Answering a Research Methods Assignment
Three anonymised submissions, three briefs reconstructed from activity slides, and one finding that runs through both sides: every fault that recurs in the answers has a counterpart in the material that set them.
Three briefs, none of them supplied as a brief
The three documents supplied under assignment filenames are answers, not tasks. Nothing in the material is a brief document. Each task had to be traced back to the weekly activity slide whose wording the answer matches - a reconstruction by content, and an inference rather than a statement of the source.
THE THREE TASKS, AS RECONSTRUCTED FROM THE ACTIVITY SLIDES
| Answer | The task it matches | Stated length | Deliverable asked for |
|---|---|---|---|
| A | Post your thoughts on two quotations about qualitative and quantitative data, quoted by the slide from page 4 of a 1994 methods text | 225 words | A posted opinion |
| B | Two ethics case studies: what should a researcher refused access do, and discuss your thinking about a covert researcher's reasoning | 300 words for the second case; none stated for the first | A judgement on each case |
| C | Design a data collection tool specific to your research study and discuss why you think this is suitable | 600 words | An instrument, plus a rationale |
The stated lengths are source requirements. Across the weeks these counts vary by a factor of nearly three with no stated basis, and one task carries none at all.
What came back
THE THREE ANSWERS COMPARED
| A | B | C | |
|---|---|---|---|
| Stated length | 225 words | 300 words (second case only) | 600 words |
| Body length | about 217 words | about 285 words, covering both cases | about 590 words |
| References listed | 2 | 2 | 3 |
| In-text citations | 0 | 2 | 1 |
| Citations matched to the list | not applicable | 0 of 2 | 0 of 1 |
| References that are automated web captures | 2 of 2 | 0 of 2 | 3 of 3 |
| References with a usable author | 1 of 2 | 1 of 2 | 0 of 3 |
| Answers the prompt as worded | Partly - the closing thesis sentence reverses the argument the body makes | No - the prompt cannot be answered as set | Partly - the design half is not attempted |
| Deliverable the brief asked for | Delivered | Delivered | Not delivered |
Body lengths are whitespace-token counts of body text excluding headings and reference lists, derived for this library. No submission declares its own length and the source states none.
Every answer is short, and none of them by much
Three in-text citations, and not one of them in a reference list
This is the finding that survives every other reading of these three documents. Two of the three have no in-text citations at all, so a reader cannot tell which claim rests on which source. The third has one, in a bracketed numeral style, pointing at a list that is not numbered and naming an author who is not in it.
The one submission that does cite in the body cites twice, in author-date form with page numbers - and its reference list contains two entirely different works. Both citations look careful. Neither can be followed.
The reference lists have their own problems: an entry whose title field reads "No page title", entries reproducing the page title in the author field, a locator given as a 72-page range, an author field carrying initials and no surname, a several-hundred-character search address, and one time-limited signed download link that carried a five-minute expiry and stopped working within minutes of being generated.
The same faults, in the material that set them
None of this is peculiar to student writing, and treating it as a story about carelessness would miss what the record actually shows. Set the two side by side.
THE FAULT IN THE ANSWERS, AND THE SAME FAULT IN THE MATERIAL
| Fault | In the three answers | In the teaching material |
|---|---|---|
| Citations that lead nowhere | Three in-text citations across three documents; zero appear in any reference list | One week cites a 1994 methods text with a precise page number and omits it from that week's reference list |
| Lists that do not match the content | No list contains the work its body actually leans on | One week lists five works, cites ten others that are not listed, and lists one that is never cited |
| Bare addresses standing in for references | Query-string addresses and one expiring signed download link | The data collection week's whole reference slide is a single bare address that is the source of nothing in that week; three of its set readings are addresses with no author, year, title or page range |
| Sources that cannot be identified | Three entries reproduce the page title in the author field, so none has a human author | The prescribed text is called "Text" throughout four weeks; no author, title, year or publisher is ever given, and a single surname is disclosed once |
| Borrowed material not marked as borrowed | A sixteen-item list introduced with an attribution line and then set out as running prose without quotation marks | A week teaches transcription by reproducing an article's abstract with no framing sentence and no task attached; the journal is not named and three works cited inside it cannot be followed up |
| Terms used and never defined | "Design thinking" appears in two of the three, undefined and unconnected to the argument | Probes, interview guides, standard deviation, spread and statistical assumptions are all used and defined nowhere |
| Text defects that change the meaning | Three sentences across the three where one wrong word reverses the sense | A slide sentence with its verb missing; an instruction reading "stay focussed on the primary focus"; an attribution running together as "sourced fromhttp" |
When the brief makes the answer impossible
Each of the three answers inherits something from the task it was set. Three of those inheritances are structural, and they are worth knowing because the same thing happens to a practitioner handed a vague terms of reference.
Three inherited defects
A prompt asking about reasoning that was never printed
The second ethics case asks the student to discuss the researcher's reasoning. The case as printed states no reasoning. The answer discusses whether the research was justified instead, which is the nearest available task. The case material is on Covert Research and Negotiating Access.
One word count across two tasks
Two case studies sit on consecutive slides and only the second carries a length. The answer divides about 285 words roughly evenly between them. Whether that is 5% short or half the required length for one case cannot be determined from the brief.
A deliverable the material never shows
The design-a-tool task asks for an instrument, and no instrument of any kind appears anywhere in the four weeks supplied. The answer contains none either. That case is worked through on Designing a Data Collection Instrument.
There is a fourth, subtler one. The brief for the first task quotes two lines second-hand from page 4 of a 1994 text, and gives no full reference for that text. The answer drops the intermediary and attributes both lines straight to the people who said them. That is a citation fault, and it is also the only move the slide left available.
What the three answers get right
- All three reach a position rather than surveying considerations. Each one commits: a synthesis of two opposed quotations, an explicit recommendation and judgement on two cases, a stated preference for one collection method with reasons.
- All three relate the task to a research topic of their own rather than answering in the abstract.
- All three cite something, however imperfectly - which is more than several slides of the material that set them do.
- One introduces a distinction the material never supplies, between covert work done for internal organisational purposes and covert work done for publication, and turns an ethics problem into a design problem by recommending the constraint be anticipated at site selection.
- One reaches forward into content it had not yet been taught, and names a coding approach from an external source because the subject supplied none.
How to answer one of these
Nine moves, most of which cost minutes
Reconstruct the brief in writing
Copy the task wording to the top of your draft. Where a single sentence sets two things - design and discuss, or two cases under one count - split the word budget explicitly and say how you split it.
Answer the prompt as worded
If it cannot be answered as set, say so in one sentence, state what you answered instead and why. An acknowledged substitution is a defensible move; an unacknowledged one reads as a misunderstanding.
Cite what you actually read
Where the task quotes something from a text you have not read, either find the original or cite it as quoted in the intermediary. Attributing straight to the originator claims a reading you did not do. See Critical Summary and Paraphrasing.
Close the loop both ways
Every in-text citation appears in the list; every listed work is cited in the body. This is a two-minute check and it is the single most common failure in all three of these documents and in four weeks of the material.
Edit machine-generated references
Delete "No page title" entries, replace page titles sitting in author fields, trim query strings to a stable address, and never cite a link that expires. See Harvard Reference Anatomy.
Mark borrowed material as borrowed
Quotation marks and a locator for quoted text; genuine restatement for a paraphrase. A long list introduced by an attribution line and then run on as your own prose is not marked, whatever was intended.
Define every term you import
If you bring a term in from outside the subject, define it at first use and cite where it came from. An undefined term used once and never returned to adds nothing to an answer.
Read it aloud once
All three of these documents contain a sentence where one substituted word reverses the meaning - and in one case the reversed sentence is the thesis. Silent proofreading does not catch these; reading aloud does.
Spend the word count
Not one of the three used its allowance. If you are under, check the instruction you have not yet carried out before you go looking for something to pad.
Two-minute check before you submit
- The task wording is answered, clause by clause, including the clause you find least convenient
- Every in-text citation is in the list and every listed work is cited
- No reference contains a machine-generated artefact, an expiring link or a title where an author should be
- Any quoted block is inside quotation marks with a locator
- Every imported term is defined at first use
- The piece has been read aloud once, for sense rather than for typing errors
- You are inside the word count and not more than a little under it
What to carry forward
- Three in-text citations across three answers, none of them in a reference list - and four consecutive weeks of the material with no functioning reference list either.
- Where a brief is ambiguous or asks for something it never models, the answer inherits the defect. Name the inheritance rather than absorbing it silently.
- Answer the prompt as worded, or say plainly what you answered instead and why.
- Machine-generated references need editing before they are references. Nothing else on this page is as cheap to fix.
- Whatever you model your work on passes its faults to you. Check the model before you copy its habits.
Frequently asked questions
Why reconstruct the briefs instead of quoting them?
No brief document exists in the supplied material. The three files carrying assignment names are answers. Each task was traced to the weekly activity slide whose wording it matches, which makes the briefs on this page an inference from content rather than a statement of the source, and they are marked as such.
Is a short answer a serious problem?
Not in itself - the three shortfalls here are roughly 1.7%, 3.5% and 5%, which no marker would dwell on. What matters is that in each case the unused words sit next to a part of the task that was not carried out. Treat being under the count as a diagnostic rather than as the fault itself.
What is the single most common referencing fault in these documents?
In-text citations that appear in no reference list: three citations, zero matches. The second is unedited machine-generated entries - page titles in author fields, a title reading "No page title", query-string addresses and one link that expired five minutes after it was created. Both are fixable in the last few minutes of writing.
If the teaching material cites badly, does that excuse the same fault in an answer?
No, and that is not the argument. It changes the diagnosis rather than the standard: work that has never seen a functioning reference list is unlikely to produce one. The practical response is to take your referencing model from a source that does it properly and to say which one you used.
What should I do when a brief asks something that cannot be answered?
Say so, in one sentence, and state what you answered instead. One of these three answers faces a prompt asking about reasoning that the printed case never states; it answers a different question without acknowledging the substitution, which reads as a misunderstanding rather than as the judgement it actually was.
How do I mark a long borrowed list properly?
Quotation marks around the quoted text, a locator to the page or section, and a reference entry that lets a reader find it. If the list is long, introduce it, quote it as a block and attribute it in the same sentence - and do not shift individual items into the first person, which turns a quotation into a claim of your own.
References and source attribution
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford. Listed in one of the supplied submissions and in an earlier week's references of the teaching source.
- Miles, M. & Huberman, M. 1994, Qualitative data analysis - cited on the week 7 activity slide with a precise page number for two quotations, and absent from that week's reference list.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Naoum, S. G. 2013, Dissertation Research & Writing for Construction Students, 3rd ed., Routledge.
- Trochim, W. M. K. 2006, Research Methods Knowledge Base - the data collection week's only listed reference, given as a bare address that is the source of nothing in that week.
- The supplied teaching source: the weekly activity slides of weeks 7, 9, 10 and 11, and three student submissions supplied with that material, used here anonymised.
Suggested questions for Ask KEVOS
- Check my draft: does every in-text citation appear in my reference list?
- Rewrite this machine-generated reference as a proper entry.
- The brief asks two things in one sentence - help me split my word budget between them.
- How should I word an acknowledgement that a prompt cannot be answered as set?
- Give me a two-minute pre-submission check for a research methods assignment.