Measuring constitutional preferences: A new method for analyzing public consultation data

Area of Science:

• Political Science
• Computational Social Science
• Constitutional Law

Background:

• Public consultations are crucial for constitutional design but generate large, unstructured datasets.
• Existing methods for analyzing this narrative data are limited and lack standardization.
• The Comparative Constitutions Project (CCP) provides a reference ontology for constitutional topics.

Purpose of the Study:

• To develop and validate a novel methodology for analyzing public consultation data on constitutional topics.
• To compare citizen concerns in public consultations with topics present in global constitutions.
• To address the practical challenges of analyzing voluminous, narrative data from constitutional design processes.

Main Methods:

• Utilized a comprehensive reference ontology from the Comparative Constitutions Project (CCP).
• Applied natural language processing (NLP), specifically sentence-level semantic similarity, to classify consultation responses.
• Analyzed a large dataset (265,000 responses) from Chile's 2016 public consultations.

Main Results:

• Successfully classified citizen responses according to constitutional topics using NLP.
• Enabled a comparison between topics prioritized by citizens and those in existing global constitutions.
• Demonstrated the feasibility of analyzing large-scale narrative data from public consultations.

Conclusions:

• The developed methodology offers a scalable and standardized approach to analyzing public consultation data.
• This technique can reveal citizen priorities in constitutional design and facilitate cross-national comparisons.
• The approach has significant implications for researchers, policymakers, and constitution drafters involved in public participation.