16
such as the OECD and UNESCO. This report too does not use a fully integrated measurement
framework.
Data and Methodological Framework
While centralized global development database such as WDI includes a rich set of indicators of
child health (e.g. stunting, wasting, under-weight and under-nourishment), education related
outcome indicators only relate to self-reported literacy rates. However, one exception is the
WIDE dataset which does not contain country-level information on poverty and income level.
To this end, a hybrid dataset that contains student learning data for a wide range of countries in
the world along with information on educational and economic development of the country has
been constructed. This data set is used primarily to describe OIC wide trends in learning
outcomes and input quality. In specific cases (e.g. for measures of accountability among
teachers), this has been complemented by data used in published studies.
Trend analysis of learning outcomes is primarily based on performance in PISA, TIMSS, SACMEQ,
PIRLS and EGRA and has a greater emphasis on secondary school students who participated in
TIMSS and PISA. In the absence of comparable data on learning outcomes for primary and pre-
primary education, EGRA and EGMA data is used to comment on learning levels in early grade.
This is completed by analysis based on PIRLS and TIMSS grade 4 which help assess learning level
among children in upper primary grades. For the vast majority of OIC countries, internationally
comparable data is not available. Discussion on these countries is based on input specific
indicators of quality such as PTR and proportion of trained teachers. Desk review of national
assessment of student performance is used to comment on education quality in these countries.
Since majority of the OIC countries don’t participate in any major international assessment of
learning outcomes, additionally data on youth literacy is used which is widely available for most
OIC countries. For these reasons, the measures of quality vary throughout the report based on
the underlying data source.
The following issues need to be kept in mind when interpreting findings of our descriptive trend
analysis at the country or region level.
First, most OIC countries with no comparable data on learning outcomes are lowor lowermiddle
income. Therefore, the report does not draw comparison of participating countries by income
groups. Instead, for comparison purposes, other non-OIC countries are grouped into OECD and
non-OCED countries. While the majority in the OIC sample has a high poverty rate, those
participating in TIMSS and PISA are middle or high income countries or aspiring to be high
income countries in the near future.
Many OIC countries have explicit targets to achieve OECD average scores in international
student assessments. Major national policy documents of OIC member states such as Saudi
Arabia’s Vision 2030, Jordan’s NCHRD 2016 report and Egypt’s Sustainable Development
Strategy have adopted indictors relating to achieving a certain performance benchmark in
international assessments such as TIMSS and PISA. For these reasons, despite some differences
in income, a comparison of OIC with OECD and non-OECD countries is meaningful.
Second, participation in TIMSS and PISA among OIC countries vary over time – some countries
joined late while some have withdrawn from the recent round of assessment. This again affects
trend analysis. Given the variation in participation rate, no attempt has been made to restrict
comparison to the same group of OIC countries in the analysis based on TIMSS and PISA.