Please visit our most recent publication: Hill, HZ and Pitt, JH. Failure to Replicate: A Sign of Scientific Misconduct? Publications 2014, 2, 71-82; doi: 3390/publications2030071
You might also be interested in Pitt JH and Hill HZ. Statistical Detection of Potentially Fabricated Data. Posted on arXiv: 1311.5517, and on Figshare: http://dx.doi.org/10.6084/m9.figshare.858921
See also a recent talk "A Spreadsheet Program for Use in the Detection of Anomalous Numerical Data of the Type Frequently Encountered in Cell and Radiation Biology Colony Survivals" presented at the Biennial Meeting of the American Society for Photobiology in San Diego, June 19, 2014 and a poster presentation of similar name and abbreviated content that was presented at the Annual Meeting of the Radiation Research Society in Las Vegas, NV September 21-24, 2014.
Attempts to Publish
The scientific atmosphere over the past decade or so has undergone a sea change. Over the centuries there have been a number of cases of fraud but until recently, these have been considered to be rare occurrences. In the last decade or two, however, the scientific community has become more aware of instances of scientific misconduct and out and out fraud. Articles have appeared in the news sections of Science and Nature, cases are reported almost weekly in the NIH Guide and the post eRetraction Watch (http://retractionwatch.wordpress.com) reports retractions on more than a daily basis. Of course not all of these are the result of fraud or misconduct, but many are. So I thought, of course, my paper – co-authored by Joel Pitt – would be a welcome addition to the list of revelations. Wrong.
As of this writing, our efforts have been rejected by 9 journals. Only 3 have gotten past the editorial office. The first submission went to Science on 6/15/11. It was rejected by the Editor on 7/6/11. In this first attempt, I anonymized the players. Nature makes a preliminary judgment based on a prepublication synopsis which I submitted on 7/13/11. It was rejected by the Editor on 7/14/11. Next, I set my sights on PLoS ONE, but before doing this, on the advice of my daughter’s partner, a lawyer, I consulted a law firm that specialized in defamation law. This little sally back into the judicial arena set me back about $20,000 more, but I learned a lot. The attorneys pointed out that the case was so transparent that there was no point in anonymizing. However, my strategy should be to present the facts and let the reader draw whatever conclusions were to be drawn. I not only rewrote the paper according to their suggestions, I submitted it to the American Journal Editors for both editorial services and statistical critique (another $1000). Once vetted and certified in this manner, the paper was submitted to PLoS ONE which received it on 1/22/12 and the Editorial Office rejected it on 2/10/12. I appealed the rejection and was rejected again. The next stop was the journal Accountability in Research. By its very name, this would seem to provide the appropriate setting for our efforts. I sent it directly to one of the editors with whom I had corresponded earlier when looking for an expert in ethics (which we later decided not to employ). Here again, I hit a blank wall. Since 4 of the 8 papers in question had been published in Radiation Research I decided to apply to the Editor of that journal. In his rejection letter, the editor said that I would be allowed to withdraw my name as an author on one of the papers but that Howell would be given a chance to respond. I actually have no objection to that, in fact, I would be delighted to debate with Howell about the radiation biology aspects of the studies, but I decided to seek further before doing so. Our next stop was the Annals of Applied Statistics submitted on 3/13/12. This was the journal that published the statistical analysis of chemotherapy studies at Duke University that have led to multiple retractions and several resignations. This time, at least, the paper made it past the Editor in Chief to be rejected by an associate editor on 3/16/12. This latter had some helpful comments that might make it acceptable to them after revision. Here is the draft that was submitted to AoAS. Pitt completely revamped the statistics and developed an impressive method to analyze Bishayee's colony results and put p-values on the probabilities of finding the rounded average as one of the 3 triplicate colony counts. We submitted this new revised paper to AoAS and were rejected again but this time with permission to resubmit. The critiques by the Associate Editor (a different one) and the referee mainly had to do with the format. The referee's comments were rather odd and his/her turns of phrase were "anodyne" (his/her word). We conscientiously rewrote, resubmitted and were re-rejected. I appealed to the Editor-in-Chief who politely declined to reconsider. The entire correspondence may be found in the attached (AOAS Correspondence) should anyone care to look. Bruce Ames, famous for the Ames test for mutagenesis/carcinogenesis, co-authored a series of exchanges in the Life Sciences Forum of the FASEB Journal in 2009 (McCann, Hudes, Ames "Unusual clustering of coefficients of variation in published articles from a medical biochemistry department in India" FASEB J 23:689-708 (2009)). The challenged authors were given an opportunity to respond. This seemed eminently fair. We would certainly be happy to have Howell and Bishayee respond to anything that we might publish. I submitted, as required, an Initial Inquiry: an abstract and cover letter on December 17,2012. Within hours, we were rejected. We then submitted to Statistics in Medicine on 12/29/12. We got the bad news on 1/7/13 that "Although the problem you are working on is interesting, the degree of methodological innovation is not at the level of development that would receive a high enough priority for publication in Statistics in Medicine". Since one of the Howell-Bishayee papers had been published in Acta Oncologica we decided to give that one a try. Success (sort of) at last: I submitted an abstract for a poster to be presented at the Biennial Meeting of the American Society for Photobiology in Montreal, June 23-27, 2012. It was accepted and my poster was displayed with much interest. Slapped in the face again: I submitted an abstract for a poster to be presented at the Annual Meeting of the Radiation Research Society in San Juan, PR from 9/30 – 10/3/2012. That abstract was rejected. I believe it is very rare for a presentation of a member in good standing to be turned down. Curious.
7/18/12 The talk was attended by graduate students, faculty, post-docs and staff. There was, over all, amazement that someone could get away with producing so much suspicious data and that the authorities did such a poor job of policing the research.
Our next approach
The letter to Micron had some rather unexpected repercussions. Shortly thereafter, I received a letter from the medical school Dean threatening me with dismissal if I did not "cease and desist". So much for scientific truth and integrity. Pests like me should be shot down even iif they are exposing, to say it politely, some anomalous data backing up figures. I responded (cf my letter in response) but I have not ceased and desisted -- cf the article about me in Nature (the URL is on the home page). On October 10, 2013, we submitted a re-edited, revamped version of our paper to "The American Statistician" (if you would like to see it, you can follow the link by clicking on the title or view it on ArXiv or FigShare (see above). After 4 months of waiting, I finally corresponded with the secretary who told me it was back from the referees and things looked favorable. More time passed and I finally queried the Editor. Our correspondence and the reviews are included in this Link. The reviews (2 from referees and 1 from the Associate Editor) are the best I have seen in a long time. But it appears that Dr Christensen, the EIC, really did not want to publish the paper so he sent it to Terence Speed, he who destroyed the NIH statistician, James Mosimann, in the Baltimore-Imanishi-Kari case and has even testified for OJ Simpson. Speed's report was the kiss of death. I have been corresponding with a scientist who was very involved in the Baltimore-Imanishi-Kari case and he said "Speed is dishonest". So we soldier on. But I have to say, it was reassuring to see that we were finally getting serious comments and consideration by 3 obviously astute referees (the referee's report for Annals of Applied Statistics was some kind of a joke).
While we were chewing our nails waiting to hear from The American Statistician, we decided to try another tack. Reading the Instructions to Authors of the BioMed Central Journal BMC Medical Research Methodology, it actually looked as though it might be possible to submit to them the same paper that had been submitted to The American Statistician. so we gave it a try, being absolutely open with them about what we were doing. Well, it did not work but in their rejection notice they did say that if the American Statistician rejected it, their editorial office had expressed interest in reviewing it. So when we were rejected by The American Statistician, we did submit it to the BMC journal and there it has sat since May 23, 2014. I did receive a private email from a person who acknowledged being a referee on the paper, with suggestions for making it better and stating that s/he hoped they would publish it -- but no word yet (September 8).
Meanwhile, in January, 2014, I learned that Grant Steen, a person well-known for his interest in scientific publishing and misconduct, was organizing an on-line special issue of the MDPI journal Publications entitled "Misconduct in Scientific Publishing". We composed a new article entitled "Failure to Replicate: A Sign of Scientific Misconduct" and submitted it for the 28 February deadline. It was subjected to 6 reviews before finally being accepted and published on-line on September 1, 2014 (follow the same link for "Misconduct..."). So, at long last, we have published some of our analysis (the focus of this paper is quite different from the pending submission to BMC Medical Research Methodology which deals with Dr Pitt's statistical model and our analysis of all of the numerical data from the lab spanning a nearly 10 year period). While this acceptance and publication is most gratifying, it is disturbing that MDPI is on Beall's list of predatory publishers. Our interactions with the journal editorial office were extremely professional and we did not pay any publication fees. At this writing, I have requested that Dr Beall take the MDPI Publications journal off his toxic list and he is considering it.
Howell and Bishayee Publications
Here follow the list of papers co-authored by Howell and Bishayee in order of journal acceptance, followed by the grant application and renewal. Citations are as of 7/13/13.
1.Howell RW, Goddu SM, Bishayee A, Rao DV (1998) Radioprotection against lethal damage caused by chronic irradiation with radionuclides in vitro. Radiat Res 150: 391-399. Accepted 5/21/98. Cited 16 times, most recent Dec, 2012.
Isotopes involved: P-32, tritiated water, Iodine-125, Iodine-131, tritiated thymidine
2.Bishayee A, Rao DV, Howell RW (1999) Evidence for pronounced bystander effects
caused by nonuniform distributions of radioactivity using a novel three-dimensional
most recent Mar, 2013.
Isotope involved: tritiated thymidine
3.Goddu SM, Bishayee A, Bouchet LG, Bolch WE, Rao DV, et al. (2000) Marrow toxicity of 33P-versus 32P-orthophosphate: implications for therapy of bone pain and bone metastases. J Nucl Med 41: 941-951. Accepted 9/14/99. Cited 17 times, most recent in 2012.
Animals: GM-CFC (bone marrow cell colonies) survivals: There are 9 experiments in which the terminal digits of the colony counts are not significantly different from uniform but the Z-score for expected number of triples containing the average is 6.69 for a p-value of less than 10 raised to the minus 9 power.
4.Bishayee A, Rao DV, Bouchet LG, Bolch WE, Howell RW (2000) Protection by DMSO against cell death caused by intracellularly localized iodine-125, iodine-131 and polonium-210. Radiat Res 153: 416-427. Accepted 12/23/99. Cited 38 times, most recent 3/2013.
Isotopes involved: tritiated water, Iodine-125, Iodine-131, Polonium-210
5.Bishayee A, Rao DV, Srivastava SC, Bouchet LG, Bolch WE, et al. (2000) Marrow-sparing effects of 117mSn(4+)diethylenetriaminepentaacetic acid for radionuclide therapy of bone cancer. J Nucl Med 41: 2043-2050. Accepted 5/30/00. Cited 28 times, most recent in 2/2013.
Animals: GM-CFC (bone marrow cell colonies) survivals: See #3. Cannot tell which experiments were used in which paper.
6.Bishayee A, Rao DV, Howell RW (2000) Radiation protection by cysteamine against the lethal effects of intracellularly localized Auger electron, alpha- and beta-particle emitting radionuclides. Acta Oncol 39: 713-720. Accepted 7/6/00. Cited 7 times, most recent 12/2012.
Isotopes involved: tritiated water, Polonium-210, Iodine-125, Iodine-131, also cysteamine alone
7.Bishayee A, Hill HZ, Stein D, Rao DV, Howell RW (2001) Free radical-initiated
and gap junction-mediated bystander effect due to nonuniform distribution of
incorporated radioactivity in a three-dimensional tissue culture model.
Isotope involved: tritiated thymidine
8. Howell RW, Bishayee A (2002) Bystander effects caused by nonuniform distributions of DNA-incorporated (125)I. Micron 33: 127-132. Available on-line 9/01. Cited 29 times, most recent 3/2013.
Isotope involved: Iodine-125
of Radioactivity. Submitted for the 11/1/99 deadline. Funded 7/1/99-6/30/05 Link p30
of Radioactivity RENEWAL. Funded 7/10/06-6/30/11
NON SEQUITUR (c) (2009-11) Wiley Miller Ink, Inc. Used by permission of Universal Uclick. All rights reserved.