Methodology and attribution

PyQCA is a software implementation, not a substitute for methodological judgment. Researchers remain responsible for calibration decisions, directional expectations, truth-table thresholds, remainder handling, interpretation, and robustness assessment.

Related methods

Core QCA

PyQCA uses established QCA concepts including truth tables, necessity, sufficiency, complex/parsimonious/intermediate solutions, consistency, and coverage. The R QCA package and Dusa’s published work are important methodological references.

Threshold sweep

qca.ThresholdSweep is a Python-native adaptation of the algorithmic structure published by ThSQCA. Its MIT attribution is preserved in THIRD_PARTY_NOTICES.md.

Machine-learning-enhanced QCA

qca.mlqca is methodologically informed by Huang’s mlQCA protocol. PyQCA independently implements the workflow using its own data contracts, QCA engines, search, reporting, and stability analysis.

Source and license audit

See the repository files:

• THIRD_PARTY_NOTICES.md for third-party attribution;

• PROVENANCE_AUDIT.md for the source-origin review;

• CITATION.cff for software citation metadata;

• LICENSE for the MIT License.

Key mlQCA article

Qin Huang (2025), “Towards machine-learning enhanced QCA: optimizing coverage and empirical significance”, Quality & Quantity, 59, 4259-4281. https://doi.org/10.1007/s11135-025-02146-2