eduKateSG · How Education Works · Comparative Education · Teacher Pay, Status and Working Conditions · 1 October 2026
Comparing teachers across countries looks simple until the measures are defined. Salary can mean a starting statutory scale, pay after fifteen years, top-of-scale pay or actual compensation. Work can mean classroom teaching hours, contractual working time or self-reported total hours. Status can mean social prestige, professional autonomy, qualification requirements, job satisfaction or whether teachers themselves believe society values their work. Put those measures into one ranking and the comparison becomes impressive-looking nonsense.
This guide owns WEP195 inside eduKateSG’s comparative-education estate. It complements Teacher Compensation & Salary Structures, Teacher Time and Teacher Retention. Its job is not to publish a teacher-country league table. It is to show how comparable evidence can reveal different professional architectures.
The main current sources are OECD’s newly released Education at a Glance 2026 salary chapter, which uses 2025 salary data with a print-data cut-off of 17 June 2026, the corresponding teacher time chapter, and TALIS 2024 for teacher-reported working conditions, status and satisfaction. These sources measure different things and should remain separate.
A few current examples show why the distinction matters. Education at a Glance 2026 reports an OECD-average statutory salary of USD 66,583 in PPP terms for a lower-secondary teacher with the most prevalent qualification after 15 years of experience. Germany is among the highest-salary systems in the available data. Korea shows a steep career progression: OECD reports lower-secondary statutory starting pay at about 24% below similarly educated workers, but pay after 15 years at about 29% above them. The United Kingdom and United States show large subnational variation, so one national headline number hides substantial internal differences. None of those facts establishes which system is ‘best’.
Teacher pay is not one number
Teacher pay is not one number is mainly about distinguishing starting salary, mid-career statutory salary, top-of-scale salary and actual earnings. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is comparing the same career stage and education level across systems. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is comparing one country’s starting salary with another country’s average actual salary. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the measure and career stage are explicit. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
PPP conversion solves one problem, not every problem
PPP conversion solves one problem, not every problem is mainly about using purchasing-power parity to improve cross-country comparability while retaining institutional differences. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is reading OECD equivalent-USD tables as purchasing-power comparisons rather than exchange-rate take-home pay. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating PPP dollars as local cash salary. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the reader knows what the converted figure represents. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Statutory salary and actual salary differ
Statutory salary and actual salary differ is mainly about separating official pay scales from observed compensation including bonuses or allowances. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is using statutory salary to study progression and actual salary to study realised earnings. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is mixing the two in one league table. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that comparisons use like measures. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Relative earnings matter for recruitment
Relative earnings matter for recruitment is mainly about comparing teacher pay with alternative earnings available to similarly educated workers. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is asking whether a graduate sacrifices income by entering teaching. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is looking only at absolute salary without the local labour market. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the comparison includes an appropriate professional benchmark. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Career progression shapes long-term attractiveness
Career progression shapes long-term attractiveness is mainly about examining how pay develops across years rather than only the entry point. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is Korea’s large progression from starting salary to the top scale. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is calling a system high-paying or low-paying from the starting salary alone. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the career trajectory is visible. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Subnational variation matters
Subnational variation matters is mainly about recognising that national labels can hide regional pay systems. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is the United Kingdom and United States showing substantial subnational salary variation. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is publishing one UK or US number as if every jurisdiction uses it. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that country comparisons preserve federal or devolved structure. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Qualification requirements affect comparability
Qualification requirements affect comparability is mainly about checking whether systems expect bachelor’s, master’s or other professional preparation. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is comparing teacher earnings with workers of similar educational attainment. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is ignoring qualification differences when judging pay. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that salary is interpreted alongside entry requirements. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Teaching level affects pay
Teaching level affects pay is mainly about separating pre-primary, primary, lower-secondary and upper-secondary scales. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is noting OECD averages increase with education level taught. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is comparing a primary figure with an upper-secondary figure. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the school level is named. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Teaching hours are not total working hours
Teaching hours are not total working hours is mainly about distinguishing classroom instruction from preparation, marking and administration. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is using OECD D4 for official teaching time and TALIS for self-reported total work. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming a lower teaching load means a shorter work week. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that both time measures are visible. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Working-time rules differ by system
Working-time rules differ by system is mainly about recognising that some systems regulate annual teaching hours while others specify broader working time. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is comparing contractual teaching time only within its definition. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating one country’s regulation as equivalent to another’s school-level practice. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the institutional rule is explained. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Non-teaching work is part of the profession
Non-teaching work is part of the profession is mainly about counting planning, feedback, collaboration, administration and student support. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS finding that teaching is less than half of total workload in many systems. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is describing non-teaching hours as time away from the real job. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the wider professional task is represented. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Status is partly social perception
Status is partly social perception is mainly about using teacher reports of whether society values the profession as one status indicator. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is Singapore TALIS teachers reporting relatively high perceived social value. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating prestige as measurable by salary alone. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that status is treated as multidimensional. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Status is also institutional
Status is also institutional is mainly about considering selectivity, preparation, autonomy, career pathways and public trust. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is a system combining professional training with structured progression. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is reducing status to ceremonial praise. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that institutional design enters the comparison. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Salary satisfaction differs from salary level
Salary satisfaction differs from salary level is mainly about distinguishing perception from nominal or PPP pay. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is teachers in one system reporting high salary satisfaction without necessarily having the highest OECD salary. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is using satisfaction to infer objective compensation ranking. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that subjective and objective measures stay separate. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Terms of employment extend beyond salary
Terms of employment extend beyond salary is mainly about including benefits, work schedule, security and other conditions. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS questions about satisfaction with employment terms excluding salary. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is calling pay the whole employment package. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that the analysis includes non-salary conditions. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Work intensity matters
Work intensity matters is mainly about examining pace, deadline pressure and fragmentation alongside hours. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is two teachers working similar hours under different interruption patterns. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is using weekly hours as the only workload measure. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that intensity is considered independently. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Administrative workload can drive stress
Administrative workload can drive stress is mainly about recognising low-value or fragmented work as a distinct condition. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS showing administrative work is a common stress source across systems. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming administrative time is harmless because it is fewer hours than teaching. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that stress indicators are interpreted with task type. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Marking demands vary by assessment culture
Marking demands vary by assessment culture is mainly about linking feedback workload to subject, assessment and school expectations. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is a system with frequent written assessment creating different correction demands. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is comparing marking hours without assessment context. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that assessment architecture is part of the explanation. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Classroom discipline changes working conditions
Classroom discipline changes working conditions is mainly about recognising that teaching context affects demand. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS examining disciplinary climate as a workload and stress condition. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating teacher stress only as a salary problem. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that school conditions enter the model. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Student diversity requires support capacity
Student diversity requires support capacity is mainly about connecting inclusion and adaptation expectations to time and resources. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is teachers differentiating instruction while maintaining common goals. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is calling all variation a teacher-effort problem. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that system support is considered. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Leadership affects retention
Leadership affects retention is mainly about seeing school-level management as part of working conditions. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is a principal clarifying priorities and protecting collaboration time. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming national salary policy determines every retention decision. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that school factors remain visible. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Professional learning is a benefit and a workload
Professional learning is a benefit and a workload is mainly about recognising training can improve practice while consuming time. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is teachers participating in useful development with protected time. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is adding professional learning on top of unchanged responsibilities. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that development is judged by both impact and workload. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Autonomy can raise professional status
Autonomy can raise professional status is mainly about allowing teachers discretion within curriculum and standards. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is teachers adapting examples and assessment based on evidence. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is equating autonomy with absence of accountability. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that professional discretion and responsibility are linked. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Mentoring changes early-career conditions
Mentoring changes early-career conditions is mainly about supporting beginning teachers while they build routines and judgement. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is reduced or supported workload paired with mentor feedback. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is expecting novices to perform like veterans immediately. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that entry conditions are included in retention analysis. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Job satisfaction is not the inverse of workload
Job satisfaction is not the inverse of workload is mainly about allowing meaningful work and high demand to coexist. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is Singapore teachers reporting high job satisfaction alongside long self-reported working hours. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming satisfied teachers have no workload problem. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that multiple indicators are read together. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Career intentions are not identical to actual attrition
Career intentions are not identical to actual attrition is mainly about using plans to leave as a warning signal rather than a forecast. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS reports of younger teachers intending to leave within five years. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating intention percentages as future exit rates. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that intentions are labelled correctly. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Teacher shortages require local diagnosis
Teacher shortages require local diagnosis is mainly about avoiding one universal explanation. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is a region struggling to recruit a specific subject despite acceptable national averages. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is blaming shortage on salary alone. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that subject, geography and conditions are investigated. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
High salaries do not remove all workforce problems
High salaries do not remove all workforce problems is mainly about recognising that pay is one lever among several. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is a well-paid system still facing workload or ageing-workforce issues. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming salary determines satisfaction mechanically. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that multifactorial explanations are preserved. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Low relative pay can create opportunity-cost pressure
Low relative pay can create opportunity-cost pressure is mainly about seeing how competing graduate careers matter. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is teachers earning less than comparable tertiary-educated workers in many OECD systems. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is interpreting this as proof of low social value. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that economic opportunity cost is separated from social status. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Comparisons should not become league tables
Comparisons should not become league tables is mainly about using data to understand system design rather than crown winners. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is comparing Germany’s high salary levels, Korea’s steep progression, Singapore’s social-value indicators and Finland’s teaching-time share as different features. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assigning an overall best teaching system score. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that each dimension remains separate. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Germany illustrates high salary levels
Germany illustrates high salary levels is mainly about using a documented example without universalising it. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is OECD 2026 reporting Germany among the highest teacher salary levels in available data. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is concluding Germany has the best working conditions. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that pay is not confused with the whole profession. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Korea illustrates career progression
Korea illustrates career progression is mainly about using its salary trajectory as a different system feature. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is OECD 2026 reporting lower-secondary pay moving from below comparable-worker earnings at entry to above them after 15 years. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating the path as automatically attractive to every teacher. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that progression is described alongside time-to-top and conditions. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
The UK illustrates devolved variation
The UK illustrates devolved variation is mainly about showing why nation-level and subnational detail matters. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is OECD 2026 reporting a sizeable lower-secondary starting-salary range across UK subnational entities. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is using England as a proxy for the entire UK without saying so. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that jurisdiction is explicit. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
The United States illustrates state variation
The United States illustrates state variation is mainly about recognising federal systems cannot be summarised by one pay scale. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is OECD 2026 showing wide lower-secondary 15-year statutory salary differences across subnational entities. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is treating one state salary as national. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that state structure is preserved. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Singapore illustrates a status-and-workload combination
Singapore illustrates a status-and-workload combination is mainly about using TALIS 2024 to compare perceived social value, salary satisfaction and workload. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is 71% reporting teachers valued in society and 55% satisfied with salary alongside high reported working hours. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is using survey perception as proof of objective prestige. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that survey measures remain labelled. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Finland illustrates time allocation
Finland illustrates time allocation is mainly about noting how teaching can occupy a larger share of reported working time than in several Asian systems. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS showing teaching at least half of self-reported total working time in Finland. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is assuming that automatically means lower workload. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that share of time is distinguished from total hours. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Japan illustrates why satisfaction and workload must be separated
Japan illustrates why satisfaction and workload must be separated is mainly about reading low overall satisfaction in TALIS beside changing working time. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is TALIS 2024 reporting 79% overall job satisfaction and a decrease in total working time since 2018. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is turning one indicator into a verdict on the profession. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that trend, level and context are kept distinct. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
Current data need publication dates
Current data need publication dates is mainly about treating teacher-system comparisons as F3 evidence. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is using EAG 2026 for 2025 salary data and TALIS 2024 for survey conditions. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is mixing years silently. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that every quantitative comparison names the data year. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
The comparison should end in design questions
The comparison should end in design questions is mainly about using evidence to ask what a system is trying to attract, support and retain. The first discipline of comparative education is to compare the same construct. If one number includes bonuses and another does not, if one refers to primary teachers and another to upper-secondary teachers, or if one is a statutory scale while another is observed average pay, the arithmetic may be correct while the inference is wrong.
A practical example is asking whether salary progression, preparation, workload and autonomy fit the desired teaching profession. The example should be read as one dimension of system design. Teacher policy is a bundle: entry requirements, preparation, salary path, workload, autonomy, school leadership, professional learning, career structure and social expectations. A strong value on one dimension can coexist with strain on another.
The common failure is using comparison mainly to imitate another country. International comparison invites ranking because rows and columns create the appearance of a contest. But a country can have high salaries and demanding working conditions, high social prestige and recruitment difficulties, or high job satisfaction alongside substantial administrative stress. The evidence becomes more useful when contradictions are preserved instead of averaged away.
The acceptance test is that transfer requires institutional context. Every quantitative statement should retain its population, year and measure. A comparison should also end with a mechanism question: what feature of the system could plausibly produce this pattern, and what additional evidence would be needed before drawing a causal conclusion?
A current comparison snapshot without a league table
| System / evidence | What the current data illustrate | What not to infer |
|---|---|---|
| OECD average · EAG 2026 | Lower-secondary statutory salary after 15 years averages USD 66,583 PPP equivalent across OECD countries/economies with data. | This is not a global teacher salary, take-home pay or cost-of-living-free comparison. |
| Germany · EAG 2026 | Germany is among the highest teacher-salary systems in available OECD data; the country profile reports very high actual and 15-year-equivalent salary indicators. | High pay alone does not establish workload, satisfaction or educational quality. |
| Korea · EAG 2026 | Lower-secondary starting salary is reported below similarly educated-worker earnings, while 15-year salary is above that benchmark; progression is steep and time to the top is long. | One career path should not be reduced to a starting-pay comparison. |
| United Kingdom · EAG 2026 | Lower-secondary starting and top salaries vary meaningfully across subnational entities. | England, Scotland, Wales and Northern Ireland should not be treated as one identical pay system. |
| United States · EAG 2026 | OECD reports very wide subnational variation in 15-year lower-secondary statutory salaries. | One state’s teacher salary is not a national salary scale. |
| Singapore · TALIS 2024 | 71% report teachers valued in society; 55% report salary satisfaction; overall job satisfaction is 87%, while administrative work and marking are common stressors. | Survey perceptions are not the same as statutory pay or objective prestige. |
| Finland · TALIS 2024 | Teaching accounts for at least half of self-reported total working time, unusual among participating systems. | A higher teaching-time share does not tell us total workload by itself. |
| Japan · TALIS 2024 | Overall job satisfaction is reported at 79%, and total reported working time has fallen since 2018. | A single satisfaction figure cannot explain the profession’s full working conditions. |
A better method for comparing teaching professions
- Choose the teacher population and education level before collecting numbers.
- Separate statutory salary, actual salary, relative earnings and benefits.
- Use PPP conversion only for the comparison it is designed to improve.
- Keep national, devolved and subnational systems separate where governance differs.
- Pair teaching hours with total workload rather than treating either as sufficient.
- Add teacher-reported status, satisfaction and stress as separate dimensions.
- Record qualification and preparation requirements because they change opportunity cost.
- Use the same data year wherever possible and label mixed-year evidence explicitly.
- Do not infer causal effects from cross-country correlations.
- End with design questions rather than an overall winner.
Frequently asked questions
Which country pays teachers the most?
That depends on the salary measure, level of education, career stage and PPP method. Education at a Glance 2026 places Germany, Luxembourg and the Netherlands among very high statutory salary systems at several levels, but an overall ranking can still mislead.
Are Singapore teachers highly paid internationally?
The strongest directly comparable current evidence used here for Singapore is TALIS teacher-reported salary satisfaction and working conditions, not an OECD EAG statutory salary table entry. Do not splice incompatible sources into a ranking.
Does higher teacher pay improve student outcomes?
Pay can affect recruitment and retention incentives, but cross-country correlations cannot establish a simple causal relationship with student outcomes. Systems differ in selection, preparation, curriculum, workload and many other factors.
What does PPP salary mean?
PPP conversion adjusts for differences in price levels to improve purchasing-power comparison. It is not the amount a teacher receives in US dollars or a substitute for local tax and benefit analysis.
Why compare teacher salaries with tertiary-educated workers?
Because teaching often requires tertiary qualifications. Relative earnings help show the opportunity cost of choosing teaching rather than another similarly educated occupation.
Why do teaching hours and total working hours differ?
Teachers spend substantial time on preparation, marking, collaboration, administration, communication and student support outside classroom instruction.
What does teacher status mean?
It can include social value, occupational prestige, professional autonomy, qualifications, career structure, trust and teachers’ own perceptions. No single indicator captures all of it.
What does TALIS measure?
TALIS is an international survey of teachers and school leaders. Many working-time, satisfaction and stress measures are self-reported and should be interpreted as survey evidence rather than time-clock records.
Why not publish a country ranking?
Because salary, status and working conditions are multidimensional and measured differently. An overall ranking would require arbitrary weights and would hide important institutional trade-offs.
Where should current comparisons be checked?
Use Education at a Glance 2026 for current OECD system indicators and TALIS 2024 for teacher-reported working conditions.
The quiet conclusion: a teaching profession is a package of signals
A country tells prospective and current teachers what the profession means through more than salary. It signals value through who is admitted, how teachers are prepared, how quickly pay progresses, whether professional judgement is trusted, how much time is protected for preparation and feedback, how schools are led and whether career growth requires leaving classroom expertise behind.
Comparative data are useful when they make those signals visible. They become less useful when they are compressed into a winner. The better question is not which country has the best teachers’ deal. It is which combination of pay, time, status, preparation and working conditions a system has chosen—and whether that combination can attract, develop and retain the teachers its students need.