Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› psci-topic-selection
psci-topic-selection
GitHub评估实证心理学项目是否适合Psychological Science期刊及选择稿件类型。通过影响力、稳健性、开放科学和格式四个维度进行压力测试，辅助决定是投研究文章还是注册报告，避免低效投入。
Trigger Scenarios
在多个项目或框架中为Psychological Science做选题决策
收到关于工作‘增量’、‘效力不足’或‘过于小众’的反馈
需要在Research Article和Registered Report之间做出选择
计划投稿复制研究或扩展研究
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill psci-topic-selection -g -y
SKILL.md
Frontmatter
{
"name": "psci-topic-selection",
"description": "Use when deciding whether an empirical psychology project fits Psychological Science and which manuscript type to target. The journal favors concise, high-impact, well-powered work whose claims are robust and broadly relevant. Frames the question; it does not collect data."
}
Topic Selection & Fit (psci-topic-selection)
Psychological Science publishes concise, high-impact empirical psychology. The bar is a finding that is important, robust, and broadly relevant — and that can be argued in a very tight format. Use this skill to pressure-test fit before investing.
When to trigger
- Choosing among possible projects or framings for Psychological Science
- A reader said the work feels "incremental," "underpowered," or "too niche"
- Deciding between a Research Article and a Registered Report
- A replication or extension you want to place (it must fit an Article or RR, not a retired Short Report)
The fit test
A strong Psychological Science paper usually clears all four:
- Impact + breadth. It changes how psychologists understand a phenomenon and is relevant beyond one narrow paradigm. State why a psychologist in another area should care.
- Robustness. Adequately powered, with a credible design and analysis; the result should not hinge
on researcher degrees of freedom (see
psci-study-design,psci-data-analysis). - Open-science ready. You can share data and materials (or justify a documented exemption) and, ideally, preregister — these are now requirements/graded factors, not extras.
- Fits the format. The argument can be made with a ≤ 2,000-word Introduction + Discussion and a focused Method/Results. If it needs a sprawling narrative, it is the wrong venue or wrong framing.
Manuscript type
- Research Article — completed, well-powered empirical study (novel, replication, or extension).
- Registered Report (Stage 1) — strong prospective design; reviewed and accepted before data, which protects against publication bias and is ideal for confirmatory or replication work.
- Registered Report with Existing Data — confirmatory analysis on data you have not yet analyzed for this purpose; declare data provenance honestly.
- Commentary / Reply — a brief, focused response to a published paper (≤ 1,000 words).
Fit scoring — worked example (illustrative)
Score a candidate against the four gates before investing. The attention project, two framings:
Candidate A (off-fit): one N = 45 study, surprising induction effect,
no preregistration, request-only data.
Impact/breadth ~ moderate Robustness ✗ (underpowered, single study)
Open-science ✗ (request-only) Format ✓
Verdict: off-fit → reframe as a powered, preregistered package or place elsewhere.
Candidate B (strong): two preregistered studies (N = 240; N = 300),
internal replication, open data + materials with DOIs.
Impact/breadth ✓ (load-bearing premise, clinical inheritance)
Robustness ✓ Open-science ✓ Format ✓ (argues in < 2,000 words)
Verdict: strong fit → Research Article (or RR if framed prospectively).
Manuscript-type decision rules
| If the work is... | Target type | Why |
|---|---|---|
| Completed, powered, results in hand | Research Article | standard high-impact empirical slot |
| Designed but not yet run, confirmatory | Registered Report (Stage 1) | protects against publication bias; strongest credibility |
| A confirmatory test on un-analyzed existing data | RR with Existing Data | declare provenance honestly |
| A focused critique of a published paper | Commentary / Reply | brief, bounded (confirm current limit) |
| A replication of a famous fragile finding | Article or RR | Short Reports / standalone PDRs are retired |
Reviewer / editor pushback at the fit stage
- "Feels incremental" → find the decisive test or boundary condition that a broad audience inherits.
- "Too niche" → if no one outside the subarea cares, reframe the question or choose a specialty venue.
- "Underpowered and surprising" → this is the venue's cautionary archetype; add power and replication before submitting, or expect a Registered Report suggestion.
Anti-patterns
- A single small, underpowered study with a surprising effect (robustness risk)
- A niche paradigm result with no broad relevance
- Choosing a Research Article when a Registered Report would make a confirmatory claim far stronger
- A project whose data/materials you cannot share and cannot justify withholding
Output format
【Question / effect】one sentence
【Impact + breadth】who cares beyond the subarea, and why
【Robustness】powered? design credible? preregisterable?
【Open-science ready】data + materials shareable (or justified)? [Y/N]
【Type】Research Article / Registered Report (S1) / RR-Existing-Data / Commentary
【Fit verdict】strong / needs reframing / off-fit (why)
【Next】psci-literature-positioning
Supplementary resources
../../resources/external_tools.md— preregistration and power-analysis tools../../resources/official-source-map.md— accepted manuscript types and format
Version History
- 1839142 Current 2026-07-05 14:16
