Investigation Exemplars and Evaluation
Understanding what a top-band investigation looks like is essential for producing one. This lesson examines the characteristics of high-scoring investigations across different topic areas, provides structural exemplars, and guides you through the process of evaluating your own findings — a crucial element that many students neglect. The ability to critically assess your own work demonstrates the sophisticated linguistic thinking that examiners reward with the highest marks.
What Top-Band Investigations Look Like
Moderators and examiners consistently identify several qualities that distinguish top-band investigations (25-30 marks out of 30) from mid-band work (15-20 marks):
| Quality | Top Band | Mid Band |
|---|
| Research question | Precise, focused, linguistically motivated | Broad, vague, or superficially interesting |
| Data | Carefully selected, sufficient, well-presented | Adequate but sometimes poorly selected or too much/too little |
| Analysis | Multi-level, detailed, interpretive, theoretically informed | Single-level (usually lexis only), descriptive rather than analytical |
| Theory | Integrated throughout analysis, critically evaluated | Bolted on at end, used superficially |
| Evaluation | Honest, reflective, considers limitations and alternatives | Brief or absent |
| Writing | Clear, academic, accurate use of terminology | Generally competent but with inconsistencies in register or terminology |
Key Definition: Evaluation — the critical assessment of your own investigation, including its strengths, limitations, the reliability and validity of your findings, and how your conclusions relate to wider linguistic debates.
Exemplar Structure: Language and Gender Investigation
Here is a structural outline of a top-band investigation on language and gender:
Title
"An investigation into the use of hedging and boosting strategies by male and female participants in a mixed-gender university seminar"
Introduction (250 words)
- States the research question clearly
- Defines hedging and boosting with examples
- Outlines key theories: Lakoff (1975) on women's language, Holmes (1995) on hedging as a positive politeness strategy, Cameron (2007) on the social construction of gender
- Explains why a university seminar is an interesting context (academic setting, power dynamics, mixed-gender interaction)
Methodology (350 words)
- Recorded a 45-minute seminar with 8 participants (4 male, 4 female) plus the tutor
- Transcribed 15 minutes of the most linguistically rich interaction
- Explained transcription conventions used
- Addressed informed consent, anonymity, and right to withdraw
- Acknowledged the observer's paradox and steps taken to minimise it
- Defined categories for hedging (epistemic hedges, approximators, shields) and boosting (emphatics, amplifiers)
Analysis (1000 words)
- Lexical analysis: Compared the frequency and type of hedges used by male and female speakers. Found that female speakers used more hedges overall (7.8 per 100 words vs 4.2), but that the types of hedges differed — women favoured epistemic modal verbs ("might," "could") while men used more approximators ("sort of," "kind of")
- Pragmatic analysis: Examined the function of hedges in context. Found that many female hedges served a relational function (maintaining group harmony, inviting others to contribute) rather than indicating genuine uncertainty — supporting Holmes over Lakoff
- Discourse analysis: Examined where hedges occurred in the discourse structure. Found that hedges were most frequent at the start of disagreements, suggesting a politeness function for both genders
- Grammar: Analysed sentence structure and modal verb usage in detail
- Throughout: integrated theory, used specific examples from data, evaluated findings against existing research
Conclusion (250 words)
- Summarised key findings
- Evaluated: partially supported the hypothesis, but with important nuances
- Discussed limitations (small sample, single context, observer's paradox)
- Suggested further research directions
Exemplar Structure: Child Language Acquisition Investigation
Title
"An investigation into the acquisition of past tense morphology by a child aged 2;6 to 3;0"
Introduction (250 words)
- Focuses on the overgeneralisation of past tense -ed (e.g., "goed," "runned")
- Outlines the U-shaped learning curve (correct irregular forms → overgeneralisation → correct forms again)
- References Berko's Wug test, Brown's morpheme acquisition order, and the nativist vs constructivist debate
Methodology (300 words)
- Recorded the child (the student's younger sibling) during play sessions, mealtimes, and story time over 6 weeks
- Transcribed selected 5-minute extracts from each session (6 extracts total)
- Obtained parental consent; child given a pseudonym
- Coded all past tense verb forms as: correct regular, correct irregular, overgeneralised, or other
Analysis (1100 words)
- Morphological analysis: Tracked the frequency of different past tense forms across the 6-week period. Found a clear increase in overgeneralisation from Week 1 (10% of irregular verbs) to Week 4 (45%), followed by a slight decrease
- Lexical analysis: Identified which irregular verbs were overgeneralised first (less frequent ones like "dig" → "digged") vs retained in correct form (highly frequent ones like "went," "was")
- Syntactic analysis: Examined the sentence contexts in which past tense forms appeared — simple sentences vs complex
- Pragmatic analysis: Considered the functions of past tense usage — narrating, reporting, imaginary play
- Linked findings to theoretical debates: the pattern supported nativist claims about rule-learning, but the role of frequency (high-frequency irregulars retained) also supported usage-based approaches