Measuring Value and ROI of Academic Libraries: The IMLS Lib Value Project

Editor's Notes

• #3: Also University of Pittsburgh, Drexel, 4 universities in NY
• #10: Academics read a lot—they report on average reading 25 articles, 8 books or book chapters, and almost 12 other publications. This appears to be pretty close to reading in the US and Australia. On average across all disciplines. 295-300 per year compares to 280 per year 23.3 per month in 2005 in the U.S.Books or book chaptersOther publications include web sites, conference proceedings, gov docs
• #11: Source of book readings is different—purchased or from publisher is main source; library is second.
• #12: <1% responded they read the article “elsewhere”For ALL responses (n=1162): Library was only 1.7%, Office/Lab was 60.2%, Home was 27.9%, and traveling was 10%
• #13: Value of virtual visits to the library.
• #14: Question: Did you obtain your article through a print or electronic source?2011 Frequencies: Electronic: 658 (print=49)2005: Electronic: 394
• #15: Outcomes from reading are another important explicit value that go beyond just sheer amount of reading. From nearly 900 respondents in the United States (Tenopir et al, 2009a), readings were found to have many important outcomes. Faculty members told us that their Readings have profound outcomes. They:• Inspire new thinking (55%/54 of readings)• Improve results (40%/38 of readings)• Narrow, broaden, or change the focus (27%/28 of readings)• Resolve technical problems (12%10 of readings)• Save time (12%/10 of readings)• Lead to faster completion of the task at hand (7% /5 of readings)• Foster collaborations (6%/4 of readings)• Waste the time of the reader (<1%/<1 of readings)
• #47: Our data will not match reported ARL ebook data since we report titles acquired to ARL and publisher use stats record volumes acquired.Also data for 2008 is sketchy for use data.
• #48: **Note #New Ebooks for 2008 is a rougher estimate. Cost per PRINT book in the CARLI consortia 2003-2008 was $64.55.Spiro and Henry note that while “the initial costs of acquiring electronic contents may be higher (than print), the long-term costs will be lower” IN fact, my data point to lower e content initial costs AND low long-term costs.
• #49: ** Some vendors have only one way of reporting use – downloads/views of the entire book rather than individual sections. This is rare, only a few vendors, but it means that that use is a view of the entire book. **Since use data gathering could only get to 82%, it’s not 100%!! Not all use was able to be counted.
• #50: 7 of the 40 ebook vendors we purchase ebooks from were not able to supply ebook use data. Also note that Open Access ebooks are not currently providing use data (just like open access ejournal publishers). One of the many problems with these data is that we often buy huge ebook packages at the end of a fiscal year, or they get turned on then – those are counted as ebooks available or purchased in that year, but actually can’t really be used until the next fiscal year. AND, total use for earlier years are less reliable (undercounting) due to unavailable data.*Total uses per year generally undercounted because only publishers for which we have data--which actually excludes most Ebooks. According to ARL data, our largest Ebook collections come from free resources that don't track usage for us: Early English Books Online, ECCO, Archive of Americana, Monograph print study (CARLI Consortia) found a range of cost/use from $18.00 to $35.00 for print monos (five years, FY2003-2008)
• #51: Ebsco only started providing ebook use data in 2011,and very spotty too! Not useful. No password, despite a few months of trying, from Archives of Americana. EEBO is used mostly as a database for locating where print copies of the books are located. American Council of Learned Societies had only a few titles that were used a LOT. Over 7,000 uses of one book in a few months, obviously used for a text for a class. E-reserves basically. Again, return users not unique users (but not that there’s anything wrong with that).
• #52: So 11% of ebook holdings account for nearly 50% of use; not exactly the 80/20 rule, but close! Wiley switched to normal COUNTER data in calendar year 2009, and the way they report usage prior to that means that the #Ebooks Total for FY2008 is incomplete. So very likely if we had the full number you wouldn't see that 20% to 10% dip in %Ebooks Used by Publisher from 2008-2009 that you see on the graph now--the %used in 2008 for Wiley is realistically lower than 20%.
• #53: Large proportion of unused ebooks is due to buying large package deals; more selective buying, or PDA ebooks might result in a more thoroughly used collection. Very much “just in case” instead of “just in time.” However, in a different, broader study of monograph use, print monographs had a “not used” rate of 33% (CARLI consortia) over a 5-year period. Other studies have shows that “not used” rates of up to Approval books at Penn State had a non-circ rate of 31% and UIUC’s was 40% (approximately a 2.5-year study period).we continue to buy “Just in Case” books in e format, just like we do in print. BUT, lower costs overall for e books. We can afford to have a broader collection since there are lower costs associated with ebooks. More on that to come later in this presentation.
• #54: This is a high number of AVERAGE uses per ebook. And trending upwards. Due perhaps to multiple users possible per ebook, or returning users who “check out” the ebook each time they use it. ???
• #55: Are these repeat users or new users? Probably some of both, but the lack of a distinct “comet tail” of single uses per single book means that ebooks are used differently than print books. Repeat users equivalent to renewals in print circulation?
• #56: Paul N. Courant and Matthew “Buzzy” Nielsen in: The Idea of Order: Transforming Research Collections for 21st Century Scholarship. Council on Library and Information Resources (CLIR), Washington, D.C. Accessed online at: http://www.clir.org/pubs/reports/pub147/pub147.pdf.
• #57: To summarize this section,
• #60: IRB APPROVAL RECEIVED FIRST! Elsevier generously offered an incentive of a $100 Amazon gift voucher to each participant. The response was overwhelming. Within a week nearly 400 UIUC volunteers had signed up to participate. Elsevier asked us to stop recruiting on October 15, and they eventually weeded the number of UIUC participants to 129 faculty and Ph.D. students.
• #61: Very few Humanists and Social Scientists. Self-selecting? Lots of chemists (and other physical scientists and engineers) who have been widely exposed to ebooks for the past 5-6 years.Social Science + Humanities + Interdisciplinary = 8 0r less than 1 percent. So looking at results by discipline is not valid. I included a few just to see how those 7 responded!
• #65: Monthly or weekly or daily basis = about 76%. Somewhat surprising for scientists? I don’t think so, but there is a sense out there that scientists and engineers don’t use books. Wrong!Daily or weekly basis: 47.3% or nearly half of users are “frequent users”
• #67: Nothing new here! Can cite other studies with similar results.
• #68: Asked to check as many boxes as apply. Not wildly different uses and needs from the different disciplines – in fact very much the same across the disciplines. Ebook use behaviour is not determined by discipline or subject of the ebook.
• #69: Note here that the earlier, self-reported “Behavior” pretty closely matches what they value. So we may be able to say that in the future behavior is a predictor of value, which makes sense. People don’t spend time doing things they don’t value. Even when you are rewarding them for their participation.
• #70: 67.4% say they either “Need to have” or it would be “nice to have” the ebook they viewed. Based on 129 persons reading over 800 ebooks
• #71: Comment box following the valuation question elicited the most responses, 435 comments. That comment box was simply labeled “Please elaborate” So we saw in the previous slide that the response was mostly positive in terms of value (70/30 split basically). However, more comments that followed the question were negative, although by a small margin. Because the overwhelming negative comment was about the lack of relevant material, not about format, we can see that this model has “legs” and can keep running. Content and access needs to improve. But users are ready.More negatives than positives, most focusing on lack of relevant material and not having access when neededTotal responses totaled 435
• #73: YMMD Your Mileage May Differ! USE IS LOCALLY INFLUENCED!
• #74: “Non-Standard” here means “confusing to users!”