Chapter 3. Measuring deficits in youth well-being (Module 1)

This module provides guidelines to identify and quantify the main challenges facing youth in terms of well-being. It describes the five selected dimensions to analyse youth well-being (health, education and skills, employment, civic participation and empowerment, and some elements associated with overall life evaluation, feelings and meaning), and proposes a number of quantitative indicators that could be used to measure and analyse the extent of the deficits affecting youth in each dimension. The final part of this module introduces a method to calculate the number of youth suffering from multiple deprivations: the Youth Multi-dimensional Deprivation Indicator (Y-MDI).

  

Several frameworks have been proposed to measure the different aspects of youth well-being. Among them, it is worth mentioning the UN World Programme of Action for Youth, with its 15 priority areas and recommended indicators (UN DESA, 2010); the Global Youth Wellbeing Index, developed by the International Youth Foundation (www.youthindex.org); the Youth Development Index of the Commonwealth Youth Programme (youthdevelopmentindex.org); and the World Bank toolkit to support youth at risk in middle-income countries (World Bank, 2008). The Organisation for Economic Co-operation and Development (OECD) has long promoted the measurement of well-being and has embedded the notion at the core of policy making.

Following the OECD well-being adjusted framework (Boarini, Kolev and McGregor, 2014), the present toolkit considers youth well-being in terms of both objective and subjective measures (see Box 3.1). It further focuses on five dimensions of well-being that are deemed to be especially relevant for youth and the various transitions occurring in this particular period of life: health, education and skills, employment, participation and empowerment, and some elements associated with overall life evaluation, feelings and meaning (Table 3.1). See Annex 3.A1 for the full list of proposed indicators, their definitions and sources of data.

Box 3.1. The OECD well-being framework and the adjusted framework for developing countries

The OECD well-being framework identifies three pillars for understanding and measuring people’s well-being. First, it emphasises that people’s perceived experiences of their life (quality of life) is as important as objective measures (material conditions). Second, it focuses on well-being outcomes as opposed to well-being drivers (e.g. health conditions and not the number of surgical interventions). Third, it looks at inequalities across population groups and across the full range of well-being dimensions, rather than limiting to material conditions (i.e. poverty). The framework considers 11 dimensions of well-being and four capitals to ensure sustainability. The adjusted framework adapts the well-being measurement indicators to the context of developing countries and considers the capitals as a dynamic “systems” of collective goods and resources, emphasising the interdependencies across these various types of resources.

Figure 3.1. The OECD well-being framework
picture

Sources: OECD (2011), How’s Life? Measuring Well-being; Boarini, Kolev and McGregor (2014), Measuring well-being and progress in countries at different stages of development: Towards a more universal conceptual framework.

Table 3.1. Youth well-being dimensions

Health

Education and skills

Employment

Civic participation and empowerment

Life evaluation, feelings and meaning

  • Mortality and morbidity

  • Sexual and reproductive health (SRH)

  • Substance abuse

  • Personal health satisfaction

  • Participation and progression

  • Completion and attainment

  • Learning achievement

  • Education satisfaction

  • Social and emotional skills

  • Access

  • Quality

  • Job satisfaction

  • Social capital

  • Civic and political engagement

  • Crime and violence

  • Life evaluation, feelings and meaning

Health

Adolescence is a stage full of opportunities to adopt a healthy lifestyle, and youth are often portrayed as the very emblem of health. Yet, it is also a time of exposure to certain risky behaviours in terms of sexual relations and substance abuse. Basic youth health indicators usually include mortality and morbidity, SRH and substance abuse. These provide an objective measure of the health situation and are best complemented by subjective measures of personal health satisfaction (Table 3.2).

Table 3.2. Youth well-being indicators: health

Mortality and morbidity

SRH

Substance abuse

Subjective indicator

  • Mortality rate

  • Disability-adjusted life years (DALYs)

  • Adolescent birth rate

  • Maternal mortality ratio

  • Prevalence of sexually transmitted infections (STIs)

  • Prevalence of illicit drug use

  • Adolescent alcohol consumption

  • Adolescent tobacco consumption

  • Personal health satisfaction

Mortality and morbidity. Mortality followed by morbidity are unquestionably the worst outcomes affecting youth, as well as the rest of the population. There are many causes leading to morbidity and death among youth. Among the most important globally are road injury, communicable diseases (HIV/AIDS, respiratory infections), malnutrition (iron-deficiency anaemia), depression, suicide and violence (WHO, 2014). To measure the extent of morbidity and death among youth, two major indicators from the World Health Organization (WHO) can be used: the youth mortality rate and the youth rate of DALYs.

The youth mortality rate considers all causes of death. The youth mortality rate should therefore be disaggregated at least by major cause groups: i) communicable, maternal, perinatal and nutritional conditions; ii) non-communicable diseases; iii) unintentional injuries; and iv) intentional injuries. The suicide rate (death by self-harm) in particular should be considered to emphasise the importance of mental illness among youth and its potentially devastating effect. The youth mortality indicator can be complemented by the youth DALYs rate, which combines mortality and morbidity data. The latter is a measure of the years of healthy life lost due to ill health, disability or premature death from all causes. It estimates the gap between the current health status and an ideal health status, with the young population living to an advanced age free of disease and disability.

SRH. In adolescence, individuals start to become sexually sensitised and active. Risky behaviours that lead to early initiation and unprotected sex can lead to a variety of adverse outcomes, from STIs to unintended pregnancies that can lead to maternal mortality. The selected indicators on SRH reflect the three major challenges to adolescents and youth. The first two – adolescent birth rate and maternal mortality ratio – are specific to women and are part of the United Nations (UN) Sustainable Development Goals (SDGs) indicators. Globally, maternal conditions come in second among causes of mortality in adolescent girls (WHO, 2014). The third indicator, which concerns both men and women, measures the prevalence of youth STIs, of which HIV is the most acute and pervasive. According to WHO, HIV/AIDS is the second leading cause of mortality among adolescents, and HIV/AIDS-related deaths are estimated to have increased since the new millennium (2014). Besides HIV, the most widespread STIs include syphilis, gonorrhoea, chlamydia and trichomoniasis, which are currently curable, and hepatitis B, herpes simplex virus (HSV or herpes) and human papillomavirus (HPV), which are incurable, as is HIV.

Substance abuse. At an age when individuals are in the process of forming their identity, adolescents are more easily influenced by others and face higher risks for engaging in abuse of substances, such as tobacco, alcohol and illicit drugs, with immediate and lifelong negative consequences. These indicators give an idea of the number of adolescents or youth who have consumed any of these substances in the recent past. Early interventions can prevent addictions and subsequent negative impacts on youth development. It has been demonstrated, for instance, that the use of substances in adolescence is associated with poor cognitive development, mental health disorders, injuries, violence, non-communicable diseases and premature death. Barring death, these adverse effects can extend into adulthood (WHO, 2014).

Box 3.2. Teenage pregnancy in Côte d’Ivoire

Teenage pregnancy is widespread phenomenon in Côte d’Ivoire, which increased drastically from 19.6% in 2005 to 31.9% in 2012. This trend is particularly worrisome as early pregnancy will affect education and future employment prospects for these girls.

Figure 3.2. Adolescent pregnancy rate, by place of residence
picture

Source: OECD Development Centre (2017), Examen du bien-être et des politiques de la jeunesse en Côte d’Ivoire.

Education and skills

Youth education and skills are a dimension crucial for work and life. Objective educational outcomes usually focus on the following areas: i) participation and progression; ii) completion and attainment; and iii) learning achievements. They can be complemented by a subjective measure of general satisfaction with the educational system (Table 3.3).

Table 3.3. Youth well-being indicators: education and skills

Participation and progression

Completion and attainment

Learning achievement

Subjective indicator

  • Enrolment in secondary education

  • Enrolment in tertiary education

  • School life expectancy

  • Drop-out rate

  • Graduation from lower secondary education

  • Educational attainment

  • Mean years of schooling

  • Literacy rate

  • Poor academic performance

  • Education satisfaction

Participation and progression. School enrolment is the main indicator on youth participation in education. This indicator should be computed for different levels of education – e.g. primary, secondary and tertiary – to identify gaps at different stages. Alternatively, indicators on school attendance could be used. The latter are likely to provide a better reflection of education participation, as not all enrolled youth are necessarily attending school. Another participation measure, school life expectancy assesses the number of years youth can expect to spend within different levels of education. One of the worst outcomes youth can experience is to leave school prematurely without completing their education, as identified by school drop-out rates. These should be measured for different levels of education to assess the stages at which school survival is more threatened.

Completion and attainment. Educational attainment usually refers to the highest level of education attained or completed. Like other education indicators, it is based on the International Standard Classification of Education (ISCED) classification, which ranges from no schooling to tertiary education. Completion and attainment give a good indication of the level of formal skills acquired by youth. Completion and attainment can also be measured by the less precise mean years of schooling: the average number of completed years, excluding years spent repeating individual grades. Substantial progress has been achieved worldwide to give children greater access to education, yet school completion is far from universal, especially at the lower secondary level (United Nations Educational, Scientific and Cultural Organization [UNESCO], 2014). To document and monitor this last phenomenon, it is particularly recommended to rely on the gross graduation rate for lower secondary education. Gross graduation rate refers to the total number of graduates (regardless of age) at the specified level of education, divided by the population at the theoretical graduation age from the specified level.

Box 3.3. Dropouts in Cambodia

The combination of high dropout rates and low enrolment in the next level leaves a small pool of young people in school. The Youth Well-being Review of Cambodia estimates that 63% of youth of secondary and tertiary school age have left the education system. In 2015/16, although gross enrolment rates for primary school were 110%, only 80% of these students completed the primary education cycle. Enrolment and completion rates continue to cascade down in lower and upper secondary education. Low-education is a key driver of high level of skills mismatch and poor working conditions for youth.

Figure 3.3. Gross enrolment and completion rates for different levels, 2010/11 and 2015/16
picture

Source: OECD Development Centre (2017), Youth Well-being Review of Cambodia.

Learning achievement. The most basic indicator on learning achievement is the youth literacy rate, which measures the percentage of youth who can both read and write with understanding a short, simple statement on their everyday life. While essential, literacy is only one of the cognitive skills youth need to develop and to succeed in life. Several international surveys have been implemented to assess the proficiency of students in a number of cognitive skills. For instance, OECD Programme for International Student Assessment (PISA) surveys evaluate the performance of youth age 15 in reading, mathematics and science. Scores can be used to determine the percentage of low performers in terms of learning achievement in these different areas.

Employment

The passage from schooling to the labour market may be one of the most critical transitions in a young person’s life. Not all youth have the chance to transition quickly into employment and/or access decent work. The ILO defines decent work as opportunities for work that is productive and delivers a fair income, security in the workplace and social protection for families, better prospects for personal development and social integration, freedom for people to express their concerns, organize and participate in the decisions that affect their lives and equality of opportunity and treatment for all women and men. For some, it takes a long time to secure decent jobs. In the meantime, they may be stuck in untenable situations: unemployed, underemployed or working poor-quality jobs. Others may not have even started the transition and remain inactive, with no future plans to work. A successful school-to-work transition is relatively short and leads to quality jobs. The level of job satisfaction is another important measure of youth well-being in the workplace (Table 3.4).

Table 3.4. Youth well-being indicators: employment

Access

Quality

Subjective indicator

  • Labour market transition stages

  • Not in education, employment or training (NEET) rate

  • Labour market transition length

  • Employment status

  • Informal employment rate

  • Vulnerable employment rate

  • Poorly paid

  • Time-related underemployment rate

  • Qualifications mismatch

  • Job satisfaction

Access to employment. The first recommended indicator on access to employment considers the three stages of transition to the labour market: transition not yet started, in transition, transition completed. The second proposed indicator focuses on a particularly vulnerable group among youth who have not yet successfully transitioned to employment; NEET youth are neither working nor studying. The NEET rate is a good overall indicator on the underutilisation or potential underutilisation of youth labour, especially when combined with the underemployment rate. However, the NEET rate aggregates individuals with very different situations and challenges and needs to be interpreted and analysed with care. Specifically, analysis of NEET youth should look separately at i) unemployed non-students (preferably defined using the relaxed definition of unemployment, which includes discouraged job-seekers in addition to unemployed workers looking for work); and ii) inactive non-students (those neither in the labour force nor in education or training [NLFET] who have either future plans to work or no intention to look for work). The third indicator, labour market transition length, looks at the time it takes to get a job. It is worth noting that a short transition is not necessarily a good thing, especially in low-income countries where the informal sector is large, and youth end up taking the first, potentially poor job they are offered. Long transitions, on the other hand, are more common in developed or emerging economies, where youth unemployment rates are high and transition time, in extreme cases, can last up to three years.

Quality of employment. Job quality is a major concern among youth, especially in developing countries. Several indicators cover the multiple facets of quality of employment, aiming to capture the most recurrent features of poor-quality youth jobs, namely informal employment, vulnerable employment, low earnings, time-related underemployment and skills mismatch (International Labour Organization [ILO], 2013). These poor job attributes are often interrelated. For instance, Shehu and Nilsson show that informality among youth is associated with lower remuneration, job dissatisfaction (the majority of informally employed youth express the desire to change their employment situation), skills mismatch (undereducation), and underemployment, both in terms of working time and income (2014). Informality is a complex and highly heterogeneous phenomenon often difficult to fully capture from existing data sources in developing countries. Alternatively, vulnerable employment can be used as a proxy for informal employment. Vulnerable employment encompasses forms of employment (e.g. own-account and contributing family work) that are, in most cases, characterised by poor working conditions. Indicators on the quality of employment are complemented by a subjective measure, which informs on the share of youth who report being dissatisfied with their jobs.

Civic participation and empowerment

Youth participation and empowerment are important rights for the well-being of youth. Participation looks at a young person’s level of engagement in political and social activities. Empowerment is a broad concept touching on many critical areas of youth inclusion and well-being, including social and human empowerment, economic empowerment, women’s empowerment and political empowerment (MSActionAid-Denmark, 2010). Generally speaking, it can be seen as the full capacity of a young person to exercise his or her rights as a citizen. As put forward in the United Nations Development Programme (UNDP) Youth Strategy 2014-2017, youth empowerment requires guaranteeing their rights to participate in government decision making and processes at the national, sub-national and local levels (UNDP, 2014). Participation and empowerment are, however, only made possible by the existence and enforcement of legal frameworks that protect youth rights to participate in civic and political activities and to freedom of expression. In areas where crime rates are high or social norms induce young people to criminal and risky behaviours, it is unlikely that social capital can be built and young people can engage in civic and political activities. This dimension therefore includes measures of crime and violence.

The present toolkit considers three different areas of well-being related to youth participation and empowerment: i) social capital; ii) civic and political engagement; and iii) crime and violence (Table 3.5).

Table 3.5. Youth well-being indicators: civic participation and empowerment

Social capital

Civic and political engagement

Crime and violence

  • Social network support

  • Trust in institutions

  • Civic engagement

  • Voices among youth

  • Voting rate

  • Rate of formal contact with the police/criminal justice

  • Rate of detainment

  • Homicide rate

  • Rate of victims of assault

  • Gender-based domestic violence

Social capital. Youth who lack social support networks are isolated and often excluded from society and, as such, experience a form of deprivation in well-being. To measure the extent of this deficit, an indicator for the rate of youth who receive social network support is proposed, defined as those who have relatives or friends on whom they can count in case of need.

Civic and political engagement. Civic engagement can take many forms, including volunteering to an organisation, donating money to a charity or helping a stranger in need of help. These three criteria are used to define the rate of youth civically engaged. The indicator on political engagement relates to youth voter turnout. It measures the percentage of youth who declare that they vote when elections take place, be it at the local or national level. An additional indicator can be proposed to account for youth who voice their opinions. It is proxied by the percentage of youth who have recently voiced their opinion to a public official.

Crime and violence. Criminal and violent behaviours can lead to severe adverse consequences for both perpetrators and victims. Two indicators are proposed as regards perpetrators. The rate of juveniles brought into formal contact with the police or the criminal justice system relates to those who are suspected, arrested or cautioned. The rate of juveniles detained refers to those who are held in prisons or penal or correctional institutions. As regards victims, the two indicators selected focus on the worst scenarios: homicide and non-fatal assault resulting in serious bodily injury. An indicator on gender-based domestic violence is further recommended, given its overall significant extent – especially among adolescent and young girls – and the serious and long-lasting damages it can cause victims, from restricted physical integrity and poor health outcomes to increased vulnerability and poverty (OECD, 2014).

Box 3.4. Youth crime and violence in El Salvador

In El Salvador youth are both offenders and victims of crimes. The country is strongly marked by violence and insecurity, due to rivalling gangs. These gangs (or maras) have become criminal structures with a high level of territorial power. After the end of the truce (2012-15) between the two largest gangs and government forces, the level of violence and insecurity started rising again, becoming a majo concern for the population and social cohesion.

The violence in El Salvador manifests itself in homicides, kidnapping, extortion and other forms. Young people are affected morally and physically by the level of violence and insecurity: in 2013 18.4% of adolescents (age 13 to 17) reported being physically attacked in the last 12 months. Young girls are particularly vulnerable to sexual aggression. The young age of most gang members and the often negative portrait of youth in media coverage influences the image of youth in the country, stigmatising those who live in neighbourhoods with high crime rates. This climate makes free movement difficult and limits community activities (sports, arts, public spaces, etc.) which erodes the social fabric.

Table 3.6. Number of young people victim of crime in El Salvador, 2014, per 100 000 minors

Total

Male

Female

Homicide rate, youth

n.a.

208.1

13.8

Homicide rate, all ages

n.a.

120.2

8.7

Rape, age 18-30

1.52

n.a.

2.9

Sexual aggression, age 0-17

15.01

1.5

29.1

Statutory rape

56.4

0.9

114.0

Source: OECD Development Centre (2017), Youth Well-being Review of El Salvador.

Life evaluation, feelings and meaning

Life evaluation, feelings and meaning is an important subjective dimension of well-being. This dimension is too often overlooked in spite of its importance, which is increasingly recognised (Table 3.7). Life evaluation, feelings and meaning matter, first because youth are the best judges of their life situations, and second because how youth evaluate, experience and give sense to their lives has direct implications to their behaviours and subsequent outcomes (Boarini, Kolev and McGregor, 2014).

Table 3.7. Youth well-being indicators: life evaluation, feelings and meaning

Life evaluation

Feelings

Meaning

  • Dissatisfied with life

  • Negative feelings

  • Low sense of purpose in life

Life evaluation. Life evaluation provides important information on the overall personal assessment of youth life quality. A positive life evaluation is typically associated with other positive well-being outcomes, including, for instance, higher social connections and productivity at work (Boarini, Kolev and McGregor, 2014). The indicator selected captures the percentage of youth who are currently dissatisfied with their lives as a whole.

Feelings. Measures of youth’s feelings, be they positive (e.g. joy, contentment, pride) or negative (e.g. fear, fatigue, anxiety), are good complements to the information provided by life satisfaction, since they are less affected by memory-biases and map more directly onto youth’s daily activities. One relevant indicator is the rate of youth with negative feelings, defined as the percentage who experienced at least one of the following feelings during most of the day before the interview: anger, depression, sadness, stress or worry.

Meaning. Having a meaning in life, beliefs or spirituality – sometimes referred to as having a purpose in life – is important to young people and something from which they draw strength to cope with the many challenges they face (Boarini, Kolev and McGregor 2014 2014). The related indicator is the share of youth who feel that their lives do not have an important purpose. A low sense of purpose in life is detrimental for youth well-being, both in the present and in their future lives. It is often associated with social isolation, poor non-cognitive skills and mental disorders, such as depression.

Youth Multi-dimensional Deprivation Indicator (Y-MDI)

The above assessment gives a good picture of the situation of the youth population in a country in each of the five well-being dimensions. However, the analysis is limited to a snapshot per sector. In reality, many young people are affected by multiple dimensions and are at risk for falling into multiple deprivations. Multiple deprivations reinforce each other, making it even more difficult for a young person to overcome his or her deficits and perform in any one dimension. Discourse on measuring the various dimensions of well-being has been prevalent since the 1990s with the UNDP’s Human Development Report (UNDP, 1990). In more recent years, the OECD has launched its Better Life Initiative (OECD, 2011). This framework has also been applied to the developing context by Boarini, Kolev and McGregor (2014), whose work also serves as a comprehensive review of the development of the discourse on multi-dimensional measurement over the years. This section introduces a method to calculate the number of youth suffering from multiple deprivations and presents a number of tools to analyse the multi-dimensionality of youth well-being.

A multi-dimensional indicator complements the dashboard of youth well-being indicators by providing a simpler and more concise way to monitor the overall well-being of youth and measure progress over time (OECD/EU/JRC, 2008; Alkire and Foster, 2007). This can also be useful for policy advocacy (Saltelli, 2007). The indicator allows identification of overlaps in deprivations, which helps to promote synergies among different sectoral policies, bringing a more cohesive approach to youth-related interventions (Boarini, Kolev and McGregor, 2014). Box 3.5 illustrates how a well-designed youth health policy can have positive effects on multiple dimensions of well-being.

Box 3.5. Kenya’s Primary School Deworming Project (PSDP): An example of effective multi-dimensional youth health policy targeting

In many developing countries, school absenteeism presents a sizable challenge. Qualitative surveys and interviews often mention challenges, such as cost, distance and a lack of school uniforms and other necessities. Education policies typically focus on subsidies and other provision strategies to boost school attendance, with mixed success.

Health policies for children focus on a wide range of risk factors. Among them, 600 million school-aged children worldwide are troubled by intestinal worms. While mild infections often go unnoticed, severe infections can cause iron-deficiency anaemia, malnutrition and stunting, all of which typically have lasting negative effects over the life cycle. Oral deworming pills have proven tremendously effective; they kill most types of worms at a price of USD 0.50 per child per year. The key challenge and pressing issue is sustainable, widespread provision, as the pills only prove effective in preventing rapid spread if a sufficient number of children takes them.

Between 1998 and 2001, the non-governmental organisation (NGO) International Child Support ran the PSDP in western Kenya. It had strong effects in the areas of health, education and future employment. It reduced worm infections by 70%. It reduced school absenteeism by 25% and increased girls’ pass rates in Kenya’s national primary school exit programmes by 9.5 percentage points. Finally, men who were dewormed as children worked 3.4 more hours per week, spent more time in entrepreneurial activities and were more likely to work in higher-wage manufacturing jobs than their untreated peers.

Source: J-PAL 2012.

The construction of the Y-MDI involves three steps. Step 1 defines the youth well-being conceptual framework against which youth deprivations will be measured. Step 2 selects the specific indicators within each dimension that will be measured. Step 3 calculates the final headcount in deprivations and overlaps to generate the Y-MDI.

Step 1. Define the youth well-being conceptual framework

The first step in the construction of the Y-MDI is to define the conceptual framework against which youth well-being dimensions will be measured. The capabilities approach describes well-being in terms of freedom to achieve desired ways of living (i.e. capabilities), and well-being may be seen as a set of inter-related capabilities. According to this approach, youth should be given the opportunities to be and do what they have reason to value (Sen, 1998; Deneulin and Shahani, 2009). A complementary framework, called the asset-based approach, views youth as change agents who are capable of creating their own solutions in the absence of public policies. In this approach, a public institution’s main role is to provide an environment where youth are included and able to use their assets (skills, knowledge, initiatives) (McKnight and Kretzmann, 1996). Both approaches argue for the provision of a minimum of capabilities, which unlock youth well-being outcomes.

The above well-being concepts provide the framework which will guide the normative decisions for the construction of the Y-MDI. First, the framework defines which dimensions are needed to achieve the minimum conditions for well-being. It is important that each dimension is clearly distinguished from another and does not overlap in measurement. Second, it differentiates well-being needs or capabilities for youth aged 15-29 using a life cycle approach. For example, youth aged 15-17 need different capabilities than those aged 18-29, the former being school age and the latter likely being in the labour force. Third, it determines what type of measures best capture these capabilities. Indicators can measure both well-being outcomes (e.g. educational attainment) and inputs such as the freedom to go to school (e.g. access to education). Furthermore, indicators can be both objective and subjective measures. The recent discourse on well-being is converging around the view that well-being is both the satisfaction of objective needs and wants, as well as the quality of life which youth experience, i.e. subjective measures (Boarini, Kolev and McGregor, 2014).

The Y-MDI selects four dimensions as the minimum well-being conditions for youth: education, employment, health and civic participation. For comparability, objective output measures are chosen when data are available. Subjective measures are used only when objective measures are not available. The data of all the dimensions need to be available within one dataset so the same youth are measured on all the indicators. This makes Y-MDI a challenging exercise when household surveys only capture a few dimensions.

Step 2. Select the indicators

The second step in the construction of the Y-MDI is to select the indicators to measure well-being within each of the four dimensions. The decision on the number of indicators per dimension is guided by the rule of parsimony (i.e. explaining as much information with the least number of indicators). There are no strict criteria on the number of indicators to include per dimension, but it is important that the selected indicators capture the different ways a youth can be deprived within the dimension. For example, for education, indicators should capture the educational attainment but also the quality of education received. Selecting too few indicators risks missing important well-being aspects, whereas too many will cause overlap among them. The number of indicators per dimension should be more or less the same to ensure a balanced measurement across dimensions.

Each selected indicator needs to be statistically robust, meaning that it has variance, is valid and is internally consistent with the other indicators in the dimension (De Neubourg et al., 2012):

  • Variance means that, within the sample, there needs to be both youth who are deprived and not deprived of that indicator. If nearly all youth score similar values, it is arguably not a discriminating indicator.

  • Validity means that the indicator should capture what it intends to capture (no systematic measurement error). Lack of validity is a common problem with subjective data, where youth tend to under-report unpopular behaviours (e.g. criminal acts) and over-report socially desirable behaviours (e.g. voting when non-compliance is punished). Also, when facing data constraints, using proxies can be a solution; however, they should exclusively capture the concept that they intend to measure. Otherwise, they may bias the results. There is no way to test for validity statistically. Therefore, the selection of an indicator should have solid argumentation.

  • Internal consistency means that the indicators within a dimension are complementary measures of deprivation. This can be checked statistically by cross-tabulating each selected indicator with every other indicator within the same dimension. Summing up the percentage of youth who are deprived in both indicators simultaneously plus those who are not deprived in both indicators captures the overlap between the two indicators. Internally consistent indicators have a sufficient degree of overlap – meaning that they measure the same construct; for example, years of schooling and literacy both capturing education – while at the same time measuring something unique within the dimension.

Determining the threshold where a youth is considered deprived is a difficult exercise due to diverging views on the definition of deprivation. Using internationally-agreed definitions or national legislation is the best way to go. International standards often use a rights-based approach, where minimum values are given to fulfil basic rights. For example, the SDGs state that all boys and girls should complete quality primary and secondary education. This can be operationalised using an indicator on educational attainment that sets a threshold at lower secondary education as the minimum level.

Missing data and non-applicable data are important issues to consider. Missing data occur when survey respondents miss information for no intended reason; non-applicable data are when respondents miss information because the question did not apply to him or her. Both are problematic because lacking values on any one indicator makes it impossible to measure the respondent on all other indicators.

  • In case of missing variables, the respondent cannot be analysed and is therefore excluded from the analysis. If an indicator has more than 5% missing values, the indicator is in danger of being contaminated with bias and therefore should not be used (Schafer, 1999).

  • In case of non-applicable values, it is important to know why respondents are not scored on the indicator. Because non-applicable values are not missing values, these youth are still included in the analysis by counting them as not deprived for the indicators that do not apply. This imputation can undercount deprivation. If severe undercounting is expected, the non-applicability can be taken as a sign of diverging needs among different age groups and serve as a basis to make a life cycle split. Non-applicable indicators should be kept to a minimum.

There are three methods to identify those deprived within a dimension. One is by counting those who are deprived in any one indicator within a dimension (union approach). The other is by counting those deprived in all indicators in a dimension (intersection approach). The third method assigns different weights to each indicator, thus creating a ranking among them. In practice, this last often creates contentions on the various methods of assigning weights (Decancq and Lugo, 2013). For this toolkit, a union approach is preferred, as the indicators are designed to capture a minimum well-being state. Thus, a youth deprived on one indicator is sufficient to be considered deprived on the whole dimension (Alkire and Foster, 2011).

The Y-MDI measures the minimum level of functionings within each dimension. Youth should have a minimum level of quality education (SDG 4.1, 4.4 and 4.6), be able to access high-quality employment once they are in the labour force (SDG 8.6 and 8.8), enjoy healthy lives and have access to SRH (SDG 3.4 and 3.7), and be able to participate as citizens with access to associations and information (SDG 16.6 and 16.10). Table 3.8 shows the basic indicators and thresholds of the Y-MDI.

Table 3.8. Y-MDI basic indicators and thresholds

Dimension

Description

Indicators (15-17)

Indicators (18-29)

Education

Early school leavers* and lower secondary school attainment

Not enrolled and not completed lower secondary education

Not completed lower secondary education

Education quality

Literacy rate (or other national cognitive test results)

Employment

Unemployment and inactivity

Not in employment, education or training

Job quality and child labour

Working more than 48 hours per week

Paid less than 60% of median wage

Working in hazardous jobs with low safety standards

Health

Ability to function in daily physical and mental tasks

Experiencing difficulties in basic physical and mental tasks or having a permanent disability

Access to SRH services

Rate of STIs

Civic participation

Freedom to join an association

Barriers to join associations

Access to information

Access to the Internet

* Any individual not having finished lower secondary school is considered an early school leaver.

Step 3. Identification and sub-analysis

The third and final step in the construction of the Y-MDI is the calculation and analysis. The Y-MDI can answer four broad questions on youth well-being at the national level:

  • What percentage of youth are multi-dimensionally deprived? This is answered by the calculation of the headcount ratio provided by the Y-MDI. This headcount is calculated by summing the number of youth who are multi-dimensionally deprived and dividing this number over the total number of youth within the age group of interest. This provides the percentage of youth within the country that are multi-dimensionally deprived.

  • Of all youth with at least one deprivation, what is the headcount of those with one deprivation, two deprivations, etc.? This is answered by calculating separate headcount ratios for youth with one deprivation, those with two, etc. This gives an impression of the amount of youth with different numbers of deprivations.

  • Which dimensions overlap each other? This is answered by the calculation of deprivation overlaps, which expresses which deprivations youth experience simultaneously. This overlap can be illustrated by combining a maximum of three dimensions in a Venn diagram. The diagram comprises a maximum of seven different groups of youth, consisting of those with no overlap (three groups), those with overlap in two of the dimensions (three groups) and those with overlap in all three dimensions (one group).

  • What is the most prevalent form of deprivation among young people? This is answered through the calculation of headcounts for each deprivation separately and portraying the values in a radar chart (also referred to as spider graph).

The Y-MDI was calculated for Viet Nam using 2015 data from the ILO school-to-work transition survey (SWTS). The indicator shows that 11% of youth aged 18-29 and 10% of youth aged 15-17 were deprived in at least two dimensions (Figures 3.4 and 3.5). Younger youth experienced deprivations mostly in employment, whereas older youth showed the most deficits in education (Figures 3.6 and 3.7). Finally, for the younger youth, employment deprivation strongly overlapped with deficits in education, which could imply that they entered vulnerable employment or became unemployed after dropping out of education (Figure 3.8). Interestingly, for older youth, deprivation in employment was, in most cases, not interconnected with deficits in education, showing that the deprivation did not necessarily stem from a lack of education and vice versa. Rather, those with high educational attainment still experienced deficits in employment (Figure 3.9). This is in line with research findings from the Vietnamese Ministry of Labour, Invalids and Social Affairs (MOLISA) showing that youth with the highest unemployment rates are those with tertiary degrees (MOLISA, 2016).

Figure 3.4. Share of youth deprived in one or more dimensions, among age group 15-17, Viet Nam, 2015 (%)
picture
Figure 3.5. Share of youth deprived in one or more dimensions, among age group 18-29, Viet Nam, 2015 (%)
picture
Figure 3.6. Composition of deprivation among youth (15-17), Viet Nam, 2015
picture
Figure 3.7. Composition of deprivation among youth (18-29), Viet Nam, 2015
picture
Figure 3.8. Overlaps in deprivations among youth aged 15-17, Viet Nam, 2015
picture
Figure 3.9. Overlaps in deprivations among youth aged 18-29, Viet Nam, 2015
picture

Source: OECD Development Centre (2017), Youth Well-being Policy Review of Viet Nam.

ANNEX 3.A1. Youth Well-being Indicators
Table 3.A1.1. Youth health

Main indicators

Definition

International databases

Raw data sources

Prevalence of STIs

Percentage of individuals aged 15-24 reporting a STI in the 12 months preceding the survey among those who ever had sexual intercourse.

Demographic and Health Surveys (DHS) Program – STATcompiler

DHS

Adolescent birth rate

Annual number of births to women aged 15-19 per 1 000 women in that age group.

Global SDG Indicators Database

Civil registration data, household surveys (DHS, Centers for Disease Control and Prevention [CDC]-assisted reproductive health surveys, etc.), population census

Mortality rate, by major cause group

Number of deaths per 100 000 individuals aged 15-29 by main causes, including maternal conditions, mental and behavioural disorders, self-harm (suicide) and violence (interpersonal, collective violence and legal intervention).

WHO – Global Health Estimates (GHE), cause-specific mortality estimates

Vital (death) registration data

Secondary indicators

Definition

International databases

Raw data sources

Adolescent alcohol consumers

Percentage of individuals aged 15-19 who have consumed any alcohol in the past 12 months.

WHO – Youth and alcohol data

National surveys; international surveys: World Health Survey, STEPwise approach to surveillance; Gender, alcohol, and culture: An international survey; European Cancer Anaemia Survey

Adolescent tobacco consumers

Percentage of individuals aged 13-15 who are current users of any tobacco product (those who consumed any smokeless or smoking tobacco product at least once in the past 30 days prior to the survey).

WHO – Global youth tobacco survey (GYTS) data

GYTS

Annual prevalence of illicit drug use by drug type

Percentage of youth (no standardised age range across countries) who used at least once in the past year i) heroin; ii) cocaine; iii) cannabis; iv) amphetamines; or v) ecstasy.

United Nations Office on Drugs and Crime (UNODC) – Drug use statistics

Annual Reports Questionnaire, school surveys, youth risk behaviour surveys, household surveys, government sources

Maternal mortality ratio

Annual number of female deaths from any cause related to or aggravated by pregnancy or its management (excluding accidental or incidental causes) during pregnancy and childbirth or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, per 100 000 live births.

Global SDG Indicators Database

Civil registration data, household surveys, population census, sample or sentinel registration systems, special studies

Rate of DALYs, all causes

Years of healthy life lost due to ill health, disability or premature death from all causes per 100 000 individuals aged 15-29.

WHO – GHE, Global Burden of Disease data

Vital (death) registration data, Global Burden of Disease 2010 study

Personal health satisfaction

Percentage of individuals who are satisfied with their personal health.

Gallup World Poll data

World Poll surveys

Table 3.A1.2. Youth education and skills

Main indicators

Definition

International databases

Raw data sources

Net enrolment rate in secondary education

Number of students enrolled in secondary education in the theoretical age group for secondary education, expressed as a percentage of the official school-age population corresponding to secondary education.

UNESCO Institute for Statistics database

School register, school survey or census for data on enrolment by level of education; population census or estimates for school-age population

Gross graduation ratio from lower secondary education

Number of graduates (regardless of age) from lower secondary education, expressed as a percentage of the population at the theoretical graduation age for that level of education.

UNESCO Institute for Statistics database

School register, school survey or census for data on graduation by level of education; population census or estimates for school-age population

Drop-out rate

Cumulative drop-out rate to the last grade of i) primary education; and ii) lower secondary general education.

UNESCO Institute for Statistics database

National population census, household surveys, labour force surveys

Literacy rate

Percentage of youth aged 15-24 who can both read and write with understanding a short, simple statement on their everyday life (generally, literacy also encompasses “numeracy”: the ability to make simple arithmetic calculations).

UNESCO Institute for Statistics database

National population census, household surveys, labour force surveys

Secondary indicators

Definition

International databases

Raw data sources

Educational attainment

Percentage distribution of youth aged 25-29 according to the highest level of education attained or completed, with reference to International Standard Classification of Education (ISCED) classification (e.g. no schooling, primary, secondary, tertiary)

UNESCO Institute for Statistics database (available for the population over age 25)

DHS, UNICEF Multiple Indicator Cluster surveys, population census, household surveys, labour force surveys

Gross enrolment ratio in tertiary education

Number of students enrolled in tertiary education, regardless of age, expressed as a percentage of the official school-age population corresponding to tertiary education (five-year age group, starting from the official secondary school graduation age).

UNESCO Institute for Statistics database

School register, school survey or census for data on enrolment by level of education; population census or estimates for school-age population

Proficiency in reading, mathematics and science

PISA performance scores in reading, mathematics and science among students age 15.

OECD – PISA surveys

OECD PISA surveys, other international surveys: Trends in International Mathematics and Science Study, Programme on the Analysis of Education Systems, Southern and Eastern Africa Consortium for Monitoring Educational Quality, Latin American Laboratory for Assessment of the Quality of Education

Satisfaction with the educational system

Percentage of youth aged 15-24 who are dissatisfied with the educational system or the schools in the city/area where they live.

Gallup World Poll data

World Poll surveys, World Values Survey (WVS) wave 6: Percentage of youth age 29 and under who think that the formal education system in their country does not provide people with skills/training to i) find employment; ii) perform their jobs well; or iii) start a business

Social and emotional skills: ability to pursue goals, work with others and manage emotions

Abilities to pursue goals (perseverance, self-control, passion for goals), work with others (sociability, respect, caring) and manage emotions (self-esteem, optimism, confidence).

Young Lives, OECD Longitudinal Study of Children’s Social and Emotional Skills in Cities (launch expected after 2020)

School life expectancy

Number of years a person of school entrance age can expect to spend within the specified level of education (secondary and tertiary).

UNESCO Institute for Statistics database

School register, school survey or census for data on enrolment by level of education; population census or estimates for school-age population

Mean years of schooling

Average number of completed years of education among youth aged 15-24, excluding years spent repeating individual grades.

UNESCO Institute for Statistics database (available for the population age 25 and over)

National population census, household and/or labour force surveys

Table 3.A1.3. Youth employment

Main indicators

Definition

International databases

Raw data sources

NEET rate

Share of youth aged 15-29 not in education, employment or training.

NEET includes inactive and unemployed non-students.

ILO/Understanding Children’s Work (UCW) YouthSTATS,

ILO YouthSTATS

School-to-work transition survey (SWTS), labour force survey; population census and/or other household surveys with an appropriate employment module

Employment status

Distribution of workers aged 15-29 by employment status: i) employees; ii) employers; iii) own-account workers; iv) members of producer co-operatives; and v) contributing family workers.

ILO/UCW YouthSTATS,

ILO YouthSTATS

SWTS, labour force survey; population census and/or other household surveys with an appropriate employment module

Informal employment rate

Share of workers aged 15-29 who are in informal employment.

Informal employment: Jobs in the formal sector, informal sector or households, which lack basic social or legal protections or employment benefits. For operational reasons, the concept is measured as the number of persons employed in informal employment in their main job.

ILO/UCW YouthSTATS

SWTS, labour force survey; population census and/or other household surveys with an appropriate employment module

Secondary indicators

Definition

International databases

Raw data sources

Share of inactive non-students

Share of youth aged 15-29 neither in the labour force nor in education or training (different from NEET in that it captures only economically inactive non-students, i.e. those that are not actively seeking jobs).

ILO/UCW YouthSTATS,

ILO YouthSTATS

SWTS, labour force survey; population census and/or other household surveys with an appropriate employment module

Share of unemployed non-students

Share of youth aged 15-29 unemployed and not in education or training (different from NEET in that it captures only unemployed non-students, i.e. those actively seeking jobs).

ILO/UCW YouthSTATS,

ILO YouthSTATS.

SWTS, labour force survey; population census and/or other household surveys with an appropriate employment module

Vulnerable employment rate

Share of workers aged 15-29 who are either own-account or contributing family workers.

ILO/UCW YouthSTATS,

ILO YouthSTATS

SWTS, labour force survey; population census and/or other household surveys with an appropriate employment module

Poorly paid

Share of own-account workers and paid employees aged 15-29 with below-average wages or income.

Available in ILO publications (Global Employment Trends for Youth 2013: A generation at risk)

SWTS, labour force survey with earnings module, establishment survey on employment and earnings

Time-related underemployment rate

Employed youth aged 15-29 who are willing and available to increase their working time and worked fewer hours than a specified time threshold during the reference period (usually 40 hours per week).

ILO/UCW YouthSTATS

SWTS, labour force survey, household survey with an employment module, administrative records

Qualifications mismatch

Skills mismatch between the job that a worker aged 15-29 does and their level of educational qualification, measured by applying the normative measure of occupational skills categories from the International Standard Classification of Occupations.

Undereducated/well-matched/overeducated: Young workers in a particular group who have a lower/the assigned/higher level of education.

Available in ILO publications (Global Employment Trends for Youth 2013: A Generation at Risk, Work for Youth [W4Y] project)

SWTS, labour force survey, household survey with an employment module, establishment survey

Job satisfaction

Share of workers aged 15-29 who are satisfied (reported being very satisfied or somewhat satisfied in the SWTS) with their jobs.

Available in ILO publications (Global Employment Trends for Youth 2013: A Generation at Risk, W4Y project)

SWTS; labour force survey and other household surveys with an employment module that includes information on job satisfaction

Distribution of youth aged 15-29 by stage of school-to-labour market transition

Transitioned: Youth currently employed in i) a stable job, whether satisfactory or non-satisfactory; OR ii) a satisfactory but temporary job; OR iii) satisfactory self-employment.

In transition: Youth i) currently unemployed (relaxed definition); OR ii) employed in a temporary and non-satisfactory job; OR iii) in non-satisfactory self-employment; OR iv) inactive and not in education or training, with intention of looking for work later.

Transition not yet started: Youth i) still in school and inactive (inactive student); OR ii) inactive and not in education or training (inactive non-student), with no intention of looking for work.

Negative outcomes: Youth in transition AND transition not yet started (inactive non-students with no intention of looking for work.

Available in ILO publications (Global Employment Trends for Youth 2013: A Generation at Risk, W4Y project)

SWTS; labour force survey and other household surveys with an employment module that includes information on job satisfaction

Table 3.A1.4. Youth civic participation and empowerment

Main indicators

Definition

International databases

Raw data sources

Interpersonal trust

Percentage of youth aged 15-29 who declared trusting people.

WVS – Online Data Analysis

WVS wave 6

Institutional trust

Percentage of youth aged 15-29 who declared being confident in institutions, such as national government, civil services, police, military, judiciary system or media.

Gallup World Poll data,

WVS – Online Data Analysis

World Poll surveys, WVS wave 6

Secondary indicators

Definition

International databases

Raw data sources

Rate of civic engagement

Percentage of youth aged 15-29 who, in the past month, have volunteered their time to an organisation, donated money to a charity, or helped a stranger or someone they did not know who needed help.

Gallup World Poll data

World Poll surveys

Rate of social network support

Percentage of youth aged 15-29 who declared that they have relatives or friends they can count on to help them if they were in trouble/in case of need.

Gallup World Poll data

World Poll surveys

Voicing opinions

Percentage of youth aged 15-29 who have voiced their opinion to a public official in the past month.

Gallup World Poll data

World Poll surveys

Voting rate

Percentage of youth age 29 and under who declared that they usually or always vote when elections take place (local or national).

WVS – Online Data Analysis

WVS wave 6

Rate of juveniles brought into formal contact with the police and/or criminal justice system

Number of juveniles age 17 and under suspected, arrested or cautioned (all crimes) per 100 000 juveniles.

UNODC – Statistics on criminal justice

Crime Trend Survey, national surveys, scientific literature

Rate of juveniles detained

Number of juveniles age 17 and under held in prisons, penal institutions or correctional institutions per 100 000 juveniles.

UNODC – Statistics on criminal justice

Crime Trend Survey, national surveys, scientific literature

Homicide rate

Number of victims aged 15-29 of intentional homicide (unlawful death purposefully inflicted on a person by another person) per 100 000 youth.

UNODC – Homicide statistics

United Nations Survey of Crime Trends and Operations of Criminal Justice Systems (UN-CTS)

Victims of assault

Number of victims aged 15-29 of assault (physical attack against the body of another person resulting in serious bodily injury, excluding indecent/sexual assault, threats, slapping/punching and assault leading to death) per 100 000 youth.

UNODC – Statistics on crime (only available for the total population); WHO – Global school-based student health surveys (GSHS)

UN-CTS; GSHS: Percentage of youth aged 13-17 years who have been physically attacked at least once in the past 12 months

Gender-based domestic violence in the lifetime

Percentage of women aged 15-29 who have experienced physical and/or sexual violence from an intimate partner at some time in their lives.

OECD Social Institution and Genders Index

DHS, International Violence Against Women Survey

Table 3.A1.5. Youth life evaluation, feelings and meaning

Main indicators

Definition

International databases

Raw data sources

Dissatisfaction with life

Percentage of youth aged 15-29 who are currently dissatisfied with their lives as a whole: youth who reported less than 5 on a scale from 1 (completely dissatisfied) to 10 (completely satisfied).

Gallup World Poll data

World Poll surveys

Negative feelings

Percentage of youth aged 15-29 who experienced at least one of the following feelings during most of the day before the interview: anger, depression, sadness, stress or worry.

Gallup World Poll data

World Poll surveys

Secondary indicators

Definition

International databases

Raw data sources

Low sense of purpose in life

Percentage of youth aged 15-29 who feel that their lives do not have an important purpose.

Gallup World Poll data

World Poll surveys

References

Alkire, S. and J. Foster (2011), “Counting and Multidimensional Poverty Measurement”, Journal of Public Economics, Vol. 95/7-8, Elsevier Ltd., Amsterdam, pp. 476-487.

Alkire, S. and J. Foster (2007), “Counting and Multidimensional Poverty Measures”, Oxford Poverty & Human Development Initiative Working Paper, No. 7, University of Oxford, Oxford.

Boarini, R., A. Kolev and A. McGregor (2014), “Measuring Well-being and Progress in Countries at Different Stages of Development: Towards a More Universal Conceptual Framework”, OECD Development Centre Working Papers, No. 325, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jxss4hv2d8n-en.

Decancq, K. and M.A. Lugo (2013), “Weights in Multidimensional Indices of Wellbeing: An Overview”, Econometric Reviews, Vol. 32/1, Taylor & Francis, Abingdon, pp. 7-34.

De Neubourg, C. et al. (2012), “Step-by-Step Guidelines to the Multiple Overlapping Deprivation Analysis (MODA)”, Innocenti Working Paper, No. 2012-10, UNICEF Office of Research, Florence.

Deneulin, S. and L. Shahani (eds.) (2009), An Introduction to the Human Development and Capability Approach: Freedom and Agency, Earthscan, London.

ILO (2015), “What Does NEETs Mean and Why Is the Concept so Easily Misinterpreted?”, Technical Brief Note, No.1, International Labour Organization, Geneva.

ILO (2013), Global Employment Trends for Youth 2013: A Generation at Risk, International Labour Organization, Geneva.

J-PAL (2012), “Deworming: A Best Buy for Development”, J-PAL Policy Bulletin, Abdul Latif Jameel Poverty Action Lab, Cambridge, MA.

Jütting, J. and J.R. de Laiglesia, (eds.) (2009), Is informal Normal? Towards More and Better Jobs in Developing Countries, OECD Development Centre, OECD Publishing, Paris.

McKnight, J. and J. Kretzmann (1996), Mapping Community Capacity (Revised), Institute for Policy Research, Northwestern University, Evanston.

MOLISA (2016), “Over a Million Unemployed in Quarter 2/2016”, webpage, Ministry of Labour, Invalids and Social Affairs, Socialist Republic of Viet Nam, Ha Noi.

MSActionAid-Denmark (2010), Empowerment Guide, Mellemfolkeligt Samvirke ActionAid-Denmark, Copenhagen.

OECD (2014), Social Institutions and Gender Index: 2014 Synthesis Report, OECD Publishing, Paris, www.genderindex.org/sites/default/files/docs/BrochureSIGI2015.pdf.

OECD (2011), How’s Life? Measuring Well-being, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264121164-en.

OECD/EU/JRC (2008), Handbook on Constructing Composite Indicators: Methodology and User Guide, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264043466-en.

Saltelli, A. (2007), “Composite Indicators Between Analysis and Advocacy”, Social Indicators Research, Vol. 81/1, Springer Publishing, New York, pp. 65-77.

Schafer, J.L. (1999), “Multiple Imputation: A Primer”, Statistical Methods in Medical Research, Vol. 8/1, Sage Publishing, Thousand Oaks, pp. 3-15.

Sen, A. (1998), Development as Freedom, Oxford University Press, Oxford.

Shehu, E. and B. Nilsson (2014), Informal Employment Among Youth: Evidence From 20 School-to-Work Transition Surveys, International Labour Organization, Geneva.

UN DESA (2010), World Programme of Action for Youth, United Nations Department for Economic and Social Affairs, New York.

UNDP (2014), UNDP Youth Strategy 2014-2017: Empowered Youth, Sustainable Future, United Nations Development Programme, New York.

UNDP (1990), Human Development Report 1990, Oxford University Press, Oxford.

UNESCO (2014), Teaching and Learning: Achieving Quality for All, Education for All Global Monitoring Report 2013/14, United Nations Educational, Scientific and Cultural Organization, Paris.

WHO (2014), Health for the World’s Adolescents: A Second Chance in the Second Decade, World Health Organization, Geneva.

World Bank (2008), Supporting Youth at Risk: A Policy Toolkit for Middle Income Countries, World Bank, Washington, DC.