ワンクリックでManusで任意のスキルを実行

alterlab-qca

スター46

フォーク6

更新日2026年7月2日 13:45

Runs Qualitative Comparative Analysis — crisp-set (csQCA), multi-value (mvQCA), and fuzzy-set (fsQCA) — for small-to-medium-N configurational research: calibrating raw data into set membership, building and refining a truth table, and Boolean minimization into conservative / parsimonious / intermediate solutions with consistency and coverage. Because there is no maintained Python QCA library, it shells out to R's QCA package (calibrate, truthTable, minimize) via Rscript and documents that dependency honestly rather than faking a Python API. Use when the request mentions QCA, fsQCA, csQCA, configurational or set-theoretic analysis, necessary/sufficient conditions, truth tables, or calibration of conditions. For net-effect estimation of a single treatment prefer alterlab-causal-inference; for interpretive analysis prefer alterlab-qualitative-methods. Part of the AlterLab Academic Skills suite.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

AlterLab-IEU

AlterLab-IEU/AlterLab-Academic-Skills

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

関連職種SOC

SOC 職業分類に基づく

データサイエンティストコンピュータ・数学職·SOC 15-2051

ファイルエクスプローラー

3 ファイル

SKILL.md

readonly

このリポジトリの他の Skills

同じリポジトリ

alterlab-skill-finder

AlterLab-IEU/AlterLab-Academic-Skills

The AlterLab front door and multi-agent launcher — routes a task to the right AlterLab skill(s) when the user invokes the suite without naming one, and for a multi-stage goal (or on the keyword 'alterflow', aliases 'alterresearch' / 'ultralab') it CLARIFIES the goal with a few questions, SELECTS the skills the task needs, and runs a dynamic multi-agent workflow composing them (via alterlab-workflow-orchestration, alterlab-research-pipeline, or alterlab-ssci-orchestrator). Triggers on 'use AlterLab skills', 'which AlterLab skill for X', 'is there an AlterLab skill for…', a multi-stage research goal, 'alterflow …', or any generic AlterLab request where the user does not know skill names. It always asks clarifying questions before executing a multi-step run. Use when someone references AlterLab generically, describes a multi-stage goal, or fires the alterflow keyword; when the user already names a specific skill, defer to that skill directly. Part of the AlterLab Academic Skills suite.

2026-07-0246

alterlab-meta-analysis

AlterLab-IEU/AlterLab-Academic-Skills

Runs quantitative meta-analysis — computes effect sizes (Hedges' g / standardized mean difference, log odds/risk ratios) with their variances, pools them under fixed-effect and random-effects models, quantifies heterogeneity (I-squared, tau-squared, Cochran's Q), draws forest and funnel plots, and tests publication bias (Egger's regression, trim-and-fill) — using statsmodels.stats.meta_analysis in Python or the field-standard R metafor via Rscript. It enforces PRISMA reporting and the random- vs fixed-effect decision. Use when pooling effect sizes across studies, running a systematic review's quantitative synthesis, or assessing heterogeneity and publication bias. For finding and screening the literature prefer alterlab-deep-research; for a single study's statistics prefer alterlab-statistical-analysis. Part of the AlterLab Academic Skills suite.

2026-07-0246

alterlab-missing-data

AlterLab-IEU/AlterLab-Academic-Skills

Handles missing data with principled methods — forces an explicit MCAR / MAR / MNAR mechanism statement, then applies multiple imputation by chained equations (MICE) with Rubin's-rules pooling of estimates and standard errors, or full-information maximum likelihood (FIML) where a likelihood/SEM model applies. Uses statsmodels MICE / MICEData in Python or the field-standard R mice via Rscript, and warns that single (mean/regression) imputation and scikit-learn's IterativeImputer return one completed dataset without Rubin's-rules pooling, so they understate standard errors if used as multiple imputation. Use when a dataset has missing values, when choosing an imputation strategy, or when reporting how missingness was handled. For general modeling on complete data prefer alterlab-statistical-analysis; for latent-variable models with FIML prefer alterlab-sem-psychometrics. Part of the AlterLab Academic Skills suite.

2026-07-0246

alterlab-multilevel-models

AlterLab-IEU/AlterLab-Academic-Skills

Fits and reports mixed-effects / multilevel / hierarchical models for clustered, nested, longitudinal, and repeated-measures data — random intercepts and slopes, variance components and the ICC, cross-level interactions, and GLMMs (logistic/Poisson) — using statsmodels MixedLM and bambi (Bayesian on PyMC) in Python, or the field-standard R lme4 / glmmTMB / brms via Rscript. It enforces the reporting items reviews find under-reported: full fixed + random specification, centering, variance components + ICC, estimation method, assumption checks, model comparisons, and effect sizes. Use when data are grouped/nested (students in schools, repeated measures, panel/longitudinal) and the question concerns within- vs between-cluster variation. For general single-level regression prefer alterlab-statsmodels; for panel fixed effects used for causal identification prefer alterlab-causal-inference. Part of the AlterLab Academic Skills suite.

2026-07-0246

alterlab-qualitative-analysis

AlterLab-IEU/AlterLab-Academic-Skills

Analyzes qualitative data as a dispatched pipeline module — codebook development, thematic / framework / content analysis, and inter-coder reliability computed correctly (Krippendorff's alpha as primary via the krippendorff package or a bundled stdlib nominal calculator with bootstrap CIs; Cohen's / Fleiss' kappa via statsmodels) with 95% CIs and thresholds (alpha >= .80 reliable, .667-.80 tentative). It BRANCHES by design: coefficient-based ICR for codebook / content-analytic coding, versus consensus-and-reflexivity for reflexive thematic analysis where a statistic is not the right criterion. Supports human-vs-LLM double-coding with an alpha check against a human gold standard. Use when coding interviews or open-ended text, building a codebook, or reporting intercoder reliability. For topic modeling / embeddings / supervised text classification prefer alterlab-text-as-data; for the reflexivity gate prefer alterlab-ssci-reflexivity-gate. Part of the AlterLab Academic Skills suite.

2026-07-0246

alterlab-ssci-orchestrator

AlterLab-IEU/AlterLab-Academic-Skills

Coordinates a social-science study through the stage-gated methods pipeline — question then design-gate then measurement-gate then sampling-gate then the right analysis module then inference-gate then reporting — holding a single YAML Design Passport that each gate reads and appends, and enforcing gate order with PASS / WARN / BLOCK semantics (design-gate and inference-gate are fail-closed). It is thin: it routes and holds the artifact, it does not run analysis itself. Use when a researcher describes a whole social-science study end to end, asks to run the methods pipeline or a full methodology review, or wants the design-to-inference workflow coordinated rather than a single step. For a single stage trigger that gate directly (alterlab-ssci-design-gate, -measurement-gate, -sampling-gate, -inference-gate); for multi-agent execution mechanics see alterlab-workflow-orchestration. Part of the AlterLab Academic Skills suite.

2026-07-0246

name	alterlab-qca
description	Runs Qualitative Comparative Analysis — crisp-set (csQCA), multi-value (mvQCA), and fuzzy-set (fsQCA) — for small-to-medium-N configurational research: calibrating raw data into set membership, building and refining a truth table, and Boolean minimization into conservative / parsimonious / intermediate solutions with consistency and coverage. Because there is no maintained Python QCA library, it shells out to R's QCA package (calibrate, truthTable, minimize) via Rscript and documents that dependency honestly rather than faking a Python API. Use when the request mentions QCA, fsQCA, csQCA, configurational or set-theoretic analysis, necessary/sufficient conditions, truth tables, or calibration of conditions. For net-effect estimation of a single treatment prefer alterlab-causal-inference; for interpretive analysis prefer alterlab-qualitative-methods. Part of the AlterLab Academic Skills suite.
license	MIT
allowed-tools	Read Bash(Rscript:*)
compatibility	Requires R with the QCA package (install.packages('QCA'); optionally SetMethods). NO maintained Python QCA library exists, so this skill shells to R via Rscript — declare R as a dependency. No API key. On the Anthropic API (no runtime install) document the R requirement and degrade gracefully.
metadata	{"skill-author":"AlterLab","version":"1.0.0","depends_on":"alterlab-ssci-design-gate (configurational vs net-effects), alterlab-qualitative-methods; audited by alterlab-ssci-inference-gate"}

QCA — Set-Theoretic, Configurational, Equifinal

Skill type: ANALYSIS MODULE. Qualitative Comparative Analysis (Ragin) treats cases as configurations of conditions and asks which combinations are consistently sufficient (or necessary) for an outcome. It is not net-effects regression: it assumes equifinality (several paths to the same outcome), conjunctural causation (conditions work in combination), and asymmetry (the path to Y is not the mirror of the path to not-Y).

Core Mission

QCA IS ABOUT COMBINATIONS AND SET RELATIONS, NOT AVERAGE MARGINAL EFFECTS. CALIBRATE, THEN MINIMIZE.

When to Use This Skill

"Run a fuzzy-set QCA / fsQCA on my configurational data."
"Which combinations of conditions are sufficient for the outcome?"
"Calibrate these variables into set membership and build a truth table."
"Test necessary and sufficient conditions across my ~20 cases."

Does NOT Trigger

The request is really about…	Route to	Why not this skill
The net effect of one treatment (DiD/IV/RDD/regression)	`alterlab-causal-inference`	Correlational/marginal effects, not set relations.
Whether a configurational design is even appropriate	`alterlab-ssci-design-gate`	Design routing, upstream.
Interpretive/thematic analysis of qualitative data	`alterlab-qualitative-methods`	Not set-theoretic.
Plain descriptive statistics	`alterlab-statistical-analysis`	No calibration/minimization.

The R path (verified — QCA package, no Python equivalent)

There is no maintained Python QCA library (the only PyPI option, scpQCA, is stale and non-standard). The field standard is Adrian Dușa's R QCA package. Shell to it via Rscript. The three verified calls:

library(QCA)

# 1. Calibrate raw columns into fuzzy (or crisp) set membership.
#    thresholds for the direct method = c(exclusion, crossover, inclusion), ascending.
df$A <- calibrate(raw$A, type = "fuzzy", thresholds = c(4, 6, 8))

# 2. Build the truth table.
#    incl.cut = consistency cutoff; n.cut = min frequency for an observed row.
tt <- truthTable(df, outcome = "OUT", conditions = c("A","B","C"),
                 incl.cut = 0.80, n.cut = 1, show.cases = TRUE, sort.by = "incl")

# 3. Boolean minimization.
#    conservative = no remainders; parsimonious = include "?"; intermediate = supply dir.exp.
sol_c <- minimize(tt, details = TRUE)                       # conservative
sol_p <- minimize(tt, include = "?", details = TRUE)        # parsimonious
sol_i <- minimize(tt, include = "?", dir.exp = c(A=1,B=1,C=1), details = TRUE)  # intermediate

Fit parameters printed: inclS (sufficiency consistency), PRI, covS (raw coverage), covU (unique coverage); necessity uses inclN, RoN. pof() tests parameters of fit for a specific expression; SetMethods adds necessity/sufficiency XY plots, robustness, and clustered-data diagnostics. Full runnable template and the calibration guidance: references/qca_in_r.md.

The discipline QCA demands

Calibration is a theoretical act, not a mechanical one. Anchor the three thresholds (full-out, crossover, full-in) on substantive knowledge; avoid membership exactly 0.5 (it is undefined for logical operations). Justify each anchor.
Report both the parsimonious and intermediate solutions, and be explicit about which logical remainders (counterfactuals) the intermediate solution used — that is where analytic choices hide.
Consistency before coverage. A solution term needs high sufficiency consistency (inclS, commonly ≥ .80) before its coverage is worth interpreting; report PRI to catch simultaneous subset relations in Y and not-Y.
Necessity and sufficiency are separate analyses — test necessary conditions before minimizing for sufficiency; do not read a sufficiency solution as a necessity claim.

Output Template

CALIBRATION:  <per condition: anchors (excl, crossover, incl) + justification>
TRUTH TABLE:  <rows, incl.cut, n.cut, contradictions handled>
SOLUTIONS:    conservative / parsimonious / intermediate (state remainders used)
FIT:          per term inclS, PRI, covS, covU; overall solution consistency + coverage
CLAIM:        configurational (combinations X are sufficient for Y); NOT "X raises Y on average"

References

references/qca_in_r.md — full runnable Rscript template, calibration guidance, fit-parameter glossary, and the graceful-degradation note when R is unavailable.

Part of the AlterLab Academic Skills suite.