About the CLCI

The Cult-Like Control Index (CLCI) is a transparent 0–40 scoring system designed to evaluate any religious, spiritual, wellness, or ideological group on the spectrum from low-control / mainstream to destructive / extreme.

The framework

The CLCI is grounded in Steven Hassan's BITE model — Behavior, Information, Thought, and Emotional control — first published in his 1988 book Combatting Cult Mind Control and refined since. We score each of the four BITE categories from 0 to 10, then add signed modifiers (-5 to +5) for factors not captured cleanly inside BITE.

Total CLCI = Behavior + Information + Thought + Emotional + Modifier (clamped to 0–40).

The four BITE categories (0–10 each)

Behavior Control — daily life regulation (dress, diet, time, relationships, sex, marriage, finances), conformity demands, and the cost of deviation.
Information Control — censorship of outside sources, deception, insider/outsider information asymmetry, and spying.
Thought Control — loaded language, black-and-white thinking, punishment of doubt, doctrine over personal experience.
Emotional Control — fear, guilt, love-bombing, phobias about leaving, shunning.

Modifiers (±5 maximum total)

Financial demands and exploitation
Leadership accountability (or lack of it)
Shunning / disconnection policies
Documented harm (legal, physical, mental)
Exit costs — practical, social, financial barriers to leaving

Modifiers can be negative too — a group with strong leadership accountability, low exit costs, and informed-consent practices earns a credit.

Grading bands

0–5 — Minimal documented control. No public-source pattern of systematic control. Voluntary participation, transparent governance, no shunning, low exit costs.
6–12 — Low Control. Some structure or expectation but no dominant pattern of behaviour, information, thought, or emotional control.
13–20 — Moderate Control. Some controlling patterns are documented; experience varies widely by branch, congregation, or era.
21–30 — High Control. Documented patterns of significant control across multiple BITE categories.
31–40 — Destructive / Extreme. Severe control with documented harm and substantial exit costs.

Confidence levels

Every group is rated with a Confidence: High (court records, peer-reviewed academic work, multiple corroborating BITE assessments), Medium (reputable journalism + ex-member testimony but limited academic study), or Low (mostly anecdotal, fragmented documentation).

Editorial principles

Spectrum, not labels. No group gets called "a cult." Every group sits on a continuum.
Sub-branches, not whole traditions. We rate "Salafist Islam (high-control sub-branches)", not Islam. We rate "Evangelical Megachurches (high-control variants)", not all Evangelicalism.
Public sources only. No private gossip, no anonymous accusations without corroboration.
Members are not the problem. Members of high-control groups are often the people most harmed. Compassion over judgment.
Updateable. Groups change. Scores change. We document the date and source for every revision.

Known limits of the framework

The CLCI is calibrated to be useful, not to be perfect. Two limits are worth naming explicitly:

The 31–40 band compresses qualitatively different harms. Because each BITE axis maxes at 10 and the modifier range is small (±5), groups with very different real-world harm profiles can score similarly at the top of the scale. The Church of Scientology (37) and the People's Temple (40, 918 deaths at Jonestown) are not 8% apart in lived consequence, but the formula has nowhere left to put the distance. Treat the 31–40 band as "destructive across multiple BITE axes", not as a granular ranking of severity. The body text and timeline of each entry carry the qualitative differentiation that the number cannot.
Modifiers are coarse. A +1 for "documented founder sexual misconduct" applies whether the documented count is one credible victim or hundreds. The modifier reasoning string on each entry tries to convey the scale qualitatively, but the numeric +1/+2/+3 is necessarily blunt.
Confidence ratings reflect documentation density, not truth. A "Low" confidence rating means "the academic and journalistic record on this group is thin", not "the patterns described are dubious". We try to add hedging language to Low-confidence entries so readers don't mistake editorial caution for editorial certainty.

What the CLCI is NOT

It is not a clinical diagnostic tool.
It is not legal advice.
It is not a court verdict.
It is not an attack on faith — many low-CLCI groups are faith communities the world depends on.
It is not a substitute for talking to a licensed therapist or contacting ICSA if you need real support.

Sources we draw on

Steven Hassan's BITE assessments at freedomofmind.com
Robert Jay Lifton's eight criteria of thought reform
Dr. Janja Lalich's "bounded choice" framework
Court records and government investigations
Peer-reviewed sociology of new religious movements
Long-form journalism (NYT, The Guardian, etc.)
Documented ex-member testimony

Want to dig deeper? Start with our explainer What Is the BITE Model? or take the 30-question self-assessment.

The framework

Total CLCI = Behavior + Information + Thought + Emotional + Modifier (clamped to 0–40).

The four BITE categories (0–10 each)

Behavior Control — daily life regulation (dress, diet, time, relationships, sex, marriage, finances), conformity demands, and the cost of deviation.

Information Control — censorship of outside sources, deception, insider/outsider information asymmetry, and spying.

Thought Control — loaded language, black-and-white thinking, punishment of doubt, doctrine over personal experience.

Emotional Control — fear, guilt, love-bombing, phobias about leaving, shunning.

Modifiers (±5 maximum total)

Financial demands and exploitation

Leadership accountability (or lack of it)

Shunning / disconnection policies

Documented harm (legal, physical, mental)

Exit costs — practical, social, financial barriers to leaving

Modifiers can be negative too — a group with strong leadership accountability, low exit costs, and informed-consent practices earns a credit.

Grading bands

0–5 — Minimal documented control. No public-source pattern of systematic control. Voluntary participation, transparent governance, no shunning, low exit costs.

6–12 — Low Control. Some structure or expectation but no dominant pattern of behaviour, information, thought, or emotional control.

13–20 — Moderate Control. Some controlling patterns are documented; experience varies widely by branch, congregation, or era.

21–30 — High Control. Documented patterns of significant control across multiple BITE categories.

31–40 — Destructive / Extreme. Severe control with documented harm and substantial exit costs.

Editorial principles

Spectrum, not labels. No group gets called "a cult." Every group sits on a continuum.

Sub-branches, not whole traditions. We rate "Salafist Islam (high-control sub-branches)", not Islam. We rate "Evangelical Megachurches (high-control variants)", not all Evangelicalism.

Public sources only. No private gossip, no anonymous accusations without corroboration.

Members are not the problem. Members of high-control groups are often the people most harmed. Compassion over judgment.

Updateable. Groups change. Scores change. We document the date and source for every revision.

Known limits of the framework

The CLCI is calibrated to be useful, not to be perfect. Two limits are worth naming explicitly:

The 31–40 band compresses qualitatively different harms. Because each BITE axis maxes at 10 and the modifier range is small (±5), groups with very different real-world harm profiles can score similarly at the top of the scale. The Church of Scientology (37) and the People's Temple (40, 918 deaths at Jonestown) are not 8% apart in lived consequence, but the formula has nowhere left to put the distance. Treat the 31–40 band as "destructive across multiple BITE axes", not as a granular ranking of severity. The body text and timeline of each entry carry the qualitative differentiation that the number cannot.

Modifiers are coarse. A +1 for "documented founder sexual misconduct" applies whether the documented count is one credible victim or hundreds. The modifier reasoning string on each entry tries to convey the scale qualitatively, but the numeric +1/+2/+3 is necessarily blunt.

Confidence ratings reflect documentation density, not truth. A "Low" confidence rating means "the academic and journalistic record on this group is thin", not "the patterns described are dubious". We try to add hedging language to Low-confidence entries so readers don't mistake editorial caution for editorial certainty.

Sources we draw on

Steven Hassan's BITE assessments at freedomofmind.com

Robert Jay Lifton's eight criteria of thought reform

Dr. Janja Lalich's "bounded choice" framework

Court records and government investigations

Peer-reviewed sociology of new religious movements

Long-form journalism (NYT, The Guardian, etc.)

Documented ex-member testimony