WARNING - the Computer Science ethics system is obsolete

Ethics projects which have already been submitted will be processed.

New projects WILL NOT be considered.

Use the central University Ethics system instead.

For Researchers

Principle Investigator / Independent Scholar / Other Researcher: There are some issues that really only relate to PIs CI's and the like, and so if you are one of these, this is the page for you!

It is your job to make sure your research projects are entered into the ethics system. This is because the university wants to keep a record of all projects, as part of the process you should also make sure that you check the 'participants' box if you are using human participants at any point.

It maybe that you have decided to not get ethical approval even though you are using participants in your study. If this is the case you should be aware that in this case you are not legally covered or insured and will technically be liable in the event of anything going wrong.

If anything goes wrong contact the CS Ethics Liaison immediately.

In General, you don't need to know everything but keep in mind:

In summary then, we can see that the following list of key principles should be taken into account in any research resign be it in the field, quasi-experimental, or within the laboratory.

Competence: Keep up to date, know your limitations, ask for advice;
Integrity: Have no axe to grind, or desired outcome;
Science: Follow the Scientific Method;
Respect: Assess you participants autonomy and capability of self-determination, treat participants as equals, ensure their welfare;
Benefits: Maximising benefits and minimising possible harms according to your best judgement, seek advice from your organisations ethics committee;
Justice: Research should be undertaken with participants who will benefit from the results of that research;
Trust: Maintain trust, anonymity, confidentiality and privacy, ensure participants fully understand their roles and responsibilities and those of the experimenter; and finally
Responsibility: You have a duty of care, not only to your participants, but also to the community from which they are drawn, and your own community of practice.

Remember: If you wish you can also use the manual paper versions of the application form; instructions are also available.

Useful Links

Our CS Ethics Cue Card: Use this cue card to guide you through the kind of questions you need to think about when making an application;
Central University Ethics Resource: Look at the University resource for a more general guide;
Make an Ethics Application: Make your application online (this is also printable too);
Ethics Tweets: Any problems or system errors - have a look at our system tweets; and
Ask a Question via the STAFF ONLY Computer Science Q and A System: Ask other staff members; and
Ask a Question via the Computer Science Q and A System: Still unsure? Ask us - and everyone else - a question for clarity.

Example Application from the SCWeb² Project - Thanks to the Leverhulme Trust (F/00 120/BL)

Application for Ethical Approval. The proposal was submitted for approval to the Central Ethics Committee as we wanted to work with human participants, and further, those over 65 years of age;
We attended the panel in which 10 committee members questioned us for approximately 15 minutes. This was helpful as they were each experts in their field and had different views on the kinds of methods we should employ along with the kinds of analysis that should take place and the most appropriate statistical methods to use;
Panel Decision. A letter giving the panels decision took around a week to reach us - and there where only three minor things that needed to be changed;
We made these changes and submitted them back to the Ethics Committee Secretary (Dr. Stibbs) who approved them without further recourse to the committee;
Ethics Approval. We received the approval letter, again, within about a week; and
Started work about three weeks later - with the final data collection being finished in October 2010.

Guidelines

The American Psychological Association's (APA), 'Ethical Principles of Psychologists and Code of Conduct';
The United States Public Health Service Act (Title 45, Part 46, Appendix B), 'Protection of Human Subjects';
The Belmont Report, 'Ethical Principles and Guidelines for the Protection of Human Subjects of Research';
The Council of International Organisations of Medical Sciences, 'International Ethical Guidelines for Epidemiological Studies'; and finally
The World Medical Association's, 'Declaration of Helsinki - Ethical Principles for Medical Research Involving Human Subjects'.

Ethical Principles... in a little more detail

Each set of research standards proposes a slightly different set of principles by which ethical research is to be conducted. However, there are a number of commonalities which are covered throughout all of these standards. It is these commonalities that make up the seven specific principles which will be outlined in more detail in this section. These principles are mainly culled from the American Psychological Association's ethical guidelines along with the ethical guidelines proposed by the Beaumont Report. This is mainly because they are repeated, and in some cases expanded upon, in other more recent work, however, the core principles remain the same.

You Must Be Competent: This first principle relates not so much to the methodology that will be used, or the direct ethical considerations of how that methodology will affect the participants. Rather, it relates directly to the competency of the computer scientist designing the research and with those who will deliver that methodology and analyse the resultant data (in the case of double-blind trials). This places the onus on the computer scientist to make sure that they do not overstep the bounds of their competency, and indeed, that they understand the domains under which, and principles to which, they were trained. If there is an area for which you are not fully trained or that does not have recognised professional standards, you should exercise careful judgement and take appropriate precautions, in some cases be overly cautious, to protect the welfare of those with whom you work and the participants with whom you will be conducting the research. You should also make sure that you maintain your knowledge of the relevant scientific research area and information relating to your computer scientist specialism. This may be through membership of professional and governing bodies; in the United Kingdom this may be through the British Computer Society (BCS), in the United States the Association of Computer Machinery (ACM) or the Institute of Electrical and Electronic Engineers (IEEE), or in Europe via one of the chartered engineering Institutes. Indeed, each country and region normally have their own governing body for computer science; and human computer interaction as a sub discipline. However, in the absence of such a governing body guidance from one of the various experimental psychology associations, most notably the American Psychological Association, may be useful.
You Must Have Integrity: It is often easy in CS evaluations, or scientific work in general, to have a preconceived notion of the desired outcome. Indeed, I use the term 'desired' deliberately in this context, to show how easy it is to influence work in a way which supports the investigators preferred outcome (there should be no desired outcome). Removal of this desire is one of the main reasons why the 'blind method' is used when conducting trials. Blinding is a basic tool to prevent a researcher's cognitive bias causing them to unconsciously influence the participants of an experiment. It is a significant threat to a study's internal validity, and is therefore typically controlled using a double-blind experimental design. A double-blind trial describes an especially stringent way of conducting an experiment in an attempt to eliminate subjective bias on the part of both experimental subjects and the experimenters. This use of controls recognises two things: firstly, that there may be an implicit bias imposed by the computer scientist; but also, that the computer scientist, or indeed the company for which they work, may require a certain outcome. It is this final point which needs to be moderated with regard to the ethical procedures of the work, and this is why integrity is such an important part of the process; and a key component of ethically sound methodologies. Striving for personal research integrity enables you to understand: (1) that you must not be concerned with the resultant outcome of the work; (2) that both a seemingly positive, or negative, outcome with regard to an evaluation is valid and tells us more about the interface or the system than we knew before; (3) that by trying to create a valid experiment, free of bias in all aspects of possible influence, enables your work to actually have meaning; and that (4) being driven only by an incorrect desire to validate a bad system or interface will, in the end, not address the needs of the user, or the long-term business case of the organisation conducting the trials.
Conform to Scientific Principles: Belief can alter observations; those with a particular belief will often see things as reinforcing their belief, even if to another observer they would appear not to do so. The scientific method is a body of techniques for investigation and knowledge acquisition which removes the need to believe through the application of empiricism. To be scientific, a method of inquiry must be based on the gathering of observable, empirical, measurable, and refutable evidence, and be subject to specific principles of reasoning. The essential elements of the scientific method are observations, definitions, and measurements of the subject of inquiry; theoretical, hypothetical explanations of observations and measurements of the subject; and reasoning including logical deduction from the hypothesis or theory; the principle means of validation for all of these elements is testing by experimentation which produces observable results. This means that by conforming to scientific principles the research methodology can be validated within a well understood framework. The scientific principles discussed, ensure that the evaluation and testing of the interface and system is performed to maximise the validity and generalisability of the results. Without these safeguards, built into the scientific method, there can be no reliable test that the resultant outcomes represent a truthful understanding of the users interactive behaviour. Therefore, if you do not conduct your evaluations in a scientific manner then your results may be incorrect, and the stress, which you have placed your participants under, will have been for nothing.
Respect Your Participants: At all times you should show a high degree of respect for your human participants within the research setting, including treating them with dignity and respecting their autonomous choices. You must not presume that you know better or try to elicit responses from your participants with which they do not agree. Not every participant will be capable of self-determination or self autonomy. They may exhibit a diminished capacity in some regard, for instance they may be very young, or have some cognitive or learning difficulties. In this case you should still treat them with the utmost respect. In addition, if you have any suggestion that your participant has some form of diminished autonomy, or that they do not fully understand any aspect of the study, or that they cannot give proper consent, you must look for other participants or the agreement of their guardian. As a computer scientist you must accord appropriate respect for the rights, dignity and worth, of all participants, this includes an individual's right to privacy confidentiality self-determination and autonomy being mindful that legal and other obligations may lead to inconsistency and conflict within sight of these rights. You should also be aware of differences being age, gender, race, and ethnicity and be specifically aware of any cultural aspects which may influence, or affect a person's dignity, or rights, or where culture may exert an implicit force when undertaking experimental research. As part of this respect you should also make sure that your participants have an appropriate understanding of the purpose of the research and why their participation is useful and beneficial. Additionally, you should make them aware of any monitoring or recording devices, within the experimental environment, and how the data collected from those devices will be used and stored. Finally, you should not apply any external force in the recruitment of participants and their participation within the study should be entirely voluntary. This means that at all times you must be more concerned with your participants welfare, and the welfare of the researchers undertaking the experimental work.
Maximise Benefits: The maximum for increasing benefits and reducing possible harms is that first proposed by the medical profession: 'do no harm'. Further, the Hippocratic Oath requires physicians to benefit their patients according to their best judgement. Maximising benefits and minimising possible harms is one of the most important aspects of research within the human sciences and more specifically within the ethical domain. At all times the experimenter must keep in mind the possible harm's that could occur within the experimental setting. In this case, most ethical committees require a risk assessment to be undertaken before the experiment can proceed. The main problem with understanding benefits is that they can seldom be estimated accurately in advance and in most research the benefit is to science as opposed directly to the individual, or community from which they are drawn. Because the cost to participants and the benefits to science cannot easily be understood, especially at the outset of the research, it falls upon the experimenter the designer of the research study to take a morally responsible decision. Luckily, the research ethics committee can also provide a secondary check regarding this, sometimes tricky, judgement. Finally, another factor to consider when assessing benefits and harms are not those as applied to the individual, but to the society or institution from which individuals are drawn. For example, with regard to studies involving some kinds of religious institutions; there is a very delicate balance between understanding the benefits to science, weighed against the harm that may come to the celebrants, of the religion, or the society which surrounds it, if through this research the institution is diminished or destroyed.
Ensure Justice: Threats to justice often occur due to be inherent power differential between experimenter and research participant. In this case, it is the duty of the experimenter (the computer scientist in this case) to make sure that this is not the case and that participant selection, proceeds were possible, within the populations or communities which will directly benefit from the proposed research. As has been discussed previously, research was often undertaken with participants who would not directly benefit from the results of that research. This means that participants may undergo a number of experimental methodological procedures, but as either individuals or as a subgroup, would never be able to reap the benefits of the resultant understandings gained from those experiments. We have already seen that one of the guiding motivations behind empirical work with human subjects is that the work should do more good than harm, or that more people should benefit than suffer. There are a number of accepted just ways of disputing burdens or benefits: (1) to each person an equal share; (2) to each person according to individual need; (3) to each person according to individual effort; (4) to each person according to societal contribution; and (5) to each person according to merit. Of course these distributions of justice may be difficult in the real world especially when participant selection is by random probabilistic methods. On solution would be to bound the sample frame, or the population under investigation, such that the population from which the representative random sample is drawn receives the benefits and burdens equally. In addition, there should be appropriate procedures to ensure that experimenters, assistants, and participants alike have adequate access mechanisms to address any possible concerns regarding the research itself. So then, in CS, the application of justice is markedly easier than in other subjects. This is mainly because the participants within the study are drawn directly from the population to which that study is applicable; there are normally, no invasive procedures, or procedures which might cause psychological distress or harm and therefore the benefits often outweigh the burdens. The only real burden within an CS evaluation is that of time and effort, and these could be suitably compensated with a financial incentive.
Maintain Trust: Researchers must maintain a high degree of trust between themselves, the participants, and the experimenters who will be enacting the experimentation. This trust is based upon the agreement about what will, or will not, be involved within the experimentation, and how the data collected from that experiment will, or will not, be used. Aspects of confidentiality and privacy should be addressed from the outset and this should include the anonymising of individual user data and the protection of the identity of the participants. This protection must also be applied over data which may, in combination, be used to identify of the participant. For instance, data which suggests a male, over 85, being treated for a heart condition at Manchester Royal Infirmary, first admitted in June 2006, may enable the identity of that user to be derived. While some HCI specialists may not see the importance of maintaining privacy and confidentiality within the user data itself it is important to understand that the use to which this data may be put, by unscrupulous individuals, is often not self-evident. As well as privacy and confidentiality, trust must be maintained throughout the process so that the participants fully understand their roles and responsibilities and those of the experimenter. This enables them to feel safe and comfortable within the research setting so that a more accurate experiment can be undertaken. If a certain level of trust is not maintained then the participant may become agitated or stressed and these two factors may become a confounding variable within the experimental research.
Social Responsibility: Social responsibility is final principle discussed here and in some respects is not directly relevant to the practice of human computer interaction. However, the computer scientist should be aware of the professional and scientific responsibilities to both the community, and the society, in which they work and live. This includes not just the local organisation or institution of which they belong, but the wider population within the area and also the more distributed community of practice of which they are part. Indeed, if the computer scientist brings disrepute onto the CS community by unethical conduct, then the cause of CS, in general, is diminished. As we have already seen, professional ethics, competency, and integrity is key to the practice of CS, and such is its importance that the professional bodies representing the CS specialist take professional conduct very seriously. More importantly we have, as computer scientists, a responsibility to the society in which we live and work. Again, when undertaking scientific research we must strive to advance human welfare and the science of human computer interaction, we must strive to avoid the misuse of our work, or the misunderstanding of the results of that work (this may also be in a journalistic or media setting), we must comply with the law, if the moral and ethical compass of the practitioner agrees. And we must encourage the development of the law and social policies that serve the interests of humans interacting with computers from a software, systems, or ergonomic perspective.