Ethics in Economic and managEmEnt sciEncEs : a rEsEarchEr ’ s rEsourcE

In an international research climate of increasingly demanding ethical review, based on a biomedical model, reflection on best practices in social, behavioural and economic science research is necessary. It is widely believed that these sciences cannot be held to the same practical requirements as those for biomedical research, although the principles of ethical research are surely universal. This article considers the ethical requirements, principles and guidelines directing research in the social, behavioural and economic sciences, recognised in the national and international arena. By means of a systematic review of available best practices, it is anticipated that general guidelines for social, behavioural and economic science research could be developed and offered to researchers in these fields. Specific consideration is given to the unique characteristics of social, behavioural and economic science research.

…common sense and good scientific practices are vital first steps in ethical problem solving in social and behavioural research.
The absence of data on risks and wrongs in social scientific research does not prove that subjects go unscathed".
-J. Michael Oakes (2002: 449), writing on risks and wrongs in social science research.

Introduction: Social, behavioural and economic science research and ethics
The opening quotations in this manuscript outline three pertinent characteristics of social, behavioural (which covers most research under 'management sciences') and economic science (SBES) research.It is considered to be largely of a nature that is highly unlikely to permanently scar participants in any physically or psychologically meaningful way.Good ethics is just good research; adhering to ethical principles is likely to improve the quality of research, not detract from it.However, good research is not necessarily ethical.Finally, considering the previous statements, it would also be wrong to assume that there are no risks in SBES research.This document is an attempt to explore the ethical issues faced by SBES researchers, to reconcile the seeming disparities mentioned, and to propose some basic guidelines towards better ethical research in the social, behavioural and economic sciences.Overtly, it is precisely the non-physical nature of SBES research that causes unease for those who have to review such research.Indeed, the probability of the appearance of physical harms, as in biomedical research, is much more easily estimated than the probability of the appearance of non-physical harms such as psychological discomfort (National Bioethics Advisory Commission, 2001).SBES research is typically described as 'minimal risk' research (Sieber, Plattner & Rubin, 2002;Thompson et al., 2006).Such minimal risk may be defined as where '…the probability and magnitude of harm or discomfort anticipated in the proposed research are not greater, in and of themselves, than those ordinarily encountered in daily life' (Code of Federal Regulations, 2005).
Researchers have responsibilities, including that of safeguarding the physical, social and mental well-being of the participants in their research.Particularly relevant to SBES research is the anticipation that participants may gain new insight into themselves that they find disconcerting, that some action during participation makes them feel violated in some way, and the basic need that, where possible risks do exist, they should be avoided as far as possible (British Sociological Association, 2002).
The Belmont Report (in Sieber, 2004) outlined three basic principles for human subject research: beneficence, respect and justice.Beneficence, simply defined, means maximising the good, minimising the bad and acting in a spirit of goodwill.If any benefits are to accrue from the research, the primary beneficiaries should be the participants.Respect relates not only to respecting the wishes of potential participants, but also to protecting the interests of vulnerable participants (for example, the mentally or physically frail).Respecting participants' wishes implies, for example, that participants should be free to refuse to participate, or to decide to withdraw at any time, without reproach.Justice relates mostly to a fair distribution of risks and benefits among researcher(s) and participant(s), if such risks cannot be completely avoided.
Although the principles outlined above originated in a longer history of biomedical research, it should be noted that SBES research is distinct in the sense that the extent of interaction (i.e.interference) of the researcher with the participant is often minimal.Dingwall (2008: 3) argues that research methods employed in human and social sciences (HSS) research are those employed by ordinary people in their normal daily lives (author's emphasis): '… observing other people, asking them questions, reading documents or looking at pictures'.Clearly, the threat posed by such SBES research cannot be linearly compared with biomedical interventions such as experimental gene therapy or drug trials.
Given a perceived rigid adherence to inappropriate biomedical models, this author set out to investigate, by means of a literature review, best practices and guidelines in existing frameworks to guide SBES research.This review took the form of several electronic and manual searches of available ethical codes and guidelines, which are discussed in this article.The objective of this work is thus to provide SBES researchers with a resource for considering and addressing ethical issues in their own research.

Informed consent in SBES research
Lindegger and Richter (2000: 313) describe informed consent as the 'cornerstone of clinical trials' and a 'fundamental requirement' in studies to test HIV/AIDS vaccines.'Informed consent describes an interactive process in which individuals or their surrogates agree to voluntarily participate in a research study after the purpose, risks, benefits and alternatives have been thoroughly described and understood' (Marshall, 2007: 23).What must be assessed, however, is the meaning of informed consent in SBES research, and whether stringent adherence to this requirement is that fundamental to the so-called 'soft' sciences.Wassenaar (2008) notes that SBES research must include some method or technique for ensuring informed consent.Alsmadi (2008) draws an important distinction between the rights to privacy, confidentiality and anonymity.In providing informed consent, the participants waive their right to privacy, in the sense that they are volunteering information for public use.However, this does not imply a waiver of the rights to confidentiality and anonymity.
Informed consent relates to potential participants in the research being aware of '…reasonably foreseeable factors that may be expected to influence their willingness to participate such as potential risks, discomfort, or adverse effects' (American Psychological Association, 2002).Different authors have noted that consent should be seen as ongoing and evolving, rather than as a once-off event (Brody, 2001;Dingwall, 2008;Smythe & Murray, 2000).It is a matter to be continually negotiated with the participant as the project develops.Haverkamp (2005) places the responsibility for monitoring and renegotiating ongoing consent on the researcher.Informed consent has also been described as 'preventive ethics' (Parker, 1995, in Lindegger & Richter, 2000: 313).Clearly, the implication is that, by obtaining informed consent, certain general pitfalls in terms of research ethics are recognised, and can be avoided.
A noted exception to written consent, informed consent means that a consent form would be the only identifying link between the participant and the data (Common Federal Policy, 1991).The National Science Foundation (http://www.nsf.gov/bfa/cpo/policy/hsfaqs.htm) has interpreted the so-called 'common rule' (i.e. the Common Federal Policy) to mean that '… when the subject can readily refuse to participate by hanging up the phone or tossing out a mailed survey, the informed consent can be extremely brief'.Haverkamp (2005) also recommends that researchers make clear in what form data will be stored, and who will have access to it.In the same way that guests showing inappropriate behaviour are asked to leave, researchers in the social, behavioural and economic sciences run the risk of being asked to leave if participants deem their behaviour or questions inappropriate (Dingwall, 2008).The best defence for normal, healthy, adult participants is a simple refusal to participate.Marshall (2007) points out that language, literacy levels, beliefs about who has the authority for decision-making and beliefs about nature itself could be determinants in the process of obtaining informed consent.Alsmadi (2008) also notes that special caution must be exercised in obtaining (parental) consent when working with children, or when working with students or employees, who may be more vulnerable to coercion.

What social, behavioural and economic sciences can learn from HIV vaccine trials
Lindegger and Richter (2000) describe informed consent as that where a) all the relevant information regarding the proposed research is communicated; b) the information is understood sufficiently by the prospective participant for him to make an informed decision; c) there is no form of coercion; and d) the participant expressly agrees to participate, usually in written form.
The fundamental aim of informed consent is to assist individuals in making a decision about participating or abstaining from specific research.The nature of this process should be to assist individuals in making decisions that '… are truly in their own best interest' (Lindegger & Richter, 2000: 314).In SBES research, the aims and objectives of the investigation should be clearly outlined, as should issues such as who will have access to the data, where and how it will be stored, and what will be done with the results.Possible risks associated with participation should be considered.These possible risks are described as an '…invasion of privacy, loss of confidentiality, psychological trauma, indirect physical harm, embarrassment, stigma, and group stereotyping' (Oakes, 2002: 449).Other risks are those that threaten '…a subject's personal standing, privacy, personal values and beliefs, their links to family and the wider community, and their position within occupational settings, as well as the adverse effects of revealing information that relates to illegal, sexual or deviant behaviour' (Economic and Social Research Council (ESRC), 2005: 21).Given this broad definition of 'risk', participants should be told what methods will be employed to obtain information from them.
While full disclosure of the technical aspects of a research protocol during the process of obtaining informed consent would satisfy the legal requirements, this does not suggest that it satisfies the moral requirement.Lindegger and Richter (2000) rightly point out that, in the case of HIV vaccine trials, the participants ought to understand to a far greater extent the implications of participation (which could include, inter alia, the discovery of one's own HIV status).In economic and management science research, obtaining written informed consent or full disclosure prior to commencing an investigation may also alter the behaviour the researcher is actually interested in studying.One may think here of examples such as purchasing behaviour or an investigation into deviant behaviours, such as organisational theft or bullying.Clearly, outlining the purpose of the investigation beforehand and requiring express written consent may alter the very behaviour one is interested in studying.
On the topic of coercion, it speaks for itself that it can never be tolerated or seen as acceptable.Although this statement seems self-evident, one needs to critically consider the true extent of the freedom participants are allowed in, for example, an organised and scheduled survey in an organisation, or even in a classroom situation.

Issues arising within culturally diverse settings
Important cultural differences exist in terms of individuals' willingness to share what may be considered personal information.Fuentes (2004) points out that what is considered a 'benefit' varies across cultures.Dingwall ( 2008) also relates how obtaining signed consent forms in an Asian culture was considered deeply offensive, presenting the researcher as disrespectful and lacking in trust.Sieber, Plattner and Rubin (2002) maintain that having to sign an 'agreement' may be perceived as an indication that one's word is not enough.They also relate previous bad experiences associated with signing forms -native Americans lost land in precisely this fashion.The US National Science Foundation (http://www.nsf.gov/bfa/cpo/policy/hsfaqs.htm)notes that cultural norms and lifestyles may have implications for obtaining informed consent.Van den Hoonaard (2001) also notes the important cultural characteristic continuum of individualismcollectivism. In collectivist settings, placing the interest of the individual before that of the group would probably be seen as an affront.
Where individuals are defined, for example in terms of group membership and social relations, the consent of individuals other than the actual participant might be indicated (for example the life partner, elders in the community or tribal leaders).However, researchers differ in their opinions on the extent to which these 'significant others' should be involved.While some insist on the principle of first-person consent with the possible supplemental approval of significant others (Ijsselmuiden & Faden, 1992;Olivier, 1995), other consider wider participation to be essential (Richter, Lindegger, Abdool-Karim & Gasa, 1999).Marshall (2007) calls for researchers to investigate the influence of cultural, social and political factors on community representatives' decisions as far as participation in research is concerned.Marshall (2007) notes important provisions for doing multinational health research in resource-poor settings, which seems applicable to the multi-cultural South African context.These provisions include respect for the cultural traditions of participants or participant communities, promoting collaboration between researchers from resource-rich and resourcepoor settings and continuous community involvement, training of stakeholders in research ethics as well as an ethical review of research protocols, developing participantappropriate ways of gaining informed consent (such as verbal and in the participant's mother tongue), approaching consent as an ongoing process, providing feedback to participants and communities, anticipating possible conflicts and developing plans for their resolution.The ESRC (2005)  Deception should be minimized, and, when necessary, the degree and effects must be mitigated as much as possible.Researchers should carefully weigh the gains achieved against the cost in human dignity.To the extent that concealment or deception is necessary, the researcher must provide a full and accurate explanation to participants at the conclusion of the study, including counseling, if appropriate.
Deception can be justified only when no physical or psychological harm will be caused (Zikmund, 2003).Harm, in this sense, may be taken to include, inter alia, '…harm to self-esteem, stress, (and) future employability…' (Alsmadi, 2008: 157).Here, harm may also be seen as equivalent to risk, the meaning of which was clarified above.
In South Africa, the National Department of Health ( 2004) has issued specific research ethics guidelines.In these guidelines, research involving deception, concealment or covert observation is specifically addressed (under Point 10).Where research involves the deception of identifiable participants, covert observation or the nondisclosure of the purpose of the research, it can be said to be unethical to conduct such research.However, in instances in which deception, covert observation or the non-disclosure of the purpose of the research is essential to the research being conducted, the researcher must satisfy a research ethics committee that: full disclosure of the purpose and/or methodology would threaten the scientific validity of the project; the extent of deception can be defined; alternative methodologies would not achieve the same outcomes; additional risk is not introduced for participants; disclosure will follow the initial deception; and the relationship between researchers, research and participants should not be negatively affected by the deception.Finally, participants should also have the opportunity to withdraw their results after the deception has been made known.

Sensitive questions
Answers to sensitive questions in survey research are seen as being particularly prone to distortion and are simply wrong, and management of this phenomenon within the social sciences is desirable (Barnett, 1998).An important distinction in survey research bears on whether the researcher is interested in behaviours or attitudes, as the associated responses to these two categories of items are quite different.For behavioural items, the response can theoretically be compared to some actual, objective behaviour.In contrast, attitudinal items often invoke a qualitative dimension, which cannot be compared with actual occurrences.In general, it can be seen that the more sensitive the question, the greater the underreporting of the intended measured attitude or behaviour (Barnett, 1998).Barnett presents four strategies for managing this phenomenon: a.Guarantee anonymity.Singer (1978, in Barnett) Millstein (1987, in Barnett) has also indicated no effect for the mode of administration when comparing face-to-face interviews, interactive computer interviews or selfreported questionnaires.
d. Alternative methodologies, such as focus groups, the nominative technique, vignettes and randomised response techniques may also be employed.
Although one may state that 'sensitivity is in the eye of the beholder', Barnett (1998) notes that 'typical' sensitive dimensions include sexual practices, alcohol and drug use, AIDS, income and criminal behaviour.In organisational research, it is noted that issues such as business ethics, bankruptcy and transgressions of organisational standards are sensitive.The final recommendation from Barnett (1998) is that sensitivity should be defined by the participants, and within a particular context.SBES researchers would do well to consider beforehand the possible consequences of their research for all those (individuals, departments, and organisations) involved.Employing focus groups prior to full roll-out can also assist in contextually defining the sensitivity and acceptability of questions.Richardson and Godfrey (2003) note the important distinguishing feature of an interview as being that of an emotional bond developing between the interviewer and the interviewee.

The interview as instrument
In ensuring the protection of interviewees, the latter should be aware that they may stop the interview, that the researcher may choose to stop the interview, and that correct procedures ought to be applied in both closing an interview and referral to social or other mental health services (Richardson & Godfrey, 2003).It is also important to recognise that the interview is inherently exploitative -the researcher wishes to gain information from the interviewee for his or her own purposes, although this is obviously acceptable if it is correctly handled.Reciprocity in the relationship is important in making sure of this (Richardson & Godfrey, 2003: 349).
When it comes to interviews, Richardson and Godfrey (2003) also stress the ongoing nature of the informed consent process.Interviewees may, for instance, give consent only at the end of the interview, or be allowed to view the specific content from the interview (the excerpts) that will be used in the research.In an interview setting, informed consent would entail the interviewee being assured of the purpose of the research; what the information gathered will be used for; how their anonymity will be protected; and who will have access to the information (Richardson & Godfrey, 2003).

The important differences between social, behavioural and economic and biomedical sciences
The American Association of University Professors (AAUP; Thompson, Elgin, Hyman, Rubin & Knight, 2006) has expressed serious concern about the practice of requiring ethical approval for research involving human subjects that poses no serious risk of harm to participants.However, the recommendation is not that social science(s) be exempt from ethical review, because it is possible that social science research could pose serious psychological harm to participants.Paradoxically, certain biomedical research poses no significant risk to participants (for example, data-gathering by survey).The call is to consider the methodology rather than the discipline to be evaluated.As such, the recommendation is that '…research whose methodology consists entirely of collecting data by surveys, conducting interviews, or observing behaviour in public places…' be exempt from review (Thompson et al., 2006: 3) et al., 2006: 3).The AAUP lashes out particularly at the 'paternalism' (Thompson et al., 2006: 4) of Institutional Review Boards for calling a halt to research where it is deemed that even the completion of sensitive items may be stressful to participants.Rightly stated, also, autonomous individuals who provide informed consent should be able to opt out of the research if they find the survey items sensitive.Naturally, the normal exemption of special or vulnerable populations (such as children, individuals in national service or incarcerated) still applies.Solomon (2005), however, notes at least 10 contributions that the social sciences could make to bioethics.These are noted as recognising the gaps between practice and ideals; assisting in evaluating personal aptitude for ethical analysis; investigating and recognising the institutional/environmental context in which ethics takes place; encouraging moral accountability; investigating cause and effect relationships and predictive values assumed in bioethics; clarifying the applicability of ethical principles in multicultural contexts; recognising the relevance of ethical principles in new contexts and to social phenomena; and the implied new moral problems, and greater elaboration of existing, identified problems.Solomon (2005: 40) notes that biomedical research can indeed benefit from social science research in that the latter can provide the role of '…context, intentionality, and outcomes…' in research.

Best practice guidelines 5.1 A European example
Freed-Taylor (1994) outlines the following principles for social science research, which is a summary of various European codes.Acceptance of responsibility implies that all parties to the research should carry equal responsibilities, and (possible) harms and benefits should be anticipated before the research starts.The conduct of research should be such that it contributes to both a positive relationship between researchers and participants and future research.Issues like anonymity have to be clarified and (possible) harms avoided.Compliance with legislation is an obvious essential.Information gleaned should be presented to the scientific community, and the limitations in terms of its interpretation should be noted.Cross-cultural research requires ethical approval (and legal compliance) in all participating countries.Vulnerable groups raise special issues of informed consent and potential risk.'Vulnerable' participants are not clearly defined, but have been noted to include '…children, prisoners, pregnant women, mentally disabled persons, economically or educationally disadvantaged persons' (Common Federal Policy, 1991).In addition, this category includes '…racial minorities…the very sick, and the institutionalized' (National Commission, 1979).Weijer and Emanuel (2000) consider participants to be vulnerable if they are not in a position to provide informed consent owing to their position (such as being in prison), or not possessing adequate intellectual faculty (such as children or the mentally ill).Resolution of conflicts, should they arise, should consider the contribution of relevant ethical bodies, professional associations or colleagues.Where research involves interviews, Research Ethics Board (REB) approval of the proposed interviewing techniques must be sought, and the free and informed consent of the interviewee needs to be ensured.

A Canadian example
Article 3.1 requires REB approval for the collection of information by means of personal interviews, which may be described as including such means as face-to-face, telephonic or electronic encounters, or individualised questionnaires which the researcher uses to gather materials for such purpose as a biographical study or other research involving specific personalities (Canadian Tri-Council Policy Statement, 1998: 3.3).
Additionally, the Policy defines personal information as … information relating to a relatively identifiable person who has a reasonable expectation of privacy.It includes information about personal characteristics such as culture, age, religion and social status, as well as their life experience and educational, medical or employment histories.However, Article 1.1 (c) is excluded from REB review research that is exclusively based on publicly available information.This includes documents, records, specimens or materials from public archives, published works and the like, to which the public is granted access.
As a general rule, the best protection of confidentiality of personal information and records will be achieved through anonymity.If the data being stored are truly anonymous, the research project will need only minimal REB scrutiny (Canadian Tri-Council Policy Statement, 1998: 3.2).
In an academic context, the following provision is important: … when records of prisoners, employees, students or others are used for research purposes, the researcher should not provide authorities with results that could identify individuals, unless the prior written consent of the subjects is obtained.Researchers may, however, provide aggregated data that cannot be linked to individuals to administrative bodies for policy decision-making purposes (Canadian Tri-Council Policy Statement, 1998: 3.4).

UNESCO
The United Nations Educational, Scientific and Cultural Organization developed a set of ethical guidelines for international research conducted under their programme entitled Management of Social Transformations (MOST) (www.unesco.org/most/ethical.htm).Importantly, these guidelines explicitly hold themselves as not an alternative to the 'scientific and professional judgement' of the researcher(s).The guidelines do, however, place ethical responsibility primarily on the principal researcher(s).They also require the principle of beneficence to be maintained, potential benefit to be a consideration in the initial choice of research topic and the consequences, and especially risks and use of the research to receive due consideration.The guidelines also call for competence in researchers and the explication of their personal ethical stance; compliance with 'customs, standards, laws and regulations'; and respect for a host culture.The guidelines also include familiar requirements, such as informed consent; avoidance of harm and coercion; undue intrusion; confidentiality; and access to and preservation of results.Lastly, the guidelines also make recommendations in terms of methodological issues, and stress reporting the results with integrity, respecting the work of other researchers and full disclosure in publication(s).The survey you have received is interested in studying … (project description in layman's terms).By completing this survey, you agree that the information you provide may be used for research purposes.Know that you are free to decide not to complete the survey, although your data cannot be replaced by anyone else's.The survey will, however, be completed anonymously, and we as researcher(s) will have no way of connecting the information you provide to you personally.We do not foresee that you are likely to experience any negative consequences due to completing this questionnaire OR We foresee the following consequences in completing the questionnaire … (outline anticipated risks and harms).Even so, the researcher(s) undertake to keep the individual information provided herein confidential, not to let it out of their possession, and to analyse results only at the group level.It is hoped that the information we gain from this survey will help us in … (state anticipated outcomes of the project).However, should you have questions or concerns, please contact the principal investigator … (provide a name and suitable contact information).

A South African example
What should be included in a typical cover letter/ project description?According to the guidelines proposed by Lindegger and Richter (2000), it would entail giving participants a reason for the research, how the information would be used, the anticipated outcomes of the research, and its (possible) implications for participants personally.It has also been noted that obtaining informed consent may not necessarily be the first action taken in the process of the research, and could even be the last (see 2.1).However, when information gathered and used for research purposes is directly related to specific individuals (such as in qualitative research), written, informed consent should be applied as the golden standard.The reality is, though, that such consent may only be available verbally (such as in the case of illiterate participants), and should ideally be continuous in nature, while the research agenda develops in interaction.

Cultural sensitivity
Given the plural character of South African society, special care should be taken to ensure cultural sensitivity, whether the research is qualitative or quantitative in nature.Quantitative research, such as in surveys, should be made available in, and at a level of language that is appropriate to the potential respondent.As a general practice, English is the language of surveys, although only a small percentage of South Africans speak it as a first language.
In qualitative research, great care has to be exercised in obtaining informed consent.Indications are that it is not the individual alone, but probably also the community that has to be considered.When participants are to be asked questions, there should be sensitivity to culture, religion, language, lifestyle and any other differences.

Education
In South Africa, there is a two-fold responsibility in research ethics relating to education, and institutions of higher learning must accordingly provide (additional) training in ethics for researchers.Ethics, especially in the way they relate to management and economic sciences is in all likelihood not treated with the earnest attention it deserves, on account of its ostensibly innocuous nature.Further, graduates in these disciplines often register with professional bodies, and their practice as professionals is then overseen by an external 'watchdog' (e.g.industrial psychologists or chartered accountants).However, in academia, and given the much-less regulated South African context, a greater responsibility falls on institutions that initiate research, and professional training in the ethics of research remains the responsibility of the institution.
Researchers themselves, often working in settings characterised by great resource disparity and differences in levels of formal education, have a tremendous responsibility to educate participants not only about research and its value, but also about research ethics, if the good will of the community and the future of research are to be secured.In this case, the concept of education would include learning that the results of research should be communicated to participants, and are not solely for publication in a subject-specific academic journal.Also in the academic context, consideration of who the 'primary participant' is would suggest that a more senior and experienced study supervisor should be aware of, and sensitive to, ethical issues, and be able to guide learners in this respect.

Special issues
The ethics of research is also enhanced by cognisance of issues relating to validity in terms of research execution.When it comes to asking difficult or sensitive questions, employing deception, covert observation or concealment and using interviews as a data-gathering instrument, all these issues were discussed under point 3 above.Researchers are referred to this section for some basic guidelines on appropriate ethical considerations.Hunter (2008) notes the development of an increased awareness of the need for ethical review of research involving human subjects.This includes such considerations as journal requirements of ethical approval of publishable research; requirements of external funding agencies; the fear of liability; and a general increased awareness of the ethical implications of research.It has also been noted (De Vries, De Bruin & Goodgame, 2004) that it is particularly researchers in the social, behavioural and economic sciences (SBES) who have lost patience with research ethics committees!Given this climate, Du Bois ( 2004) rightfully acknowledges that compliance 'for the sake of it', or, even worse, in pursuit of the avoidance of penalty, in principle runs counter to ethical review.Wolcott (1994) has also warned of the threat that a mechanistic adherence to rules poses to real ethical issues.It is the express intention of this paper to make research ethics more accessible to the management and economic sciences researcher.It is intended to bring about a greater understanding of the need for the ethical review of research, but ultimately to present guidelines for researchers that would encourage ethical consideration in the execution of research.

Conclusion
Dingwall (2008: 3) is of the opinion that humanities and social science (HSS) research poses: '…at most…a potential for causing minor and reversible emotional distress or some measure of reputational damage … risks that research participants are well able to assess for themselves'.This attitude of placing responsibility solely on willing participants seems inadequate.It has been noted above (see point 5) that research ethics may also rest on factors such as the competence and professionalism of the researcher.Oakes (2002) interprets the US Code of Federal Regulations, title 45 part 46 (2005), as stating that all survey research is exempt from ethical review, unless identifying information is collected and harm may follow from its disclosure.The Code of Federal Regulations, title 45 part 46, (2005) exempts research: … involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures or observation of public behaviour, unless (i) information obtained is recorded in such a manner that human subjects can be identified, directly or through identifiers linked to the subjects; and (ii) any disclosure of the human subjects' responses outside the research could reasonably place the subjects at risk of criminal or civil liability or be damaging to the subjects' financial standing, employability, or reputation.
Other exempt research is 'anonymous and harmless surveys'; 'observational studies'; 'surveys on groups; organizations and associations'; research that deals with the 'collection or analysis of existing data'; and 'publicly available data sets or those stripped of identifying information' (Oakes, 2002: 457-459).

End notes
1 The ESRC (2005: 7) defines 'human participants' as 'including living human beings, human beings who have died recently (cadavers, human remains and body parts), embryos and foetuses, human tissue and bodily fluids, and human data and records (such as, but not restricted to medical, genetic, financial, personnel, criminal or administrative records and test results including scholastic achievements)'.

3 The axis of validity and being ethical 3.1 Deception, concealment and covert observation
presents helpful case studies as part of their Research Ethics Framework (REF), from which one important question arises: 'How will you respond in explaining your research, if you are accused of being unethical?'