Diagnostics

The diagnostics test the model

They check for income bias, duplicate indicators, weak evidence and scores that change sharply when inputs change.

computed 2026-08-2950 countries

Four of nine dimensions track income per head

At 0.7 or more, a correlation with log GDP per capita counts as tracking income. These results use 50 countries; treat coefficients as hints.

Dimension correlation with GDP per capita

Anticipation	0.873	0.921	50
Agency	0.850	0.861	50
Adaptability	0.818	0.829	50
Learning	0.725	0.734	50
Building	0.641	0.661	50
Trust	0.627	0.711	35
Experimentation	0.622	0.754	49
Coordination	0.467	0.583	44
Shared Purpose	0.451	0.461	46

Every dimension survives without income-linked data

10 indicators meet the 0.7 correlation threshold. This table shows the shift after removing them. Every dimension keeps at least one measured indicator.

Score shift when wealth-correlated indicators are removed

Agency	8.31	45
Learning	8.09	47
Adaptability	6.43	46
Building	3.86	41
Coordination	0.00	0
Trust	0.00	0
Experimentation	0.00	0
Shared Purpose	0.00	0
Anticipation	cannot be scored	52

Some indicators track income

This compares each indicator with log GDP per capita and with the registry's prior.

Indicator correlation with GDP per capita

Secure internet servers	Anticipation	I	0.895	0.60
Output per worker	Building	O	0.893	0.90
Individuals using the internet	Agency	I	0.879	0.55
Human Capital Index	Learning	C	0.878	0.50
Fixed broadband subscriptions	Adaptability	I	0.849	0.60
Scientific articles	Anticipation	O	0.797	0.40
Financial account ownership	Agency	I	0.789	0.50
Researchers in R&D	Anticipation	I	0.788	0.50
Tertiary enrolment	Learning	I	0.785	0.50
Statistical performance	Anticipation	C	0.763	0.25
Electricity transmission losses	Adaptability	O	0.691	0.40
Generalised interpersonal trust	Trust	P	0.669	0.25
Procedures to start a business	Agency	C	0.644	0.15
R&D expenditure	Anticipation	I	0.631	0.50
Time to start a business	Agency	C	0.598	0.20
Resident trademark applications	Experimentation	O	0.574	0.30
Labour force participation	Adaptability	I	0.572	0.15
Border compliance time to export	Coordination	C	0.547	0.30
Credit to the private sector	Agency	I	0.543	0.55
Resident patent applications	Experimentation	O	0.527	0.45
Tolerance of entrepreneurial failure	Experimentation	P	0.478	0.10
Economic fitness	Building	O	0.464	0.45
High-technology exports	Building	O	0.459	0.40
New business density	Agency	C	0.405	0.35
Income inequality	Shared Purpose	O	0.379	0.20
Public education expenditure	Learning	I	0.350	0.30
Early-stage entrepreneurial activity	Experimentation	C	-0.296	0.15
Firms offering formal training	Learning	C	0.256	0.20
Budget execution fidelity	Coordination	C	0.218	0.15
Vocational share of secondary	Learning	I	0.218	0.10
Tax revenue	Shared Purpose	O	0.209	0.30
Manufacturing value added	Building	O	0.178	0.20
Time to enforce a contract	Trust	C	0.173	0.20
Unemployment rate	Adaptability	O	0.158	0.25
Electricity connection speed	Building	C	0.026	0.25

One indicator can drive the income signal

Each row removes one indicator. A positive delta means the dimension tracks income more without it.

What each indicator does to its dimension's correlation with GDP per capita

Border compliance time to export	Coordination	0.515	0.218	+0.297
Output per worker	Building	0.641	0.442	+0.199
Generalised interpersonal trust	Trust	0.367	0.173	+0.194
Income inequality	Shared Purpose	0.379	0.209	+0.171
Fixed broadband subscriptions	Adaptability	0.818	0.664	+0.154
Human Capital Index	Learning	0.725	0.617	+0.108
Tertiary enrolment	Learning	0.725	0.637	+0.088
Individuals using the internet	Agency	0.850	0.778	+0.072
Economic fitness	Building	0.641	0.578	+0.063
Electricity transmission losses	Adaptability	0.818	0.770	+0.048
Resident trademark applications	Experimentation	0.563	0.517	+0.047
Financial account ownership	Agency	0.850	0.811	+0.039
Secure internet servers	Anticipation	0.873	0.838	+0.036
Statistical performance	Anticipation	0.873	0.843	+0.030
Resident patent applications	Experimentation	0.563	0.543	+0.020
Tolerance of entrepreneurial failure	Experimentation	0.563	0.550	+0.013
Time to start a business	Agency	0.850	0.841	+0.009
Procedures to start a business	Agency	0.850	0.843	+0.007
Firms offering formal training	Learning	0.725	0.718	+0.007
Tax revenue	Shared Purpose	0.379	0.379	+0.001
Scientific articles	Anticipation	0.873	0.874	0.000
Credit to the private sector	Agency	0.850	0.862	-0.012
High-technology exports	Building	0.641	0.657	-0.016
R&D expenditure	Anticipation	0.873	0.890	-0.017
Researchers in R&D	Anticipation	0.873	0.890	-0.017
Early-stage entrepreneurial activity	Experimentation	0.563	0.583	-0.020
Labour force participation	Adaptability	0.818	0.838	-0.020
Public education expenditure	Learning	0.725	0.749	-0.024
Unemployment rate	Adaptability	0.818	0.845	-0.027
Budget execution fidelity	Coordination	0.515	0.547	-0.032
New business density	Agency	0.850	0.884	-0.034
Electricity connection speed	Building	0.641	0.692	-0.051
Vocational share of secondary	Learning	0.725	0.782	-0.057
Manufacturing value added	Building	0.641	0.705	-0.064
Time to enforce a contract	Trust	0.367	0.669	-0.302

Some indicator pairs overlap

Pairs at 0.85 correlation or above. They are candidates for review, not automatic removals.

Redundant indicator pairs

Researchers in R&D	Scientific articles	0.916
R&D expenditure	Researchers in R&D	0.890
Human Capital Index	Fixed broadband subscriptions	0.872
Researchers in R&D	Human Capital Index	0.865
Secure internet servers	Financial account ownership	0.860
Secure internet servers	Human Capital Index	0.860

The dimensions remain distinct

Dimension pairs sorted by correlation. Pairs at 0.9 or above are candidates for review.

Dimension pair correlations

Anticipation	Agency	0.834
Anticipation	Adaptability	0.785
Trust	Adaptability	0.772
Agency	Adaptability	0.766
Agency	Trust	0.733
Anticipation	Trust	0.732
Agency	Experimentation	0.699
Anticipation	Shared Purpose	0.654
Adaptability	Building	0.651
Anticipation	Experimentation	0.644
Anticipation	Learning	0.639
Experimentation	Building	0.638
Anticipation	Building	0.632
Experimentation	Adaptability	0.621
Agency	Building	0.604
Agency	Learning	0.602
Learning	Adaptability	0.599
Trust	Shared Purpose	0.595
Anticipation	Coordination	0.546
Trust	Building	0.536
Learning	Shared Purpose	0.528
Learning	Experimentation	0.524
Trust	Experimentation	0.518
Coordination	Building	0.504
Agency	Coordination	0.498
Adaptability	Shared Purpose	0.483
Coordination	Shared Purpose	0.480
Trust	Learning	0.476
Agency	Shared Purpose	0.475
Coordination	Trust	0.419
Coordination	Adaptability	0.401
Coordination	Learning	0.388
Coordination	Experimentation	0.362
Learning	Building	0.325
Building	Shared Purpose	0.288
Experimentation	Shared Purpose	0.165

Evidence is strongest for Anticipation and weakest for Trust

Subjectivity share counts perception proxies and indicators with no dataset. Trust is the weakest: two of its eight indicators are observed and its mean confidence is 0.13.

Measurement quality by dimension

Anticipation	7	5	2	0.70	0.60	0.29
Agency	8	6	2	0.73	0.58	0.25
Building	8	5	2	0.61	0.53	0.38
Learning	7	5	2	0.69	0.49	0.29
Adaptability	8	4	4	0.50	0.47	0.50
Experimentation	8	4	4	0.32	0.23	0.63
Shared Purpose	7	2	4	0.27	0.22	0.71
Coordination	8	2	3	0.23	0.15	0.75
Trust	8	2	3	0.21	0.13	0.88

Every indicator family has data behind it

A dimension can ask two questions under one name, and several readings of one question are not several independent signals. Scoring counts every observed indicator equally either way. This table says which family the evidence came from.

Indicator families inside a dimension

Trust	Institutional	3	1
Trust	Social	2	1

0% of observed values clamp at the frame edge

0 of 1684 observed cells fall outside the frame and clamp to 0 or 100. These are historical or late-arriving values.

Clamped cells by country

These indicators have no data

The model specifies them, but no adequate dataset covers them. They stay in the registry and lower confidence.

Indicators with no adequate dataset
			Why it is missing
Adaptability	Long-term unemployment share	ILOSTAT	Closer to reallocation speed than the headline rate: it asks whether people who lose work find new work. ILOSTAT publishes it; the World Bank API does not carry it.
Adaptability	Export diversification	UNCTAD	UNCTAD publishes the concentration index and it is computable. Another good candidate for the next adapter.
Adaptability	Disaster preparedness and recovery	INFORM / UNDRR	INFORM is largely a hazard-exposure index, so using it here would measure geography rather than capability.
Adaptability	Institutional responsiveness	none	No dataset exists. Measurable in principle from legislative and regulatory timestamps, which no one has assembled comparably.
Agency	Adult digital skills	ITU	ITU collects this but coverage across these ten countries is broken and the task lists differ by year.
Agency	Perceived control over life	World Values Survey	WVS wave 7 covers most of these countries but fieldwork years differ by up to six years, and Singapore is thin.
Anticipation	Government foresight capacity	none	No comparable international dataset. OECD and UN work is descriptive, not scored. Envisioning would have to build this.
Anticipation	Long-horizon research share	OECD MSTI	Published by OECD for members only. Missing for India, South Africa, Brazil in comparable form.
Building	Large project delivery	none	The single best measure of execution capability and there is no comparable international dataset. Assembling one is a defensible Envisioning project. Building mixes industrial output with state delivery capacity, and this indicator carries the delivery side: the other six indicators cannot see a national programme that was delivered. Documented deliveries are recorded in data/evidence and never scored, see D20.
Building	Firm scale-up rate	OECD DynEmp	OECD business dynamics work covers members irregularly. Nothing comparable for India, Brazil or South Africa.
Coordination	University-industry collaboration	Global Innovation Index	GII carries it, but the underlying item is a WEF executive opinion survey whose microdata is not inspectable. Excluded on principle 4.
Coordination	Civil society strength	V-Dem	V-Dem core civil society index is inspectable and would fill this. Not yet wired; it needs its own ingestion adapter.
Coordination	Public-private collaboration	none	No comparable dataset. PPP investment databases measure infrastructure finance, not collaboration capacity.
Experimentation	Venture capital investment	OECD / national VC associations	Still a gap after a direct check on 2026-08-26. The OECD SME and Entrepreneurship Financing scoreboard is the only inspectable aggregate and it carries venture capital for 6 of these 16 countries, in national currency rather than as a share of GDP, latest year 2022. Brazil, India, South Africa and Singapore are all absent, so wiring it would score the rich half of the set and lower coverage for the rest. Commercial databases cover the world and are not inspectable. Read A1 before treating this dimension as measured.
Experimentation	Regulatory sandbox activity	none	Countable from primary sources but nobody maintains a comparable register. A realistic candidate for Envisioning to build.
Experimentation	University spinouts	national TTO associations	Reported nationally with incompatible definitions of what counts as a spinout.
Experimentation	Business share of R&D	UNESCO UIS	UIS publishes this and it is a good candidate for the next ingestion adapter.
Learning	Adult learning participation	Eurostat / OECD PIAAC	The best available measure of continuous learning capacity, and it exists only for European and OECD members.
Learning	Research citation impact	OpenAlex	Computable from OpenAlex, which is open and inspectable. This is the highest-value gap to close next.
Shared Purpose	Sense of national belonging	World Values Survey	Culturally loaded. High national pride is not the same as capacity for collective action and must not be read as such.
Shared Purpose	Volunteering	Gallup World Poll / CAF	CAF publishes country figures but the underlying Gallup microdata is proprietary, so it fails the inspectability rule.
Shared Purpose	Political polarisation	V-Dem	V-Dem political polarisation is inspectable and would fill this. Pluralism is the target, so only hostile polarisation should count against a country.
Shared Purpose	Civic participation	World Values Survey	Behavioural rather than attitudinal, so it is the item worth prioritising if only one survey measure can be harmonised.
Trust	Trust in public institutions	OECD Trust Survey / WVS	The OECD survey covers members only. Mixing it with WVS items for India and South Africa would break comparability.
Trust	Cooperation beyond the in-group	World Values Survey	The distinction that makes this dimension worth measuring. Same licensing obstacle as generalised trust.
Trust	Court case clearance rate	CEPEJ / OECD / national court statistics	Whether a court finishes what it starts, counted from case records rather than asked in a survey. D23 named it as the observable replacement for the two retired WGI composites and it has stayed unfilled since. CEPEJ publishes it for Council of Europe members and OECD for its own members, so no single publisher reaches this country set, and a harmonised series needs the project second source adapter. See D57.

These datasets were rejected

Retired rows stay in the registry and lower confidence like gaps. The reason for each rejection is recorded.

Datasets examined and rejected
			Why it was rejected
Building	Logistics infrastructure quality	World Bank	Retired 2026-08-26. An LPI sub-index, so the same freight-forwarder survey as logistics performance and the same problem. Electricity connection speed stays as the observable infrastructure measure in this dimension. See D23.
Coordination	Government effectiveness	World Bank	Retired 2026-08-26. The WGI aggregate the opinions of experts and firms, and in this set the four WGI series correlate with each other between 0.93 and 0.98 while correlating with log GDP per capita at 0.91. That is one perception of national wealth counted in three dimensions. Cross-agency delivery records are the replacement and they are a declared gap. See D23 and A4.
Coordination	Regulatory quality	World Bank	Retired 2026-08-26. Same measurement as government effectiveness, filed under a different name, correlating with it above 0.93. See D23 and A4.
Coordination	Logistics performance	World Bank	Retired 2026-08-26. The LPI is a survey of international freight forwarders, so it records how a country looks to global logistics firms. A small country with a small port scores low whatever its state can organise, which is artefact A9. Container throughput and border time are observable and stay. See D23.
Shared Purpose	Voice and accountability	World Bank	Retired 2026-08-26. Artefact A5: it measures the democratic channel while Shared Purpose asks whether people can see themselves in a common project. Singapore scored 20.9 while being one of the most effective collective actors in the set. Voter turnout, volunteering and civic participation are the observable replacements and all are declared gaps. See D23 and A5.
Trust	Rule of law	World Bank	Retired 2026-08-26. A WGI perception composite correlating with control of corruption above 0.95 and with log GDP per capita at 0.83. Court throughput and case clearance rates are the observable replacement and are a declared gap. See D23 and A4.
Trust	Control of corruption	World Bank	Retired 2026-08-26. A WGI perception composite. It measures reputation for corruption, which is not the same thing as corruption, and it tracks income closely. Prosecution and audit records are the observable replacement and are a declared gap. See D23 and A4.
Trust	Intentional homicide rate	World Bank	Added 2026-08-26 by D23 as the observable replacement for two retired WGI perception composites, and retired 2026-08-27 by D44 when the diagnostic built in D42 measured what it was doing. It raised this dimension's correlation with log GDP per capita by 0.288, from 0.096 to 0.385, the largest single wealth contribution anywhere in the model. It was added to fix wealth contamination and it was the wealth contamination. The dataset is sound and the objection is to what it measures here: homicide is driven heavily by organised crime, a society can be physically safe while trusting very little, and across this country set the variation it carries is mostly income. Behavioural trust measures remain a gap.