UNESCO Institute for Statistics (UIS) indicator `READ.LOWERSEC.WPIA` measures adjusted wealth parity in the proportion of students reaching at least minimum reading proficiency at the end of lower secondary education. It is not a reading score and not the minimum-proficiency percentage itself. The indicator compares the underlying proficiency rate for the poorest wealth quintile with the rate for the richest wealth quintile and expresses the disparity on an adjusted 0–2 parity scale. A value of 1 is parity, and 0.97–1.03 is commonly treated as the parity band.
The 2022 dataset contains 77 official country and territory observations. Every one of them is below 0.97; none falls inside or above the parity band. China, Macao Special Administrative Region has the highest value at 0.91673, followed by China, Hong Kong Special Administrative Region at 0.86575 and Estonia at 0.80004. Cambodia is lowest at 0.18578, followed by the Dominican Republic at 0.24371 and the Philippines at 0.26979. The mean is 0.526 and the median 0.526.
Table of Contents
What the adjusted wealth parity index compares
The wealth dimension compares the poorest and richest quintiles of the relevant population. For this reading indicator, the underlying measure is the percentage of students at the end of lower secondary education who achieve at least the minimum proficiency level in reading. The adjusted wealth parity index summarizes how similar those two wealth-group percentages are.
Because minimum proficiency is an outcome where higher values are preferable, a parity value below 1 indicates disparity in favor of the wealthier group. A value above 1 would indicate the opposite direction after adjustment. All 77 observed values are below 1, so the published cross-section consistently points toward an advantage for the richest quintile relative to the poorest quintile.
Parity does not mean high reading achievement
A WPIA of 1 would mean the poorest and richest quintiles have the same underlying proficiency rate. It would not tell us whether that shared rate is high or low. Two groups can be equally low-performing and still be at parity, or they can both have very high proficiency and also be at parity.
This distinction makes WPIA an equity measure rather than an achievement-level measure. To judge actual learning levels, the sex- or wealth-specific proficiency percentages, or the overall national proficiency rate, are needed in addition to the parity index.
None of the 77 observations reaches the parity band
UIS adjusted parity measures are commonly interpreted with 0.97–1.03 as a band close to parity. In this dataset, all 77 observations are below 0.97. There are no observations in the parity band and no observations above 1.03.
The result does not mean the size of the gap is the same everywhere. Values range from 0.18578 to 0.91673. The entire observed sample lies on the same side of parity, but the distance from parity varies enormously across countries and territories.
Macao SAR is closest to parity in the observed data
China, Macao Special Administrative Region has the highest WPIA at 0.91673. China, Hong Kong Special Administrative Region follows at 0.86575, Estonia at 0.80004, Ireland at 0.78564, and the United Kingdom of Great Britain and Northern Ireland at 0.78151.
Only 3 observations reach 0.80 or higher, and 14 reach 0.70 or higher. Even the highest observation remains below the 0.97 lower boundary of the parity band. This shows that measurable wealth-related disparity remains even among the countries closest to parity in the available 2022 data.
| Country or territory | 2022 WPIA |
|---|---|
| China, Macao Special Administrative Region | 0.91673 |
| China, Hong Kong Special Administrative Region | 0.86575 |
| Estonia | 0.80004 |
| Ireland | 0.78564 |
| United Kingdom of Great Britain and Northern Ireland | 0.78151 |
| Japan | 0.77001 |
| Singapore | 0.75288 |
| Viet Nam | 0.74865 |
| United States of America | 0.74720 |
| Türkiye | 0.71397 |
Cambodia is farthest from parity
Cambodia has the lowest value at 0.18578, followed by the Dominican Republic at 0.24371, the Philippines at 0.26979, Jordan at 0.27032, El Salvador at 0.27149, Mongolia at 0.27892, North Macedonia at 0.29364, Bulgaria at 0.29743, Morocco at 0.30297, and Indonesia at 0.30898.
A value close to zero indicates a large disparity in the direction favoring the richest quintile. But 0.20 does not mean the poorest-quintile proficiency rate is 20%. The index is derived from a comparison between two subgroup rates, and the original percentages are needed to recover the actual proficiency levels.
| Country or territory | 2022 WPIA |
|---|---|
| Cambodia | 0.18578 |
| Dominican Republic | 0.24371 |
| Philippines | 0.26979 |
| Jordan | 0.27032 |
| El Salvador | 0.27149 |
| Mongolia | 0.27892 |
| North Macedonia | 0.29364 |
| Bulgaria | 0.29743 |
| Morocco | 0.30298 |
| Indonesia | 0.30898 |
The mean and median are both about 0.526
The simple mean is 0.526 and the median 0.526. The first quartile is 0.417 and the third quartile 0.660. The middle half of the observed countries and territories therefore falls between roughly 0.417 and 0.660.
The mean and median being almost identical suggests that the center of the observed distribution is not driven by just a few extreme cases. The more important fact is the distance from parity: even the median observation is only a little above 0.5, well below the reference value of 1.
Many observations are below 0.5
There are 17 observations below 0.40 and 37 below 0.50. Another 26 lie between 0.50 and under 0.70, while only 14 are at 0.70 or higher. These bands are descriptive groupings for this article, not official UIS performance thresholds.
The official interpretive reference remains parity at 1 and the 0.97–1.03 parity band. Arbitrary cutoffs such as 0.5 should not be turned into pass/fail categories. They are useful only for showing how the observed distribution is spread.
Why the adjusted index uses a 0–2 scale
A conventional parity index divides the value for the likely disadvantaged group by the value for the likely advantaged group. That ratio is easy to read below 1, but it becomes asymmetric and unbounded above 1. UIS therefore applies an adjustment when the ratio exceeds 1 so that the scale is symmetric around 1 and bounded between 0 and 2.
This also means adjusted values above 1 should not be read as literal multiples. The index is designed to show direction and distance from parity. The current dataset happens to contain no values above 1, but the adjusted-index identity is still essential to interpreting the indicator correctly.
The underlying outcome is minimum reading proficiency
The base learning indicator asks whether students at the end of lower secondary education achieve at least a minimum proficiency level in reading. That makes this series a learning-equity indicator rather than an enrolment or completion indicator.
A country can have high lower-secondary completion but still show a large wealth gap in reading proficiency. Conversely, participation and completion rates can be lower while the assessed subgroup displays a smaller wealth gap. Access, progression, learning, and equity are separate dimensions of education systems.
The index cannot explain why the wealth gap exists
A low WPIA establishes the presence of a large wealth-related disparity but does not identify its cause. Household resources, school quality, location, language, digital access, tutoring, health, nutrition, and many other factors may be associated with learning gaps.
Causal explanations require additional student- and school-level evidence. The current dataset should therefore be used to describe the observed pattern of equity rather than to assign a specific cause to any country’s index.
Country observations are not weighted by student population
Each country or territory appears as one national observation regardless of the size of its student population. The simple mean reported here is therefore an unweighted average of the 77 country-level indices, not a global student-weighted wealth-parity estimate.
Coverage is also incomplete. The official 2022 data contain 77 national observations, while many catalogue units are not reported. Missing countries remain missing and must not be treated as zero, because zero would represent an observed extreme value rather than absence of data.
A single year cannot establish improvement or decline
The 2022 cross-section cannot show whether a country is moving toward parity or away from it. It also cannot tell whether a particular assessment cycle or change in coverage influenced the observed value.
A trend analysis should follow the same `READ.LOWERSEC.WPIA` indicator across comparable years and, where possible, pair the index with the underlying poorest- and richest-quintile proficiency rates. That combination would show both the size of the equity gap and the absolute learning level of each group.
Official source and interpretation
The official source is the UNESCO Institute for Statistics. This comparison uses 77 NATIONAL observations for 2022 from indicator `READ.LOWERSEC.WPIA`. UIS defines adjusted parity indices by comparing the likely disadvantaged group with the advantaged group and transforming ratios above 1 so the index is symmetric around 1. The official parity methodology is described at UNESCO UIS parity indices, and the indicator values are available through the UNESCO UIS Data API.
The central interpretation is that WPIA measures wealth equity in a learning outcome, not reading proficiency itself. A value of 1 represents parity, values below 1 point toward an advantage for the richest quintile for this positive outcome, and countries without observations remain missing rather than being assigned zero.
Frequently Asked Questions
What does a WPIA of 1 mean?
It means the underlying indicator values for the poorest and richest wealth groups are at parity; 0.97–1.03 is the commonly used parity band.
Does a WPIA of 0.5 mean the poorest group has 50% reading proficiency?
No. WPIA is an adjusted parity index derived from two subgroup proficiency rates, not the proficiency percentage itself.
Did any of the 77 observations reach parity in 2022?
No. All 77 official observations are below 0.97.
Can missing countries be treated as zero?
No. Missing source observations remain missing and must not be zero-filled.
Related Articles
Green Map creates custom-edited map images using open geographic data sources such as geoBoundaries, Natural Earth, OpenStreetMap, and government open data. These maps are edited visual materials, not raw data files, and are provided for education, documents, presentations, and graphic reference.
