Ethnography: from a scene to a defensible account
A vivid scene can hide a weak inference. Learn to preserve what was seen, identify whose explanation is being reported and revise an account when a contrary episode appears.
Course topic: Ethnographic Methods and Fieldwork
Study route and exercise: University Reading List editorial project. How we curate · Sources checked:
Scope and responsibility
The free starting point teaches classroom autoethnographic writing; it does not replace a full fieldwork-methods course. All practice here uses fictional notes, with no recruitment, covert observation or personal disclosure required.
Choose your starting point
Start with a short scene
Use Tombro for observation-to-writing exercises before designing a whole study.
Plan a whole project
Use Hammersley and Atkinson for the linked problems of question, access, relations, analysis and ethics.
Improve an existing notebook
Use Emerson, Fretz and Shaw when writing and interpreting fieldnotes is the immediate bottleneck.
Compare the three books
Teaching Autoethnography: Personal Writing in the Classroom
Melissa Tombro · 2016
Free writing workshop
- Choose it when
- You need practice turning a bounded scene into a reflective account.
- Use another route when
- You need comprehensive sampling, access and research-governance guidance.
- Selected reading
- Read “Writing about Spaces and Events”, “The Interview Process” and “Conducting Observations”.
- Limits and trade-offs
- Personal writing offers situated insight; it cannot make another person’s motive directly observable.
Try before choosing
Rewrite a fictional queue scene without using “shy”, “hostile” or “welcoming”.
Check your work: Retain action, sequence and attributed speech; move psychological labels into a separate possible-interpretation column.
Ethnography: Principles in Practice
Martyn Hammersley; Paul Atkinson · Edition: 4 · 2019
Reflexive project design
- Choose it when
- You need to connect a question to cases, access and analysis.
- Use another route when
- Your only need is stylistic revision of a short scene.
- Selected reading
- Read the named research-design, field-relations, analysis and ethics sections against your selected edition’s contents.
- Limits and trade-offs
- A whole-project account demands decisions about evidence and access, not just a polished narrative.
Try before choosing
Specify a question about workshop entry and two observations that could challenge your preferred answer.
Check your work: Each observation could actually count against the explanation; “observe more” is insufficient.
Writing Ethnographic Fieldnotes
Robert M. Emerson; Rachel I. Fretz; Linda L. Shaw · Edition: 2 · 2011
Fieldnote craft
- Choose it when
- You have jottings but lose sequence, speakers or uncertainty when expanding them.
- Use another route when
- You need a general anthropology survey or a complete interview-method manual.
- Selected reading
- Read the discussions of writing scenes, members’ meanings, coding and composing an ethnographic account.
- Limits and trade-offs
- Rich description is constructed through choices; a detailed sentence is not independent corroboration.
Try before choosing
Expand three invented jottings and mark every remembered quote versus paraphrase.
Check your work: No quotation marks are added where wording was not supplied; inference remains distinguishable from description.
Check your preparation
Is a gatekeeper’s explanation an observation?
The utterance is an observed event in the packet; its explanation of others’ motives is an attributed claim requiring further support.
Must I collect data from real people?
No. Use the complete invented packet. Actual research needs appropriate consent, access and institutional arrangements beyond this reading exercise.
What is a negative case?
An episode inconsistent with the current explanation, not merely a disappointing result. State what the explanation predicts before selecting it.
Six steps with a concrete output
Outputs marked ready: 0 / 6
These are your own study marks, not an assessment. They are not sent to this site.
-
Bound the scene
Selected reading: Tombro: spaces/events and observations.
Make this: Create a four-row table of time, action, attributed speech and uncertainty from the packet.
Check your work: Every entry identifies its source; missing speech is not invented.
-
Separate explanations
Selected reading: Fieldnotes: members’ meanings.
Make this: Write distinct gatekeeper and analyst explanations of the same wait.
Check your work: Neither voice is silently promoted to the other.
-
Make a rival testable
Selected reading: Ethnography: design and cases.
Make this: Derive one predicted episode under token access and one under novice exclusion.
Check your work: Predictions differ on a novice who holds a token.
-
Use the contrary episode
Selected reading: Fieldnotes: coding and analysis.
Make this: Update the blanket exclusion claim using the admitted novice.
Check your work: The revision preserves the observed refusals while narrowing the claim.
-
Inspect access and voice
Selected reading: Ethnography: field relations and ethics.
Make this: List whose perspective is absent and a paper-only question that would help.
Check your work: A proposed question is not reported as an interview already conducted.
-
Write a bounded account
Selected reading: Tombro: putting it together; Fieldnotes: writing.
Make this: Produce 250 words with evidence, rival, counterepisode and unresolved issue.
Check your work: The account describes this packet, not the prevalence of a community norm.
Worked exercise
Who is being excluded?
Invented notes from a repair workshop: 09:00 a sign says “Bench access requires a numbered token; tokens were issued yesterday.” At 09:02 two self-described first-time visitors without tokens wait outside the bench area. At 09:04 the coordinator says, “New people prefer watching.” At 09:06 another first-time visitor presents a token and is admitted. At 09:08 an experienced visitor without a token is asked to wait. Assess “novices are barred from the benches” and the coordinator’s motive claim.
Assumptions
Treat these five entries as the entire packet, accurately recorded, with “first-time” and “experienced” supplied. No unrecorded interactions or actual participants are implied. A descriptive rule, an attributed explanation and a causal account are different claims.
- The wait and admission episodes are observable actions in the packet. “Prefer watching” is the coordinator’s account; no waiting visitor’s preference was recorded.
- A blanket novice ban predicts that even a token-holding novice is refused. The 09:06 admission contradicts that prediction. The experienced visitor’s wait also prevents equating every wait with novice status.
- The posted token condition fits all recorded outcomes. It remains possible that yesterday’s token distribution disadvantaged newcomers, but the packet contains no distribution evidence.
Answer and interpretation
The blanket novice-ban claim fails. Token status explains the recorded admission pattern more economically, while unequal access to tokens remains an open rival question. The coordinator’s preference explanation is unconfirmed, not disproved merely because another mechanism fits.
Test the result
Do not turn five selected episodes into an exclusion rate. State a useful missing item: how yesterday’s tokens were offered and to whom.
Change an assumption
Change the admission evidence
Replace the sign with “Any visitor may use a free bench.” Keep the two waiting novices and the coordinator’s statement, but replace the later episodes: at 09:06 the coordinator refuses a first-time visitor at an empty bench, saying “Only regulars”; at 09:08 an experienced visitor is admitted without a token.
- The token-rule explanation no longer has either a posted condition or matching admission evidence. It cannot be carried over from the first packet.
- The explicit “Only regulars” refusal plus admission of an experienced visitor supports a role-based barrier in these episodes, contrary to the open sign.
- The earlier “prefer watching” statement still does not establish the waiting visitors’ preferences. The sign may be aspirational or inconsistently enforced; neither possibility cancels the refusal.
Answer and interpretation
The revised packet supports an enacted regulars-only barrier more strongly than a token barrier. It still does not establish how often the practice occurs, who authorized it or what every visitor wanted.
Test the result
Name the exact changed entries. Competing accounts of policy versus practice remain possible, but an account claiming no role-based refusal must address the supplied direct refusal.
Common mistakes
Narrative as proof
A fluent reconstruction can join events that the notes never connected. Keep sequence and source visible.
Reflexivity as confession
Describe how position and access shape the available evidence; personal disclosure is not itself validation.
One countercase erases all inequality
The admitted novice refutes a universal ban; it does not prove equal opportunity to obtain tokens.
Before moving on
- Separate actions, attributed explanations and analyst claims in a supplied scene.
- Use a rival prediction and a contrary episode to revise an explanation.
- Produce a bounded account with an explicit missing perspective and access question.
Sources and edition checks
Sources checked:
- Tombro: official open book
Tombro’s 2016 open book is CC BY-NC-SA 4.0 except credited exceptions; its classroom autoethnographic focus is narrower than a whole fieldwork course.
- Hammersley and Atkinson: fourth edition
The fourth-edition methods text connects reflexivity to design, access, relations, analysis, writing and ethics. Use named sections in the selected copy.
- Emerson, Fretz and Shaw: second edition
The 2011 second edition treats fieldnotes as writing and analysis, including revision and coding; publisher description is not free access to the whole book.
These are proposed study activities, not official course assignments or a claim that every book has been read cover to cover.