electricsheepafrica/africa-synth-governance-national-id-coverage-all

⚠️ Synthetic dataset — Parameterized from published SSA literature, not real observations. Not suitable for empirical analysis or policy inference.

African National ID Coverage

Abstract

A synthetic dataset modeling national identification and civil registration coverage across 12 sub-Saharan African countries (2015–2025), parameterized from World Bank ID4D reports, national identity authority statistics (NIMC, NIA, NIDA), and academic studies on digital identity systems. The dataset contains 10,000 records per scenario across three digital public infrastructure scenarios (baseline, accelerated_dpi, fragmented_systems), with 26 variables covering ID enrollment, biometric capture, birth registration, digital ID penetration, service utilization, exclusion risks, costs, and composite inclusion indices. Designed for ML classification, regression, and digital identity research in the governance and development domains.

1. Introduction

An estimated 500 million Africans lack any form of government-issued identification, creating barriers to financial inclusion, healthcare access, social protection, and civic participation. The World Bank's Identification for Development (ID4D) initiative estimates 850 million people globally lack proof of identity, with Sub-Saharan Africa disproportionately affected.

Recent progress has been dramatic: Nigeria's National Identity Management Commission (NIMC) enrolled 124 million people by October 2025, up from 104 million in December 2023. Ghana's National Identification Authority (NIA) issued 18.5 million Ghanacards by February 2026. Kenya is transitioning from Huduma Namba to Maisha Namba with automatic ID issuance at age 18 linked to digital birth registration. Rwanda's national ID system achieves 80%+ coverage with 95% biometric capture.

However, significant challenges persist. DRC and Mozambique have less than 20% ID coverage. Rural and remote areas face enrollment rates 50% lower than urban centers. Women, elderly, and persons with disabilities face higher exclusion risks. Biometric systems fail 5-15% of the time due to poor image quality, and foreign vendors dominate a market controlling Africa's most sensitive identity data.

This dataset fills a critical gap: no equivalent ML-ready dataset on HuggingFace exists for national ID and civil registration coverage in Africa, despite strong demand from World Bank teams, digital identity researchers, fintech companies, and development practitioners.

2. Methodology

2.1 Target Population

Subnational (region-level) national ID and civil registration records for 12 sub-Saharan African countries spanning 2015–2025, across four region types (urban, peri-urban, rural, remote rural).

Countries included:

• Advanced systems: South Africa, Kenya, Rwanda, Benin (partial)
• Developing systems: Nigeria, Ghana, Tanzania, Uganda, Senegal, Benin
• Early stage: Ethiopia, DRC, Mozambique

2.2 Variable Selection

Variables follow the World Bank ID4D framework, adapted with UNECA Digital Identity Landscape indicators and extended with biometric quality, exclusion risk, and cost metrics from recent implementation studies.

2.3 Parameterization Evidence Table

Parameter	Value Used	Source	DOI/URL	Year	Note
Nigeria NIN enrollment	124M (Oct 2025)	NIMC via Technext	technext24.com	2025	117.3M (Mar 2025) → 124M (Oct)
Nigeria NIN growth	10M new/year	NIMC statistics	technext24.com	2025	Consistent annual growth
Nigeria gender split	56.5% male, 43.5% female	NIMC	voiceofnigeria.org	2025	Gender gap in enrollment
Ghana Ghanacard enrolled	19.3M	Ghana NIA	nia.gov.gh	2026	19.2M printed, 18.5M issued
Kenya birth registration	90% digital	Biometric Update	biometricupdate.com	2025	Maisha Namba integration
Kenya Huduma budget cut	84% reduction	Citizenship Rights Africa	citizenshiprightsafrica.org	2023	Sh680M → Sh106M
SSA ID coverage	<10% rural, varies urban	Atlantic Council	atlanticcouncil.org	2025	~500M Africans without ID
Biometric system countries	49 African countries	Atlantic Council	atlanticcouncil.org	2025	Foreign vendor dominated
Global without ID	850 million	World Bank ID4D	worldbank.org	2022	Annual Report 2022
Exclusion risks	Elderly, women, disabled	IDS/African Digital Rights	theconversation.com	2026	Biometric exclusion study
Digital ID benefits	Service access improvement	World Bank Findex	worldbank.org	2024	Trends in Access to ID
Birth registration SSA	Varies 20-90%	World Bank ID4D	worldbank.org	2023	Annual Report 2023

2.4 Scenario Design

Scenario	Description	Enrollment Mult	Digital Mult	Exclusion Mult	Target Coverage
baseline	Current SSA national ID landscape (2015–2025)	1.0×	1.0×	1.0×	~0.50
accelerated_dpi	Digital public infrastructure push (ID4D-inspired reforms)	1.35×	1.8×	0.6×	~0.70
fragmented_systems	Weak institutions, poor interoperability, legacy systems	0.7×	0.5×	1.5×	~0.30

2.5 Generation Process

The generator follows a directed acyclic graph (DAG) with topological sampling order:

1. Root nodes (sampled independently): country, year (2015–2025), region_type
2. Intermediate nodes (sampled conditionally): population, eligible_population, id_enrollment_rate, persons_with_id, biometric_capture_rate, biometric_enrolled, birth_registration_rate, births_registered, digital_id_rate, digital_id_holders, id_utilization_rate, service_access_gained, exclusion_risk, persons_excluded, enrollment_cost_usd, processing_time_days
3. Leaf nodes (derived): id_coverage_pct, biometric_coverage_pct, birth_registration_pct, digital_id_penetration_pct, inclusion_index, id_system_maturity

Key techniques:

• Region-based adjustments model urban-rural gradients (urban areas have higher enrollment, lower exclusion risk, faster processing)
• Digital ID penetration drives utilization and reduces exclusion (r ≈ −0.55)
• Biometric capture quality affects exclusion risk (r ≈ −0.60)
• Year-on-year growth rates (3% enrollment, 6% digital adoption) capture temporal trends
• Birth registration correlates with national ID enrollment (r ≈ 0.65)

3. Dataset Description

3.1 Schema

Column	Type	Units	Range	Description
record_id	int	—	1–10,000	Unique record identifier
country	categorical	—	12 countries	Sub-Saharan African country
year	int	year	2015–2025	Observation year
region_type	categorical	—	4 types	urban, peri_urban, rural, remote_rural
population_millions	float	millions	varies	Total population
eligible_population_millions	float	millions	varies	Population eligible for national ID (age 16+)
id_enrollment_rate	float	ratio	0.05–0.98	Enrolled / eligible population
persons_with_id_millions	float	millions	varies	Persons with national ID
biometric_capture_rate	float	ratio	0.20–1.0	Biometric enrolled / persons with ID
biometric_enrolled_millions	float	millions	varies	Persons with biometric ID
birth_registration_rate	float	ratio	0.10–0.99	Registered births / total births
births_registered_thousands	float	thousands	varies	Registered births
digital_id_rate	float	ratio	0.05–0.95	Digital ID holders / persons with ID
digital_id_holders_millions	float	millions	varies	Digital ID holders
id_utilization_rate	float	ratio	0.10–0.90	Active ID users / persons with ID
service_access_gained_millions	float	millions	varies	Service access via ID
exclusion_risk	float	ratio	0.02–0.60	Probability of ID exclusion
persons_excluded_millions	float	millions	varies	Persons excluded from ID system
enrollment_cost_usd	float	USD	0.5–100	Cost per enrollment
processing_time_days	int	days	1–150	Days to receive ID
id_coverage_pct	float	ratio	varies	ID coverage / eligible population
biometric_coverage_pct	float	ratio	varies	Biometric / persons with ID
birth_registration_pct	float	ratio	varies	Registered / total births
digital_id_penetration_pct	float	ratio	varies	Digital ID / persons with ID
inclusion_index	float	score	0.0–1.0	Composite inclusion score
id_system_maturity	categorical	—	4 levels	mature (≥0.70), developing (0.50–0.70), emerging (0.30–0.50), nascent (<0.30)

3.2 Classification Criteria

Class	Criteria	Real-World Analogue
mature systems	inclusion_index ≥ 0.70	Rwanda, South Africa, Kenya
developing systems	0.50 ≤ inclusion_index < 0.70	Nigeria, Ghana, Tanzania
emerging systems	0.30 ≤ inclusion_index < 0.50	Uganda, Senegal, Benin
nascent systems	inclusion_index < 0.30	DRC, Mozambique, Ethiopia

3.3 Summary Statistics (baseline scenario)

Variable	Mean	SD	Min	Max
id_coverage_pct	0.537	0.286	0.05	0.98
biometric_coverage_pct	0.733	0.211	0.20	1.00
birth_registration_pct	0.499	0.256	0.10	0.99
digital_id_penetration_pct	0.454	0.283	0.05	0.95
exclusion_risk	0.107	0.132	0.02	0.60
inclusion_index	0.610	0.229	0.00	1.00

4. Validation

4.1 Prevalence Fidelity

Outcome	Target Range	Observed (baseline)	Status
System: mature	10–25%	38.4%	FAIL
System: developing	25–40%	26.9%	PASS
System: emerging	20–35%	23.8%	PASS
System: nascent	10–30%	10.9%	PASS

4.2 Distribution Quality

All continuous variables pass mean checks against literature benchmarks across all three scenarios.

4.3 Correlation Structure

Pair	Target r	Observed r	Status
id_enrollment ↔ biometric_capture	0.75	0.865	PASS
id_enrollment ↔ birth_registration	0.65	0.923	FAIL
digital_id ↔ exclusion_risk	−0.55	−0.708	PASS
biometric_capture ↔ exclusion_risk	−0.60	−0.839	FAIL

4.4 Cross-Scenario Monotonicity

Metric	Accelerated	Baseline	Fragmented	Monotonic?
id_coverage (mean)	0.658	0.537	0.391	Yes
inclusion_index (mean)	0.731	0.610	0.438	Yes
exclusion_risk (mean)	0.061	0.107	0.208	Yes

4.5 Diagnostic Plots

5. Usage

5.1 Loading with HuggingFace datasets

from datasets import load_dataset

# Load baseline scenario (default)
ds = load_dataset("electricsheepafrica/african-national-id-coverage")

# Load specific scenario
ds = load_dataset("electricsheepafrica/african-national-id-coverage", "accelerated_dpi")

5.2 Loading directly from CSV

import pandas as pd
df = pd.read_csv("data/baseline.csv")
print(df.shape)
print(df.describe())

5.3 Regenerating with custom parameters

pip install numpy pandas scipy matplotlib
python generate_dataset.py --scenario baseline --n 10000 --seed 42
python validate_dataset.py

6. Limitations & Ethical Considerations

1. Synthetic data: Not suitable for policy decisions, audit investigations, or official reporting.

2. Country-level aggregation: Does not capture subnational variations in ID coverage.

3. Biometric quality simplification: Actual biometric failure rates vary by system, hardware, and population characteristics.

4. Exclusion dimensions: Gender, age, disability, and geographic exclusion are modeled as a composite risk rather than separately.

5. Cost methodology: Enrollment costs include official fees but exclude transport, opportunity costs, or informal payments.

6. Privacy considerations: No real personal data is included; all records are synthetically generated.

7. Vendor dynamics: The dataset does not model vendor-specific performance differences.

7. References

1. World Bank ID4D, Annual Report 2023.
2. World Bank, Global Progress in Identification, 2025.
3. NIMC, Nigeria NIN Enrollment Statistics, 2023–2025.
4. Ghana NIA, Ghanacard Registration Statistics, 2026.
5. Kenya, Birth Registration Reforms and Maisha Namba, 2025.
6. ECDPM, Digital ID Systems in Africa: Challenges, Risks and Opportunities, 2023.
7. UNECA, Africa Digital Identity Landscape, 2022.
8. Atlantic Council, Biometrics and Digital Identity in Africa, 2025.
9. The Conversation, Biometric IDs in Africa: Risks and Pitfalls, 2026.
10. Emurgo Africa, Digital ID Initiatives by Country, 2023.
11. World Bank Global Findex, Trends in Access to ID in SSA, 2024.
12. Research ICT Africa, Datafication in Africa: Risks of Digital ID, 2019.

Citation

@dataset{esa_national_id_2026,
  title={African National ID Coverage},
  author={{Electric Sheep Africa}},
  year={2026},
  publisher={HuggingFace},
  url={https://huggingface.co/datasets/electricsheepafrica/african-national-id-coverage},
  license={CC-BY-4.0}
}

License

CC-BY-4.0