The English Language NEA has a split personality, and that is its point. Half of it asks you to behave like a linguist: collect real language data, analyse it with the frameworks you have studied, and answer a question you designed yourself. The other half asks you to behave like a writer: produce a piece of original writing in a chosen genre, and then explain, in a commentary, exactly why you made the language choices you made. Together the two halves count for 20% of the A level, and they reward different strengths — which means a student who is weaker in one half can still build a strong overall mark. This guide explains both halves, their word counts, and how to make them work together. For the wider picture of coursework across subjects, see our guide to writing your A Level NEA.
One point of honesty first. Both halves must be your own authenticated work, produced under your centre’s supervision. Nobody can legitimately write either half for you. What good help can do — and what this guide tries to do — is teach you the method: designing an investigation that data can actually answer, and writing a commentary that shows linguistic understanding rather than merely describing what you wrote.
The two halves, with the AQA figures

Boards differ, so confirm your specification — but the AQA version, the most widely taught, is the clearest illustration of the structure. The NEA, called Language in Action, appears in the A level only; there is no NEA at AS. It forms 20% of the overall marks, out of 100 in total, assessed by teachers and moderated by AQA. It consists of a language investigation of 2,000 words, excluding data and quotations, marked out of 50; and a piece of original writing and a commentary, 750 words each excluding quotations, marked out of 50, with 25 marks for each. Other boards split the work differently — Edexcel’s coursework centres on original writing pieces with a commentary, and OCR’s combines an investigation with a poster-style presentation of it — so treat the AQA figures as the model, not as a universal rule, and check your own board’s handbook before planning.
The practical implication is that you are managing two deadlines inside one component. Students who spend all their time on the investigation and dash off the original writing, or vice versa, lose marks they could have banked. Timetable both halves from the start, and treat the commentary as a piece of analytical writing in its own right, not as an afterthought to the creative piece.
The language investigation: a question data can answer
The investigation is the closest thing in A Level English Language to a mini-dissertation. You choose an area of language that interests you, formulate a question or hypothesis about it, collect your own data — spoken, written or digital — and analyse that data using the language levels and frameworks from your course: lexis, grammar, phonology, pragmatics, discourse, and the social and contextual factors that shape language use.
The question is everything, and the most common failure is a question that data cannot answer. “How does gender affect language?” is a field, not a question: it has no defined data set, no comparison, and no way to tell when it has been answered. “Do male and female sports commentators on televised football use different frequencies of evaluative adjectives when describing refereeing decisions?” is a question: it names a data set, a variable, a feature to count and compare, and a context narrow enough to be manageable in 2,000 words. Narrow until you can state, in one sentence, what data you will collect, what you will count or compare, and what result would count as an answer.
Collect data with a method, not by accident. Decide your sample before you collect it: how many texts, speakers or posts, from where, over what period, and selected by what rule. A sample chosen because it was easy to find — your friends, one forum thread, three episodes — is a convenience sample, and you must acknowledge its limits honestly in the write-up rather than pretending it represents a population. Record your data accurately, transcribe speech carefully, and keep the raw data separate from your analysis: examiners want to see the evidence, not only your summary of it.
Analysing the investigation: frameworks, not impressions
Analysis earns marks when it applies frameworks systematically, not when it offers impressions about language in general. Work through your data level by level, and let the patterns in the data drive the structure: if your data shows a striking difference in modality between two speaker groups, that difference becomes a section, supported by counted examples and short, well-chosen quotations from the data. Quantify where counting is meaningful — frequencies, proportions, comparisons between groups — but never let a table do the analysis for you: every figure needs an interpretation, and every interpretation needs a data example.
A workable investigation structure runs: introduction and question, with a brief rationale and any hypothesis; methodology, stating your data, sample, collection method and analytical approach, including limitations; analysis, organised by finding rather than by data item; and conclusion, answering the question directly and stating how far the answer generalises beyond your sample. The methodology section is not administrative filler: on AQA, the investigation is marked across AO1 (methods and terminology), AO2 (analysis of meanings) and AO3 (contextual understanding), and a methodology that justifies its choices — why this sample, why these frameworks, what the limits are — is evidence for all three.
Be honest about what the data does not show. A small sample, a single context, or an ambiguous result are not failures; they are findings, provided you say so and explain what a larger or different study might test next. Examiners trust an investigation that knows its limits far more than one that over-claims from ten questionnaire responses.
Original writing: choose a style model, then earn the commentary
The original writing half asks you to write in a defined genre — on AQA, under one of three broad purposes: the power of persuasion, the power of storytelling, or the power of information — based on a style model you have studied. The style model matters: it is the published example whose genre conventions, register, structure and language features you are deliberately drawing on, and your commentary must show that you understand those conventions, not merely that you can write fluently.
Choose a style model and a genre you can actually imitate with understanding. A political speech, a travel article, a short story opening, an opinion column: each has conventions of address, structure, lexis and syntax that a reader recognises, and your piece should deploy them deliberately. Write the piece first, but keep notes on your choices as you go — the metaphor you chose and why, the sentence length pattern you used for pace, the point at which you shifted register for effect — because those notes become the commentary, and a commentary reconstructed weeks later from memory is always thinner.
The commentary, 750 words on AQA, is analytical writing about your own writing. It should link your choices to the style model and to linguistic concepts: explain what you borrowed from the model and what you adapted, justify your language choices with terminology, and evaluate what worked and what did not. A commentary that merely summarises the plot of your piece, or that says “I used emotive language to persuade the reader” without showing how and where, earns little. Quote your own piece, name the feature, and explain its intended effect on the intended audience — that three-step move, repeated across your main choices, is the commentary’s engine.
Presentation and the traps to avoid
Keep the two halves cleanly separated in presentation: investigation, data and analysis in one place; original writing and commentary clearly labelled in the other, with the style model identified. Watch the word counts: on AQA, the investigation’s 2,000 words exclude data and quotations, and the original writing and commentary are 750 words each excluding quotations from your piece and the style model. Quotations are not a way to smuggle extra words past the limit, and examiners check.
Finally, protect authenticity. Your data must be genuinely collected by you, your writing genuinely written by you, and your commentary genuinely about your own choices. Teacher feedback within your centre’s rules is permitted, but the work submitted must be yours, and your centre will authenticate it as such.
How to get guidance on your Language NEA
If your investigation question will not narrow to something data can answer, your analysis keeps sliding into impression, or your commentary describes your piece instead of justifying it, help at the planning stage is worth far more than rescue at the deadline. Our support here is coaching, not substitution: we help you sharpen an investigable question, plan a defensible method, and comment on drafts you have written, with model material you can study and learn from — both halves stay yours, as they must. Support is priced from £20 per 1,000 words of material reviewed or prepared. To start, send your brief — your board, your question or genre, your data or style model so far, and your deadline — and place an order through the site, email team@unitedacademichelp.uk, or message us on WhatsApp for a quick reply about what help would genuinely move your coursework forward.
Frequently asked questions
How much is the English Language NEA worth? It counts for 20% of the A level. On AQA it is 100 marks in total: 50 for the investigation and 50 for the original writing and commentary together.
What are the two halves? On AQA: a language investigation of 2,000 words excluding data and quotations, and a piece of original writing (750 words) with a commentary (750 words). Other boards split the work differently, so check your specification.
What makes a good investigation question? One that names a data set, a feature to compare, and a context narrow enough to answer in 2,000 words — and that can be answered, or partly answered, by data you can actually collect.
What is a style model? A published example of the genre you are writing in, whose conventions you study and deliberately draw on. Your commentary must show how your choices relate to it.
What should the commentary do? Justify your language choices with terminology, linked to the style model and linguistic concepts, quoting your own piece and explaining intended effects — not summarise what you wrote.
Can someone write my NEA for me? No. Both halves must be your own authenticated work under your centre’s supervision. Legitimate help is guidance — question and genre advice, method planning, and feedback on drafts you have written.
