[LAW-LIB:59575] RE: In search of bad online searches

From: Bryan M. Carson (bryan.m.carson@gmail.com)
Date: Thu Jul 23 2009 - 16:54:16 PDT

  • Next message: Kent Olson: "[LAW-LIB:59576] Re: In search of bad online searches"

    Folks,

    There was recently a discussion on law-lib about using bad searches to teach students how to construct good searches. This led me to post a request to another discussion list (libref-l) asking for similar examples. I have compiled the results from both lists, along with related threads dealing with fake history and a hoax quotation. The results from both lists are compiled below.

    Bryan M. Carson

    *****************************************************************************

    With my biology students I often stress the need to search both the common and scientific name of an organism to get results. The example I usually use is "killer whale" vs. "orca" - the latter almost always gets significantly more results, and of course, searching both with the Boolean 'OR' is best.

     Jenna Ryan
    Louisiana State University
    jryan1@lsu.edu

    *****************************************************************************

    I use the example of trying to find articles about "Martin Luther King Jr." but getting information about Martin Luther and his relationship with the King during the Reformation. Also, I had a friend who would use the example of trying to find information about the history of cowboys in Texas and finding articles about the football team.

    Leslie M. Haas
    Loyola University Chicago
    lhaas@luc.edu

    *****************************************************************************

    In my library we have nursing students searching for topics that have no relation to each other. I can't remember the exact search but I mimic one that goes along the line like "pregnancy AND bone fractures". I explain that these topics have plenty of results individually but not very many useful ones together.

    In the history end I demonstrate the difficulties of searching for historical eras or years. Many students want to search for "women's rights AND 1800s". I explain how searching for eras are difficult because there are so many ways of referring to time periods.

    Andy Heiz
    amh_library@YAHOO.COM

    *****************************************************************************

    One that worked well for me in Academic Search Premier was "civil war" and causes. It pulled up a lot of articles about Africa and Bosnia, and the students and I got to chat about why that happened and the importance of choosing the right database. I also use words that have two meanings but are spelled the same as examples: Indian (Native Americans or Indians), China (the country or the porcelain), and bass (the guitar, fish, and beer).

    Annie Smith, MLS
    Utah Valley University Library
    smithan@uvu.edu

    *****************************************************************************

    I think it is interesting to do a search that is misspelled and compare it to one that is spelled correctly. If one were to search "sexual harassment" (correct spelling) and compare the results with a search for "sexual harrassment (incorrect spelling). I find it interesting how many items come up when searching with the incorrect spelling. You can do this on an article database, such as Academic Search Complete, as well and often get a few results with the incorrect spelling. Students might assume that is all there is, since a few results (instead of zero) appear.

    Dr. John Jaeger
    Dallas Baptist University
    johnja@DBU.EDU

    *****************************************************************************

    One I can think of is the difference between www.whitehouse.gov and www.whitehouse.com. Be prepared for the difference. An explorer’s hint is to key in www2 at the beginning of headings and see where you get.

    Niema Jones
    thinker_i@hotmail.com

    *****************************************************************************

    These are not exactly bad examples, but some of my favorites:

    In history, I tell the students to look for terms that are relevant to the time they are interested in -- for instance, in different eras you will get different retrieval for shell shock, battle fatigue, and post-traumatic stress disorder. Also, that they should not look for anything under the heading World War I until World War II had started.

    In a more advanced search, we look for Martin Luther, and get a lot about him, about Dr. Martin Luther King, and about elementary education because of all of the schools named after Dr. King. This gives a chance to talk about the Boolean operator NOT and the law of unintended consequences.

    Phyllis Bratton
    pbratton@JC.EDU

    *****************************************************************************

    An example on the limits of Boolean.

    Several years ago, I was helping a student who was researching Black soldiers in the U.S. West. We did a keyword search using "black" and "soldiers" and "west" and among the items that came up was a collection of Rudyard Kipling stories which contained "Soldiers three," " East and West" and "In Black and White." Which is why I often refer to Boolean as a shotgun search, sometimes you hit the target but you can get a lot of scatter.

    Cindy Rosser
    Waco-McLennan County Library
    cynthiar@ci.waco.tx.us

    *****************************************************************************

    I use our online catalog to find examples of non-productive searches. Within our Admin mode, I can retrieve actual searches done by patrons (with no patron identification, of course). You can see false drops, bazillions of hits, and zero hits, as well as misspellings, the wrong first name with the last name, and synonyms that don't work.

    Gretchen Pearson
    Pearson@LEMOYNE.EDU

    *****************************************************************************

    I'm not sure if this is the type of thing you mean, but if you search Martin Luther King Jr on Google, one of the top 5 results is from an organization run by David Duke of the KKK. The site looks deceptively legitimate at a glance. In fact we had an interview in which the librarian had to do a library instruction class for the library staff as if we were students, she chose MLK as the subject and did a search on Google and apparently hadn't looked at it closely because she pulled it up as if it were a legit site.

    And if you search for just about anybody famous by name or any topic by one word (abortion, genocide, etc.), which most undergrads want to do, in any database you will get far too many hits the majority of which won't be helpful without adding additional search criteria.

    Susan R. Barber
    Christopher Newport University
    sbarber@cnu.edu

    *****************************************************************************

    Subject: [LAW-LIB:59297] In search of bad online searches
    Date: Thu, 18 Jun 2009 17:31:19 -0400
    From: Kent Olson <kolson@virginia.edu>

    Yesterday I was reviewing a law student's Westlaw research. She was looking for cases and articles about conservators' power of attorney in New York. When her initial search: "power of attorney" "conservator" turned up a few hundred thousand documents, she modified it: "power of attorney" "conservator" "new york" and instead got almost three million documents.

    This particular instance is an example of the Googlization of search -- adding more words narrows a Google search but does the opposite in Westlaw. It's probably a familiar problem for many of us. But it got me thinking about using really unsuccessful searches as a teaching tool. We show our students how to construct good Lexis and Westlaw searches, but it could also be a good classroom exercise or assignment to give them a series of bad searches and ask "What's wrong with these searches? What would you do to fix them?"

    Legal writing instructors sometimes hand out lists of error-riddled citations for students to correct, but I don't recall seeing anything similar with bad online searches. And so I'm soliciting your help. If you have a record of any memorable searches, please send them my way. I'll compile a "Kids Write the Darndest Searches" list (without identifying contributors, unless you ask for attribution) and share the results with the group.

    Kent Olson
    University of Virginia Law Library
    kolson@virginia.edu

    *****************************************************************************

    Kent

    I work in the law firm world. My comments are based on experiences in several AM law firms.

    I take the time to comment because the situation you describe is emblematic of the frustration we all feel when working with summer and incoming associates in the library.

    It is not a personal comment directed toward you or your colleagues.

    The problem is that that your student was using a keyword approach for a legal concept. The fact that she added the keyword NY to the search probably means she was in the wrong database and she probably hadn't though much about jurisdiction before she started either. At least she could have used the segment for NY.

    Law students are coming to us without a good handle on generalized sources in which to begin research nor an overall sense of jurisdiction and controlling authorities.

    Jeff Cohan
    Jcohan@fchs.com

    *****************************************************************************

    I concur with Jeff's comments. It is not enough to simply think about query construction. I bang into my student's heads the concepts of database content and selection until they're sick of me. And, as Jeff notes, that database selection is part of the jurisdiction/controlling authority decision that is fundamental to doing competent legal research.

    I am wondering what database(s) Kent's student was working in. I just replicated the 2 searches in the ALLCASES database and adding "new york" narrowed the results significantly, which is what I expected.

    Matt Morrison
    mmm72@cornell.edu

    *****************************************************************************

    And what about Boolean operators...that is another thing students have no concept about, and of course Google has not helped in this area. If the student had used the proper connectors the search would have been more successful.

    Dawn Urquhart
    Librarian, Aird & Berlis LLP
    durquhart@airdberlis.com

    *****************************************************************************

    Of course, the main problem with the search is, if entered as Kent described it, it or'd all the phrases. Of course each addition is going to keep increasing the results. It's as Kent said. The same technique decreases the number of hits on Google.

    Students today are coming to us with the Google search strategy in hand. The idea of using an or, and, pre or +, w/ etc. is foreign to them.

    What I had never thought about until I saw Kent's message is that running the same search in Google versus Westlaw is that the consequences are exactly opposite each other. If we take this one step further, think about the results you get if you use the same search in LexisNexis. I would guess no results since what is the likelihood of this phrase? >"power of attorney" "conservator" New York.

    Mariann Storck
    Legal Information Specialist/Library Manager
    U.S. Attorneys Office
    1225 17th St, Ste 700
    Denver, CO 80202
    mariann.storck@usdoj.gov

    *****************************************************************************

    I agree with Mariann, and I work in a law firm library. Kent's point was that the incoming law students and lawyers are all members of the Google generation and apply that searching technique to other sources in a "one size fits all" way. We experienced legal researchers know it doesn't work that way, particularly with Westlaw or Lexis searches. Those who teach students and attorneys need to keep it in mind and I thank him for pointing it out. I also look forward to some examples of bad searches that can be used in a teaching context.

    Lauri R. Flynn
    Gunderson Dettmer LLP
    1200 Seaport Blvd.
    Redwood City, CA 94063
    lflynn@gunder.com

    *****************************************************************************

    Although I don't normally post to the list, I wasn't sure if anyone had seen this recent survey of UK studies on legal research behavior (http://legalinformatics.wordpress.com/tag/calr/). Some really interesting findings, and it certainly highlights the challenges that so many law librarians, other educators, agencies, and firms face with research training issues for this current crow of law students.

    Jason E. Wilson
    Vice President
    Jones McClure Publishing
    jwilson@jonesmcclure.com

    *****************************************************************************

    This is a really intriguing thread. Thanks for posting the question, Kent.

    I’ve taught at law firms and in a law school and have found that students seem to be genuinely interested in understanding Boolean logic. My course is always a work in progress, but lately I've started by explaining three manifestations of Boolean – full, implied and template. Next we explore how Google, Lexis and Westlaw translate those three types of Boolean into their systems. I think the comparison helps them think about the impact a few thoughtful choices can have on their search results. The biggest take-away for them (I hope!) is to use Google’s Advanced Search as their default in order to make the use of the powerful template Boolean a bit easier.

    Ellen Callinan
    ellen.callinan@gmail.com

    *****************************************************************************

    I must chime in on this topic. It's not just the students...

    This is my 10th year of teaching internet research full time to lawyers and they have NO idea of what Boolean and proximity connectors are used for.

    We start each 6 hour CLE seminar with those concepts and the lawyers who think they know how to use the Internet (that is "Googling" to most of them) are simply amazed (and quite grateful to learn these basic concepts).

    Carole Levitt J.D., M.L.S.
    President, Internet For Lawyers
    clevitt@netforlawyers.com

    *****************************************************************************

    I have an article from a very long time ago that speaks to the issue of student searching that may be of interest to some on this list.

     These articles may also be of interest:

     

     David P. Dillard
    jwne@TEMPLE.EDU

    *****************************************************************************

    Interesting article sent to me by one of our library staff members:
    http://tech.yahoo.com/news/ap/20090512/ap_on_hi_te/eu_ireland_wikipedia_hoaxer

    Dr. John Jaeger
    Doctoral Research and Reference Librarian
    Dallas Baptist University
    johnja@DBU.EDU

    *****************************************************************************

    Interesting (and not entirely surprising) that Wikipedia came out smelling the best in this test.

    Louise E. Alcorn
    West Des Moines Public Library
    louise.alcorn@wdm-ia.com

    *****************************************************************************

    Date: Fri, 19 Jun 2009 19:51:03 -0700
    From: LIBREFED <librefed@KENT.EDU>
    Subject: REFQ: "Fake History" question

    I have a patron who is looking for examples of "fake history". He wants examples where an someone claimed a historical fact, and later it was proved that the "fact" was based on falsified evidence. These examples should be unambiguous, clear cases of fraud - not simply differences of opinion - and not inteneded as a joke, parody, satire or fiction. He is particularly looking for examples from the 20th or 21st centuries, but 19th century might be ok.

    I've been going round and round all day on this, and I can't think straight any more. Does anyone out there have any ideas on this? Any help would be appreciated.

    Mary Buelow
    Hedberg Public Library
    Janesville, WI 53545
    MBuelow@Hedbergpubliclibrary.org

    *****************************************************************************

    Dear Mary Buelow:

    The case of the crystal skulls is an interesting one. When I worked at the US Geological Survey Library we would occasionally get questions about the skulls found among Mayan ruins, etc. Actually, they were 19th century fakes. See: http://www.si.edu/opa/insideresearch/articles/V9_CrystalSkulls.html

    Speaking of which, there are plenty of examples in the art world.

    Another case is the Russian insistence that the Allies open up a "Second Front" during World War II to relieve German pressure on them. However, the Russians were not at war with either Japan or Italy. At that time, when the Russians insisted on a second front, the US was fighting on seven fronts, and the British on nine fronts. However, even today people still speak of the "Second Front" in WWII.

    The Kensington Runestone is another artifact that is dismissed as a hoax. See: http://en.wikipedia.org/wiki/Kensington_Runestone

    R. Lee Hadden
    Geospatial Information Library (Map Library)
    U. S. Army Geospatial Center
    Robert.L.Hadden@usace.army.mil

    *****************************************************************************

    The bust of Nefertiti was recently found to be a fake.

    Brenda Myc
    Dragonlocke87@AOL.COM


    -- 
    Bryan M. Carson, J.D., M.I.L.S., Ed.D.
    Special Assistant to the Dean for Grants & Projects/
    Coordinator of Reference & Instructional Services
    Western Kentucky University Libraries
    Author, "The Law of Libraries and Archives" (Scarecrow Press)
    
    1906 College Heights Blvd. #11067
    Bowling Green, Kentucky  42101-1067
    Phone: 270-745-5007; Fax: 270-745-2275
    bryan.carson@wku.edu | bryan.m.carson@gmail.com
    
    All original content copyright 2009 Bryan M. Carson



    This archive was generated by hypermail 2b29 : Thu Jul 23 2009 - 16:56:54 PDT