Guidance Notes for Reviewers and Policymakers on Covert, Deceptive and Surveillance Research

(1) Do Not Assume These Research Approaches to be Inherently Unethical

For the reasons discussed throughout this collection and for those summarised below, these approaches to research cannot be assumed to be inherently unethical. The ethical principles by which research must be judged depend on answers to a range of questions such as:

• (a)

  Who has a right to know the information that is being obtained?

• (b)

  Who has a right to control access to that information?

• (c)

  Whose interests must be protected and why?

It cannot be assumed that all information is the private property of the individual from whom it is obtained. In some circumstances, the information is public, and withholding information about it may be unethical in itself.

The use of deception in research is widely thought to be problematic. The very word ‘deception’ implies an intention deliberately to mislead people who are participating in research. Many writers have supposed that covert research is intrinsically deceptive. Covert research is simply research which is not fully disclosed, and there are many situations where complete disclosure will not take place. That is not necessarily ‘deceptive’. Several chapters in this open access volume have discussed compelling ethical reasons why full disclosure is not ethically required for particular types of research. Those who assume covert research to be deceptive have generally been influenced by a particular class of sociological research, covert participant observation, which has at times led into illegal, unethical or at least ethically questionable activities. The examples they cite have included participation in sexual activity in public toilets, football hooliganism, neo-fascist political organisations and gang culture. The researchers producing such evidence needed to offer compelling reasons for why some elements of deception and covert work were necessary in these situations.

(2) Do Not Treat Covert, Deceptive and/or Surveillance Research ‘Approaches’ as a Set of Homogeneous Practices

Covertly observing people in private may be a breach of their rights, although there may be cases where this is justified. Observing people in public is less controversial, although still may impinge on people’s reasonable expectations of privacy. Surveillance is not covert when ‘consented’ to by individuals seeking to purchase goods and services. And if the ‘purposes’ of surveillance are fully conveyed to those being surveilled it cannot be seen as deceptive. For these reasons, the details of what researchers plan to do and how they plan to do it have to be examined. Surveillance activities which are not necessary to the research should be removed, and efforts taken in all cases (covert, surveillance and deceptive) to mitigate potential harms which could arise. This may include, for example, pixelating faces of people when monitoring CCTV for learning about footfall patterns in public areas.

Deception has been widely used in research and in many different ways. Just as the term has a range of meaning in general conversation, it means different things in research contexts. Examples include the following:

• A medical treatment trial divides people into two groups, half of which unknowingly receive a placebo. For example, everyone might be told that they are receiving an experimental drug and then one group is given a placebo.

• A psychologist tells research subjects that the research is being done to examine one psychological trait, when in fact it is being done to examine another one. Subjects might be told they are experimenters, while accomplices may pose as subjects.

• An experiment is set up in which the naïve subject is thrust into interactions with the researcher’s accomplices, people who are party to the deception.

• A journalist secures an interview with a politician on the understanding that it will be used for a profile, when the actual intention is to subject the politician to public criticism.

• A researcher poses as a member of the public to discover how people are treated by an agency.

Only in the medical case is research being undertaken in the private sphere. It is possible to find examples of journalism which intrude on private life – for example, ‘blagging’, or pretending to celebrities’ service providers that they are authorised to get personal information – but such practices are unquestionably unethical. Furthermore, in deceptive cases, the research subjects should be informed of the truth after the research has taken place.

(3) Consider the Vital Role of ‘Context’

The ethical considerations to be applied to covert and/or surveillance research as well as any research involving deception must be assessed in terms of the context in which the research is conducted. This refers to the concepts of ‘situated ethics’ (where the ethical values and principles accepted in a specific situation or context must be given due consideration in any overall ethical assessment) and what is known as ‘researcher positionality’ (where the motives, intentions and skills of the individual researchers or organisations must be taken into consideration).

Most professional codes of guidance identify at least four main areas of ethical concern. These include:

• (1)

  the ‘research relationship’, which is the responsibility to the research’s funders or sponsors;

• (2)

  the relationship of the researcher to the research participants or to human subjects;

• (3)

  responsibilities to the researcher’s profession and to other researchers; and

• (4)

  responsibilities to the wider society.

The objections to these research approaches are mainly focussed on the second and third of these, but some stretch to the fourth. At the same time, justifications for covert research (including surveillance research) may well be given in the context of the fourth. In all cases of deception in research, the consequences in these four domains must be considered and assessed.

(4) Engage in a Full Examination of the Approach

Surveillance and/or covert research or any element of deception should be assessed in terms of: who is doing what to whom, in what situation and for what purpose? This is a way of expanding the context of the research actions being assessed. In all research, there is a possibility that the actions of the researcher will change the behaviour of the research subject or participant, and a process of full disclosure may defeat the object of the research. The effect of revealing one’s status as a researcher might be to alter the behaviour of the people being observed, and that would be self-defeating.

The presence of a researcher as participant might, for example, indicate in itself a level of support for the activity being researched (whether the other participants are aware of this researcher or not), such as acting as encouragement to anti-social behaviour (arising from a larger supportive group reducing inhibitions of the participants). None of those risks, serious as they are, is directly attributable to deceit as such.

Circumstances may arise where researchers mislead people innocently: for example, telling them that the research will be used to examine one problem, when it may turn out to be used for another; that the research will be uniquely part of one research project, when often data are recycled into others; or that the research will be used to improve treatment when it turns out that it does not do so. These are less ethically problematic than cases of deliberate deception, but proportionate efforts should be made to make the truth known to the research participants once the research concludes.

(5) Be Aware of the ‘Methods’ Available to Engage in Covert and/or Surveillance Research

The techniques available to conduct covert research or surveillance as well as ways of using forms of deception are developing all the time as the technologies develop. The ethical considerations to be borne in mind when using new technologies must keep up with developments in technology. Ethics appraisal must take into account the specific way that data are being gathered and the nature of those data. Some covert methods of data gathering include: a researcher directly observing the actions of subjects; a researcher indirectly observing the actions of subjects (such as via CCTV or the use of audio recording devices); a researcher participating in a group, community or organisation primarily to conduct research and not disclosing their research role to the members of the community; a researcher deliberately concealing some or all elements of the research (i.e., not ‘fully informing’ the subjects/participants under study); automated data gathering and analysis that is controlled by an algorithm. While remaining aware of these methods, also be aware of their limitations and realistic alternatives to the proposed research methods.

(6) Consider How the Growth of Data Analytics Has Implications for the Degree of Covert and/or Surveillance Research Made Possible

Assessing the ethical implications for covert actions requires reviewers to remain up-to-date with the rapid growth and increasing technological sophistication of Big Data. While making possible increasingly global coverage, the potential for drilling down to specific individuals and/or their communities is also enhanced. Public and political tolerance of such innovations will depend upon the uses to which they are put and that should assist with any ethics appraisal. Reviewers should consider the potential for misuse of any piece of research and ensure that researchers are aware of and have taken steps to mitigate these concerns in their work.

(7) Consider the Range of Advice and Guidance Available to Help in Assessing the Ethics of Any Proposal Containing Covert and/or Surveillance Research

Both surveillance and covert study might violate a principle of ‘prior’ informed consent – in which case it might be possible to seek consent from observed subjects/participants retrospectively. This might prove difficult with surveillance research that is supposed to be anonymous. This would require identifying and finding the people surveilled, or who contributed the data. Informed consent, however, is only requisite when the information being obtained is private; it is neither ethically required, nor ethical to require it, when such information is neither private nor reasonably expected to be such, for example, in the evaluation of the conduct of government. As with all methods, every effort should be made to ensure people’s dignity and autonomy are promoted, and risks to subjects, researchers, the research community and society in general are minimised. In all cases, to protect researchers and research agencies, national and international law should be respected.

(8) Consider in Detail How ‘Standard’ Ethical Research Principles, Such As Anonymity, Confidentiality and Consent, Are To Be Managed

Anonymity and confidentiality are sometimes used as methods to protect the interests of research subjects. In some contexts, research subjects are not identifiable, and seeking informed consent would compromise the integrity of that anonymity. In cases where the participants are known to the researcher, researchers might not have total control over anonymity – especially when participants themselves engage in disclosure acts, or given modern data analytics technologies which enable personal identification to take place. Similarly, when researchers receive information about illegal acts ‘in confidence’, to retain that confidence might require the researcher to contravene a law and may violate the rights of others.

People cannot give prior informed consent if they do not know what they are consenting to, and the whole point of deception is to ensure that they do not know what is going on at the time when it is happening. The question is whether the research is of such a character that prior informed consent is required. Public policy research includes many circumstances where the consent of the subject is not required: they include some actions in the public domain, actions to make governments accountable or aspects of organisational research.

This does not address the criticism that the absence of consent fails to treat people with respect. If the person affected has no right to consent, part of the objection to deception falls – but the remaining part, that this is not treating people with respect, is difficult to avoid. Deceiving people about the purpose of research, and enlightening them afterwards, treats them as ‘naïve subjects’.

If people do not know what is happening, they cannot exercise a meaningful choice about it which is a challenge to their autonomy. But a restriction on liberty is not the primary concern. It is that sometimes research inflicts costs and harms on research subjects, for example, by the deliberate infliction of stress or anxiety, and the right to consent afterwards is not much consolation.

Deception certainly betrays trust. Whenever researchers negotiate the terms of a contract with a research participant, they make undertakings, explicitly or implicitly: engaging in deception about their role is a breach of the particular rights and duties that are created in the process. However, reasons exist (discussed fully in Iphofen & O'Mathúna, 2022) to ethically justify doing so, even while such research remains risky and must be very carefully reviewed. Where deception has occurred, the correct information, and the reasons for the deception, should be made explicit to the participants when their involvement finishes.

(9) Consider the Public Interest

A major justification often given for covert and/or surveillance research is that it offers the only way of gathering data that may be in the public interest. The reviewer’s role is to assess, in as much detail as possible, whether the proposed research engagement can be considered truly ‘in the public interest’. Note that what is in the public interest is not always the same as what the public finds interesting.

Some (like Homan, R. (1991). The ethics of social research. London: Macmillan) argue that deception raises legitimate concerns over the professional responsibilities of researchers:

• deception may ‘pollute the research environment’, leading people to be suspicious of researchers;

• deception is bad for the reputation of research and researchers; and

• use of deception may legitimate deception to be more widely used.

Additionally, there is the risk that:

• the habit of deception may infect the researcher’s behaviour – it could ‘become a way of life’ and

• the strain of maintaining a deception may be damaging to the researcher.

It is not in the public interest for trust in research and researchers to be undermined. Therefore, deception in research must be justified, carefully reviewed and disclosed to participants when their participation concludes.

(10) Ensure You Are as Informed as Possible by the Available Literature and Illustrative Case Studies So That You Can Give Careful Consideration to the ‘Promises’ Made By Any Research That Is Covert, That Requires Surveillance and/or Entails Significant and Ethically ‘Risky’ Deception

Thoroughly examine any research proposal that appears (implicitly or explicitly) to require deceptive, covert and/or surveillance practices. Information about the ethical risks entailed in these research approaches is clear and easily available. A recommended source is the open access volume for which this guidance was developed and is cited in the Preamble.

(1) Do Not Assume Covert, Deceptive and/or Surveillance Research To Be Inherently Unethical

For the reasons discussed throughout this collection and for those summarised below, these research approaches cannot be assumed to be inherently unethical. Policies based on such research can be ethically justified. The ethical principles by which this form of research must be judged depend on answers to a range of questions such as:

• (a)

  Who has a right to know the information that is being obtained?

• (b)

  Who has a right to control access to that information?

• (c)

  Whose interests must be protected and why?

These questions have been considered in greater depth throughout the open access volume within which this guidance is contained. It cannot be assumed that all information is the private property of the individual from whom it is obtained. In some circumstances, the information is public and should be used to inform public policy. Not allowing public information to inform public policy may be unethical in itself.

(2) Do Not Treat Covert, Deceptive and/or Surveillance Research ‘Approaches’ as a Set of Homogeneous Practices

Covertly observing people in private may be a breach of their rights, although there may be cases where this is justified. Observing people in public is less controversial, although still may impinge on people’s reasonable expectations of privacy. Surveillance is not covert when ‘consented’ to by individuals seeking to purchase goods and services. And if the ‘purposes’ of surveillance are fully conveyed to those being surveilled it cannot be seen as deceptive. For these reasons, the details of what researchers plan to do and how they plan to do it have to be examined. Surveillance activities which are not necessary to the research should be removed, and efforts taken in all cases (covert, surveillance and deceptive) to mitigate potential harms which could arise. This may include, for example, pixelating faces of people when monitoring CCTV for learning about footfall patterns in public areas.

(3) Deceptive Research Practices Cover a Variety of Approaches That Must Be Understood before Deciding to Use or Not Use Such Results

Deception has been widely used in research and in many different ways. Just as the term has a range of meaning in general conversation, it means different things in research contexts. Examples include:

• A medical treatment trial divides people into two groups, half of which unknowingly receive a placebo. For example, everyone might be told they are receiving an experimental drug and then one group is given a placebo.

• A psychologist tells research subjects that the research is being done to examine one psychological trait, when in fact it is being done to examine another one.

• An experiment is set up in which the naïve subject is thrust into interactions with the researcher’s accomplices, people who are party to the deception. Subjects might be told they are experimenters, while accomplices may pose as subjects.

• A journalist secures an interview with a politician on the understanding that it will be used for a profile, when the actual intention is to subject the politician to public criticism.

• A researcher poses as a member of the public to discover how people are treated by an agency.

Using deception in research is ‘risky’ as it can lead to various types of harms, and often has negative connotations, which should be taken into account by policymakers. Despite this, these approaches can be ethically justified if the knowledge gained from such research is important, especially for public policy, and if the harms involved are not excessive and mitigated against as much as reasonably possible. Research involving deception must be carefully and thoroughly examined to determine whether the deception was justified in order to decide whether or not to use its findings in policymaking.

(4) Policymakers Should Consider the Vital Role of ‘Context’

The ethical considerations to be applied to research involving covert, deceptive and surveillance methods must be assessed in terms of the context in which the research is conducted. These factors include the importance of culture and history, and how this impacts the assessment of research. This refers to the concepts of ‘situated ethics’ (where the ethical values and principles accepted in a specific situation or context must be given due consideration in any overall ethical assessment) and what is known as ‘researcher positionality’ (where the motives, intentions and skills of the individual researchers or organisations must be taken into consideration).

(5) Engage in a Full Examination of the Detailed Approach Used in the Research

Research with covert, deceptive and surveillance methods should be assessed in terms of: who is doing what to whom, in what situation and for what purpose? This is a way of expanding the context of the research actions being assessed. In all research, there is a possibility that the actions of the researcher will change the behaviour of the research subjects or participants, and a process of full disclosure may defeat the object of the research. These assessments require advisors with requisite skills and may take some time, which should be taken into account when planning policy development.

(6) Be Open and Transparent If Research Using Covert, Deceptive or Surveillance Methods Was Used to Inform Policy

The use of these methods should be acknowledged when describing the studies that were used to inform policy. Attempts to hide or ignore dimensions of a study that might be seen by some as raising ethical questions are likely to generate further controversy beyond that of the research itself. Not disclosing such aspects of studies might lead some to question or even distrust the integrity of the policymaking process. Transparency and open discussion about the studies used to inform policy, along with addressing questions some might have about them, will help to offset such controversies.

(7) Provide a Clear Rationale to Explain How Research Using Covert, Deceptive or Surveillance Methods Was Determined as Suitable to Inform Policy

If covert, deceptive or surveillance research is used to inform policy, the process and criteria by which such assessments were made should be openly available. This both helps to inform users of the policy about how specific studies were included or excluded, and helps to provide information to the public on how such assessments can and should be made. Such a process should be determined and put in place ahead of time for those making and influencing policy that is likely to be informed by research using these methods.

(8) Data Analytics May Be Used to Inform Policy, But They Must Be Used Carefully and Transparently

Big Data raises a number of relatively new ethical quandaries which are only just starting to be carefully and thoroughly evaluated. This open access volume provides some such detailed assessments. Policymakers should develop guidelines and regulations that ensure data are used ethically and appropriately. In turn, policy should be developed on the basis of ethically justified Big Data research, especially when they have been collected using covert, deceptive or surveillance methods. Much further work needs to be carried out in this area to ensure data are an asset and not a liability for society.

(9) Policymakers Should Seek Community Input on Acceptable Research Practices

Communities and cultures will differ in their evaluation of the acceptability of covert, deceptive and surveillance methods in research. Policymakers should seek input from the communities impacted by their policies (‘stakeholders’) so that these views are taken into account. This will also help to ensure that policies may be more likely to be seen as acceptable by the community. Since these views may change over time, especially following prominent events related to these forms of research, this input should be sought on a regular basis. The mechanisms used for gathering this input can also be used by policymakers to explain their rationale in using or not using such types of research.

(10) Ensure You Are as Informed as Possible by the Available Literature and Take Account of Illustrative Case Studies

Information about the ethical risks entailed in these research approaches is clear and easily available. As recommended above, see Iphofen and O'Mathúna (2022).

SUGGESTIONS FOR ADDING TO AND/OR IMPROVING THIS GUIDANCE SHOULD BE SENT TO: https://prores-project.eu/contact-us/