Transcribing Spoken Data: A-Level English Language (7702)
Welcome to your study guide on transcribing spoken data! In your AQA A-Level English Language course, Component 3 is the Non-Exam Assessment (NEA), titled Language in Action (Component Code: 7702/C). This component contributes \(20\%\) of your total A-Level (out of 100 marks across the NEA portfolio) and includes an independent Language Investigation of \(2,000\) words.
If you choose to investigate spoken language—such as how gender influences conversation, how language is used in the workplace, how parents interact with children, or how regional accents differ—you cannot simply rely on standard written summaries. You must turn audio into written text using rigorous linguistic methods. This process is called transcription.
Don't worry if transcription seems daunting or technical at first! Once you understand the basic symbols and conventions, it becomes a powerful toolkit for unlocking how speech truly works.
---1. Spoken Data in the Language Investigation
Why Transcribe?
Spoken language vanishes the moment it is uttered. To analyse it across language levels (such as phonology, grammar, pragmatics, and discourse structure), linguists create a visual, printed record: a transcript. A transcript captures not just what was said, but how it was said, including pitch, pauses, overlaps, and volume.
Everyday Analogy: Think of a transcript like a musical score. A standard script just gives the actor the words. A musical score tells the performer exactly how long to hold a note, when to pause, how loud to play, and when two instruments play at the exact same moment.
Data Volume vs. Analytical Depth
A very common trap for students is gathering hours of recorded speech and trying to transcribe all of it. AQA examiners strongly advise against this!
• Quality over quantity: Transcribe short, manageable, and highly relevant extracts.
• Depth of analysis: A closely analysed \(2\)-minute recording will earn significantly higher marks than a superficial look at a \(30\)-minute conversation.
• Word count rules: Your transcripts, data tables, and keys are placed in the Appendix and do not count towards your \(2,000\)-word investigation limit.
Key Takeaway: Choose focused spoken samples. Transcribing small, data-rich interactions allows you to explore multiple linguistic frameworks in depth within your \(2,000\)-word limit.
---2. Standard Transcription Conventions (Jeffersonian / CA System)
In linguistic research and AQA A-Level English Language, spoken discourse is usually recorded using modified Conversation Analysis (CA) conventions, also known as Jeffersonian transcription (developed by Gail Jefferson). Whenever you transcribe spoken data, you must provide a clearly labelled Transcription Key at the beginning of your transcript or in your methodology.
Core Transcription Symbols and Definitions
1. Micropause: (.)
A tiny pause of less than a split second (typically \(< 0.2\text{--}0.5\) seconds). It often marks a breath, a momentary hesitation, or a micro-boundary between grammatical units.
2. Timed Pause: (1.2) or (2)
A measurable silence recorded in seconds or tenths of a second. Timed pauses are crucial when analysing turn-taking, dominance, hesitation, or tension.
3. Overlapping Speech: [ ] or //
Square brackets or double slashes indicate the exact point where two speakers talk simultaneously.
Example:
Speaker A: I really thought that was [brilliant]
Speaker B: [no way] it was awful
4. Latching: =
An equals sign shows latching, meaning there is no audible gap or pause at all between two turns or two utterances.
5. Emphatic Stress: underlining or CAPITALS
Underlined words or capital letters show heightened vocal emphasis, increased volume, or a sudden rise in pitch (e.g., I said no or I said NO).
6. Elongation / Sound Stretching: : (colon)
Colons indicate that a vowel or consonant sound has been stretched or prolonged (e.g., ye::s or so:::). More colons represent a longer stretch.
7. Cut-off / False Start: - (hyphen)
A hyphen represents a truncated word, an abrupt stop, or a self-interruption (e.g., I wen- I came home early).
8. Intonation Contours: ? or ↗ (rising) and . or ↘ (falling)
Punctuation in transcription does not follow written grammar rules. A question mark or upward arrow indicates rising intonation (often signaling a question or seeking agreement). A period or downward arrow indicates falling intonation (signaling completion or finality).
9. Non-Verbal / Paralinguistic Features: ((double brackets))
Used to describe contextual sounds and actions that are not spoken words, such as ((laughter)), ((sighs)), or ((coughs)).
10. Breathing Patterns: ·hhh (in-breath) and hhh (out-breath)
A raised dot followed by 'h's represents an audible in-breath (inhalation). 'h's without a dot represent an audible out-breath (exhalation).
11. Uncertain Transcription: (word) or ( )
Single parentheses indicate words that the transcriber is unsure of. Empty parentheses indicate indecipherable speech.
12. Phonetic / Accent Features: /IPA symbols/
When your investigation focuses on accent, dialect, or regional phonology (such as glottal stops or the BATH/TRAP vowel split), use the International Phonetic Alphabet (IPA) rather than informal spelling.
Key Takeaway: CA/Jeffersonian conventions allow you to capture the dynamic, real-time nature of spoken interaction. Always stick to standard symbols and include a key!
---3. Structuring Spoken Data in Your NEA Report
Where Does the Data Go?
To keep your report professional, academically rigorous, and within the \(2,000\)-word limit, follow this layout structure:
1. In the Methodology Section:
• Data Gathering: Explain what equipment was used and how audio was captured.
• Ethics and Consent: Confirm that you secured informed consent and protected participant anonymity.
• System Justification: State which transcription system you used (e.g., modified Jeffersonian) and justify why it was appropriate for your research aims.
• Fidelity: Discuss the limitations of audio clarity or background noise.
2. In the Appendices:
• Place your full, completed transcripts here.
• Number every line: Numbering lines (1, 2, 3...) is essential. This allows you to cite exact evidence in your main analysis (e.g., "As seen in Speaker B's latching turn on line 24...") without quoting huge chunks of text.
4. Pitfalls and Examiner Warnings
Avoid these common mistakes highlighted by AQA examiners:
1. Avoid "Eye Dialect" (Colloquial Spelling):
Do not write words like "wudva", "innit", or "gonna" arbitrarily. Informal spellings can introduce personal bias and obscure grammatical structures. If you are studying phonological variation, use standard orthography alongside IPA symbols (e.g., writing "butter" and noting the glottal stop as \(/ˈbʌʔə/\)).
2. Never Treat Speech as "Broken Writing":
Spontaneous speech naturally contains false starts, fillers, micropauses, overlaps, and non-standard grammar. Never describe these as "mistakes," "bad English," or "poor grammar." Instead, analyse them as functional features of spontaneous real-time processing, turn-taking management, or politeness strategies.
3. Don't Forget the Key:
A transcript without a transcription key loses academic validity. Make sure every symbol used in the transcript is explicitly defined in your key.
4. Don't Ignore Paralinguistics and Context:
A pause or an overlap can completely change the pragmatic meaning of a sentence. Always transcribe pauses, interruptions, and non-verbal actions (like laughter) to understand the full social context.
5. Quick Reference Summary
Memory Aid for Core Symbols (The "P-O-L-I-S-H" Checklist):
• P - Pauses: (.) for micropause, (1.5) for timed silence.
• O - Overlaps: [ ] to show simultaneous speech.
• L - Latching: = for speech connected with zero gap.
• I - Intonation: ? or ↗ (rising) and . or ↘ (falling).
• S - Stress & Sound Stretching: Underlining or CAPS for emphasis; : for elongation.
• H - Halts & Hesitations: - for abrupt cut-offs / false starts.
Final Tip for Success: When you write your analysis, always link your transcription observations back to your research question, theories, and concepts (such as power, gender, occupation, or language change). The transcript is your evidence; your linguistic analysis is what earns the marks!