Tuesday, September 30, 2008

CINAHL on EBSCO vs. CINAHL on Ovid

As it gets closer to the end of the year and libraries with CINAHL Ovid are forced to switch to CINAHL EBSCO, we are seeing more frustration from librarians regarding the differences in searching the two different search platforms.

I admit I am a die hard Ovid user. But I have been playing around with CINAHL on EBSCO ever since they released EBSCO 2.0. There are some things I really like about EBSCO's platform, but there are some things I really hate and find clunky when I am searching CINAHL. I am not the only one who feels this way, just browse through some recent MEDLIB-L threads and you can see plenty of others comparing searching on the two systems. David Dillard wrote very extensively about the differences between the two systems and how they handle multi database searching, nesting terms, field searching, and truncation.


Here are some of the things I noticed that have been giving me minor headaches:


There is a big difference for saving, print, emailing, and downloading. While I am not a big fan of Ovid's Citation Manager box and I tend to like EBSCO's folder system for selecting citations to print, email, save, etc. EBSCO's has not built into its system a method for saving all results in the set. EBSCO's system is set up for the individual searcher who can easily go through the results on their own and select relevant ones while conducting the search. It is not set up for librarians who conduct a search and then want to send the results to a patron. I just went to the eye doctor yesterday and my prescription changed, so perhaps I am just not seeing it, but there is no icon or link to select all search results. Therefore the librarian must click on every single citation to save to the folder. A slightly faster method to this would be to display 50 citations and then click on the link to add 1-50 to the folder. Then you click on the folder to send the citations via email. This is annoying and time consuming.


PDFs email out separately. If you are emailing a search that contains citations with full text articles in PDF form, those PDF articles and their citation are sent as individual emails. This method of emailing results is the default. Again this great for the individual searcher who will expect to see multiple emails in their account, but it is a poor method for a librarian sending a search results email to a patron. It is best for librarians conducting searches for patrons to uncheck the send PDF and send HTML as separate emails boxes. Leave the box, "current search history" checked. This will send all of the citations in one email and provided a persistent URL to each citation that the patron can click on and view their full text options.


While I take issue with EBSCO's method for emailing PDFs and HTMLs, please note that citations emailed from OVID are STILL lacking persistent URLs. This improvement is WAY OVERDUE from Ovid.


A few people have mentioned that thesaurus searching or searching using the subject heading is very clunky (at best) in CINAHL. I completely agree. This is one area that Ovid by far outshines EBSCO. We have the "Suggest Subject Terms" box checked as our default setting so when you type in a term like heart attack you are taken to the subject mapping screen where you can either check the box to search it as a keyword or check the box for Myocardial Infarction. Once you check a box (or boxes) you must click the search database button at the top of the screen. This is a little confusing because people tend to want to click on the browse button instead. The search box next to the browse button for searching the thesaurus not for specifically for searching or adding terms to your search. It is a poor location and poor design for this tool because it is easily mistaken as an add to search button.

If you want to add more terms to your original search then you must hit the Clear button at the top of the search button and type in a new term. This is extremely confusing! Everybody's first inclination is to type your new term in the box that says, "Add the following to your search: Term(s)." If you do this you will not be sent to the thesaurus and you will be doing a keyword search!

Some libraries have tried to prevent this confusion by using Advanced Search as their default screen. Advanced search gives the user multiple search boxes. This really doesn't help the problem, it just adds more search boxes to the confusion. EBSCO still defaults to searching terms as keywords when there is already a term in the first box. So if you want it to search your term as a subject heading then you must ALWAYS delete whatever is in the first/top search box and put the new term in.

Since you are doing a lot of single term searching (to ensure that you are searching using subject terms), you will have to click on the Search History/Alerts to combine your individual subject searches together. For example if you did three separate subject searches on cholesterol, myocardial infarction, and congestive heart failure, you must click on the Search History/Alerts link, delete everything in the search box and then check the boxes representing each subject search and combine them appropriately. If you do not delete everything in the search box prior to checking the boxes and clicking the "Add" button you will get whatever term is in the search box (usually the one leftover from your most recent search) combined with the items you checked. This is frustrating.

I am really going to miss Ovid's easy subject mapping and combining of terms.

Every database company creates their own platform and user interface. The methods you would use to search PubMed may not always be appropriate for Ovid Medline and vice versa. In the past librarians had the option to search CINAHL through different companies, Ovid, EBSCO, Proquest, and they usually picked the interface that was the right price and met their searching needs. Soon people will only be able to use EBSCO and they must adjust their methods and style of searching to best accommodate the interface. Not every difference is a shortfall, some things are better and some things are just different. EBSCO has a lot of nice features that Ovid and other platforms do not have such as the hover abstract and persistent URLs in emailed results. Only by experimenting with the systems will we learn the best methods for ourselves and our users.



Labels: , , ,

6 Comments:

At 11:55 AM, Anonymous Anonymous said...

Too true, Krafty. I share the exact same frustrations with EBSCOhost that you do. The continual need to hit the clear button, the inconvenient "search database" button when you are looking at mapping, ditto, ditto. I am also frustrated by the difficulty of what was an easy procedure in Ovid--send the search history and the search results via email to a user or even print it out. I can send one and then send another but for librarians it is important for our users to see and understand how we did their search so that we got THESE results. It is part and parcel of evidence based information practices. We won't and need the transparency of what terms we used and how we combined them. Yes, it keeps us on our toes to conduct a good and thorough search (which we should be doing anyway)but is necessary. I ususally advise my searchers when I train them to always save their search history and print it with their search so that when they come to me we can work together to see if they missed anything or so that they can recreate their search again later. EBSCOhost is much better than it used to be but still VERY clunky and difficult to use correctly.

 
At 1:19 PM, Anonymous Anonymous said...

I use CINAHL when helping faculty and students in our nursing program. This includes doing a lot of instruction. I am committed to teaching them to use the subject headings, but I believe the design of the EBSCO interface actually deters that. I also want them to drill down to the trees to take advantage of subheadings, the hierarchy, etc., but the layout of the interface does not encourage that. "Search history/alerts" could hardly be in smaller print. Nor could "Search Options" to get to all limits. Frustrating, frustrating, frustrating and a disservice to nursing.

 
At 7:22 PM, Anonymous Graham Spooner said...

Like The Krafty Librarian, we have had access to both Ovid & EBSCO versions of CINAHL and would certainly endorse the observations and comments made (and others who have left comments). Naturally, effective access to CINAHL is a BIG issue for us here at the College of Nursing in Australia.

However, one other thing EBSCO does better is the Explode function. In mapping to headings, if there are no narrower terms then Explode does not provide the box to tick. Ovid puts it there even if the term is already at the bottom of a tree.

 
At 12:23 PM, Anonymous Heather Ganshorn said...

Also, their RSS feeds usually either don't work at all to begin with, or stop working after a couple of months. Really stupid syndication? Really stressful syndication? I've started referring to the product as Ebsco 0.2

 
At 4:18 PM, Blogger Kevin said...

Graham: I use the EBSCO RSS service heavily and have not had any problems. If I do have any questions though I have found their support to be very helpful.
Check them out here:
http://support.ebscohost.com

 
At 4:02 PM, Blogger ML said...

Krafty,
I second your complaints. We've set up the interface so that users default to the Advanced Search and have only one search box in that interface; mapping to subject heading is also on by default. It's still clumsy to teach, though, and there is still that tempting, screen-hogging array of limits that users will see at the start of a search (novice searchers want to apply these to every search term, rather than waiting to apply them to a final set of combined terms). Regarding citation management, being able to manage no more than 50 citations at a time is a huge problem. We do a fair number of exhaustive searches, and those typically yield 100's of citations - who has the time to repeatedly download small sets of same?
Bottomline: EBSCO has done some tinkering, but their new interface is poorly designed, from a general usability standpoint, and from the stand point of expert searchers.

 

Post a Comment

<< Home

RSS Button Subscribe to this feed.
Creative Commons License
This work is licensed under a Creative Commons Attribution 2.5 License.
       
 
The Krafty Librarian has been a medical librarian since 1998. She is currently the medical librarian for a hospital system in Ohio. You can email her at: