Monitoring Children's Rights:
Indicators for Children's Rights Project
What do we know about children?
[Back to index]
The Childwatch International Monitoring Children's Rights Project is a complex endeavour that involves researchers, experts and activists from many different countries and at many different levels. The project was originally designed jointly with the Director of Childwatch International, Per Miljeteig, without whose continued interest and support, professional, intellectual and personal it would not have made progress. Many of the fundamental ideas and much of the inspiration came from prior experiences in the European Centre 'Childhood as a Social Phenomenon' project, particularly the work of Angelo Saporiti, who has also had considerable input into the development of this project. The Global Advisory Committee of the project has provided valuable support and advice. Special thanks are due to the Chair, Geraldine Van Bueren, and to Ferran Casas. Considerable progress was made through a collaboration with UNICEF's New York Headquarters in the context of the 1996 Congress Against the Sexual Exploitation of Children, and a lively pre-Congress workshop in Stockholm with Mark Connolly, Kusum Gopal, Janet Herran and Heather Montgomery.
The greatest debt, however, is due to the members of the country case study teams in Senegal, Nicaragua, Thailand, Vietnam and Zimbabwe. In particular I have to single out, the pioneering members of the pilot study team in Senegal, Fatou Fall Dia, Abdoul Karim Gueye, Nafy Diagne Gueye and Sidy Gueye; the often heated debates between members of the Nicaraguan team, Alejandra Guido, Gladys Matute, Luis Moro and Mireille Vigil; the consolidating work carried out by the core members of the Zimbabwe team, Israel Chokuwenga, Linda Dube, Joshua Kembo, Precious Moyo and Ratidzai Sharon Nkomo, and the sterling work of interpretation, questioning and consistent companionship of Cao Manh Thang in Vietnam.
Monitoring Children's Rights Project
The collection of childhood social indicators began in the 1970s. Under situations of increasing complexity and rapid social change, childhood indicators, like other social indicators, require more and better data to demonstrate what is happening and why it is happening. Pressure to increase global information about life situations of children has increased steadily since 1979, the United Nations International Year of the Child. Since then, academic studies of children have increased and, a decade later, the Convention on the Rights of the Child was adopted by the United Nations. The vast majority of nations have now signed and ratified the Convention, which means that their governments are now obliged to supply regular reports on the situation of children's rights in their country to the Committee on the Rights of the Child, which was established to monitor the implementation of the Convention.
To ensure effective implementation of the Convention on the Rights of the Child, so that children really benefit from the protection it gives, practical indicators, based on reliable statistical or other data-gathering methods, are required. These must be easy to collect, interpret and use, not only for the Committee on the Rights of the Child, but also for UNICEF and specialized agencies of the United Nations as well as national governments and the NGO community.
The Committee on the Rights of the Child has called for indicators related to the various articles of the Convention, that could meet some basic requirements such as validity, objectivity, sensitivity, comparability, accuracy and disaggregation, and has appealed to the UN system, NGOs and the research community for assistance.
Childwatch International has designed a demonstration project to meet the needs expressed by the Committee to analyse further specific needs for indicators to monitor the implementation of the Convention on the Rights of the Child and suggest how the status of the various rights could be expressed through objective data. Through a series of country case studies, the project has developed a strategy for identification and development of the appropriate indicators.
The results of the project will be presented in Country Case Reports, as well as a Summary Report that combines the results and experiences of the study as a whole, with recommendations for future actions and activities. A Manual on how to use selected indicators based on easily available quantitative and qualitative data in the monitoring of the Convention for use by governments, communities, NGOs and child rights advocates will also be produced. This present volume is intended to act as a background document to explain the basic premises of the project.
As pointed out above, verifiable information and accurate data are essential for the implementation of the Convention on the Rights of the Child. Any monitoring process must be based on information, as the Committee on the Rights of the Child has recognised in calling for reliable indicators for children's rights to be used in the reporting process by countries that have signed and ratified the Convention. It is only by knowing the actual situation of children that efficient steps to improve their lives can be taken, monitored and evaluated.
The project is based on the premise that many indicators for children's rights can be developed on the basis of already-existing data. It is not necessary to construct entirely new systems of indicators, but rather to explore the possibilities of adapting systems that are already in place. Thus monitoring children's rights should become part of the normal mechanisms of government and not a special effort once every five years, when reports to the Committee become due. A routine system of children's rights indicators would also serve other purposes, for both non-governmental and governmental sectors, not only showing the conditions of children's lives on a daily basis, but also monitoring the impacts of programmes designed to improve their situation as well as providing the possibility of assessing rapidly the effects of natural or man-made disasters.
A good deal of information about children already exists, but much of it is not collected with children in mind, with children as the units of observation, or in ways that reveal much about children's rights. The data are scattered around different agencies and seldom centralised. Information about an individual child is often fragmented throughout the system, with facts about school attendance in one place, about family life in another and health records appearing in several different data bases. Not only is the information not coordinated, much of it, although regularly collected, recorded, published and archived, is seldom put to use in the interests of children. Reports to the Committee on the Rights of the Child skim the surface of available data, without taking advantage of the full range of information, and without paying attention to the national contexts of children's lives.
The objective of the Childwatch project is to aid the process of moving towards a situation in which the fullest possible information about children's lives can be used to improve their future enjoyment of their rights. It argues that, while the process of constructing mathematical indicators demands certain technical and statistical skills, it must be based on the initial work of indicator development, outlined in this manual, which requires less obviously technical, but equally painstaking, sociological, cultural and conceptual labour.
Monitoring children's rights has two objectives:
Monitoring children's rights has five components:
[Back to index]
Monitoring children's rights around the world
The project is designed to fit into the overall process within the field of human rights to develop indicators for use in monitoring the various international human rights treaties, particularly the Covenant on Economic, Social and Cultural rights.
Current discourses on the Convention on the Rights of the Child (CRC) tend to be dominated by legal standard-setting. When it comes to setting policy goals and interventions for implementing these standards, it is appropriate to explore the meanings they have in different cultural contexts. One danger of accepting that other systems of concepts and beliefs can be equally rational in their own terms, is that this can lead to the argument that it is impossible to make cross- cultural comparisons, in which case it would be impossible to establish universal human rights law except through cultural imperialism. The Childwatch International Monitoring Children's Rights Project, however, takes the position that international standards can be set, not through establishing which system is 'best', nor by negotiating some kind of lowest possible denominator, but rather by moving the debate to another level in which, as a sociologist has put it, 'different institutions, embodying different conceptual schemes, may be illuminatingly seen as serving the same social necessities' (1).
As has been widely recognised, the CRC contains many linguistic and cultural traps and has been criticised at times for being at best inapplicable outside Western countries or at worst as an instrument of cultural imperialism (2). The most notable example is probably that of Article 3 (1) on the best interests of the child. Although this article is usually taken to be fundamental in both Western and non-Western settings, in the former it is associated with legal and social work practice, while in the latter it often means something akin to 'children first' (3). Not for nothing is Philip Alston's book The best interests of the child subtitled 'Reconciling culture and human rights'. As he points out, the CRC, perhaps more than any other international human rights instrument, has revealed a need to seek 'approaches which involve neither the embrace of an artificial and sterile universalism nor the acceptance of an ultimately self-defeating cultural relativism.' (4)
Resistance to the very idea of cultural relativism, as the African human rights lawyer Abdullahi Ahmed An-Na'im points out, is often based on the suspicion that it 'denies to individuals the moral right to make comparisons and to insist on universal standards of right and wrong' (5). Yet individuals, however outraged they may be by certain customs and practices, especially in the fields of child abuse and exploitation, cannot set either local or international standards by themselves and more than respect for culture can be used, as Alston says, as a 'trump card' (6). All customs and standards are social, the product of interactions within groups, whether these are small-scale or what is sometimes called the 'human rights' or 'international' community. Moreover, as An-Na'im further reminds us, cultures do not simply differ from group to group, they are also dynamic, flexible and often internally inconsistent or ambivalent (7). Reality is socially constructed and values constantly negotiated. The ideational structure of international human rights law is itself a new cultural product.
The CRC presents some fairly obvious difficulties. One example is Articles 16 (1) on privacy, which has little relevance in countries where individual, physical privacy does not have high cultural value. Another is the use of the phrase 'periodic review of placement', in Article 25, which is hard to envisage in the context of traditional fostering systems, as in West Africa (8). More deep-rooted difficulties can be experienced with respect to Article 1, the definition of the child, not simply in legal terms, with respect to the ages at which certain activities can be permitted or expected, but also in terms of the cultural definitions of the ages and stages of childhood. Even in the English-speaking world there is not single word that can be meaning-fully applied to infants, toddlers, primary school children, early teenagers and youth under the age of 18 years. Indeed, some terms may be exclusively applied to either girls or boys. In addition, there are wide cultural variations in expectations of the level of maturity expected at different ages or stages, which also has implications for the interpretation of Article 12 (CRC). Thus the process of developing a children's rights monitoring system begins with a critical evaluation of the text of the Convention in the local language and in conjunction with any regional children's rights instruments, such as the Organisation of African Unity Charter on the Rights and Welfare of the African Child. The main questions asked are:
This leads to a structured set of indicators within each family rather than a list of unrelated indicators.
[Back to index]
In 1992, the first examples of an entirely new genre of text were produced as States Party to the Convention on the Rights of the Child began to report to the Committee set up under Article 43. Although the Committee produced guidelines, these texts are extremely diverse, although they fall into three superficial textual categories:
These first reports vary enormously in length and also show considerable differences in terms of the quantity and type of both data and supplementary information provided. Although the purpose of the reporting system is to monitor the progressive achievement of children's rights within each country that is a State Party, and the terms monitoring and indicators are currently very much part of the milieu of international child welfare, attempts to set up monitoring systems or develop appropriate indicators rarely appear within these texts, in most cases they are not even mentioned.
Perhaps one of the more interesting comparisons between reports is the relative concentration on:
Inevitably these are country reports, which have more to say about states than about the children whose rights are supposedly the subject of the texts. This influences the structure and content of a report to the Committee to a greater extent than any definition of the child or measures taken to provide for or protect children. It is not unusual for even the word 'child' to appear for the first time quite a considerable way in to the text. Mexico is unusual in mentioning children in the first sentence and Costa Rica is the only report in the first twenty received by the Committee that uses the actual words of a child.
It is also quite clear that states in general tend to provide numerical data with respect to the functions they have of either replacing or supplementing the family - such as orphanages, juvenile detention centres, pre-school care and of course schooling. Families are mentioned, often in the way in which they secure the rights of children so that the state does not have the same requirement to monitor, as in the Pakistan report which states that the joint family system performs 'many of the functions that are otherwise carried out by the social security system in the West'. With respect to hard data on children - let alone indicators that might be used as the baseline for a monitoring process - governments appear only to have data on children in institutions - schools, prisons, orphanages - on the health of children under five years old, but not on the families they live in despite much discussion of the ideal family. There is almost no date on the work they do particularly under the age of 10 (much less the value to the GNP).
[Back to index]
Regrouping articles of the Convention
The Convention on the Rights of the Child provides a framework for monitoring in two ways. In the first place, it contains 40 articles that can be monitored. It might be tempting to think of finding an indicator for each article, but this would be a false approach. Many of the articles cover broad areas, such as health and economic exploitation, which require more than one indicator, and some articles have overlapping concerns, for which indicators might be shared.
Thus it makes more sense to group the articles, consider the range of phenomena they cover in particular social and cultural contexts, and develop corresponding concepts. These concepts then point to a series of questions that will lead to appropriate data.
The interpretation of the Convention in any social and cultural context will vary, just as ideas about children and childhood vary across cultures and societies. This is reflected in the ways articles are regrouped within the Childwatch process, following the critical evaluation of the text. Regrouping follows national, cultural interpretation and thus matches the way in which children are thought about in society and reflected in the way data about them are collected and presented.
The Convention was largely drafted by Western countries, this means that there are several areas where interpretation is difficult, particularly in developing countries. Chief among these are:
However, each of the country case studies has produced regroupings of the articles of the Convention according to national ideas about such topics as rights, childhood, family and citizenship. The contrast between these regroupings and those provided in the more dominant forms of regroupment derived from Northern settings can be seen in Figure 1.
Figure 1: Some regroupments of the articles of the Convention on the Rights of the Child
|UNICEF:||Basic principles, Survival and development, Protection|
|Children's Rights:||Provision, Protection, Participation (with or without basic principles)|
Some regroupments from Country Case Studies:
|Senegal:||Basic principles, Provision, Protection, Participation|
|Vietnam:||Elementary rights, Provision, Protection|
|Nicaragua:||Human development, Children's rights, Adult responsibilities|
|Thailand:||Basic rights of children; Extension and support rights; Protection, Participation|
In addition to this use of the Convention as a framework for monitoring children's rights in general, it can be used to focus on particular issues or problems. This is particularly useful when considering how to monitor difficult areas, such as the sexual exploitation of children, which are often described as unmeasurable. Grouping articles around the key article (Article 34) provides avenues through which to seek data through which the sexual exploitation of children can be measured and monitored (Figure 2).
Figure 2: Using the Convention as a framework to monitor the sexual exploitation of children
|1, 16||Children, childhood, children's dignity (innocence), privacy|
|34, 35||Prostitution, traffic, pornography|
|12 (13,14,15)||Consent, power and maturity|
Linked articles, relating to provision
|3, 4||Reasonable expectations|
|5,8, 19, 21,22||Family|
|26, 27, 30||Community and state|
[Back to index]
Indicator is a fashionable word these days and means many things to many people. Even academic books and articles invariably start with long sections on definition, and no two definitions are alike. When people who work in programmes and projects talk about indicators, it sometimes sounds as if all they need to do is develop indicators and their work will be done. Indicators often sound like a magic solution to all problems. On the other hand, some experts often make it sound as if indicator work is extremely technical and requires considerable statistical or mathematical training. Although advanced indicator work may require these kinds of skills, it is not difficult to understand the principles of indicator construction and to be able to seek, find and evaluate the kinds of data and information required.
It is practical to begin with the simple phrase 'an indicator indicates'. An indicator just shows you something or points in a particular direction. Thus, a pain in the stomach tells a doctor there is something wrong 'down there', but he needs more information to know if it is appendicitis, the onset of labour, or just wind. This makes it possible to understand what an indicator does. But a more specific definition is required in order to develop actual indicators. One operational definition is:
A social indicator is the elaboration of an empirical determination of a concept representing a social phenomenon (or some of its attributes). (9)
This may sound like a complicated definition, but it becomes easy if some key words are examined. These words are: social, concept, measurement and statistics.
The word 'social' refers to everything in social life, so it is also so unspecific that it is also about nothing.Thus it is better to begin with the purpose for which a social indicator will be used than to try to define the word social. A social indicator is something used to assess the living conditions of individuals and social groups, and changes in these conditions between different places and at different times. Assess implies value judgements - there are no objective assessments. Any evaluation has a historical, cultural and social basis and also implies that judgements are being made with respect to a particular reference point, or standard.
An indicator stands for something else - a concept. It is more important to be clear about the concept, so that it can be measured, than to be accurate about the techniques of measurement.
To measure something means transforming abstract ideas and concepts into concrete phenomena. If, for example we take the term 'Child Labour' it can be given an operational definition by asking questions such as:
What do we mean by the idea or concept 'Child Labour'?
How do we define 'Child Labour' in a concrete sense?
How can we capture the concrete phenomenon of 'Child Labour'? By establishing and defining some practical rules:
Do we only include children who are paid for their activities?
Do we include children who work with or for their parents wihout payment?
What about school activities?
Measurement is a technical question, but before measuring one must have an operational concept of what is to be measured.
Thus, although technical refinement is necessary at a later stage, the first part of indicator construction is a relatively simple process (Figure 3).
Figure 3: The process of constructing a social indicator
|Phenomenon||What do you want to measure?|
|Concept||How can this phenomenon be defined?|
|Data||What information best describes and/or measures this concept?|
|Indicator||How can this information be used to measure phenomenon/concept?|
It is important to realise that indicators are only tools to help understand and follow that is happening. The events are more important than the indicators, just as children are more important than statistics.
It is also important to understand the difference between numbers, statistics and indicators, because the frequent emphasis on the importance of numerical (quantitative) data often leads to confusion. Numbers, in themselves, may mean very little -- or even be misleading. For example the statement 'There are 1 million sexually abused children in Asia' is often quoted. Yet this is an unsubstantiated number. To turn it into a statistic, one would need to add some qualifying information, such as:
According to estimates made by [A], on the basis of [B] type of calculation, in [Year 1], there were 1 million sexually abused children in Asia [under X years of age].
In other words, an indicator requires elements of comparison (time, place and proportion):
TIME This statistic is [C%] more (or less) than the calculation made on the same basis in [Year 2]; PLACE [D%] more (or less) than estimated by the same method in the same year in [for example, Africa]; PROPORTION And represents [E%] of the total population of Asian children in the same age group.
Thus, indicators are just tools for aiding understanding and not the answer to all questions. They are measurements or descriptions: if you start recording them regularly, they tell you how things are changing over time.
Statistical offices and governments often use a standard list of indicators to measure national progress. But there is currently no universal list that can be used succesfully in all countries. It is important to know the purpose and the context of monitoring before selecting and constructing indicators. In other words, there is a standard process rather than a standard list.
This does not mean discarding all existing indicators, but it does mean some rethinking. In many cases a long list of indicators is used simply because the statistics are available. A children's rights monitoring system has to begin by considering a number of possible sources of data, testing them out and then selecting one, or a limited set, that can be easily used. Good indicators are clear, easy to keep track of and record over time (a doctor taking a patient's temperature has a very successful indicator, the process doesn't take long and a trainee nurse can do it on her first day, mothers can do it for children, patients can do it for themselves)
Good indicators are also cost-effective (to use the same medical example again, thermometers are cheap, temperature charts can be made with a ruler and a sheet of paper). Cost-effectiveness (and speed of putting into practice) also means looking for possible indicators within existing data sets or by building on existing data sets. For example, information about working children can be obtained by proxy from school enrolment, absenteeism and drop-out figures; or by adding extra questions to routine government data collection, such as:
(i) Labour force or household surveys: add the question "What work is done by children in this household?"
(ii) Medical records of accidents: "Where did this accident happen -- at school, at home, at work, in the street?"
The answers to both these questions could provide considerable interesting information about child work.
To fulfil the criteria for good indicators the data must be the best available.
[Back to index]
Indicators for children's rights
Childhood indicators have been defined as:
statistical time series that measure changes (or constancies) in the conditions of children's lives and in the health, achievement, behaviours, and well-being of children themslves. They are numbers that tell something significant about how today's children live and how we as a society are raising them. (10)
The collection of childhood social indicators began in the 1970s. Under situations of increasing complexity and rapid social change, childhood indicators, like other social indicators, require more and better data to demonstrate what is happening and why it is happening. Pressure to increase global information about life situations of children has increased steadily since 1979, the United Nations International Year of the Child. Since then, academic studies of children have increased , as we have seen, the Convention on the Rights of the Child has led to a search for children's rights indicators, which will show the extent to which the provisions of the Convention are being achieved in each country.
The Convention does not establish norms for all the rights it contains, but the Committee on the Rights of the Child, which has responsibility for monitoring the implementation of the Convention, has called for a valid system of indicators to give information on:
From the point of view of monitoring children's rights, three levels of indicator are required:
It is also important to remember that the Convention on the Rights of the Child is unique among international human rights instruments in providing for economic, social and cultural rights as well as civil and political. This means that the tenor of human rights work has changed from the idea of seeking and denouncing violations, as has been the historical case with civil and political rights, the idea of which has all too often been used as a political and economic weapon. Monitoring economic, social and cultural rights is viewed more as a matter of tracking progresive achievment of internationally-set standards, than pointing the finger at alleged infractions,which can only produce defensive and negative attitudes on the part of accused governments. The idea is that monitoring is a support for government in the proper process of governance and the development of the next generation of citizens, rather than a means of locating and denouncing violations of children's rights.
More importantly from the point of view of the Childwatch project is the fact that, for monitoring the Convention on the Rights of the Child an exceptionally wide range of indicators is required at different levels of statistical sophistication, and in areas that have previously been largely outside the usual focus of indicators work on economic issues. To quantitative, economic measures of development are now added the more qualitative aspects of quality of life; to the idea of the violation of political and civil rights is now added the notion of progressive realisation of freedoms and enjoyment of richer human existence.
Traditionally children's indicators have been used to rank nations in terms of their performance against a number of criteria, mostly related to health and education, as in the UNICEF publications The state of the world's children, and The progress of nations. This can be taken further. For instance, comparing figures in tables from The State of the World's Children, 1995, shows relationships between government expenditure in different sectors and the outcomes for children. Indicators such as these can be used for children's rights work, because they show the extent to which governments prioritise children's issues (Article 4 of the Convention on the Rights of the Child).
The basic difference between childhood indicators and indicators for children's rights can be illustrated by the following example:
In 1989, 80% of Zimbabwean children under 1 year of age were immunised against all antigens except measles.
Children's rights indicator:
In 1989, 20% of Zimbabwean children under 1 year of age were not protected by vaccination. Their rights to health, survival and development were not being met.
But which groups of children are in this 20%?
Children's rights indicators answer this question by considering two special dimensions: disaggregation and children-centered statistics.
[Back to index]
Indicators for children's rights are a particular kind of childhood indicator. They show differences between groups of children with respect to whether particular rights are being violated or not met. All indicators should be:
Factual: Mean the same to everyone; Valid: Measure what they claim to measure; Verifiable: Can be checked; Sensitive: Reflect changes; Reliable: Not be subject to change themselves; Comparable: Capable of being used accurately in more than one situation; Accurate: Use the best possible data.
Thus national Under-Five-Years Mortality Rate (U5IMR) is a childhood indicator that tells you about the health and life chances of children under five. If it is tracked over time, it indicates the impact of various health promoting activities, and even national wealth. But it is also important to know if, for example, girl babies die more frequently than boy babies, or that a disproportionate number of children of a particular ethnic group die before their fifth birthday, or what differences in U5IMR there are between children in a remote jungle area and children in the capital. Knowing which groups of children do not get the right to life, survival and development is not only a rights issue, it also helps to plan and target programmes. Breaking statistics down to see which groups are most affected by violation or non-achievement of rights is achieved by disaggregating existing data, or building appropriate disaggregations into new data, and is vital for child rights indicators. It shows inequalities between groups of children.
This relates to the important principle of non-discrimination (Article 2 of the Convention) and can be explored as part of the process of critical reading of the Convention, in order to decide which disaggregations are appropriate for a national monitoring system. The questions to ask in this respect are:
For boys ands girls generally?
For different socio-economic groups?
For different races, ethnicities or religions?
According to where they live (province, district, eco-area, urban/rural)?
To answer these questions it is necessary to look at several sources of data critically and this may raise a third question:
Key factors to look for are:
Mortality and morbidity rates, and nutritional status:
Which may correlate with provision of potable water, immunisation, health service provision;
School drop-out, absenteeism and repetition rates:
Which may correlate with provision of schools and teachers as well as with child work figures and the rates of entry into secondary and higher education.
[Back to index]
Children may be subject to a double inequality:
In recent years, historians and sociologists have pointed out that ideas about childhood and children are social constructions and that the current notions used in the North and exported to the South, through means such as the Convention on the Rights of the Child, are relatively recent constructions, dating from around the eighteenth century. The idea of childhood underlying the standards set by the Convention on the Rights of the Child is derived from an ideal that became common among higher socio-economic groups in Europe and North America during the nineteenth-century. Although this ideal is by no means typical of the countries in which it originated, it is implicit in most international writing about and programmes for children. It assumes that children will be cared and provided for in a nuclear family, will play and go to school rather than work, and have few responsibilities and duties in their families or in communities.
As the twentieth century has progressed...highly selective, stereotyped perceptions of childhood - of the innocent child victim on the one hand and the young deviant on the other - have been exported from the industrial world to the South. They have provided a focal point for the development of both human rights legislation at the international level and social policy at the national level in a wide range of countries. (11)
Thus another question to be asked at the start of the monitoring process is: What is the definition (or definitions) of childhood in this country?
Ideas to take into consideration:
What words are in common use, by people and by organisations, to talk and write about people under 18 years of age?
What do these words mean? In terms of the age groups to which they refer, and in terms of the expectations of rights, duties and abilities of people in these groups?
Would these expectations be different for boys and girls, in urban and rural areas, among poor children and rich children?
In the cultures and languages of ethnic groups, childhood may have different meanings.
What ethnic groups should be taken into account?
What are the words these groups use for people under 18 years of age?
What do these words mean in terms of the age groups to which they refer, and in terms of the expectations of rights, duties and abilities of people in these groups?
Would these expectations be different for boys and girls?
How true is this ideal of childhood for most children in X?
To what extent is this ideal implicit in writing about and policies for children in X?
Most statistics are not collected with children as the unit of observation The unit of observation and analysis may be adults, households or institutions. To find out what is really happening to children it may be necessary to recalculate the data so that children are the units of observation and childhood is the unit of analysis. Even though information about children is collected, the data are not generally children-centred. Children are not considered as individuals or as a social group neither in scientific literature, (psychology, pedagogy, and sociology), nor in social research and official statistics.
In most childhood data, the unit of analysis is the family and developmental problems are to the forefront. The asymmetric relations that define children out of social life are taken for granted. Statistically, a vast amount of data that could be collected about children's lives - time budgets, econmic activities, domestic duties and so forth - tend not to exist. Children are studied with respect to childhood institutions, such as the school or the family, but not with respect to the system of production or the labour market. This brings with it a requirement to make significant qualitative changes a the level of surveys and also to carry out secondary analysis of existing material. According to this approach, there are both first and second order biases in official statistics with respect to children:
Because children are discriminated against in society they are discriminated against in statistics;
A discriminative selection is imposed on existing data. Information about children is collected and recorded but it is neither presented nor tabulated as such in official publications: ultimately children are not transferred from the level of observation to the level of analysis.
What are children-centred statistics? Some contenders:
1. Number of births
It might be assumed that the number of births would be a children-centred statistic, but this is just a number that says nothing about the status of children or childhood. Even in a time series, number of births is not a very good indicator, because it gives no idea of the relationship between number of births and size of population.
2. Crude birth rate (1)
This is is an indirect childhood indicator. Because it has an element of proportion, crude birth rate can act as an indicator of the status of children, especially in a time series.
3. Total fertility rate (2)
This is a more accurate indirect childhood indicator that enables one to make certain comments about the social value of children over time and in different places or groups.
4. Proportion of children in the population
The proportion of children in the total population does not say much, it is a simple comparison between populations in a time series.
Children-centred statistics focus on children in statistical calculations. Table 1, which shows statistics from Italy, is significant because it shows how children-centred statistics differ from family-centred statistics. When families are presented in statistics according to the number of children, the children appear to be attributes of families (like number of radios or number of dogs). In the columns labelled 'children' in the same table, the difference is notable. The frequently cited statistic (in Italian media) that '41.8% of Italian families have only one child' might be interpreted to mean that '41.8% of Italian children are only children'. But in reality only 19.9% of Italian children are only children, the rest enjoy the company of at least one sibling.
(1) Annual number of births per 1,000 population. [Back]
(2) The number of children that would be born per woman if she were to live to the end of her childbearing years and bear children at each age in accordance with prevailing age-specific fertility rates.[Back]
Table 1: Children, family and siblings, Italy, 1983 (000) (12)
Number of children
Families with children
In developing countries, the importance of children-centred statistics can also be seen in the figures given for single parent households, which are used for many purposes. This is an adult-centred, household-based statistic. It is sometimes possible to discover from it how many children live in one-parent households. But you can rarely tell how many live in a one-parent-one-child, one-parent-two-child, one-parent-three-child family, despite the fact that this makes a good deal of difference to the risk factors involved and has implications for planning and targeting resources in prevention programmes.
"Most public statistics as well as most social accounting and social survey studies -- when reporting about children -- do so with reference to their parents' position in the socio-occupational structure."
"The taken-for-granted way in which fertility data are presented as data on children is symptomatic of an adult-oriented twist to statistical material....Typically we have information about age-specific birth rates, for example, but we do not know the average ages of parents for current cohorts of children, nor do we know if they are changing."
"..statistics show how many unemployed men have children, but not how many children live in households in which the father is unemployed." (13)
Statistics are either collected or presented with adults or institutions (schools, health services) in mind. But it is possible in many cases to return to the raw data and recalculate children-centred statistics.
[Back to index]
One disaggregation that is often forgotten is age. This does not just refer to the differences between children and adults, or children and the elderly. It also refers to differences between age groups within childhood. Sometimes it is necessary to break down the group 0-18 years into smaller age groups. A typical example is to examine separately the mortality and morbidity rates for children under 12 months of age because children are most vulnerable at these ages. But it is also important in other times of childhood. For example, it is useful to break down school attendance and drop out rates by year of age between seven and twelve, because there are particular ages at which children are most likely to begin to go out to work, or to the streets - usually about eight years old for boys for example.
One difficulty is that information on children is not normally recorded or presented in ways that make it easy to compare data from different sources. The age groups used usually reflect the needs of the organisation or agency collecting the information. Thus, census offices everywhere, presents the entire population in five-year age groups; Ministry of Education statistics reflect the priorities of schools; the Ministry of Health is oriented towards illnesses and health services; the household survey shows children as attributes of households and the Ministry of Labour is interested in the legal workforce.
It is not only the case that each culture/nation has its own construction of childhood, there are also cross-cutting (vertical) separate constructions of childhood pertaining to particular ministries. In order to make meanignful comparisons between different state agencies, let alone between different nation states, certain questions have to be asked of existing data:
What different ages are given for the end of childhood in different legal documents? How can these different ages be tabulated?
What age groupings are used for data collection and presentation by different agencies? (include government, IGO, NGO)
How easy is it to compare figures between agencies (children in school compared to children who are working for example)?
What are the major problems encountered in trying to compare age groups between agencies?
How could these problems be solved?
It is important that the age groupings used to collect and present information about children are the same. In most countries the age groups depend on the perspective of the agencies involved in collecting and presenting data for different purposes. Thus for example:
An indicator system requires comparisons to be made between information from these different sources, as well as for calculations to be made for the proportion of children in any age group that is, for example, enroled in school or involved in a vaccination programme. This means that, in order for a government to be able to fulfil its obligations under the Convention on the Rights of the Child, all agencies collecting information about children must make raw data available to the ministry or agency responsible for reporting to the Committee on the Rights of the Child. In order to monitor children's rights for all the articles of the Convention and the whole country it is necessary to establish a single frame of reference with respect to age groups, which can be used to collect and present data from all ministries and agencies responsible for information about children. The rationale behind this is not that all Ministries and other agencies should present all data in these age groups, but that the age groups in which data and collected and presented can be reconciled between data sets. This would mean that eventually:
[Back to index]
Notes and references
1. MacIntyre, A., 1970 "The idea of a social
science" in Wilson, B.R. (ed.) op cit.
2. See for example, Boyden, J., "Childhood and
the policy-makers: a comparative perspective on the globalization of childhood",
in James, A., and Prout, A., (eds.) Constructing and reconstructing childhood,
Brighton, Falmer Press: 184-215.
3. Even the translation into European languages other
than English can result in a change of meaning. For example, the official
Spanish translation, "el interest superior del niño" cannot
be understood in the same way as "the best interests of the child",
and likewise turns the issue from legal to moral discourse. See Ennew, J.
1996 forthcoming, "Manual para la recopilacion y analisis de la informacion
sobre derechos de la infancia", San Jose de Costa Rica, Instituto Interamericano
de Derechos Humanos.
4. Alston, P., 1994, The best interests of the
child: reconciling culture and human rights, Florence, UNICEF and Oxford,
5. An-Na'im, A. A., 1992, "Taking a cross-cultural
approach to defining international standards of human rights: the meaning
of cruel, inhuman or degrading treatment or punishment", in An-Na'im,
A. A. (ed.), Human rights in cross-cultural perspective: A quest for
consensus, Philadelphia, University of Pennsylvania Press, pp. 19-43,
6. Alston op. cit, p. 20.
7. Ibid, p. 27.
8. See for example Goody, E., 1970, "Kinship
fostering in Gonja: deprivation or advantage?" in, Mayer, P. (ed.),
Socialisation: the approach from social anthropology, London, Tavistock,
9. Saporitti, A. "A methodology for making children count", in Qvortrup, J.,
Bardy, M., Sgritta, G., & Wintersberger, H., Childhood Matters: Social theory, practice and politics,
Aldershot, Brookfield USA, Hong Kong, Singapore, Sidney, Avebury
10. Zill II, N., H. Sigal and O. G. Brim Jr., 1983,
"Development of childhood social indicators", in Zigler, E. F.,
S. L. Kagan and E. Klugman (eds.) Children, Families and government:
perspectives on American social policy, Cambridge, Cambridge University
11. Boyden, J., "Childhood and the policymakers:
a comparative perspective on the globalization of childhood", in James,
A. and Prout, A., (eds.) Constructing and deconstructing childhood: contemporary
issues in the sociological study of childhood, Basingstoke, The Falmer
Press, 1990, 191.
12. Source: Angelo Saporitti, training materials
for the Childwatch Monitoring Children's Rights Project in Vietnam.
13. Qvortrup, J., 1991, Childhood as a social
phenomenon, Eurosocial Report Vol. 36, Vienna European Centre, pp. 17,
22, & 23.
[Back to index]