Making Everyone Count: Data Limitations and Measurement Challenges - - PDF document

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡

Making Everyone Count: Data Limitations and Measurement Challenges for Arab States in the era of the 2030 Development Agenda

Rita Maalouf, MPH, Romesh Silva, PhD, and Hala Youssef, PhD.

September, 2017

ABSTRACT The 2030 development agenda has been accompanied by calls for a 'data revolution'. In this paper we review data gaps, data quality issues and methodological challenges associated with 2 critical indicators in population health in Arab States. We focus our review on the under-five mortality rate (U5MR) and contraceptive prevalence rate (CPR). Both U5MR and CPR were core indicators in the Millennium Development Goals and are also core indicators in the Sustainable Development Goals (SDGs). Our gap analysis focuses on reviewing data availability by way of surveys, censuses and administrative data for 21 Arab countries between 1990 and 2015. We then discuss population data preparedness for population- related SDGs in 3 Arab states – Egypt, Lebanon, Libya, and Oman – to exploring the diverse challenges faced by the region. We conclude by discussing the demographic measurement challenges in the SDG era with specific attention on subnational disaggregation challenges and their implications for national statistics systems.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡

• 1. ¡ Introduction

In the lead up to 2015, which marked the conclusion of the Millennium Development Goals (MDGs) and the launch of the 2030 development agenda, there has been some useful discussion on the completeness and quality of the evidence-base for monitoring of development progress. At the global level, United Nations (2015) noted that in 2014, 79%

• f low-/middle-income countries had at least 2 data points for 16 or more of the 22 official

MDG indicators. Yet, many countries have weak administrative data systems and under- developed national statistical systems. Thus, there is still heavy reliance on household surveys (such as the demographic and health surveys, and multiple indicator cluster surveys) to estimate basic demographic and social indicators. Smith (2015) in his review of ESCWA country experiences of monitoring the MDGs, noted that ``only one third of the MDG indicators could be measured in the region using

• fficial statistics alone.'' Further, most Arab countries produced data for less than half of

the MDG indicators. His study noted that the major challenges that affected MDG monitoring in the region were data gaps, non-compliance with methodological standards and lack of available disaggregated data. The newly-established Sustainable Development Goals (SDGs), associated with the 2030 development agenda, lay out an ambitious set of 17 goals, 269 targets, and 230 indicators (United Nations, 2015). This is a considerable expansion from 8 goals, 21 targets, and 60 indicators in the MDGs. Further, given their call to `leave no behind', the SDGs outline that most indicators need to be disaggregated by sex urban/rural, where possible, but also by income, gender, age, race, ethnicity, and disability. In this vein, there have been recent calls for a `data revolution' to address the notable data gaps and data quality challenges in many low- and middle-income countries (Gonzales Morales et al., 2015). But much work remains to be done - including basic methodological development. The most recent report by the UN Secretary General on SDGs notes that 60% of the SDG indicators are classified as having an established methodology with either widely available or insufficient data and

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ 40% of indicators are classified as having a methodology currently under development (United Nations, 2016). In the Arab region, there are several additional statistical challenges associated with the monitoring of population dynamics and development progress. These include the effect of conflict, instability and forced displacement on national statistics systems in the region, as well as the measurement challenges associated with large labor migration flows (El-Zein et al., 2016). The 2030 development agenda and its associated sustainable development goals outline an even more ambitious set of goals, targets and indicators than the millennium development goals (SDGs). Gaps in data availability pose challenges to national statistical systems and the regional statistical system in monitoring and evaluating regional, national and subnational progress towards the SDGs. Across the Arab region, there are notable gaps in availability of basic indicators on demographic and social conditions. In our assessment of SDG data readiness in the Arab region, we focus on two core SDG indicators - namely, the under-five mortality rate, and the contraceptive prevalence rate.

• 2. ¡ A Recap of Development Status at the Conclusion of the MDGs Era

Global overview of MDG progress: In 2015, the achievements of the MDG period were summarized globally by “The Millennium Development Goals Report 2015”. The report noted global achievements made on several goals between 1990 and 2015; the number of people living in extreme poverty declined by more than half, the literacy rate among youth aged 15 to 24 increased to around 91%, the global U5MR declined by more than half etc. (United Nations, 2015). However, countries and regions showed different rates of progress. Although Sub-Saharan Africa achieved notable reduction (28%) in the percentage of people whose income is less than 1$ a day, it failed to achieve this target. A global 50% reduction in child mortality did not meet the MDG target of two third

• reduction. Despite having saved millions of lives, this substantial reduction was not

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ equally distributed by wealth quintiles and residence; children in the poorest households and those living in rural areas are the most vulnerable (United Nations, 2015). The need to report on these goals, targets, and indicators required additional effort to that involved in achieving the MDGs. Developing countries struggling with reporting at the beginning of the MDG era were aided by an increasing number of household surveys to report on a variety of indicators (United Nations, 2015). However, the major challenges cited at the end of 2015 were the lack of timely and disaggregated data in various parts of the world. Regional overview of MDG progress: At the launch of the MDG era, 22 Arab countries committed to achieving the targets by

• 2015. However, this commitment was poorly translated as most countries lacked national

policies and frameworks to achieve these goals. Several regional and international efforts were initiated to support and provide resources for monitoring and reporting. Every country selected a specific sub-set of indicators to report on which prevented a comprehensive comparison across the region (Smith, 2015). The final regional review for the Arab region was conducted in 2013 by the United Nations and the League of Arab states. Progress made prior to 2010 had stopped or even reversed for some indicators citing political transitions and conflicts as the possible cause (UN., & LAS, 2013). The report suggests that extreme poverty rates in Syria are expected to revert to their 1990s’ levels maybe even higher by 2015 (2013). Progress towards reducing by three-quarters the maternal mortality ratio has also been slowed between 2010 and 2015. By 2014, Lebanon and Oman were among the countries who achieved this goal and Egypt, Lebanon, Oman, Qatar, Saudi Arabia, and Tunisia achieved Goal 4 of reducing by two-thirds the Under-5 Mortality Rate (World Health Organization, 2014). By the end of 2015, Arab countries were at different stages of development and reporting. Data gaps and Methodological Challenges experienced during MDG Monitoring process:

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ All MDG related reports, global or regional, raised serious concerns about the quality of available data, the periodicity of data collection and reporting of indicators. Several conclusions were drawn from a study of Arab regional experience with sustainable development monitoring and evaluation experience commissioned by United Nations- Economic and Social Statistics Division (UN-ESCWA) (Smith, 2015).

• ¡ Official statistics in the region can produce around one third of the MDG

indicators while the rest are produced using international assistance and surveys demonstrating the inability of the past statistical systems to produce official estimates needed for monitoring, evaluation and accountability.

• ¡ Indicator data are not collected on an annual basis with substantial lag period

between data collection and reporting. Rather it is collected sporadically either based on official statistics, data collected by international agencies and/or modelled data.

• ¡ Major reliance on international assistance during that period for producing data

and indicators. International assistance has been mostly centered on filling the data gaps in the short term. Assistance is needed for a long-term construction or re- construction of statistics infrastructure.

• ¡ Previous Global, Regional and National MDG reports did not stress on the

importance of having data of good quality to monitor and evaluate progress towards the MDGs and most importantly, hold accountable, government officials and responsible parties for the progress, or lack of progress, towards these goals. As an example, the Saudi Arabian Sustainable Development Report was of questionable data quality, with gaps, short time series, and inaccurate data.

• ¡ The lack of nationally produced disaggregated data and reliance on national

household surveys is also an important challenge since evidence shows that data based on models or sporadic data collection fails to include older adults and groups of variable ethnicities (Stuart et al., 2015). Lack of global and regional report at the end of 2015—making it difficult to define starting points for post 2015 agenda.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ Several of these challenges require long-term action plans especially when building a powerful and sustainable national statistical infrastructure. Given that countries concluded their work or reporting for the MDGs at various points prior to 2015, additional challenges arise such as: what is the starting point?

• 3. ¡ The 2030 Development Agenda and Arab States

Despite the considerable progress that has been done by the Arab Countries during the MDG era, challenges cited previously not only halted the progress on several MDG Goals but, their persistence adds substantial weight to the already laborious process of achieving and reporting the SDGs. Conflict, fragile states and migration: While the MDGs have been criticized for excluding children and institutionalized individuals (Jerven, 2014), the following themes discussed have been barely mentioned in the 2030 Development agenda. The aim of the following section is to shed on the light on the implication of such themes before and during the 2030 Agenda. Since 2011, several Arab countries experienced different forms of instability and almost half were affected by war, conflict or disasters (United Nations, 2016). Countries such as Tunisia and Egypt went through a phase of changing regimes with minimal violence compared to countries such as Syria, Iraq and Yemen enduring extremely violent civil wars. These events have contributed to having 18% of all cross- border refugees residing in the Mashreq region consisting mainly of Syrian, Iraqi and Palestinian refugees (El-Zein et al., 2016). The ongoing conflicts are generating population movements in addition to health and environmental burdens for the neighboring countries staling economic and social development (El-Zein et al., 2014). The limited resources used for development in refugee receiving and conflict affected countries are enormously stretched in the face of a crisis which is possibly shifting into a protracted situation with uncertain results and consequences. Conflict has been linked to reversal or halt in the progresses achieved. The World Bank estimates that poverty reduction is slowed by 2 to 7% for every 3 years of violent conflict (Jones & Rodgers,

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ 2011) and the lack of progress in several Arab countries after the year 2010 (UN., & LAS, 2013). In addition, some studies are demonstrating a minimum of 5-year reduction in life expectancy among Syrians (Mokdad et al., 2016). The 2030 Development Agenda was designed during a period of turmoil in the Arab States and several other regions in the world with around 65.6 million displaced individuals by 2016 based on official figures by UNHCR. An important form of migration in this region is Labor migration. The highest percentage

• f labor migrants out of all residents is found in the Gulf Cooperation Council (GCC)

countries; almost half of the residents are non-nationals with limited availability of data disaggregated by nationality (Fargues, 2011). In addition to that, climate driven migration during periods of drought and flooding stresses on the infrastructure noting that these phenomena are experienced differently between urban and rural population (El-Zein et al., 2014). The effects of such events on individuals and their impact on social development has been barely mentioned in the post 2030 Agenda despite the widely-proven links. Mobile individuals seem to be stuck in a hard-to-escape wheel: because of their mobility they are unlikely to take part in any nationally tailored development plan, escape the routine data collection regarding certain development indicators, and hence their status and needs are not reflected in any updated development plan. Subnational monitoring and evaluation challenges: The set of Goals selected require substantial resource mobilization to meet data checkpoints and ultimately the deadline. However, countries facing at least one of the above-mentioned challenges are forced to redirect their resources in the way they seem fit. Lebanon, for example, a country hosting 1.3 million Syrian refugees registered with UNHCR in additional to its fragile institution. The themes discussed are not limited to the Arab States but are worldwide challenges that should have been acknowledged while designing a global roadmap. Jerven in his working paper estimated that $27 billion dollars were spent between 1990 and 2015 to support MDG surveillance and $254 billion dollars will be the estimated cost

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ to report on 169 targets via complementary surveys in the Post-2015 ear (2014). This number, although believed to be an underestimate, is quite large to be invested in surveys complementing a roadmap while it has not been proven that roadmaps, such as the MDGs

• ptimized the use of available resources (Jerven, 2014).

The challenges faced by monitoring the MDGs will unfortunately affect the SDG monitoring process as efforts have not yet allowed the upgrade from complete reliance on sporadic data collection or model based estimates to a comprehensive and up to date statistical infrastructure.

• 4. ¡ Data Gaps, Data Quality Deficits & Measurement Issues

SDG Data Availability We review the data availability of the CPR and U5MR indicators for the MDG monitoring period – identifying notable variation in data availability across the 21 Arab States. Several nationally representative surveys contain a module on reproductive health including the CPR among married women (or women in union) between the ages of 15-49. Based on the available point estimates, the Family Planning Estimation Tool (FPET) produces national or sub-national CPR estimates. The data discussed below has been retrieved from the FPET. The availability of data points (DP) across all the UN-ESCWA countries is explored in Figure 1. The range of DP availability extends from Egypt with14 DP in 15 years to the UAE with only 1 DP. Half of the countries have less than 6 DP for the same time period including Syria, Yemen, Libya, Iraq, Kuwait, Qatar, Oman, Mauritania, Kingdom of Saudi Arabia, UAE, Sudan and South Sudan. Interestingly, we observe that a number of high- income countries in the region have some of the most limited data on CPR and largest levels of uncertainty associated with their CPR estimates between 2000 and 2015. This suggests that in the Arab region, financial resources are not the main limitation for several governments when it comes to strong population health statistics. Figure 1 also displays the average relative size of the half-confidence interval of annualized national CPR estimates. Djibouti has the smallest uncertainty level (10.3%) compared to South Sudan (79.7%). Half of the countries have uncertainty levels greater

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ than 20% including the GCC countries. Most countries with an uncertainty level greater than 20% have less than 6 DP across 15

• years. This may translate into further inaccuracies when generating modeled estimates and

projections. CPR data was also retrieved from the State of the World Population (SWOP) 2015 and compared to estimates from the FPET in Figure 1A. Although most countries have modeled estimates matching with the latest survey generated estimate, the Syrian Arab Republic, Comoros, and Djibouti have different internationally produced estimates by the end of the MDG era. It is important to note that the errors observed are not systematically associated with one method; i.e. a single method is not systematically increasing or decreasing the CPR estimate.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ In Syria, the 2015 SWOP estimate seems to be consistent with the latest national CPR figure around 58% in 2006. Despite this match, it is hard to believe that pre-war CPR estimates remained constant throughout a war that destroyed significant health infrastructure and possibly the delivery and availability of contraceptive methods. On the other hand, the FPET estimates for Comoros and Djibouti seem to be a more realistic figure as they are comparable to the latest national figures in these 2 countries at around 19 and 23 % respectively. The inconsistency sometimes lies in the basic data differences which is either due to different data sources with different timing in original measurements, registration gaps or different projection methodologies. In many cases consultation with the country's national statistical bodies is required for checking the validity. Interestingly, ¡the ¡3 ¡countries ¡with ¡shown ¡in ¡Figure ¡1a ¡have ¡uncertainty ¡levels ¡under ¡20% ¡ with ¡5 ¡to ¡6 ¡DP ¡across ¡15 ¡years ¡based ¡on ¡Figure ¡1. ¡This ¡suggests ¡that ¡modeled ¡estimates ¡ may ¡also ¡be ¡failing ¡under ¡relative ¡certainty ¡levels.

58 24 24 68 13.7 66 66.2 68.2 58.3 19.4 23 67

10 20 30 40 50 60 70 80

Syria Comoros Djibouti Morocco

Figure 1a: CPR Estimate variability across methods and countries.

SWOP-CPR-any-method-2015 FPET-any-method-2015 Latest survey estimate

MICS, 2006 DHS, 2012 MICS, 2006 PAPFAM , 2011

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ CPR estimates generated by the Demographic Health Surveys (DHS) were analyzed for Jordan, Comoros, Egypt, Mauritania, Morocco, Sudan, Tunisia, and Yemen. Aggregate data were available and accessible as well as disaggregated data by age, residence, education, wealth quintile, number of living children and other. Confidence intervals were not available for this specific indicator, and hence the certainty of these estimates could not be generated. In summary, internationally produced aggregate CPR estimates exist for all the countries in the UN-ESCWA region. Estimates for half of the countries are based on a reasonable number of data points (greater than or equal to 5 within a period of 15 years) and reasonable uncertainty (less than or equal to 20%). However, inconsistent model technique generating significantly different estimates may not provide the intended support to countries unable of generating their own estimates. Moreover, these estimates may well be used as a starting point for the Post-2015 era in the absence of nationally generated

• estimates. In the case of such staggering differences: how can countries successfully plan

for the 2030 Agenda? Figure ¡2 ¡provides ¡an ¡overview ¡of ¡the ¡range ¡of ¡Under ¡5 ¡Mortality ¡(U5MR) ¡rate ¡data ¡ availability ¡in ¡the ¡Arab ¡states ¡between ¡1990 ¡and ¡2015. ¡

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ 15 countries have at least 5 data points between the period of 1990 and 2015 with a level

• f uncertainty lower than 20%. Similarly, to the CPR estimates, Kuwait and the UAE,

high-income countries, have 1 and 4 DP respectively, the lowest regionally, while Egypt has the highest with 16 DP. U5MR estimates generated by the Demographic Health Surveys (DHS) were analyzed for Jordan, Comoros, Egypt, Mauritania, Morocco, Sudan, Tunisia, and Yemen. Aggregate data were available and accessible as well as disaggregated data by sex, mother’s age at birth, residence, education, wealth quintile, birth order, etc. Confidence intervals were not available for this specific indicator, and hence the certainty of these estimates could not be generated.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡

• 5. Case Study: Implementation of the 2030 Agenda in Egypt, Lebanon,Libya, and

Oman There are notable challenges in collecting comprehensive and high quality data on the demographic and social situation in the Arab region. These include substantial gaps in data availability at the national level, but also at subnational levels (most notably for disaggregation by sex, urban/rural, national/non-national status). In this section, we present a comparative analysis for 1 Mashreq country (Lebanon), 1 Maghreb country (Libya), 1 North African country (Egypt) and 1 GCC country (Oman). We choose these countries given that they straddle the three main sub-regions in the region and have notably different health and statistics systems. The first factor examined was the country’s latest census because it provides denominators to many SDG indicators. The last census conducted in Lebanon was in 1932 with no plans

• f conducting a new one soon. All Lebanese official statistics rely on projections or

fragmented data which may not be as accurate as census data. Egypt, Libya and Oman had censuses in the past 10 years and plans for to take part in the 2020 Census round; for Egypt and Oman already conducted. With the global advocacy to enhance the status of Civil registration and Vital Statistics (CRVS) system, a review of countries system in the region has been conducted by the World Health Organization. The Lebanese CRVS has been labeled as Weak with incomplete vital registration while the rest of the countries have functional but inadequate

• systems. This evaluation is the first step in strengthening the CRVS systems in the region

(WHO, 2013). Because of the regions heavy reliance on surveys, the latest survey with disaggregated data was also examined as a factor as it could provide denominators. All countries have recent nationally representative surveys with disaggregated data.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ Disaggregated health data are a tool to ensure that the 2030 Agenda does not leave anyone

• behind. Hence, the countries’ compliance with the reporting of disaggregated health data

was also examined. All countries provide disaggregated data to a certain extent except for Libya. Since the Agenda has an inclusive nature and will require substantial resource mobilization, raising awareness about its importance may facilitate linking the Agenda to National Strategies as well as enhancing the participatory decision making process. At the moment, all countries have not yet launched awareness campaigns, however, Egypt has designed a social media plan to compliment the dissemination of their own National Statistical plan. Another important factor is the development of a National Statistical Development Strategy (NSDS) in countries to promote the renaissance of their statistical infrastructure to monitor and evaluate progress. A strategy is under development in Libya and Egypt, while a National Spatial Data Infrastructure has been launched in Oman. No strategies have been proposed in Lebanon yet. Given that SDGs have the highest chance of being achieved under the umbrella of government commitment, a national plan to achieve the SDGs will ensure the implementation of activities, monitoring and timely measurement of indicators. All 4 countries have not yet managed to finalize a National SDG strategy; however, Egypt and Oman are in stages of preparation of a strategy. Lebanon and Libya have not yet launched similar activities. Finally, for countries planning a strategy, budget allocation is very important to ensure government commitment. Egypt and Oman did not yet reach that part during strategy negotiation efforts. Arab countries are more different than alike when it comes to their preparedness for the post-2015 Agenda. Because these differences reflect their limited or poorly distributed

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ resources, the same level of progress and reporting cannot be expected from all countries. Different form of support and planning should be proposed to shift the major load of data reporting from aid-supported surveys to country statistical monitoring. Further development of CRVS systems, household surveys and censuses as core building blocks of national statistical systems is required in addition research on data integration of multiple data sources to improve precision and consistency of SDG indicator estimates. Table 1: Comparison matrix of 4 select Arab countries. Factor Egypt Lebanon Libya Oman Latest Census 2017 1932 2006 2014 Status of Civil registration and Vital Statistics System Functional but Inadequate Weak Functional but Inadequate Functional but Inadequate Latest Nationally representativ e surveys (ex: DHS, MICS ) EDHS, 2014 MICS, 2009 PAPFAM, 2014 MICS,2014

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ Disaggregate d health indicators generated by the governments Gaps are identified and made for by strengthening health systems and digitalizing CRVS at both thesubnationa l and district levels Disaggregated maternal and neonatal mortality rates by nationality Under development Disaggregated data by region, nationality, gender, and age groups are provided through the National Center for Statistics and Information Data Portal Raising public awareness about the 2030 Agenda Is in process, a communicatio n plan has been set in place and is under implantation. The 2030 SDS plan was distributed to all public with the daily newspapers. A social media campaign is taking place. Underdevelopme nt Underdevelopme nt Underdevelopme nt National Statistical Developmen t Strategy (NSDS) Not Available Underdevelopme nt The National Centre for Statistics & Information mandated the establishment of Oman National Spatial Data Infrastructure (Oman NSDI) launched in 2014

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ National plan to achieve the SDGs with multi- stakeholder approaches and tailored to national/ sub-national / local contexts A council led by the ministry of planning and administrative reform is working specifically

• n data issues

with all stakeholders. CAPMAS is representing the Ministry

• f Planning

and administrative reform. Subnational plans are yet to be in place. None Underdevelopme nt Budgeting strategy to Finance the plan NA N/A N/A N/A

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡

• 6. ¡ Discussion ¡and ¡Conclusions ¡

¡ ¡ There ¡have ¡been ¡wide-­‑ranging ¡calls ¡for ¡a ¡‘data ¡revolution’ ¡to ¡underpin ¡the ¡2030 ¡ development ¡agenda ¡and ¡ensure ¡that ¡everybody ¡counts. ¡Yet, ¡there ¡are ¡major ¡data ¡gaps, ¡ in ¡relation ¡to ¡the ¡core ¡population-­‑related ¡SDG ¡indicators ¡across ¡the ¡Arab ¡States ¡region. ¡ Further, ¡even ¡when ¡population ¡data ¡are ¡available, ¡there ¡are ¡considerable ¡data ¡quality ¡ issues ¡-­‑ ¡most ¡notably ¡concerning ¡national ¡representativeness ¡and ¡imprecision ¡of ¡ estimated ¡indicators. ¡Given ¡the ¡variation ¡in ¡country ¡contexts ¡across ¡the ¡region ¡-­‑ ¡with ¡ many ¡countries ¡affected ¡by ¡protracted ¡conflict ¡or ¡large ¡refugees ¡and ¡mix ¡migration ¡flows ¡-­‑ ¡ this ¡necessitates ¡customized ¡approaches ¡to ¡address ¡SDG ¡monitoring ¡and ¡review ¡

• challenges. ¡The ¡2020 ¡census ¡rounds, ¡recent ¡CRVS ¡system ¡strengthening ¡initiatives, ¡and ¡

upcoming ¡national ¡and ¡international ¡household ¡surveys ¡across ¡the ¡region ¡provide ¡a ¡ crucial ¡opportunity ¡for ¡improving ¡SDG ¡indicator ¡monitoring ¡on ¡pop-­‑related ¡issues. ¡ ¡ ¡ Standard ¡approaches ¡to ¡demographic ¡measurement ¡and ¡population ¡estimation ¡are ¡not ¡ well-­‑suited ¡to ¡settings ¡beset ¡by ¡substantial ¡population ¡upheaval ¡or ¡large ¡migration ¡flows ¡ — ¡without ¡detailed ¡data, ¡which ¡are ¡rarely ¡available ¡in ¡such ¡settings. ¡In ¡that ¡vein, ¡for ¡ substantial ¡parts ¡of ¡the ¡region ¡it ¡continues ¡to ¡be ¡difficult ¡to ¡even ¡measure ¡the ¡population ¡ size, ¡structure ¡and ¡composition ¡as ¡well ¡as ¡Tier ¡I ¡SDG ¡indicators. ¡Further, ¡modeled-­‑based ¡ estimation ¡and ¡projection ¡techniques ¡to ¡such ¡settings ¡are ¡difficult ¡to ¡calibrate ¡in ¡a ¡way ¡ that ¡captures ¡the ¡underlying ¡population ¡dynamics ¡and ¡equally ¡difficult ¡to ¡validate. ¡ ¡ For ¡many ¡parts ¡of ¡the ¡region, ¡a ¡hybrid ¡approach ¡that ¡focuses ¡on ¡rapid ¡population ¡and ¡ demographic ¡measurement ¡in ¡conjunction ¡with ¡national ¡statistical ¡systems ¡strengthening ¡ is ¡required. ¡This ¡will ¡require ¡innovation ¡in ¡the ¡areas ¡of ¡measurement ¡of ¡population ¡and ¡ demographic ¡indicators ¡in ¡conflict-­‑affected ¡societies ¡and ¡improvements ¡in ¡migration ¡flow ¡ and ¡stock ¡data. ¡It ¡will ¡also ¡require ¡better ¡utilization ¡of ¡available ¡data ¡-­‑ ¡particularly ¡using ¡ data ¡integration ¡methods ¡that ¡combine ¡multiple ¡data ¡sources ¡and ¡estimation ¡methods. ¡ This ¡is ¡likely ¡to ¡entail ¡the ¡trial ¡and ¡use ¡of ¡hybrid ¡approaches ¡that ¡combine ¡multiple ¡types ¡

• f ¡data ¡— ¡traditional ¡population ¡data ¡(collected ¡via ¡surveys, ¡censuses, ¡and ¡

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ adminisytrative ¡data ¡systems), ¡but ¡also ¡new ¡data ¡sources ¡(collected ¡via ¡remote ¡sensing, ¡ satellite ¡imagery, ¡and ¡new ¡technologies), ¡and ¡perhaps ¡even ¡qualitative ¡information. ¡ However, ¡these ¡realities ¡and ¡customized ¡approaches ¡need ¡to ¡be ¡reflected ¡in ¡national ¡ statistical ¡development ¡strategies, ¡2030 ¡development ¡agenda ¡implementation ¡plans ¡and ¡ national/international ¡financing ¡for ¡development. ¡ ¡ ¡ ¡ ¡ ¡

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ References ¡ El-Zein, A., Jabbour, S., Tekce, B., Zurayk, H., Nuwayhid, I., Khawaja, M., ... & Hogan,

• D. (2014). Health and ecological sustainability in the Arab world: a matter of
• survival. The Lancet, 383(9915), 458-476.

El-Zein, A., DeJong, J., Fargues, P., Salti, N., Hanieh, A., & Lackner, H. (2016). Who's been left behind? Why sustainable development goals fail the Arab world. The Lancet, 388(10040), 207-210. Fargues, P. (2011). Immigration without inclusion: non-nationals in nation-building in the Gulf States. Asian and Pacific Migration Journal, 20(3-4), 273-292. Jerven, M. (2014). Benefits and costs of the data for development targets for the Post-2015 Development Agenda. Data for Development Assessment Paper Working Paper,

• September. Copenhagen: Copenhagen Consensus Center.

Jones, G. A., & Rodgers, D. (2011). The World Bank's World Development Report 2011

• n conflict, security and development: a critique through five vignettes. Journal of

International Development, 23(7), 980-995. Mokdad, A. H., Forouzanfar, M. H., Daoud, F., El Bcheraoui, C., Moradi-Lakeh, M., Khalil, I., ... & Wagner, J. (2016). Health in times of uncertainty in the eastern Mediterranean region, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet Global Health, 4(10), e704-e713. Stuart, E., Samman, E., Avis, W., & Berliner, T. (2015). The data revolution: Finding the missing millions. Overseas Development Institute. Smith, R. (2015). Measuring Sustainable Development in the Arab region: A review of country experiences and recommendations for monitoring and evaluation post-

• 2015. Beirut: UN-ESCWA.

United Nations, (2015). The millennium development goals report. Washington: United Nations. UN., & LAS. (2013). The Arab Millennium Development Goals Report: Facing Challenges and Looking Beyond 2015.

DRAFT ¡PAPER ¡ ¡ ¡ ¡ ¡ ¡ ¡ Please ¡do ¡not ¡cite ¡or ¡circulate ¡

¡ World Health Organization. (2013). Strengthening civil registration and vital statistics for births, deaths and causes of death: resource kit. World Health Organization. (2014). Progress report on the achievement of the health- related Millennium Development Goals and global health goals after 2015.