Skip to main content

Educational tool only. All groups exist on a spectrum of control. Individual experiences vary. Based on publicly available reports, ex-member accounts, court records, and expert analyses — not medical or legal advice.

CLCI Hub
Start here

If you are researching a group or the field

Entry point for academic researchers, students, and policy researchers using the CLCI Hub dataset and methodology pages.

For: Academic and policy researchers, students, dissertation writers.

Last reviewed 21 May 2026Editorial policySource policy

Introduction

CLCI Hub is an editorial reference rather than a research instrument, and the distinction matters when citing it in academic or policy work. The pages below cover what the dataset is, what its known limits are, how to cite it, and where to find the structured exports.

Start with the methodology and limits

Read /methodology, then /methodology/source-hierarchy and /methodology/scoring-limitations. The honest framing of what CLCI scores do and do not measure is the most important context for a researcher arriving from outside the cult-recovery field.

Then the data layer

/research/dataset has the dataset overview; /research/data-dictionary is the field-by-field reference; /research/methodology-limitations documents known biases and coverage gaps; /research/downloads has the CSV and JSON exports.

Citation

/research/citation-guide has the recommended citation forms. We ask that you cite the methodology-limitations page alongside any aggregate use of CLCI scores.

Where the catalogue is thin

  • Anglophone coverage is stronger than non-Anglophone.
  • Recent online-first groups are under-represented.
  • Structured evidence is sparse for most entries (free-text sources are the majority).
  • Per-entry confidence ratings are coarse (three levels, no probability backing).

FAQ

Can I use CLCI scores as a numeric variable in regression?
We do not recommend it without acknowledging the limits. CLCI scores are editorial ordinal-ish ratings; treating them as interval-scale measurements without caveats overstates their precision.
Are the exports versioned?
The exports are regenerated at every build. /research/changelog summarises site-level dataset changes; per-group changes are aggregated on /updates.

This page is educational and not legal, medical, or clinical advice. See the Legal Disclaimer. Found something wrong? Submit a correction.