Nigeria’s Population: Sources, Estimates, and Uncertainty

Short Answer

Nigeria’s population is not one exact, continuously updated count. Census results, incomplete vital registration, surveys, and model-based gridded estimates answer different questions at different dates. Read any figure with its year, geography, definition, and method—and treat gridded estimates as models with uncertainty, not as a census.

Nigeria’s population is often quoted as a single total, but the available evidence points to a more careful answer: different Nigerian population figures come from different source systems, refer to different dates, and carry different kinds of uncertainty. A census is a direct enumeration of people in a territory at approximately the same time [2]. A projection carries a census-based population forward. Civil registration records births and deaths as they occur, but Nigeria’s National Bureau of Statistics states that compulsory registration of births and deaths has not been implemented as legislated since 1979 [2]. Gridded products such as WorldPop’s Nigeria releases are model-based estimates at about 100 m resolution, with a reference period around 2019 in version 2.0, and they map people only where residential settlements are predicted [3]. These numbers are not interchangeable. The useful question is not simply “What is Nigeria’s exact population?” but “Which Nigerian population, for what date, at what geography, and produced by which method?”

What the main source types measure

A census tries to count every person in the territory at roughly the same time; the National Bureau of Statistics describes this as direct enumeration [2]. It provides a stock of population and can support detailed structural information, but it is periodic and becomes less current at local scales as people are born, die, and move. The WorldPop methods chapter notes that polio eradication and yellow fever vaccination campaigns in Nigeria have been based on census results from 2006, and that even where censuses are conducted at standard intervals, population estimates can quickly become inaccurate at local scales [1].

Vital statistics come from registering important events, especially births and deaths [2]. When registration is complete, it can update a census between censuses. The Nigerian statistical source notes that compulsory registration has not been implemented as legislated since 1979, so vital statistics alone cannot be treated as a complete continuous count [2].

Surveys and modelled estimates fill gaps. The modelling work presented for WorldPop’s Nigeria version 1 is described as a hierarchical Bayesian framework designed to account for uncertainty using sparse survey data [1]. That is a different kind of evidence from a census: it combines data and assumptions to produce an estimate, and it should be read as such.

Gridded population estimates allocate population to small cells. WorldPop’s Nigeria version 2.0 provides 3 arc-second, approximately 100 m grid cells with national coverage and estimates for age-sex groups; it is an update of version 1.2 and is based on more recent settlement information and a different regional boundary definition [3]. Its model-based estimates most likely represent the period around 2019, matching when satellite imagery was processed to generate building footprints [3].

How a gridded estimate is built—and why that matters

WorldPop’s Nigerian products are not census tables printed onto a map. Version 2.0 uses settlement information and building footprints derived from satellite imagery, and maps populations only where residential settlements are predicted [3]. That design is useful for local planning: it shows where people are likely to live within a national territory, not merely how many people a national total contains.

The method also introduces specific limits. The reference period is a modelling choice tied to imagery and inputs, not a real-time clock [3]. The version 2.0 release explicitly changed the regional boundary definition relative to version 1.2 [3]. The earlier version 1.2 release statement gives an original release on 10 July 2019 and a release date of 29 March 2020, showing that dataset versions are separate products with their own dates [4]. The Nigeria version 1 modelling work used a Bayesian framework to account for uncertainty [1]. In practice, a gridded value is best interpreted as a modelled estimate with an associated method, inputs, and version—not as a headcount taken on the ground.

Where uncertainty enters Nigerian population figures

  • Temporal uncertainty. A figure tied to the 2006 census and a gridded estimate for around 2019 are not measuring the same population at the same time. The WorldPop methods chapter notes both the use of 2006 census results in Nigerian vaccination planning and the loss of local accuracy over time [1].
  • Geographic and boundary uncertainty. Version 2.0 used a different regional boundary definition from version 1.2 [3]. Subnational totals should not be compared across versions without checking boundary and date.
  • Coverage and settlement-mask uncertainty. Populations are mapped only in areas where residential settlements are predicted [3]. A cell outside that mask may reflect the product’s coverage rule rather than a verified absence of people.
  • Input and model uncertainty. Settlement maps, building footprints, survey data, and model assumptions all affect estimates. The Nigeria version 1 modelling framework explicitly accounts for uncertainty [1].
  • Definitional uncertainty. A national total, an adult population, an age-sex group, and a grid-cell population are different measures. WorldPop version 2.0 includes age-sex group estimates [3], so a total and an age-specific estimate may have different uncertainty.

Comparing the main source families

The table is a reading aid, not a ranking of sources. Its purpose is to show why a “population of Nigeria” figure needs a source label.

Common source families for Nigerian population figures and their main cautions
Source familyWhat it isUseful forMain caution in Nigeria
CensusDirect enumeration of people at approximately one time [2]Historical counts and benchmarksPeriodic; local relevance declines between censuses [1]
Civil registration and vital statisticsRecords of births, deaths, and other vital events [2]Continuous updating when completeCompulsory registration had not been implemented as legislated since 1979 [2]
Surveys and modelled estimatesEstimates built from sample data and models [1]Filling data gaps and expressing uncertaintyDepends on survey inputs and modelling assumptions [1]
Gridded population modelPopulation allocated to small grid cells, often about 100 m [3]Local distribution and age-sex estimates [3]Model-based reference period, settlement mask, and version/boundary differences [3]

Common mistakes when reading a Nigeria population figure

  • Treating a census as current. A 2006-based figure cannot be assumed to describe 2019 or later; the source explicitly notes local inaccuracy over time [1].
  • Treating a gridded estimate as a census. WorldPop version 2.0 is model-based, with a reference period around 2019 and a settlement mask [3]; it is not a direct enumeration.
  • Comparing versions without checking metadata. Version 2.0 used more recent settlement information and a different regional boundary definition from version 1.2 [3].
  • Reading blank space as zero population. The product maps people only where residential settlements are predicted [3]; a blank area may reflect the model’s coverage rule, not a verified absence of people.
  • Quoting more precision than the source supports. An estimate of a grid cell or national total should carry its year, geography, and method; extra decimal places do not remove model uncertainty [1][3].

A practical checklist for using Nigeria population data

  1. Name the question. Is it a historical count, a current planning estimate, an age-sex breakdown, or a local distribution? The source type should match.
  2. Record the reference date. State whether the figure is a census count, a projection, or a model estimate, and give its reference period. For WorldPop version 2.0, that period is around 2019 [3].
  3. Check the geography. Confirm national versus state or local coverage and the boundary version; version 2.0 used a different regional boundary definition from version 1.2 [3].
  4. Inspect the coverage rule. For gridded products, ask whether non-residential or unmapped areas are excluded [3].
  5. Identify the method and inputs. Was the figure directly enumerated, derived from registration, or modelled from survey and settlement data [1][2][3]?
  6. State uncertainty. Use the dataset’s own uncertainty information where available; the Nigeria version 1 modelling framework was designed to account for uncertainty [1]. If no range is available, avoid implying exactness.
  7. Compare only compatible figures. Same year, same boundary, same definition, same source family—or explain the difference in the text.

What can and cannot be concluded

The evidence supports a clear conclusion: Nigeria’s population data are plural, not singular. A census count, a civil-registration series, a survey-based model, and a gridded population surface each answer a different question. The evidence also supports caution: census results age, registration is incomplete, and gridded estimates depend on settlement mapping and model assumptions [1][2][3]. What the supplied sources do not provide is a single current headcount that can be treated as exact for every local area. Students and teachers should therefore avoid presenting any Nigeria population figure without its date, source, geographic scope, and method. The best short sentence is usually: “This is a census count, projection, or model-based gridded estimate for this area and period, produced by this source.”

FAQ

Does Nigeria have an exact current population count?

The supplied sources do not provide a single continuously updated exact count. Census counts are periodic, registration is incomplete, and gridded products are model-based estimates for a reference period, such as around 2019 for WorldPop version 2.0.

What is the difference between a census and a gridded population estimate?

A census directly enumerates people at a date. A gridded estimate models how population is distributed across small cells, often using settlement and satellite inputs, and carries uncertainty from its method and reference period.

Why do WorldPop Nigeria maps show population only in some areas?

WorldPop Nigeria version 2.0 maps populations only where residential settlements are predicted. A blank area may reflect that coverage rule rather than a verified zero population.

Can I use the 2006 census as a current figure?

Not without adjustment. The WorldPop methods chapter notes that Nigerian vaccination planning has relied on 2006 census results and that population estimates can become inaccurate at local scales over time.

Why might two Nigeria population estimates differ?

They may use different reference dates, boundary definitions, settlement inputs, survey data, or modelling methods. WorldPop version 2.0 used more recent settlement information and a different regional boundary definition than version 1.2.

References

  1. [1] WorldPop. 8 Nigeria (v1) | WorldPop Book of Methods. Accessed November 16, 2026. https://wpgp.github.io/bookworm/ch9.html
  2. [2] National Bureau of Statistics, Nigeria. Population and Vital Statistics. Accessed November 16, 2026. https://www.nigerianstat.gov.ng/download/21
  3. [3] WorldPop and National Population Commission of Nigeria. Bottom-up gridded population estimates for Nigeria, version 2.0. 2021. https://doi.org/10.5258/soton/wp00729
  4. [4] WorldPop. NGA_population_v1_2_README.pdf. Release statement, 29 March 2020. https://data.worldpop.org/repo/wopr/NGA/population/v1.2.1/NGA_population_v1_2_README.pdf

Related Terms

Leave a Reply

Your email address will not be published. Required fields are marked *

Browse more of Country Facts