“There are huge benefits to be gained from producing statistics that are familiar to, and usable by, governments and other development partners.“
When the Kenya National Bureau of Statistics published labour force statistics from its 2015–16 integrated household survey, it found that 72% of working-age Kenyans were employed. In Turkana County, the rate was 62%. However, this national survey excludedrefugee camps. A survey commissioned by UNHCR and the World Food Programme in 2016 found that only 16% of households in Kakuma refugee camp – located in Turkana County – reported having at least one employed person in the household.
These two statistics are hard to compare for several reasons. The national survey asks about employment at an individual level, the refugee survey at a household level. And while the national survey, implemented with technical support from the World Bank, aligns with the definitions relating to labour statistics as prescribed by international bodies and standards, the refugee survey asks the simple question ‘How many in the household have work?’, making comparison even more difficult.
While implementation of both surveys was technically robust overall, this example illustrates how surveys implemented by humanitarian organisations are often developed with a specific humanitarian purpose in mind – in this case, to explore options for targeting humanitarian assistance. In pursuing these valid objectives, international statistical standards and best practices are all too often forgotten or discarded as not applicable or overly complex.
However, satisfying the primary humanitarian purpose of a survey while at the same time aligning it with statistical standards comes with huge benefits. In fact, at a time where both the importance of national leadership and of humanitarian– development collaboration in addressing forced displacement is widely acknowledged, and where protracted situations constitute the vast majority of forced displacement, every humanitarian survey that does not ‘speak the language’ of government and development partners reflects a missed opportunity.
Benefits of alignment
Aligning humanitarian survey work with international statistical standards and best practices will allow the insights produced from the data to be used more effectively in policy dialogue and advocacy, because the survey statistics will be based on concepts that decision-makers are familiar with. Moreover, collecting data in a way that is aligned with national/official survey methods allows for some degree of comparison between forcibly displaced and national or local host communities, even if the latter are not explicitly included in the survey sample. Furthermore, applying tried and tested (and readily available) statistical standards can reduce the cost and complexity of survey design. Finally, the use of internationally established and recognised measurement practices can increase the attractiveness of the data to researchers for secondary use, thereby enhancing the impact of and return on any investment in data collection.
Concerns that using these standards in humanitarian survey work could be burdensome or impracticable in light of resource and capacity constraints are understandable but ultimately unfounded in most cases – especially when it comes to contexts of protracted displacement (as opposed to emergencies). Since these international standards have been carefully developed to apply in contexts as diverse as Norway and the Central African Republic, and across the full socioeconomic spectrum within these countries, they are also robust in forced displacement settings. The standards are well supported by useful documentation and usually come with guidance material aimed at data practitioners from a variety of backgrounds. And finally, since the marginal cost of expanding an interview by a few questions is negligible in most cases, their application often does not increase survey costs in any notable way.
Good practice in Kenya
UNHCR, in collaboration with the World Bank, conducted two further household surveys in Kenya’s Kalobeyei and Kakuma refugee camps, in 2018 and 2019 respectively. This time, the survey questionnaire was modelled largely on the national integrated household survey, which in turn aligns with a wide variety of statistical standards. The results from the surveys allow for direct comparison of the camps’ population with their national hosts. We now know that at the time of the surveys 37% of the working age population in Kalobeyei camp were employed, compared with 62% in Turkana County and 72% in Kenya overall. We also learned that 58% of refugees in the camp live below the national poverty line, as compared with 72% of the population of Turkana County and 37% across Kenya.
The governor of Turkana County, Josphat Nanok, welcomed the comparable statistics with the words “government now has data!”, and outlined how they would be used to inform national and sub-national policymaking, including in incorporating refugees and asylum seekers in the national education system. The governor also stressed that the Kalobeyei survey would inform the national statistical office’s decision to extend its national household survey to the refugee camps.
The decision to better align humanitarian surveys with international statistical standards is hardly a trade-off at all, especially in protracted situations. The returns far outweigh the concerns. As more humanitarian surveys incorporate these standards, the methodological divide between humanitarian surveys and their government and development equivalents will shrink. In parallel, learning from humanitarian survey work will increasingly feed into the development and refinement of the survey standards themselves.
 KNBS (2018) Labour Force Basic Report, 2015/16 Kenya Integrated Household Budget Survey bit.ly/KNBS-survey-746
 UNHCR/WFP/Kimetrica (2016) Refugees Vulnerability Study Kakuma, Kenya bit.ly/UNHCR-WB-Kimetrica-Kakuma-2016
 Such as those developed under the auspices of the UN Statistical Commission or those specified in SDG indicator metadata sheets.
 . Humanitarian agencies are increasingly moving towards an ‘open microdata’ approach. For example, UNHCR in 2020 launched its Microdata Library, where much of the survey microdata it collects is available to external data users in anonymised form. https://microdata.unhcr.org
 UNHCR/World Bank (2020) Understanding the Socioeconomic Conditions of Refugees in Kalobeyei, Kenya: Results from the 2018 Kalobeyei Socioeconomic Profiling Survey bit.ly/UNHCR-WB-Kalobeyei-2018 Report for the Kakuma survey report forthcoming in March 2021
 . Speech delivered at Global Refugee Forum 16 December 2019. Audio recording at: bit.ly/GRF-recording
Felix Schmieding is a Senior Statistician with the Center. His earlier work includes assignments with UNHCR, UNDP, and the UN Statistics Division.