This post is part of a series, which serves to provide hands-on information and resources for authors and editors.
Retractions – the removal of an article from the academic record at any time after its publication – are probably every author´s nightmare. Whether they result from genuine misconduct or honest error, in any case they spotlight how research projects have failed. From 1997 to 2012, the rate of retractions rose sharply, and since 2012 appears to have leveled off at about 4 retractions per 10,000 publications.
While these numbers might not sound very high, the effects for the community can be detrimental and far-reaching. As an example, retracted manuscripts between 1992 and 2012 were found to account for a loss of approximately $58 million in funds by the National Institutes of Health (NIH). Furthermore, the risk of retracted research being inadvertently spread is high and may have dangerous implications. Articles that have been long retracted – like an infamous study from 1998, which falsely claimed a relationship between the MMR vaccine and autism – sometimes haunt academic literature and the public discourse for decades.
As stated above, ethical violations do not always happen on purpose but can also occur simply due to a lack of knowledge. If you are reading this article, you most likely don’t have any criminal intent but simply want to ensure best publishing practices and avoid misconduct, the rejection or even retraction of your article. That’s certainly a good requisite, so let’s move on!
Know Who to Trust
Some people would claim that we have entered a kind of “post-truth” era, where emotions and personal beliefs are valued higher than scientific facts. Whether that is completely true or not – if we want to counteract this development, the public needs to be able to trust the work that researchers are doing. In order to nurture this trust, academic journals have to comply with a number of rigorous and transparent standards. A legitimate, non-predatory publisher will make sure to check all the relevant boxes to safeguard the quality, integrity and reliability of the content they publish.
But beware of the predatory ones! They prioritize self-interest at the expense of scholarship, usually provide false or misleading information, deviate from best editorial and publication practices, lack transparency, and use aggressive and indiscriminate solicitation practices. If you are unsure whether a journal is predatory or legitimate, check their website thoroughly. Also, check if the journal is a member of DOAJ, COPE, OASPA, or STM. Look into the journal’s editorial policy as well as the editorial board, and read through past issues of the journal. In case of doubt, do not submit!
The aforementioned Committee on Publication Ethics (COPE) is the entity you should be aware of when looking for detailed resources on academic misconduct. The UK-registered charity was once founded as a “self-help group” for journal editors, but authors can also benefit from the resources they provide.
In the following, let’s go through the 4 most common types of unethical publishing and see how to avoid them!
- Plagiarism & Duplicate Publication
- Unethical Research
- Fabrication & Falsification
- Authorship Misconduct
1. Plagiarism & Duplicate Publication
This practice certainly has been the downfall for more than one researcher (or politician) in the past – appropriating another person’s intellectual property as their own. It does not matter what kind of format the source is (abstracts, grant applications, speeches, research papers…), whether it is published or unpublished, and in what language it is written – when using the words, thoughts, ideas or expressions of others, they need to be attributed properly.
Plagiarism can also have copyright implications. Be aware that copyright infringement, whether intentional or unintentional, can be grounds not only for retraction, but can also result in significant penalties. If you want to reproduce a figure that has already been published in another work, do not forget to ask the holder of the copyright for permission, or, even easier, opt to use open access content (note, however, that each license offers different reuse rights).
Self-plagiarism, i.e. using large text chunks of your own previously published manuscripts is also something you should avoid, as it creates redundancy and possible copyright infringements. Instead, cite the original publication, build on your previous work, and present it in a new way. The most blatant form of self-plagarism would be duplicate publication, i.e. publishing all results or a substantial part of your study more than once (usually in another journal) without adequate cross-referencing or justification. On that note, please never submit your manuscript to more than one journal at a time.
Upon submission, most journal editors will run your manuscript through a plagiarism checker. These tools provide them with a percentage value of similarity to other publications. Some universities or institutions enable their students and academic staff to use fee-based plagiarism prevention software like iThenticate or Turnitin as well. If that is not the case for you but you would still like to check your own work, there are free plagiarism checkers you can find online, e.g. Plagiarisma and Grammarly.
Lastly, be aware that preprints can lead to high similarity values in plagiarism checks. Thus, when you submit a manuscript, always remember to mention – if applicable – the existence of related preprints (e.g. on arXiv or SocArXiv). Although journal editors will check the sources of similarity, a short note in the cover letter can help prevent misunderstandings or (unfounded) suspicions of plagiarism.
2. Unethical Research
As a general rule, research on humans and animals should minimize possible harm and maximize possible benefits. Such risk of harm to participants exists in many types of research, and it may be not only physical but also of psychological, emotional, social, financial, or legal nature. Particular carefulness is needed with vulnerable and disadvantaged groups or with children. In a research publication, it is therefore of utmost importance to verify that your research met all relevant ethical standards; that you have obtained ethical approval for the study by your institution’s review board if needed; and that subjects have given informed consent where appropriate, i.e., they have been clearly and adequately informed that they are free to decide whether or not to participate in the research and whether data collected from and about them will be included in analysis.
The gold standard for ethical research involving human subjects was set by the World Medical Association in in 1964 with the Declaration of Helsinki (last amended in 2013). Further guidance on the research on humans and informed consent rules can be found from APA’s Ethics Code and the Guidelines for Research Ethics in the Social Sciences, Humanities, Law and Theology by the Norwegian National Committee for Research Ethics in the Social Sciences and the Humanities (NESH).
If you are located inside or outside the European Union and collect data from European data subjects, you should also be aware that the General Data Protection Regulation applies to you. It applies as long as the data remains identifiable, both directly (e.g., by name or address) and indirectly (e.g., by geographic location, which, combined with other information, may allow for possible identification). Researchers must obtain the consent of the data subject from the outset or perform pseudonymization. Regardless of where your research is taking place: always keep in mind to not collect personal/sensitive data if it is not necessary for your research project!
The use of non-human subjects in research has always been the subject of heated debate. Every researcher should carefully consider the question of whether animals really need to be used for the research when planning their studies. Before animal testing can be conducted, the research protocol and sample sizes must be reviewed by the scientist’s animal ethics committees. In most cases, these committees follow the 3-R rules:
- Replace animal testing wherever possible with other methods such as in vitro biological systems or mathematical modeling
- Reduce the number of animals used as much as necessary; use only as many animals as needed to obtain reliable data (read the literature carefully! Perhaps your intended experiment has already been performed and reported)
- Refine the study to minimize the animal suffering and improve welfare
Once your study has been approved, it is imperative that you work with the local animal welfare committee, which will oversee that animals are housed in appropriate facilities, properly handled by trained personnel, and have access to veterinary care.
Experiments on humans and/or animals that have not received appropriate ethical approval and/or were conducted under unethical conditions must not be published. This issue is particularly relevant in the medical area, but also in psychology, social sciences and education. If your institutional/national framework does not have a research ethics committee which you can consult, explain the situation and support your explanation with evidence. Clearly state how you have ensured that your research meets all ethical and legal requirements.
On a practical note, you should add your ethics declarations in the Methods section and/or in a specific paragraph at the end of the manuscript. Specifically, medical journals will often ask you to provide a separate ethical statement, which might also contain information about conflicts of interest, research data availability, author contributions and other details. Always make sure to carefully read the instructions for authors to find out what format and data are required by the journal to which you are submitting your work.
3. Fabrication & Falsification
It goes without saying that the invention or manipulation of experimental data are strictly taboo for any honorable researcher. Trust and good faith are the principles on which science and academic publishing depend; at the same time, this is what makes systems vulnerable to abuse.
A study on risk factors for scientific misconduct by Daniele Fanelli et al. suggests that academic culture, peer control, money-based publication incentives, and national misconduct policies may influence the scientific integrity of researchers. This breeding ground has even given rise to actual factories for the systematic production of falsified research results – so-called paper mills, which sell the authorship of already accepted papers to researchers. In a recent survey of nearly 7,000 scientists in the Netherlands, a considerable 8% of the participants confessed to having falsified or fabricated data at least once in the previous three years. Yet, we may be only seeing the tip of the iceberg.
One important and troubling aspect of falsification is image manipulation. Because wide-scale screening of this issue is still lacking, it’s difficult to make a definite statement about its prevalence. However, a study published in the journal mBio found that 3.8% of 20,600 published papers with Western blots (an analytical technique in molecular biology) were found to contain manipulated images. Half of these seemed to be the result of fraudulent behavior.
Of course, it is acceptable to format an image to match a journal’s requirements, i.e. to compress it or to convert it into another file format. Other minor adjustments, like cropping empty space from the edges, are also acceptable as long as they do not have any influence on the interpretation of the image. However, you should be very careful with setting the brightness and contrast of photos, for example. If you really have to make changes, make sure that you apply them to the whole image and that you do not lose the subtle differences between the image’s pixels. Try to choose the best, most representative photos from the beginning and to keep modifications at an absolute minimum.
Authors should be aware that more and more publishers are now starting to use software which automatically identifies manipulated and possibly unreliable or plagiarized images, so the probability of getting caught is growing.
There is also increased consideration of requiring authors to submit their raw data at the time of submission or even making it publicly available. So don’t forget to always keep your lab book neat or maybe even think about an electronic alternative!
4. Authorship Misconduct
Putting your and your co-authors’ names at the top of a manuscript is a big deal. Doing so implies that all of you take responsibility and have contributed substantially to the reported work. (Learn more about authorship criteria and guidelines within different research specialties from COPE).
Make sure to not leave out researchers who should be included. At the same time, do not include so-called “guest” or “gift authors”, i.e. people who have not contributed significantly to the manuscript. The latter is still quite a common phenomenon in academia but considered ethical misconduct nonetheless.
Always make sure that each co-author has seen and approved the submission. Communicate with each other and agree on the order of author names. The most common way is to order the names by relative contribution. In the humanities, however, it is sometimes by seniority or alphabetic. In any case, avoid jostling for authorship positions after submission.
Last but not least, make sure to disclose any (potential) conflicts of interest you and your co-authors might have. These could be of a financial, personal or professional nature. You might have received funding from industry sources or have personal financial interests in the published work. The publication process could also be influenced by non-financial competing interests in the form of personal relationships with organizations and individuals, memberships or advisory roles. Omitting these ties could potentially damage your credibility as a researcher.
So Much For Now
We hope that these guidelines have given you more confidence in dealing with scholarly publication ethics. By sticking to these main ethical principles, you are doing your part to maintain research integrity and public trust in science. Congratulations for your commitment and good luck with publishing your work!
[Title image by Twomeows_IS/iStock/Getty Images Plus]