Retraction: Mistake or Misconduct?

THE ETHICIST: RESEARCH

Ethics in the Academy of Management

 

Columnist: Lorraine Eden, Visiting Professor (The Ohio State University) and Professor (Texas A&M University), leden@tamu.edu

Date: October 28, 2013

 

Subject:  RETRACTION: MISTAKE OR MISCONDUCT? 

 

KEY INSIGHT:  Seeing a journal article with the word “RETRACTION” written in diagonal watermark across the front page is probably a shock to most management and business scholars. Not only is the percentage of articles withdrawn from publication across all disciplines very small (the estimate is less than .02% per year), the number of retracted articles in business and management journals is small relative to those in, for example, biomedical journals. Recently, however, a number of our well-known journals including Journal of Management Studies, Organization Science and Strategic Management Journal – even the Journal of Business Ethics! — have posted retraction notices. Why are articles retracted? I discuss the various categories of article retraction, look at retraction in the context of business and management journals, provide examples of retraction categories,  and end with questions for discussion.

 

1. INTRODUCTION

 

Are you familiar with the names of these individuals: Joachim Boldt, Ulrich Lichtenthaler, Naoki Mori,  Diederik Stapel? Probably not – unless you work in the same research area as they do. They have been incredibly prolific scholars, with many more publications than most other researchers. Unfortunately, their publication records have been too good to be true and many of their publications have now been withdrawn – retracted – by the journals.  Because their number of retracted articles is so large, these individuals have been identified as “repeat offenders” and some (Stapel, for example) have even become household names (Bhattacharjee, 2013).

 

Where do article retractions appear? If you said “in bio-medical journals”, you would be right, but that is only partly correct. Retracted articles appear in journals from all disciplines across the board; and they can be found in our own management and business journals, including Strategic Management Journal, Journal of Management Studies, Organization  Science, Research Policy, and (even!) the Journal of  Business Ethics.

 

In this blog posting, I want to explore categories of article retractions and provide examples, look at retractions in business and management journals, and suggest a reading list for those interested in this topic for teaching and/or research purposes.

 

First, it is important to separate “corrections” and “expressions of concern” from retractions (ICMJE, 2013).  Corrections can occur in the form of typos or non-consequential mathematical mistakes; the normal process is to publish the correction and link it to the original article.

 

Expressions of concern are typically published by journals when they are not sure of the conduct or integrity of the work. For example, a “repeat offender” (an author who has had multiple papers retracted) may lead other journals to question his/her other publications and to publish an expression of concern as a warning to readers. For example, after Shigeaki Kato (a former endocrinology researcher at the University of Tokyo) had five articles retracted, an expression of concern was issued by Molecular and Cellular Biology about five other of his papers that were published in that journal (Retraction Watch, May 30, 2013).  An expression of concern can be used to indicate that the journal editors have started an investigation into a particular article. Since the time from initiation to completion of a retraction can take years, such a statement can be an “early warning signal” to readers that a particular article may be seriously flawed (Jasny, 2011).

 

Retraction, on the other hand, is the withdrawal of a previously published article from a journal.  This is “science’s ultimate post-publication punishment: retraction, the official declaration that a paper is so flawed that it must be withdrawn from the literature” (Van Noorden, 2011b: 26).

 

Retractions are very rare; most scholars who work in this area estimate retractions are only .02% of published papers, that is, 2 in 10,000 (Van Noorden, 2011b: 27).  A low percent, however, may represent the proverbial “tip of the iceberg” since Van Noorden notes that prior surveys suggest “1-2% of scientists admit to having fabricated, falsified or modified data or results at least once” (p.27).

 

Retractions are a huge amount of work for journal editors and publishers. Typically, the decision to retract an article is taken after extensive consultation among the journal editor(s), publisher and the author(s).  Because the issues are so sensitive and involve potential damage to author reputation, employment and income, investigations tend to take place in secret. The average time to retract a published article is estimated as two years, longer when a senior scholar is involved (Chen, Hu, Mllbank and Schultz, 2013: 239).  Normally, but not always, the journal publishes a formal retraction statement explaining the reason or reasons for withdrawal, and the article appears with a large “retracted” watermark across the front page or entire article (see the example on this page).

 

2.  WHY JOURNAL ARTICLES ARE RETRACTED – MISTAKE OR MISCONDUCT?  

 

The key reasons why a journal article is so flawed that it must be withdrawn from publication boil down to two: author misconduct or mistakes. A small percentage of retraction cases involve publisher errors, but by far the most common are author related. For example, of the 4,232 retracted publications over 1980-2010 identified in the PubMed and Web of Science databases, according to Grieneisen and Zhang (2012), only 9 % of the 3,631 papers where reasons were given for the retraction identified the cause as “publisher error”.

 

In Box 1, I categorize the main types of article retractions. My list is a compilation and interpretation of lists proposed and used by other authors. I separate author from publisher errors, and then separate the author category into three main types: research misconduct, distrust of data or interpretations, and publishing misconduct.

 
Box 1: Categories of Article Retraction 

 

1.       Author error

a. Research misconduct

i.  Data fraud (data falsification, fabrication or manipulation; intentionally biased research design, data used without permission)

ii. Inaccurate or misleading reporting of results

iii.  Other research misconduct (failure to obtain legally required oversight such as institutional board approval, ethical   problems with research)

b.  Distrust data or interpretations

i.   Honest error (incorrect data, calculation   errors)

ii.  Findings cannot be replicated

iii. Published data or interpretations no longer considered valid or reliable by some or all the authors (e.g. unexplained   data irreproducibility, experimental artifacts discovered post-publication)

c. Publishing misconduct

i.   Plagiarism from the works of others

ii.  Redundant publication (duplicate publications,   self-plagiarism, failure to disclose or acknowledge original publications)

iii. Authorship issues (failure to consult or   inform listed authors, excluding authors who contributed substantially to the work)

iv.  Vague copyright issues or legal concerns

2.       Publisher error

a.  Accidental duplicate publication

b.  Accidental publication of version without   final author corrections

c.  Published in wrong journal or wrong issue

3.       Other and Unspecified Reasons

Source: Author’s integration and   revision of retraction categories identified in Fang, Steen and Casadevall   (2013: Figure 1), Grieneisen and Zhang (2012: Table 2) and Wager and Williams   (2011: Table 1).

 

A key dispute among researchers who study retractions is what percentage of author errors are due to misconduct (deliberate intention to deceive) versus mistakes (inadvertent, unintended errors). There are at least two issues here: (1) imputing motives from actions, and (2) defining what is and is not a fraudulent action.

 

First, motives, of course, are difficult to identify from actions. Box 1 makes it clear how difficult it is to separate mistakes from misconduct because of our inability to distinguish action from motive.  In Box 1, I would probably identify “mistakes” as including 1.b.i and possibly other parts of 1.b; whereas “misconduct” would most likely include all of 1.a, parts of 1.b, and all of 1.c.  Others might put more of these actions into the mistake category.

 

As an example, suppose the reader or journal editor can see an author error (for example, duplicate paragraphs in two publications), but cannot tell what motivated the author’s action. Honest error or deliberate fraud? Most authors, when faced by an editor questioning them about the duplication, will argue they made an honest mistake and no intentional fraud was involved. Moreover, editors are also reluctant to attach motive to action – at least in print — fearing possible retribution such as being sued for defamation of character by the author. As a result, retraction statements tend to be “safe summaries” of “the facts” without much detail, in order to avoid implying anything about the author or authors’ motivations for their actions.

 

Second, there have been several attempts in the literature to separate mistakes from misconduct. Below, I review recent empirical work on this topic.

 

Steen (2011) hypothesized that inadvertent error papers should be randomly distributed throughout the literature; whereas deliberately fraudulent papers would be quite different: non-random, clustered, targeting particular journals, with larger numbers of co-author teams.  He separated the 788 English-language papers that had been retracted from the PubMed database between 2000 and 2010 into three categories: fraud (197), research error (545) and unknown (46).  Research fraud was defined so as to only include data fabrication or falsification; plagiarized and self-plagiarized papers, for example, were classified as research error. As a result, the research fraud category was defined quite narrowly.  Looking at the data, Steen found clear evidence of deliberate fraud, concluding that “papers retracted because of data fabrication or falsification represent a calculated effort to deceive” and that “such behavior is neither naïve, feckless nor inadvertent” (p. 113).  More than 50% of the fraudulent papers had a first author who had written other fraudulent papers; whereas less than 20% of the erroneous papers had a first author who was a repeat offender.

 

In Fang, Steen and Casadevall (2012), misconduct is defined so as to include fraud, suspected fraud, duplicate publication (self-plagiarism), and plagiarism. The “error” category, in this paper, is therefore much smaller and more likely closer to the definition of mistakes, when compared to Steen (2011).  The authors found that 67% of 2,047 retracted papers indexed in PubMed since 1973 were due to misconduct, while only 21% were due to error. Their percentages, however, are based on all retracted articles, including those where no reason was given for the retraction. In Table 1 below, I recalculate the % distribution using the smaller denominator of articles where reasons for retraction were provided. As Table 1 shows, almost one-quarter of retracted articles in Pub-Med were listed as author error compared to three-quarters as author misconduct.

Table 1: Retracted Articles in PubMed, 1977-2011 (Fang et al. 2012)
Articles by Type* Number Distribution (category as % of   articles with reasons)
All articles   2,047
Articles with provided reasons   1,865
  • Fraud (fabrication/falsification)
    697  37.4
  • Suspected fraud
    192  10.3
  • Plagiarism
    200  10.7
  • Duplicate publication
    290  15.5
  • Error
    437  23.4
  • Other
    108    5.8

No reason given

    182
*   Articles can be classified in more than one category.

Source: Calculations based on data in Fang, Steen and Casadevall (2012, Table 2).

 

The broadest statistical analysis of retracted articles to date, by Grieneisen and Zhang (2012), analyzes a database of 4,232 retracted articles in the PubMed and Web of Science (WoS) databases between 1928 and 2011. By including WoS journals, this paper branches out of the bio-medical literature to include the social and physical sciences. The addition of WoS articles is very useful because the authors can capture retractions across a much wider array of disciplines, and can also perform subgroup analysis comparing WoS with PubMed journals. The article also differentiates between author and publisher errors, and breaks the reasons for retraction into nine categories.  See Table 2 for a summary of their results.

Table 2: Retracted Articles in PubMed and Web of Science, 1980-2010
Articles by Type* Number Distribution (category as % of   articles with reasons)
All articles 4,232
Articles with provided reasons 3,631
  • Fraudulent/fabricated data
     602     16.6
  • Distrust data or interpretations
     915     25.2
  • Other research misconduct
     123      3.4
  • Plagiarism
     796     21.9
  • Duplicate publication
     562     15.5
  • Authorship issues
     271      7.5
  • Unspecified “copyright issues”
     44      1.2
  • Other publishing misconduct
     100      2.8
  • Publisher error
     328      9.0
No reason given      601
*   Articles can be classified in more than one category.
Source: Calculations based on data in Grieneisen and Zhang (2012, Figure 3).

 

It is interesting to compare Table 1 (based on Fang et al.’s data) and Table 2 (based on Grieneisen and Zhang’s data). While there are differences in the way the authors grouped the data (see the discussion posted on Retraction Watch) , perhaps the most important difference between the two studies is that Table 2 includes retractions in WoS journals and Table 1 does not.  The percentage of retractions for plagiarism is twice as high in the dataset that includes WoS journals (21.9% versus 10.7%; whereas the duplicate publication percentages are the same (15.5%). The percentage for fraudulent/fabricated data, on the other hand, is half as large (16.6% versus 37.4%).  This suggests that biomed journals may be more likely to be plagued by author research problems (data fraud); whereas other journals may see more author publication problems (plagiarism and duplicate publications).

 

The Grieneisen and Zhang (2012: 1) article also highlights the role played by repeat offenders – authors with more than one article retraction – noting that “15 individuals account for more than half of all retractions due to alleged research misconduct” . Some authors have so many retractions that they can completely skew total retraction numbers for a particular discipline, university and country (e.g., Joachim Boldt’s 88 retracted articles in anesthesiology; Adrian Maxim’s 48 retractions in electrical engineering); see Table 4 in Grieneisen and Zhang (2012).

 

In addition to the phenomenon of repeat offenders, some researchers have argued that the profession is shifting in ways that encourage greater probability of fraudulent behavior. Honig, Lampel, Siegel and Drnevich (2013), for example, argue that academia is tilting toward treating research as an entrepreneurial activity where authors are more likely to attempt to “game” the system. Honig et al. (2013) argue this is due to the pressures and rewards involved in today’s publish-or-perish environment, for example, where tenure and promotion depend on the number of top-tier publications one has in hand.  Steen (2011) provides supportive empirical evidence; finding that research fraudsters are more likely to target top-tier journals. With the publication-to-submission ratio in our top journals well below 10%, these authors suggest that scholars will be more likely to cut ethical corners in order to increase their chances of successful publication, especially in top-ranking journals.  The rise in self-plagiarism, “slicing and dicing  into the smallest publishable unit”, coercive citation, and manipulation of data and results are not surprising in this environment.  Similar points are made in, for example, Elliott, Marquis and Neal (2013). When coupled with large financial rewards and/or release time from teaching for high-publishing faculty, the incentives for research misconduct can be significant, as witnessed in several Asian universities (Ching, 2013; The Economist, 2013b). In a JIBS editorial, Eden (2010), and an earlier The Ethicist posting, I also raised these concerns; see Scientists Behaving Badly: Insights from the Fraud Triangle (July 27, 2011).

 

It is also important to point out that research misconduct carries with it a variety of direct and indirect costs. The old adage that “one bad apple spoils the bunch” exemplifies the worry that research misconduct taints and devalues all research, creating a “market for lemons” (Cottrell, 2013). Retractions are also frequently referred to as the “tip of the iceberg”, which promotes the view that research is tainted and, similar to littering, may encourage others to engage in misconduct on the grounds that “everyone does it”.

 

A recent article by Chen, Hu, Mllbank and Schultz (2013) focuses on the costs that retracted articles pose to other scholars and research in general by examining how retracted articles are cited in subsequently published research.  Their visuals make it abundantly clear that high-profile retracted articles that are tightly networked into a research area can cause enormous damage to the whole area.  Moreover, the damage done to co-authors and to PhD students writing dissertations built on fraudulent or fabricated data provided by their chair can be career threatening; see, for example, the Diederik Stapel case where data fraud occurred in at least 55 papers, many with co-authors, and 10 PhD dissertations under his supervision (Bhattacharjee, 2013; Flawed Science, 2012).

 

 

3.  RETRACTIONS IN BUSINESS AND MANAGEMENT JOURNALS

 

This brings me to our own organization – the Academy of Management – and the journals in which our members publish. What do we know about retraction rates in our business and management journals? The only study to date on this topic is Karabag and Berggren (2012). They used four databases (Business Source Premier, Emerald, Science Direct and JSTOR) to scan for retracted articles in economics and management, finding a total of 31 articles in management journals (see their Table 1) and an even lower number (6) in economics journals (see their Table 2).  Of the 31 management journal retractions, eight were publications involving Ulrich Lichtenthaler (who now has 12 retractions according to Retraction Watch).  Once repeat offenders are removed, the number of retracted articles falls considerably.

 

Karabag and Berggren are puzzled by the low number of retracted articles and provide some possible explanations.  A key reason they give is that the business and management journals do not have explicit code of ethics in place to handle either plagiarism or research dishonesty.  It is true that most of our business and management journals have been “late to the table” at adopting explicit “rules of the game”. Perhaps this is because the large publishers (e.g., Wiley, Elsevier) have set up ethics codes to which all of their journals are expected to subscribe.  The adoption of software to catch plagiarism and self-plagiarism, such as CrossCheck, is also fairly recent.

 

However, as a former Editor-in-Chief of the Journal of International Business Studies (JIBS), I was much distressed to discover that Karabag and Berggren were unaware that JIBS has had an explicit Code of Ethics for Authors, Editors and Reviewers in place since July 2007, and been a member of COPE, the Committee on Publication Ethics (www.cope.org) since 2007 also. My editors and I developed the JIBS Code of Ethics with the explicit goal of NOT having to make ex post journal retractions. Our argument was straightforward. By setting up clear ex ante rules of the game with formal dispute settlement procedures, we hoped to deter and catch research fraud BEFORE the papers were published in JIBS. Wide publicity through JIBS editorials, ethics workshops for doctoral students and junior faculty, and JIBS Paper Development Workshops were also used (and continue with the current editorial team) to disseminate best ethical practices. In terms of enforcing these norms and practices, my editorial team did have a number of difficult cases that covered many of the types of author error identified in Box 1 (primarily plagiarism, self-plagiarism and authorship issues). However, these were handled almost exclusively at the pre-publication stage.  Give me ex ante over ex post rules any day!

 

The AOM journals have also joined COPE and established guidelines for handling ethical dilemmas in AOM publications. New policies are in place at AMJ for dealing with plagiarism, using the software program CrossCheck (Colquitt 2012), and for identifying data overlap with other published and unpublished work (Colquitt 2013).  If Honig et al. (2013) and Steen (2011) are correct, top-tier journals such as AMJ are likely to be inundated with submissions that now carry a higher likelihood of having research fraud attached.  This suggests our journal editors and reviewers need to develop and practice better “trust, but verify” procedures in order to protect research integrity.

 

Reliance on software programs like CrossCheck alone, however, is unlikely to be enough. They do not help with non-publishing related forms of research misconduct such as data manipulation or fabrication.  Even with plagiarism and self-plagiarism cases, as one journal editor reminded me after reading this blog post, software is a “rather unsatisfactory first step in a longer process”.  It may help with the most egregious cases, but comes at a high cost and not only in terms of time needed for already busy editors.

 

I expect these ethical norms and policies in place at JIBS and AMJ will soon be followed by most business and management journals. My guess is the subsequent isomorphic behavior, however, will vary considerably. The response by some journals, as one reader of this post suggested to me, may be quite weak (for example, the journal’s publisher has joined COPE so the journal puts a COPE membership stamp on its home page, but does little else to disseminate or enforce ethical norms and practices). Others may take a strong stand with the journal introducing its own code of ethics, which is widely communicated to editors, authors and reviewers.

 

There will also be differences between norms and practices since it requires much more effort (and involves more risk) to actually enforce ethical norms than to create and publish them. Reviewers and readers must be prepared to be whistle blowers, identifying suspected cases of author error (both mistakes and misconduct). Journal editors and publishers must have dispute settlement procedures in place – and be willing to follow them through – even up to the point whereby the journal may actually have to identify a published article as belonging in the .02% of articles that have been retracted from publication. Indeed, at least one paper (Marusic, Katavic & Marusic, 2007) has attempted to categorize journals’ responsibility for addressing and enforcing research ethics by doing a SWOT (strength, weakness, opportunity, threat) analysis!  The 2012 report on the Stapel case, Flawed Science (2012), provides a good sense of the work involved.

 

 

4. EXAMPLES OF RETRACTION STATEMENTS

 

It may be helpful to understand retraction categories if I share some examples. The statements below are lightly paraphrased versions of the original published statements attached to a number of retracted articles. The original articles are referenced in brackets. I have paraphrased the statements in order to generalize them by taking out the author, journal and article specifics.  YVIP stands for “year, volume, issue, pages”. All of these retraction statements are examples of author error.

 

  1. Data Fraud (Trampe, Stapel & Siero, Journal of Consumer  Research, 2011)

It has come to our attention that TITLE by AUTHOR, which appeared in JOURNAL (YVIP), contained fraudulent data that had been manipulated and at times fabricated by the author. This has been determined by a joint investigation by the Universities of XXX. We are therefore informing our readers that this article has been retracted. We apologize for any problems that the publication of this article may have caused.

 

  1. Data Fabrication (Marx & Stapel, European Journal of Social Psychology,  2012)

The following article from JOURNAL (AUTHOR 1, AUTHOR 2, TITLE, JOURNAL, YVIP) has been retracted by agreement between AUTHOR 1, the journal Editor-in-Chief and the publisher. The retraction has been agreed following the results of an investigation into the work of AUTHOR 2. The Committee has determined that this article contained data that was fabricated by AUTHOR 2. His co-author, AUTHOR 1, was unaware of his actions, and not in any way involved.

  1. Statistical Errors (Lichtenthaler, Strategic  Management Journal, 2012)

The following article (AUTHOR, TITLE, JOURNAL, YVIP) has been retracted by agreement between the authors, editors and publisher. The article is retracted at the authors’ request due to material technical errors which have rendered many of the article’s conclusions incorrect. The first author takes responsibility for these statistical errors.

 

  1. Statistical Errors and Duplication (Lichtenthaler, Journal of Business Venturing,  2008)

This article (AUTHOR, TITLE, JOURNAL, YVIP) has been retracted at the request of the Editor-in-Chief and the author. The author contacted the Editor-in-Chief about statistical irregularities in this article in DATE. The Editor-in-Chief thoroughly investigated this article and other preceding papers from the same database. On this basis, the Editor-in-Chief made the decision to retract the paper. The grounds for retraction are an error in statistical analyses, an omitted variable bias, and a “new” measure that was not “new” because it was already used in AUTHOR, TITLE, JOURNAL ,YVIP. These errors undermined the review process and are too substantial for a corrigendum. Please our publisher’s policy on Article Withdrawal.

 

  1. Statistical Errors (Ernst, Lichtenthaler & Carsten,  Journal of Management Studies,  2011)

The following article from JOURNAL (AUTHOR 1, AUTHOR 2, AUTHOR 3, TITLE, JOURNAL, YVIP) has been retracted by agreement between the authors, the journal’s editors and the publisher. The article is retracted due to errors in the reported empirical results, which form part of the basis for the conclusions drawn by the authors in the study. While the second author did not collect the data, he takes the responsibility for these technical errors.

 

  1. Cannot Reproduce Results (Lee et al., Science, 2013)

As a result of additional experiments, we wish to retract our paper (AUTHOR, TITLE, JOURNAL, YVIP). Specifically, we have not been able to consistently reproduce the results shown in Figure X. We have also discovered critical errors in Figures Y and Z.  Although we recognize that some parts of this paper may remain valid, we note that key parts of the paper depend on the results of these figures. For these reasons, we retract the main conclusion of the paper.

  1. Self-Plagiarism (Salam, Journal of Business Ethics, 2009)

The Editors and publisher regret to report that the paper published by AUTHOR as TITLE in JOURNAL (YVIP) is nearly identical to that published earlier by SAME AUTHOR as TITLE in JOURNAL (YVIP). This is a serious violation of publication ethics which according to our Policy on Publishing Integrity warrants a retraction notice to be published in the journal and a ban from publishing in any of the journal’s publications for an initial period of x years.

  1. Authorship Errors (Lunsford, Analytical Letters, 2011)

We, the editor and publisher of JOURNAL, are retracting the following article (AUTHOR, ARTICLE, JOURNAL, YVIP). The author’s institution has conducted an investigation into the authorship of this article, and established that the claim of sole authorship is not justified. This constitutes a breach of warranties made by the author with respect to authorship. We note we received, peer-reviewed, accepted and published the article in good faith based on these warranties, and censure this action.  The retracted article will remain online to maintain the scholarly record, but it will be digitally watermarked on each page as RETRACTED.

 

  1. Duplication & Statistical Errors      (Lichtenthaler, Research Policy,  2010)

The article (AUTHOR, TITLE, JOURNAL, YVIP) has been retracted at the request of the Editors-in-Chief. After discussions with the author about concerns raised by readers concerning papers he published earlier in JOURNALS in YEARS, the Editors have decided that the current article should be retracted. There are two main grounds for this retraction. (1) The author failed to disclose (through specific citations, or through a mention in the Acknowledgements section, or in a covering letter to the Editor) the existence of other closely related papers by the same author. In the absence of this information, the referees and editors involved in handling the paper were misled as to the level of originality of the paper. If they had been aware of these parallel papers, they would almost certainly have concluded that each of the two papers in question did not represent a sufficiently substantial and original contribution to knowledge in its own right to merit publication in a leading journal.  (2) In this paper and other closely related papers, the author has been inconsistent in his treatment of the variables.  In particular, variables treated as important in one paper are disregarded in a parallel paper, and vice versa. …. This raises severe doubts as to the validity and robustness of the conclusions.  If the referees and editors involved in handling the paper had been aware of this (i.e. if their attention had been drawn to the other closely related papers and they had spotted this inconsistency), they would undoubtedly have rejected the paper on methodological grounds.

 

  1. Duplication & Statistical Errors (Lichtenthaler,  Ernst & Hogel, Organization Science, 2010)

This article (AUTHOR, TITLE, JOURNAL, YVIP) is being retracted after an assessment that the work violates our publication standards in two important respects. First, the citation to highly related prior work by the first two authors is quite incomplete. As a result, it was not possible to assess the novelty of the work. In addition, there is reason to believe that key results in the paper would not hold if variables included in this related work had been incorporated into the analysis.

  1. Plagiarism (Rosoi, Applied Economic Letters, 2012)

The following article has been retracted from publication in JOURNAL (AUTHOR 1, TITLE, JOURNAL, YVIP). This article substantially reproduced the content of the following paper (AUTHORS 2,3 & 4, TITLE, JOURNAL, YVIP). The Editors and publisher note that submission of a paper to JOURNAL will be taken to imply that it represents original work, not previously published, and that it is not being considered elsewhere for publication.

  1. Plagiarism (Geh, Journal of  Business Ethics,  2012)

The editors and publisher regret to report that the paper published by AUTHOR as TITLE in JOURNAL (YVIP) includes several passages (about x percent) that duplicated passages published earlier by AUTHOR B in TITLE (JOURNAL, YVIP).  This is a violation of publication ethics, which according to our Policy on Publishing Integrity warrants a retraction of the article and a notice to this effect to be published in the journal.

 

 

5. A SHORT BIBLIOGRAPHY ON JOURNAL RETRACTIONS

 

I attach below a short bibliography of the pieces I found most useful on journal retractions. The website Retraction Watch run by Adam Marcus and Ivan Oransky is also highly recommended (http://retractionwatch.wordpress.com). The process for finding retracted articles is particularly well described in Chen, Hu, Mllbank and Schultz (2013) and Grieneisen and Zhang (2013).

 

  • Bhattacharjee, Y. 2013. The mind  of a con man. New York Times,   April 26.
  •  Chen, C., Hu, Z., Mllbank, J., & Schultz, T. 2013. A visual analytic study of retracted articles in scientific literature. Journal of  the American Society for Information Science and Technology, 64(2): 234-253.
  •  Ching, N. 2013. Fame is fortune in Sino-science. Nautilus: Science Connected, 5. http://nautil.us/issue/5/fame/fame-is-fortune-in-sino_science
  •  Colquitt, J.A. 2012. From the editors: Plagiarism policies and      screening at AMJ. Academy of Management Journal, 55.4:  749-751.
  •  Colquitt, J.A. 2013. From the editors: Data overlap policies at AMJ. Academy of Management Journal, 56.2: 331-333.
  •  Corbyn, Z. 2012. Misconduct is the main cause of life-sciences retractions.  Nature, 490 (October): 21. http://www.nature.com/news/misconduct-is-the-main-cause-of-life-sciences-retractions-1.11507
  •  Cottrell, R.C. 2013. Scientific integrity and the market for lemons. Research Ethics 0(0): 1-12
  •  Eden, L. 2010. Letter from the Editor-in-Chief: Scientists behaving badly. Journal of International  Business Studies, 41: 561-566. http://www.palgrave-journals.com/jibs/journal/v41/n4/pdf/jibs20109a.pdf
  •  Eden, L. 2011. Scientists behaving badly: Insights from the fraud triangle. The Ethicist. July 27. http://ethicist.aom.org/2011/07/scientists-behaving-badly-insights-from-the-fraud-triangle/
  •  Elliott, T.L., Marquis,  L.M,      & Neal, C.S. 2013. Business ethics perspectives: Faculty plagiarism and fraud. Journal of Business Ethics, 112.1: 91-98.
  •  Fang, F.C., & Casadevall, A. 2011. Editorial: Retracted science  and the retraction index. Infection and Immunity,79(10): 3855-3859.
  •  Fang, F.C., Steen, R.G., & Casadevall, A. 2012. Misconduct  accounts for the majority of retracted scientific publications. PNAS, 109(42):  17028-17033.
  •  Flawed Science: The fraudulent research practices of social psychologist Diederik Stapel.  2012. Report of the Levelt Committee, Noort Committee and Drenth Committee, Universities of Tilburg, Amsterdam and Groningen.  November 28.  https://www.commissielevelt.nl/wp-content/uploads_per_blog/commissielevelt/2013/01/finalreportLevelt1.pdf.
  •  Grieneisen, M.L, & Zhang, M. 2012. A comprehensive study of retracted articles from the scholarly literature. PLoS ONE 7(10): e44118.  doi:10.1371/journal.pone.0044118.
  •  Honig, B., Lampel, J., Siegel, D., & Drnevich, P. 2013. Ethics in the production and dissemination of management research: Institutional failure or individual fallibility? Journal  of Management Studies.  doi: 10.1111/joms.12056.
  •  ICMJE (International Committee of Medical Journal Editors). 2013. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals:  Publishing and Editorial Issues Related to Publication in Medical Journals: Corrections, retractions and “expressions of concern”.       http://www.icmje.org/publishing_2corrections.html
  •  Jasny, B.R. 2011. The Science retraction experience. Science.  http://publicationethics.org/files/u661/Jasny_Science%20presentation_final.pdf
  •  Karabag, S.F., & Berggren, C. 2012. Retraction, dishonesty and plagiarism: Analysis of a crucial issue for academic publishing and the inadequate responses from leading journals in economics and management disciplines. Journal of Applied Economics and Business Research, 2(3): 172-183.
  •  Martin, B.R. 2013. Editorial: Whither research integrity? Plagiarism, self-plagiarism and coercive citation in an age of      research assessment. Research Policy, 42: 1005-1014.
  •  Marusic, A., Katavic, V., & Marusic, M. 2007. Role of editors and journals in detecting and preventing scientific misconduct: strengths,  weaknesses, opportunities and threats. Med  Law, 26: 545-566.
  •  National Academy of Sciences, National Academy of Engineering, and Institute of Medicine of the National Academies. 2009. On Being a Scientist, Third Edition. The National Academies Press. Washington, DC.  http://www.nap.edu/download.php?record_id=12192
  •  Retraction  Watch. 2013. Five Kato papers subject to an expression of concern, plus, a statute of limitations on correcting the literature? May 30.
  •  Steen, R.G. 2011. Retractions in the scientific literature: do authors deliberately commit research fraud? Journal of Medical Ethics, 37: 113-117.
  •  The Economist. 2013a. How science goes wrong: Problems with scientific research.  The Economist. Oct 19.
  •  The Economist. 2013b. Looks good on paper: Scientific research. The Economist. September 28.
  •  Van Noorden, R. 2011a. The reasons for retraction. Nature Newsblog. October 6. http://blogs.nature.com/news/2011/10/the_reasons_for_retraction.html
  •  Van Noorden, R. 2011b. The trouble with retractions.  Nature, 478 (October 6): 26-28.
  •  Wager, E., Barbour, V., Yentis, S., & Kleinert, S. 2009. Retractions: Guidance from the committee on publication ethics (COPE).  Available at:  http://publicationethics.org/files/u661/Retractions_COPE_gline_final_3_Sept_09__2_.pdf
  •  Wager, E., & Williams, P.  2011. Why and how do journals retract articles? An analysis of      Medline retractions 1988-2008. Journal of Medical Ethics, 37: 567-570.
  •  Williams, P., & Wager, E.  2013. Exploring why and how journal editors retract articles: Findings from a qualitative study. Science  and Engineering Ethics, 19:1-11.

 

 

6.  QUESTIONS FOR DISCUSSION

 

1.  Have you ever read two articles (either both published or one published and the other in the publication process) and realized there was substantial overlap that looked to you like plagiarism/self-plagiarism? If so, what did you do about it and what happened?

 2. As a reviewer or editor, have you ever been faced with a manuscript submission with what appears to be an example of research misconduct?  How did you handle it?

3. Why do you think the number of retracted articles in our business and management journals is (apparently) so low? Is an entrepreneurial culture for research partly to blame, as argued by Honig et al. (2013)? Is it the lack of ethics codes and plagiarism software, as argued by Karabag and Berggren (2012)?

4. Should journals have their own formal Code of Ethics or is membership in COPE sufficient to deter research misconduct?

 5. What are your views on plagiarism detection software like CrossCheck? Should it be used on all journal submissions? 

 6. If a case of research misconduct appears after publication, who should be responsible for deciding whether misconduct occurred?  Should it be up to the journal? What role should the home university(ies) of the author(s) play?  The journal publisher?

7. Should examples of research misconduct be treated differently from author error, in particular, should the term “retraction” be used for author error?

8.  How should journals treat article retractions? Should they provide detailed descriptions that explain why the retraction happened, or simply list the retraction?

 9.  Which of the retraction statements in section 3 do you find most/least helpful and why?

10. How should journals treat repeat offenders? Should they be banned from publishing in our journals for a fixed time period, and if so, how long?

 

 

7.  ACKNOWLEDGEMENTS

My interest in this topic was first stimulated by an earlier post by Kim Eggleton on The Ethicist in October 2011, “The increase of retractions”. Thank you also to John Cantwell, Kathy Lund Dean, Benson Honig, Anne Tsui and Erik van Raaij for their helpful comments on the first draft of this blog post. I add my usual caveat that the views expressed here are my own and my responsibility.

Finally, an apology for the period where there were no postings on THE ETHICIST. I am on sabbatical leave for the academic year 2013-2014 from Texas A&M University, and am fortunate to be visiting in the Department of Management and Human Resources at The Ohio State University (Go, Buckeyes!).  Because of the time involved in moving from Texas to Ohio, I fell behind in my commitments, including THE ETHICIST.  I am now back on schedule, and welcome your comments as Kathy Lund Dean, Paul Vaaler and I – together with other members of the AOM Ethics Education Committee – attempt to pique your interest and stimulate your questions about ETHICS IN THE ACADEMY.

 

PDF version downloadable HERETHE-ETHICIST-BLOG-RESEARCH-2013-03-RETRACTIONS-FNL

 

——————————————————————————————————————————-

The Ethicist Terms of Use: AOM, contributors to THE ETHICIST, and AOM officers, staff and volunteers accept no responsibility for the content of all postings on THE ETHICIST, including the opinions and information posted or circulated by users on THE ETHICIST.  The content of all postings is solely the responsibility of the users. AOM cannot warrant the accuracy of any information posted on THE ETHICIST and disclaims all warranties with regard to information circulated on THE ETHICIST.  This disclaimer includes all warranties of merchantability and fitness

VN:F [1.9.22_1171]
Rating: +2 (from 2 votes)
Retraction: Mistake or Misconduct?, 100% based on 2 ratings

10 thoughts on “Retraction: Mistake or Misconduct?

  1. Benson Honig

    Lorraine, thank you for writing such an important and comprehensive view of ethics in and around AOM. One important ‘weapon’ against fraud is to begin encouraging replication and extension studies. The new Academy journal ‘Discoveries’ will hopefully make inroads. I also noted a special issue call at SMJ. These are important signs that our discipline is beginning to mature- and recognize the need for verification. Thank you again – Well Done!
    Benson

  2. Bernard Forgues

    Lorraine,
    Thank you for your very useful clarification about retractions.
    Just for the record, in example number 12, the name of the person involved is Eugene Z. Geh, not Getz. At least, this is how it is written in Retraction Watch and other outlets.
    Bernard

  3. Jennifer Leigh

    Thank you for this overview of the issue, the typologies, the references, and the discussion questions. This is a wonderful resource for planning discussions in advanced methods classes with master’s and doctoral students.

  4. Lorraine Eden

    Glad you found the blog post helpful – and apologies for the typo of Getz for Geh. I would correct this in the PDF version, but it changes the link. Will correct in the blog itself. thanks!

  5. Lorraine Eden Post author

    I did fix it in the PDF version too — and added another cite to the recent Economist article on this topic (which several individuals had flagged to me). This version has a November 11, 2013, date at the end of the article.

  6. Lorraine Eden Post author

    New paper on retractions posted at SSRN available for download:
    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2361984
    ============
    The Retraction Penalty: Evidence from the Web of Science
    Susan F. Lu
    University of Rochester – Simon Business School
    Ginger Zhe Jin
    University of Maryland – Department of Economics; National Bureau of Economic Research (NBER)
    Brian Uzzi
    Northwestern University – Kellogg School of Management
    Benjamin F. Jones
    Northwestern University; National Bureau of Economic Research (NBER)

    Simon School Working Paper No. FR 13-30

    Abstract:
    Scientific articles are retracted at increasing rates, with the highest rates among top journals. Here we show that a single retraction triggers citation losses through an author’s prior body of work. Compared to closely-matched control papers, citations fall by an average of 6.9% per year for each prior publication. These chain reactions are sustained on authors’ papers (a) published up to a decade earlier and (b) connected within the authors’ own citation network by up to 4 degrees of separation from the retracted publication. Importantly, however, citation losses among prior work disappear when authors self-report the error. Our analyses and results span the range of scientific disciplines.

  7. Benson Honig

    We definitely have had citation cartels in management research. Without systematic replication, we’ll never know what should or should not be retracted.

  8. Lorraine Eden Post author

    Retraction Watch has now flagged five articles in LEADERSHIP QUARTERLY written by Fred O. Walumbwa (Florida International University) as having been retracted by the journal. The five articles (and corrigendum) being retracted from publication are:

    Fry, L. W., Hannah, S. T., Noel, M., & Walumbwa, F. O. 2011. Impact of spiritual leadership on unit performance.Leadership Quarterly, 22(2): 259-270. 4

    Corrigendum to “Impact of spiritual leadership on unit performance” The Leadership Quarterly, 22 (2011) 259-270.

    Peterson, S. J., Walumbwa, F. O., Avolio, B. J., & Hannah, S. T. 2012. The relationship between authentic leadership and follower job performance: The mediating role of follower positivity in extreme contexts. Leadership Quarterly, 23(3): 502-516.

    Walumbwa, F. O., Wang, P., Wang, H., Schaubroeck, J., & Avolio, B. J. 2010. Psychological processes linking authentic leadership to follower behaviors. Leadership Quarterly, 21(5): 901-914.

    Zhang, X., Walumbwa, F. O., Aryee, S., & Chen, Z. X. 2013. Ethical leadership, employee citizenship and work withdrawal behaviors: Examining mediating and moderating processes. The Leadership Quarterly, 24(1): 284-297.

    Avolio, B. J., Rotundo, M., & Walumbwa, F. O. 2009. Early life experiences as determinants of leadership role occupancy: The importance of parental influence and rule breaking behavior. The Leadership Quarterly, 20(3): 329-342.

    See: http://retractionwatch.com/2014/02/07/leadership-journal-to-retract-five-papers-from-fiu-scholar/

Comments are closed.