This article was originally reviewed for my statistics and research methods class.
Lowery, R. (2011). A Visualization Tool for Atlas Collection Assessment. Journal of Map & Geography Libraries: Advances in Geospatial Information, Collections & Archives, 7:2, 138-153.
This article examines the use of visualization tools in collection assessment, particularly for hard-to-assess and non-circulating materials such as atlases. The author examines the University of Illinois Chicago collection of geographic materials to determine their range and depth, as well as look for gaps within the collection.
I found this study very straightforward – the author had a simple yet effective study that was clearly explained and seems easily replicable. The idea of using visualization tools to understand collections seems creative and a pushing-forward of traditional assessment methods (which I know admittedly very little about, so it’s possible that I am completely mistaken). As a Digital Libraries person, I’m frequently amazed (and frankly overwhelmed) at the variety of online tools and code snippets that could be utilized in creative ways to help us better understand data – but it also seems like these tools have a fairly low penetration into non-digital-librarian activities. Hopefully studies like this will spark more thought about new ways of examining and understanding library collections.
This study also seemed helpful in terms of enabling librarians to understand collections that don’t have the usual circulation data, or that are unique in other ways (such as the finding aid that the author mentions). On the other hand, the process does have some drawbacks that might make it more suitable for use in conjunction with other assessment methods, rather than as a stand-alone.
For instance, the data cleaning process, while fairly straightforward, might not be as easy as one thinks. Obviously, there are the usual stopwords, but it was clear even from this study’s relatively simple results that refining data is not necessarily a one-step process. In addition, I’d be wary of using an algorithm I don’t fully understand to examine data. I think it might be an interesting experience to find a collection with which one is extremely intimate and then look at the ways the data does or doesn’t show things based on how it is refined.
Lastly, I found the literature in this study to be slightly lacking. I found the discussion of various ways collections are traditionally assessed to provide good rationale for why this new tool might be a more effective measure, but I would have liked to see other examples of visualization tools being used in this manner. Perhaps those do not exist?
This was a clear, well-written article that examines an interesting prospect for collection analysis and development. Although this study focused on atlas collections at an academic library, and so is primarily directed at collection development librarians at those institutions, the author made a strong case for the use of these kinds of visualizations for assessment at other kinds of institutions and with other types of resources.