Social Research update logo
Issue 20 Summer 2000

Social Research Update is published quarterly by the Department of Sociology, University of Surrey, Guildford GU2 7XH, England. Subscriptions for the hardcopy version are free to researchers with addresses in the UK. Apply by email to

Safety in Social Research

Gary Craig, Anne Corden and Patricia Thornton

Dr Gary Craig is Professor of Social Justice at the University of Hull. Drs Anne Corden and Patricia Thornton are Senior Research Fellows at SPRU, University of York.

Written comments on the proposed Safety Code can be sent to Gary Craig at CASS, University of Hull, Hull, HU6 7RX or to the Social Research Association at

There is raised awareness of the dimensions of risk that may be associated with professional situations which involve close social interaction between people in a private setting. People who work in this way include doctors, solicitors, officers of local and central government, teaching staff, insurance and estate agents. Depending on the particular situation, there are various components of risk for the professionals involved:

Risk is carried on both sides, however. The person receiving the professional service may also be at risk of physical harm or abuse, or in danger of being exploited, frightened or psychologically harmed. Not all professional people behave properly, and some are inadequately trained to deal with the situations they encounter.

If the professional person has an employer or manager, they should understand the risks involved on both sides, consider the balance of responsibilities, and act or intervene to protect employees and their clients.

Several professions are developing or already have guidance or codes of practice for their staff, to help maintain their own personal safety. The social work profession’s guidance for members has to take account of issues such as potential allegations of child abuse and apparently increasing levels of physical attacks against social workers (see for example the Campaign for Safety in Social Work promoted by the magazine Community Care throughout 1999). Many teaching and higher education professional bodies now provide guidance to their members (see for example the Association of University teachers circular LA/5674, February 1996).

A major contribution has come from the work of the Suzie Lamplugh Trust, established after the disappearance and presumed murder of Suzie Lamplugh when she was working as an estate agent. This charity publishes a guide for employees (Suzie Lamplugh Trust n.d.) which has been widely influential.

It is perhaps surprising that there has not been wider discussion about issues of personal safety within the social research profession. Social research often requires researchers to work on their own, and interviewers involved both in survey work and qualitative enquiry are often engaged in one-to-one relationships, usually conducted in private situations. The topics of enquiry may be sensitive and invoke strong feelings among those people participating (Lee 1993).

The authors have contributed to the development of framework codes of conduct for their own social research units to help to protect the safety of their staff. Kenyon and Hawker (2000) uggested that a set of guidelines might be helpful, and argued for further discussion around this topic. Such ideas are reflected in a small but developing literature on the subject within and outside the UK (Arksey and Knight 1999; Paterson et al. 1999).

Who should be concerned with safety?

The Suzie Lamplugh Trust explains that safety at work is a dual responsibility for the employer and the employee. It may be hard to clarify how such responsibilities should be shared, however, and hard to ensure that each party meets its responsibilities. Employers of researchers are generally universities or research institutes and under the terms of the Health and Safety at Work Act and the subsequent Management of Health and Safety at Work Regulations (which came into force in 1993, implementing a European Union Framework Directive) they have a general ‘duty of care’ to their employees. For all normal purposes a university personnel office might legally stand as the employer, but the duty of care might be regarded as reverting to the research manager, who is usually the head of the unit or the research budget holder. A number of situations arise, in the context of social research, where responsibility for safety might be contested. For example, how far does this duty of care extend when researchers conduct interviews alone on the street at night in an unfamiliar town? What might be regarded as reasonable precautions and who should be responsible for taking which of them?

Another tricky issue is when does fieldwork start and finish? Fieldwork may involve travelling, and using a private car may be the only option, or the safest choice, for researchers in remote or potentially dangerous contexts. Not all universities have insurance policies which cover the use of private cars for travel to work, however. If a private car is used, who is responsible for meeting the costs of damage in the event of accident, or violence done to the vehicle? More generally, are researchers insured for personal injury in all situations related to the conduct of fieldwork? These issues may be particularly pertinent for researchers studying phenomena at the boundaries of criminality; working with potentially higher risk groups such as ex-offenders or people with a history of psychological disturbance, or simply exploring issues where the threat of violence is greater, for example, working across sectarian divides, or studying homophobic violence.

It is clear that there are a number of hard decisions to make. The next step is how to implement the decisions. Managers or employers can put in place some elements of safety with written guidelines; provision of safety aids; and procedures by which fieldwork can be monitored. However, the amount of control which can be reasonably exercised over a researcher in the field may be limited. A code of safety may be brought to the researcher’s attention within the contractual framework of their employment, but much more may need to be done to encourage staff to remember and use the guidelines. Ways of making such guidelines ‘stick’ will include general awareness raising at various levels. Safety issues should feature in the training of all new research staff, and be included in induction packs or staff handbooks. More experienced staff may be among those who are prepared to take greater risks and there will be a need for continued reminders and reinforcement throughout a researcher’s career. Supervisors may need to take staff through procedures and remind them, with each new bout of fieldwork. There may be a role for University Safety Officers, or ‘outside bodies’ such as the Suzie Lamplugh Trust advisers, to visit research units occasionally to bring fresh perspectives and maintain interest in issues. It will always be important to remind research staff that if they ignore their employers’ procedures and policies for health and safety at work, they may be considered negligent should an incident occur.

Unexpected or unpleasant situations may arise for the researcher, however, even when they are following guidelines, and the effects of physical or emotional violence or the threat of violence may be traumatic. There is need to consider how far employers have responsibility to ensure availability of suitable debriefing, stress counselling or therapeutic help after bad experiences, and how research units might access such professional help. Safety must also be considered from the perspective of research respondents. Ethical guidelines developed variously by the Social Research Association (SRA 1999), the British Sociological Association (BSA undated), and other professional bodies, remind researchers of the need to protect the interests of potential respondents. This includes respect for privacy, ensuring informed consent and confidentiality, and awareness of the possibilities of doing harm. Ethical considerations, however, sometimes emerge in complicated ways. One of the authors conducted research with children where it was possible that undisclosed child abuse might be revealed. Interviewers had to protect themselves and the integrity of their research and at the same time make it clear to the children that they would not remain inactive if ‘they thought anyone was in danger’.

According to the SRA ethical guidelines ‘the principle of informed consent is ... an expression of belief in the need for truthful and respectful exchanges between social researchers and human subjects’. There are in reality no firm safeguards in place to prevent respondents from exploitation by researchers and it may be appropriate to consider whether some form of professional accreditation would be an appropriate move in this direction. This might offer the possibility of sanctions against manipulative or deceitful behaviour on the part of researchers but would be unlikely to prevent other more oppressive forms of abuse.

Race, culture and gender

Issues of race, culture and gender may have significant impact on researchers’ safety, particularly in violently divided societies. Lone female researchers are, in general, more vulnerable than lone males. Even where the threat of physical violence is not the issue, certain more orthodox cultures may find women researchers unacceptable and react with hostility. Certain racialised contexts may make the conduct of non-ethnically-matched interviewing especially hard. Potentially risky situations of this kind may sometimes be avoided by asking respondents in advance about preferences and expectations. Trying to match the gender or race of interviewer and respondent raises resource issues, and is an important matter for the design and commissioning stage of the research.

Additional subtle issues of culture arise, including the use of body language, the way researchers dress, and the acceptability or not of physical contact. Researchers may inadvertently increase the risks to themselves associated with the research. GPs have been offered advice about ‘unsafe’ territory (Sandell 1998) but researchers are often left to rely on their own judgement or intuition in arranging and conducting an interview. The quality of much social research depends on establishing the appropriate distance between researcher and respondent - a distance neither over-familiar nor too detached. It may not always be easy, even with prior briefing, to know what that distance should be.

Who pays for safety?

There are resource implications attached to the conduct of research that is both ethically robust and safe. Some research funders now acknowledge and accept a budget component to cover mobile phones, personal alarms and phone cards; the use of hire cars or taxis, and appropriately priced and located overnight accommodation. Some dimensions to the costing of safety have more significant implications. In situations where it is inappropriate for researchers to work on their own, research managers have to consider their working in pairs, working with friends or colleagues who are paid simply to ‘wait outside’, or to conduct interviews only in public places. All of these tactics may be costly, but funders are likely to take seriously arguments that the quality of research suffers if researchers feel vulnerable and frightened.

Suggestions for a Code of Practice

In order to stimulate discussion and progress the matter, we have considered what a code of practice for the safety of social researchers might look like, and offer a first draft. The code is designed for research funders, employers, research managers and researchers conducting fieldwork. The aims are to point out safety issues which need to be considered in the design and conduct of social research in the field and to encourage procedures to reduce the risk. The intention is not to be alarmist about potential dangers but to minimise anxieties or insecurities which might affect the quality of the research. The suggested code covers:

Developing the Code of Practice

These suggestions will need to be developed to take account of the different types of social research and the context in which it takes place. For example, the focus on safety in carrying out one-to-one interviews in private settings needs to be broadened to cover observational and ethnographical research. We have written from the perspective of working in fairly well-resourced university research centres, and acknowledge that issues facing independent researchers and those working in small units may be different. The next step is to consult widely on the draft Code. We have sent the full draft to the Social Research Association and hope that it will lead in consulting with its membership, with the aim of producing a consolidated Code which the SRA can formally endorse. Such a Code would not be set in stone but can be developed or amended in the light of practice.


Arksey, H. and Knight, P. (1999) Interviewing for Social Scientists, London: Sage.

BSA (undated) Statement of Ethical Practice, British Sociological Association.

Kenyon, E. and Hawker, S. (2000) ‘Once would be enough’: some reflections on the issue of safety for lone researchers, International Journal of Social Research Methodology, 2, 4, 313-27.

Lee, R.M. (1993) Doing Research on Sensitive Topics, London: Sage.

Paterson, B., Gregory, D. and Thorne, S. (1999) ‘A protocol for researcher safety’, Qualitative Health Research, 9, 2, 259-69.

Sandell, A. (1998) Oxford Handbook of Patients’ Welfare: a doctor’s guide to benefits and services, Oxford: Oxford University Press.

SRA (1999) Social Research Association Ethical Guidelines, Appendix B in SRA Directory of Members, London: Social Research Association.

Suzie Lamplugh Trust (undated) Personal Safety at work: guidance for all employees

Social Research Update is published by:

Department of Sociology
University of Surrey
Guildford GU2 7XH
United Kingdom.

Telephone: +44 (0) 1 483 300800
Fax: +44 (0) 1 483 689551

Edited by Nigel Gilbert.

Summer 2000 © University of Surrey

Permission is granted to reproduce this issue of Social Research Update provided that no charge is made other than for the cost of reproduction and this panel acknowledging copyright is included with all copies.