Survival to Grade 4 of primary education ranged from 18.59% to 99.85% among 62 reporting countries and territories in 2023. The UNESCO Institute for Statistics (UIS) records 99.85% for Mexico, 99.82% for Chile and 99.81% for Bulgaria. The lowest available values are 18.59% for Palestine and 28.26% for the British Virgin Islands. These figures cover boys and girls together.
The indicator estimates how many pupils who entered the first grade of primary education are expected to reach the fourth grade. It is not a test score, a four-year graduation rate or necessarily the survival rate to the final grade of primary school. Primary cycles have different lengths in different systems. This distinction matters when comparing territories that follow different school structures.
Across these 62 observations, the median is 97.12% while the unweighted mean is 91.75%. Thirty-nine reporting places reach at least 95%, and 49 reach at least 90%. The coverage is deliberately restricted to records with an available 2023 observation; it should not be mistaken for a complete ranking of the world’s countries.

Table of Contents
What survival to Grade 4 actually measures
UIS indicator SR.1.G4.CP concerns the proportion of a first-grade school cohort expected to progress to the fourth grade of primary education. The value is presented as a percentage of the original cohort. Because the series combines both sexes, it cannot on its own answer whether boys or girls have a higher likelihood of progressing through those grades.
For a practical assessment of schooling access, this survival measure should be read alongside a separate indicator of entry into the first grade. Strong continuation among pupils who have enrolled does not establish that children who never enrolled had an opportunity to attend. Entry and subsequent progression describe different points in the education pathway and may call for different responses.
UIS describes a reconstructed-cohort method: information on enrolment and grade repetition for two successive school years is used to infer how pupils move from one grade to the next. It is a statistical representation of progression, not necessarily a longitudinal record tracking each individual student from the first day of school to Grade 4.
A value of 98% means that, within the indicator’s assumptions, roughly 98 of every 100 first-grade entrants are expected to reach Grade 4. It does not demonstrate that those pupils passed fourth-grade examinations, that every child in the population enrolled initially, or that their reading and mathematics performance is strong. Enrolment, progression, completion and learning outcomes are separate measurements.
Which reporting places have the highest values?
Mexico leads the supplied set with 99.85%. Chile follows at 99.82% and Bulgaria at 99.81%. Macao and Poland each round to 99.79%, followed by Denmark at 99.78%. Slovakia reaches 99.74%, Estonia 99.72%, Switzerland 99.71%, Norway 99.69% and Finland 99.66%.
When many reported rates sit close to the upper boundary of the scale, a country ranking is a blunt tool for identifying remaining educational needs. A more informative next step may be to examine differences within each country or continuation in later grades, provided official measures with matching definitions and coverage are available.
These are extremely close results. The table orders places using the unrounded values, so two territories displaying the same two-decimal figure can still occupy different positions. A few hundredths of a percentage point cannot establish a meaningful difference in educational effectiveness without information on uncertainty, collection practices and wider indicators.
Oman records 99.65%, Austria 99.54%, Turkmenistan 99.47%, Paraguay 99.38% and Azerbaijan 99.34%. High values occur across distinct geographical settings. That observation describes the distribution of the submitted series; it does not establish that geography, national income or any particular school policy caused the measured outcome.
A rate close to 100% indicates that a reconstructed first-grade cohort is expected to retain almost all its members through the fourth grade. Even such a figure leaves unanswered how many children were never enrolled in the first place and whether those reaching the grade acquired the learning skills expected of them.
Why the low observations need careful interpretation
The two lowest figures, Palestine at 18.59% and the British Virgin Islands at 28.26%, are much lower than the central group. Niger is next at 62.61%, followed by Côte d’Ivoire at 68.16%, India at 74.87% and Togo at 75.29%. Four places fall below 70%, while six are below 80%.
Those contrasts justify looking more closely at each series but do not explain what happened to individual pupils. This table contains no separate measurements of school closures, household migration, curriculum rules, grade repetition policy or reporting coverage. Selecting one explanation from these possibilities would go beyond the evidence.
Smaller territories and places experiencing substantial student mobility may require special attention to school transfers and the way a statistical cohort is reconstructed. UIS itself cautions that consistency of coverage and enrolment and repeater reporting is important. Here the original observations have been retained; no attempt has been made to replace low readings with estimates or to reinterpret a missing value as zero.
For Palestine, the reference year is 2023, and the measure is a statistical cohort estimate. It would be inappropriate to attribute the entire numerical result to events after the reference period or to treat it as a count of known individual dropouts. The underlying administrative process would have to be studied before making any such claim.
Why the mean falls below the median
The median is 97.12%, obtained from the two central values after sorting all 62 observations. The arithmetic mean is 91.75%, about 5.37 percentage points lower. This occurs because the distribution contains a relatively small number of notably low values despite most reported places having high survival rates.
An unweighted average can move when a later release adds or removes reporting territories, even if every country shared by both releases remains unchanged. A sound trend check would distinguish a summary of all places available in each release from a comparison using only a consistently observed set. Otherwise a change in coverage can be mistaken for a change in school progression.
In fact, 49 of the 62 values reach 90% or more; 39 reach at least 95%; and 24 reach at least 98%. These counts give more information about the distribution than a single mean. They also show why the very low observations should not be portrayed as typical of every country in the selected set.
The average here weights each country or territory equally, regardless of the size of its entering first-grade cohort. India and a small island territory each contribute one observation. The result is therefore not a population-weighted rate for the world’s pupils. Obtaining such a rate would require comparable cohort sizes and sufficient geographic coverage.
The span between the highest and lowest values is 81.26 percentage points. A range highlights extremes but says nothing by itself about where most observations lie. Looking at the range together with the median and threshold counts gives a more reliable summary of this uneven distribution.
Patterns within broad regions
Several reporting countries in the Americas record high values: Mexico 99.85%, Chile 99.82%, Paraguay 99.38%, Bolivia 98.14% and Costa Rica 96.46%. Guatemala registers 94.06% and Honduras 89.01%. These comparisons illustrate variation within the continent, rather than a single rate that applies to every country in it.
The available Asian observations are also varied. Macao records 99.79%, Oman 99.65%, Azerbaijan 99.34%, the Philippines 96.70%, Malaysia 91.61% and India 74.87%. The Republic of Korea is 98.95%. Individual examples should not be combined into an Asian average unless an appropriate and comprehensive regional dataset is constructed.
Across the African places in this selection, Seychelles is 96.49%, Morocco 95.50% and Mauritius 94.05%, compared with Senegal at 81.90%, Cameroon at 80.53%, Côte d’Ivoire at 68.16% and Niger at 62.61%. The spread is visible even without making any claim about countries that are missing from the available records.
European observations include Bulgaria at 99.81%, Denmark 99.78%, Slovakia 99.74%, Switzerland 99.71%, Norway 99.69% and Austria 99.54%. Geographic proximity alone does not explain their differences. The figures cannot identify the effects of school administration or demographic composition without supplementary evidence.
Full table of the 62 reporting places
The table below contains every supplied 2023 value, sorted from highest to lowest. Three-letter identifiers distinguish territories that might otherwise have ambiguous names. Percentages are rounded to two decimals for display only; all calculations and ordering used the underlying full-precision numbers.
| Rank | Country / territory | Grade 4 survival rate, 2023 |
|---|---|---|
| 1 | Mexico (MEX) | 99.85% |
| 2 | Chile (CHL) | 99.82% |
| 3 | Bulgaria (BGR) | 99.81% |
| 4 | Macao (MAC) | 99.79% |
| 5 | Poland (POL) | 99.79% |
| 6 | Denmark (DNK) | 99.78% |
| 7 | Slovakia (SVK) | 99.74% |
| 8 | Estonia (EST) | 99.72% |
| 9 | Switzerland (CHE) | 99.71% |
| 10 | Norway (NOR) | 99.69% |
| 11 | Finland (FIN) | 99.66% |
| 12 | Oman (OMN) | 99.65% |
| 13 | Austria (AUT) | 99.54% |
| 14 | Turkmenistan (TKM) | 99.47% |
| 15 | Paraguay (PRY) | 99.38% |
| 16 | Azerbaijan (AZE) | 99.34% |
| 17 | Republic of Korea (KOR) | 98.95% |
| 18 | Israel (ISR) | 98.92% |
| 19 | Sweden (SWE) | 98.80% |
| 20 | Lithuania (LTU) | 98.62% |
| 21 | Georgia (GEO) | 98.39% |
| 22 | Hungary (HUN) | 98.34% |
| 23 | Bahrain (BHR) | 98.22% |
| 24 | Bolivia (BOL) | 98.14% |
| 25 | Hong Kong (HKG) | 97.87% |
| 26 | Belarus (BLR) | 97.86% |
| 27 | Dominican Republic (DOM) | 97.74% |
| 28 | Jordan (JOR) | 97.63% |
| 29 | Belize (BLZ) | 97.60% |
| 30 | Algeria (DZA) | 97.19% |
| 31 | San Marino (SMR) | 97.17% |
| 32 | Latvia (LVA) | 97.06% |
| 33 | Philippines (PHL) | 96.70% |
| 34 | Ecuador (ECU) | 96.55% |
| 35 | Seychelles (SYC) | 96.49% |
| 36 | Costa Rica (CRI) | 96.46% |
| 37 | Fiji (FJI) | 95.70% |
| 38 | Morocco (MAR) | 95.50% |
| 39 | Moldova (MDA) | 95.47% |
| 40 | Maldives (MDV) | 94.75% |
| 41 | Guatemala (GTM) | 94.06% |
| 42 | Mauritius (MUS) | 94.05% |
| 43 | El Salvador (SLV) | 93.64% |
| 44 | Ukraine (UKR) | 93.59% |
| 45 | Venezuela (VEN) | 93.56% |
| 46 | Syrian Arab Republic (SYR) | 92.97% |
| 47 | Cuba (CUB) | 92.49% |
| 48 | Malaysia (MYS) | 91.61% |
| 49 | Kiribati (KIR) | 90.65% |
| 50 | Albania (ALB) | 89.79% |
| 51 | Honduras (HND) | 89.01% |
| 52 | Cayman Islands (CYM) | 87.52% |
| 53 | Nauru (NRU) | 83.00% |
| 54 | Senegal (SEN) | 81.90% |
| 55 | Rwanda (RWA) | 81.27% |
| 56 | Cameroon (CMR) | 80.53% |
| 57 | Togo (TGO) | 75.29% |
| 58 | India (IND) | 74.87% |
| 59 | Côte d'Ivoire (CIV) | 68.16% |
| 60 | Niger (NER) | 62.61% |
| 61 | British Virgin Islands (VGB) | 28.26% |
| 62 | Palestine (PSE) | 18.59% |
What this snapshot cannot establish
A single year does not indicate whether Grade 4 survival improved or deteriorated. Assessing change requires equivalent observations across multiple years, together with checks for changes in definitions, school structures and reporting scope. It is not possible to attribute a shift in a school system to a reform using this one cross-section.
For monitoring purposes, it helps to keep four questions separate: who enters school, who advances between grades, who finishes primary education, and what pupils learn. This survival series contributes to the progression question only. Before recommending a policy response, the remaining questions would need their own appropriately defined observations rather than being inferred from this rate.
The combined-sexes indicator also gives no direct measure of the gender gap. Separate female and male observations for the same year and definition would be needed. Likewise, reaching Grade 4 is not equivalent to primary-school completion, progression to secondary education, or reaching minimum levels of reading and numeracy.
UIS notes that reconstructed cohort estimates rely on assumptions about pupil-flow rates, which may not remain unchanged for an actual cohort’s entire school life. Transfers, repetition and consistency of school enrolment statistics can complicate comparisons. It is particularly important not to read minute differences between high-rate countries as precise measurements of school quality.
There is also a coverage limit. The sample contains 62 reporting countries and territories, not a complete global census. Some territories may be absent because of missing or excluded observations. A missing row is neither a value of zero nor proof of poor survival. That boundary must be retained when interpreting ranks and averages.
Official source and calculations
The values come from the UNESCO UIS indicator SR.1.G4.CP for reference year 2023. The UIS definition of survival rate by grade explains the reconstructed-cohort calculation and the main comparability limitations.
For the table, each of the 62 numerical readings has been preserved in its original precision and displayed as a percentage with two decimal places. The mean is the sum divided by 62; the median is the average of the 31st and 32nd ordered readings; and the 81.26-point range is the maximum minus the minimum. Chart and table draw from precisely the same observations.
No non-reporting place was assigned a value, and no causal model or population-weighted global rate was generated. The appropriate conclusion is limited but useful: within the 2023 reporting set, progression to Grade 4 was estimated to be very high in most places, with several unusually low observations that call for separate contextual analysis.
Frequently Asked Questions
Which reporting country has the highest survival to Grade 4 in 2023?
Mexico records 99.85%, the highest value among the 62 available 2023 observations. This is not a complete worldwide ranking.
Does Grade 4 survival measure completion of primary school?
No. It estimates the share of a first-grade cohort expected to reach Grade 4. Passing final examinations and finishing the primary cycle are different outcomes.
Is 91.75% the global average for children?
No. It is an unweighted arithmetic mean for 62 reporting places; it is not weighted by the number of pupils in each place.
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.





