Discussion Vocabulary Diagnosis
Theory basis
Science Research Writing Unit 4.4 (Useful Words and Phrases) — categorised into 5 functional groups — and Unit 4.5 (Modal Verbs) — 6 categories of modality.
5 functional categories (from 4.4.2)
- Map to literature / knowledge — positioning phrases (consistent with, in contrast to, comparable to, in line with, distinct from, contrary to...) and verbs (confirm, contradict, extend, complement, challenge, support, refute, substantiate, resemble...)
- Refine / explore implications — hedging (plausible, potential, tentative; postulate, hypothesise, speculate, theorise; it is conceivable that..., we cannot rule out..., this is reinforced by...)
- Achievement / contribution — "happy words" (accurate, novel, effective, robust, significant, valuable, valid) and "!-substitutes" / "very happy words" (compelling, crucial, remarkable, striking, unprecedented, vital)
- Current and future work — (currently, at present, promising, worthwhile, urgent; further work is needed, future studies should, we recommend, would be of interest)
- Applications / use / applicability — (to apply, to enable, to implement, to generalise, to lead to, to realise, to utilise; applicable, feasible, suitable, viable, practicable)
6 modal-verb categories (from 4.5)
- Able — can / could / be able to (and perfect forms)
- Possible / optional — may / might / could (and perfect forms)
- Expected / likely / probable — should / ought to
- Obvious / impossible — must / cannot (and perfect forms)
- Advisable / recommended — should
- Necessary / essential — must / need to / have to
⚠️ The same form can carry different meanings (e.g., "should" = expected OR advisable; "can" = able OR possible). Context disambiguates.
What this skill diagnoses
- Use of high-frequency academic expressions — does the writer use the right phrase for each function (map / hedge / contribute / future / apply)?
- Modal verb appropriateness — is the modal matched to the intended epistemic stance?
- Colloquialism avoidance — is the register academic throughout (no "a lot of", "we think", "kids", etc.)?
- Terminology consistency — same concept named with the same term throughout?
- Discipline-specific lexis — does the vocabulary match psychology conventions (e.g., "participants" not "subjects" in modern usage, "self-report" not "survey answers")?
Diagnostic signals
| Signal | Means |
|---|---|
| "A lot of studies have shown..." | Colloquial — replace with "A substantial number of studies..." or "Many studies..." |
| "We think that..." | Colloquial hedging — use "We propose / suggest / argue that..." |
| "Kids / subjects / patients" (when "participants" is the field norm) | Discipline-specific register issue |
| "Proved that..." (for a single study) | Overclaim — use "our results suggest / indicate / show that..." |
| "This may can cause..." | Modal stacking — "may" and "can" overlap; pick one |
| "Must" used to express possibility (when "may/might" is meant) | Modal category error — "must" = obvious, not possible |
| Same concept called "stress", "pressure", "strain" within one paragraph | Terminology inconsistency |
Output format
## Vocabulary Diagnosis
### Academic expression coverage (per 4.4.2 categories)
- Map to literature: [✓ / weak / ✗]
- Refine implications (hedging): [✓ / weak / ✗]
- Achievement/contribution language: [✓ / weak / ✗]
- Future work language: [✓ / weak / ✗]
- Applications language: [✓ / weak / ✗]
### Modal verb audit (per 4.5 categories)
- [list each modal found, with intended meaning, with whether it fits]
### Colloquialism hits
- [list with sentence_ref and suggested replacement]
### Terminology consistency
- [list any within-text drift]
### Top issues (severity: critical / major / minor)
1. ...
References (to be filled in Phase 2)
references/academic-vocabulary.md— the high-frequency academic expression list (extracted from 4.4.2)references/modal-verbs.md— the 6-category modal verb guide (extracted from 4.5)references/rubric.mdreferences/checklist.mdreferences/examples/