Policy on Research Integrity

1. Principles of Research Integrity

As a world-class comprehensive university with research activities spanning all major disciplines, the University of Hong Kong realises the vital importance of research integrity. All members of the University must observe the highest standards of professional conduct and must comply with the following principles of research integrity in pursuing their research activities:

honesty in the conduct and communication of research;
objectivity and openness;
duty of care;
fairness in giving credit and appropriate acknowledgement; and
responsibility for nurturing researchers of future generations.

All researchers of the University are committed to the principle of honesty in conducting research and in communicating research findings to the research community and the public. Honesty is required in presenting research goals and intentions, and in reporting procedures and findings. Such presentation and reporting must be full and fair. Objectivity of research requires maintenance of accuracy in the collection and reporting of data. Conclusions must be based on facts capable of verification, and researchers should be impartial and transparent in the handling of data. Research findings should be made accessible to the research community for verification. All researchers have a duty to care for the human participants, the animals, and the environment under study. They must be fair in giving credit for the work of other researchers who participate in the research. They have a responsibility in supervising and nurturing research students and young researchers, who will be researchers of future generations. Positions of seniority or responsibility should never be abused so as to put pressure on colleagues or research students, for example, to forego their right to proper acknowledgement of their contribution to the research or publication in question.

In pursuing their research activities, members of the University should adhere to good research practices; and should not be engaged in research misconduct such as plagiarism, fabrication, falsification or unauthorised use of data, improper ascription of authorship, and non-disclosure of potential conflict of interest. Relevant ethical approval must be obtained before the commencement of data collection. Misconduct or alleged misconduct in research will be referred to the relevant authorities for follow up, and may result in disciplinary action.

2. Good Research Practices

In pursuing their research activities, all members of the University should commit to adhering to good research practices, which reflect the principles of research integrity. It is essential that the design of research projects follows all relevant ethical guidelines, and researchers should only participate in work which conforms to accepted ethical standards and which they are competent to perform. When in doubt they should seek assistance with their research from their senior colleagues or peers. Research postgraduate students should refer to the book Research Integrity published by the Graduate School. Members of the University may also refer to the following documents, from which many of the ideas of this document are taken:

1.	Code of Conduct for Research Integrity (European Science Foundation, March 2011; from the website: http://www.esf.org/activities/mo-fora/research-integrity.html);
2.	ORI Introduction to the Responsible Conduct of Research (Revised Edition, August 2007), by Nicholas H. Steneck, published by the Office of Research Integrity, USA (http://ori.hhs.gov/documents/rcrintro.pdf);
3.	Singapore Statement on Research Integrity, developed at the 2nd World Conference on Research Integrity, July 2010, in Singapore, as a global guide to the responsible conduct of research (http://www.singaporestatement.org).

2.1   Responsible conduct of research

All research should be conducted in a responsible manner. The design of the research should be well-considered, and research procedures should be carried out with care. The most appropriate research methods should be employed, and conclusions should be based on critical and unbiased analysis of factual evidence. Research findings should be reported fully and in an objective manner.

The design of the research should aim at maximizing the benefits and reducing the risks to participants of the research. There should be an appropriate balance between benefits and risks, and the well-being of the participants involved, especially those from vulnerable groups (such as young children and the mentally disadvantaged), should be protected. Rights of participants to make decisions for themselves should be respected, and researchers should not exert undue influence on participants’ decisions. Researchers should avoid conduct that may directly influence or distort research results. In collecting information, researchers should be sensitive towards differences in ethnic origin, culture, social class, religion, gender and age, and the information collected should be used only for the purpose of the investigation. Data collection through deception should be avoided, and if it is absolutely necessary for deception to be employed, the researcher must ensure that no more than minimal risk is involved. Participants should be informed of the deception after data collection, and the data may be used only with the participants’ consent after the event. The privacy of participants must be protected and all personal information should be treated in a confidential manner.   In particular, so far as personal data are concerned, the Personal Data (Privacy) Ordinance and the data protection principles in Schedule 1 of that Ordinance should be complied with. If there is any doubt about issues regarding personal data, assistance and advice should be sought from the University’s Data Protection Officer.

2.2   Publication-related conduct

Research results should be published in a timely manner, and publications should be based on objective analysis and interpretation of accurate research data in an honest and transparent manner. Authorship should only be based on contribution to the research proper, including contribution to the design of the study, data collection, data analysis, and reporting; and should not be for merely administrative roles. Guest authorship (i.e. including authors who have not contributed to the research) or ghost authorship (i.e. not including individuals who have contributed) are not acceptable. All authors take full responsibility over all the content of the publication, and if they are responsible only for specific parts of the research or publication, this should be clearly specified in the publication as appropriate. The criteria for the order of authors appearing in the publication should take into consideration the relative contributions of the authors or prevailing international practice of the discipline, and should as far as possible be agreed by all involved at the beginning of the research.

Where appropriate and with their permission, names of individuals or organisations which have made significant contributions to the research and the roles they played in the project should be acknowledged in publications. These include funding agencies, sponsors, and research collaborators and assistants who do not meet the authorship criteria. Important works on which the research is based, and other academics who have contributed intellectually to the research should be appropriately cited or acknowledged. Publication of the same (or substantial parts of the same) work in different journals is acceptable only with the consent of the editors of the journals and where proper reference is made to the first publication. In the author’s CV such related articles must be indicated as such and not give the impression that they are distinct research outputs when they are in fact the same. Articles published in special/symposium issues should be clearly specified.

In reviewing manuscripts submitted to journals or other publications, confidentiality must be observed. Editors and reviewers should never make use of the writing or the data in the submitted manuscripts without the explicit permission of the author.

In communicating research findings to the general public (for example, through popular media), the same standards of honesty, objectivity and fairness should be observed. The significance or the practical use of the research findings or products, as well as the limitations of the findings, should be presented in a truthful manner, and the reported findings should not be misleadingly selective or exaggerated.

2.3   Care and safety

All human and animal research subjects must be treated with care and respect. In the design of the research, sufficient provisions should be made for ensuring the safety of the human participants in the data collection. Risks and distress to participants should be estimated and kept to a minimum. The safety, health and welfare of the community and of all people relating to the research must be safeguarded.

Animals should be used in research only when alternative means have been found to be not viable or adequate. Where possible, the use of animals in research should be replaced by other methods such as computer simulation. If research involving animals is really necessary, the research design should aim at reducing the number of animals used to obtain the same amount of information, and no more than the minimum number of animals necessary to ensure scientific and statistical validity should be used. Harm or distress inflicted upon the animals must be outweighed by the benefits expected from the research, and measures should be taken to keep such harm or distress to a minimum.

The physical and cultural environment should also be respected and treated with care, and measures need to be taken to minimize any necessary harmful impact on the environment. Researcher should take the sustainability of environmental resources into consideration in the design of the research. All resources should be utilized and deployed efficiently, and waste should be minimized.

2.4   Proper data handling

Research data must be collected in an ethical manner, and there should be clear record on the data collection process. Data should be kept in a secured and accessible form, preferably in more than one medium and at more than one location, and be documented and archived for a substantial period of time (at least 5 years, and preferably 10 years) to allow for verification and replication by other researchers. The University’s Policy on Management and Sharing of Research Data must be observed.

Individual researchers should be able to hold copies of the data for their own use. Nevertheless, it should be understood that retention solely by the individual researcher provides little protection to the researcher or the University in the event of an allegation of falsification of data. In some cases there may be restrictions on the retention of data obtained from limited access databases or in a project carried out under a contract. In such cases, a written indication of the location of the original data or key information regarding the limited-access database from which it was extracted must be kept in the department or research unit.

Under the Personal Data (Privacy) Ordinance, researchers must ensure that the personal data of participants are properly protected, and disclosure is not allowed without the consent of the participants concerned. If data of a confidential nature are obtained, for example from individual patient records or by questionnaires, confidentiality of the information must be observed to protect the informants. The methods of data acquisition and the use to which the data will be put must be approved by the appropriate Ethics Committee (the Human Research Ethics Committee for Non-Clinical Faculties (HRECNCF), the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (IRB), or the Committee for the Use of Live Animals in Teaching and Research (CULATR)). The researchers must not use such information for their own personal advantage or that of a third party. In general, research results and methods should be open to scrutiny by colleagues within the University and, through appropriate publication, by the profession at large, provided that the confidentiality of individuals is adequately protected, especially in accordance with any laws relating to data protection.

2.5   Disclosure of conflict of interest

In order not to jeopardize the trustworthiness of research results, any relevant or potential conflict of interest - whether personal, financial, academic or political - should be identified and declared. Conflict of interest should be declared in research proposals, publications or other forms of dissemination of findings, etc.

In reviewing research proposals and editing or reviewing research publications, a reviewer or an editor who has a relevant conflict of interest should abstain from the decision making process. For potential, minor or unavoidable conflict of interest, it should be unambiguously declared.

2.6   Compliance with relevant regulations

All experiments or other forms of research project, where the University has an interest, involving consent of human participants and the use of vertebrate animal subjects must obtain prior approval of the appropriate Ethics Committee. Informed consent of human participants should be appropriately documented. Clients or sponsors should be alerted to the ethical and legal obligations of the researcher, and to the possible restrictions this may impose on the research.

Research should, of course, be conducted in compliance with applicable laws, safety and health guidelines, which are designed to ensure safety and the welfare of human participants and laboratory animals.

2.7   Contract research

Secrecy may be necessary for a limited period in some contract research. However, that necessity must be clearly stated in the contract and fully understood by all parties. Where a research student is involved in working on such a contract, the researcher concerned must ensure that the party sponsoring the research understands and accepts unequivocally that the student’s thesis will not be restricted from publication, though there may be scope for not releasing certain types of data confidential to the sponsor. Consultation with the Policy Board of Postgraduate Education should be undertaken before any restrictive agreement is reached. Researchers conducting contract research must ensure that the policy on Use of the University’s Name and Visual Identity is complied with to avoid commercial exploitation of the University’s name. If the contract involves intellectual property, the University’s Intellectual Property Rights Policy should be adhered to, and the Technology Transfer Office (TTO) should be consulted.

2.8   Reporting irresponsible research practices

Members of the university should report to the authorities concerned any research misconduct or suspected misconduct (refer to the document Procedures for Dealing with Alleged Staff Misconduct in Research). This includes plagiarism, abuse of data, improper ascription of authorship, non-compliance with regulations, and other forms of improper research practices that are deemed to be unacceptable by the academic community.

2.9   Special needs in different disciplines

In some disciplines there may be special areas which require specific regulation, for example, animal and human experimentation and the handling of hazardous materials. The rules for these activities must form part of the general policy of ethics and safety for each department. The University Director of Safety should be consulted in these cases.

3.   Research Misconduct

Research misconduct is unethical practices that seriously deviate from those that are commonly accepted within the academic community for proposing, conducting, or reporting research. It can appear in many forms, and may lead to disciplinary action. The following list only intends to provide a general indication of the most obvious kinds of misconduct and is not exhaustive.

3.1   Plagiarism and Self-plagiarism

Plagiarism is direct copying of textual material or wilful use of other people’s data and ideas, and presenting them as one’s own without acknowledgement;
Self-plagiarism is reuse of one’s own data or previously written work in a ‘new’ publication without acknowledging that the data set has been used or written work has been published elsewhere.

References to what could constitute plagiarism may be found in the booklet What is Plagiarism?

3.2 Abuse of data – this includes

Fabrication of data – making up data and/or results where none have been obtained, and publishing them as if they were real;
Falsification of data – manipulating research materials, equipment, or processes, or changing or omitting data or results such that the research is not accurately represented in the research record;
Unethical collection of data – collecting data through exploitation of vulnerable or disadvantaged groups, or unnecessarily infringing upon the privacy of participants; the data collection process harmful to participants, or putting them to more risk than necessary;
Unauthorised use of data – infringing the data ownership rights of others, or using data involving human participants without their informed consent.

3.3 Improper Ascription of Authorship

The over-riding principle for authorship of a research output is the intellectual contribution to the research process and not merely administrative involvement. Author and co-authors should have significant participation in conceiving, executing or interpreting at least part of the research reported. The research team should agree on which individuals should be named as co-authors, and the order in which their names appear in publications.
One particularly serious offence is when senior staff (such as heads of department or supervisors) coerce colleagues or students into allowing the former to take the credit of the research in question as their own, either wholly or partly, and not acknowledging or giving proper credit to the latter. This is a failure of leadership and of moral responsibility.
Misleading ascription of authorship includes the listing of authors without their permission, attributing work to those who have not in fact contributed to the research, and the lack of appropriate acknowledgement of work primarily produced by a research student or any associate. Due recognition of all participants is an important part of a proper research process. Authors should ensure that the work of research students, research assistants, and all support staff is properly acknowledged. It does not matter whether the researchers were employed or otherwise paid for their work.
Each author must endorse the whole work. The authors of the research output should read the final paper and agree that each of them has met the minimum requirements for authorship. It is unethical to claim authorship without reading and approving the final draft in its entirety. All of the authors are equally responsible for the contents of the research output; if the contents are bogus then all authors carry the blame. Responsibility cannot be shifted from an academically senior author to an academically junior one, and vice versa.
The unattributed re-presentation of any research output whether for research or teaching in a language other than the original is unacceptable.

3.4 Non-Disclosure of Potential Conflict of Interest

Disclosure of any potential conflict of interest is essential for the responsible conduct of research. Non-disclosure is regarded as unethical behaviour.
A researcher’s affiliation with, or financial involvement in, any organisation or entity with a direct interest in the subject matter, or in the provision of materials for the research, must be included in a full acknowledgement.
Members of committees responsible for the allocation of research or conference grants should not participate in any way in the determination of their own applications, or normally those of students whom they supervise.
The source of funding for research work should always be acknowledged, unless the donor requests anonymity and such request is approved by the University.

3.5 Non-compliance with regulations – this includes

Commencing data collection for research involving human participants or animals before ethical approval is obtained from the appropriate Ethics Committee;
Failure to comply with the conditions of the ethical approval;
Failure to comply with the contract terms;
Research not reasonably conducted according to the approved research design;
Infringement of another person’s copyright, patents, trade-marks, computer software, etc., with respect to any form of research output (refer to the Intellectual Property Rights Policy of the University);
Breaching of the relevant laws, or safety and health guidelines