Protecting Participants from Harm
The source defines harm two words wide and the application form behind it asks four categories wide. Between them sits a threshold phrase that does more work than any other clause in the document — and a control-group problem that project researchers create without noticing.
What the source counts as harm
A second statement widens that definition into something more usable. Describing the design stage, the source says the research must be designed so that a participant does not suffer physical harm, discomfort, pain, embarrassment, or loss of privacy.
Five named outcomes, and they do not sit on one scale. Sorting them is worth doing, because the ones a project researcher will actually cause cluster at one end.
THE FIVE NAMED OUTCOMES AND WHERE THEY ARISE IN PROJECT RESEARCH
| Named outcome | Kind of harm | How a project study produces it |
|---|---|---|
| Physical harm | Bodily | Rare in organisational research. Real where fieldwork involves construction sites, plant, remote locations or night work |
| Pain | Bodily | Effectively absent here. It is in the definition because the definition is not written only for this field |
| Discomfort | Ambiguous — bodily or psychological | A long interview at the end of a shift; questions about a failed project the participant led |
| Embarrassment | Psychological and social | Being identified in a finding; a quotation read by peers; being seen to have held a view the report calls mistaken |
| Loss of privacy | Informational | The single most likely harm in this field. Identifiable data, recognisable quotations, and disclosure to anyone outside the study |
The five outcomes are the source's; the categorisation and examples are added here.
The source also locates harm specifically, stating that it is most likely to arise from use of data collected by the researcher and from privacy and confidentiality of information — treated in The Four Ethical Principles, with the privacy standards themselves in Confidentiality and Anonymity.
The four risk categories your application must address
Four categories, one threshold, and an instruction to detail steps rather than assert that risks are low. Take them one at a time: they fail differently and are mitigated differently.
The four categories in a project setting
Bodily risk from taking part
Usually nil for interviews and surveys. Non-trivial where collection puts someone on a site, in a vehicle, or working outside normal hours because that is when you could get access.
Distress produced by the subject matter
Project research is full of failure. Asking someone to account for an overrun, a cancelled programme or a redundancy round can reopen something that is not academic to them.
Standing, relationships and position
The category that dominates workplace research. Being identified as the person who criticised the sponsor or the delivery model changes how someone is treated long after your study is filed.
Exposure to liability or breach
Most often left blank, and most likely to bite. A participant describing a defect, a variation or a non-conformance may be producing a statement with contractual or regulatory consequences.
The threshold: beyond those normally encountered in everyday life
The phrase appears twice in the risk section — once for participants, once for the researcher — and it changes what the question means. Three things follow.
- It is a comparator, not an absolute. The form does not ask whether your study carries risk. It asks whether the risk exceeds ordinary life. Every study carries some risk, and saying so is not an admission of a problem.
- It does not license risk below the line. Risks that sit under the threshold still exist and still have to be handled well; they simply do not trigger the requirement to detail mitigation steps and support facilities in the application.
- The baseline is not fixed. What is normally encountered in everyday life differs between a site supervisor, a contract administrator and a person on a probationary contract. The comparator moves with the population you are studying.
That third point is where judgement enters, and the form requires you to exercise it without telling you how. An hour of questions about project decisions is unremarkable for a senior manager who spends the week defending decisions. The same hour, for someone whose contract is under review, is not the same event.
What the application asks you to do about risk
The instruction is not to declare risks. It is to detail the steps that will be taken to address, manage and/or minimise them. Those three verbs are not synonyms, and separating them turns a vague answer into a specific one.
Three responses the wording distinguishes
Address — remove the risk from the design
Change what you do so the exposure does not arise. Drop the question that produces the legal admission. Interview off site. Collect the variable in aggregate rather than individually. This is the strongest response and the cheapest to make before collection begins.
Minimise — reduce the exposure that remains
Shorten the instrument, narrow the population, lower the identifiability class, restrict who sees raw data, or remove the detail that makes a quotation attributable. Each of these leaves the risk in place and makes it smaller.
Manage — plan for the risk when it materialises
Decide in advance what happens if a participant becomes distressed, discloses something serious, or asks to stop. Managing is what remains once you have addressed and minimised what you can, and it has to exist on paper before the first session.
Support — name the facilities available
The form names three by way of example: counselling, debriefings or referrals. Which of them exist for your participants is a factual question about your setting, and finding out is part of preparing the application.
Researcher safety is a separate obligation
The application asks about you in your own right, in the same shape: if there are any potential safety implications for yourself as the researcher, beyond those normally encountered in everyday life, indicate how these will be addressed.
It is not an afterthought elsewhere either. The design-stage list of ethical issues names the researcher's right to safety alongside participants' rights, and the five considerations for designing research include protecting the safety of the researcher and the team. That phrasing is wider than you: assistants, transcribers and co-researchers are inside it.
Where researcher risk actually sits
- Lone fieldwork — site visits, remote locations, unfamiliar premises
- Interviews conducted after hours or in private settings
- Exposure to distressing material across many interviews in sequence
- Being told something you may be obliged to act on
- Career and reputational exposure when researching your own employer
Responses that fit an application answer
- A named person who knows where you are and when you are expected back
- A rule about which settings you will not interview in alone
- A cap on how many sessions you run in one day or week
- A decided position on what you will do with a serious disclosure
- Separation between your research role and your line role, agreed in writing
The last item in each column is the one particular to this readership: insider research is the normal case for a practising project manager, and the risk it creates for the researcher is occupational rather than physical. The related consent problem is treated in Voluntary Participation and Informed Consent, and the wider list of rights in Ethical Issues at Each Research Stage.
The right to service and the no-treatment control group
This is the one harm on the page produced by good methodology rather than by carelessness. A control group is what makes an effect attributable, so withholding the thing under study is the mechanism rather than a side effect — and the source's point is that the people it is withheld from experience it as a withheld benefit.
Project researchers meet this more often than the clinical framing suggests, because delivery organisations run comparisons constantly and call them something else.
WHERE A NO-TREATMENT CONTROL APPEARS IN DELIVERY RESEARCH
| What it is called in the organisation | What it is methodologically | What the excluded group loses |
|---|---|---|
| Pilot on two teams before wider rollout | Treatment and control | Access to a tool or method believed to make the work easier |
| Phased deployment by region or package | Staggered treatment with an untreated comparison | Earlier access, and any advantage that comes with it |
| Training offered to one cohort and evaluated against another | Classic no-treatment control | A development opportunity that appears on a record and may affect progression |
| Coaching or mentoring trialled with selected project managers | Selective treatment | Individual support of direct career value |
| A new reporting process run in parallel with the old | Comparison group | Usually nothing — the exception worth checking for before you assume a problem exists |
The right-to-service problem is the source's. These delivery translations are added here; the source does not enumerate settings.
Where an intervention is evaluated, the design questions and the ethics questions are the same questions asked twice — see Evaluation and Action Research Methodologies, and Correlation and Experimental Methodologies for what a control group is doing at all.
Showing that you have designed harm out
What a defensible risk section contains
- A statement under each of the four categories, including the ones you assess as nil, with the reason
- An explicit comparison against the threshold, naming the population you compared to
- For each risk above the threshold, what you removed, what you reduced, and what you will manage if it occurs
- The support facilities actually available in your setting, checked rather than assumed
- A stopping rule: what ends a session, what happens to data already collected, and who is told
- A researcher safety answer covering location, timing, lone working and anyone assisting you
- Where a comparison group exists, what those participants are told and whether they receive the intervention later
What to carry forward
- Harm is physical and psychological, and the standard bites on the risk of harm, not only on harm that occurs.
- Answer all four categories — physical, emotional, social and legal — including the ones you assess as nil, and give the reason.
- The legal category is the one most often left blank and the one that surfaces latest, because a recorded interview is a document.
- The everyday-life threshold is a comparator whose baseline moves with the population, and the source never calibrates it — write down the comparison you made.
- Address, minimise and manage are three different responses. An application that only identifies risks has answered half the question.
- Researcher safety is a separate question with the same threshold, and the duty extends to the team, not just you.
- A no-treatment control group can curtail access to a beneficial service. The source names that problem and does not resolve it; state your position deliberately.
Frequently asked questions
My study is interviews and a survey. Is there any risk to declare?
Almost certainly yes, in the social and legal categories rather than the physical one. Being identified as the person who criticised a delivery model, or describing a defect in a way that has contractual consequences, are risks to wellbeing arising from participation. Answer all four categories, and where you assess one as nil say so and give the reason rather than leaving the field empty.
What does 'beyond those normally encountered in everyday life' actually mean?
It sets a comparison rather than a zero-risk rule: the question is whether participation exposes someone to more than ordinary life already does. The supplied source states the threshold and never explains how to calibrate it, and the baseline plainly differs between a senior manager and someone on a probationary contract. Record which population you compared against and why, so a panel can see the reasoning.
Do I have to consider risks to myself?
Yes, and separately. The application asks whether there are safety implications for you as the researcher beyond those normally encountered in everyday life and how they will be addressed. The design-stage list also names the researcher's right to safety, and the wider design considerations extend it to the research team, so anyone assisting you is inside the obligation.
Is a pilot with two teams a control group problem?
Methodologically it is a treatment and comparison design, so the right-to-service issue applies whatever the organisation calls it. If the thing being piloted is believed to be beneficial, the teams excluded from it may reasonably feel their access has been curtailed. The supplied source names this problem and offers no solution, so you need a stated position — most often sequencing the excluded group in after measurement.
What should I do if a participant becomes distressed during an interview?
The supplied material contains no incident procedure, no reporting obligation and no stopping guidance, so nothing here can be quoted as the requirement. What the form does require is that you detail steps to manage risks and name support facilities such as counselling, debriefings or referrals. Decide and write your stopping rule before the first session, and confirm with your approving body what it expects to be told.
Does protecting participants from harm just mean protecting their data?
Data handling is the largest part of it in this field — the source says harm is most likely to arise from use of data collected by the researcher and from privacy and confidentiality issues. But the risk question is wider than data: emotional distress from the subject matter, social consequences of being identified, legal exposure from what is said, and physical risk in field settings are all separately named.
References and source attribution
- Trochim, W. M. K. 2006, Research Methods Knowledge Base (http://www.socialresearchmethods.net/kb/probform.php) — the source of the excerpt on the system of ethical protections, including the risk-of-harm and right-to-service passages quoted on this page.
- Cooper, D. & Schindler, P. 2008, Business Research Methods, 10th ed., McGraw-Hill — the text the supplied source draws on for its treatment of research ethics.
- Quinlan, C. 2011, Business Research Methods, 1st ed., Cengage Publishing, Chapter 3 — the prescribed reading accompanying the ethics material.
- Swinton, J. 2009, 'Why bother with research ethics?', Journal of Community Nursing, vol. 23, no. 9, pp. 25–27.
- The supplied teaching source: consolidated weekly teaching notes and slide material on research ethics, including the eight-part ethics application structure reproduced as an example from a government department, which supplies the four risk categories and the everyday-life threshold.
Suggested questions for Ask KEVOS
- Take my interview schedule and tell me which questions create social or legal risk for the person answering.
- Write a risk section covering all four categories for a study of my own delivery team, including the ones I assess as nil.
- Draft a stopping rule for what I do if a participant becomes distressed, and what happens to their data.
- My pilot excludes three teams from a new scheduling tool. Help me state a position on the right-to-service problem.
- What researcher safety measures should I describe for interviews conducted on site and after hours?