interesting things. I have found, for example, that our total number of searches peaks twice a year, following our user education programme very closely. I also worked out that if you look at the total number of OPAC searches conducted in terms of TP’s total student population, there is average of 3.6 searches per student per month. Of course, not all of our students even know what OPAC is, but it is fun to manipulate these figures.
So, have I convinced you to start analysing your OPAC transaction logs yet? You may come up with some surprising results and, even if you don’t, this type of exercise is a really good way to get you thinking about OPAC in a different light. And who knows, you may be that very special person who finally develops an OPAC to rival Google!
References
Blecic, D. D., Bangalore, N. S., Dorsch, J. L., Henderson, C. L., Koenig, M. H., & Weller, A. C. (1998). Using transaction log analysis to improve OPAC retrieval results. College & Research Libraries, 59, 39-50.
Moulaison, H. L. (2008). OPAC queries at a medium-sized academic library: A transaction log analysis. Library Resources & Technical Services, 52(4), 230-237.
Peters, T. A. (1989). When smart people fail: An analysis of the transaction log of an online public access catalog. Journal of Academic Librarianship, 15(5), 267-273.
Contributed by Debby Wegener, Temasek Polytechnic Library