How do we present the past to users today?

[Jody DeRidder]

Lucy Barber of NHRPC raises some questions for us in her blog post:  How do we bring the past to learners of tomorrow?

In particular, she asks readers:

• What are some of your favorite websites that provide access to historical records?
• If you design websites for historical records or use them, what do you expect?
• What do you think we need to be doing today for the “Learners of Tomorrow” in terms of historical records?

If our online content is not useful for current and upcoming users, clearly we must find ways to make it so.  I contend (and research thus far supports) that different user groups prefer different presentations.  Advanced researchers have been found to prefer access to digital items within the context of the finding aid.  Undergrads prefer item-level description search and retrieval (though in one study I did, the ones with no previous digital library experience actually found items quicker via the finding aid).  Research I've read indicates that grade school levels respond best to image icons for browsing (and depend on browsing over search), and the younger ones prefer primary colored icons. 

My most recent study of faculty researchers indicates that search capabilities are *not* currently meeting their needs, and they are thus heavily dependent upon refining and limiting capabilities, and upon browse functionality.

Who are your primary audiences, and what do you know about what they need and want?

--jody

[Elizabeth Joan Kelly]

Our primary audiences for in-archives research and virtual reference requests about our collections are primarily undergraduates and outside researchers (mostly "advanced researchers"). However, we do not know if these demographics are the same for our digital collections. I'm aware of the California Digital Library's project using a one-question survey asking Calisphere users for their role (teacher, student, independent researcher). I'm interested in other methods of determining who is using digital collections.

Last year I did a small usability study with staff and students to find out whether users could find what they wanted in one of our digital collections which employs pretty minimal metadata. Ultimately they could, but this sort of study is a little artificial because we were directing users towards a collection they might not have gone after rather than seeing how they search when they're actually doing their own research. Still, I was happy to see that users were able to find materials by searching within the digital collection. 

Our finding aids are just displayed in our catalog and on our website as PDFs. We provide links to digital collections in the catalog records and on the webpage where we display all of our collections but not in the finding aids themselves. Jody, did your researcher study involve EADs or another format for the finding aid? 

In answer to your questions--we have a good idea of who our in-person researchers are but not who our digital researchers are. We get feedback from our in-person researchers on what collections they'd like to see online and have done some minimal testing to see if (one) collection is find-able and usable. We need to know more about who our primary audiences are and what their individuals needs are.

[Jody DeRidder]

Hi Elizabeth  --

   The study to which I was referring compared the use of accessing digitized items via the EAD finding aid, versus searching for them individually.   In this study, the items linked into the finding aid had minimal (auto-generated) MODS records, which pretty much only provided context, number of pages, and rights statements.

   Focusing on testing our own interfaces is the primary way most usability studies have been performed, and it's an important start.  Another level would be to test the findability in major search engines.  A further level would be to study how researchers and scholars locate what they are seeking;  I have seen only a couple of studies on this latter aspect.  They seem to indicate that, as in seeking other answers, people turn first to those around them and to familiar sources.  Secondarily, they may seek further (such as library databases);  as a last resort (it seems) they contact librarians.  

If this is really true, then it behooves us to make our content more findable in web search engines such as Google, using common search terms.  Efforts I've seen in that direction utilize site maps, static browse interfaces that link into the databases, and schema.org encodings to help the search engine usefully index the contents of the pages crawled. 

Jody L. DeRidder
Head, Digital Services
University of Alabama Libraries
Tuscaloosa, AL 35487
Phone: 205.348.0511