cosgrove.indd Se ing the Record Straight: How Online Database Providers Are Handling Plagiarism and Fabrication Issues John Cosgrove, Barbara Norelli, and Elizabeth Putnam As a result of several well-publicized media scandals involving plagiarism and outright fabrication, the authors question whether online database providers are appending or linking corrections to original stories. Unfortu- nately, in most cases, the answer is no. In this study of how some major database providers (EBSCO, LexisNexis, ProQuest, and Thomson/Gale) handle highly publicized cases of plagiarism and fabrication, only Lex- isNexis was found to append corrections, and even then inconsistently. Databases are not alone in this oversight; even the Web sites of the publications involved were unreliable in their treatment of corrections of their own articles. cademic librarians try in good faith to direct undergradu- ates to legitimate, accurate resources. When scandals surface that call into question the trust- worthiness of those resources, it is par- ticularly troubling. Not only are these “reliable” resources in fact unreliable, but librarians’ recommendations to use them are consequently compromised. A recent spate of scandals involving plagiarism and fabrication has rocked the New York Times and other periodicals. Although it is true that the newspaper and magazine publishers involved have made efforts to correct the record by publishing corrections, retractions, and apologies, researchers using tangible library resources (print and microform) may miss these emendations. Articles in bound periodicals and microform are separated from corrections that appear a week or a month or a year later by pages, volumes, and reels. Unlike print and microform, the virtual record can be updated easily through the use of linking. Consider the case of Slate writer Jay Forman’s 2001 article “MonkeyFishing,” which contained several dubious claims.1 Not only does Slate have a correction notice appended to Forman’s article, but the John Cosgrove is the Humanities and Access Services Librarian in the Lucy Scribner Library at Skidmore College; e-mail: jcosgrov@skidmore.edu. Barbara Norelli is the Social Sciences and Instructional Services Librarian in the Lucy Scribner Library at Skidmore College; e-mail: bnorelli@skidmore.edu. Elizabeth Putnam is the Social Sciences and Exercise Science Librarian in the Lucy Scribner Library at Skidmore College; e-mail: eputnam@skidmore.edu. 136 mailto:eputnam@skidmore.edu mailto:bnorelli@skidmore.edu mailto:jcosgrov@skidmore.edu whole saga of how Forman’s deceptions came to light, including editor Michael Kinsley’s initial defense of Forman, his later apology, and links to outside sources that broke the story about the deception, is readily available and linked from Forman’s article. As an exclusively online publication, Slate did not need to clean up a paper trail. However, it is important to note that it was relatively easy to append a correction to Forman’s article and link it to Kinsley’s precisely because Slate is online. At a time when many academic li- braries have access to more publications through online sources than they could comfortably house in their buildings, librarians need to know that they can rely on publishers and database pro- viders to deliver accurate, timely, and, if need be, corrected information. This article explores the current practices of several major database providers, includ- ing EBSCO, LexisNexis, ProQuest, and Thomson/Gale, regarding corrections, retractions, and errata and argues that they should be more proactive in their treatment of known plagiarism and fabrication by alerting users to these problems. Background How libraries and database providers should treat plagiarized or forged pub- lications has been a subject of debate for years. On more than one occasion, libraries have been encouraged to remove retracted printed materials from their shelves; some acquiesce, others resist.2 As more libraries acquire greater amounts of their collections from electronic sources, the option of compliance no longer rests in their hands but, rather, in the hands of database providers. Alarmingly, the literature reveals a great disparity not only in how database providers handle retractions, but also in opinions as to Se ing the Record Straight 137 whose responsibility it is to make readers aware of problematic materials. Medical publications and databases, most notably the National Library of Medicine’s (NLM) MEDLINE, have led the way in terms of clearly stated policy, procedure, and action regarding retractions. NLM’s retraction policy for MEDLINE (2002) explains that the data- base will index retractions for previously indexed articles, link the original article citation and its subsequent retraction, and add a “Retracted Publication” label to the retracted article’s “Publication Type” field.3 However, MEDLINE’s careful policy and diligent linking have not been rep- resentative of the database industry as a whole. Nancy Garman found that many databases, including Business Dateline, UMI’s Periodical Abstracts Plustext, and Magazine Index, all accessed by Garman through Dialog, do not link retraction notices to previously indexed articles.4,5 Marydee Ojala’s research in 1996 showed inconsistencies even in MEDLINE cover- age; although the database providers DIALOG and Ovid included a particular retraction notice in their coverage of MEDLINE, LexisNexis’s did not.6 Peter Hernon and Ellen Altman point out that alerts or links to retractions are far from common in most databases.7 Although these studies were con- ducted in the mid- to late-1990s and thus reflect database policies and procedures that may have been updated since that time, a more recent controversy revolved around the retraction procedures of Elsevier Science’s ScienceDirect database. Elsevier’s removal of a Human Immunol- ogy article, which was deemed “entirely inappropriate,” from the database in 2001 led to an outcry among librarians and other concerned parties.8,9 As a re- sult of the debacle, Elsevier revised its withdrawal policy, which now states that 138 College & Research Libraries Elsevier will not remove articles from its database but, instead, will link retraction notices to original, retracted articles and add a retraction watermark to the PDF article pages.10 Although plagiarism was not the reason for the withdrawal of the Human Immunology article, it does high- light the issue of concern for maintaining the historical record. Whose responsibility is it to ensure that retractions or corrections are clearly and easily available to readers? Some recom- mend that editors read and filter the mate- rial appearing in their publications more meticulously.11 Even reviewers are lauded for their plagiarism-detection abilities, as seen in several cases of plagiarism detec- tion by Library Journal and World Literature Today reviewers.12,13 Others recommend that database providers and online pub- lishers set explicit policies for handling retractions and also take advantage of linking capabilities to pull articles and their retractions together.14 It is still not a perfect system. As a group, medical librarians seem willing to accept responsibility for alert- ing their patrons to retracted materials. A 1998 survey by Carole Hughes indicated that 41 percent of academic medical li- braries identified and/or labeled retracted publications for their patrons. Hughes stated that “academic medical libraries have a responsibility to inform library users of information regarding retracted publications.”15 Hernon and Altman dealt quite explicitly with the important role librarians can play in mediating between patrons and problematic literature. They have warned librarians not to rely too heavily on database linkages and en- couraged them to take an active role in educating their users about misconduct in the literature.16 As shown in the Else- vier case, the demands of librarians and scholars can compel database providers to adjust their policies. March 2005 Previously published research shows that (1) the scientific, particularly medical, community is taking the lead in ensur- ing that retraction notices are available to readers, and (2) database coverage of, and linking among, retraction notices and retracted articles was spo y at best, par- ticularly in the 1990s. The authors of this paper ask, then, how is the nonscientific community dealing with corrections and how are databases treating corrections currently? Methodology The authors selected one low-profile (Lyne e Holloway) and several high-pro- file incidents of plagiarism, fabrication, or error in newspaper and magazine articles of different dates. Profiled in the appendix are six writers at four differ- ent publications, for whom the authors looked at a total of eight original articles and eight correcting articles. The articles were reviewed between July and Septem- ber 2003 and again between October 28 and November 6, 2003, to confirm that the status of the information had not changed. The selected incidents of misconduct are: Low Profile: Lyne e Holloway, New York Times (error); High Profile: Mike Barnicle, Boston Globe (plagiarism); Jayson Blair, New York Times (fabrication) 2 articles; Rick Bragg, New York Times (plagiarism); Ja- net Cook, Washington Post (fabrication); Stephen Glass, New Republic (fabrication) 2 articles. Although only a small number of writers and articles were selected for review, it was assumed that well-known incidents would be a reasonable test of how database providers are handling corrections. If providers are not process- ing the most highly publicized incidents of published misconduct, it is highly http:literature.16 http:together.14 http:meticulously.11 http:pages.10 TABLE 1 Number of Articles Analyzed by Publication Publication Original Articles Correcting Articles Total Boston Globe 1 1 2 New Republic 2 2 4 New York Times 4 41 8 Washington Post 1 1 2 Total 8 8 16 1. Both Blair articles were corrected by the same article. For simplicity’s sake the authors counted this as two articles. unlikely that they are doing so for lesser- known incidents. The databases utilized for this study were EBSCO’s EBSCOhost Newspaper Source & Academic Search Elite, Lexis- Nexis Academic News, ProQuest News- papers & Research Library, and Thom- son/Gale’s Infotrac Custom Newspapers (accessed through the New York State Library’s NOVEL project). The authors accessed the database records for the selected articles and corrections to check which databases: • index the article (“original article”) and the article that corrected it (“correct- ing article”); • append the correcting article to the original article (or to the record for the correcting article where only citations and abstracts are available); Se ing the Record Straight 139 • link the correcting article to the original article (in full-text databases). The authors also reviewed the Web sites of the newspapers/mag- azines that carried the original articles to see how the publica- tions themselves were handling circumstances where corrections or retractions were needed. In addition, the database provid- ers EBSCO, LexisNexis, ProQuest and Thomson/Gale were contacted by e-mail and asked to respond to the following questions: • Do you have a policy regarding the treatment of articles/documents that are identified as plagiarized, fraudulent, or factually incorrect within your data- base? • If so, what is your policy? • Whose responsibility is it for mak- ing readers/database users aware of problematic materials? Results Table 1 shows that for each of the eight original articles analyzed, eight correct- ing articles were available, or at the very least indexed, in the publications Boston Globe, New Republic, New York Times, and the Washington Post. Of the four database providers ex- amined, the authors found that only TABLE 2 Availability of the Original and Correcting Articles (in each Publication) by Database Provider Database Provider Boston Globe New Republic New York Times Washington Post EBSCO 0/2 4/4 8/81 0/2 Thomson/Gale 0/2 0/4 6/8 0/2 LEXIS-NEXIS 2/2 4/4 8/8 2/2 ProQuest 2/2 4/4 8/8 0/2 Total 4/8 12/16 30/32 2/8 1. Citations and abstracts, but no full text available. Total 12/16 6/16 16/16 14/16 48/64 140 College & Research Libraries LexisNexis had all sixteen origi- nal and correcting articles avail- able in their database and, most important, was the only provider to append some of the correcting articles. (See tables 2 and 3.) LexisNexis appended the four New York Times correcting articles as well as the one Washington Post correcting article, for a total of five of the eight articles analyzed. (See table 4.) LexisNexis did not append the correcting articles from the Boston Globe or New Republic. In contrast, although ProQuest was second in terms of article availability in the database with seven of the eight articles, it did not include any appended correcting articles. EBSCO included six of the eight analyzed articles but also did not append any correcting articles. Thomson/Gale had the least number of articles available, with three of the eight and no appended correcting articles. None of the full-text database providers linked the correcting article to the original article. (See table 5.) The responses from the database providers with regard to faulty articles (due to plagiarism, fabrication, or other forms of misconduct) residing on their databases almost uniformly pin the re- TABLE 4 Corrections by Publication in LEXIS-NEXIS1 Publication Original Articles Analyzed Corrections Appended Boston Globe 1 0 New Republic 2 0 New York Times 4 4 Washington Post 1 1 Total 8 5 1LEXIS-NEXIS was the only database provider that ap- pended corrections. March 2005 TABLE 3 Availability of the Articles (Total) by Database Provider Database Provider Original Article Available Correcting Article Available Total EBSCO 6/8 6/8 12/16 Thomson/Gale 3/8 3/8 6/16 LEXIS-NEXIS 8/8 8/8 16/16 ProQuest 7/8 7/8 14/16 Total 24/32 24/32 48/64 sponsibility on the publisher or licensor to handle the situation. With the permission of the database providers, their comments on this ma er are included here: EBSCO: “At EBSCO Publishing, we rely on our full-text publishing partners to initially identify problematic and/or plagiarized content, as EBSCO does not have editorial license to change or remove content at will. In a situation where a pub- lishing partner notifies EBSCO, we work with the partner to remedy the situation (be it complete removal of the full text and/or a note in the bibliographic record) to the satisfaction of the publishing part- ner.”—Bethany Close, Product Manager, Secondary Databases, e-mail message to author, February 5, 2004 LexisNexis: “Unfortunately, we have no way of knowing the veracity of an article. If directed by a publisher to re- move, append a correction, or replace a document, we will. We do not to my knowledge alert customers to this and I don’t believe a policy exists. We will take our direction from the licensor and look to them for indemnification.”—Tif- fany Guard, Product Manager, Lex- isNexis Academic, e-mail message to author, January 15, 2004 ProQuest: “Our policy for the Pro- Quest databases is to follow the instruc- tion of the publisher in regard to the treatment of articles that are identi- fied as plagiarized, fraudulent, or factually incorrect.”—L. Suzanne BeDell, Vice President, Publisher, e-mail message to author, Decem- ber 19, 2003 T h o m s o n / G a l e : “ T h o m - s o n G a l e d o e s n o t s u p p o r t p l a g i a r i s m a n d d o e s n o t have control over this unethical be- havior. However, we strive to edu- cate our customers, students, and teachers on plagiarism, intellectual property, fair use, copyright, and the consequences of plagiarism.”—Kimberly Wilkes, Inside Academic Representative, e-mail message to author, March 4, 2004 Discussion It is apparent even from this small sample of articles that most of the major online database providers are not appending corrections to articles later exposed as containing plagiarism, fabrication, or er- rors. Of the resources analyzed (EBSCO- host, LexisNexis Academic, ProQuest, and Thomson/Gale Infotrac), only Lex- isNexis appended corrections, and then inconsistently. None of the database providers used links for corrections. Although it is true that the structure of a particular database (LexisNexis, for instance) may make static links more difficult to create than appending corrections, it is a shame that the most elemental characteristic of online resources—the ability to link—is so underutilized within the databases themselves. Criticism one might level at the data- base providers is undercut to some degree by the fact that some of the publications themselves are doing such a poor job with online corrections. It is unconscio- nable that the Boston Globe is selling Mike Barnicle’s article, “I Was Just Thinking,” (appendix, author 1) as it originally ap- Se ing the Record Straight 141 TABLE 5 Corrections Linked or Appended by Database Provider Database Provider Original Article and Correcting Article Linked Correction Appended EBSCO 0/6 0/6 Thomson/Gale 0/3 0/3 LEXIS-NEXIS 0/8 5/8 ProQuest 0/7 0/7 Total 0/24 5/24 peared, still not properly crediting George Carlin for the contents (especially because they fired Barnicle over the ma er) and giving the buyer of the plagiarized article no warning about the problems. Similarly, the Washington Post ought to warn buyers before they purchase Janet Cooke’s article (appendix, author 4) about the scandal that caused her to lose her Pulitzer prize and not simply put a correction at the top of the already purchased article.17 The database providers might reasonably say: “If the publications don’t care, why should we?” There are other circumstances where corrections or links might be appropriate. Here are three examples, which range from reasonably similar to the cases pro- filed in this study to, admi edly, cases that are very difficult to correct (even in electronic databases): • On January 4, 2002, an article by Fred Barnes, “Stephen Ambrose, Copy- cat,” appeared on the Weekly Standard’s Web site in anticipation of its publication in the January 14th print issue. Barnes pointed out striking similarities between Ambrose’s latest book, The Wild Blue: Men and Boys Who Flew the B-24s over Ger- many (2001), and a 1995 book by Thomas Childers, Wings of Morning: The Story of the Last American Bomber Shot Down over Germany in World War II. The New York http:article.17 142 College & Research Libraries Times reported Barnes’s story on January 5th (David D. Kirkpatrick, “2 Say Stephen Ambrose, Popular Historian, Copied Passages”). The very next day, Ambrose apologized, admi ing that he had taken lines from Childer’s book without proper notation but said it had been inadvertent. Subsequently, Ambrose was accused of not properly crediting sources in several of his books. None of the databases linked or ap- pended Kirkpatrick’s article to Tom Fer- rell’s September 9, 2001, New York Times book review of Ambrose’s book. In this case, the “correction” is not a reflection on the book reviewer but, rather, on the fact that an article appeared in the same publication (New York Times) correcting the subject of the review (Ambrose’s book). • Brian Walski was fired when the Los Angeles Times learned that a Walski photograph that appeared on the front page on March 31, 2003, was actually a composite of two photos. On April 2, 2003, the Times ran an editor’s note along with the doctored photo and the two original photos. None of the databases included the Walski image (or any of the other images associated with the front page article), although two (Lexis-Nexis Academic and ProQuest Newspapers) included captions for it with no mention of the scandal. In this case, the photo was not reproduced in the databases, so one might say “no harm, no foul.” However, there also was no correction for the cap- tion indicating that the photo it referred to had been doctored. • On April 11, 2003, CNN Chief News Executive Eason Jordan wrote a New York Times op/ed piece, “The News We Kept to Ourselves,” in which he disclosed that for years CNN had not reported atroci- ties commi ed by the Iraqi government because of fears of reprisals. March 2005 In this case, corrections are moot. Be- cause this is a ma er of omission and not commission, there is nothing to “correct.” How many transcripts would need to be “corrected,” and what would the correc- tions say? How many years of transcripts should be linked to a copy of Jordan’s admission? The complexity of these cases (espe- cially CNN’s) need not be a deterrent to database providers trying to provide some level of editorial control, however. In budget-constrained times, database providers should consider value-added editorial services to differentiate them- selves from their competitors. Resources such as CQ Researcher that are centered on editorially driven topics shine in their treatment of contemporary issues raised by these three cases. If one of the major database providers with contracts to provide full-text partnered with CQ, the authors suspect that many libraries would be interested in the results. As it stands now, Lexis-Nexis Academic clearly has an edge over the other data- bases considered here simply because it is currently be er (but certainly not perfect) at connecting corrections to the articles it corrects. Conclusion Finding reliable materials using on- line databases is difficult enough for students, especially undergraduates, without having to navigate easily fixed pitfalls. The articles in this study are those most obviously in need of a cor- rection or a link to a correction—articles identified by the publications themselves as being flawed by error, plagiarism, or fabrication. Academic librarians in- struct students to carefully evaluate the literature in their campuses’ database resources. Unfortunately, it is not practi- cal to expect undergraduate students to routinely search at the level necessary to Se ing the Record Straight 143 uncover corrections and retractions nor do librarians commonly have the time to teach those skills. It also is unlikely that a citation index, a la Shepard’s, will be developed for news or popular press resources, however valuable it might be. The obvious solution, then, is for database providers to take responsibility for connecting articles and corrections. Where flawed articles and the articles that correct them are both available in a database, at minimum, we should expect the corrections or citations for the cor- rections to be appended to the erroneous articles or for the original and correcting articles to be linked. The authors urge publishers and database providers to routinely label and link corrections. As shown in the Elsevier case, the demands of librar- ians and scholars can compel database producers to adjust their policies. The authors encourage librarians to speak to their database representatives and voice their concerns. The authors have already spoken to theirs. Notes 1. Jay Forman, “MonkeyFishing: Cruel and Unusual? Or Good Sporting Fun?” Slate (June 8, 2001). Available online from h p://slate.msn.com/id/109707/. Correcting article by Michael Kinsley, “MonkeyFishing: Slate Apologizes,” Slate (June 25, 2001). Available online from h p://slate.msn. com/id/110932/. 2. See Michael Rogers, Norman Oder, and Andrew Albanese, “Columbia Urges Libs. to Pull Book,” Library Journal 125 (Jul. 1, 2000): 16–20, which describes an incident in which a Columbia University dean requested that libraries remove a book containing “improper citations”; and T. Sco Plutchak, “Sands Shi ing beneath Our Feet,” Journal of the Medical Library Association 90 (Apr. 2002): 161–63, which discusses a le er from Elsevier Science to subscribers of Human Immunology requesting that they remove pages containing an “entirely inappropriate” article [re- trieved 19 November 2003]. Available online from h p://www.pubmedcentral.gov/articlerender. fcgi?artid=100760. 3. “Fact Sheet: Errata, Retraction, Duplicate Publication and Comment Policy for MEDLINE,” National Library of Medicine Web site (Mar. 6, 2002) [retrieved 27 October 2003]. Available online from h p://www.nlm.nih.gov/pubs/factsheets/errata.html. 4. Nancy Garman, “Errors, Corrections, Retractions,” Online 22 (Sept./Oct. 1998): 6–7 [retrieved 27 October 2003 from ProQuest ABI/INFORM Global database]. 5. The authors have not performed a subsequent check of these databases to see if they cur- rently provide retraction links. 6. Marydee Ojala, “Oops! Retractions, Corrections, and Amplifications in Online Environ- ments,” Searcher 4 (Jan. 1996): 30 [retrieved 4 November 2003 from Thompson/Gale Business & Company Resource Center database]. 7. Peter Hernon and Ellen Altman, “Misconduct: Infecting the Literature, But Do We Really Care?” Journal of Academic Librarianship 25 (Sept. 1999): 402 [retrieved 31 July 2003 from EBSCO- Host Academic Search Elite database]. 8. The liblicense-l listserv served as a forum for much of the discussion; its archives are avail- able at h p://www.library.yale.edu/~llicense/ListArchives/. See threads “When can an article be ‘withdrawn’?” “Removal of articles from ScienceDirect,” “Vanishing Act,” “Elsevier’s Vanishing Act,” “Vanishing Act—continued,” “Questions: RE: Vanishing Act—continued,” “Elsevier’s Van- ishing Act—The Real Issue,” “Elsevier revised policy on article removal,” and “Elsevier policy on article removal” between May 2002 and February 2003. 9. See Andrea L. Foster, “Elsevier’s Vanishing Act,” Chronicle of Higher Education 49 (Jan. 10, 2003): A27 [retrieved 20 Oct. 2003 from EBSCOHost Academic Search Elite database]; and ———, “Elsevier Announces New Procedures for Retracting Online Articles,” Chronicle of Higher Education 49 (Feb. 28, 2003): A35 [retrieved 13 Oct. 2003 from EBSCOHost Academic Search Elite database]. 10. “Elsevier Policy on Article Withdrawal,” Elsevier Web site [retrieved 20 October 2003]. Available online from h p://www.elsevier.com/homepage/about/ita/editors/newsitems/with- drawl_policy.htm. The link has subsequently moved to h p://www.elsevier.com/wps/find/li- brariansinfo.librarians/libr_policies#removal. 11. Marcel C. LaFolle e, Stealing into Print: Fraud, Plagiarism, and Misconduct in Scientific Pub- 144 College & Research Libraries March 2005 lishing (Berkeley: Univ. of California, 1992), 193–94. 12. Francine Fialkoff, “There’s No Excuse for Plagiarism,” Library Journal 118 (Oct. 15, 1993): 56 [retrieved 11 June 2003 from EBSCOHost Academic Search Elite database]. 13. William Riggan, “Plagiarism and Reviewer/Editor Responsibility,” Journal of Information Ethics 6 (spring 1997): 34–38. 14. Ojala, “Oops! Retractions, Corrections, and Amplifications in Online Environments,” under “Further and Final Thoughts.” 15. Carole Hughes, “Academic Medical Libraries’ Policies and Procedures for Notifying Library Users of Retracted Scientific Publications,” Medical Reference Services Quarterly 17 (summer 1998): 37. 16. Ellen Altman and Peter Hernon, Research Misconduct: Issues, Implications and Strategies (Greenwich, Conn./London: Ablex, 1997); ———, “Misconduct.” 17. In a less-publicized case, for months a er a wire story reporting the dismissal of Gregory M. Jones by the Roswell Daily Record on June 17, 2003, for inserting quotes from the comedy Caddyshack into a golf article without citing them as such, the article was available on the paper ’s Web site without any correction. At least, they did not charge the reader for the privilege! Se ing the Record Straight 145 APPENDIX Author 1: Mike Barnicle, Boston Globe, Summer 1998—Plagiarism Barnicle was the Boston Globe’s leading columnist (of 25 years) when scandal broke in June/August 1998. Reacting to allegations of fabricated persons and quotes in Barnicle’s articles, in June of 1998, the Globe did an extensive investigation of his stories and found no problems. Then in August, Barnicle was found to have plagiarized from a George Carlin book, Brain Droppings, in his August 2nd column. Article: “I Was Just Thinking,” Boston Globe (August 2, 1998). Correcting Article: Mark Jurkowitz, “Globe Asks Barnicle for His Resignation; Cites Misrepresentations by Columnist,” Boston Globe (August 6, 1998). • EBSCOhost: Newspaper Source: Indexing not available for that date. • Lexis-Nexis Academic: News: Article and correcting article available; no correction appended or link to correcting article (as of October 28, 2003). • ProQuest: Newspapers: Article and correcting article available; no correction appended or link to correcting article (as of October 28, 2003). • Thomson/Gale Infotrac: Custom Newspapers: Publication not indexed. • B o s t o n G l o b e We b s i t e : A r t i c l e a n d c o r r e c t i n g a r t i c l e a va i l a b l e . On Thursday, October 23, 2003, the authors purchased the article from the Globe’s online archive. There was no correction and Carlin is not credited in any way, nor is there any mention of the fact that Barnicle was dismissed, in part, because of the article. Author 2: Jayson Blair, New York Times, October 2002–April 2003 (and earlier)—Fabrication Blair was a New York Times reporter who fabricated, misrepresented, and plagiarized multiple times over the course of his five-year career before resigning on May 1, 2003. A New York Times investigation found that Blair fabricated elements in 38 out of 73 stories in a seven-month period and led to the May 11, 2003, article detailing Blair’s erroneous work. Article: “A NATION AT WAR: THE SPOUSES: A Couple Separated by War While United in Their Fears,” New York Times (April 15, 2003) Correcting Article: Dan Barry, David Barstow, Jonathan D. Glater, Adam Liptak, Jacques Steinberg. Research support by Alain Delaqueriere and Carolyn Wilder, “Cor- recting the Record: Times Reporter Who Resigned Leaves Long Trail of Deception,” New York Times (May 11, 2003) • EBSCOhost: Newspaper Source (Citation & abstract only): Article (under slightly different title does not include “A Nation at War: The Spouses”) and correcting article (under slightly different title does not include “Correcting the Record”) both cited; no correction appended or link to citation for correcting article (as of October 28, 2003). • Lexis-Nexis Academic: News: Article and correcting article available; appends a correction to this article (refers to the May 11, 2003, article); correction appended May 11, 2003. • ProQuest: Newspapers: Article (under slightly different title does not include “A Nation at War: The Spouses”) and correcting article (under slightly different title does not include “Correcting the Record”) available; no correction appended or link to correcting article (as of October 28, 2003). • Thomson/Gale Infotrac: Custom Newspapers: Article and correcting article 146 College & Research Libraries March 2005 (under slightly different title does not include “Correcting the Record”) available; no correction appended or link to citation for correcting article (as of October 28, 2003). • New York Times Web site: Article and correcting article available; appends a cor- rection to this article. Article: “A NATION AT WAR: VETERANS; In Military Wards, Questions and Fears from the Wounded,” New York Times (April 19, 2003). Correcting Article: Same as above. • EBSCOhost: Newspaper Source (Citation & abstract only): Article (under different title: “Pains of Wounded Torment the Psyche As Well As the Body”) and correcting article (under slightly different title does not include “Correcting the Record”) both cited; no correction appended or link to citation for correcting article (as of October 28, 2003). • Lexis-Nexis Academic: News: Article and correcting article available; appends a cor- rection to this article; refers to the May 11, 2003 article; correction appended May 11, 2003. • ProQuest: Newspapers: Article (title does not include “A Nation at War”) and correcting article (under slightly different title does not include “Correcting the Record”) available; no correction appended or link to correcting article (as of October 28, 2003). • Thomson/Gale Infotrac: Custom Newspapers: Article (under slightly different title does not include “A Nation at War”) and correcting article (under slightly differ- ent title does not include “Correcting the Record”) available; no correction appended or link to correcting article (as of October 28, 2003). • New York Times Web site: Article and correcting article available; appends a cor- rection to this article. Author 3: Rick Bragg, NewYork Times, June 2002–May 2003—Plagiarism Pulitzer Prize-winning journalist Bragg resigned from the New York Times on May 28, 2003 a er he failed to a ribute parts of his June 15, 2002, oyster fisherman article to work by an unpaid intern, J. Wes Yoder. Article: “An Oyster and a Way of Life, Both at Risk,” New York Times (June 15, 2002) Correcting Article: Editors’ Note (Metropolitan Desk) New York Times (May 23, 2003) • EBSCOhost: Newspaper Source (Citation & abstract only): Article cited; could not find a citation for the correcting article; no correction appended or link to correct- ing article (as of October 28, 2003). • Lexis-Nexis Academic: News: Article and correcting article available; appends a correction to this article; correction appended May 23, 2003. • ProQuest: Newspapers: Article and correcting article available; no correction appended or link to correcting article (as of October 28, 2003). • Thomson/Gale Infotrac: Custom Newspapers: Indexing not available for that date. • New York Times Web site: Article and correcting article available; appends a cor- rection to this article. Author 4: Janet Cooke, Washington Post, April 1980—Fabrication Cooke’s Pulitzer Prize-winning story about an 8-year-old heroine addict in Washington, D.C., was fabricated. The character was a composite, quotes and events made up. It turned out that Cooke also lied to the Post about her educational credentials. Cooke was asked to resign and did so in April of 1981. Article: “Jimmy’s World: 8-Year-Old Heroine Addict Lives for Fix.” Washington Post (September 28, 1980) Se ing the Record Straight 147 Correcting Article: “The Players: It Wasn’t a Game; The Reporter: When She Smiled, She Dazzled; When She Crashed … ; The Story: First the Idea, and Finally the Presses Rolled; The Publication: ‘Jimmy’ Hit Washington Like a Grenade, and Bounced; The Doubts: From the Very First Moment, Some Suspected the Worst; The Ombudsman: A er the Agony, the Reappraisal; The Prize: Of Fiefdoms and Their Knights and Ladies of Adventure; The Confession: At the End, There Were the Questions, Then the Tears; The Pressures: Heat and the Achievers Both Have a Tendency to Rise; The Conclusions: Once Again, a Fail-Safe System Proves the Exception.” Bill Green, Washington Post (April 19, 1981) • EBSCOhost: Newspaper Source: Indexing not available for that date. • Lexis-Nexis Academic: News: Article and correcting article available; appends a correction to this article; correction appended on January 17, 2001. • ProQuest: Newspapers: Indexing not available for that date. • Thomson/Gale Infotrac: Custom Newspapers: Publication not indexed. • Washington Post Web site: Article and correcting article available. On Thursday, October 30, 2003, the authors purchased the article from the Post’s online archive. There was no note about the scandal on the citation. The user discovers that the article is flawed only a er purchasing it. Author 5: Stephen Glass, New Republic, 1995–1998—Fabrication Glass fabricated parts of twenty-seven New Republic articles. He was fired in 1998 a er an internal investigation prompted by reporters at the Web site of Forbes magazine, Forbes Digital Tool. Glass wrote a fictional novel based on his own story, The Fabulist, for which Simon & Schuster paid him a $100,000+ advance. Article: “Peddling Poppy,” New Republic (June 9, 1997) Correcting Article: “To Our Readers,” New Republic (June 29, 1998) • EBSCOhost: Academic Search Elite: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • Lexis-Nexis Academic: News: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • ProQuest: Research Library: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • Thomson/Gale Infotracf: Indexing not available. • New Republic Web site: Appears that the correction is available, but Glass’s article is not. The authors did not purchase subscription. Article: “Hack Heaven,” New Republic (May 18, 1998) Correcting Article: “To Our Readers,” The New Republic (June 1, 1998) • EBSCOhost: Academic Search Elite: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • Lexis-Nexis Academic: News: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • ProQuest: Research Library: Article and correcting article both available; no correction appended or link to correcting article (as of November 6, 2003). • Thomson/Gale Infotrac: Indexing not available. • New Republic Web site: Appears that the correction is available, but Glass’s article is not. The authors did not purchase subscription. 148 College & Research Libraries March 2005 Author 6: Lynette Holloway, New York Times, July 2003—Error On July 7, 2003, the New York Times ran a story by Lyne e Holloway about Steven Got- tlieb, reporting that he had lost control of his company, TVT Records. A week later, the Times ran an extensive follow-up by Diana B. Henriques, as well as a correction notice in an Editor’s Note. These noted that the “main premises [of Holloway’s article]…were based on fundamental misunderstandings” and also corrected several factual errors. Article: “Force on Music Charts, and in Court, Faces Struggle,” New York Times (July 7, 2003) Correcting Articles: Diana B. Henriques, “Music Executive Is Still in Control of TVT Records,” New York Times (July 14, 2003) and “Editors’ Note,” New York Times (July 14, 2003). • EBSCOhost: Newspaper Source (Citation & abstract only): Article and correcting articles cited; no corrections appended or links to citations for correcting articles (as of October 28, 2003). • Lexis-Nexis Academic: News: Article and correcting article available; appends a correction to this article; correction appended July 14, 2003. • ProQuest: Newspapers: Article and correcting articles available; no corrections appended or links to correcting articles (as of October 28, 2003). • Thomson/Gale Infotrac: Custom Newspapers: Article and correcting articles avail- able; no corrections appended or links to correcting articles (as of October 28, 2003). • New York Times Web site: Did not have correction notice on Holloway’s article as of July 15, 2003 (when Lexis-Nexis had already appended the correction); correction is now posted with abstract (article for sale). Publication: 1: 2: 4: 1_2: Total: 8: 16: Database Provider: Total_2: EBSCO: 68: 88: ProQuest: 88_2: Total_3: 4864: