A higher-ed journal has retracted a recent paper by a New Jersey scholar who failed to adequately cite one of her sources.
Trouble is, the researcher did reference the article more than once — raising the question of whether a retraction, rather than a correction, was the right move.
The paper was written by Lynne Kowski, a professor of mathematics at Raritan Valley Community College in New Jersey, and it appeared online in November 2013 in the Community College Journal of Research and Practice.
Here’s the abstract of the article, “Mathematics Remediation’s Connection to Community College Success:”
The purpose of this study is to test the efficacy of college-level remedial math programs, specifically in community colleges. In this study, chi-square independence test and logistic regression were used to compare the long-term academic outcomes of 1,169 first-year community college students. Findings show that students requiring remediation in mathematics and successfully remediated into a college-level math class, experienced comparable outcomes to those not requiring math remediation. This indicates that remedial math programs can be highly effective at resolving skill deficiencies.
But as the retraction notice indicates, Kowski’s paper was missing a number:
We, the Editor and Publisher of the Community College Journal of Research and Practice, are retracting the following article, ‘‘Mathematics Remediation’s Connection to Community College Success,’’ by Lynne E. Kowski published in Community College Journal of Research and Practice, 38(1), 54–67, 2014.
The aforementioned article failed to reference the following article: Bahr, P.R. 2008. Does mathematics remediation work?: A comparative analysis of academic attainment among community college students. Research in Higher Education, 49(5), 420–450. DOI: 10.1007=s11162-008-9089-4
This action constitutes a breach of warranties made by the author with respect to originality. We note that 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The retracted article will remain online to maintain the scholarly record, but it will be digitally watermarked on each page as RETRACTED.
Bahr is Peter Bahr, of the University of Michigan.
Kowski tells us that she does not agree with the retraction, and that she feels she should have been offered a chance to correct her error:
The retraction states I did not reference the Bahr 2008 article, yet I did reference the article multiple times and in the reference list. Unfortunately there was one area where I forgot to reference him in my conclusions.
I received an email after the decision was made, saying my work was not “original”. It was only after I asked what they meant by my work not being “original”. I gave them a rebuttal stating that this was my original research. It was not until then that they gave details about the missed references.
They only gave me an offer of what I could state to include in the retraction.
Kowski added that she might have pushed back had she had more publishing experience:
As I have only published once before and never encountered this, I did not give any additional comment. I was too shocked and hurt and afraid I would only make the situation worse.
Also, since the decision was already made and I was never offered an opportunity to fix my error, I did not ask.
We contacted the journal to find out why it thought retracting the paper was a better course than issuing a correction and will update this post if we get an answer.
I wonder what would be Prof. Bahr’s comment to this.
This does look bad – for the editor. For consolation I can only think of Michael Eisen’s excellent tweet on the thread #SixWordPeerReview
@mbeisen Jan 21
You forgot to cite my papers
Yet the paper was cited, but not sufficiently often according to the editor. Doesn’t the editor know that highly cited papers are cited in the Introduction? Citations elsewhere in papers fail to lead to over inflated metrics.
I await the editorial response with a high degree of curiosity. The retraction may require a new category of the “Hard done by” variety.
It looks like the Editor and Publisher of the Community College Journal of Research and Practice were too stringent. Based on the information provided above, I think a correction would be a more suitable response. I also think that the Editor and Publisher should retract their retraction.
Lo Mein – If I found myself in the situation described (as the under-cited author), I would support a correction but not a retraction.
If you look at the abstract of the paper (I do not get more) and at Wikipedia, you get some deja vu:
From the abstract of the paper:
“Proponents of remediation feel that postsecondary remediation fills an important role by providing opportunities to rectify disparities generated in primary and secondary schooling to acquire the prerequisite competencies that are crucial for engaging in college-level coursework.”
From Wikipedia, “Remedial education”:
it fills an important niche in U.S. higher education by providing opportunities to rectify disparities generated in primary education and secondary schooling, to develop the minimum skills deemed necessary for functional participation in the economy and the democracy, and to acquire the prerequisite competencies that are crucial for negotiating college-level coursework.
From the abstract of the paper:
On the other hand, critics argue that remediation diminishes academic standards and devalues postsecondary credentials.
From Wikipedia, “Remedial education”:
On the other hand, critics argue that taxpayers should not be required to pay twice for the same educational opportunities, that remediation diminishes academic standards and devalues post-secondary credentials,
The thing with wikipedia is you have to check when the content was added. It could be that someone else plagiarized the paper for the Wikipedia entry. Or it could be that Ms. Kowski also created the Wikipedia entry at the same time as she was writing the paper, without thinking of the ramifications.
I do wonder if there is more to the story. On its face, this is a bad decision by the editor, but maybe there is more.
In the pursuit of excellent science, I checked the Wikipedia entry. It was already there in 2011, the paper is from 2014. But quite an original excuse: Wikipedia took this from me!
Last sentence of the abstract:
“This indicates that remedial math programs can be highly effective at resolving skill deficiencies..”
Last sentence of the abstract of Bahr (2008):
“which indicates that remedial math programs are highly effective at resolving skill deficiencies.”
It has happened before, believe me, although more often with images than text. Someone posts a free image to Wikipedia, someone else copies it for a web site, then a third party (usually a wikipedia editor with a self-assessed expertise in copyright) decides the Wikipedia photo was stolen from the web site and deletes it. (For a time I was deeply involved in administration there.)
For this reason (among others) I never wrote about my own research interests there.
OK, after checking some more, that bit from Wikipedia is actually a long quote from Bahr (2008). If the Kowski article hews so closely to Bahr, then the Kowski article could be argued to be insufficiently original, even if she did cite him a lot.
On the face of it, this seems like a poor decision from the journal, and (if the author’s account is accurate), the journal did not specify the reason for the retraction in its initial correspondence, which I would count as highly unusual, and perhaps indicative of reticence on the journal’s part to state the reason for the retraction, which is a very weak reason in my view. I certainly won’t be submitting to this journal if it cannot respond with some more compelling justification for its behavior.
Bahr pg 440:
prior research indicates that the likelihood
of successful remediation in math declines sharply with increasing skill deficiency
at college entry (Bahr 2007, n.d.; Hagedorn and Lester 2006). Although not related directly
to this pattern, one also might anticipate that students who exhibit the poorest math skills at
college entry benefit less from remediating successfully in math than do remedial math
students who exhibit stronger math skills at college entry. One might ask, does remediating
successfully in math benefit students equally across the varying levels of initial math skill
deficiency? For example, do basic arithmetic students who remediate successfully experience
academic outcomes that are comparable to those of intermediate algebra students
who remediate successfully, and do both of these groups achieve the various academic
outcomes at rates that are similar to those of successful college math students? In other
words, is remediation equally efficacious at every level of mathematical under preparation
Kowski pg 58
Prior research indicates that the likelihood of successful remediation in math declines sharply
with increasing skill deficiency at college entry (Bahr, 2008). One also might anticipate that
students who exhibit the poorest math skills at college entry benefit less from remediating
successfully in math than do remedial math students who exhibit stronger math skills at college
entry. One might ask, does remediating successfully in math benefit students equally across the
varying levels of initial math skill deficiency? For example, do basic arithmetic students who
remediate successfully experience academic outcomes that are comparable to those of intermediate
algebra students who remediate successfully, and do both of these groups achieve the various
academic outcomes at rates that are similar to those of successful college math students? In other
words, is remediation equally successful at every level of mathematical under preparation?
———-
Bahr pg 442, discussing the findings of his research,
This is a remarkable finding, as it indicates that remediation has the capacity to
fully resolve the academic disadvantage of math skill deficiency, at least as far as these
outcomes are concerned. Thus, as it pertains to students who remediate successfully in
math, the primary goal of remediation clearly is being achieved.
Kowski pg 64, discussing her study,
This is a remarkable finding, as it indicates that remediation has the capacity to resolve the academic
disadvantage of math skill deficiency, at least as far as these outcomes are concerned.
Thus, as it pertains to students who remediate successfully in math, the primary goal of remediation
is being achieved at the college selected for this study.
———-
Bahr pg 445
At least three important implications for educational policy may be drawn from this work.
First, as noted earlier, when mathematics remediation works, it works extremely well.
Thus, although critics of remediation might continue to argue that this ‘‘second chance’’ is
a ‘‘waste’’ of resources (Reising 1997), they may not argue that remedial math programs
are failing to meet their objective for students who remediate successfully.
Nevertheless, critics legitimately may question the global success of programs in which
three-quarters of the students who start a journey towards college-level math never arrive
at that destination. Moreover, as noted earlier, it is those students who have the greatest
deficiencies who are the least likely to remediate successfully. Thus, criticism of remedial
programs may be justified on two fronts: comparatively few remedial math students
remediate successfully, and those students who do remediate successfully are disproportionately
those who require the least assistance.
Note, however, that the focus of the analysis presented here is not on the effect of
remedial math coursework in general, but on the effect of remediating successfully in math.
Therefore, one implication that should not be drawn from these findings is that remedial
coursework is detrimental to the academic outcomes of some students. This distinction
between the effect of remedial coursework and the effect of remediating successfully is an
important one because studies that examine the effect of remedial coursework face a
quagmire of problems associated with controlling confounding background characteristics
of students
Kowski pg 66
At least three important implications may be drawn from this work. First, as noted earlier,
when mathematics remediation works, it works extremely well. Thus, although critics of remediation
might continue to argue that it’s a waste of assets and funds, they may not argue that
remedial math programs are failing to meet their objective for students who remediate successfully.
Nevertheless, critics legitimately may question the global success of programs in which the
majority of the students who start a journey towards college-level math never arrive at that destination.
Moreover, as noted earlier, those students who have the greatest deficiencies are the
least likely to remediate successfully. Note, however, that the focus of the analysis presented
here is not on the effect of remedial math coursework in general, but on the effect of remediating
successfully in math. Therefore, one implication that should not be drawn from these findings is
that remedial coursework is detrimental to the academic outcomes of some students. This distinction
between the effect of remedial coursework and the effect of remediating successfully
is an important one.
(This last one is the concluding paragraph of Kowski’s report.)
I assume Kowski actually studied the academic records she claimed at that her data results and tables are accurate.
Kowski states “Unfortunately there was one area where I forgot to reference him in my conclusions.” But, Rolf Degen has found two additional instances of text in Kowski’s abstract that are very similar to text derived from an article that, according StrongDreams, was authored by Bahr (I assume that there was no attribution to Bahr in those two instances from her abstract). If this is correct, then Kowski forgot to reference Bahr in more than one area. Given the above, one has to wonder whether there more segments of unattributed text exist in Kowski’s paper. The presence of a few unattributed segments of text does deserve a correction rather than a retraction. But, as is often the case with these types of situations, the question arises: Where does one draw the line?
And there is the typical pattern: slightly altering the sentences in order to cover up one’s track against plagiarism software.
The journal seems to be misusing the term “reference”.
The paper correctly referenced Bahr et al. But it seems it did not correctly indicate that it was quoting from Bahr et al.
Many plagiarized papers do include references to the source, in my experience. Sometimes they do so quite meticulously, with a reference at the end of every copied sentence.
It makes no difference – you copy-and-paste, without quotes, it’s plagiarism.
If there really is a reference at the end of every copied sentence, isn’t it more likely that the author was just unfamiliar with the proper use of quotation marks? I mean in such a case they are quite clearly not trying to appropriate someone else’s idea, or they wouldn’t have added all those references.
If someone is really meticulously referencing each sentence I would call a lack of quotation marks a grammar error, not plagiarism.
Then it is inadvertent plagiarism – possibly the only kind of believable inadvertent plagiarism (as opposed to ‘I forgot that I downloaded it instead of writing it’, Wolpert style). If it’s a genuine mistake then fair enough, no harm done, but it’s still technically plagiarism.
Also, there aren’t references at the end of every copied sentence. See above, copy-pasted as is.
Indeed in some academic practice (I know about Latin America, in earth sciences/natural resources sciences) a citation immediately at the end of the sentence is usually taken to mean that the entire sentence is from the cited source, it’s an implicit quotation mark. I teach research ethics at my University in Netherlands and have to inform such students of the European/N American standard of quotation. But in N American practice, from high school on, it’s clearly stated that anything not the author’s original creation (including synthesis, re-interpretation) must be in quotes, complete or suitably marked with … or [added] if necessary, and cited.
The examples quoted above go way beyond this “misconception”!
StrongDreams and Rolf Degen have now provided evidence of (apparently) unattributed text from the Bahr (2008) source. Miguel Roig has added the most pertinent question: “Where does one draw the line?” But in my small analysis here, I wish to also offer up some feed for thought regarding another angle and aspect of this story.
THE RISK AND A SOFT WARNING
The implication of this story could be more serious than some believe and may, together with the recent Beall-induced retraction (http://retractionwatch.com/2014/01/20/jeffrey-beall-scores-a-retraction/#comments), signal a turning point in which authors whose papers were pulled (retracted) based on minor (limited) acts of plagiarism (or textual recycling) may begin to fight back against editors, journals and/or publishers, with the claim that coercion was used. Allow me to elaborate.
Kowski states: “They only gave me an offer of what I could state to include in the retraction.” This suggests, as I have claimed all along, that publishers and editors tend to be pushy, forcing authors to feel guilty and giving them the exact text that they must use in a retraction notice, even possibly against their will. Could this explain why so many retraction notices we see at RW published by publishers that lack transparency and details, could themselves be employing unethical, questionable or forceful principles or methods to obtain statements by authors? Seriously, think about it, why do we see so many standardized self-plagiarized retraction notices with the exact same wording? Why do we see little fight back from scientists except for those with an inflated super-ego?
At this point, I wish to give a Wikipedia definition for coercion (http://en.wikipedia.org/wiki/Coercion) (I did not check the source of the Wiki definition):
“Coercion /koʊˈɜrʃən/ is the practice of forcing another party to act in an involuntary manner by use of intimidation or threats or some other form of pressure or force, and describes a set of various different similar types of forceful actions that violate the free will of an individual to induce a desired response. These actions can include, but are not limited to, extortion, blackmail, torture, and threats to induce favors. In law, coercion is codified as a duress crime. Such actions are used as leverage, to force the victim to act in a way contrary to their own interests. Coercion may involve the actual infliction of physical pain/injury or psychological harm in order to enhance the credibility of a threat. The threat of further harm may lead to the cooperation or obedience of the person being coerced.”
In other words, coercion is illegal, or at minimum, ethically and morally suspect. This could add a nasty twist to retractions in favor of authors that only plagiarize slightly, or unintentionally. If the scientific community senses that in some cases retractions are being FORCED, or that authors are being FORCED to use specific wording, imposed by the editors, journal or publisher, then I think we are entering a dangerous new phase where the very ethics of those forcing the retraction note could start to come under very close scrutiny, and counter-attacks. Although the publishers are trying to show an iron-fist against the lack of ethics, to correct the literature and to instill firm ethical standards in publishing, the methods being used are starting to look, in some cases, highly questionable.
Let’s not forget, most of these publishers have very powerful legal teams and their smooth-talking PR teams will quickly force the hands of authors. So, in most instances, the authors will cave in, and accept, without much of a fight, the imposed wording for the retraction note. This because most scientists will fear for their jobs and careers and thus wish to seek the path of least resistance, even if it is a forced retraction, to resolve the issue. A forced retraction, through imposed wording, would give editors, journals and publishers the higher moral ground, even if it used unscrupulous methods. It is this paradoxical situation that we now have to begin analyzing moving forward, and retrospectively in already issues retraction notices.
THE SUGGESTIONS
I know that the vigilantes and hard-liners among the retraction-hungry commenters will loathe my assessment, but I warn (and predict) that there will be an aggressive fight-back by authors and academics against this type of retraction unless the following is done for all future and past retraction notices:
a) a FULL and transparent explanation of what the infraction was. In the case of provable figure manipulation, figure duplication, or data manipulation, there is no doubt that these cases of plagiarism and self-plagiarism should have a zero-tolerance policy and approach, namely a retraction. Even so, the retraction note MUST reflect the following information: a) what the academic “crime” was, in plain-speak, without any euphemisms. If there was figure, data or table duplication, then call it that. If there was data manipulation, then call it that. If there was serious plagiarism, then call it that. But quantify it. See my notes below on a possible solution.
b) The editors should issue their uncoerced statement, and uncontrolled by the publisher. The publisher should issue a separate, or standard, statement. And the authors should issue their own statement. If editors wish to collude (or cooperate) with publishers, then that is their decision. But an author MUST never be forced to state something that he/she does not want to state, and a publisher MUST NEVER indicate the exact words that an author must use. If yes, then this is coercion, and is a potentially criminal act (independent of the guilt of the author).
I suggest that retraction notices that are incomplete and that lack transparency and details be themselves corrected, or retracted for the failure to declare all facts. If we draw a parallel to authors that submit a paper to a journal in which facts are hidden, this is considered by the publishers to be unethical, so why should incomplete or untransparent retraction notices be viewed as any different? Double-standards (actual or perceived) by publishers could cause their downfall if academics perceive it this way.
Community College Journal: A CASE STUDY
If editors and publishers do not provide full, transparent and quantitative reasons for a retraction, then we should be wary of foul play, the lack of professionalism or experience on behalf of the editors or publisher, or possibly even outside interference or bias (none can be ruled out unless proved otherwise). Rolf Degen and StrongDream’s acute analysis indicates that perhaps there were several instances of plagiarism or similar from the Bahr (2008) paper (yet to be confirmed), in which case my own analysis might need to be adjusted (or cancelled). Recently, retractions based on the issue of a few sentences (= let’s say between 2 and 5 sentences) being similar in context to another source, but not necessarily the original source, has left the academic community perplexed, the ethics societies such as COPE, WAME, CSE, ICMJE and ORI silenced or cautious, and those whose papers were retracted by such aggressive actions stunned.
Why not use a rule as simple as that used in sports, like soccer/football? A yellow card would be a warning, in which a correction would be issued indicating which text had failed to reference the source, and what the exact source was. In the same correction note, the journal/publisher would also state that this correction serves as a first warning and that this action is referred to as plagiarism, without fear of using the “P” word. Even the soccer-rule could be a little harsh for such a minor offence, and we could then use the baseball rule in which you get two strikes and then you’re out on the third (i.e., a retraction if you have been accused of plagiarism three times). Then, for a new innings, you get another three chances. The retraction by Community College Journal does show a rather iron-fisted approach, but which could make sense if figures were manipulated, if figures were duplicated, or if data was fabricated. But for such cases of textual recycling, there should be a call to common sense. Excessive actions like this one draw criticism by the scientific community as foul play, excessive, unfair, and possibly even unprofessional. And, in an extreme case where the author can prove that the editor or publisher FORCED them to make a statement against their will, or forced them to use words that they did not want to use, then we could potentially be dealing with a criminal act, coercion. Unbalanced and rushed decisions like this not based on common sense will perhaps have a negative effect on the journal. Instead of sending a message to the scientific community that justice will be met with a fair hand, based on rational thought, common sense and fair, humane and calculated responses, it is showing a more than zero-tolerance (i.e., excessive) approach. Future authors would fear submitting to this journal and/or publisher, even if they were quite confident in the content of their papers’ text simply because one slight accidental or careless slip would land up in a retraction, rather than something more fair like a correction.
Surely, in this case, the message that the journal, editors and publisher are sending to Kowski are too harsh? They will engender bitterness in our young scientists (my critics will argue that it will engender caution and greater care in writing), and send the wrong message that science is intolerable and does not accept mistakes. I do not know her, but I sense that Kowski was being quite frank and honest (and bitter) in her remarks that she was naïve, that the error, which she was not aware of, was serious, and that she was concerned about the coercive nature of the wording forced upon her by the editor. I say, use the football or baseball rule, and give her a second chance. Using a more fair and balanced system using the rough guidelines I suggest above, I am convinced that over time, some scientists will learn and reform within a fair and rational publishing framework, while serial plagiarists or frauds will (hopefully) fall by the wayside.
Retractions are here in force, it seems. And they are here to stay. They form part of the publishing panorama. However, there are still many unknowns, and even seasoned editors and established journals are now starting to feel the heat, as are scientists who have never been held up to as a high a level of scrutiny by academics and society as now. Yet, rather than surging forward with an ultra-aggressive approach to seek blood-thirsty justice, I call on the more radical pool to exercise moderation in punishment, because, should the methods being employed by the issuing authority be found to be in any way morally or ethically suspect, justice will be turned on its head, and the academic community will revolt en masse.
If the journal (any journal) decides to retract, and the authors do not agree, that lack of agreement should be included in the notice. Other than that, I’m not sure I follow your arguments about coercion etc. If the authors believe the study is sound they can resubmit it elsewhere.
Here, where 4 paragraphs are copied nearly word for word without citation or quote marks, what really is the remedy? Potentially the study could be re-written, but at the heart, she would still be re-using someone else’s ideas and conclusions. Also, is the journal taking the risk that there are other undiscovered sections copied from other sources? And Ms. Kowski’s response was not, “I accidentally quoted Bahr 4 times extensively and asked for the chance to rewrite” (the Doris Kearns Goodwin defense?) but only “I omitted one citation.” I don’t think she understand why the article as written is a problem.
(Scientifically speaking, she could have made a very different and interesting conclusion. The Bahr study was of something like 20,000 students getting remedial math at a major university, her study is a couple hundred students getting remedial math at a community college. She could have concluded that remedial math is effective even when given at what would normally be considered to be a lower quality setting. Instead she just copy/pasted Bahr’s conclusions.
I think the only remedy needed in *this* case is for the editor to issue a fuller retraction notice, such as,
“Sections of 4 separate paragraphs, including the abstract and concluding paragraph, were re-used nearly verbatim from Bahr 2008. In our editorial judgement, this issue could not be corrected by adding additional citations and quotation marks. Therefore we have marker the paper as retracted.”
There is an easy remedy: They could simply replace the name Lynne Kowski by the name of the plagiarized author, Peter Bahr, and let the paper stay as it is. This kind of swap was already done once with a math paper:
http://zbmath.org/?q=an:1257.11089
The whole affair again reminds us of the necessity of a good and free plagiarism software. When will Google deliver its Plagifind?
While Dr. or Ms. Kowski unfortunately chose to, or erred, in lifting passages from Dr. Bahr’s work to describe her rationale and conclusions, she apparently did do some original research using the student records from her community college, and her findings agreed broadly with Dr. Bahr’s findings which used student records from a major school. I would encourage her to rewrite and try to re-publish (with full disclosure to the new editor so the new manuscript can get a proper peer review).
I think the strange wording in the retraction is due to the fact that the editors wanted to be polite and not humiliate her in public view. They did not want to write: “That whole piece is pilfered together from front to back. “
I agree with the view of Editors and Publishers abusing authors, sometimes. I think researchers and RW have not been sufficiently aware of these problems. The current assumption seems to be that only authors are the bad apples, and retractions are effective means to eliminate them. This view reflects a lot of ignorance and prejudice, and of course it is great for those in power (editors and publishers). More critical thinking, please!
I agree with Jaime that “Community College Journal of Research and Practice” is shooting itself in the foot. By moving in this direction, sooner or later they will receive letters like this one:
http://angryphysics.blogspot.mx/2011/07/turning-tables.html
Holy moly, this was like a little story or a movie. First it’s about the hard done by Author and the aloof star chambery Editors. Then the clues gradually acrue, at the hands of the doughty lit searchers. And it turns out the Author is actually the nefarious culprit. Wow! But the Editors come out badly too. The Stumblebum Authorities . . . .
I didn’t even know cut and paste was a kind of plagiarism until I read Neuro Skeptic’s recent post. Disturbing. You realize of course that if Kowski had been doing this properly in her conclusions she wouldn’t have copied Bahr’s wording, she would have quoted him then said “and guess what? I got the same results, how cool is that?” or words to that effect.
Many thanks to the doughty lit searchers. Truly impressive.