Methods note: fieldwork
What should a survey do about places field teams cannot safely reach?
Every recent national survey in Nigeria has lost areas to insecurity. How the surveys handled it, what their headline figures then describe, and the rules we follow.
Published October 2026. Sources checked 11 October 2026.
In short
- The 2024 Nigeria DHS could not complete 20 of its 1,400 clusters because of insecurity, half of them in Zamfara. Losses are no longer concentrated in the north-east.
- Surveys handle this differently: the DHS drops the areas, while the 2018/19 Living Standards Survey and the 2018/19 GHS-Panel wave replaced them, and the national poverty estimate for 2018/19 leaves out Borno. The same state can mean different things in different surveys.
- A high response rate does not show coverage, because it is calculated only where teams went. What matters is saying plainly which population an estimate describes.
What recent surveys lost
Insecurity or inaccessibility removed areas from every major Nigerian household survey we reviewed.
Areas not reached in Nigeria's main household surveys
Nigeria DHS 2024
- Selected
- 1,400 clusters
- Not reached
- 20, for insecurity
- Where, and how it was handled
- Zamfara 10, Borno 3, Benue 2, Imo 2, Katsina, Kaduna and Niger 1 each. Excluded, not replaced1
Nigeria Mini-DHS 2025–26
- Selected
- 876 clusters
- Not reached
- 5, for insecurity
- Where, and how it was handled
- Zamfara, Borno, Bauchi, Niger and Kwara, 1 each. Excluded2
Nigeria DHS 2018
- Selected
- 1,400 clusters
- Not reached
- 11, for 'law-and-order situations'
- Where, and how it was handled
- Zamfara 4, Sokoto 3, Katsina 2, Lagos 1, Borno 1. Dropped3
MICS 2021
- Selected
- 1,850 survey and 337 immunisation-survey areas
- Not reached
- 128 'inaccessible' areas
Living Standards Survey 2018/19
- Selected
- 60 areas per state
- Not reached
- 139 areas, about 6%, could not be listed
- Where, and how it was handled
- Replaced from the same state and sector. Borno sampled from the 359 of 788 areas judged accessible; 7 of its 60 then not reached and not replaced6
GHS-Panel, Wave 5
- Selected
- 518 areas
- Not reached
- 8, for conflict
- Where, and how it was handled
- Zamfara, Katsina, Kebbi and Sokoto. Not visited7
| Survey | Selected | Not reached | Where, and how it was handled |
|---|---|---|---|
| Nigeria DHS 2024 | 1,400 clusters | 20, for insecurity | Zamfara 10, Borno 3, Benue 2, Imo 2, Katsina, Kaduna and Niger 1 each. Excluded, not replaced1 |
| Nigeria Mini-DHS 2025–26 | 876 clusters | 5, for insecurity | Zamfara, Borno, Bauchi, Niger and Kwara, 1 each. Excluded2 |
| Nigeria DHS 2018 | 1,400 clusters | 11, for 'law-and-order situations' | Zamfara 4, Sokoto 3, Katsina 2, Lagos 1, Borno 1. Dropped3 |
| MICS 2021 | 1,850 survey and 337 immunisation-survey areas | 128 'inaccessible' areas | Borno sampled only from 7 of its 27 local government areas; Borno results labelled '7 LGAs'4,5 |
| Living Standards Survey 2018/19 | 60 areas per state | 139 areas, about 6%, could not be listed | Replaced from the same state and sector. Borno sampled from the 359 of 788 areas judged accessible; 7 of its 60 then not reached and not replaced6 |
| GHS-Panel, Wave 5 | 518 areas | 8, for conflict | Zamfara, Katsina, Kebbi and Sokoto. Not visited7 |
- In the 2024 DHS and the 2023 GHS-Panel visit, the north-west accounted for most losses: 10 of the 20 clusters lost in the 2024 DHS were in Zamfara1, and all 8 areas the GHS-Panel could not visit were in four north-western states7. The 2025–26 Mini-DHS lost one cluster in each of five states across three zones2.
- Borno is handled differently in each survey. The 2024 DHS lost three Borno clusters1. The 2021 MICS drew Borno's sample from seven local government areas holding an estimated 29% of the state's population, an estimate that the report says does not account for people displaced into them4. The national poverty figure for 2018/19 leaves Borno out altogether8.
Why a 99% response rate does not mean full coverage
Drop or replace?
The DHS sampling manual says that where areas may have to be excluded for 'extreme inaccessibility, violence or instability', the issue needs to be considered 'at the very beginning of the survey, before the sample is drawn', and that excluded areas 'should constitute a coherent domain', because a survey with scattered exclusions 'is difficult to interpret and to use'. For households, it says 'no changes or replacements are allowed in the field'10.
The Living Standards Survey and the GHS-Panel in 2018/19 replaced inaccessible areas with others from the same state, sector or zone, and then said plainly what the result means: the sample is 'representative of areas of Nigeria that were accessible' and 'will not reflect conditions in areas that were undergoing conflict'6,11. Replacement keeps the sample size. It does not bring back the people who live where teams could not go.
Elsewhere in the region the losses have been larger. Burkina Faso's 2021 DHS completed 514 of 600 clusters, and its report says results for the Sahel and Est regions must be interpreted with caution and not compared directly with other regions12. Mali's 2018 DHS completed 345 of 379 clusters and surveyed only the urban part of Kidal13. Cameroon's 2018 DHS completed 432 of 470 and attached the same caution to its South-West region14.
What remote methods can and cannot add
Where teams cannot go, humanitarian monitoring relies on key informants. REACH's hard-to-reach assessments in north-east Nigeria interview people who have recently left, or been in contact with, an inaccessible settlement, and combine their answers at settlement level; REACH states that the findings 'are not statistically generalisable'15. IOM's Displacement Tracking Matrix counts displaced people through key informants at local government, ward and site level16. Both are valuable for knowing who is where. Neither is a sample survey, and neither yields a margin of error.
Phone interviews do not automatically reach these areas either. REACH's 2020 needs assessment in the north-east could not cover five Borno local government areas, 'due to insecurity and a lack of phone signal'17.
Third-party monitors can go where an agency's own staff cannot. A study of their use in Afghanistan, Somalia and Syria found frequent irregularities in their data and a transfer of risk to the monitors, and concluded that third-party monitoring is 'most useful as a last resort measure or as a complement' to an agency's own checks18.
The field team's safety
The people at risk include the field team. The Aid Worker Security Database records 31 major security incidents affecting aid workers in Nigeria in 2024 and 10 in 202519. It does not measure risk to survey teams specifically, and we found no Nigerian data that does. Guidance from fragile settings includes keeping interviews short where travel is dangerous (a 60-minute limit was advised in Somalia) and relying on locally resident interviewers whom the community trusts20.
Our practice
These are the rules every Probatio Africa study follows. Read them as commitments you can hold us to.
- We assess the security of every selected area before fieldwork, and decide any exclusion before the sample is drawn, not in the field.
- We do not replace a sampled area in the field. If an area becomes unreachable, we record it, report it by state, and state the population the estimates then describe.
- We report coverage and response separately: how many selected areas were reached, and the response rate within them.
- Where key-informant or remote data fill a gap, we label them as indicative and keep them out of any estimate that carries a margin of error.
- A team can stop work in its area without penalty if it judges the situation unsafe, and resumes only when the field lead signs off.
What the evidence cannot settle
- How much bias excluded or replaced areas introduce into national figures. None of the Nigerian survey reports we read estimates it.
- How closely key-informant estimates match household-survey estimates for the same settlements. We found no validation study.
- How great the risk to survey field staff is. There are data on aid workers, but we found none on survey teams.
Sources
- 1.Federal Ministry of Health and Social Welfare, National Population Commission and ICF (September 2024). Nigeria Demographic and Health Survey 2023–24: Key Indicators Report. Section 2 describes the sample, fieldwork and the clusters not reached.https://dhsprogram.com/pubs/pdf/PR157/PR157.pdf
- 2.Federal Ministry of Health and Social Welfare, National Population Commission and ICF (June 2026). Nigeria Mini-Demographic and Health Survey 2025–26: Key Indicators Report.https://dhsprogram.com/pubs/pdf/PR166/PR166.pdf
- 3.National Population Commission and ICF (May 2019). Nigeria Demographic and Health Survey 2018: Key Indicators Report. Copy held in the Nigeria Governors' Forum repository.https://ngfrepository.org.ng:8443/bitstream/123456789/3145/1/NDHS%202018.pdf
- 4.National Bureau of Statistics and UNICEF (August 2022). 2021 Multiple Indicator Cluster Survey (MICS) & National Immunization Coverage Survey (NICS): Survey Findings Report. Sampling frame on p. iii; sample design in section 2.2.https://www.unicef.org/nigeria/media/6316/file/2021%20MICS%20full%20report%20.pdf
- 5.National Bureau of Statistics and UNICEF (August 2022). 2021 Nigeria MICS and NICS: Statistical Snapshots.https://nigerianstat.gov.ng/download/1241210
- 6.National Bureau of Statistics (Accessed 11 October 2026). Nigeria Living Standards Survey 2018–2019: sampling.https://nigerianstat.gov.ng/nada/index.php/catalog/68/sampling
- 7.National Bureau of Statistics and World Bank (Accessed 11 October 2026). General Household Survey, Panel 2023–2024, Wave 5: study description.https://microdata.worldbank.org/catalog/6410/study-description
- 8.National Bureau of Statistics (2020). Frequently Asked Questions on the Nigerian Living Standards Survey (NLSS) 2018/19.https://nigerianstat.gov.ng/download/1093
- 9.World Bank Microdata Library (Accessed 11 October 2026). Nigeria Demographic and Health Survey 2024: study description. Reproduces the survey's own documentation, including sample design and response rates.https://microdata.worldbank.org/catalog/8029/study-description
- 10.ICF International (September 2012). Demographic and Health Survey Sampling and Household Listing Manual.https://dhsprogram.com/pubs/pdf/DHSM4/DHS6_Sampling_Manual_Sept2012_DHSM4.pdf
- 11.National Bureau of Statistics and World Bank (Accessed 11 October 2026). General Household Survey, Panel 2018–2019, Wave 4: study description.https://microdata.worldbank.org/index.php/catalog/3557/study-description
- 12.INSD and ICF (July 2022). Enquête Démographique et de Santé du Burkina Faso 2021: rapport des indicateurs clés.https://dhsprogram.com/pubs/pdf/PR139/PR139.pdf
- 13.INSTAT Mali, via the World Bank Microdata Library (Accessed 11 October 2026). Mali Enquête Démographique et de Santé 2018: study description.https://microdata.worldbank.org/catalog/3526/study-description
- 14.Institut National de la Statistique, Cameroon, via the World Bank Microdata Library (Accessed 11 October 2026). Cameroon Enquête Démographique et de Santé 2018: study description.https://microdata.worldbank.org/catalog/3717
- 15.REACH (2022). Assessment of Hard-to-Reach Areas: Population Movement and Communication, Adamawa and Yobe, Northeast Nigeria, July–September 2022 (factsheet).https://repository.impact-initiatives.org/document/impact/24754f09/REACH_NGA_Factsheet_Population-movement-and-Communication__Hard_to_Reach-Assessment-in-Northeast-Nigeria_September_2022.pdf
- 16.IOM Displacement Tracking Matrix (February 2020). Nigeria Displacement Report 31. Cited for its methodology section.https://displacement.iom.int/sites/g/files/tmzbdl1461/files/reports/Nigeria%20-%20Displacement%20Report%2031%20%28February%202020%29.pdf
- 17.REACH (2020). Nigeria Multi-Sector Needs Assessment 2020: executive summary.https://repository.impact-initiatives.org/document/reach/cd5f942f/Nigeria-MSNA-2020-Executive-Summary.pdf
- 18.E. Sagmeister and J. Steets, with A. Derzsi-Horváth and C. Hennion, GPPi, SAVE resource paper (October 2016). The use of third-party monitoring in insecure contexts: lessons from Afghanistan, Somalia and Syria.https://assets.publishing.service.gov.uk/media/5f82d1cf8fa8f5044f333664/SAVE__2016__The_use_of_third-party_monitoring_in_insecure_contexts.pdf
- 19.Humanitarian Outcomes, Aid Worker Security Database (Data verified to 31 December 2025). Total incidents by country. A live database; figures can be revised.https://www.aidworkersecurity.org/incidents/report/country
- 20.J. Hoogeveen and U. Pape, in Data Collection in Fragile States (World Bank and Palgrave Macmillan) (2020). Fragility and innovations in data collection. Open access.https://link.springer.com/chapter/10.1007/978-3-030-25120-8_1
How to cite this note
Probatio Africa (2026). What should a survey do about places field teams cannot safely reach? Methods note. Sources checked 11 October 2026. https://probatioafrica.com/evidence/methods/surveying-insecure-areas.
Our text is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0): you may copy, share and adapt it, including in paid work, if you credit Probatio Africa and link to this page. Quotations from laws, standards and other sources keep their owners' terms; cite the original when you reuse one. If you find an error, write to enquiries@probatioafrica.com. We correct errors quickly and record each correction, with its date, at the foot of the note.
Related notes
Every note follows the same format and is checked against its sources.
Sampling
How do you draw a sample in Nigeria when the last census was in 2006?
A usable frame exists, but national surveys are split between old and new frames, and every population total is a projection. What that means for any estimate, and how we handle it.
Read the noteSurvey modes
When can a phone survey stand in for a household visit?
Phone surveys are fast and cheap, and in Nigeria they can reach most households. They miss particular people, and weighting narrows the gap without closing it.
Read the note