World Education Forum > Press Kit >
Good Statistics Count
( -pdf )
The quality of the 183 country reports to be presented at the World Education Forum has exceeded all expectations.
“The message that educational planning must be based on reliable, up-to-date statistics is being heard,” says Warren Mellor, co-ordinator of the global EFA 2000 Assessment, the biggest review of education ever undertaken. “The collecting and using of data for sound policy-making is now on the agenda in most of the world’s countries.”
National co-ordinators spent two years struggling to come up with recent and honest data. They discussed draft country reports at more than fifteen regional and sub-regional workshops and received technical assistance. But, as Mellor points out, “the Assessment is not an end it itself, it is just the beginning of a long process.”
The Assessment was primarily designed to help countries with future planning and policy-making on basic education. It was never intended to create an international “league table” comparing countries’ educational performance. Nevertheless, governments sometimes manipulate data in order to make their results look as flattering as possible.
A Rosy Picture
“Certain governments tend to use statistics ‘in the best way’,” comments Svein Osttveit, Executive Secretary of the Education for All Forum. “Everyone agrees education is important, but some political leaders hesitate to show lack of progress or, in some cases, that things are getting worse.” Denise Lievesley, Director of the UNESCO Institute for Statistics agrees. “It is not surprising that political leaders hesitate to show lack of progress and want their countries to show themselves in the best light possible.” She hopes that the promotion of policy-making based on evidence will begin to have an impact on the quality of data.
For years, a “goals-reporting culture” dominated efforts to review progress in education, according to Jim Irvine of UNICEF. In the mid-decade review of education in 1996, Irvine says, country reports submitted by governments painted so rosy a picture of the achievements as to invite criticism. Apart from producing doubtful “official” figures, this practice produced statistics that were not only dishonest but also virtually useless for measuring changes, for better and worse.
A Culture of Transparency
The EFA 2000 Assessment has attempted to reverse this trend and to promote a culture built on transparency and openness. It began by refining and harmonizing the measurement tools. Eighteen statistical indicators were selected on the six Education for All goals (see box overleaf). These indicators include enrolment figures, public spending on education, pupil-teacher ratios, repetition rates and the ratio of female to male literacy. Countries were asked to present not just national averages but sub-regional data to show gaps in urban and rural figures as well as data for the smallest sub-district or village.
“Our team has been working long nights checking the consistency and plausibility of the data received,” says Lievesley, “If there are concerns, they have been raised with the country and with the EFA regional organizers.” However, Lievesley admits, “we have to be sensitive because statistical systems within countries are not always as advanced as we would like them to be.”
A positive trend is the emergence of a new attitude to reporting among governments. In the Arab States, to cite one example, Victor Billeh, regional co-ordinator of the EFA 2000 Assessment, explains that “in the past, transparency was taboo in the Arab States. It was considered shameful to admit that levels were not as high as expected. This barrier has now been crossed.” All country reports have been published and shared in the region and they are also available on the Internet. “The information now being exchanged is fuller than ever before,” Billeh says.
The EFA 2000 Assessment has also revealed the limitations of quantitative data. “We have been rather poor at getting information that enables us to judge the quality of education, the efficiency of the system and whether resources are being used effectively,” comments Lievesley. “From now on, we need to look at collecting data in different ways, not just concentrating on official sources but, for example, employing surveys on out-of-school children. We also need to work more closely with non-governmental organizations who may have access to local data.”
Education specialists agree that although policy should not be driven by statistics, it should be informed by statistics. Today, the relevance of data to policy is under-recognized in many countries, and too much information collected by government ministries just sits in offices and is not distributed in accessible and useful ways.

The Six Education for All Goals

Expansion of early childhood care and development activities, including family and community interventions, especially for poor, disadvantaged and disabled children.

Universal access to, and completion of, primary education (or whatever higher level of education is considered as “basic”) by the year 2000.

Improvement in learning achievement so that an agreed percentage of an appropriate age cohort (e. g. 80 per cent of 14 year-olds) attains or surpasses a defined level of necessary learning achievement.

Reduction of the adult illiteracy rate (the appropriate age group to be determined in each country) to, say, one half its 1990 level by the year 2000, with sufficient emphasis on female literacy to significantly reduce the current disparity between male and female illiteracy rates.

Expansion of provision of basic education and training in other essential skills required by youth and adults, with programme effectiveness assessed in terms of behavioural changes and impacts on health, employment and productivity.

Increased acquisition by individuals and families of the knowledge, skills and values required for better living and sound and sustainable development, made available through all education channels including the mass media, other forms of modern and traditional communication, and social action, with effectiveness assessed in terms of behavioural change.

Source: Framework for Action to Meet Basic Learning Needs, 199