Research Integrity and Publication Ethics

In scientific research, misconduct can take many forms and involve the actions of many different stakeholders — authors, editors, reviewers, and publishers. All parties must understand what is expected of them and be informed about industry best practices, guidelines, and codes of conduct.

AIP Publishing defines misconduct as any conduct that violates or compromises the ethical standards and expectations determined by the scientific publishing community. This includes cases where there is a breach of legal duty, responsibility, or principle of law. As a publisher committed to advancing the physical sciences, AIP Publishing is responsible for clearly defining what is expected of authors, providing transparent policies and procedures, and taking appropriate and timely action when needed.

Editors and staff use resources such as those from the Committee on Publication Ethics (COPE) to address ethical concerns, and any potential publication misconduct is investigated and resolved according to these best practices.

Table of Contents

References and Citations

Any statement where an observation, derivation, or argument has been previously reported should be accompanied by the relevant citation. References should be clearly relevant to the manuscript and should not contain unnecessary self-citations or citations to other authors or journals for the purpose of boosting citation metrics. Per COPE’s guidelines, this constitutes ethical misconduct and may result in the rejection of a manuscript.
If an author suspects that a peer reviewer or editor is requesting citations to their manuscript to inflate citation metrics, they should notify the Editor or Publisher of their concern and should never add recommended citations unless they are both relevant and necessary.

Plagiarism

Plagiarism is the unauthorized and unacknowledged use of someone else’s words, ideas, processes, or results in a way that misleads others into thinking they are original to the author.

The severity of plagiarism ranges across a broad spectrum. Plagiarism is most serious when intentional but may also result from honest error, lack of awareness, or poor judgment. Regardless of intent, it can harm an author’s reputation and career and may carry legal consequences when copyright is violated.
The most effective way to prevent plagiarism is through accurate attribution and citation of all sources used in one’s writing. Although citation styles may vary across scientific disciplines, their purpose is to direct readers to the original source of the information used.

Best Practices

When deciding if attribution or citation is needed, authors should keep the following in mind:

  • Always acknowledge and include a citation for the source of any ideas, evidence, and supporting data that contributed to the work, even if it is the author’s own previous work (see Text Recycling).
  • When using verbatim sections of text from another work, authors should clearly indicate this by using quotation marks or block quotes in addition to citing the source.
  • When paraphrasing or summarizing the work of others, authors should cite the original work and clearly indicate where their own thoughts and expression begin.
  • Authors should cite all relevant papers in the field that influenced the current research. Failing to conduct a comprehensive review and omitting key references to prior work may be perceived as plagiarism, particularly when earlier studies have made original contributions to the topic.
  • Always cite the primary source and not subsequent sources that have been derived from the primary source.
  • Authors should ensure that any source cited meaningfully supports their argument, hypothesis, or statement, and that it has been interpreted accurately.
  • Citations must be clear and comprehensive enough to allow readers to access the original work by including archived links or DOIs when possible.
  • Credit should be given and permission obtained for any material reused from unpublished work.
  • When adapting figures and images, an author must obtain permission from the original copyright holder and provide documentation of that permission. Making cosmetic or incidental modifications is not enough to constitute a new work for the purposes of copyright.

All manuscripts submitted to AIP Publishing journals are automatically screened through Crossref Similarity Check, powered by CrossCheck iThenticate to identify content overlap with other works. These reports are carefully reviewed by the Editorial team to determine whether the overlap is acceptable. Editors may send the manuscript back to the authors to address the overlap or reject the manuscript for violating AIPP’s policies on plagiarism.

Text Recycling

Text Recycling (also Self-Plagiarism) is the practice of authors reusing portions of text from their own previous publications without formally quoting or citing them. All work that can be drawn on, even one’s own, should be properly attributed and cited.

  • Text recycling in hypotheses, results, or discussion can mislead the reader by presenting ideas as original when they are not.
  • Text recycling in methods may imply that a method is new when it is not.
  • Text or image recycling anywhere may violate another publisher’s copyright. While some degree of overlap may be necessary, this should always be properly cited and annotated to make it clear what comes from earlier papers and what is new.

Best Practices

The overarching guideline to avoid harmful text recycling is that it should always be clear where ideas, methods, and even particular phrasing come from. Any element not original to the current article needs to be identified as such to allow readers an accurate view of the present work.

Duplicate Publication and Dual Submission

Duplicate Publication (also Redundant Publication) is the publication by an author or group of authors of more than one article that is identical in the hypothesis, results, discussion, or conclusions, although there may be superficial differences in the articles (such as a revised title and/or abstract, changed author list, or redrawn figures). Duplicate publication can be a type of text recycling/self-plagiarism or the result of dual submission.
Dual Submission (also called simultaneous or double submission), is the submission of manuscripts that are identical in the essentials (as described for duplicate publication) to multiple journals at the same time or while the manuscript is already under consideration with a journal. Dual submissions will be rejected.

Dual/Duplicate Publication Special Cases and Exceptions

Re-publication in another language of articles that are scientifically similar or identical is permissible in some cases, if the editors deem that re-publication in another language will allow access to significant scientific results by a considerable number of readers for whom the results of the article would otherwise be poorly accessible. The publisher of the original article(s) must be notified and give permission, and the author(s) must obtain the consent of the editor of the new publication. If the article is republished, the secondary publication must make it clear to readers via a statement at the end of the abstract, and the original article should be cited in the references.
Theses and Dissertations are exempt from the restrictions of duplicate publication because they are intended for limited circulation. However, if the thesis or dissertation is published under a copyright agreement, authors must ensure that any subsequent publication in a journal does not violate these terms, including the use of data gathered during the thesis research. For authors who wish to reuse content from their own AIPP article in their thesis or dissertation, more information can be found here.
Preprints are also exempt from the above restrictions of duplicate publication. AIP Publishing allows posting of preprints on noncommercial preprint servers, such as arXiv, bioRxiv, and ChemRxiv. Please see our Web Posting Guidelines for more details. Authors are asked to provide the DOI of a preprint posted on a server upon submission. Submitting to a publication or platform that provides post-publication peer review is not considered a preprint and precludes submission to any AIP Publishing journal.

Salami Slicing and Segmented Publication

Salami slicing (also data fragmentation) and segmented publication occurs when the results of a single study are published across multiple articles when a single article would be appropriate. This is usually combined with a failure to provide information about other publications or submissions from the study in question. It can be difficult to distinguish between legitimate reasons for producing multiple articles from one study and attempts to falsely increase publication output.
Valid reasons include scenarios such as

  • Publishing noteworthy preliminary results in a short-form format (such as a Letter) and publishing complete results later in a full-length article
  • Interesting or unexpected results may cause researchers to continue the study or to reformulate the model being used, leading to additional publications

As these examples show, multiple publications can be a natural part of the research process and within the expected bounds of publication practices. In contrast, segmented publication results from an intentional effort to inflate a researcher’s publication record. Segmented publication is unethical because it can mislead readers into believing that multiple studies were conducted, can result in duplicate (redundant) publication, and does a disservice to readers by causing them to search through multiple articles to get a complete description of a study and its results.

When salami slicing is suspected in a submission, depending on the extent of overlap, transparency by the authors, and amount of new content, Editors may return the manuscript for revision or reject it.

Best Practices

When unsure about whether a submission may be viewed as a segmented publication, authors should avoid ethical issues by properly citing and disclosing to the Editor any other submissions and prior publications that are relevant to the newly submitted work.

Data Fabrication, Falsification, and Image Manipulation

Research integrity requires reporting conclusions based on accurately recorded data for all relevant observations. Data fabrication, falsification, and inappropriate image manipulation represent serious research misconduct as they present false or manipulated data and scientific results.

Fabrication refers to making up results or data, recording or reporting them, and presenting them as if they were real. This may occur when:

  • A researcher falsely claims that an experiment or laboratory procedure was conducted when it was not, or when no study was performed at all.
  • A researcher copies results from previous work and publishes them as original findings.

Falsification refers to the alteration of research materials, processes, protocols, or equipment, or changing or omitting data or results. Falsification includes:

  • Removing outliers without a stated reason and without confirmation by an approved statistical test
  • Disregarding inconvenient or conflicting data and results
  • Selective reporting of findings
  • Omitting and changing data points
  • Reporting of anticipated results that have not been observed at the time of submission
  • Image manipulation outside of the proper and acceptable practice for improved readability (see below)

Image Manipulation: When is it appropriate to edit images?

AIP Publishing refers to the guidelines on Digital Images and Misconduct by the Council of Science Editors (CSE) and the STM’s Recommendations for handling image integrity issues. Authors should retain all original image files and must make them available for assessment if needed.

Acceptable:

  • Images can be technically improved for readability. Adjustments of brightness, contrast, and color balance are proper and acceptable when applied to both the whole and the control image provided that they do not obscure, eliminate, or misrepresent any information present in the original.
  • When using any image manipulation, authors should report at the time of submission how the image was manipulated. Authors should mention in the figure caption, and/or refer to the text of the article in the caption any image manipulation for the benefit of the readers.
  • The grouping of images from different parts of the same sample, or from different samples, fields, or exposures, must be made explicit by the arrangement of the figure (e.g., dividing lines) and in the text of the figure legend.

Unacceptable:

  • Any image adjustment that affects the interpretation of the data is considered fraudulent.
  • No specific feature within an image may be enhanced, obscured, moved, removed, or introduced.
  • In cases of questionable figures, if the original data cannot be produced by an author when asked to provide it, acceptance of the manuscript can be revoked.

Best Practices

All coauthors of a submitted manuscript are encouraged to:

  • Preserve the integrity of the research data record, i.e., maintain a clear and complete record of the acquired data and all processes and characterizations performed. Authors should refer to their institution’s practices for proper and acceptable data keeping, whether in the form of hardcopy or electronic laboratory notebooks.
  • Discuss and explain the way data have been collected, processed, and analyzed.
  • Understand acceptable, unacceptable, and fraudulent practices of image manipulation.
  • Review the images included in the manuscript prior to submission and compare them to the original image.

Maintaining the Integrity of the Publishing Process

Researchers must not engage in any form of misrepresentation or manipulation during the publishing process, including but not limited to

  • Using third-party services to create, submit, or review a manuscript
  • Recommending peer reviewers with whom they have an existing personal or professional relationship
  • Collaboration with others to inflate citation metrics or exchange positive peer reviews
  • Inappropriate authorship practices, such as the deliberate inclusion of individuals who do not meet the criteria for authorship

AIPP uses both automated and manual checks during the publication process to assess the legitimacy of manuscript content, peer review reports, and authorship. If AIPP finds that the publishing process is manipulated or compromised, corrective actions will be taken, such as

  • Rejection or withdrawal of manuscripts that fail integrity checks
  • Removal of reviewers who provide fake or biased reports from the peer review process
  • Retraction of articles published through fraudulent and unethical practices

In certain cases, AIPP may decline to share detailed information with involved parties regarding integrity checks and findings, such as when manipulation of the publishing process is suspected.

Human and Animal Ethics

For any manuscript describing research involving human participants or animal subjects, the Author Declaration must include a statement that the authors obtained ethics approval (or a statement that it is not required). This statement must include the names of the ethics committee or institutional review board (IRB), approval ID numbers, and a statement that any human participants gave informed consent before participating in the study.

Ethical Conduct of Research using Human Participants

AIPP adheres to the standards used by international organizations such as WMA Declaration of Helsinki-Ethical Principles for Medical Research Involving Human Participants and expects authors to do the same, for example,

  • Approval from an IRB or ethics committee must be received before starting the study and the appropriate protocols followed
  • Authors must obtain informed consent from human participants for study participation and publication
  • Authors must maintain the confidentiality of human data unless explicitly given permission to share or publish these data

Additionally, AIPP refers to COPE’s position on ethical standards for research related to vulnerable groups and individuals.

Animal Testing

Any study that involves the use of live animals must include a statement in the Methods section describing the relevant laws and institutional guidelines that were followed, including if the study was exempt from institutional approval or approval was not required. Failure to meet these requirements or to otherwise adhere to ethical standards in the use of animal test subjects may be grounds for rejection.

Authors who would like more guidance on the accepted standards for treatment of animals in research should consult the following resources:

International Guiding Principles for Biomedical Research Involving Animals

Guidelines for Animal Research: Reporting of In Vivo Experiments (ARRIVE)

Please be aware that individual journals may have specific requirements listed on their editorial policies page.