Using language data and research findings (dictionaries, corpora, tables, graphs, statistics)

Using Language Data and Research Findings in Paper 2 Section A

Welcome to your study guide for Using Language Data and Research Findings! If you are preparing for AQA A-Level English Language (7702) Paper 2, Section A (Language Diversity and Change), this guide will walk you through exactly how to use real linguistic evidence—such as dictionaries, corpora, statistics, and empirical research—to build top-mark evaluative essays.

In Section A of Paper 2, you are given an essay prompt asking you to "Evaluate the idea that..." regarding either Language Diversity or Language Change. There is no stimulus text provided in the exam. This means you must bring your own toolkit of data, studies, and statistics to prove your points.

Don't worry if handling statistics and research studies feels intimidating at first! Think of linguistic data like evidence in a detective case: rather than simply saying "language changes over time" or "men and women speak differently", data allows you to prove how, why, and to what extent language behaves the way it does.

1. Dictionaries and Lexicographical Data

Dictionaries are not just books that define words; to a linguist, they are historical records that show how language is standardised, codified, and continuously updated.

Key Functions of Dictionaries in Linguistic Essays:

• Codification and Standardisation: Recording spelling, grammar, and definitions to create a shared national standard.
• Tracking Lexical Loss and Semantic Shift: Demonstrating how words enter the language (neologisms), change meaning (broadening, narrowing, amelioration, pejoration), or fall out of use (obsolescence).
• Prescriptivism vs. Descriptivism: Highlighting the debate between laying down strict "correct" rules versus recording language as it is naturally used.

Core Lexicographical Evidence to Memorise:

• Samuel Johnson's Dictionary of the English Language (1755): Johnson originally set out to fix and stabilize the English language (a prescriptivist goal), but famously realized during his work that language is dynamic and cannot be frozen in time, reflecting the shift toward descriptivism.
• The Oxford English Dictionary (OED): Built on historical principles, the OED tracks the first recorded usage of words, their changing definitions through time, and marks obsolete words.
• Word of the Year (e.g., Oxford, Collins): Annual selections illustrate how cultural, political, and technological developments drive rapid lexical change.

Quick Review: Dictionaries provide concrete historical proof of codification and semantic change. Use Johnson (1755) and the OED to show that language change is natural and ongoing, rather than a sign of "decay".

2. Language Corpora and Computational Data

A corpus (plural: corpora) is a large, principled collection of electronically stored real-world texts (both spoken and written) used by linguists to perform quantitative and qualitative analysis.

Major Corpora to Reference:

• British National Corpus (BNC): A 100-million-word collection of late-20th-century spoken and written British English. It provides an empirical snapshot of real British language use across diverse regional and social contexts.
• Corpus of Contemporary American English (COCA): A massive, continuously updated corpus of American English useful for examining modern lexical trends and grammatical shifts.
• Google Books Ngram Viewer: A digital tool that charts the frequency of words and phrases across millions of digitised books spanning multiple centuries.

Key Corpus Metrics to Know:

• Word Frequency: The raw or normalised count of how often a word or phrase appears over time (e.g., tracking the statistical decline of archaic modal verbs like thee/thou or the rise of multi-word verbs).
• Collocations: Words that frequently appear together (e.g., "heavy rain" vs. "strong rain"). Studying collocations helps reveal cultural associations and semantic shifts.
• Concordances (Key Word in Context / KWIC): A display format showing the target word aligned in the centre with its immediate surrounding text, allowing linguists to analyze grammatical patterns and patterns of usage.

Key Takeaway: Corpora move your essay beyond guesswork. Instead of claiming a word is "rare" or "becoming popular," corpus data provides measurable, objective proof of usage over time.

3. Quantitative Research Findings and Empirical Studies

To score highly in AO2 (critical understanding of concepts and issues), you need a bank of specific empirical studies. Below is a structured summary of the key research datasets across diversity and change.

A. Regional and Social Variation Studies:

• William Labov – Martha's Vineyard (1961): Investigated the diphthong centralisation of /au/ and /ai/. Quantitative findings showed that local fishermen aged 31–45 centralised diphthongs most heavily to establish local identity and resist tourism (covert prestige).
• William Labov – New York Department Stores (1966): Investigated post-vocalic /r/ across three stores (Saks, Macy's, S. Klein). Findings showed that use of prestige /r/ increased with socioeconomic class and in careful speech, demonstrating conscious social stratification.
• Peter Trudgill – Norwich Study (1974): Recorded the non-standard pronunciation of the `-in'` suffix versus standard `-ing` ([n] vs. [ŋ]). Findings revealed that lower-working-class speakers used non-standard `-in'` significantly more, and men in all classes under-reported their standard usage (favouring covert prestige), whereas women over-reported standard usage (seeking overt prestige).
• Malcolm Petyt – Bradford Study (1985): Investigated `h`-dropping at the start of words. Found a direct statistical correlation with social class: upper-middle-class speakers dropped `h` only \(12\%\) of the time, whereas lower-working-class speakers dropped `h` \(93\%\) of the time.
• Lesley Milroy – Belfast Study (1980): Studied high-density and low-density social networks. Found that individuals with tight, closed networks (dense and multiplex) maintained high rates of non-standard, vernacular linguistic features regardless of gender.

B. Gender and Interaction Studies:

• Robin Lakoff (1975): Proposed the Deficit Model, arguing women's language contains features such as hedges, tag questions, and super-polite forms reflecting social insecurity (largely based on qualitative observation).
• Don Zimmerman and Candace West (1975): Conducted a small-scale recording study of cross-sex conversations. Reported that \(96\%\) of all interruptions were made by men, proposing the Dominance Model.
• Pamela Fishman (1983): Analysed 52 hours of conversations between couples. Found that women asked three times as many questions as men and used supportive minimal responses to keep conversations going (termed conversational shitwork).
• Deborah Cameron (2007 – The Myth of Mars and Venus): Provided a critical modern review challenging rigid gender difference models. Cameron argues that gender differences in speech are minimal and that variation is driven by social context, job roles, and identity construction rather than biological gender.

C. Dialect Levelling and Contemporary Language Change:

• Paul Kerswill and Ann Williams – Milton Keynes Study: Studied phonological variation in the "new town" of Milton Keynes. Data demonstrated dialect levelling—the reduction of marked regional dialect forms across generations as children adopted a levelled, standardized south-eastern variety rather than their parents' regional accents.
• Multicultural London English (MLE): Research highlights how inner-city youth from diverse ethnic backgrounds share innovative lexical, grammatical, and phonological features (e.g., glottal stops, the pronoun man, and fronted vowels), demonstrating language change driven by multicultural contact.

Key Takeaway: Always mention the linguistic variable (e.g., [h]-dropping, post-vocalic /r/, tag questions, interruptions) and the numerical trend observed in the study.

4. How to Integrate Data into Section A Evaluative Essays

In the exam, your goal is to evaluate competing explanations. Follow these steps to ensure your data supports a strong academic argument:

Step 1: Triangulate Quantitative Data with Qualitative Explanations

Never present numbers in isolation. Pair statistical patterns (quantitative) with social reasons (qualitative):
• Data: Petyt found \(93\%\) `h`-dropping in the lower-working class vs. \(12\%\) in the upper-middle class.
• Explanation: Explain this through social stratification, overt prestige (aiming for standard forms to signal status), and covert prestige (using non-standard forms to signal group solidarity).

Step 2: Challenge Prescriptive Narratives

Use corpus and lexicographical findings to dismantle complaints about "falling standards":
• When prescriptivists claim that youth slang or dialect levelling is "ruining" English grammar, counter with evidence from Kerswill, Williams, or BNC data to show that language variation is systematic, rule-governed, and natural.

Step 3: Evaluate Methodological Limitations

Show examiners high-level critical thinking (AO2) by evaluating how data was gathered:
• Point out that early studies like Zimmerman and West (1975) used very small, non-representative sample sizes (only 11 white, middle-class conversations).
• Contrast older, deficit-oriented studies with modern research like Deborah Cameron to show how linguistic consensus has evolved.

5. Common Pitfalls to Avoid in Section A

• Pitfall 1: "Name-Dropping" without Linguistic Detail.
Don't: "Trudgill proved that men and women talk differently."
Do: "Trudgill's 1974 Norwich study investigated the non-standard alveolar [n] variant in the `-ing` suffix, demonstrating that working-class males used higher frequencies of non-standard forms due to covert prestige."

• Pitfall 2: Relying on Anecdotes.
Avoid basing claims on personal experience (e.g., "People in my town drop their T's because..."). Always anchor your points in established linguistic concepts, dialect surveys, or corpus evidence.

• Pitfall 3: Forgetting Linguistic Frameworks (AO1).
Ensure you explicitly name the linguistic level you are discussing: phonology (sounds), lexis/semantics (words and meanings), grammar/syntax (sentence structures), or pragmatics (contextual meaning).

• Pitfall 4: One-Sided Arguments.
Because every Section A question begins with "Evaluate the idea that...", you must explore multiple perspectives. Use your data to provide evidence for a view, and then introduce counter-evidence from another corpus or linguist to weigh up the conclusion.

Quick Revision Checklist

Before entering the exam, ensure you can confidently write about:
1. Samuel Johnson (1755) and the OED as markers of standardisation and change.
2. Corpora (BNC, COCA, Ngram Viewer) and key metrics (frequency, collocations, concordances).
3. At least three regional/social empirical studies (e.g., Labov, Trudgill, Petyt, Milroy).
4. At least three gender/interaction studies and their critiques (e.g., Lakoff, Zimmerman & West, Fishman, Cameron).
5. Modern change/levelling research (Kerswill & Williams, MLE).