From jde-skills
Guides analysis to meet Journal of Development Economics (JDE) empirical norms: clustered inference, attrition bounds, measurement error, and disciplined heterogeneity for field/survey data in low- and middle-income settings.
How this skill is triggered — by the user, by Claude, or both
Slash command
/jde-skills:jde-data-analysisThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- The identification is settled but the estimation, inference, or heterogeneity analysis is unconvincing
JDE referees are experienced with the realities of field and survey data in developing countries — clustered sampling, panel attrition, noisy self-reports, seasonality, and small effective sample sizes. Analysis that ignores these reads as naive. Hold the work to these standards:
Because JDE's replication policy lets editors or referees request data, programs, and computational details at the review stage, every number in a table must be reproducible from a script the day you submit. Build the analysis to be auditable, not just presentable.
Hypothetical: a cluster-randomized health-extension experiment, 80 villages, ~30 households each, with 12 percent endline attrition.
| Referee objection | The JDE-norm response |
|---|---|
| "SEs ignore the clustered randomization" | Re-cluster at the randomization unit; wild-cluster boot / RI |
| "Differential attrition biases the effect" | Report attrition by arm; add Lee bounds; show the bound holds |
| "Spillovers violate SUTVA in your villages" | Distance-ring / treated-neighbor checks; bound the leakage |
Treat this skill as an executable review pass, not a prose hint. First lock the development constraint, identification, welfare or distribution margin, and implementation context; then judge whether the current manuscript answers the venue's real reader: development economists who expect a development mechanism, credible design, and policy-relevant external validity.
claim / evidence / risk / manuscript location rows, so the next agent can edit rather than rediscover the issue.resources/official-source-map.md has been checked for volatile rules and the manuscript has one concrete fix for the largest venue-specific risk.【Estimator】+ why it fits the design
【Inference】clustering level / few-cluster method / MHT
【Attrition】rate, differential? bounds?
【Measurement】key variable construction + error handling
【Heterogeneity】pre-specified vs exploratory
【Robustness done / missing】[...]
【Reproducible from code today?】[Y/N]
【Next step】jde-tables-figures
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin jde-skillsStress-tests causal identification strategies (RCT, DID, IV, RDD) for Journal of Development Economics manuscripts in low- and middle-income settings.
Guides construction and auditing of empirical pipelines for the Journal of Human Resources—sample construction, causal estimates with correct clustering, robustness checks, and comparative estimation against prior work.
Builds labor economics analysis samples from CPS/ACS/register data, runs wage decompositions and AKM models, and handles standard errors and robustness checks for JOLE replicability.