Plagiarism and Research Integrity
The source's remedy has a second half most readers skip — it protects you against accusations, not only against the offence. That distinction turns referencing from a formatting job at the end into a records discipline that starts the day you open the first article.
The definition, clause by clause
It is worth reading slowly, because almost every clause forecloses a defence that people reach for. The wording is doing more work than a first pass suggests.
WHAT EACH CLAUSE COVERS AND WHAT IT RULES OUT
| The source's words | What it covers | What it forecloses |
|---|---|---|
| use and/or presentation | Both taking the material and putting it in front of a reader | The distinction between working from something and quoting it. Both are named |
| somebody else's work or ideas | Arguments, findings, structures and framings — not only sentences | The belief that rewording is by itself sufficient. An unattributed paraphrase is inside the definition |
| as your own | The absence of attribution, however it came about | Intent as a defence. The definition does not mention deliberateness at any point |
| a most serious offence | The source's own severity rating | Treating referencing as a style or formatting matter |
| plagiarism and/or accusations of plagiarism | Two separate exposures, only one of which is about your conduct | The assumption that being innocent is the same as being able to demonstrate it |
| everything you take from any and every source | Every source type, with no stated exception | Any private category of material that quietly does not need a reference |
| learn to reference properly | A skill acquired by practice over time | Referencing as a single task performed once, at the end |
Column one is the source's wording. Columns two and three are analysis added here.
The severity language is worth registering separately. Across the four principles the supplied material sets out, no other is called a most serious offence — see The Four Ethical Principles for how it sits alongside the other three.
The offence and the accusation are two different exposures
The remedy clause says you can avoid plagiarism and/or accusations of plagiarism. That is not a rhetorical flourish. It names a second thing that can happen to you, and it is the one most likely to happen to a careful researcher.
Avoiding the offence — about your conduct
- You did not present anyone else's work or ideas as your own
- Every argument you took, you attributed
- Every phrase you kept, you marked as quoted
- This is settled by what you did
Avoiding the accusation — about your evidence
- You can show which source each claim came from
- You can produce the note, with its reference, that the paragraph was built from
- You can distinguish your own summary of an argument from your own contribution
- This is settled by what you kept
The two come apart because of time. Between the week you read an article and the week you write the paragraph there may be six months, three drafts and two changes of research question — and by then the difference between an idea you formed and an idea you absorbed is not something memory reconstructs.
Referencing is a capture problem, not a writing problem
The supplied material treats bibliographic records as a separate discipline, with its own instruction to capture the full reference the moment you find the work. The connection drawn here is that the same discipline is also the whole mechanism of the plagiarism defence — the source does not make that link explicitly.
Put plainly: you cannot reference at write-up what you did not record at reading. The mechanics of building the record are in Note-Taking and Bibliographic Records.
Where attribution is won or lost
At the moment you find the work
The full reference is captured then, in the form you will need later, not a title and a vague memory of the journal. Everything downstream depends on this step and nothing recovers it.
At the moment you take a note
Each note carries its source and its page or section. A note without its source is not a shortened note — it is unusable material that will either be dropped or become a risk.
At the moment you write the note
Mark in the note itself whether what you wrote is a direct quotation, a paraphrase of the author's argument, or your own reaction to it. Three different things that look identical a month later.
At the moment you draft
Carry the marker across with the sentence. Attribution added in a later pass is attribution reconstructed from memory, which is exactly the process that produces wrong citations.
At the moment you revise
Paragraphs get merged, split and moved. Each move is an opportunity for a claim and its citation to separate. Check attribution after structural edits, not only after writing.
How citation discipline fails: a worked example
Supplied alongside the teaching material is a short reflective post — roughly 300 words — in which a student sets out the steps and ethical issues they would address before beginning research on a technical research and development topic. It is described here anonymised, and it is an illustration from one document, not evidence about anything.
The thinking in it is sound. It frames ethics as risk management, ties accountability to the topic's own uncertainty, and sets out a five-item personal list: reducing risk and protecting participant wellbeing, informed consent for stakeholders, confidentiality protection, avoiding misleading research practices, and the right to withdraw at any time.
The referencing is a different story, and it fails in four observable ways that between them cover most of how attribution comes apart.
FOUR OBSERVABLE CITATION WEAKNESSES IN ONE SHORT POST
| What is observable | How it appears on the page | What it costs |
|---|---|---|
| In-text citations with no author and no date | A claim attributed to a web source that carries neither a name nor a year | Nothing can be traced. On the source's own definition, material that cannot be attributed to anyone reads as yours |
| One reference given as a bare URL | A link sitting in the reference list with no author, title, year or access information | Not a Harvard reference and not a durable one. When the address changes the evidence goes with it |
| Bracketed number markers used alongside author-date | Markers in the shape of [A] and [B] in a document that also carries names and years | Two citation systems in one document, neither applied consistently, and the marker system is not the one required |
| A numbered marker with no numbered list to match | A work cited by number where no correspondingly numbered reference list exists | The citation points nowhere. A reader cannot identify the work being relied on |
Observations from a single supplied student post, anonymised. An illustration of a failure pattern, not data about students in general.
None of these four is dishonesty. Each is haste, a system half-remembered, or an article found late and pasted in. And each produces precisely the artefact an accusation feeds on: a claim in your work with no traceable owner.
The observation is worth keeping for one reason. It locates the risk in schedule pressure, which is the same place the ethics case studies in the same material locate theirs — see Ethics Application Failures: What Goes Wrong.
The sources people do not think of as sources
The clause is "any and every source". In workplace research the unattributed material is almost never a journal article, because journal articles arrive already looking like things that need referencing. It is the material that arrives looking like work.
SOURCE TYPES THAT GET MISSED, AND WHAT TO CAPTURE AT THE MOMENT OF USE
| Source type | Why it gets missed | Capture at the time |
|---|---|---|
| An internal methodology or process document | It feels like the organisation's furniture rather than somebody's work | Title, issuing unit, version, date, and where the copy you used came from |
| A standard, framework or body-of-knowledge guide | So widely used that it reads as common knowledge | Title, issuing body, edition or version, year, and the clause or section relied on |
| A consultant's deck or vendor white paper | Delivered in a meeting rather than found in a search | Author or firm, title, date delivered, and the fact that it is unpublished |
| A colleague's unpublished analysis | It was given to you informally, often verbally | Name, role, form of the material, date, and their agreement to be cited |
| A definition or figure taken from an intranet page | No author is visible, so it appears to belong to nobody | Page title, site, date accessed, and a saved copy — the page will change |
| A table or diagram you redrew rather than copied | Redrawing feels like authorship | The original source of the structure, plus a note that you adapted it |
The clause is the source's; the source types and the capture elements are synthesis added here for a workplace research setting.
There is a practical difficulty behind that table. The supplied material's worked reference patterns cover a journal article and a book, and nothing else — see Harvard Reference Anatomy: Journal Articles and Books. The source types most likely to be missed are also the ones with no worked pattern to copy, which is why capturing the raw elements matters more than getting the punctuation right on the day.
Integrity beyond attribution, and where the source stops
Two structural facts about the supplied material set the limits of what it can tell you about integrity.
The first is that plagiarism does not appear on the source's own list of unethical behaviours. That list names violating non-disclosure agreements, breaking participation confidentiality, misrepresenting the results, deceiving people, invoicing irregularities, avoiding legal liability and lying — see Unethical Behaviour in Research. Plagiarism sits on the principles list instead, and the two are never cross-referenced.
The second is that of the four principles, plagiarism is the only one with a stated remedy. Attribution is therefore the only integrity control the source operationalises.
Checking a draft before you submit it
Run this over a finished draft rather than a paragraph. Failures cluster in whole sections — the one written fastest, the one revised most, the one imported from an earlier document.
Attribution check on a complete draft
- Every claim that is not yours has a citation, and every citation has a matching entry in the reference list
- Every reference-list entry appears at least once in the text
- One citation system throughout — no bracketed markers surviving in an author-date document
- No entry in the reference list is a bare URL; each has an author or issuing body, a title, a year and access detail
- No in-text citation is missing an author or a year
- Every direct quotation is inside quotation marks and carries a page or locator
- Every paraphrased argument is attributed, not only the quoted phrases
- Internal documents, standards, decks and intranet pages you relied on are referenced like any other source
- Redrawn tables and figures name the original they are adapted from
- Sections merged or moved during revision have been re-checked, because that is where claims and citations separate
What to carry forward
- The definition covers ideas as well as words and never mentions intent. Rewording without attribution is inside it, and so is doing it by accident.
- The remedy names two exposures. Avoiding the offence is about what you did; avoiding the accusation is about what you kept.
- Referencing is won at reading time. A note without its source is not a shorter note — it is unusable material.
- Mark each line of a note as quotation, paraphrase or your own thought. It costs a keystroke and it is the distinction memory does not preserve.
- The four observed weaknesses — no author, no date, a bare URL, mixed citation systems, a marker pointing nowhere — are haste, not dishonesty, and they produce exactly the artefact an accusation needs.
- "Any and every source" catches internal documents, standards, decks and intranet pages, which are also the source types with no worked reference pattern in the supplied material.
- Attribution is the only integrity control the source operationalises. The rest of the integrity questions it raises are left open.
Frequently asked questions
Is it still plagiarism if I did not mean to do it?
On the source's definition, the question does not arise. It defines plagiarism as the use or presentation of somebody else's work or ideas as your own, and says nothing about deliberateness. That is why the remedy is framed around referencing everything from any and every source rather than around honest intent — the control has to work whether or not you meant anything by it.
Does paraphrasing count if I have rewritten it completely?
The definition covers work or ideas, so an argument put entirely in your own words still belongs to whoever made it. Rewording removes the quotation marks; it does not remove the attribution. The practical test is whether a reader could tell which parts of your paragraph are your contribution and which are somebody else's reasoning.
Do I need to reference internal company documents?
The source's clause is everything you take from any and every source, with no stated exception, and its definition turns on whose work it is rather than where it was published. An internal methodology document, a standard, a consultant's deck or a colleague's unpublished analysis is somebody's work. Capture the issuing body, title, version and date at the time you use it, because those documents change without notice.
What is wrong with using numbered citation markers?
Nothing, in a venue that uses them. The problem observed in the supplied student post is mixing — bracketed markers alongside author-date citations in one document, in a subject that requires the Harvard system, with one marker pointing to a numbered list that does not exist. Consistency is the requirement, and using the system your assessment specifies is the other half of it.
Why is plagiarism not on the source's list of unethical behaviours?
The material does not say. The seven named behaviours concern confidentiality, contracts, money, deception and misrepresentation, and plagiarism appears on the separate list of four ethical principles instead. The two lists are never cross-referenced in the supplied material, so treat the omission as a feature of how the material is organised rather than as a judgement about severity — the same source calls plagiarism a most serious offence.
How do I prove I did not plagiarise if I am accused?
With records, not recollection. The evidence that answers an accusation is the dated note carrying the source it came from, the marker distinguishing quotation from paraphrase, and a reference captured at the time you found the work. None of that can be produced after the fact, which is why the source's remedy is a habit rather than a check.
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 — listed with the ethics material, and relevant to the referencing of online and unpublished sources discussed here.
- The supplied teaching source: consolidated weekly teaching notes and slide material on research ethics, which supplies the plagiarism principle quoted on this page, and an anonymised student forum post supplied with the same material, from which the observed citation weaknesses are drawn.
Suggested questions for Ask KEVOS
- Check my draft chapter for claims that carry no citation and citations with no reference-list entry.
- My reference list mixes bracketed numbers with author-date. Convert it to a single consistent system.
- List every internal or unpublished document type in my workplace study that needs a reference, and what to capture for each.
- Help me design a note-taking template that marks quotation, paraphrase and my own thinking separately.
- I have a paragraph I cannot trace to a source. Walk me through the options and what to do next.