On average across countries with true cohort data, 72% of students who enter upper secondary education graduate from any programme within its theoretical duration. Two years after the end of the theoretical duration, average completion increases to 81%. For countries with cross cohort data, the average completion rate is 83%.
The completion rate (within the theoretical duration of the programme plus two years) of students in a general upper secondary programme (86%) is higher than for students in a vocational one (70%), on average across countries with data.
In all countries with available data, women have higher completion rates than men in upper secondary education. The gender gap decreases with time, as men take longer to complete their programmes.
Education at a Glance 2020
Indicator B3. Who is expected to complete upper secondary education?
Highlights
Context
Upper secondary completion rates measure how many of the students who enter an upper secondary programme ultimately graduate from it. One of the challenges facing education systems in many countries is students disengaging and consequent dropping out of the education system, meaning that they leave school without an upper secondary qualification. These young people tend to face severe difficulties entering – and remaining in – the labour market. Leaving school early is a problem for both individuals and society. Graduating with excessive delays is another source of concern, raising the issue of a later entry into the labour market and hence delaying the time when they are typically able to start contributing financially to society.
This indicator is restricted to initial education only, meaning it only captures students who are entering upper secondary education for the first time. For those students, it measures the successful completion of upper secondary programmes and the proportion of students still in education at two specific points: 1) the theoretical duration of the programme they entered; and 2) two years after the end of the theoretical duration. The difference between these two time frames sheds light on the extent to which students tend to graduate “on time” (or within the amount of time expected given the theoretical duration of the programme). It also allows completion rates by gender and programme orientation to be compared.
Like graduation rates, completion rates do not indicate the quality of upper secondary education; they do, however indicate to some extent the capacity of this education level to engage students to complete their programmes within a specific period.
Other findings
For nearly all countries, the completion rate is higher for upper secondary general programmes than for vocational ones, within the theoretical duration. In Estonia and Norway, the difference in completion rates is at least 30 percentage points higher for general programmes than for vocational ones.
On average, 3% of students who enter an upper secondary general programme are still in education two years after the end of the theoretical duration of the programme and 12% have not graduated and are no longer enrolled.
In some countries and economies, upper secondary students transfer between programme orientations before graduating, meaning that they could graduate from a programme orientation that is different from the one they entered. In the Flemish Community of Belgium, Chile and Iceland, at least 10% of students who enter an upper secondary general programme graduate from a vocational one. Similarly, in Brazil, Iceland, Israel and Norway, at least 10% of upper secondary students graduate from a general programme after entering a vocational one.
In all countries with available data, the completion rate of students who entered a general programme is higher than that of students who entered either type of vocational programme.
Figure B3.1. Distribution of upper secondary students by their status at the end of the theoretical duration of their programme and two years later
True cohort only
1. Students enter a general upper secondary education programme and only split into general and vocational programmes after one or more academic years
2. Year of reference 2017.
3. The data refer only to the Flemish Community.
4. Year of reference 2013 for the theoretical duration and 2015 for the theoretical duration plus two years.
Countries are ranked in descending order of the share of students who graduated from any programme.
Source: OECD (2020), Table B3.2. See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en)
Note
Completion, graduation and attainment rates are three different measures. Completion describes the percentage of students who enter an upper secondary programme for the first time and graduate from it a given number of years after they entered. The restriction to first-time entrants to upper secondary education means that adult-education programmes and students entering upper secondary education again after their initial schooling are excluded. For example, students who enter a vocational upper secondary programme after having completed a general upper secondary programme are not captured by this indicator.
This measure of upper secondary completion rates should not be confused with upper secondary graduation rates. The graduation rate represents the estimated percentage of people from a certain age cohort who are expected to graduate at some point during their lifetime. It measures the number of graduates from upper secondary education relative to the country’s population. For each country, for a given year, the number of students who graduate is broken down into age groups (for example, the number of 16-year-old graduates divided by the total number of 16-year-olds in the country). The overall graduation rate is the sum of these age-specific graduation rates.
A third indicator in Education at a Glance uses the notion of educational attainment (see Indicator A1). Attainment measures the percentage of a population who have reached a certain level of education, in this case those who graduated at least from upper secondary education. It represents the relationship between all graduates (in the given year and previous years) and the total population.
Analysis
Completion rates for true cohort and cross cohort data
Completion rates are calculated using two different methods, depending on data availability. The first method, true cohort, follows individual students from entry into an upper secondary programme until a specified number of years later. Completion is then calculated as the share of entrants who have graduated in that time frame. The second method, cross cohort, is used when data on individuals are not available. It calculates completion by dividing the number of graduates in a year by the number of new entrants to that programme a certain number of years before, where the number of years corresponds to the theoretical duration of the programme.
Because of the difference in methodologies, caution must be exercised when comparing true cohort and cross cohort completion rates. On the one hand, countries with true cohort data are able to report exactly how many students from a given entry cohort have graduated within a specific time frame. That means that the true cohort completion rate includes students who graduated before or exactly at the end of the time frame (even if they graduated from a different upper secondary programme than the one they began) and excludes students who took longer than the time frame to graduate.
On the other hand, the number of graduates used in the cross cohort calculation corresponds to the total number of graduates of an upper secondary programme in a given calendar year. Thus, it includes every student who graduated that year, regardless of the time they took to successfully complete the programme. As an example, consider a programme with a theoretical duration of three years. Completion rates will then be calculated using the graduation cohort in 2018 and an entry cohort two academic years earlier, in 2015/16. For countries with cross cohort data, the graduation cohort in 2018 will include students who entered in 2015/16 and graduated on time (within three years) as well as all others who entered before 2015/16 and graduated in 2018. As a result, in countries where a significant share of students take longer to graduate, the cross cohort method will overestimate completion rates compared to the true cohort method, for which the time frame is limited.
The theoretical duration of upper secondary programmes may vary across countries. Therefore, despite having the same reference year for graduates (2018 unless specified otherwise), the year used for the entry cohort differs across countries. Please see Annex 3 (https://doi.org/10.1787/69096873-en) for more information on each country’s theoretical duration of upper secondary programmes.
True cohort completion
On average across countries and economies with true cohort data, 72% of students who enter upper secondary education graduate from any programme within the theoretical duration of the programme. Two years after the end of the theoretical duration, the average completion rate increases to 81%. The completion rate increases between the theoretical duration and two years on, but for some countries and economies the increase is substantial. Notably, the completion rate at this level increases by at least 15 percentage points in Austria, the Flemish Community of Belgium, France, Norway, Portugal and Switzerland (Table B3.1).
A significant difference in completion rates between the shorter and longer time frames is not necessarily a negative outcome. It could reflect a more flexible upper secondary system, where it is common for students to transfer between different programmes or programme orientations, thus delaying their graduation. In the Flemish Community of Belgium, for example, 19% of students who enter a general upper secondary programme graduate instead from a vocational programme within the theoretical duration of their original programme plus two years. In Iceland and Norway, the opposite pathway is more common: more than 20% of students who enter a vocational programme transfer and graduate instead from a general programme (Table B3.2).
More generally, in countries that provide broad access to upper secondary education, flexibility may be important to give students more time to meet the standards set by their educational institution. In countries where upper secondary education is restricted either by admissions criteria or because students from disadvantaged backgrounds have less access to this level, completion rates may be higher because of the selection bias.
Nevertheless, students who graduate after excessive delays, or who leave the system without graduating are indeed a source of concern. Analysing how many of the students who are still in education by the theoretical duration leave the education system within the following two years may shed light on whether these students are delayed because of system characteristics or because they are falling behind and at risk of dropping out.
On average across countries and economies with available data, 51% of students who entered an upper secondary programme have graduated from a general programme and 24% from a vocational programme by the end of its theoretical duration. About 16% were still in education (even if at a different level) and 12% were no longer enrolled and had not graduated from any upper secondary programme. The picture evolves quite considerably two years after the end of the theoretical duration of the programme, as many of those who were still in education either graduate or leave the system. At this point, on average, 55% of students have graduated from a general programme and 28% from a vocational programme. Some 3% are still in education and 16% are no longer enrolled and have not graduated (Figure B3.1).
Cross cohort completion
Cross cohort completion rates take into account all graduates in a given academic year, regardless of the time it took them to complete the programme. As a result, cross cohort completion rates tend to be considerably higher than true cohort completion rates. Although they cannot be used to assess whether students are graduating with excessive delays, cross cohort completion provides valuable information on the share of students who are not graduating at all.
On average across the eight countries that submitted cross cohort data, 83% of students who enter an upper secondary programme complete it. There is, however, a wide variation among countries, ranging from 65% in Costa Rica to 96% in Korea (Table B3.1).
The completion rate pattern by programme orientation shows that in most countries with available data, cross cohort completion is higher in general programmes than in vocational programmes, except in the Slovak Republic (Figure B3.2). On average, the cross cohort completion rate is 10 percentage points higher for general programmes, ranging from 2 percentage points in Japan to 22 percentage points in Spain (Table B3.1).
Figure B3.2. Cross cohort completion of upper secondary education by programme orientation at graduation (2018)
Countries are ranked in descending order of the general programme completion rate.
Source: OECD 2020, Table B3.1. Ad-hoc survey on completion rates. See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
Completion rate by programme orientation
The flexibility to transfer between upper secondary programmes is important to ensure that students do not get locked into a programme that does not reflect their interest or ability. However, in most countries with true cohort data, students tend to graduate from the programme they entered: 73% of entrants to upper secondary general programme graduate from the same programme and 4% graduate from a vocational programme within the theoretical duration. Similarly, 58% of entrants to upper secondary vocational programmes graduate from the same programme and only 4% have gained a general qualification within the theoretical duration.
In all countries with true cohort data, except Israel and Switzerland, the completion rate within the theoretical duration for students who enter a general upper secondary programme is higher than for students who enter a vocational one. On average across countries with true cohort data, the completion rate for general programmes within the theoretical duration is 76%, compared to 62% for vocational programmes. In Estonia and Norway, the completion rate for general programmes is at least 30 percentage points higher than that for vocational programmes. The completion rates of vocational programmes within the theoretical duration range from 41% in Iceland to 94% in Israel. For countries with cross cohort data, the figures range from 53% in Costa Rica to 93% in Japan and Korea (Table B3.1).
In most countries, the difference in completion rates between the two orientations does not change significantly after two years following the end of the theoretical duration. One notable exception is Norway, where the gap reduces by 12 percentage points between the shorter and longer time frames. In contrast, the gap actually increases by 10 percentage points in France and by 17 percentage points in Portugal as the completion rate for general programmes increases considerably more than that of vocational programmes during the two years after the end of the theoretical duration (Table B3.1).
For the first time, the ad-hoc survey on upper secondary completion rates disaggregates vocational programmes into those which give access to tertiary education and without access to tertiary education (but may give direct access to post-secondary non-tertiary education). This further disaggregation is meant to shed light on the different pathways through upper secondary education but also on the differences in completion rates between these vocational programmes.
Figure B3.3. Completion rate of upper secondary education within the theoretical duration plus two years, by programme orientation at entrance (2018)
1. Year of reference 2013 for the theoretical duration and 2015 for the theoretical duration plus two years.
2. Year of reference 2017.
3. The data only refer to the Flemish Community.
4. The standard errors are included when data are provided through a survey.
Countries are ranked in descending order of completion rate of students who entered a general programme (for true cohort, by the theoretical duration plus two years).
Source: OECD (2020). See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
The entry into general upper secondary programmes may be subject to stricter admission criteria than into vocational ones (Box B3.1).In all countries and economies with available data, the completion rate of students who entered a general programme is higher than that of students who entered either type of vocational programme (with or without access to tertiary education). However, many countries have substantial differences in completion rates between vocational programmes. In the Flemish Community (Belgium), France, Italy, Latvia and Switzerland, students who entered a vocational programme without access to tertiary education are considerably less likely to complete upper secondary education than those who entered one with access to tertiary education. In contrast, the difference in completion rates between vocational programmes is low in Austria (Figure B3.3).
Among countries with cross cohort data, completion is also higher for general programmes than for vocational programmes. The average completion rate for general programmes is 87%, compared to 77% for vocational ones. The largest difference is found in Spain, where the completion rate for general programmes is 22 percentage points higher than for vocational programmes. One exception is the Slovak Republic, where completion is higher in vocational programmes than in general ones (Table B3.1).
As many countries aim to develop their upper secondary vocational programmes in the hope of better preparing students for the labour market, the comparatively lower completion rate for these programmes is concerning. It highlights the challenge faced by educators and policy makers alike of not only attracting students to vocational tracks, but also of supporting them through successful completion. Some countries have been successful in considerably increasing completion rates in vocational programmes and diminishing the gap between vocational and general programmes, however (Box B3.2). It is important to note, however, that there is a wide variation in size, duration and even completion rates of vocational programmes across countries.
Box B3.1. Transition between lower and upper secondary education
The transition from lower secondary to upper secondary education is an important step in students’ academic trajectory. Ensuring a smooth transition helps to foster higher achievement gains and prevent students from dropping out (OECD, 2011[1]).
After having completed lower secondary education, students face many options: transition courses between lower and upper secondary level, basic vocational education, general education courses or no immediate entry into upper secondary level. Whatever the programme orientation chosen, the rate of immediate transition to upper secondary level shows the percentage of students who complete lower secondary education and start upper secondary education straight away. This rate varies widely across countries with available data, reaching almost 100% in Belgium, Japan, Korea, Latvia and Slovenia (metadata questions on policies and system characteristics). However, lower rates of immediate transition are not necessarily a negative outcome; they could reflect a more flexible educational system in which it is common for students to re-enter education later on, thus delaying their entry into upper secondary education.
In most countries with available data, successful completion of lower secondary level is sufficient to give students access to upper secondary education. However, when a national end-of-year examination is required to move from one level to another, some education authorities have introduced stricter conditions for entry into general upper secondary programmes than into vocational ones. In Norway, students must pass an exam with a minimum score to enter a general programme, whereas they need an educational agreement with a company to enter a vocational pathway. In Iceland, longer general programmes require higher grades than short ones but no specific exam is needed to enter upper secondary education.
Overall, countries have a distinct structure for lower and upper secondary education. Some studies have highlighted the benefits of combining primary, lower and upper secondary education in terms of school belonging (OECD, 2011[1]), which in turn is inversely related to depression, social rejection and school problems (Anderman, 2002[2]). However, there is no one-size-fits-all solution, and [...] the choice of a particular pattern depends on the characteristics of the national educational system.
Completion rate by gender
In every country with available data (both true and cross cohort), women are more likely than men to complete upper secondary education, both within the theoretical duration and two years after (Table B3.1). On average across countries and economies with true cohort data, 76% of women graduated from upper secondary education within the theoretical duration of the programme, compared to only 68% of men. The difference in completion rates between women and men by the theoretical duration is at least 11 percentage points in the Flemish Community of Belgium, Iceland, Italy, Norway and Portugal. The gender gap narrows two years after the theoretical duration of the programme, when the completion rate among women increases to 84% and among men to 78% (Table B3.1).
In all countries and economies except Finland and Sweden, the gap in completion rate between men and women narrows or remains the same within the two years after the end of the theoretical duration of programmes, meaning relatively more men tend to delay graduation. Many factors may contribute to this delay, one of which is the higher incidence of grade repetition among men, who are more likely than women to repeat a grade even after accounting for students’ academic performance and self-reported behaviour and attitudes (OECD, 2016[3])
The difference between upper secondary completion rates for women and men tends to be smaller among countries with cross cohort data. On average, the completion rate for women is 4 percentage points higher than for men, and the difference reaches 8 percentage points in Slovenia and Spain.
The gender gap also varies considerably depending on the programme orientation at entrance. In all countries with true cohort data, the completion rate of women is higher than that of men, whatever their programme orientation (Figure B3.4), except in Lithuania for students who entered a vocational programme. While the gender gap in favour of women tends to be similar for students entering a general or vocational programme (7 percentage points) within the theoretical duration plus two years, the completion rate of men in vocational programmes is equal or significantly higher than that of women in some countries (Table B3.1).
Figure B3.4. Completion rate of upper secondary education within the theoretical duration plus two years, by gender and programme orientation at entrance (2018)
1. Year of reference 2013 for the theoretical duration and 2015 for the theoretical duration plus two years.
2. Year of reference 2017.
3. The data only refer to the Flemish Community.
4. The standard errors are included when data are provided through a survey.
Countries are ranked in descending order of the completion rates of women who entered a general programme.
Source: OECD (2020). OECD (2020), Table B3.1. See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
Box B3.2. Trends in completion rates of upper secondary education, by programme orientation, within the theoretical duration plus two years
Increasing the number of students who complete upper secondary education is a priority for many education policy makers. However, this is a challenging goal, which may require changes at the system, school and classroom levels. Figure B3.5. Trends in completion rates of upper secondary education within the theoretical duration plus two years, by programme orientation shows trends in completion rates broken down by programme orientation. The reference years used for the trend comparison in each country vary according to data availability (as indicated below the country’s name on the horizontal axis), and therefore cross-country comparisons cannot be made from these data.
It is, however, possible to observe that the Flemish Community (Belgium), Finland, France and Norway have been able to increase completion rates over recent years for both general and vocational programmes in upper secondary education. In all four countries, the completion rate for vocational programmes has increased by more than for general programmes. In France, the total upper secondary completion rate increased by 8 percentage points between 2005 and 2017, driven mostly by an increase of 10 percentage points in the completion rate for vocational programmes. A sharp increase in completion rates for vocational programmes can also be observed in the Flemish Community of Belgium and Finland from 2010 to 2017 and in Norway between 2012 and 2018. In the Netherlands and Estonia, an increase in completion rates for vocational programmes was accompanied by a decrease in completion rates for general programmes.
In Sweden, an upper secondary school reform in 2011 may help explain the trend between 2010 and 2018. This has meant, among other things, that higher demands have been introduced for completion/graduation.
Figure B3.5. Trends in completion rates of upper secondary education within the theoretical duration plus two years, by programme orientation
How to read this figure: In France, the completion rate for total upper secondary education increased by 8 percentage points from 2005 to 2017. In Sweden, it decreased by 3 percentage points from 2010 to 2018.
Note: Completion rate by the theoretical duration of the programme plus two years.
1. The data only refer to the Flemish Community.
Countries are ranked in descending order of the percentage-point change in completion rates of upper secondary programmes.
Source: OECD 2020 ad-hoc survey on completion rates. See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
Definitions
The true cohort method requires following an entry cohort through a specific time frame, which in the case of this survey corresponds to the theoretical duration N and the theoretical duration plus two years (N+2). Only countries with longitudinal surveys or registers are able to provide such information. Panel data can be available in the form of an individual student registry (a system including unique personal ID numbers for students) or a cohort of students used for conducting a longitudinal survey.
The cross cohort method only requires the number of new entrants to a given ISCED level and the number of graduates N years later, where N corresponds to the theoretical duration of the programme. Under the assumption of constant student flows (constant increase or decrease in the number of students entering a given ISCED level throughout the years), the cross cohort completion is closer to a total completion rate (i.e. the completion rate of all students, regardless of the time it took them to graduate). As such, in countries where a large share of students do not graduate “on time” given the theoretical duration of the programme, the cross cohort completion may be more comparable to longer time frames of the true cohort completion.
The theoretical duration of studies is the regulatory or common-practice time it takes a full-time student to complete a level of education. True cohort completion is measured within two time frames: by the end of the theoretical duration and by the end of the theoretical duration plus two years. The theoretical duration always refers to the programme in which the student originally entered upper secondary education. This means that even if a student transfers to a different programme with a different duration they will still be registered according to the programme in which they originally entered the level. Please see Annex 3 (https://doi.org/10.1787/69096873-en) for information on each country’s theoretical duration for general and vocational upper secondary programmes.
The programme orientation can refer either to the programme in which the student originally entered upper secondary education or to the programme from which the student graduated. Both types of analysis are included in the indicator. The titles, subtitles or axis titles of the figures (and tables) will clarify which programme is being disaggregated by programme orientation. Only programmes sufficient for level completion are included. Four programme orientations are considered in the analysis:
general programmes (ISCED-P 343 and 344)
vocational programmes without access to tertiary education (ISCED-P 353)
vocational programmes with access to tertiary education (ISCED-P 354)
combined vocational programmes (ISCED-P 353 and 354).
The reference year for the survey is 2018 and refers to the academic year 2017/18 in countries where the academic year runs from Sept-June. For countries submitting true cohort data, the reference year should be two years after the end of the theoretical duration of the programme. For example, if a programme has a duration of two years, the cohort reported must have entered upper secondary education in the academic year 2014/15. Their status is then recorded by the end of the theoretical duration of the programme (academic year 2015/16) and two years later (academic year 2017/18). For countries submitting cross cohort data, the year of reference corresponds to the reference year for the graduate data. Reference years that differ from 2018 will be clearly indicated throughout the indicator (even if not noted below the charts in this paper).
Methodology
Data on completion rates refer to the academic year 2017/18 and were collected through a special survey undertaken in 2019. Countries could submit data either using either the true cohort or cross cohort methodology.
The completion rate for both methods is calculated as the number of graduates divided by the number of entrants N or N+2 years before (where N is the theoretical duration of the programme).
For countries that submit true cohort data it is also possible to calculate the share of students still in education and the share of students who have neither graduated nor are still enrolled – all of which is calculated within the timeframes of N and N+2. Both shares are calculated by dividing the number of students in the given situation by the number of new entrants N or N+2 years before.
References
[2] Anderman, E. (2002), “School effects on psychological outcomes during adolescence”, Journal of Educational Psychology, Vol. 94/4, pp. 795-809, http://dx.doi.org/10.1037/0022-0663.94.4.795.
[4] CEDEFOP (2011), The Benefits of Vocational Education and Training, Publication Office of the European Union, Luxembourg.
[3] OECD (2016), Education at a Glance 2016, OECD Publishing, Paris, https://dx.doi.org/10.1787/eag-2016-en.
[1] OECD (2011), Reviews of National Policies for Education, Improving Lower Secondary Schools in Norway, Reviews of National Policies for Education, OECD Publishing, Paris, https://doi.org/10.1787/9789264114579-en.
Indicator B3 Tables
Table B3.1 Completion rate of upper secondary education, by programme orientation at entrance and gender (2018)
Table B3.2 Distribution of entrants to upper secondary education by programme orientation and outcomes after the theoretical duration and after the theoretical duration plus two years (2018)
Cut-off date for the data: July 19th, 2020. Any updates on data can be found on line at http://dx.doi.org/10.1787/eag-data-en. More breakdowns can also be found at http://stats.oecd.org/, Education at a Glance Database.
StatLink: https://doi.org/10.1787/888934163496
Table B3.1. Completion rate of upper secondary education, by programme orientation at entrance and gender (2018)
Completion rate of full-time students who graduated from any programme
Note: The data presented in this table come from an ad-hoc survey and only concern initial education programmes. For true cohorts, the reference year (2018, unless noted otherwise) refers to the year of graduation by the theoretical duration plus two years. See Definitions and Methodology sections for more information.
1. Year of reference 2017.
2. Students enter a general upper secondary education programme and only split into general and vocational programmes after one or more academic years.
3. Year of reference 2013 for the theoretical duration and 2015 for the theoretical duration plus two years.
Source: OECD/UIS/Eurostat (2020). See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
Please refer to the Reader's Guide for information concerning symbols for missing data and abbreviations.
Table B3.2. Distribution of entrants to upper secondary education by programme orientation and outcomes after the theoretical duration and after the theoretical duration plus two years (2018)
True cohort only
Note: The data presented in this table come from an ad-hoc survey and only concern initial education programmes. See Definitions and Methodology sections for more information.
The columns for "not graduated and not enrolled" may include students who left the country before graduation.
Students who continued their studies in the adult-education system are included in the columns for “not graduated and not enrolled”
1. Year of reference 2017.
2. Students enter a general upper secondary education programme and only split into general and vocational programmes after one or more academic years.
3. Year of reference 2013 for the theoretical duration and 2015 for the theoretical duration plus two years.
Source: OECD/UIS/Eurostat (2020). See Source section for more information and Annex 3 for notes (https://doi.org/10.1787/69096873-en).
Please refer to the Reader's Guide for information concerning symbols for missing data and abbreviations.