academic-writing-standards（学术写作标准）

Academic Writing Standards

This skill provides expertise in academic writing standards for peer-reviewed research papers, ensuring clarity, rigour, and adherence to scientific writing conventions.

Core Writing Principles

Clarity and Directness

Prioritise:

• Clarity over eloquence
• Precision over persuasion
• Simple constructions over complex ones
• Active voice wherever possible

Avoid:

• Unnecessary adjectives and adverbs
• Overstatement and hyperbole
• Excessive qualification ("very", "clearly", "significantly", "novel")
• Complex punctuation where simpler alternatives work

Style Transformations

Examples of preferred style:

Wordy: "The results clearly demonstrate that the novel approach significantly outperforms existing methods"
Better: "The approach outperforms existing methods"

Complex: "The model—which incorporates multiple data sources; including case counts, hospitalisations, and genomic data—provides insights"
Better: "The model incorporates case counts, hospitalisations, and genomic data. It provides insights"

Passive: "It was found that the infection rate was increasing"
Active: "We found the infection rate increased"

Hedged: "It appears that the results seem to suggest that there might be a relationship"
Direct: "The results suggest a relationship"

Punctuation Simplification

Avoid semicolons when possible:

Avoid: "The model includes three components; case counts, delays, and reporting rates"
Better: "The model includes three components: case counts, delays, and reporting rates"
Or: "The model includes three components. These are case counts, delays, and reporting rates"

Avoid excessive em-dashes:

Avoid: "The approach—which we developed over three years—shows promise"
Better: "The approach shows promise. We developed it over three years"

Simplify nested clauses:

Avoid: "The method, which incorporates data from multiple sources, including surveillance systems, which track cases daily, and laboratory reports, provides estimates"
Better: "The method incorporates data from surveillance systems and laboratory reports. It provides estimates"

Formatting Standards

Document Structure

• One sentence per line in markdown format
• Maximum 80 characters per line
• UK English spelling (favour, colour, modelling, analyse)
• No trailing whitespace
• No spurious blank lines

Mathematical Notation

• Use proper LaTeX formatting in appropriate contexts
• Define all notation clearly on first use
• Keep mathematical exposition accessible

Citation and Reference Standards

Citation Format Checking

Common formats to verify:

• Pandoc markdown: [@author2024], [@author2024; @other2023]
• Multiple citations: [@first2024; @second2024]
• In-text citations: @author2024 showed that...

Reference Integrity

Check for:

• Placeholder citations: [@placeholder], [@TODO], [@CITE]
• Malformed citations: Missing brackets, typos in citation keys
• Dangling references: Citations in text without corresponding bibliography entries
• Unused references: Bibliography entries never cited

Citation consistency:

• Verify citation keys follow consistent naming (e.g., authorYear, author_year)
• Check citation formatting matches throughout document
• Ensure proper use of et al. in multi-author citations

Bibliography Verification

When.bib file available:

• Cross-reference every citation against bibliography
• Check for missing entries
• Verify citation keys match exactly
• Note any formatting inconsistencies in bibliography

When.bib file unavailable:

• Flag that references cannot be fully verified
• Suggest author independently verify all citations
• Check citation formatting consistency in text

Originality and Attribution

Identifying Potential Issues

Flag text that:

• Uses distinctive phrasing that may be borrowed
• Contains technical descriptions matching common sources
• Includes sequences of concepts in specific order suggesting copying
• Lacks clear paraphrasing when discussing others' work

Not plagiarism checking:

• Cannot definitively identify plagiarism
• Flags passages requiring author verification
• Suggests paraphrasing where appropriate
• Encourages proper attribution

Proper Paraphrasing Guidance

Poor paraphrasing:

Original: "The model incorporates a hierarchical Bayesian structure with conjugate priors"
Poor: "The approach uses a hierarchical Bayesian framework with conjugate priors"

Good paraphrasing:

Better: "We used Bayesian hierarchical modelling with conjugate prior distributions"

Common Writing Issues

Overused Qualifiers

Remove or replace:

• "clearly", "obviously", "evidently" → Often unnecessary, let evidence speak
• "very", "quite", "rather" → Use stronger base word
• "significantly" → Reserve for statistical significance
• "novel", "new" → Show novelty through comparison, don't claim it
• "state-of-the-art" → Demonstrate through benchmarking

Vague Language

Replace with specifics:

Vague: "The model performed well"
Specific: "The model achieved 95% accuracy"

Vague: "We used a large dataset"
Specific: "We used a dataset of 10,000 cases"

Vague: "Results improved substantially"
Specific: "Accuracy improved from 80% to 92%"

Redundancy

Common redundancies to fix:

• "past history" → "history"
• "future plans" → "plans"
• "end result" → "result"
• "basic fundamentals" → "fundamentals"
• "completely finished" → "finished"

Field-Specific Conventions

Epidemiology and Public Health

• Use "infection" not "case" when referring to true infections
• Distinguish "reported cases" from "infections"
• Use "reproduction number" not "R value" in formal writing
• Define abbreviations on first use: "reproduction number (R)"

Statistical Reporting

• Report confidence/credible intervals: "estimate (95% CI: lower, upper)"
• Use "uncertainty interval" for Bayesian analyses
• Report p-values accurately: "p < 0.001" not "p = 0.000"
• Distinguish statistical significance from practical importance

Computational Methods

• Use "implementation" not "coding"
• "Algorithm" for theoretical description, "implementation" for code
• Report computational resources when relevant
• Specify software versions and packages

Review Structure

When reviewing academic writing, structure feedback as:

1. Reference Issues

- Citation formatting problems - Placeholder citations - Missing bibliography entries - Inconsistencies in citation style

1. Attribution Concerns

- Passages requiring verification - Suggestions for better paraphrasing - Unclear sourcing of ideas

1. Style Improvements

- Clarity and conciseness suggestions - Active voice conversions - Simplified sentence structures - Removed unnecessary qualifiers

1. Formatting Issues

- Line length violations - Formatting inconsistencies - Spelling (UK vs US English)

When to Apply This Skill

Use these standards when:

• Reviewing academic manuscripts
• Editing research papers
• Preparing submissions to journals
• Writing methods sections
• Drafting discussion sections
• Revising based on reviewer comments

Maintain scientific rigour whilst improving readability. Always provide specific, actionable feedback with examples.