Peter Murray
Google Scholar's Dramatic Coverage Improvement Five Years after Debut -
Abstract: "This article reports a 2010 empirical study using a 2005 study as a base to compare Google Scholar's coverage of scholarly journals with commercial services. Through random samples of eight databases, the author finds that, as of 2010, Google Scholar covers 98 to 100 percent of scholarly journals from both publicly accessible Web contents and from subscription-based databases that Google Scholar partners with. In 2005 the coverage of the same databases ranged from 30 to 88 percent. The author explores de-duplication of search results by Google Scholar and discusses its impacts on searches and library resources. With the dramatic improvement of Google Scholar, the uniqueness and effectiveness of subscription-based abstracts and indexes have dramatically changed." - Peter Murray
"the conclusion cannot be clearer: libraries can seriously consider cancelling a large number of subscription-based abstracts and indexes since their unique contents and value are rapidly evaporating." - John Dupuis
I sorta predicted it in 2004. - John Dupuis
well, sure you can cancel the journal index if it's only value is the link to the article. I predict some indexes will add more value and other services to keep themselves relevant. One good example now is SciVal, which mines Scopus data for profiles of research output and strengths. Of course some of them won't figure this out, or use certification requirements to stay in business, but otherwise that's my call. - Elizabeth Brown
Unfortunately (or fortunately, I'm not sure), beancounters may regard the added value as de minimus, or at least not worth the $$. I'd never say I was happy to have been fired or that Eureka! disappeared, but the writing for most (not all) A&I non-full-text databases has been on the wall for a while. - walt crawford
A good example of added value is SciFinder. It's hard to imagine chemists giving that one up any time soon. On the other hand, I'm sure there are lots of A&I companies out there who are really sweating. And if they're not, they should be. - John Dupuis
+1 Dorothea - Mr. Gunn
I don't think it is crawling MLA directly, and I would bet that MLA is one database that it doesn't duplicate very well. Now, something like Library Lit from Wilson would be a different matter... - Joe
It is covering like 98% of the current journal literature, but I wonder about the older journal volumes that have not been scanned in, and the obscure conference papers, and book chapters, and newspapers, and lots of other stuff that hasn't been digitized. - Joe
Well, Joe, I don't think the question is "Is GS better than A&I Service X" but rather "Is GS better than X + $20K per year" or whatever. I suspect that services aimed at faculty and grad students will do better than ones aimed at undergrads if only for that reason. For ugrads, if it isn't online, it doesn't exist. For researchers, it's much less likely that one old, obscure document will be just as good as any other. - John Dupuis
Steve, maybe because the big honking amalgamation gives people the illusion that they don't have to know anything or try very hard with their searches? PubMed is free, so we don't even have to worry about canceling it. If people find what they need via Google Scholar, that's fine with me. What I worry about is people with more serious questions (i.e., docs looking for clinical answers, researchers - not just students writing assignments) *thinking* that they're finding the best papers (when they're not, necessarily) because "Google is so easy." - Rachel Walden
Sure, Steve, I'm guilty of a gross over-simplification. On the other hand (and without actually being able to check...), I suspect that the trend over the long term is downward for ILL just like for print book circulation. My point is that ugrads have different research habits and timeframes to work with that are going to make them somewhat less likely to value a particular unique document as opposed to some other document which is not quite perfect but is more readily available. - John Dupuis
A problem with GS at the moment (AFAIK) is that it has no API. I'm guessing this is part of the deal Google has cut with its sources. So doing anything with GS (like automatically constructing citation counts for a set of articles) is a screen-scraping job. Yuk. Also, unlike the others, we don't actually KNOW what the coverage is... - Chris Rusbridge
BTW Happy, the ILL staff may be busy, but I'm guessing their counts are in tens, max hundreds per day. That's probably a couple of orders of magnitude down on GS use. I don't think it disturbs the generalism: if it isn't online, it doesn't exist (pretty much)... - Chris Rusbridge
I know I'm being picky, but I did a GS search for the author, and found a "related article" citation (to an earlier article on a similar topic) from one T Roy of the Library Journal... - Chris Rusbridge
And since I'm being *very* picky, I note with sadness that yet again this article by a librarian is toll access (US$38 to me). Not that the author had a huge amount of choice in library journals... - Chris Rusbridge
seems like I went home right when the conversation got interesting... I think ILL will always be used, because overall the quality of a lot of scanned material is not that great. Plus many scholars need/want to see the physical object when they're researching a project. Our ILL activity is way up since more articles are being found through GS that we don't have, and we're a pretty well-stocked academic library. I predict ILL will remain strong and even increase as the need for unique materials expands, especially for mid-size and larger academic campuses. - Elizabeth Brown
There was a similar article a couple of years ago: Google Scholar’s Coverage of the Engineering Literature: An Empirical Study and that one is available: - John Dupuis
And Steve, the fact that GS use dwarfs ILL is both relevant and irrelevant. On the one hand, connecting people with the documents they need is one of our vital missions. On the other, resources are limited and we have to design our services and build our collections within those constraints. I would argue that we're heading into an era where we'll be able to redeploy some (much?) of what we've been spending on A&I towards other priorities. And those priorities would of course vary a lot from place to place -- some might value ILL, some buying scanned backfiles, some on other things. - John Dupuis
I agree with Steve on the last part there. Students need to be led to the view that they need to use all useful things, not just the easy to find things. - Pete's Got To Go
Steve & Pete, I'm with you 100% on that one and have the IL scars to prove it. All I would say is that GS is just as valid a tool as any other for helping students find the right document for their needs. I mostly don't teach GS, I mostly teach more subject-specific databases but frankly for a lot of the subjects I teach GS is very comparable. As for the "if it isn't online it doesn't exist" attitude, I suspect it may be somewhat more likely for science people than for others. - John Dupuis
Sure, to the extent that GS disrupts the A&I industry that's an opportunity for libraries to both save some money and teach & promote the tools that students want to use anyways. - John Dupuis
'Librarians and faculty alike often assert that "all researchers use Google Scholar." Based on this study, this is essentially correct.' Good article from the latest ISTL on science researchers: - John Dupuis
On the other hand: 'Not all traditional fee-based databases (e.g., Web of Science) and not all subject-specific article databases (e.g., PubMed), are in a "death spiral." ' - John Dupuis