From jop-skills
Defends research designs for JOP (Journal of Politics) manuscripts — covering causal identification, experimental/survey design, formal-empirical linkage, and qualitative case selection/process tracing. Does not write code.
How this skill is triggered — by the user, by Claude, or both
Slash command
/jop-skills:jop-research-designThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
JOP is **methodologically diverse** and demanding about each tradition. The design must credibly connect
JOP is methodologically diverse and demanding about each tradition. The design must credibly connect
the argument (jop-theory-building) to evidence — and, because acceptance is contingent on
replicability, it must be one a JOP replication analyst can re-run. This skill is mode-aware: pick
the section matching your work and defend it against the strongest alternative.
jop-literature-positioningjop-replication-and-data-policy).For the single strongest rival explanation, write one sentence: "If the rival were true rather than my argument, the data would look like ___; instead they look like ___." If you cannot, the design does not yet identify the contribution.
A JOP referee asks the same question of every empirical design: would the result survive if the cleanest rival story were true? Match the objection to the design move that closes it; keep diagnostics in the Online Appendix.
| Referee objection | The design answer |
|---|---|
| "Leans on selection-on-observables" | Move to a counterfactual design (DID, IV, RDD) or show E-value/Oster bounds |
| "Parallel trends is asserted, not shown" | Event-study pre-trends plot; heterogeneity-robust estimator |
| "Exclusion restriction is doubtful" | Argue exclusion substantively; report first-stage F and weak-IV CIs |
| "Case selection is convenience" | Justify by design logic; say what the case is a case of |
A hypothetical paper claims a new transparency law cut local corruption, identifying off staggered adoption across municipalities. A referee objects that early adopters were already cleaning up — a selection story. The author answers structurally: an event study shows flat pre-trends, a Callaway–Sant'Anna estimate gives a −0.6 SD drop, 95% CI [−1.0, −0.2] (illustrative), and an Oster bound shows an unobserved confounder would need to be 1.7× the observed covariates to null the effect (illustrative).
【Mode】quant-causal / experiment / formal-empirical / qualitative
【Estimand or claim】what is being identified/shown
【Key assumption(s)】and how each is defended
【Rival ruled out】the adjudication sentence
【Replicable?】fully scriptable for the JOP analyst? [Y/N]
【Next】jop-data-analysis
../../resources/external_tools.md — design/identification packages (R/Stata/Python) and CAQDAS for qualitative work../../resources/official-source-map.md — JOP methodological diversity and replicability-contingent acceptancenpx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin jop-skillsDefends research design for AJPS manuscripts: causal identification, experimental/survey design, formal-empirical linkage, case-based inference. Strengthens arguments without writing code.
Defends research design for APSR manuscripts: causal identification, case selection, process tracing, experimental design, and formal-empirical linkage.
Defends research design for BJPS manuscripts: causal identification, case selection, process tracing, experimental design, and formal-empirical linkage.