2.2 Gather outcome data

Effective human rightsmonitoring is based on reliable and credible data, and accurate analysis of this data using international human rights standards. Data collected with an unclear or biased methodology risks being dismissed, undermining the credibility of your findings, and therefore your capacity to engage in advocacy. It could even damage your reputation as an organisation.

This step is intended to give you general guidance on education outcome data.

Data sources

National governments and international development agencies regularly collect and publish education statistics obtained at intervals from various countries, as well asdisaggregated data.

There arethree types of data:

Administrative data

This primary source of education data includes information gathered from a census of schools in agiven country, with categories of data such as pupil and repeater enrolments, numbers of teachers and derived pupil-teacher ratios, pupil progression rates (promotion, drop-out), education expenditure information, and participation rates derived from the combination of enrolment and population data (gross and net intake, enrolment, completion, transition rates), as well as information on the structure of the education system (duration of primary, lower secondary, upper secondary cycle).

Administrative data typically provides limited information on the individual characteristics of pupils (such as age, sex and residence), and some information on the characteristics of their households. Since much of this data is obtained from schools, it focuses on children who attend school and not children who are out-of-school.

Census data

A population census represents a complete enumeration of the entire population. It provides basic information about population size and distribution, gender, age, language, educational status, and other characteristics. Because census data provides information on the entire population, it can be disaggregated better than survey data (eg at the regional or district level or small sub-populations), which may not be representative at the sub-national level. 

Student assessment tests

Many national governments carry out assessments of learning achievements, particularly on reading and maths competencies. Many countries also participate inregional and international student assessments, which may allow for cross-country comparisons of learning achievements.

Adapted from UNESCO (2004) Guide To The Analysis And Use Of Household Survey And Census Education Data and Education Policy and Data Center (2004) About EPDC Administrative Data.

Where to find data in international databases

Data obtained from international data sets may be easier to obtain and this data has the advantage that its indicators allow forcross-country comparisons.

In theIndicators Selection Tool, if there is data available for the indicator in international databases, we have provided a link to it. If you do not find any links to databases for some of the indicators you have selected in Step 2.1, it is probably because no international data is available for that indicator. However, you may be able to find data from national sources, such as the Ministry of Education, the National Statistics Office and UNDP Human Development Reports.

The main international databases are:

UNESCO Institute for Statistics

UIS is the leading source for international education statistics. Covering more than 200 countries and territories, the UIS database covers all education levels and addresses key policy issues such as gender parity, teachers and financing.

The World Bank EdStats Query

EdStats holds around 2,500 internationally comparable education indicators on access, progression, completion, literacy, teachers, population, and expenditures. The indicators cover the education cycle from pre-primary to tertiary education. EdStats also holds learning outcome data from international learning assessments (PISA, TIMSS, etc), equity data from household surveys, and projection data to 2050.

UNESCO’s World Inequality Database on Education

WIDE brings together data from surveys from over sixty countries to enable users to compare education attainment between countries, and between groups within countries, according to factors that are associated with inequality, including wealth, gender, ethnicity and location. 

The Right to Education Project also has a page on where to find international Data and Statistics.

The importance of disaggregated data

Under international law, States are required to collect disaggregated data based on various grounds ofdiscrimination. From a human rights perspective, collecting and analysing disaggregated data is crucial because inequalities across various segments of a population on various education outcome indicators often indicate that thechances people have to enjoy their basic rights to education are heavily shaped by the circumstances in which they are born and not by factors over which they have control. In many countries,being a girl, living in poverty, being disabled, belonging to an ethnic minority or living in rural areas radically reduces the chances of obtaining a quality education. Even more so when you fall into multiple groups and experiencemultiple forms of inequality.

For eachoutcome indicator (and relevantprocess indicators) in theIndicators Selection Tool, we have provided a list of possible levels of disaggregation.

OHCHR on disaggregated data

While disaggregated statistics are essential for addressing human rights concerns, it is not practical or feasible always to undertake disaggregation of data at the desired level. Disaggregation by sex, age, regions or administrative units, may, for instance, be less difficult than by ethnicity, as the identification of ethnic groups often involves objective (eg language) and subjective (eg self-identity) criteria that may evolve over time.

Source: OHCHR (2000) Human Rights Indicators: A Guide to Measurement and Implementation: p. 68 

Data published at the national level is particularly helpful for obtaining disaggregated data. International data sets usually do not have data disaggregated by ethnic groups or by regions / districts. For the latter, you may sometimes find data in national databases, or else be able to access data from the local government (eg the district education office). National data sets may also contain more up-to-date data than international sources.

Many States and other institutions often fail to disaggregate data by all relevant marginalised groups, making it difficult to ascertain whether outcomes differ across groups.

When disaggregated data by ethnic group is not available, it may be possible to use disaggregation by region or municipality, comparing right to education indicators data of those regions in which an ethnic minority makes up more than half of the population with those regions where they make up less than half of the population. If you are working on a marginalised group for which disaggregated data is not easily available (eg immigrants or refugees), you may need to collect your own data (see below). 

Rely primarily on existing data

In order to ascertain whether there are inequalities and deprivations in the enjoyment of the right to education, you are going to need data. In order to make your monitoring project more practicable, you should, whenever possible, select the outcome indicators for which there is existing data. A list of indicators for which you have no established data or where you lack the ability to obtain the data is not very useful. Beyond making the monitoring project more feasible, there are otheradvantages to using existing data.

When relying on existing data you should bear in mind that those who originally collected the data may have had reasons to distort that data. For instance, if resource allocation from the government to public schools is tied to enrolment, schools may report higher enrolment figures in order to obtain greater resources, producing distortions in the estimates of student enrolments and the number of teachers or classrooms. Likewise, private schools may under-report income and expenditure in order to derive greater benefits or lower their tax liability.

Therefore, whenever possible you should use a standard data source that is internationally accepted (such as those suggested for each indicator in theIndicator Selection Tool) and whenever doubts have been raised about the veracity of the data, you shouldassess the reliability of that data.

It may also be worth considering alternative sources of data, such as academic data sets or monitoring projects conducted by civil society organisations (CSOs).

If you cannot find any data, you may want to consider collecting your own.

Collect new data

For some indicators there may not be data available. It may be that no data has ever been collected for this indicator. For instance, for the indicatorspercentage of children with disabilities enrolled in mainstream schools andpercentage of children with disabilities enrolled in special schools it is likely that reliable data does not exist. It may also be the case that data is only partially available. For instance, often, statistics on enrolment rates will be available for the country but will not be disaggregated by region or by income. Or it may be that you have data but it is not of sufficient quality, eg it is too old. In such cases, you should consider whether you can collect your own data. For an example of CSOs collecting their own data, seehere.

If you decide to collect your own data, you should assess whether you have the necessary expertise to ensure that the data you collect is reliable and credible. If not, you should think about seeking professional help (see Step 1.4).