Trigger timing
- After reading through the draft, it feels like "it reads like a policy document / work report"
- Two or more instances of "is of great significance" / "provides a new perspective for..." appear in one paragraph
- The policy recommendations read like the trio of "strengthen, improve, advance"
- The self-check list includes "economic implications not elaborated"
Comparison table (blacklist → whitelist)
| Problem type | Blacklist (delete/revise) | Whitelist (replace with) |
|---|---|---|
| Empty value | "has important theoretical value" | "provides a theoretical framework for understanding X from the perspective of Y" |
| Empty significance | "has important reference value for XX" | "The results of this paper imply that the transmission channels of XX policy should incorporate YY" |
| Vague recommendations | "regulation should be strengthened / institutions improved / reform advanced" | "Regulators should establish a Y system at stage X, with a focus on Z-type firms" |
| Method showing off | "This paper uses a complex DSGE model" | "This paper constructs a DSGE model and introduces X friction to simulate scenario Y" |
| Vague contribution | "enriches the existing literature" | "for the first time uses a staggered DID identification strategy to examine the effect of X on Y" |
| Literature piling | "(Zhang San, 2020; Li Si, 2021; ...)" | Discuss separately by contribution: one paragraph on theoretical literature, one on empirical literature, and one on this paper's position |
| Tables without interpretation | "Table 3 reports the baseline regression results." | "Table 3 shows that a 1 standard deviation increase in X raises Y by 0.04 standard deviations, roughly equivalent to X% of the mean" |
| Equations without meaning | Go directly to the next paragraph after the equation | "The second term on the right-hand side of equation (3) captures the adjustment cost of X; when Y > 0..." |
| Clichéd contribution | "provides a new perspective for XX" | "Unlike the OLS used by XXX (202X), this paper..." |
Writing conventions
Tense
- Describe this paper's methods/data → present tense: this paper uses / this paper constructs
- Report empirical results → past tense: the study finds / we tested
- Explain the economic implications of the results → present tense: this implies / this finding indicates
Person
- Avoid "the author" and "we believe" in conclusive paragraphs
- Can use: "this paper finds / research shows / results show"
- In English, avoid "We feel that..." / "It is believed that..."
Humility and boundaries
- Do not write "this paper proves" — empirical research can only say "this paper provides evidence supporting"
- Do not treat statistical significance as economic significance — also reporteconomic magnitude
Four rules of Chinese academic syntax
- Split long sentences: If a sentence exceeds two lines and contains ≥3 nested "的" characters, split it into two sentences—first give the judgment, then give the conditions. This is especially true for sentences reporting measurement results: "在控制 A、B、C 并考虑 D 的情形下 X 对 Y 的影响依然显著" should be changed to "基准结论稳健。控制 A、B、C 后，……".
- Subjects must not dangle: Delete subjectless sentences such as "通过对……的分析，发现……" and change them to "本文分析……，发现……".
- Numbers and magnitudes: When reporting coefficients in the main text, also give the reference frame (percentage relative to the mean / standard deviation); strictly distinguish "约 X%" from "X 个百分点" and keep this consistent throughout.
- Integrate parenthetical citations into sentences: Author-year parenthetical citations (author, year) should be attached after the point rather than piled up at the beginning of the sentence; when one parenthetical citation contains more than three references, consider whether all of them really need to be listed.
Terminology consistency
A concept should be defined once atits first occurrence, and only the same name should be used throughout thereafter. Common pitfalls:
- "全要素生产率 / TFP / 生产率" used interchangeably—write "全要素生产率（TFP）" at first mention, and use TFP consistently thereafter
- The core explanatory variable is called "数字化转型" in the introduction, "数字化水平" in the model section, and Digital in tables—these three must converge into one Chinese name + one variable name
- "显著" is used only for statistical significance; economic "large / substantial" should not be written as "显著"
Mini comparison (before revision / after revision)
❌ Before revision: "通过使用双重差分模型进行的实证检验可以发现，政策的实施对于企业的创新水平的提升具有显著的正向的促进作用。"
✅ After revision: "双重差分估计显示，政策使企业专利申请量提高约 8%，相当于样本均值的四分之一。"
The revision removed five "的", one dangling subject, and one clichéd "显著促进", and added the economic magnitude.
One paragraph, one summary
The first sentence of each paragraph must be asummary sentence, and subsequent sentences support it. A reviewer who quickly reads the first sentence of each paragraph should be able to reconstruct the article's logic.
Incorrect:
Zhang San (2020) found that... Li Si (2021) found that... Wang Wu (2022) found that...
Correct:
Existing research has reached two opposing judgments regarding the effect of X on Y (Zhang San, 2020; Li Si, 2021). One line of research holds that..., while the other holds that... The disagreement between the two judgments lies in...
Writing self-check list
- The first sentence of each paragraph is a summary statement
- Number of blacklist phrase hits = 0
- Policy recommendations specify "who should do what"
- After each table there is an interpretation of economic meaning, not just numbers reported
- Mathematical formulas have textual explanations of economic meaning
- Tense and person are consistent throughout the text
- Significance and economic magnitude are reported together
Anti-patterns
- "具有重要的 X 意义" throughout—should not appear even once
- Policy recommendations fill a paragraph with the four verbs "加强、完善、推进、深化"
- Method description is longer than result interpretation
- Cram 5 parallel references into the introduction
[27,28,29,30,31]— split them and discuss separately
Companion materials and examples
Paired "❌ before revision / ✅ after revision" examples of 10 types of language errors are available in resources/worked-examples/12-style-before-after.md。
Output format
【黑名单命中】X 处，分别在：[...]
【段首概括率】X / 总段数
【政策建议可操作性】高 / 中 / 低
【经济含义阐述】到位 / 缺失 [...]
【时态/人称一致性】一致 / 待统一
【下一步】er-submission