As the year draws to an end, it's time for a re-run the next instalment of our series 12 things we learned in 2012 and the ever popular and long-awaited 13 things we learned in 2013. So here are 14 things from 2014! The need for brevity means I've had to simplify loads of information into a few lines - so don't take my word for it all, click the links to find out more.
#1 Children have high hopes and expectations to study for longer to get a good job and future. At the age of 12 years between three-quarters of children in Ethiopia and nine in ten in Peru wanted to go on to vocational training or university after finishing school. Such high hopes often bump up against sharp realities later on, especially for the most marginalised children. But it makes little sense to interpret the reasons for children leaving school early as a lack of understanding of the importance of a good education.
#2 Children are very able to explain many of the pressures on them and their communities which reduce their chances in life. 2014 saw the 25th anniversary of the UN Convention on the Rights of the Child (see this great, upbeat report from UNICEF). Our series of video interviews with children in Ethiopia show what holds them back: lack of food, the illness of parents, inability to afford school costs which makes children feel ashamed: poverty a connecting and contributing factor in all cases.
#3 Parents care very much about the quality of schooling their children receive. Parents are a motivated, important resource to help drive up education standards, but as this analysis from Andhra Pradesh shows, the poorest people may also be those least able to enforce accountability. The report rings an alarm bell: Poor parents recognise that they lack social power when it comes to complaining about teachers and schools! There's a policy opportunity here - in arming parents to help them get better opportunities for their children (the data revolution has potential to contribute).
#4 National differences in cognitive test scores are already in place before children start school. Our analysis compared cognitive test results at age 5 and age 8 (the Younger Cohort), and age 12 and age 15 (the Older Cohort) across all four study countries. Two things are instructive â first that national (average) differences were in place at age 5 before children start school (i.e. Vietnamese children outperform the other countries, and children in Ethiopia score the lowest). Second, at primary school age, the Vietnamese children made about twice as much progress on cognitive tests per year as in Peru and Andhra Pradesh/Telangana. The global learning crisis requires national answers.
#5 Child stunting reduced among 12-year-olds between 2006 and 2013, but in Andhra Pradesh the social groups with the highest initial stunting rate saw the least gain. The average reduction between 2006 and 20013 was 4 percentage points, and while there was reduction of 6 percentage points for the least poor children there was almost no change for the poorest. Given the impacts of stunting on child survival, he
#6 Background disadvantage in Peru is associated with children having fewer later opportunities to learn. Researchers in Peru found that socio-economic background at age 1 year old predicated childrenâs opportunities to learn (e.g. how likely it was that they would have their work books marked) at age 10. Here school is reinforcing existing inequalities through unequal provision. As the authors say, the analysis depicts: âa highly unequal system of education in which from an early age the quality of instruction 10 years later can be predicted.â The policy message: schools can widen gaps - we need to improve opportunities to learn for the poorest children.
#7 In Vietnam primary schools are encouragingly equitable in the 'value-added' they provide to learning, but differences in household background that affect opportunities to learn still count against poor children (for example in access to extra classes or learning resources beyond school). Taken together, points 6 and 7 suggest that helping poor children learn better requires addressing inequities within the school system, and targeting the underlying reasons (i.e. poverty) why poor children are less likely to able to benefit from school.
#8 Most of the private school test score 'premium' in India, disappears when we take household background into account. Previously weâve highlighted the rise of low-fee private schools in India, and that this growth is stratifying schools for disadvantaged girls and the poorest children. Childrenâs average test scores are typically higher in private schools than in government schools but aanalysis of childrenâs household circumstances shows that most of the test score difference is down to family background. Private schools tend to do better on average principally because they de facto exclude the poorest children, not necessarily because they contribute more to learning. Tellingly, however, the same analysis shows that private schools are much cheaper to run, and highlights weaknesses in accountability in government schools.
#9 The school environment is often hostile, often at the hands of teachers. Corporal punishment is banned in India,
#10 What would support learning, from outside the classroom? More positive evidence for social protection. Our evidence from the huge MGNREGA programme which provides employment in rural areas in India shows how the programme was associated with positive effects on childrenâs grade progression and cognitive tests. Interestingly, the authors note these effect sizes were similar to those commonly found for other conditional cash transfers (i.e. conditionality is not the only way to ensure children benefit). The ILO have described social protection for children as 'a right unfulfilled'. It would be a huge step forward - for child he
#11 And now for one that isn't about school; how children and families experience urban relocation. Ethiopia, in common with many African countries, is urbanising rapidly. The consequence is redevelopment and the displacement of urban communities from city centres to the outskirts. Many Young Lives households are experiencing these relocations. A study of their experiences notes both the good and the bad: prospects and expectations of better infrastructure (kitchens, toilets, more space and privacy) run hand-in-hand with fears of losing livelihood opportunities and the break-up of tightly knit communities . As a Brit this strikes a particular cord: 1950s slum clearances were supposed to generate âcities in the skyâ but often led to a concentration of disadvantage on the periphery.
#12 And one important and, perhaps, surprising fact: our latest data show that young women were more likely than young men to be enrolled in education at age 19 in both Ethiopia and in Vietnam. Gender-based analysis needs to consider both girls and boys. This analysis also shows that gaps often widen when gender is combined with poverty - so instead of thinking only about gender, it's more instructive to think about both gender linked to social background.
#13 Context, community and social environment, is critical to young people's well-being as they grow up. In 2014 considerable interest has been given to the position of girls in particular (. Eliminating harmful practices requires more than criminalisation but addressing the root causes, some of which have do to with economic vulnerability, and providing education and employment opportunities for girls. With that in mind it is worth noting across all of our study countries, the young women most likely to have had a baby by age 19 were poorer girls in rural areas.
#14 Finally: What's the future for birth cohort studies in low- and middle-income countries? Within SDG debates has come a loud call for a 'data revolution' to support the achievement of development goals . Better monitoring and more disaggregated data are essential to this endeavour, but it shouldnât stop there. A data revolution also needs to provide better policy advice, which implies not only measuring social problems but analysing their causes and consequences. How can cohort studies further that agenda? Key conclusions of a symposium hosted by UNICEF Office of Research were that whatâs needed are better networks among studies and better alignment of questionnaires and instruments to increase the collective âpowerâ of what is known about what matters for children.
Reading back over what I have written, I realise it's mostly about children's schooling, or what can support learning. Primary enrolment rates have increased sharply and while some children remain out of school (typically the poorest and disabled children), most children spend a huge amount of time at school (click here and look at the time-use figures on page 10). And for those interested in ways to deliver policy for children, increased enrolment presents a 'delivery platform' opportunity: school is a huge part of many children's lives, with its impact going far beyond learning.
As the year draws to an end, it's time for a re-run the next instalment of our series 12 things we learned in 2012 and the ever popular and long-awaited 13 things we learned in 2013. So here are 14 things from 2014! The need for brevity means I've had to simplify loads of information into a few lines - so don't take my word for it all, click the links to find out more.
#1 Children have high hopes and expectations to study for longer to get a good job and future. At the age of 12 years between three-quarters of children in Ethiopia and nine in ten in Peru wanted to go on to vocational training or university after finishing school. Such high hopes often bump up against sharp realities later on, especially for the most marginalised children. But it makes little sense to interpret the reasons for children leaving school early as a lack of understanding of the importance of a good education.
#2 Children are very able to explain many of the pressures on them and their communities which reduce their chances in life. 2014 saw the 25th anniversary of the UN Convention on the Rights of the Child (see this great, upbeat report from UNICEF). Our series of video interviews with children in Ethiopia show what holds them back: lack of food, the illness of parents, inability to afford school costs which makes children feel ashamed: poverty a connecting and contributing factor in all cases.
#3 Parents care very much about the quality of schooling their children receive. Parents are a motivated, important resource to help drive up education standards, but as this analysis from Andhra Pradesh shows, the poorest people may also be those least able to enforce accountability. The report rings an alarm bell: Poor parents recognise that they lack social power when it comes to complaining about teachers and schools! There's a policy opportunity here - in arming parents to help them get better opportunities for their children (the data revolution has potential to contribute).
#4 National differences in cognitive test scores are already in place before children start school. Our analysis compared cognitive test results at age 5 and age 8 (the Younger Cohort), and age 12 and age 15 (the Older Cohort) across all four study countries. Two things are instructive â first that national (average) differences were in place at age 5 before children start school (i.e. Vietnamese children outperform the other countries, and children in Ethiopia score the lowest). Second, at primary school age, the Vietnamese children made about twice as much progress on cognitive tests per year as in Peru and Andhra Pradesh/Telangana. The global learning crisis requires national answers.
#5 Child stunting reduced among 12-year-olds between 2006 and 2013, but in Andhra Pradesh the social groups with the highest initial stunting rate saw the least gain. The average reduction between 2006 and 20013 was 4 percentage points, and while there was reduction of 6 percentage points for the least poor children there was almost no change for the poorest. Given the impacts of stunting on child survival, he
#6 Background disadvantage in Peru is associated with children having fewer later opportunities to learn. Researchers in Peru found that socio-economic background at age 1 year old predicated childrenâs opportunities to learn (e.g. how likely it was that they would have their work books marked) at age 10. Here school is reinforcing existing inequalities through unequal provision. As the authors say, the analysis depicts: âa highly unequal system of education in which from an early age the quality of instruction 10 years later can be predicted.â The policy message: schools can widen gaps - we need to improve opportunities to learn for the poorest children.
#7 In Vietnam primary schools are encouragingly equitable in the 'value-added' they provide to learning, but differences in household background that affect opportunities to learn still count against poor children (for example in access to extra classes or learning resources beyond school). Taken together, points 6 and 7 suggest that helping poor children learn better requires addressing inequities within the school system, and targeting the underlying reasons (i.e. poverty) why poor children are less likely to able to benefit from school.
#8 Most of the private school test score 'premium' in India, disappears when we take household background into account. Previously weâve highlighted the rise of low-fee private schools in India, and that this growth is stratifying schools for disadvantaged girls and the poorest children. Childrenâs average test scores are typically higher in private schools than in government schools but aanalysis of childrenâs household circumstances shows that most of the test score difference is down to family background. Private schools tend to do better on average principally because they de facto exclude the poorest children, not necessarily because they contribute more to learning. Tellingly, however, the same analysis shows that private schools are much cheaper to run, and highlights weaknesses in accountability in government schools.
#9 The school environment is often hostile, often at the hands of teachers. Corporal punishment is banned in India,
#10 What would support learning, from outside the classroom? More positive evidence for social protection. Our evidence from the huge MGNREGA programme which provides employment in rural areas in India shows how the programme was associated with positive effects on childrenâs grade progression and cognitive tests. Interestingly, the authors note these effect sizes were similar to those commonly found for other conditional cash transfers (i.e. conditionality is not the only way to ensure children benefit). The ILO have described social protection for children as 'a right unfulfilled'. It would be a huge step forward - for child he
#11 And now for one that isn't about school; how children and families experience urban relocation. Ethiopia, in common with many African countries, is urbanising rapidly. The consequence is redevelopment and the displacement of urban communities from city centres to the outskirts. Many Young Lives households are experiencing these relocations. A study of their experiences notes both the good and the bad: prospects and expectations of better infrastructure (kitchens, toilets, more space and privacy) run hand-in-hand with fears of losing livelihood opportunities and the break-up of tightly knit communities . As a Brit this strikes a particular cord: 1950s slum clearances were supposed to generate âcities in the skyâ but often led to a concentration of disadvantage on the periphery.
#12 And one important and, perhaps, surprising fact: our latest data show that young women were more likely than young men to be enrolled in education at age 19 in both Ethiopia and in Vietnam. Gender-based analysis needs to consider both girls and boys. This analysis also shows that gaps often widen when gender is combined with poverty - so instead of thinking only about gender, it's more instructive to think about both gender linked to social background.
#13 Context, community and social environment, is critical to young people's well-being as they grow up. In 2014 considerable interest has been given to the position of girls in particular (. Eliminating harmful practices requires more than criminalisation but addressing the root causes, some of which have do to with economic vulnerability, and providing education and employment opportunities for girls. With that in mind it is worth noting across all of our study countries, the young women most likely to have had a baby by age 19 were poorer girls in rural areas.
#14 Finally: What's the future for birth cohort studies in low- and middle-income countries? Within SDG debates has come a loud call for a 'data revolution' to support the achievement of development goals . Better monitoring and more disaggregated data are essential to this endeavour, but it shouldnât stop there. A data revolution also needs to provide better policy advice, which implies not only measuring social problems but analysing their causes and consequences. How can cohort studies further that agenda? Key conclusions of a symposium hosted by UNICEF Office of Research were that whatâs needed are better networks among studies and better alignment of questionnaires and instruments to increase the collective âpowerâ of what is known about what matters for children.
Reading back over what I have written, I realise it's mostly about children's schooling, or what can support learning. Primary enrolment rates have increased sharply and while some children remain out of school (typically the poorest and disabled children), most children spend a huge amount of time at school (click here and look at the time-use figures on page 10). And for those interested in ways to deliver policy for children, increased enrolment presents a 'delivery platform' opportunity: school is a huge part of many children's lives, with its impact going far beyond learning.