The Four Ethical Principles
Three of the four are stated as guidance you can act on. The fourth is four questions the supplied source poses and never answers — and the first of them describes a recognised integrity problem that is left hanging.
Four principles, claimed as universal
Two claims are packed into that framing sentence and both are worth pulling apart. The first is universality: the source is not describing a local requirement but asserting that these hold generally. The second is that ethics has two subjects — honesty, and the rights of individuals.
Split the four against that second claim and they divide cleanly in half, which is a useful way to remember them. Two protect the people in your study. Two protect the people who read it.
THE FOUR PRINCIPLES AGAINST THE SOURCE'S OWN FRAMING
| Principle | What the source says it concerns | Which half it serves | Who it protects |
|---|---|---|---|
| Do no harm | Harm arising from use of data collected by the researcher, and from privacy and confidentiality of information | Rights of individuals | Participants, and anyone identifiable in your data |
| Free choice | Participants should not be coerced; coercion may come from feeling unable to refuse, or from a penalty for refusing | Rights of individuals | Participants, particularly those inside a power relationship with you |
| Plagiarism | Use or presentation of somebody else's work or ideas as your own; a most serious offence; avoided by referencing properly | Honesty | Other researchers, and the body of knowledge |
| Reporting research and collecting data | Four questions about dropping variables, disclosing sources of measures, transparency of methodology, and whether your interpretation is appropriate | Honesty | Readers of your findings, and anyone who acts on them |
Columns one and two reproduce the source's content. Columns three and four are analysis added here.
Do no harm — and where the source says harm comes from
That is a narrower and more specific claim than the principle's name suggests. It does not say harm comes from the interview, or from the subject matter, or from what you ask people to do. It says harm comes from the data — from its use, and from failures of privacy and confidentiality.
For a project management study that is almost certainly the right call. You are unlikely to injure anyone. You are quite likely to hold a file recording which named individuals said what about which named project, and to move it between a laptop, a cloud drive and a supervisor's inbox.
The practical consequence of locating harm in the data is that harm control for most project studies is information handling, not welfare procedure. The things you actually have to get right are listed, indirectly, in the ethics application's own requirements.
- How the data will be recorded — for example audiotape, videotape, or written notes
- What happens to the recorded data once it has been analysed
- How, where and in what form the data will be stored
- Whether the data will be individually identifiable, re-identifiable (coded), or non-identifiable
- Who, apart from you and your supervisors, will have access to the data and the results, and on what conditions
Those five come from the application structure the source reproduces, and the storage and identifiability items carry stated requirements behind them — see Research Data Storage, Retention and Ownership. The harm question in its broader form is covered in Protecting Participants from Harm.
Free choice — two shapes of coercion
Notice that the first of the two requires nothing to be done to the participant at all. It is a state of mind produced by the situation — the participant feels they cannot refuse. Nobody has to threaten anything for coercion to be present.
Coercion by position — nothing is said or done
- The participant reports to you, directly or through one line
- You are on the same delivery team and the study is about that team's work
- The participant's contract or access depends on the organisation you represent
- You are the only route to something the participant wants
- Everyone else visibly agreed
Coercion by penalty — refusal costs something
- Non-participation is recorded, noticed or reportable
- Access, information or opportunity is withheld from those who decline
- Declining requires an explanation and agreeing does not
- Participation is bundled into an activity people are required to attend
The source names the two mechanisms; the specific forms listed under each are synthesis added here for a project setting. The distinction is worth keeping because the two are fixed differently. Penalty coercion is removed by changing the process. Position coercion usually cannot be removed at all, only mediated — most often by taking yourself out of the approach entirely.
The version of this problem you will actually meet is milder and more common: a study inside your own workplace, where the participants are colleagues. That case has its own treatment in Dependent Groups and Unequal Power, and the consent mechanics are in Voluntary Participation and Informed Consent.
Plagiarism — the one the source rates most seriously
Three things are doing work in that passage. The definition covers ideas, not only words, which means paraphrasing an argument without attribution is inside it. The severity language is the strongest anywhere in the ethics material — no other principle is called a most serious offence.
And the remedy is the only concrete instruction the source attaches to any of the four principles: reference everything you take from any and every source. It also frames referencing as a skill to be learned rather than a formatting task, which is a fair description of what it takes to do it reliably under thesis conditions.
Structurally, plagiarism is the odd one out among the four. It protects nobody in your study. Participants are unaffected, the sponsor is unaffected, and no data is at risk. What it protects is other researchers and the accumulated body of knowledge you are trying to add to. The mechanics are in Plagiarism and Research Integrity, and the reference formats themselves in Harvard Reference Anatomy: Journal Articles and Books.
Reporting and collecting data — the principle that is four questions
The fourth principle is different in kind from the first three. Where those assert something you can act on, this one asks. The slides label it Validity.
The question the source leaves hanging
The first question is worth spending a paragraph on, because the shape of the problem is learnable even though the source declines to rule on it.
A variable dropped before collection is a design decision. You narrowed the study, you can say why, and it becomes a declared delimitation. The same variable dropped after collection, because of what it produced, is a decision made in the light of the result — and that changes the meaning of everything that survives, because the surviving results are now a selection rather than a set.
The source does not say this is wrong. It asks whether you can. What the rest of the supplied material happens to model, though, is instructive.
The other three questions all lean the same way, towards writing down what you did and what you decided. Disclosing the original source of measures and scales, transparency in methodology, and checking that your interpretation matches what you actually collected are three faces of the same discipline: making the reasoning visible enough for someone else to disagree with it.
Working the four into your design
The four principles are not a review you run at the end. Each one attaches to a specific decision you make early, and each one is expensive to retrofit.
Where each principle lands in your project
Evidence that each principle has been applied
- A stated identifiability class for every dataset, decided before collection
- A written description of how a participant could decline at no cost
- A reference for every idea in the draft that is not yours, internal documents included
- Variables, measures and scales named in the proposal with their original sources
- A dated log of every change to the analysis plan, with the reason for each
- Every result reported, including the ones that went the wrong way
What to carry forward
- The four divide in half: do no harm and free choice protect the individuals in your study; plagiarism and honest reporting protect the people who read it.
- The source locates harm in the data — its use, and privacy and confidentiality of information. For a project study that makes harm control an information-handling problem.
- Coercion has two mechanisms and only one of them requires anyone to do anything. Feeling unable to refuse is enough.
- Plagiarism covers ideas, not only words, and the source's clause is 'any and every source' — including internal documents that do not feel like publications.
- The fourth principle is four unanswered questions. Do not accept a summary that supplies answers the source never gave.
- The hanging question is about dropping variables after seeing results. Fix your variables in the proposal and log every later change with its reason and date.
Frequently asked questions
Can I drop a variable that produced no significant result?
The supplied source asks exactly this question and never answers it, anywhere in the material. What can be said is that a variable dropped before collection is a declared delimitation, while one dropped afterwards because of what it produced makes the surviving results a selection. If you do it, disclose it and say why, rather than deciding the question on the source's silence.
Why does the source say harm comes from data rather than from participation?
It states that harm is most likely to arise from use of data collected by the researcher and from issues around privacy and confidentiality of information. For business and project research that is a realistic assessment — the study itself rarely injures anyone, while the file recording who said what about whose project can do real damage if it is mishandled or identifiable.
Is a consent form enough to satisfy free choice?
No. The source names two mechanisms of coercion, and only one of them involves a penalty. The other is a participant who feels they are not in a position to refuse, which a form does nothing about. Where the participant reports to you or works alongside you, free choice has to be engineered into how the approach is made, not asserted on a page.
Does plagiarism apply to internal company documents?
On the source's own definition, yes. It defines plagiarism as the use or presentation of somebody else's work or ideas as your own, and specifies referencing everything you take from any and every source. Unpublished internal material — a methodology document, a consultant's deck, a colleague's analysis — is somebody's work regardless of whether it was ever published.
Why do the slides call the fourth principle Validity?
The slides use that label and add the question of whether your interpretation of the results is appropriate once the data is collected, which is a validity question rather than a reporting one. The study notes head the same material as reporting research and collecting data. Both labels are in the supplied material and the source does not explain the difference.
Are these principles a standard I can cite?
The source describes them as universal and concerned with honesty and the rights of individuals, and introduces the list with 'typically they include'. That is a teaching characterisation rather than a citation of a code. The only hard, code-backed requirements anywhere in the supplied material concern data storage, retention and identifiability.
References and source attribution
- 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.
- Whiteman, N. 2012, Undoing Ethics: Rethinking Practice in Online Research.
- The supplied teaching source: consolidated weekly teaching notes and slide material on research ethics, which supplies the four principles, the two case-study prompts, and the four unanswered questions quoted on this page.
Suggested questions for Ask KEVOS
- Classify my data as individually identifiable, re-identifiable or non-identifiable, and tell me what changes in each case.
- My participants are colleagues on my own team. How do I remove coercion by position rather than just coercion by penalty?
- Draft the section of my proposal that names the original source of every measure and scale I am using.
- One of my variables produced nothing. Help me write the limitation that discloses what I did with it.
- List every kind of internal document in my workplace study that needs a reference.