Use when defending the research design of an American Journal of Political Science (AJPS) manuscript — causal identification for observational work, experimental and survey-experimental design, formal-empirical linkage, or case-based inference. AJPS reviewers are quantitatively demanding, so identification must license the claim being made. Strengthens the design; it does not write code.
SkillScore83Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 19 of 23 test cases with the skill, and 14 of 23 without it.
Without skill61%
With skill83%
Checks it passed that the model alone missed
Routes the user to ajps-data-analysis as the next workflow step.
Demands explicitly defining what theoretical phenomenon the selected cases are a 'case of'.
Uses the bracketed section headers 【Mode】, 【Estimand or claim】, 【Key assumption(s)】, 【Rival ruled out】, 【Inference】, 【Robustness/sensitivity】, and 【Next】.
and 4 more in the full report
5 fixed14 passed either way4 still failing0 broke
Pass-rate lift
+22 pp
Tokens
+40%
Agent turns
0%
Response time
-25%
Model gemini-3.6-flashMeasured 8/15/2026v119 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
After finalizing the identification strategy, sample specs, adjudication sentence, and inference corrections in the research design stage, what is the specified next step in the submission workflow?
What was checked, without and with the skill
Checked
Without skill
With skill
Routes the user to ajps-data-analysis as the next workflow step.
Missed
Passed
case-16 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md117 linessingle file
Research Design (ajps-research-design)
AJPS publishes many methods but holds identification and inference to a high standard. The design must
credibly connect the argument (ajps-theory-building) to evidence and rule out the strongest rival.
This skill is mode-aware: pick the section that matches your work and defend it on its own terms.
When to trigger
Specifying identification, sampling, case selection, or experimental design
A reviewer questioned causal claims, a confound, external validity, or inference
Preparing a pre-analysis plan before collecting/analyzing data
Justifying why your design adjudicates the rival from ajps-literature-positioning
Quantitative / causal inference
Identification first. State the estimand and the assumptions that license a causal reading
(ignorability, parallel trends, exclusion, continuity). Defend them; do not assert them.
Designs: experiments (incl. survey/conjoint), DID/event study (use modern staggered-adoption
estimators, not naive TWFE), IV (first-stage strength, exclusion, weak-IV-robust inference), RDD
(density/manipulation tests, bandwidth robustness), matching/weighting with balance + sensitivity.
Inference: cluster at the level of treatment assignment; randomization inference for
experiments; small-cluster corrections (wild-cluster bootstrap) when clusters are few.
Sensitivity: how strong must an unobserved confounder be to overturn the result?
Experiments (lab / survey / field)
Preregister the design and primary analyses; report power / MDE; pre-specify subgroups.
Address attention/manipulation checks, attrition, balance, and ethics/IRB and consent (the AJPS
submission portal asks for human-subjects documentation — see ajps-submission).
For survey experiments: sampling frame, treatment realism, and the limits on generalization.
Formal-empirical linkage
Make the empirical test follow from the model's comparative statics, not a loose analogy.
Distinguish predictions unique to your model from those shared with rivals, and test the unique ones.
Case-based / qualitative & multi-method
Case selection justified by design logic (typical, deviant, most/least-likely, paired) — say what
the case is a case of.
Process tracing with explicit tests (hoop, smoking-gun, straw-in-the-wind); state what evidence
would have disconfirmed the argument; plan source documentation (see
ajps-replication-and-verification, qualitative path).
Use when defending the research design of a Comparative Political Studies (CPS) manuscript — cross-national/panel identification, case-based comparison and process tracing, experiments, or multi-method designs. CPS prizes comparative leverage. Strengthens the design; it does not write code.