High Frequency Crowdsourced Prices (HFCP) of Staple Food Commodities at National Scale 2024 - 2025
High frequency crowdsourced commodity prices in Nigeria
Nigeria, 2024 - 2025
Get MicrodataIdentification
NGA_2025_HFCP_v01_M
High Frequency Crowdsourced Prices (HFCP) of Staple Food Commodities at National Scale 2024 - 2025
High frequency crowdsourced commodity prices in Nigeria
| Name | Country code |
|---|---|
| Nigeria | NGA |
Price Survey [hh/prc]
Price data were collected and submitted by volunteers and enumerators from hundreds of markets across Nigeria’s 36 states and the Federal Capital Territory. Each market entry includes location metadata, allowing for spatial analysis of price variation and regional trends. The dataset was updated daily, offering a high-resolution view of price fluctuations over time. This frequency supports the detection of short-term shocks, seasonal patterns, and long-term trends. Prices are submitted by volunteers and contributors, mainly through mobile data submission platforms. Submissions undergo automated validation and technical review to ensure data quality and integrity.
The high frequency commodity prices are valuable resource for policy and planning. For instance, government agencies can use the data to inform food subsidy programs, inflation targeting, and emergency response planning. Also, the dataset can help advance relevant research and analyses on various dimensions of agrifood systems. Economists, data scientists, and academics can explore price elasticity, market integration, and food system resilience, while NGOs and international organizations can monitor food affordability and target interventions in vulnerable regions. Finally, citizens can access localized price information to make informed purchasing decisions. By democratizing access to food price data, the Nigeria Food Price Tracking initiative fosters transparency, accountability, and evidence-based decision-making in Nigeria’s food system and this dataset demonstrates the richness and the opportunity.
Sample survey data [ssd]
Individuals
Version
Version 2: Edited data, anonymized and packaged for public distribution
2025-12-10T05:00:00.000Z
This version compiles real-time crowdsourced price submissions from volunteers and trained enumerators. Outlier detection and removal are applied in order of submission receival, ensuring that for any given date, the dataset represents the exact state of data that is most reliable for that day.
Scope
The survey includes the following food products: yellow maize, white maize, garri, yam, imported rice, sorghum, white beans, brown beans, soybeans, and local rice. These items represent a range of staple crops and legumes commonly consumed in the region.
| Topic |
|---|
| Food Prices |
Coverage
National coverage
The data encompass all 36 States and the Federal Capital Territory (FCT) of Nigeria, providing comprehensive coverage at the sub-national (Admin 1) level.
The dataset includes data at multiple sub-national levels: Geopolitical Regions, States (Admin 1), and Local Government Areas (LGAs) (Admin 2).
Retail, Wholesale, and Farmgate Markets
Producers and sponsors
| Name | Affiliation |
|---|---|
| Julius Adewopo | World Bank Group |
| Samuel Lolo | National Bureau of Statistics (Government of Nigeria) |
| Godstime Osekhebhen Eigbiremolen | World Bank Group |
| Bo Pieter Johannes Andree | World Bank Group |
| Utz Pape | World Bank Group |
| Sering Touray | World Bank Group |
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| National Bureau of Statistics | NBS | National Agency | Implementation |
| Development Data Group | DECDG | World Bank Group | Funding, Implementation, Data Processing |
| Name | Abbreviation | Grant number | Role |
|---|---|---|---|
| Food Systems 2030 | FS2030 | TF0C7822 | Technical support for system set-up, data curation, and analytics in support of food and nutrition security monitoring |
| Food Systems 2030 | FS2030 | TF0C7819 | Technical support for system setup, data curation, and analytics to strengthen national food security preparedness |
Sampling
Data was collected based on citizen-science crowdsourcing approach, which allowed for self-selection by volunteers who visited nearest market to them and submitted data multiple times during the week. The cohort of crowdsourcing volunteers was formed by inviting interested citizens through social media groups, a website, and direct referrals from local stakeholders. The self-nomination of prospective volunteers involved an initial submission of basic profile information that include age, gender, educational level, and geographic information (mainly State and local government area – LGA), previous experience with data collection, and consent to participate. Although the initial expression of interest from the public represents convenience sampling (a type of non-probabilistic sampling approach), the final selection of the cohort followed a stratified random sampling that mainly focused on spatial representativeness across States. A total of 731 volunteers were enlisted to collect intraday prices at various markets across the country. An additional layer of 18 trained enumerators was added to the pool of data collectors to assess quality of the data submitted by the crowd.
Crowdsourcing modality for data collection is typically flexible, especially considering that the volunteers submitted data through mobile smartphone from their various locations in the country. It was recognized that self-selection and digital-competency bias is inherent in the modality of enlisting volunteers, as noted in previous studies. However, this is inconsequential for the purpose of price data collection, because it was imperative to focus on the data collectors’ ability to understand the questions, engage with the market vendors to obtain most accurate, and utilize mobile-based technology to successfully report the data on time.
100% of those who were enlisted to participate stayed active during the data collection period.
Survey instrument
The survey form was designed by NBS Team, based on a previous questionnaire that was co-designed with European Commission Joint Research Center (EC-JRC) for prior price data collection in 3 States in Nigeria.
The data collection questionnaire was processed through rounds of internal drafting and review by the core implementation at National Bureau of Statistics in Nigeria, sing the prior price survey template as a starting point. Final review of the form was conducted by the Data Innovation Team at World Bank for clarity and completeness.
Data collection
| Start | End | Cycle |
|---|---|---|
| 2024-12-01 | 2025-06-27 | Daily (with intraday submissions) |
Time-series [tim_ser]
Daily
- Face-to-face computer-assisted interviews [capi]
| Response Rate | Training |
|---|---|
| Enumerator Training |
The price data enumerators were trained to curate accurate and complete data from various market outlets, based on internal standard operating procedures of National Bureau of Statistics. The enumerators were trained to exercise due diligence in assessing quantities and local measure for the unit commodity prices, while offered guidance on requirements for the curated price datasets. The crowd volunteers received a standard operation protocol document to guide their price data collection as well. |
Data processing
| Type | Description |
|---|---|
| Captured electronically using smartphone and a data collection app | A mobile-based application called “Open Data Kit – ODK (www.odk.com)” was adopted and configured for the front-end data submission by volunteers. |
| Masking of spatially out-of-bounds data points | Datapoints whose coordinates (longitude and latitude) are outside of Nigerian boundary were excluded by using "select by mask" function in ArcGIS (with the Nigerian boundary shapefile) to select only datapoints within Nigeria |
| Exclusion of obviously spurious datapoints based on secondary market observations | Based on ancillary market surveys, no commodity was priced less than 100 naira per Kg or higher than 6000 naira per kg, therefore a global threshold was applied to the prices |
| Identifying of outliers | Using statistical logic, a rule (+/-2.5 *standard deviation) was applied to define the outliers in the dataset |
No data editing was conducted, so the price data are represented as submitted. Rather, we focused on applying proper statistical rule to finalize the processing of valid price data records per commodity, and confirmed through independent price observations by trained enumerators. Due to the broad geographical coverage of the crowdsourcing, price ranges varied, with periodic volatility, therefore, we applied rule-based logic that has been proposed in literatures. Commodity prices that are approximately above or below 2.5 standard deviations (i.e. ≈mean±2.5 S.D) were flagged as major outliers, on a rolling daily basis. The dataset includes a columns for the calculated standard deviation measure for each commodity on each day.
Quality standards
The data is governed by the World Bank Development Data Quality Policy
| Standard | Producer |
|---|---|
| Development Data Quality Policy | World Bank |
Data appraisal
Enumerator validation
The average daily crowdsourced prices were systematically compared with independent price observations from trained enumerators across the ten commodities. The independent enumerators were assigned by the National Bureau of Statistics (NBS) and they followed extant standard protocols for market price surveys, consistent with national statistical frameworks and international best practices for food price. Data consistency check between crowd-submitted prices and enumerator-submitted prices was performed by assessing for distribution overlap and correlation over the data collection period. The evaluation metrics included distribution overlap coefficients the - 1. Bhattacharyya coefficient to quantify the overlap between probability distributions, 2.The cosine similarity index the angular proximity between two time-series vectors, to evaluate the co-movement of temporal price trends, rather than magnitude; and 3. The overlapping quantile similarity to compare central tendencies and dispersion characteristics of compared prices, mainly assessing the similarity in the empirical quantile functions.
| Name | Affiliation | Role |
|---|---|---|
| Julius Adewopo | World Bank Group | Consultant |
2025-12-14T05:00:00.000Z
Summary of internal consistency between data sources shows that the Bhattacharyya Coefficient (BTC), Overlapping Quantile Similarity (OQS), and Cosine Similarity Index (CSI) were strongly positive for all commodities. Higher BTC and OQS values (generally > 0.80) indicate strong agreement in the shape and spread of price distributions, while CSI values reflect directional co-movement in temporal price trends. Collectively, these metrics demonstrate that the crowdsourced dataset captures the distributional and dynamic structure of market prices with high fidelity relative to the enumerator benchmark.
Access policy
Data is stored in World Bank microdata library
11
Data is stored along with this metadata in microdata library
Depositor information
| Name | Affiliation |
|---|---|
| Julius Adewopo | World Bank Group |
2025-12-04T05:00:00.000Z
Disclaimer and copyrights
Prices were collected across markets by volunteers and trained enumerators. Although the submissions have been thoroughly validated, the dataset do not replace official market prices and statistics from the focal country.
Contacts
| Name | Affiliation | |
|---|---|---|
| Julius Adewopo | World Bank Group | jadewopo@worldbank.org |
| Bo Pieter Johannes Andree | World Bank Group | bandree@worldbank.org |
Metadata production
DDI_NGA_2025_HFCP_v01_M
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| Julius Adewopo | J.A. | World Bank Group | Data Innovation Specialist |
| Samuel Lolo | S.L | National Bureau of Statistics - NBS (Government of Nigeria) | Technical Team Lead |
| Godstime Osekhebhen Eigbiremolen | G.O | World Bank Group | Economist |
| Bo Pieter Andree | B.P.J | World Bank Group | Data Scientist |
| Utz Pape | U.P | World Bank Group | Regional Economist |
| Sering Touray | S.T. | World Bank Group | Senior Economist |
2025-12-10T05:00:00.000Z
Metadata version
Version 01 (January 2026)