econ-abstract-writing

分类: 调研与分析 | 上传者: dariia-m | 下载: 0 | 版本: v1.0（最新）

撰写学术经济学论文摘要的指南。每当用户需要帮助撰写、起草、修改或构建经济学论文的摘要时，无论是实证微观、发展经济学、应用经济学还是相关领域，都可以使用此技能。当用户提到“摘要”、“论文摘要”或询问如何将他们的发现压缩为简短描述时也会触发。该技能综合了 David Evans (CGDev)、Marc Bellemare 的最佳实践以及顶级经济学期刊（AER、QJE、AEJ：Applied 等）中观察到的模式。 --- # 如何撰写经济学论文的摘要很多人只会阅读论文的摘要来决定它是否值得阅读、分享或引用。有些人甚至连头衔都过不了。摘要是最压缩的推销内容：它必须清晰、快速地告诉读者你做了什么和发现了什么。这项技能主要基于 David Evans 对顶级经济学期刊摘要的分析，并辅以 Marc Bellemare 的写作建议、对摘要可读性的实证研究以及 AER、QJE 和 AEJ：应用论文的常见模式。 ## 有效的证据在开始构建之前，有两个值得了解的经验事实： **可读性预测引用。** Dowling 和其他人检查了《经济学快报》中的摘要，发现单词更简单和句子更短的摘要与更多的引用相关。正如贝尔马尔所说：不要将缺乏可理解性与智力严谨性混为一谈。 **可访问性扩大了你的受众。** Bellemare 的经验法则：如果你的标题不令人反感，并且你的摘要对于你狭窄的子领域之外的人来说是可以理解的，那么你就已经将引用的范围扩大了十倍 - 因为许多人引用的论文他们只阅读了摘要。 --- ## 顶级经济学期刊中的核心结构摘要遵循引言公式的压缩版本。埃文斯指出了好的产品的五种成分

更新日志: Source: GitHub https://github.com/dariia-m/my_claude_skills

目录结构

当前层级: tree/main/abstract/

📄 SKILL.md 11.8 KB

SKILL.md

---
name: econ-abstract-writing
description: >
  Guide for writing the abstract of an academic economics paper. Use this skill whenever the user asks for help writing, drafting, revising, or structuring an abstract for an economics paper - whether empirical micro, development economics, applied economics, or related fields. Also trigger when the user mentions "abstract," "paper summary," or asks how to compress their findings into a short description. This skill synthesizes best practices from David Evans (CGDev), Marc Bellemare, and patterns observed in top economics journals (AER, QJE, AEJ: Applied, etc.).
---

# How to Write the Abstract of an Economics Paper

A lot of people will read no further than the abstract of your paper to decide whether it is worth reading, sharing, or citing. Some will not even get past the title. The abstract is your most compressed sales pitch: it must tell the reader what you did and what you found, clearly and fast.

This skill is based primarily on David Evans' analysis of abstracts in top economics journals, supplemented by Marc Bellemare's writing advice, empirical research on abstract readability, and common patterns from AER, QJE, and AEJ: Applied papers.

## The Evidence on What Works

Before getting to structure, two empirical facts worth knowing:

**Readability predicts citations.** Dowling and others examined abstracts in Economics Letters and found that abstracts with simpler words and shorter sentences were associated with more citations. As Bellemare puts it: do not confuse lack of intelligibility with intellectual rigor.

**Accessibility expands your audience.** Bellemare's rule of thumb: if your title is not repellent and your abstract is intelligible to people outside your narrow subfield, you have expanded the scope of your citations tenfold - because many people cite papers they have only read the abstract of.

---

## The Core Structure

Abstracts in top economics journals follow a compressed version of the introduction formula. Evans identifies five ingredients of a good introduction (motivation, research question, empirical approach, detailed results, value-added). The abstract squeezes a subset of these into 4-7 sentences. The typical pattern:

1. **(Often)** Open directly with the research question and empirical approach
2. **(Sometimes)** Open with one sentence of motivation, then jump into the research question and empirical approach
3. **(Almost always)** Spend most of the space on a detailed discussion of the results
4. **(Sometimes)** Close with one sentence on the implications of the results

That is the whole formula. Notice what is NOT in this list: sentences dedicated to the existing literature. The abstract is not the place to review what other people have done. That belongs in the introduction.

---

## Section-by-Section Guidance

### Opening: Research Question + Approach (1-2 sentences)

The most common opening for abstracts in top journals is to jump straight into the research question, often incorporating the empirical design in the same sentence.

**Pattern A - Lead with the research question and method combined:**
- "We embed a field experiment in [context] to test whether [question]."
- "We use a randomized controlled trial to study [question] in [setting]."
- "This paper estimates the effect of X on Y using [identification strategy] in [setting]."

**Pattern B - One sentence of motivation, then the question:**
Some papers open with a single sentence of motivation before pivoting to the research question. Evans found that no paper in his sample used more than one sentence for motivation in the abstract.

Examples of single-sentence motivations:
- "Despite massive investments in X, there is little evidence on its effectiveness."
- "Relative-pay concerns have potentially broad labor market implications."
- "The delivery of basic health products and services remains abysmal in many parts of the world where child mortality is high."

Each of these is exactly one sentence, and then the next sentence states the research question and design.

**What to avoid:**
- Do NOT spend multiple sentences on motivation. One is the maximum.
- Do NOT dedicate sentences to the existing literature. Papers in lower-ranked journals are more likely to do this, and it wastes precious space.
- Do NOT be vague about what you actually do. The reader should know the question and the method from the first 1-2 sentences.

### Results: The Heart of the Abstract (2-4 sentences)

On average, top-journal abstracts devote at least half their sentences (about 54 percent in Evans' sample) to describing results. This is where the abstract does its real work.

**Be specific.** About half of top-journal abstracts include specific point estimates. Even those that do not still go into substantial detail. "We find a positive effect" is not enough. "We find a 25 percent decrease in child malnutrition" is the standard to aim for.

**Structure your results logically:**

1. **Main result first.** State the headline finding.
2. **Heterogeneity or nuance second.** If there is an important qualification, subgroup difference, or surprising twist, this is where it goes.
3. **Supporting detail or mechanism third.** Additional results that reinforce the story or help explain the main finding.

**Example of well-structured results (from Ashraf et al., 2020, AER):**
- Sentence 1 (main result): Offering career opportunities attracts less prosocial applicants.
- Sentence 2 (nuance): However, the trade-off exists only at low levels of talent; marginal applicants in treatment are more talented and equally prosocial.
- Sentence 3 (downstream effects with a point estimate): These are hired and perform better at every step of the causal chain, including a 25 percent decrease in child malnutrition.

Three out of four sentences in that abstract are about results. That ratio is typical for top journals.

### Closing: Implications (0-1 sentence)

Some abstracts close with a sentence about policy implications or broader significance. This is optional. If you include it, keep it to one sentence. If your results speak for themselves, you can skip it.

---

## Length and Word Limits

**Typical length:** 4-7 sentences for papers in top journals. Shorter abstracts (4-5 sentences) and longer abstracts (6-8 sentences) follow the same basic structure, with longer ones simply having more room for results.

**Journal word limits vary:**
- AER: ~100 words (this is shorter than you think)
- Many journals: 150 words
- Some journals: up to 250 words
- Working papers: no formal limit, but keep it tight

Even when there is no word limit (as with working papers), resist the urge to write a long abstract. The discipline of compression forces clarity. If you cannot summarize your paper in 150 words, that may signal a problem with the paper itself, not just the abstract.

---

## The Abstract Formula in Practice

Here is the formula laid out as a template:

**For a 4-sentence abstract (tight, AER-style):**
1. Research question + empirical approach (1 sentence)
2. Main result (1 sentence)
3. Key detail, heterogeneity, or mechanism (1 sentence)
4. Implication or additional result (1 sentence)

**For a 6-sentence abstract (more room):**
1. One sentence of motivation (optional)
2. Research question + empirical approach (1 sentence)
3. Main result (1 sentence)
4. Important nuance or heterogeneity (1 sentence)
5. Supporting result or mechanism (1 sentence)
6. Implication (1 sentence)

**For a working paper abstract (7-8 sentences):**
1. One sentence of motivation (optional)
2. Research question (1 sentence)
3. Empirical approach and data (1 sentence)
4. Main result with point estimate (1 sentence)
5. Heterogeneity or mechanism (1-2 sentences)
6. Robustness or additional findings (1 sentence)
7. Implication or policy relevance (1 sentence)

---

## Common Mistakes (and How to Fix Them)

| Mistake | Fix |
|---------|-----|
| Opening with literature review | Open with research question or one sentence of motivation |
| Multiple motivation sentences | Cut to one sentence maximum |
| Vague results ("we find a positive effect") | Include magnitudes, point estimates, specific findings |
| No mention of empirical approach | State the identification strategy in the first 1-2 sentences |
| Too much space on method, too little on results | Aim for 50%+ of sentences on results |
| Dedicating sentences to existing literature | Remove entirely; literature belongs in the introduction |
| Abstract reads like a table of contents | The abstract reports findings, not paper structure |
| Jargon that only specialists understand | Use simpler words; readability predicts citations |

---

## Checklist Before Finalizing

Use this checklist to audit a draft abstract:

- [ ] Can a reader identify the research question from the first 1-2 sentences?
- [ ] Is the empirical approach stated clearly?
- [ ] Are the main results reported with specific magnitudes or point estimates?
- [ ] Do results occupy at least half the abstract's sentences?
- [ ] Is there zero or at most one sentence of motivation?
- [ ] Are there zero sentences reviewing existing literature?
- [ ] Would a non-specialist economist understand the main contribution?
- [ ] Does it meet the target journal's word limit?
- [ ] Have you used simple, clear language?
- [ ] Did you write the abstract after finishing the rest of the paper (or at least revise it last)?

---

## When Helping a User Draft an Abstract

When a user asks for help writing or revising an abstract, follow this workflow:

1. **Ask for the key ingredients**: What is the research question? What is the identification strategy? What are the 2-3 most important results (with magnitudes if available)? Is there a key policy implication?
2. **Determine the length constraint**: What journal are they targeting? What is the word limit? For working papers, suggest aiming for 100-150 words.
3. **Draft using the formula**: Open with question + approach (or one sentence of motivation first), then dedicate the majority of sentences to results, optionally close with implications.
4. **Audit against the checklist**: Especially check that results dominate, no literature is reviewed, and the language is accessible.
5. **Iterate**: Tighten phrasing, cut unnecessary words, ensure every sentence earns its place.

If the user provides an existing abstract, diagnose it against the formula. The most common problems are: too much motivation, literature review sentences that do not belong, vague results, and missing empirical approach. Flag these and suggest specific rewrites.

**Key editing principle:** Every word in the abstract competes with every other word. If a word does not help the reader understand what you did or what you found, cut it.

---

## Relationship to the Introduction Skill

The abstract is a compressed version of the introduction. If the econ-intro-writing skill is also available, use both together. The introduction follows the full 7-part structure (motivation, question, approach, results, value-added, optional, roadmap). The abstract compresses this into 4-7 sentences, dropping the value-added/literature discussion and roadmap entirely, and keeping motivation to one sentence at most.

Think of it this way:
- **Introduction**: the full argument for why your paper matters and what it contributes (3-5 pages)
- **Abstract**: the elevator pitch (100-250 words)

The results in both should be consistent and specific. The abstract should never promise something the introduction does not deliver, and vice versa.

---

## References

- David Evans, "How to Write the Abstract of Your Development Economics Paper," Center for Global Development, 2021. https://www.cgdev.org/blog/how-write-abstract-your-development-economics-paper
- Marc Bellemare, writing advice on abstracts and economics papers.
- Dowling et al., research on abstract readability and citations in Economics Letters.
- David Evans, "How to Write the Introduction of Your Development Economics Paper," Center for Global Development, 2020.

登录后下载/点赞/收藏 ❤ 29 | ★ 0

econ-abstract-writing

目录结构

SKILL.md

举报内容

提示