English Composition › Section O · The case study profile › Topic O.2
Writing a case study
Writing a case study means running a small piece of research yourself: asking a real question, interviewing and observing a participant, and turning your notes into a report that others can trust and build on.
The assignment: a study of language use
A common college assignment asks you to study someone you know, such as a classmate, a friend or a relative, focusing on how they use language. A good participant speaks, or at least sometimes uses, a language other than English, so you can study how their languages and cultures interact: how they shift between home and work, or how one language colours the other. Other possible focuses include body language, or how someone's language changes across texts, calls, email and face-to-face talk. Check your college's policy, get the participant's written permission, and plan a report of about 1,000 to 1,200 words.
Choosing a research question
Everything in the study follows from the research question, so it is worth revising several times. A strong question has four traits. 1. Open-ended: it has several plausible answers, not yes or no. 2. Debatable: reasonable people could disagree about the best answer, and you do not already know it. 3. Answerable: your interviews, observations and background reading can supply evidence for an answer. 4. Consequential: the answer matters to someone, such as the instructors and administrators at your college, because it could change what they do.
Worked example: sharpening a research question
Topic: how a student called Maya uses language at her café job. 1. Draft: "Does Maya use slang at work?" This is a yes-or-no question, and one shift would answer it. 2. Draft: "Is it unprofessional to use slang with customers?" This is debatable, but it asks for a value judgement that observing Maya cannot settle. 3. Revision: "How does Maya adjust her language for customers, coworkers and her manager, and what might that suggest for workplace training at the college?" It is open-ended, debatable in its explanations, answerable from interviews and observation, and consequential for the people who design training.
Background reading and interview questions
Before collecting data, learn what is already known, because a case study is meant to add to a field. Browse academic journals and trade publications; articles that many others cite are usually important, and you should read both landmark studies and recent ones. Next, learn about your audience, for example from your college's fact book or student profile. Then draft 10 to 15 open-ended interview questions, such as "How do you decide which language to use in a group chat?" Plan follow-up probes for details, record answers in notes or, with permission, on a recorder, and follow up when an answer opens a new thread.
Observing in the field
Choose a setting where the participant really uses language in the way you are studying: watching a non-gamer play a video game tells you little. You might observe in person or, with permission, through a group chat or video call. Take detailed, dated notes as you go. Useful things to record include words and phrases the participant uses often, substitutions such as emoji, slang or acronyms, shifts in volume, tone or vocabulary, and insider expressions that the participant's group understands but outsiders might not. Note exact wording when you can, since accurate quotations are your strongest evidence.
Organising the report
The introduction gives the background and context: what you studied, why it matters, a short review of earlier research, your participant and observation environment, and your methods. Most case studies then use headed sections and subsections. A practical plan is to give each criterion you examined its own section, such as greetings, forms of address or emoji. Within each, present the qualitative data, the quantitative data, any additional information, and then your analysis of that criterion. Acknowledge what you could not observe. A planning grid with one row per criterion keeps each section focused.
Worked example: a body paragraph for one criterion
Model paragraph: "Forms of address. Maya addressed her manager as 'Ms Lopez' in all nine exchanges I observed, but called coworkers by nicknames in Spanish and English. In her interview she said, 'With my boss I keep it formal, even when she jokes.' This suggests that Maya treats rank, not friendliness, as the signal for formal address. Because I observed only two shifts, I cannot tell whether the pattern holds with other managers." Analysis: 1. A heading names the criterion. 2. A count supplies quantitative data. 3. An accurate quotation supplies qualitative data. 4. The interpretation is hedged with "suggests". 5. A limitation is stated openly.
Visuals that earn their place
Visuals help with ideas or data too complex for a sentence. A bar chart compares values across separate groups or categories. A line graph shows how a value changes over time. A pie chart shows how parts make up a whole, so its slices must total 100 percent. A table sets out exact values that readers need to look up. Give every visual a clear title and labels, so readers can see its point without your explanation, and tie it to the text. Include a visual because it clarifies your data, never just because one is expected.
Drawing conclusions
Look at all your data together. Which words, choices or behaviours recur across interviews and observations? Base conclusions on consistent behaviour, not on a single striking moment. The most useful conclusions often challenge an assumption you started with, such as the idea that a participant is always more formal in English. Present your conclusion as a theory the evidence supports, explain its relevance to your audience, and name the questions it leaves open. A closing sentence such as "Whether this pattern holds for other first-year students would be worth a wider survey" turns a limit into an invitation.
Peer review and sharing
Electronic peer review, through shared documents, comment boxes and tracked changes, is convenient, gives reviewers time to think, and keeps a permanent, shareable record of feedback. Make comments specific. "Revise this" is vague. "Make this question more debatable" is somewhat helpful. A specific, clear comment says why and how: "This is a yes-or-no question; try asking how her language changes." When revising, accept or reject each suggestion individually, then remove all comments and markup from the final copy. Consider presenting at a student conference or submitting to an undergraduate journal, but only if your participant's consent covers it.
Common mistakes to avoid
First, a research question with an obvious or yes-or-no answer. Second, skipping background reading, so the study cannot show what it adds. Third, closed interview questions that invite one-word answers. Fourth, observing in a setting where the behaviour you are studying never happens. Fifth, generalising from one moment instead of a consistent pattern. Sixth, adding charts that decorate rather than clarify. Seventh, clicking "accept all" on a reviewer's changes without reading each one.
Key takeaways
Start with a research question that is open-ended, debatable, answerable and consequential. Read existing research, then draft 10 to 15 open-ended interview questions and observe the participant in a natural setting, with written consent. Introduce the study with background, context and methods; organise the body by criterion, pairing qualitative and quantitative data with analysis. Use clearly labelled visuals only where they clarify. Conclude from consistent patterns, state limits, and give and use specific feedback.
Flashcards
Flashcard
A small study of one participant's language use, with written consent
Test your understanding
Question 1 of 5
Which research question best suits a case study of one student's language use?
Question 2 of 5
A writer wants to show how many texts a participant sent each day over two weeks. Which visual fits best?
Question 3 of 5
Which peer review comment is the most helpful?
Question 4 of 5
A student saw the participant write one formal email to a manager. How should this shape the conclusion?
Question 5 of 5
When a case study's body is organised by criterion, what does each criterion's section contain?
