From public-administration-review-skills
Guides analysis and reporting for Public Administration Review manuscripts, emphasizing uncertainty, robustness, heterogeneity, and reproducibility for quantitative, experimental, or mixed-methods work.
How this skill is triggered — by the user, by Claude, or both
Slash command
/public-administration-review-skills:pubar-data-analysisThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
PAR reviewers are methodologically capable public-management scholars, and the journal endorses the
PAR reviewers are methodologically capable public-management scholars, and the journal endorses the
TOP transparency guidelines — so analyses should be reproducible and documented (see
pubar-transparency-and-data). Because PAR articles carry Evidence for Practice, every estimate
that drives a managerial takeaway must be analyzed honestly enough to bear that weight. This skill
covers execution and reporting; design decisions live in pubar-research-design.
renv.lock, requirements.txt, recorded ssc/net installs).【Main estimate】magnitude + interval + managerial meaning
【Identification check】(per research-design) result
【Robustness】specs that could break it → what held
【Heterogeneity】pre-specified? MHT-adjusted?
【Registered vs exploratory】clearly separated?
【Reproducible】master script + seeds + pinned versions? [Y/N]
【Next】pubar-tables-figures
| Analytic tradition | The check a PAR referee runs first | The fix that earns the benefit of the doubt |
|---|---|---|
| Survey / managerial experiment | Is inference randomization-based and pre-registered? | Randomization inference, pre-registered estimand, MDE reported |
| Observational causal (reform) | Is the "causal" word (and the policy advice) doing more than the design licenses? | State estimand + assumption; sensitivity to an unobserved confounder |
| Performance / administrative data | Are measures validated, and is gaming/selection ruled out? | Construct validation, reliability, selection checks |
| Mixed methods | Do quant and qual estimates actually corroborate? | Show where they agree, and own where they diverge |
A hypothetical PAR survey experiment tests whether a performance-feedback framing raises frontline managers' willingness to adopt a new reporting tool. The pre-registered ATE is +7.4 points (95% CI 3.0 to 11.8) on a 0–100 willingness scale, randomization-inference p = 0.006. An exploratory subgroup ("low-tenure managers") shows +13 points, but it was not pre-registered and after a Bonferroni adjustment across five exploratory subgroups its interval crosses zero. The disciplined write-up reports the +7.4 confirmatory effect with its interval and a managerial interpretation, flags the +13 figure as exploratory and not multiplicity-robust, and frames it as a hypothesis — so the Evidence-for-Practice point rests on the confirmatory estimate only. (All numbers illustrative.)
../../resources/external_tools.md — estimation, inference, and survey packages../../resources/official-source-map.md — TOP transparency policynpx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin public-administration-review-skillsDefends research designs for PAR manuscripts: causal inference (DiD, IV, RD, experiments), case comparisons, process tracing, and mixed methods. Strengthens design justification against reviewer critiques.
Guides data analysis and reporting for JPART manuscripts to meet expert review and data-code release requirements. Covers honest uncertainty, robustness checks, and public-administration-specific biases.
Guides analysis and reporting for APSR manuscripts to survive expert double-anonymous review. Covers honest uncertainty, robustness, heterogeneity, and reproducibility.