Methodology
The Moral Decay Index organizes evidence about American civic, institutional, economic, family, cultural, public-health, and social wellbeing into a framework for public understanding and moral reflection.
From evidence to an Index score
A composite score is a summary, not a substitute for the underlying evidence. Readers should examine the measures, sources, time periods, and limitations behind every conclusion.
Define the indicator
State precisely what social condition is being examined and why it matters.
Select evidence
Prefer original, public, methodologically transparent, and repeatable datasets.
Compare trends
Evaluate direction, magnitude, historical baseline, geography, and uncertainty.
Score and explain
Publish the result with its inputs, weights, update date, sources, and limitations.
Eight indicator areas
These areas are related but not interchangeable. Evidence may affect more than one indicator, and no single indicator defines the condition of the country.
What a score means
A higher score represents greater observed deterioration relative to the selected baseline and measures for that indicator. It does not represent the percentage of Americans who are immoral.
Until a complete calculation table is published, treat the displayed number as a provisional editorial estimate—not a validated scientific measurement.
A documented score should identify
- The measures included and why they were chosen
- Baseline and comparison years
- Geography and population covered
- Definitions and units
- Normalization formula
- Weight assigned to each measure
- Source for every input value
- Last update date and revision history
Evidence hierarchy
Preference is given to original, publicly accessible evidence from government agencies, universities, peer-reviewed journals, established nonpartisan research organizations, and reputable longitudinal surveys.
News reports and opinion pieces may provide context, but they should not be the sole support for an Index score. AI output is never treated as a source.
Review Data SourcesLimitations matter
Composite indexes simplify complicated realities. They can conceal regional variation, unequal effects among groups, measurement changes, missing data, uncertainty, and legitimate disagreement over what should be counted.
- Correlation does not prove causation.
- Simultaneous trends do not show that one caused another.
- Survey wording and definitions can change results.
- A national average may conceal local conditions.
Scope
Two levels of measurement
The Index operates at two different scales, and they do not use an identical indicator set. This is deliberate, and it is the first thing a reader should understand before comparing one number to another.
The national framework described above uses eight indicator areas. It is designed to describe the condition of the country as a whole, where measures such as political polarization and income inequality are well documented at the national level.
The State and Country Index — the interactive tracker at themoraldecayindex.com/checker — uses six indicators. Not every national measure has a comparable, consistently collected equivalent for all fifty states and for other countries. Where an indicator cannot be measured the same way in every jurisdiction being compared, including it would produce a ranking driven by data availability rather than by conditions on the ground. The six retained indicators are those with comparable sub-national and cross-national coverage.
| National framework (8 areas) | State & Country Index (6 indicators) |
|---|---|
| Mental Health Crisis | Mental Health Crisis |
| Family Stability | Family Breakdown |
| Crime & Violence | Crime & Violence |
| Substance Abuse | Substance Abuse |
| Trust in Institutions, Cultural Cohesion | Civic Disengagement — measured through voting and volunteering participation |
| Cultural Cohesion | Religious Decline — measured as attendance and affiliation rates, not belief |
| Political Polarization, Income Inequality | Not carried into the geographic tracker |
Because the two levels use different inputs, a national figure and a state figure are not directly comparable. Compare states to other states, and countries to other countries.
Reading a score
Every indicator and every composite runs 0 to 100, where a higher number means greater observed strain on that measure. The tracker groups those values into five bands so a reader is not left to guess whether 58 is high.
| Range | Band | Reading |
|---|---|---|
| 0–24 | A | Low strain |
| 25–39 | B | Below average strain |
| 40–54 | C | Moderate strain |
| 55–69 | D | Elevated strain |
| 70–100 | F | Severe strain |
Across the fifty states the composite currently ranges from 28 to 65, with a mean of 53.1. Most states fall in the C and D bands. A letter grade describes where a state sits against the other forty-nine, not a verdict on the people who live there.
A worked example
West Virginia is the clearest illustration of how the composite behaves, because its indicator scores are unusually spread out.
West Virginia — composite 65, ranked 50 of 50
| Indicator | Score | 50-state mean | Difference |
|---|---|---|---|
| Religious Decline | 38 | 50.5 | −12.5 |
| Family Breakdown | 65 | 53.0 | +12.0 |
| Crime & Violence | 48 | 52.4 | −4.4 |
| Substance Abuse | 82 | 54.9 | +27.1 |
| Mental Health Crisis | 72 | 56.0 | +16.0 |
| Civic Disengagement | 58 | 47.9 | +10.1 |
The six scores average 60.5, but the published composite is 65. The composite is not a simple arithmetic mean: an indicator in the severe band pulls the composite upward, because a single extreme condition is not adequately described by averaging it against milder ones. West Virginia’s substance abuse score of 82 is the highest single indicator value anywhere in the U.S. dataset. Its crime score, at 48, is below the national mean — which is exactly the kind of detail a composite alone would hide, and the reason every state report publishes all six components rather than only the headline number.
The same page reports year-over-year movement. West Virginia moved +5.5 points, the largest single-year change among the states. Direction is frequently more informative than position: a state in the C band moving sharply upward may warrant more attention than a state sitting stably in the D band.
Vintage and update cadence
Indicator values are drawn from the most recent full release of each underlying dataset at the time of the annual review. Those releases are not synchronised with each other — federal crime reporting, household survey data, and public-health surveillance all publish on different schedules, and some lag by more than a year. A composite therefore reflects a period rather than a single date.
- Review cycle — scores are reviewed annually, and out of cycle when a source publishes a substantial revision.
- Revisions — when a source restates prior-year figures, the affected scores are restated rather than left stale.
- Change notes — a material change to a score, a weight, or an indicator definition is published with the date and an explanation of what changed.
What is still outstanding
Stating limitations plainly is part of the standard this project asks of everyone else, so it applies here too. The complete calculation table — the specific measures feeding each indicator, their normalization formula, and the exact weight assigned to each — has not yet been published. Until it is, every score on this site is labelled a provisional editorial estimate, and should be read as an organizing device for public discussion rather than a validated scientific measurement.
That is a real limitation, not a formality. A reader who wants to check the arithmetic cannot fully do so yet. Publishing that table is the next substantive piece of work on this framework.
Scope
Two levels of measurement
The Index operates at two different scales, and they do not use an identical indicator set. This is deliberate, and it is the first thing a reader should understand before comparing one number to another.
The national framework described above uses eight indicator areas. It is designed to describe the condition of the country as a whole, where measures such as political polarization and income inequality are well documented at the national level.
The State and Country Index — the interactive tracker at themoraldecayindex.com/checker — uses six indicators. Not every national measure has a comparable, consistently collected equivalent for all fifty states and for other countries. Where an indicator cannot be measured the same way in every jurisdiction being compared, including it would produce a ranking driven by data availability rather than by conditions on the ground. The six retained indicators are those with comparable sub-national and cross-national coverage.
| National framework (8 areas) | State and Country Index (6 indicators) |
|---|---|
| Mental Health Crisis | Mental Health Crisis |
| Family Stability | Family Breakdown |
| Crime and Violence | Crime and Violence |
| Substance Abuse | Substance Abuse |
| Trust in Institutions | Civic Disengagement — measured through voting and volunteering participation |
| Cultural Cohesion | Religious Decline — measured as attendance and affiliation rates, not belief |
| Political Polarization, Income Inequality | Not carried into the geographic tracker |
Because the two levels use different inputs, a national figure and a state figure are not directly comparable. Compare states to other states, and countries to other countries.
Reading a score
Every indicator and every composite runs 0 to 100, where a higher number means greater observed strain on that measure. The tracker groups those values into five bands so a reader is not left to guess whether 58 is high.
| Range | Band | Reading |
|---|---|---|
| 0–24 | A | Low strain |
| 25–39 | B | Below average strain |
| 40–54 | C | Moderate strain |
| 55–69 | D | Elevated strain |
| 70–100 | F | Severe strain |
Across the fifty states the composite currently ranges from 28 to 65, with a mean of 53.1. Most states fall in the C and D bands. A letter grade describes where a state sits against the other forty-nine, not a verdict on the people who live there.
A worked example
West Virginia is the clearest illustration of how the composite behaves, because its indicator scores are unusually spread out.
West Virginia — composite 65, ranked 50 of 50
| Indicator | Score | 50-state mean | Difference |
|---|---|---|---|
| Religious Decline | 38 | 50.5 | −12.5 |
| Family Breakdown | 65 | 53.0 | +12.0 |
| Crime and Violence | 48 | 52.4 | −4.4 |
| Substance Abuse | 82 | 54.9 | +27.1 |
| Mental Health Crisis | 72 | 56.0 | +16.0 |
| Civic Disengagement | 58 | 47.9 | +10.1 |
The six scores average 60.5, but the published composite is 65. The composite is not a simple arithmetic mean: an indicator in the severe band pulls the composite upward, because a single extreme condition is not adequately described by averaging it against milder ones. West Virginia’s substance abuse score of 82 is the highest single indicator value anywhere in the U.S. dataset. Its crime score, at 48, is below the national mean — exactly the kind of detail a composite alone would hide, and the reason every state report publishes all six components rather than only the headline number.
The same page reports year-over-year movement. West Virginia moved +5.5 points, the largest single-year change among the states. Direction is frequently more informative than position: a state in the C band moving sharply upward may warrant more attention than a state sitting stably in the D band.
Vintage and update cadence
Indicator values are drawn from the most recent full release of each underlying dataset at the time of the annual review. Those releases are not synchronised with one another — federal crime reporting, household survey data, and public-health surveillance all publish on different schedules, and some lag by more than a year. A composite therefore reflects a period rather than a single date.
- Review cycle — scores are reviewed annually, and out of cycle when a source publishes a substantial revision.
- Revisions — when a source restates prior-year figures, the affected scores are restated rather than left stale.
- Change notes — a material change to a score, a weight, or an indicator definition is published with the date and an explanation of what changed.
What is still outstanding
Stating limitations plainly is part of the standard this project asks of everyone else, so it applies here too. The complete calculation table — the specific measures feeding each indicator, their normalization formula, and the exact weight assigned to each — has not yet been published. Until it is, every score on this site is labelled a provisional editorial estimate, and should be read as an organizing device for public discussion rather than a validated scientific measurement.
That is a real limitation, not a formality. A reader who wants to check the arithmetic cannot yet fully do so. Publishing that table is the next substantive piece of work on this framework.
The framework will evolve
Major score revisions will include an update date and an explanation of what changed. New evidence may change a score, a weight, an indicator definition, or the interpretation of a trend.
Evidence-based corrections are welcome. Send the page URL, disputed claim, and supporting source so the concern can be reviewed.
Review or challenge the work
Transparency is meaningful only when readers can inspect the evidence and question the judgment behind it.