On Friday, May 18, Lauren Pressley and I went to Perkins Library at Duke University to attend the inaugural meeting of the North Carolina Summon User Group. My last visit to Perkins Library was 20 years ago, so just getting to visit their beautiful facility and amazing coffee shop was a treat in itself! I was surprised to discover the intentional dearth of signage on their campus and was very thankful that Lauren knew where we were going.
Upon registration, we were greeted by two very friendly Duke librarians and given very nice Serials Solutions portfolios as well as name tags and name table tents. The meeting began with a lunch hosted by Serials Solution. There were about 50 in attendance with several schools represented including NC State, ECU, UNC-CH, Campbell, Duke, and others.
Eddie Neuwirth, Senior Product Manager for Summon, began the program by introducing himself (he lives in Cary!) and his colleague, Vince Pella who is a Customer Service Representative and flew in from Seattle for the meeting. Eddie began by giving an overview about what is new and what’s coming with Summon. He said that Summon is currently being used by 450 libraries in 40 countries and that 38% of the largest Research Libraries in North America are using it. They are currently approaching one billion records with more than 90 content types. They push out new releases every three weeks that can be added by local Summon administrators. We were pleased to note the newest release gave the option for the Widget and Search Box Builder to “Keep Search Refinements.” This was one of our concerns that we noted in our own recent usability study and we were pleased to see that they had addressed this problem! Also, the save and preview items icons are now visible (another concern we noted in our usability study).
He then introduced a new widget that will be rolled out this summer called the “Discipline” widget that uses Ulrich’s categories to group journal titles and it uses call numbers to group books. He noted that not everything is mapped to the Discipline widget and then he gave an example of how to use it by limiting the Discipline widget to Biology and searching for “Blood Cells” (which yielded 600,000+ hits).
He then proceeded to talk about other enhancements such as spotlighting images (digital repositories) and a recommender feature that will suggest LibGuides, databases, etc. The biggest announcement was that they are working on a product targeted for 2013 that will replace the need for OPACs. He gave a sneak preview of what the records will look like and they appear much like our current catalog records. They are also working on A&I citation displays so there will be sneak previews to the abstracts and database records. Eddie also mentioned a product they are developing called Intota that is a web-scale development product that will eventually eliminate the need for integrated library systems as we know them.
After Eddie’s talk, there was an hour of lightning rounds which were five fascinating 10-minute presentations where people from different universities presented research related to their use of Summon. The first up was Patrick Carr from ECU who presented his findings on the impact Summon has had on E-Journal use. He reported that the cost per session came to $.39 and the cost per search came to $.08. They found that the use of Sage and Springer journals increased four times. The use of Elsevier increased 10-15%, but the use of JSTOR was down 10% and the use of EBSCO journals was down significantly. In a later part of the meeting, Eddie explained that the numbers being down in JSTOR had to do with the fact that JSTOR provides limited access to the bulk of their metadata (JSTOR gives Google Scholar full access to all of their metadata). In the end, the download of full-text downloads stayed flat for pre-Summon and post-Summon at 1.4 million.
The next presenter was from Duke and he demonstrated their use of Summon to include their visual collections (scanned items from special collections). They were still working on Summon’s ability to include thumbnails which they anticipated to be working by this week.
For me, the highlight of the lightning rounds was hearing Karen Cicconne from NC State. They were a beta site for Summon and thus have three years of experience under their belts. Karen presented her findings from comparing the results of an EBSCO group cluster search to the same search in Summon. She used the exact wording and punctuation used by students in the EBSCO searches to see the result in Summon. NC State offers a group of course tools for every course taught at NC State, and each course offers an EBSCO cluster at the top of the page. She found that the EBSCO searches had consistently better returns but there were a couple of searches that did better in Summon. (Side note: one of the consistent themes throughout the day was bemoaning Summon’s inability to sort by relevance.) Her next study will be comparing the results from Google Scholar searches with Summon searches. Her initial reaction is that both are good (as a reminder, they are only searching journal articles in their Summon searches).
Next up was Ginny Boyer from ECU’s Joyner Library. She did a survey with ECU’s Health Science librarians to gauge their perception of Summon and its usefulness for the medical field. They found that while most did not use it, they did not feel strongly about it either way. They did a study to find out their proficiency with the tool and found mixed results. There seemed to be little or no success with winning over the Health Science librarians to Summon. The kicker to this presentation came with Karen Cicconne from NC State spoke up and said that research shows that people in the health sciences do not need Summon, they can find everything they need in ChemAbstracts and Medline. The key to success in medical searches is MESH headings.
The last group were Anita Crescenzi and Kim Vassiliadis from UNC’s Health Science library. They did a Summon usability studies that was very elaborate in its execution with 170 applicants vying for the $20 gift card offered in appreciation for 60-90 minutes of their time. Their results were identical to ours (which we did with student assistants in no more than an hour for each session). A couple of their observations: Summon searches titles very well, but loses relevance searching after that. Their students also mentioned the desire to limit results from the beginning and the desire to have the results grouped by format such as journal article or book.
We ended the day with discussions around tables based on common interests. Lauren and I gravitated to the instruction table where we talked with Karen Cicconne (NCSU), Emily Daly (Duke), and Sarah Steele (Campbell). We found that most of us do not teach Summon, but sometimes it is used as an example of a broad search versus a targeted search in a specialized database. Karen gave a very interesting statistic that said that they found that 74% of their students began their searches on their website using the “All” option, with 40% of second clicks going to the Summon articles, 30% going to their catalog, and 30% going somewhere else on the page. If you have time, I suggest that you take a look at NC State’s Library homepage to see how they are implementing Summon: http://lib.ncsu.edu/
I apologize for this very long reflection, but as you can see, it was a day packed with fascinating information. They are hoping to make this an annual event and I highly encourage others from ZSR to attend next year!