Quiz — Chapter 11: Citing Sources and Avoiding Plagiarism

Target: 70%+ before moving on. Answers and explanations are hidden — try each question before expanding.

Section 1 — Multiple Choice

Q1. Which is not one of the three jobs citation does? A. Establishes credibility B. Enables traceability C. Maintains intellectual honesty D. Guarantees your work passes a plagiarism checker

Answer **D.** The three jobs are credibility, traceability, and intellectual honesty (§11.1). Passing a checker is neither a purpose of citation nor a guarantee it provides — a perfectly cited document can still patchwrite, and a checker can flag honest matches. Citation exists to serve the reader (verifiability) and honesty, not to defeat a detection tool. A and B and C are the three real jobs.

Q2. You're writing a paper for an IEEE computer-science conference. Which in-text citation format is expected? A. (Patel, 2021) B. [4] C. A superscript footnote number D. (Patel 2021, 14)

Answer **B.** IEEE uses bracketed numbers assigned in order of first appearance (§11.2). A is APA (author–date); C is Chicago notes-and-bibliography (footnotes); D is Chicago author–date. Picking the style your venue expects is the whole task — and IEEE for an IEEE venue is non-negotiable.

Q3. A writer keeps a source's sentence structure and swaps about a third of its words for synonyms, then adds a citation. This is: A. An honest paraphrase, because the words were changed B. Acceptable, because the source is cited C. Patchwriting — which is plagiarism, even with the citation D. A direct quotation that's missing its quotation marks

Answer **C.** Patchwriting borrows the source's *expression* (structure + distinctive phrasing); changing some words doesn't make it yours, and the citation credits only the *idea*, not the *words* (§11.3). A is wrong because synonym-swapping isn't paraphrase. B is the most common misconception — a citation does not cure patchwriting. D is wrong because it's not verbatim, so quotation marks alone wouldn't fix it; honest paraphrase or marked quotes of the borrowed phrases would.

Q4. Which of these most clearly requires a citation in a paper for a general (non-specialist) audience? A. Water boils at 100°C at sea level B. Python is a programming language C. "Code reviews catch about 60% of defects before release" D. The Earth orbits the Sun

Answer **C.** A specific statistic from a specific study is contestable and traceable — it needs a source so the reader can verify the number (§11.4). A, B, and D are common knowledge: widely known and uncontested, so a citation would be noise (and would make the writer look unsure of basic facts). Note the boundary is audience-relative — but a *specific finding* needs a citation for any audience.

Q5. What is the single most reliable technique for preventing patchwriting? A. Run your draft through a plagiarism checker before submitting B. Change at least 50% of the words in any borrowed sentence C. Close the source, write the idea from memory, then reopen to check and cite D. Add a citation to every paragraph

Answer **C.** The close-the-source method removes the source's language from your working memory, forcing you to express your *understanding* rather than edit the source's words (§11.3). B is exactly the synonym-swapping mindset that *produces* patchwriting. A is reactive and catches only verbatim matches, not structural borrowing. D doesn't address the borrowing of *expression* at all.

Q6. "Self-plagiarism" refers to: A. Citing your own prior work too often B. Reusing your own previously-submitted or published work as if it were new, without disclosure C. Writing in a style too similar to your earlier papers D. A myth — you can't plagiarize yourself

Answer **B.** Submitting work as new when it was already submitted or published breaks the implicit claim of novelty and original effort for this context (§11.4) — and can raise copyright issues. It's real, and it surprises people. D is the misconception. The fix is disclosure and self-citation; legitimate reuse (a thesis from your own papers) is handled openly, not hidden.

Q7. An AI tool gives you a flawlessly-formatted citation to a paper that supports your claim. The most important reason to be cautious is: A. AI formatting is usually slightly wrong B. The paper may not exist — models fabricate plausible-looking citations C. AI can't access paywalled journals D. Using AI is always against the rules

Answer **B.** Language models generate *statistically plausible* text, so they produce real-looking citations (real-sounding authors, title, journal, DOI) for papers that were never written — a documented failure mode (§11.6). Every AI-suggested citation must be verified to exist *and* to support the claim before use; pasting one unverified risks citation fabrication. A is sometimes true but minor; D is false (AI use depends on context/policy); C is irrelevant to whether the source is real.

Q8. A reference manager (Zotero, Mendeley) is best described as: A. A tool that guarantees your citations are correct B. A plagiarism detector C. A tool that stores sources and formats citations — whose output you must still verify D. A style that competes with IEEE and APA

Answer **C.** Reference managers store sources and generate formatted citations in any style, making correct citation easier than incorrect — but they format whatever metadata they're given, including *wrong* metadata, flawlessly (§11.5). A overstates it (the accuracy is your responsibility); B is a different category of tool; D confuses a tool with a style. "Check the output" is the load-bearing rule.

Q9. When is direct quotation the right choice over paraphrase in technical writing? A. Whenever you're worried about plagiarism B. When the exact wording matters — a legal/standard definition, a famous phrasing, a contestable claim you want to represent verbatim C. Always — quoting is safer than paraphrasing D. Never — technical writing forbids quotation

Answer **B.** Technical writing cites findings and methods, not prose, so paraphrase is the default; quotation is reserved for the rare cases where the wording itself is the point (§11.3). A is backwards — quoting out of fear, when you should paraphrase, signals you didn't understand the source. C produces quotation-stuffed papers that read as un-digested. D is too strong; quotation has legitimate uses.

Q10. "Mosaic plagiarism" is: A. Copying one source word-for-word B. Stitching borrowed phrases from multiple sources into a paragraph, with citations sprinkled around but the language still borrowed throughout C. Using too many figures from one paper D. Translating a source and presenting it as your own

Answer **B.** Mosaic plagiarism builds a paragraph out of other people's language across several sources; it can pass a naive "did they cite anything?" check while still borrowing the *expression* throughout (§11.4). A is straightforward verbatim copying; D is the translation trap (also plagiarism, but distinct); C isn't a recognized category. The fix is the same close-the-source discipline applied across all sources at once.

Section 2 — True/False with Justification

State true or false and justify in one sentence.

Q11. Adding a citation to a patchwritten passage makes it acceptable.

Answer **False.** A citation credits whose *idea* it is, but patchwriting borrows the source's *expression* (structure and phrasing) — that requires honest paraphrase or quotation marks, so a cited patchwrite is still plagiarism (§11.3).

Q12. There is one universally correct citation style that all serious technical writing should use.

Answer **False.** A citation style is a *convention*, not a correctness standard; the right style is whichever your audience/venue requires (IEEE, APA, Chicago, ACS), and the real rule is to apply your chosen style *consistently* (§11.2).

Q13. If you genuinely can't find a source for a claim you know is true and widely accepted, the honest move is to leave it uncited rather than invent a plausible citation.

Answer **True** (with a refinement): leave it uncited but *flag the uncertainty in the prose* — "research generally finds…," "it is widely held that…" — which is the book's Tier 2 move; manufacturing a citation to look authoritative is the dishonest shortcut to avoid (§11.7).

Q14. Figures and datasets can be plagiarized, not just text.

Answer **True.** A chart, diagram, or dataset is authored intellectual work; reusing or even redrawing one without credit ("adapted from [source]") is plagiarism just as copying prose is — and the caption is the natural place to credit it (§11.4, connecting to [Ch 9](../chapter-09-visuals-and-data/index.md)).

Q15. Using a reference manager means you no longer have to check your citations for accuracy.

Answer **False.** Managers format whatever metadata they're fed, including wrong metadata, flawlessly; accuracy is non-delegable, so you must read every generated reference against the real source (§11.5).

Section 3 — Short Answer

Two to four sentences each. Model answers and rubrics below.

Q16. Explain why paraphrase is the default in technical writing while direct quotation is common in the humanities.

Model answer + rubric Technical writing cites *findings, methods, and data* — the reader needs the result, not the author's exact wording — so you restate the substance in your own words and cite it. The humanities often cite *texts*, where the precise language is itself the object of analysis, so quotation preserves what matters. Reserve quotation in technical writing for the rare cases where exact wording is the point (a standard's mandatory language, a famous phrasing, a contestable claim). **Rubric:** names the findings-vs.-prose distinction (1) and connects it to *what the field is citing* (1).

Q17. Describe the close-the-source method and explain why it prevents patchwriting better than "try to use your own words."

Model answer + rubric Read until you understand the idea, *close the source*, write the idea from memory in your own words, then reopen to check facts and add the citation. It works because patchwriting is driven by proximity to the source's language — when the original is visible your brain edits its words; when it's closed, the only thing in working memory is your *understanding*, so the words you produce are your own. It also doubles as a comprehension check: if nothing comes out, you didn't understand it yet. **Rubric:** states the steps (1) and explains the *working-memory* mechanism (1).

Q18. Name the three integrity problems that arise with AI-generated text and give the one-line rule for each.

Model answer + rubric (1) **Authorship/disclosure** — don't present AI prose as your original work where original work is expected; follow your context's policy and disclose when required. (2) **Fabricated citations** — verify that every AI-suggested source actually exists and supports the claim before using it. (3) **Outsourced thinking** — if you can't evaluate whether the output is correct, you can't legitimately present it as your work. **Rubric:** all three problems named (1.5), each with its actionable rule (1.5).

Section 4 — Applied Scenario

Q19. Below is a real source and a student's paraphrase. (a) Decide whether the paraphrase is honest or patchwriting, and justify. (b) If it's patchwriting, rewrite it as an honest paraphrase with a citation.

SOURCE (Strunk & White, The Elements of Style, Rule 17):
"Omit needless words."

STUDENT'S VERSION:
"Remove unnecessary words (Strunk & White, 2000)."
Answer + rubric (a) This is a borderline case worth thinking through: the source is only three words, and the student's version is a near-synonym substitution of all of them ("omit→remove," "needless→unnecessary"). For a three-word maxim, that's effectively patchwriting — the structure and idea are reproduced one-for-one — and the phrasing "Omit needless words" is *famous*, which argues for quoting it rather than paraphrasing. (b) The honest move is to *quote* the memorable phrase: *Strunk and White's famous rule is to "omit needless words" (Strunk & White, 2000, p. 23)* — or to genuinely reframe the idea: *Concise writing means cutting every word that doesn't carry meaning (Strunk & White, 2000).* **Rubric:** recognizes that very short, famous phrasings are better quoted than paraphrased (1); produces either a properly marked quote or a genuinely reframed paraphrase with citation (1).

Q20. You're submitting a paper to an APA-style journal. Your reference manager generated this entry; the underlying source is a real 2017 conference paper by Vaswani and colleagues titled "Attention Is All You Need." Spot what's wrong and what you must do.

Vaswani et al. (2017). attention is all you need. NIPS.
Answer + rubric Problems: "et al." should not appear in a *reference-list* entry (list all authors there — "et al." is for in-text), the title's capitalization is off (APA sentence-case still capitalizes the first letter: "Attention is all you need"), and "NIPS" is an unexpanded, informal venue abbreviation rather than the proper source (*Advances in Neural Information Processing Systems*). What you must do: *check the manager's output against the real source* and correct the metadata by hand — this is exactly the "verify the output" rule, and the errors are *yours* the moment you submit them. **Rubric:** identifies at least two metadata problems (1); states that verification/correction against the real source is the writer's responsibility (1).

Scoring & Next Steps

Score What it means Do this
< 50% Core concepts not yet solid Re-read §11.1–§11.3 (the why and the patchwriting spine). Redo Part A of the exercises.
50–70% Concepts forming, application shaky Redo Exercises Part B (the patchwriting fixes) and C1 (style conversion). Re-read §11.4 (gray areas).
70–85% Solid — proceed Move to Chapter 12. Do the Project Checkpoint if you skipped it.
> 85% Strong command Try Exercises Part E (build a workflow, audit a paper's citations). You're ready to cite for real.

The skill that matters most here isn't memorizing styles — managers handle formatting. It's telling honest paraphrase from patchwriting on sight, and knowing when you can't vouch for a source. If you nailed Q3, Q5, Q11, Q19, and the AI questions, you have the part that protects you. Onward to Chapter 12: Editing and Revision, where everything in Part II becomes a deliberate revision pass.