The Sustainable Development Goals (SDGs) pledge that “no one will be left behind” and that we endeavour to “reach the furthest behind first”. But this requires us - and especially decision-makers - to understand if public policies in education, health and social protection is actually benefiting the poorest groups of society. Indicator 1.b.1, for which Save the Children and UNICEF are custodian agencies, is helping us to do exactly that, estimating how ‘pro-poor’ public spending actually is.

**[A new post on the World Bank Data Blog presents more details on this measure and shares some key insights from the data](https://blogs.worldbank.org/en/opendata/who-benefits-from-public-spending--new-evidence-on-pro-poor-inve)**, which covers now 133 countries. Please have a look over there to learn more about how pro-poor spending helps us to realise child rights. All the data you see below is now available in the Global SDG Database as well as right here, on the Child Atlas (under ‘Financing & Accountability‘):
- [Pro-poor social spending (total)](https://www.childatlas.org/atlas?indicator1=PROPOOR_TOTAL&topic1=FA.PP)
- [Pro-poor health spending](https://www.childatlas.org/atlas?indicator1=PROPOOR_HEALTH&topic1=FA.PP)
- [Pro-poor education spending](https://www.childatlas.org/atlas?indicator1=PROPOOR_EDUCATION&topic1=FA.PP)
- [Pro-poor social spending (direct transfers)](https://www.childatlas.org/atlas?indicator1=PROPOOR_TRANSFERS&topic1=FA.PP)
(The first measure on total pro-poor spending also includes countries for which data exists only for one or two sectors - please have a look at the footnotes there, as not all data points are comparable.)

<iframe src='https://flo.uri.sh/visualisation/22438722/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe>

Sarah, 14, solving a math problem on the chalkboard in the classroom

large_CH11333744_Sarah, 14, solving a math problem on the chalkboard in the classroom.jpg

medium_CH11333744_Sarah, 14, solving a math problem on the chalkboard in the classroom.jpg

small_CH11333744_Sarah, 14, solving a math problem on the chalkboard in the classroom.jpg

thumbnail_CH11333744_Sarah, 14, solving a math problem on the chalkboard in the classroom.jpg

CH11333744_Sarah, 14, solving a math problem on the chalkboard in the classroom.jpg

A new post on the World Bank Data Blog presents more details on pro-poor public spending and introduces data for 133 countries, helping us to realise child rights and the SDG pledge to 'Leave No One Behind'.

Who benefits from public spending? New evidence on pro-poor investments

In a warming world of increasing frequency and severity of natural disasters, it is essential to find approaches and methods that track disaster’s impacts and recovery efforts. Remote sensing data and satellite imagery are powerful tools to provide almost real-time data and is increasingly used to support disaster recovery. However, could these approaches also be used in the long-term to chart the economic recovery after the disaster over time and space? Last year, a research project at Save the Children UK explored the potential use of night-time light (NTL) imagery to understand such long-term impacts.

NTL is a form of satellite imagery of the world at night (seen in Figure 1) which shows artificial light sources such as buildings, streetlamps and infrastructure. This data is increasingly easy to access, for example, the [World Bank has created a tool](https://worldbank.github.io/blackmarbler/) which allows free access to [NASA’s black marble project](https://blackmarble.gsfc.nasa.gov/) to use directly in statistical and data science tools.  

These images have been commonly used as a proxy for economic activities and development as the lights show where large urban centres are located and how they have expanded. In some studies, researchers have started connecting NTL with disasters and recovery efforts. In some studies, researchers have started connecting NTL with disaster and recovery efforts, as seen in Figure 1 below, where [Skoufias et. al. (2021)](https://www.tandfonline.com/doi/full/10.1080/19475705.2021.1879943) has plotted NTL changes before and after typhoons in the Phillipines. Others such as [Zhang et. al. (2023)](https://www.mdpi.com/2072-4292/15/17/4257) has compared regional NTL trends during disasters to disaster resilience curves.

_Figure 1: Light intensity before, during and after typhoons Haiyan and Rammasun in the Phillipines (source: [Skoufias et. al. 2021](https://www.tandfonline.com/doi/full/10.1080/19475705.2021.1879943))._

![nighlights-shocks-1.png](https://res.cloudinary.com/gwlvoices/image/upload/v1743676469/nighlights_shocks_1_b683672a1b.png)

For my [UCL Social Data Institute internship](https://www.ucl.ac.uk/social-data/internships) with Save the Children UK, I conducted a research project to explore the use of NTL to better understand the socio-economic impacts of disasters. Specifically, I looked at trends before and after disasters in a few selected case studies in Madagascar and Zimbabwe. To do this, I combined NASA’s black marble NTL data with the [Geo-coded Disaster Dataset (GDIS)](https://www.researchgate.net/publication/349366416_GDIS_a_global_dataset_of_geocoded_disaster_locations) which provided the locations of disasters. Then, I ran a few preliminary plots of the NTL before and after each disaster and some statistical tests. One plot can be seen in Figure 2, which shows the average monthly NTL a year before and a year after a flash flooding event in Goromonzi, Zimbabwe (2014). The graph shows that NTL brightness was reduced in the 6 months after the flooding event, before increasing again. I created a dataset with similar information for flood, storm and drought events in Zimbabwe and Madagascar from 2012 to 2018 (latest GDIS update).  

_Figure 2: Average NTL values before, during and after flash flooding in Goromonzi, Zimbabwe in 2014_

![nighlights-shocks-2.png](https://res.cloudinary.com/gwlvoices/image/upload/v1743676469/nighlights_shocks_2_2e507288e1.png)

Our project was limited by the short time frame available, the slow processing of nightlight data and the lack of detailed disaster locations; the database mentioned above is not very granular. More work is needed in finding statistical relationships between disasters and NTL changes and thus economic impacts. Furthermore, the type and severity of disasters likely play an important role and it may be helpful to triangulate NTL imagery with other data sources such as the census and locally collected data. 

However, we know from other studies that nightlights can play an important role in identifying long-term and significantly economic effects of disasters, as just recently shown for floods in Bangladesh ([Patel 2024](https://dev-patel.com/content/Floods.pdf)). My project has shown how NGOs, researchers, governments and other stakeholders can work with NTL data in the future – even when using quite basic statistical software and methods – and highlight the role that these innovative methods can play to track economic changes after disasters.


Zimbabwe is a lower middle income country in Eastern & Southern Africa. It ratified the UN Convention on the Rights of the Child. In 2015 the country committed to support delivery of the 17 Sustainable Development Goals by 2030 and the pledge to leave no one behind.

Zimbabwe

Madagascar is a low income country in Eastern & Southern Africa. It ratified the UN Convention on the Rights of the Child. In 2015 the country committed to support delivery of the 17 Sustainable Development Goals by 2030 and the pledge to leave no one behind.

Madagascar

large_nightlights-hero.jpg

medium_nightlights-hero.jpg

small_nightlights-hero.jpg

thumbnail_nightlights-hero.jpg

nightlights-hero.jpg

In a warming world of increasing frequency and severity of natural disasters, it is essential to find approaches and methods that track disaster’s impacts and recovery efforts. Remote sensing data and satellite imagery are powerful tools to provide almost real-time data and is increasingly used to support disaster recovery. 

Nightlights and shocks: Understanding disaster recovery pathways using night light data

With just six years remaining, the world is in a race against time to achieve the SDGs. Despite progress on some of the goals since 2015, [our new report](https://resourcecentre.savethechildren.net/document/racing-against-time-achieving-the-sustainable-development-goals-with-and-for-children-global-report/) shows that most countries are off-track to achieve key promises for children. We also used this opportunity to share key stats [on countries’ progress towards the 2030 Agenda](https://www.childatlas.org/sdg) directly in various tools of the Child Atlas. Where possible, we are looking beyond averages to demonstrate significant – and often growing – inequalities in many countries around the world.

Selected child-relevant SDG indicators illustrate the lack of progress, painting a pessimistic picture of countries’ abilities to deliver key child rights by the end of this decade. We use UNICEF’s assessment on progress towards the SDGs and transfer these into an easily understandable traffic light system (we explain our [methodology more in detail here](https://www.childatlas.org/blog/measuring-sdg-progress)). The result is our [new SDG heatmap](https://www.childatlas.org/sdg), which presents a quick overview of countries’ progress across selected SDG particularly relevant to children. Each goal can be opened to see progress against individual indicators, allowing even more detailed analyses. 

![sdg1.gif](https://res.cloudinary.com/gwlvoices/image/upload/v1727259066/sdg1_a821b02a1e.gif)

You find the same information on the Atlas page (or within country pages) whenever you choose an indicator which is part of the SDGs. For instance, we can see that most countries are off track to reach SDG 2 (we are looking at stunting in particular) – not surprising with levels of hunger globally now higher than in 2015 when the SDGs were agreed!

![sdg2.PNG](https://res.cloudinary.com/gwlvoices/image/upload/v1727259066/sdg2_9cf250f1df.png)

But national averages hide significant and consistent inequalities, and children are regularly denied their rights because of where they grow up, their gender or ethnicity, whether they have a disability, or the economic backgrounds of their parents. The SDGs provided a compelling approach to combating what are manifestly unfair inequalities, pledging that “no one will be left behind […] and we will endeavour to reach the furthest behind first”.

Unfortunately, progress in most countries does not live up to this promise. We use publicly available data on selected child-relevant indicators and calculate the annual rate of reduction since 2000 for each group on which we have data. We then classify a country as experiencing inclusive progressive, if over the last two decades they both progressed towards the targets as well as closed gaps between groups (boys and girls, rural and urban areas, and richer and poorer families). The findings from [the report](https://resourcecentre.savethechildren.net/document/racing-against-time-achieving-the-sustainable-development-goals-with-and-for-children-global-report/) summarise our findings: in most countries, we didn’t see inclusive progress and gaps between more disadvantaged groups and their better-off peers have widened in many cases. Coming back to our example of malnutrition: just 18 of 116 countries, for which we have data, have made inclusive progress to reduce stunting.

![incl_prog.PNG](https://res.cloudinary.com/gwlvoices/image/upload/v1727259066/incl_prog_5d81219252.png)

You can also find these statistics directly in the Child Atlas for selected indictors where this level of detail exists – both on the Atlas page or for individual countries. Unfortunately, almost half of all countries lack publicly available and easily accessible information on individual groups, making it often difficult to track the pledge to Leave No One Behind.

![sdg3.PNG](https://res.cloudinary.com/gwlvoices/image/upload/v1727259066/sdg3_6453c6a4a9.png)

These findings should be a wake-up call, not only for better disaggregated and granular data, but crucially to design policies and programmes that actively reducing inequalities and to live up to the pledge to Leave No One Behind – especially with only 6 years left!

Please do have a look at our report [‘Racing Against Time’](https://resourcecentre.savethechildren.net/document/racing-against-time-achieving-the-sustainable-development-goals-with-and-for-children-global-report/), which provides much more detail than the few paragraphs here. Importantly, they also include Children’s scorecards, allowing children directly to express their views on progress towards the SDGs, adding a crucial piece of accountability to the SDG progress. 

Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs

large_Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs.jpg

medium_Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs.jpg

small_Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs.jpg

thumbnail_Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs.jpg

Lila,17,Colombia-From Generation to Generation. Our past, present, and future with the SDGs.jpg

With just six years remaining, the world is in a race against time to achieve the SDGs. Despite progress on some of the goals since 2015, we find that most countries are off-track to achieve key promises for children. New visualisations allow to better understand SDG progress directly in the Child Atlas - where possible, looking beyond averages to demonstrate significant – and often growing – inequalities in many countries around the world.

Children's progress on the SDGs: Racing against time

A [record 766 million children](https://www.savethechildren.org.uk/news/media-centre/press-releases/one-third-of-children-worldwide-exposed-to-heatwaves) – one-third of the global child population – were exposed to extreme heatwaves in the twelve months from July 2023 to June 2024. The number of children affected by extreme heatwaves almost doubled from 2022/23 to 2023/24. Most children experiencing a heatwave were living in South Asia (213 million), followed by East Asia and Pacific (129 million) and West and Central Africa (117 million).
In the same period, 344 million children experienced the highest temperature recorded in their location since at least 1980. 
![heatwave_childrenaffected.png](https://res.cloudinary.com/dmhlewc3v/image/upload/v1725458264/heatwave_childrenaffected_64eb9796d5.png)

## Heatwaves risk children’s health and overall wellbeing
Heatwaves impact children’s health, education, nutrition, environment, and wellbeing in different ways, as we explain [in this brief](https://resourcecentre.savethechildren.net/document/policy-brief-extreme-heat-and-childrens-development-and-wellbeing/). Among others, extreme weather conditions are leading to an increase in child hospitalisations, respiratory conditions like asthma, and impact children’s mental health and overall development. They are also worsening existing inequalities and food insecurity.

## Where does this data come from
We estimated the number of children affected by heatwaves using satellite imagery of surface temperatures covering every part the world. First, we needed to clarify what we understand as a heatwave, as there is no universally accepted definition. For the purpose of this analysis, we are defining a heatwave as a period of at least three consecutive days with both the minimum and maximum of the daily surface air temperature exceeding a heatwave temperature threshold. This threshold is determined by the 99th percentile of temperatures observed in a specific location over the 1991-2020 reference period. In other words: we determine that a specific location experiences a heatwave if the temperatures for at least three consecutive days are above the hottest 1% of temperatures measured in the previous 30 years. 
Once this was done, we used satellite data on the daily surface air temperature from the Copernicus Climate Change Service (C3S) of the European Union to identify the locations affected by a heatwave between July 2023 and June 2024 (see map below). Finally, we combined this information with the distribution of the child population globally (using gridded child population from WorldPop and country-level estimates from the United Nations World Population Prospects). The [full methodology](https://resourcecentre.savethechildren.net/document/methodological-note-children-affected-by-heatwaves/) explains the previous steps in more detail.

<iframe src='https://flo.uri.sh/visualisation/18593854/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe>

## More heatwaves in the future
Heatwaves are likely to become more common as a consequence of climate change. [Data in the Child Atlas on the risk of heatwaves](https://www.childatlas.org/atlas?indicator1=CLIMATE_RISK_HEAT&topic1=CI.CR) (stemming from global climate models and therefore not directly comparable to the figures above) suggest that the share of children experiencing at least one heatwave per year is increasing from 26% in 2020 to 64% by the end of the century ([other studies expect even higher risks of heatwaves for children in the future](https://news.un.org/en/story/2022/10/1129852)). World leaders must therefore demonstrate that they are willing to take bold action to protect children from heatwaves and other climate-related events – [our policy briefing lays out the details of this](https://resourcecentre.savethechildren.net/document/policy-brief-extreme-heat-and-childrens-development-and-wellbeing/).

![heatwave_trend.png](https://res.cloudinary.com/dmhlewc3v/image/upload/v1725458264/heatwave_trend_2f47e7b231.png)

large_CH1888592_Jamer, 12, observes the effects of El Niño phenomenon in his Wayuú community in La Guajira, Colombia..jpg

medium_CH1888592_Jamer, 12, observes the effects of El Niño phenomenon in his Wayuú community in La Guajira, Colombia..jpg

small_CH1888592_Jamer, 12, observes the effects of El Niño phenomenon in his Wayuú community in La Guajira, Colombia..jpg

thumbnail_CH1888592_Jamer, 12, observes the effects of El Niño phenomenon in his Wayuú community in La Guajira, Colombia..jpg

CH1888592_Jamer, 12, observes the effects of El Niño phenomenon in his Wayuú community in La Guajira, Colombia..jpg

A record 766 million children—one-third of the global child population—were exposed to extreme heatwaves in the twelve months from July 2023 to June 2024. The number of children affected by extreme heatwaves almost doubled from 2022/23 to 2023/24.

Extreme heat and wellbeing: 766 million children exposed to heatwaves in 2023/24

In early August,  the international community celebrated World Breastfeeding Week, an opportunity to reflect on the benefits of breastfeeding and address the gaps in support and prevalence. Breastfeeding gives children a healthy start in life and is the single most effective intervention to prevent under-five deaths. It is also better for the planet and for family finances, as breastmilk is cheaper than formula and the ultimate zero-waste ‘product.’ In a nutshell, with breastfeeding being beneficial for both the mother and the child, it sounds like a no-brainer. [Research](https://internationalbreastfeedingjournal.biomedcentral.com/articles/10.1186/s13006-020-00326-4) shows evidence on breastfeeding having cognitive and health benefits for infants.

Breastfeeding is on the rise globally, with a significant increase in exclusive breastfeeding for the first six-months, reaching 48% in 2023 – a 10-percentage point increase over the past decade. This is close to the World Health Assembly target of 50% by 2025. But the world is not on track to achieve the 2030 target of at least 70%. Too many children still miss out on the benefits of breastfeeding, with severe consequences on child mortality, malnutrition, and overall development. On average, [more rural children are breastfed than their urban peers (54% vs 45%)](https://openknowledge.fao.org/server/api/core/bitstreams/f1ee0c49-04e7-43df-9b83-6820f4f37ca9/content/state-food-security-and-nutrition-2023/global-nutrition-targets-trends.html) and children from poorer households are generally more likely to be breastfed than their richer peers.  One of the reasons for this is a [major rise in formula breastmilk substitutes, driven by aggressive marketing from the $55 billion formula milk industry](https://resourcecentre.savethechildren.net/document/dont-push-it-why-formula-milk-industry-must-clean-its-act/), which has led urban and wealthier households to limit or abandon breastfeeding. Our [new report](https://resourcecentre.savethechildren.net/pdf/BMS-Final-Report.pdf/) exposes how digital marketing of breastmilk substitutes, particularly through social media and influencer marketing, undermines parents’ informed choice.

The Child Atlas brings together UNICEF data on the percentage of infants aged 0 to 5 months of age who were fed exclusively with breastmilk during the previous day. Let’s look at exclusive breastfeeding in three Eastern and Southern African countries, Rwanda, Ethiopia, and Lesotho , which have higher rates of breastfeeding than many other countries. Rwanda has the highest exclusive breastfeeding rate in the world at 81%, while Lesotho and Ethiopia both have 59% rates. Looking at the recent disaggregated data, we find higher rates of exclusive breastfeeding in rural households compared to urban households in all these three countries as well. 

![Exclusive Breastfeeding_comapreAndAnalyzeName.png](https://res.cloudinary.com/dmhlewc3v/image/upload/v1723823069/Exclusive_Breastfeeding_comapre_And_Analyze_Name_df49865504.png)

The trend graphs above depict an overall fall in exclusive breastfeeding in Rwanda but a significant rise in Zambia and Lesotho. When looking at disaggregated trends by residence, we observe overall progress for both groups in all three countries, but the rate of progress varied by groups and across countries. Ethiopia shows an upward trend in both rural and urban residences till 2016, when urban households had caught up with rural households with a significant increase in breastfeeding. Similarly in Rwanda we see a considerable rise till 2015 among urban children when the gap between the rural and urban breastfeeding was the lowest, but the progress for rural breastfeeding was merely staggering. Lesotho had the highest rate of progress among the three countries.

Gaps between the two groups have reduced significantly in Rwanda and Ethiopia but concerningly it is increasing markedly in Lesotho. Analysing the trends, we observe that in Rwanda progress in breastfeeding among rural households have been very slow while urban households had a higher rate of progress bringing them closer to rural rates. In Ethiopia, the rate of rural breastfeeding was higher but urban households have caught up with them, closing the gap. In Lesotho rural progress rate is higher than urban rate and the gaps are increasing with time (although the last survey suggests gaps are closing, unfortunately by reduced rates in rural areas).   

Increasing exclusive breastfeeding rates can reduce inequalities in societies. The change in exclusive breastfeeding needs to be analysed together with disaggregated data and the trajectory of sub- groups. Sure, there has been commendable relative achievements as the three countries are doing better than others but there is still work that needs to be done to change the magnitude and course of exclusive breastfeeding. 

large_CH1703842_Nyandeng, 36, and her two-year-old daughter Aluel at a stabilisation centre in Akobo West, South Sudan..jpg

medium_CH1703842_Nyandeng, 36, and her two-year-old daughter Aluel at a stabilisation centre in Akobo West, South Sudan..jpg

small_CH1703842_Nyandeng, 36, and her two-year-old daughter Aluel at a stabilisation centre in Akobo West, South Sudan..jpg

thumbnail_CH1703842_Nyandeng, 36, and her two-year-old daughter Aluel at a stabilisation centre in Akobo West, South Sudan..jpg

CH1703842_Nyandeng, 36, and her two-year-old daughter Aluel at a stabilisation centre in Akobo West, South Sudan..jpg

Breastfeeding gives children a healthy start in life and is the single most effective intervention to prevent under-five deaths. Despite progress, the world is not on track to achieve the 2030 target of at least 70%. Data in the Child Atlas shows that more rural children are breastfed than their urban peers.

From City to Countryside: Examining differences in exclusive breastfeeding

In 2015, leaders from every country in the world agreed to 17 goals to create a better future for all people, everywhere, by 2030. The Sustainable Development Goals promised that no one will be left behind by progress, and that people who are furthest behind are reached first.

New visualisations across the Child Atlas help us to monitor our progress towards these essential goals for children. Here we explain how we determine countries’ progress across different measures. 

## Past progress to the SDGs
In 2023, UNICEF has conducted an [extensive analysis to measure and benchmark progress on 48 child-relevant SDG indicators](https://data.unicef.org/resources/sdg-report-2023/). As part of this analysis, they have estimated how strong national-level progress has been towards the SDGs so far for each indicator. Comparing different performances across countries and time, recent progress has been categorised into 5 categories (very weak/weak/average/strong/very strong).

You can find information on past progress for any SDG indicator on the Atlas page or within country profiles.

## Inclusive progress
Focusing on changes in the national average hide significant and consistent inequalities, as children are regularly denied their rights because of where they grow up, their sex or ethnicity, whether they have a disability, or the economic backgrounds of their parents. The SDGs provided a compelling approach to combating what are manifestly unfair inequalities, pledging that “no one will be left behind […] and we will endeavour to reach the furthest behind first”. Ensuring that ‘all segments of society’ achieve the 2030 goals requires that those most distant from the target move faster than groups already closer to the targets. 

We analyse this process of convergence to understand if gaps are narrowed between those most impacted by inequality and discrimination, and their more advantaged peers. Using publicly available disaggregated data on selected indicators (all of which can be found directly in the Child Atlas), we calculate the annual rate of reduction since 2000 for each group on which we have comparable data.  This includes currently gender, residence and wealth. We then classify a country as experiencing inclusive progressive, if over the last two decades they 1) progressed towards the targets set out in the 2030 agenda, and 2) in that process closed gaps between boys and girls, rural and urban areas, and richer and poorer families.

You can find information on inclusive progress for selected SDG indicators on the Atlas page or within country profiles.

## Country on track to reach the SDGs
While the previous measures focus on past progress towards the SDGs, this particular one looks ahead, estimating if the country as a whole is likely to reach the goals by 2030. For this particular measure, we again rely on [UNICEF’s assessment for each of the 48 child-relevant SDG indicators](https://data.unicef.org/resources/sdg-report-2023/). We then reclassify their findings into an easily understandable traffic light system: green indicating that a country has either already met a target or needs only very low effort to get there. Yellow shows countries that require low or average levels of effort to reach the target. Red indicates countries which need high or very high levels of effort.

You can find this measure on the Atlas page or within country profiles. We also provide an easy heatmap for ‘countries on track’ for selected SDG indicators on our SDG overview page. The heatmap initially shows an overall rating for SDGs 1-6 and 16. This traffic-light-rating depends on the average progress for individual SDGs (as long as data is available for at least half of the indicators). You can explore the individual indicators by expanding each of the goals.

## Groups on track to reach the SDGs
Mirroring the analysis of inclusive progress, our last measure focuses on group-specific progress towards the SDGs. This allows us to estimate if countries are on track to deliver on the pledge to deliver the goals by 2030 for ‘all segments of society’.

Using disaggregated data for selected SDG indicators, we estimate if all groups are on track to reach the SDGs, comparing their individual rate of change in the past with the change required to reach the SDG targets. For each indicator, we can classify countries into one of the three categories: 1) all groups reach target, 2) the national average and some groups reach target, 3) the national average and most groups are off target.

You can find information on this measure for selected SDG indicators on the Atlas page or within country profiles.

## Data availability 
The lack of comparable and high-quality data even halfway through the SDGs means we can only paint an incomplete picture of where the world is delivering on their promises for children. Despite impressive progress in many statistical organisations to develop methodologies and collect data for around 230 indicators across the 17 goals, the remaining lack of data – shown through the grey areas across our visualisations – means governments and international organisations still flying blind when making decisions affecting children’s rights.

This is particularly true for disaggregated information on individual groups, making it particularly difficult to track the pledge to Leave No One Behind. Furthermore, data is consistently missing for many other communities impacted by discrimination and facing deprivations, which means our analyses of inclusive progress and ‘groups on track to reach the SDGs’ is focused only on a small subset of inequalities (residence, sex and wealth).

In 2015, leaders from every country in the world agreed to 17 goals to create a better future for all people, everywhere, by 2030. The Sustainable Development Goals promised that no one will be left behind by progress. Here we explain how we determine countries’ progress across different measures. 

Measuring progress towards the SDGs

Beyond the immediate loss of life and injuries, conflicts and humanitarian crises pose many challenges to children: they make finding work harder for parents, disrupt education, increase costs and poverty, all of which weakens the protections that should be in place to keep children safe from violence. In turn, these factors increase girls' risk of child marriage. As previous evidence shows, child marriage is often seen as a way to reduce financial pressures on families or to protect girls from other forms of gender-based violence. 

The [Child Atlas illustrates this relationship](https://www.childatlas.org/atlas?indicator1=PT_F_20_24_MRD_U18&topic1=CP.VG&indicator2=CONFLICT_AFFECTED&topic2=CI.CM), showing that many countries with high rates of child marriage are also places where a high share of children are affected by conflict (for instance Mali or Central African Republic). But this relationship is not consistent throughout: there are many conflict situations where child marriage is not as high as in more peaceful settings, and vice versa. And while the correlation shown in the Child Atlas can be indicative of a causal relationship – conflict increasing child marriage – we need more elaborate analytical tools to study this in detail. 

![child marriage_conflict.png](https://res.cloudinary.com/dmhlewc3v/image/upload/v1720712579/child_marriage_conflict_4a2c6b394f.png)

This was exactly what we did for the [Global Girlhood Report 2022]( https://resourcecentre.savethechildren.net/document/global-girlhood-report-2022-girls-on-the-frontline/) Most studies analysing the impact of conflict on child marriage focus on one particular area and use qualitative techniques, providing invaluable insights into the drivers of child marriage in humanitarian contexts. However, we were interested in quantifying this relationship across a large number of countries to understand how much child marriage might increase as a result of conflict. After almost two years, our analysis has just been published in the [academic journal SN Social Sciences](https://link.springer.com/article/10.1007/s43545-024-00940-7).

In this analysis, we first combined data across 56 countries, covering 3.5 million women and 2.1 million marriages. We then linked the household locations of these women to a geographically granular dataset covering conflict events over the last 30 years that shows whether any armed conflicts had taken place during that time and how many people had been killed. Bringing these two sets of information together, to see if child marriages were more prevalent during times of conflict.  

Our results showed that girls were 20% more likely to get married if a conflict took place in their area in previous five years. However, this finding hides a lot of variation, with the impacts varying between world regions and countries. You can read the whole study [here](https://rdcu.be/dNayG). 

large_CH1545104_Samira, aged 14, at school in Yatenga province, Burkina Faso..jpg

medium_CH1545104_Samira, aged 14, at school in Yatenga province, Burkina Faso..jpg

small_CH1545104_Samira, aged 14, at school in Yatenga province, Burkina Faso..jpg

thumbnail_CH1545104_Samira, aged 14, at school in Yatenga province, Burkina Faso..jpg

CH1545104_Samira, aged 14, at school in Yatenga province, Burkina Faso..jpg

Conflicts and humanitarian crises significantly heighten the risk of child marriage. Our research shows that girls in conflict zones are 20% more likely to marry, though the impact varies by region and country.

Girls on the Frontline: Are girls more likely to marry in conflict? 

_This blog was first published on the new [Child Benefit Tracker](https://www.childbenefitstracker.org/), a joint partnership by Save the Children, UNICEF and ILO to establish a one-stop-shop to track countries' pathways to expanding child benefits._

## How are Climate Risks jeopardizing the SDG of ending extreme child poverty?
The climate crisis is not a distant risk, but is already threatening children’s rights and their well-being here and now. Estimates suggest that [4 out of 5 children face at least one extreme climate event](https://resourcecentre.savethechildren.net/document/generation-hope-2-4-billion-reasons-to-end-the-global-climate-and-inequality-crisis/) – floods, cyclones, droughts, wildfires etc. – every year. These climate risks threaten the global goal of ending extreme child poverty. They push people further into poverty or prevent others from escaping it, undermining current efforts to alleviate poverty.
A [new report by the Global Coalition to End Child Poverty](https://static1.squarespace.com/static/56588879e4b0060cdb607883/t/656d8b2b0d371a5b7f1d6057/1701677873038/Children+in+poverty+-+Climate+crisis.pdf) looks at the nexus between climate change and child poverty. Children living in poverty are often more likely to suffer harm, as they for instance live in lower-quality housing, experiencing already food insecurity or other health issues, and may have inadequate access to information. And once affected by a cliamte disaster, children in poverty have often fewer capacities to cope with and adapt to the effects, as their families might be already monetarily poor, have more vulnerable livelihoods or limited access to social protection. A new analysis for this report illustrated the impact of extreme weather events on multidimensional child poverty, focusing on floods in Nigeria, Uganda and Malawi: children living inside a 5km buffer zone of flooded areas were on average 48 per cent more likely to experience multidimensional poverty compared to children living outside of this buffer zone, even after adjusting for other possible determinants of child poverty. As a result, climate change and child poverty reinforce each other: not only are children living in poverty more vulnerable to the effects of disaster or environmental stress, but these effects also increase their likelihood of living in poverty – establishing a vicious cycle.

## Are children living in poverty more exposed to climate shocks?
Children in low- and middle-income countries often bear the brunt of climate change and its impact – despite contributing least to it. A [study by the World Bank](https://blogs.worldbank.org/opendata/illustrating-gap-between-global-extreme-poor-and-global-emissions) showed that countries home to half of the population living in extreme poverty contribute only 3 per cent of global emissions. Our analysis in this report went even deeper and combined for the first-time detailed child poverty data on a subnational level with information on climate risks for 83 low- and middle-income countries. We find that children living in poverty are particularly exposed to the impacts of extreme weather events: 3 out of 10 children live in provinces with above-average rates of child poverty and climate risks.

<iframe src='https://flo.uri.sh/visualisation/16721285/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe>

However, we also found that the link between subnational level of child poverty and exposure to climate risks is complex: while in some world regions poorer provinces are more exposed to some climate risks, for other risks or world regions we find the opposite relationship or no link at all. This reiterates the importance to conduct national and local analyses on the link between child poverty and climate hazards.

## What policies and programmes can mitigate the impacts of climate shocks on vulnerable children, in particular?
Luckily, we already know what is needed to end child poverty. Contexts may vary significantly, but a [core policy agenda](https://static1.squarespace.com/static/56588879e4b0060cdb607883/t/634c6198f77b8b342de40f8e/1665950105327/Ending+child+poverty_a+policy+agenda.pdf) to address the key issues involves the following: 1) Making child poverty reduction a national priority; 2) Expanding child-sensitive social protection; 3) Improving access to and prioritising funding of quality public services; and 4) Promoting a decent work and inclusive growth agenda. Child-sensitive social protection programmes are particularly important to strengthen households’ resilience and coping strategies in response to climate shocks, stresses and hazards. It is particularly important to invest in strong routine social protection systems, including child benefits, and enhance their shock responsiveness. Unfortunately, statutory child benefit schemes are often lowest in the countries at the forefront of the climate crisis: across countries [considered by UNICEF to face extremely high climate risk](https://www.unicef.org/reports/climate-crisis-child-rights-crisis) only 10% of children are covered with child benefits. In contrast, countries facing somewhat lower climate risk have on average coverage rates of 39%, almost four times higher.
To withstand climate shocks, countries need to increase preparedness and build resilience - investment in social protection policy frameworks and operational systems that allow for rapid scale up to reach populations, especially children living in poverty, are essential for both!

_Read the [original blog on the Child Benefit Tracker](https://www.childbenefitstracker.org/) and explore data on child benefits around the world in our new data tool._

Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia

large_CH11009118_Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia.jpg

medium_CH11009118_Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia.jpg

small_CH11009118_Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia.jpg

thumbnail_CH11009118_Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia.jpg

CH11009118_Inonge (38) and her son Lawrence (7) inspect their failed maize crop in drought-hit Zambia.jpg

Children in poverty and their families contribute the least to the climate crisis, yet they bear the brunt of it. Using data on multidimensional child poverty and climate risks from the Child Atlas, we investigated this intersection in a recent report by the Global Coalition to End Child Poverty. Our analysis finds that children living in poverty are particularly vulnerable to the impacts of extreme weather events.

Child poverty and the Climate Crisis: A Disproportionate Burden

_On 14 February, Save the Children, UNICEF and ILO launched the [Global Child Benefits Tracker](https://www.childbenefitstracker.org/), a joint project to establish a one-stop-shop to track countries' pathways to expanding child benefits._

Social protection is a universal right and one of the most effective tools in reducing child poverty. It helps children and families to overcome barriers to accessing adequate nutrition, health, education and protection required to ensure children reach their full potential. But new data shows that just 28.1% of children aged 0-15 years globally have access to child or family benefit. 

The new data released by Save the Children, UNICEF and ILO shows a modest global increase in access to child benefits over a period of 14 years, from 20% in 2009 to 28.1% in 2023. However, progress has been unequal: in low-income countries, rates of coverage remain staggeringly low, at around 9%. At the same time, 84.6% of children in high-income countries are covered, almost ten times as many as in low-income countries. 


<iframe src='https://flo.uri.sh/visualisation/16728365/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe> 



Effective coverage with child benefits also varies significantly by world regions: we find the highest coverage of children in Western Europe, followed by North America and Eastern Europe and Central Asia. In contrast, children are least likely to be covered in West and Central Africa, Eastern and Southern Africa, or in East Asia and Pacific.

Positively, child benefits are now more prevalent in all world regions than 14 years ago. And encouragingly, rates have increased particularly strongly in some of the regions least covered. For instance, in West and Central Africa rates have almost quadrupled from 3.1% in 2009 to 11.8% in 2023. Similarly in South Asia, coverage rates increased significantly from 9.2% in 2009 to 24.3 in 2023.

<iframe src='https://flo.uri.sh/visualisation/16700503/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe>

Those regional and global figures hide some important differences between countries. In almost all low-income countries, less than 1 in 5 children are covered by child or family benefits. The exception are Malawi and Togo, the latter increased their coverage from 49% in 2019 to 74.5% in 2022. 

Mongolia is unique in being a lower middle-income country with  universal coverage of all children. In that income group, Bangladesh had some of the largest increases in recent years, raising coverage from 29.4% in 2020 to 44.5% in 2021.
And while on average countries in higher income groups have more children covered, there are surprisingly many children not receiving child or family benefits.

<iframe src='https://flo.uri.sh/visualisation/16700717/embed' title='Interactive or visual content' class='flourish-embed-iframe' frameborder='0' scrolling='no' style='width:100%;height:600px;' sandbox='allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation'></iframe>

Child benefits are a critical form of social protection, intended to promote the long-term wellbeing of girls and boys, including children with disabilities. Paid in cash or tax credits, child benefits are essential for reducing poverty and building resilience. This support is particularly important in a world with overlapping crises and insecurity. Unfortunately, countries most at risk are often those that are least covered. The availability of statutory child benefit schemes are often lowest in the countries at the forefront of the climate crisis: for instance, across countries [considered by UNICEF to face extremely high climate risk](https://www.unicef.org/reports/climate-crisis-child-rights-crisis) only 10% of children are covered with child benefits. In contrast, countries facing somewhat lower climate risk have on average coverage rates of 39%, almost four times higher.

Save the Children, ILO and UNICEF have developed the [Global Child Benefits Tracker](https://www.childbenefitstracker.org/) to bolster efforts to monitor and reduce exactly those gaps in child benefit coverage. And while those headline figures are also included in the Child Atlas, the Tracker much more detailed information and background on child-sensitive social protection programmes. To learn more about the important role child benefits play in helping children to fulfil their potential and additional resources, please visit [www.childbenefitstracker.org](https://www.childbenefitstracker.org/).

Children at their home in Balkh Province, Afghanistan

large_CH1699199_Abdul-Ahad, 50, with his children at their home in Balkh Province, Afghanistan.jpg

medium_CH1699199_Abdul-Ahad, 50, with his children at their home in Balkh Province, Afghanistan.jpg

small_CH1699199_Abdul-Ahad, 50, with his children at their home in Balkh Province, Afghanistan.jpg

thumbnail_CH1699199_Abdul-Ahad, 50, with his children at their home in Balkh Province, Afghanistan.jpg

CH1699199_Abdul-Ahad, 50, with his children at their home in Balkh Province, Afghanistan.jpg

Social protection is a universal right and one of the most effective tools in reducing child poverty. It helps children and families to overcome barriers to accessing adequate nutrition, health, education and protection required to ensure children reach their full potential. But new data shows that just 28.1% of children aged 0-15 years globally have access to child or family benefit. 

1.4 billion children missing out on basic social protection

1 billion children live in multidimensional poverty. This means that more than 4 in 10 children are severely deprived of rights that constitute poverty, for example because they don't go to school, don't have access to health care, are malnourished, live in overcrowded conditions or lack basic clean water and sanitation. [Our previous work](https://content.iospress.com/articles/statistical-journal-of-the-iaos/sji220060) suggested that the COVID-19 pandemic and its socio-economic impact may have resulted in an additional 100 million children living in poverty.

# Measuring child poverty via national household surveys
How do we know this? Our estimates are based mainly on household surveys such as [Demographic and Health Surveys (DHS)](https://dhsprogram.com/) and [Multi-Indicator Cluster Surveys (MICS)](https://mics.unicef.org/). These surveys are conducted every few years in most low- and middle-income countries and ask thousands of households about living conditions, access to education or health care, and many other things. However, these questions cannot be asked of every family in a country, but only of a small sample that is representative of the whole population. As a result, we can use these surveys to estimate the prevalence and depth of child poverty (and many other things) both nationally and down to a sub-national level (in most countries this is states or provinces). Those surveys are extremely valuable to understand children’s lives in different contexts and the source for many indicators in the Child Atlas.

However, these data leave us in the dark about where exactly poor children live, as we cannot zoom in further than these sub-national levels. But if we don't know which districts or villages are particularly affected, how can local governments, civil society and NGOs know that their programmes and activities are reaching those most affected by inequality and discrimination? It also limits our understanding of the links between child poverty and climate change or conflict, as we often simply do not know whether children living in poverty are also living in a conflict zone, or are particularly vulnerable to natural disasters or other environmental risks. We therefore need a much more detailed understanding of the distribution of child poverty.

# Mapping poverty locally with machine learning models
Poverty maps and small area estimates are one way of addressing this challenge. The idea is to combine household survey data with satellite imagery (e.g. night lights), information on public and private infrastructure (road density, location of critical infrastructure, travel time to hospitals, mobile phone towers and connection speed, etc.) and other possibly relevant variables (altitude, vegetation, pollution, etc.). A machine learning model then takes much of the original household survey information (from those surveys where GPS information gives us the approximate location of the household) and tries to identify a relationship between these households and a wide range of possible influencing factors. For example, do poorer children tend to live in areas with less light, have longer journeys to health facilities, or live in areas with less road density? The algorithm then uses these insights to predict poverty at a very granular level. These predictions are then tested against 'ground truth' (often another dataset or an unused part of the original household survey data) to understand how well it predicts actual poverty levels.

Such poverty maps are not new, and small area estimates of monetary poverty or wealth exist in many countries. However, this has not been done for multidimensional child poverty for a large number of countries. This leaves us guessing where the poorest children live.

# Limitations and potential of poverty maps
Save the Children has been working with fellows from the University of Warwick's Data Science for Social Good initiative and UNICEF to fill this data gap. A newly developed and innovative machine learning model estimates child poverty across sub-Saharan Africa, following the idea outlined above. The end result is a granular map (in our case hexagons with an average area of 5.16 km²) of the prevalence and depth of child poverty across the region. You can read our [methodological summary here](https://res.cloudinary.com/dmhlewc3v/image/upload/v1697544091/Microestimates_of_Child_Poverty_in_Sub_Saharan_Africa_b52c2b529a.pdf?updated_at=2023-10-17T12:01:32.380Z), and the full model is open source for anyone to use and improve. 

Any such analysis can only be an estimate. This is very much the case with our model, and the quality of our predictions varies somewhat from country to country (especially as we are trying to estimate across a continent and not just a single country). We add prediction intervals to the dataset to illustrate this uncertainty. But we also need to be careful how we use such estimates and make sure we don't jump to conclusions based on very small estimates that may have large confidence intervals attached to them. 

But this should not detract from the value of these estimates: for the first time we have a detailed understanding of how much child poverty varies within countries in sub-Saharan Africa, and where exactly we would expect to find the poorest children. And it shows what is possible when using a machine learning model and a wealth of (publicly available) data – something we hope many countries will follow to produce their own small area estimates that can then be used to inform policy and programming decisions.

# The path forward: refining poverty maps and linking them to policy
This is only the beginning of the journey. While the level of poverty may change over time, the spatial distribution of where the poor live can be quite persistent. This means that such poverty models can indeed have a relatively long shelf life. However, we would still like to better understand how the distribution changes over time, as this helps us to analyse and track child poverty in times of crisis, such as climate shocks, food inflation or conflict. We are also working to better understand the links between child poverty and climate risks, where this type of data can provide valuable insights – more on that soon!

_Update August 2024: Unfortunately, we are no longer able to show the interactive graph with microestimates of child poverty. If you are still interested in the data, please reach directly out to us!_

large_CH169655_Panashe Katiza (13).jpg

medium_CH169655_Panashe Katiza (13).jpg

small_CH169655_Panashe Katiza (13).jpg

thumbnail_CH169655_Panashe Katiza (13).jpg

CH169655_Panashe Katiza (13).jpg

1 billion children live in multidimensional poverty, but traditional data sources often leave us in the dark about where exactly poor children live. Save the Children has been working with fellows from the University of Warwick's Data Science for Social Good initiative and UNICEF to fill this data gap, producing for the first time micro-estimates of multidimensional child poverty across Sub-Saharan Africa.

Mapping Child Poverty: Using Machine Learning to provide a more Granular Picture of Child Poverty in Sub-Saharan Africa

# Data is essential to deliver children’s rights
With less than 7 years left, the world is way off-track to reach many of the Sustainable Development Goals for children set out in the 2030 Agenda for Sustainable Development. Progress for child rights is being threatened by multiple, overlapping crises with disproportionate impacts on children’s lives. Data to understand this situation is crucial: it helps us to make better decisions, allocate resources effectively and hold policymakers to account for their actions.

# Explore children's different situations
The Child Atlas is Save the Children's new data platform to visualise, compare, analyse and understand children’s outcomes globally. The platform puts high-quality, publicly available data at your fingertips. For the first time, we bring together population-based data on key child development outcomes across multiple thematic areas in one integrated platform. This is complemented by data on risks and crises, public policies and financing decisions, and other contextual information.

- **Explore the different ways inequality affects children’s lives:** The Child Atlas includes disaggregated data where available, emphasising the inequalities in key child development indicators between different groups of children. This data is crucial to support the pledge to Leave No One Behind, allowing us to better understand the intersecting experiences of inequality, and to monitor rates of improvement. In both the Atlas and the country profiles you will find disaggregation by gender, place of residence, wealth, disability and sub-national regions. However, the availability of disaggregated data varies considerably by indicator and data source - you are most likely to find disaggregated data for child outcomes [measured through household surveys](http://www.childatlas.org/blog/count-every-child) (health, nutrition, education, child protection, poverty).

![child-poverty-candleStick.jpg](https://res.cloudinary.com/dmhlewc3v/image/upload/v1694430455/child_poverty_candle_Stick_311277200a.jpg)

- **Understand the relationship between different topics:** Data helps us to draw attention to threats to progress, describe emergent risks and crises, and highlight policies and financing decisions that help to advance children’s rights. Importantly, the Child Atlas helps us to better understand how risks, crises and public policies overlap with children’s outcomes across health, education, protection, and broader well-being. For example, is under-five mortality more concentrated in countries with lower coverage of basic health services? (The answer seems to be yes, as our platform shows. Although we can only show correlations, not causality!) Being able to directly compare child outcomes with potential risks (climate, conflict, hunger, etc.) or policy choices (e.g. public investment in children) is essential if we want to take a more holistic and systemic perspective on children's rights. You can do this directly in the Child Atlas by selecting a secondary topic.

![childmortality_antenatalcare.jpg](https://res.cloudinary.com/dmhlewc3v/image/upload/v1694430463/childmortality_antenatalcare_ae562afd42.jpg)

# Blogs and AI to make sense of the numbers
Throughout the Child Atlas, we worked hard to present data in an easy-to-use format. A growing number of blogs will tell specific data stories, provide [methodological backgrounds](http://www.childatlas.org/blog/count-every-child), or highlight innovative data projects we are working on. We also use AI-generated explainers throughout the Child Atlas to help make sense of the numbers, which are clearly marketed with a small 'AI' icon. While these help to explain the data in more detail and draw out some key insights, this is an experimental feature and sometimes they are getting things wrong (so please check any such statements).

# Tell us what you think
We are very excited to finally share the Child Atlas with you! We hope that this platform will help you to describe and highlight the situation of children through data - informing strategic decisions, supporting policy or programme interventions and providing key evidence to hold decision-makers accountable. 
We see great potential for the Child Atlas and are already working on improvements. Please get in touch to tell us what you think, how this platform helps you in your work and what you would like to see included in the future. You can write to us directly at [childatlas@savethechildren.org.uk](mailto:childatlas@savethechildren.org.uk). 

large_CH1344346_Mateo, 10, lives with his family in a resettlement camp after losing his home to floods.jpg

medium_CH1344346_Mateo, 10, lives with his family in a resettlement camp after losing his home to floods.jpg

small_CH1344346_Mateo, 10, lives with his family in a resettlement camp after losing his home to floods.jpg

thumbnail_CH1344346_Mateo, 10, lives with his family in a resettlement camp after losing his home to floods.jpg

CH1344346_Mateo, 10, lives with his family in a resettlement camp after losing his home to floods.jpg

The Child Atlas is Save the Children's new data platform to visualise, compare, analyse and understand children’s outcomes globally. The platform puts high-quality, publicly available data at your fingertips. 