This chapter explores teachers’ perceptions of their working conditions in primary and upper secondary education by examining contract conditions and arrangements, working hours, job security and stability. The chapter also looks at reports on satisfaction with the profession, school, salaries and terms of employment. It also describes the levels and sources of stress and makes a comparison across levels to detect the most pressing needs at each level. The chapter concludes by examining the risk of attrition for teachers leaving the profession.
Teachers Getting the Best out of Their Students
From Primary to Upper Secondary Education
TALIS
6. Building fulfilling working conditions, well-being and satisfactory jobs
Abstract
Highlights
Primary education
Female teachers are more likely than their male colleagues to work part-time, be stressed, work longer hours and report that they would like to leave teaching in the next five years.
Fixed‑term contracts are relatively scarce in primary education: 8% have fixed‑term contracts for more than one year, and 8% have a fixed‑term contract for one year or less.
Compared to their colleagues in lower secondary education, teachers in primary education spend more total working hours (1 hour more) and teaching hours (2.9 hours more), and fewer hours in non‑teaching tasks (such as planning and marking).
Almost half of teachers (47%) are satisfied with their salaries and 65% are satisfied with the terms of their employment (e.g. benefits, work schedule). Number of working hours is one of the main predictors of teachers’ satisfaction with their employment.
The main sources of stress as reported by teachers are “having too much administrative work to do” (47%); “being held responsible for students' achievement” (47%); “addressing parent or guardian concerns” (42%); and “addressing parent or guardian concerns” (42%). In contrast to lower secondary, “maintaining school discipline” and “modifying lessons for students with special needs” seems to be a particular source of stress.
Twenty-one percent of teachers in primary education stated they would like to leave teaching in the next five years. Teachers satisfied with their salaries and terms of employment, and whose school provides opportunities to engage in discussion and has a collaborative culture are less likely to state they will leave teaching.
Upper secondary education
Female teachers are more likely than their male colleagues to be stressed and to report that they would like to leave teaching in the next five years.
Fixed‑term contracts are relatively scarce in upper secondary education: 10% have fixed‑term contracts for more than one year, and 10% have a fixed‑term contract for one year or less.
Teachers in upper secondary education spend more hours than their colleagues in lower secondary education in non‑teaching tasks such as planning and marking,
The share of teachers satisfied with the terms of employment and salary is higher in upper secondary than in lower secondary education. One of the main drivers of teachers’ satisfaction with terms of employment is the number of working hours and classroom composition; in particular, the share of low achievers in the classroom.
The main sources of stress as reported by teachers are “being held responsible for students' achievement” (47%); “having too much marking” (46%); and “having too much administrative work to do” (45%). In contrast to lower secondary, “being intimidated or verbally abused by teachers” seems to be a significant source of stress.
Almost a quarter of teachers (23%) in upper secondary education stated they would like to leave teaching in the next five years. This is a higher share than in lower secondary education. Teachers satisfied with their salaries and terms of employment, and who consider the school provides opportunities to engage in discussion and has a collaborative culture are less likely to state they will leave teaching.
Introduction
The notion of promoting teacher professionalism in order to build a highly skilled workforce has been ingrained in the policy orientations of education systems worldwide (Schleicher, 2018[1]). That being said, education systems need to support teachers as professionals. This points to the necessity of providing stimulating, rewarding and supporting working conditions that will allow teachers to be effective in their work. Indeed, the quality of teachers’ working conditions has increasingly become part of educational system inspections as teacher turnover has become a major concern (OECD, 2019[2]; Viac and Fraser, 2020[3]).
Working conditions can be understood as elements of the working environment affecting the daily work of teachers such as terms of employment (i.e. type of contract and working hours); workload (e.g. class sizes); reward and incentives structure (e.g. salaries); and supportive and collaborative working structures (OECD, 2019[2]). These elements can have an effect on teachers’ job satisfaction, stress levels and their commitment to stay in the profession, which translates indirectly into the supply and retention of teachers (Viac and Fraser, 2020[3]). As such, the working conditions of teachers play an important role in empowering the individual and collective professional capacity of school leaders (OECD, 2019[2]).
Good working conditions such as the adequate allocation of resources along with supportive and collaborative working environments can improve teachers’ overall well‑being, job commitment and efficiency (Bakker and Demerouti, 2007[4]; Borman and Dowling, 2008[5]; Cochran-Smith, 2004[6]; Collie and Martin, 2017[7]; Hakanen, Bakker and Schaufeli, 2006[8]; Mostafa and Pál, 2018[9]). In contrast, unmanageable job demands and stressful working conditions can lead to low job satisfaction and well‑being (Collie, Shapka and Perry, 2012[10]; Desrumaux et al., 2015[11]), lower levels of job commitment (Klassen et al., 2009[12]; Skaalvik and Skaalvik, 2016[13]) and burnout (Betoret, 2009[14]). They can also generate motivation to leave the profession (Skaalvik and Skaalvik, 2018[15]) and ultimately lead to actual attrition (Weiss, 1999[16]).
Although Teaching and Learning International Survey (TALIS) data were collected prior to the COVID‑19 pandemic, many of the topics covered in this chapter invite reflections on teachers’ working conditions during the crisis. School closures, along with the need to adopt a hybrid form of instruction mixing digital learning and some form of face‑to‑face interaction, have changed teachers’ work organisation. The changes can affect everything from time allocation for lesson planning and instruction to the type of working contracts (OECD, 2021[17]). The data presented in this chapter is helpful in understanding different conditions in countries and economies to better inform policy implementation.
Analyses using TALIS 2018 have already provided a description of the main features of the working conditions in lower secondary education but less is known about primary and upper secondary education. Indeed, features such as class sizes, remunerations and teaching hours fluctuate across educational levels. Therefore, it is pertinent to observe whether the levels of teachers’ job satisfaction, stress and job commitment change specific to each education level. The chapter will make use of TALIS 2018 indicators to describe the working conditions of teachers in primary and upper secondary education and highlight the main differences with lower secondary education. It will start by providing a description of the contractual job security and distribution of working hours. The following section explores the perceptions of teachers and school leaders of their work by reporting on satisfaction with their current working environment and their profession. Next, the chapter looks at occupational stress among teachers and school leaders as well as the sources of stress. The final section deals with the risk of attrition among teachers.
Job security and flexible time arrangements
The type of contract conditions (i.e. permanent and fixed‑term employment) and working‑time arrangement (i.e. full‑time or part‑time) play an important role in schools’ management of their resources (Bertoni et al., 2018[18]) while also affecting the attractiveness of teaching jobs (OECD, 2019[19]). The use of different contract modalities and work‑time arrangements carry positive and negative aspects that need to be balanced carefully.
Fixed‑term contracts can introduce some necessary flexibility to educational systems and allow schools to manage their human resources (Bruns, Filmer and Patrinos, 2011[20]). They allow schools to adjust staffing needs to respond to changing student demographics, keep long‑term financial commitments in check and facilitate the evaluation of a teacher’s skill and ability to grow before offering a permanent contract (OECD, 2019, p. 220[2]). That being said, in some national contexts, teachers under fixed‑term contracts may not benefit from the same statutory rights as their colleagues on permanent contracts when it comes to professional development opportunities or career and salary progressions. In addition, fixed‑term contracts can result in a perceived lack of stability by the teaching staff if temporary teachers are at the risk of being replaced by a teacher with a permanent appointment. This can hamper long‑term relationships teachers foster with their colleagues, leaders, students and the community, and affect school climate and collaboration (Latifoglu, 2016[21]).
Working‑time arrangements (i.e. full‑time or part‑time) allow teachers seeking to balance their work load at differences stages of their career and give schools the possibility to adjust to changing students numbers and capacities (Boeskens and Nusche, 2021[22]; OECD, 2019[2]). That being said, full‑time or part‑time may play an important role in teachers’ satisfaction and well‑being as part‑time is not a voluntary decision but the result of a shortage of permanent positions in some systems. In those scenarios, teachers may feel the need to work in multiple schools to fulfil a full‑time schedule (OECD, 2019[19]; OECD, 2014[23]). Although TALIS 2018 did not collect data on the reasons why teachers chose to work part‑time, TALIS 2013 showed that almost half of the teachers in lower secondary education working part‑time did not have the option of working full‑time (OECD, 2014[23]). In contrast, full‑time teachers may wish to work part‑time but cannot afford to or are concerned about losing their job (Sharp et al., 2019[24]).
The COVID‑19 pandemic may also have an impact on teachers’ working arrangements in that additional staff has also been often required to provide extra support to students or compensate for prolonged teachers’ absences. A recent survey of the OECD countries’ responses to the COVID‑19 pandemic showed that only a fraction adapted their recruitment practices to minimise the impact of school closure and enable school reopening. Nine of the 28 OECD countries who monitored recruitment practices recruited temporary teachers and/or other staff to support lower‑secondary students in need during the pandemic in 2020 (OECD, 2021[17]). Box 6.1 shows two of these countries, Japan and Wales (United Kingdom), which show how recruitment policies may be used to respond to the COVID‑19 pandemic.
Box 6.1. Staffing policies responding to the COVID‑19 pandemic
Japan: Reinforcing human resources in schools
Japan committed considerable extra resources to support the reopening of schools after the initial lockdown measures. Firstly, the Ministry of Education, Culture, Sports, Science and Technology (MEXT) reinforced human resources for schools that were dividing classes for staggered attendance. Additional teachers enabled multiple smaller classes to run among students in the final years of primary and lower secondary education in order to allow them to receive a sufficient amount of in‑person teaching. Between one and three extra instructors were employed per school to assist classroom teachers and, in regions with a high infection rate, school support staff were assigned to schools to support lesson planning, parental contact and COVID‑19 related administrative tasks. Finally, school counsellors and social workers were assigned on a school‑by‑school basis. Japan recruited these extra staff from the pool of retired teachers, tutors, university students and other education‑related staff or actors in the community. To facilitate the process, Japan eased qualification requirements for instructors, recognising temporary special licenses.
Wales (United Kingdom)
Wales recruited 600 extra teachers and 300 additional teaching assistants throughout the academic year 2020/21. These additional staff provided support to students at the end of upper secondary school as well as disadvantaged and vulnerable learners of all ages. They supported learners taking national end‑of‑cycle examinations in 2021 and those most affected by school closures. Professional learning resources was provided to support new and existing teachers. Staff were recruited on a one‑year fixed‑term contract and expected to move into educational roles in the following school year. The support package, provided at a school level often included extra coaching support, personalised learning programmes, and additional time and resources for exam‑year pupils.
Source: OECD (2020[25]), Lessons for Education from COVID‑19: A Policy Maker’s Handbook for More Resilient Systems, https://dx.doi.org/10.1787/0a530888-en.
This section presents findings prior to the COVID‑19 pandemic from the TALIS 2018 study on work modalities in primary and upper secondary education. Teachers responded about their contract duration (i.e. “permanent”; “fixed‑term for more than 1 year”, “fixed‑term for less than 1 year”) and whether they were employed on a full‑time basis (i.e. 90% or more of full‑time hours) or on a part‑time basis (i.e. between 71% and 90% of full‑time hours or between 50% and 70% of full‑time hours or less than 50% of full‑time hours).
Contract conditions and working‑time arrangements in primary education
On average across the TALIS participants in primary education, 84% of teachers have permanent contracts, 8% have fixed‑term contracts for more than one year, and 8% have a fixed‑term contract for one year or less (Table 6.1). Overall, there are no large differences in these percentages across participants. The United Arab Emirates is the only country with less than half of teachers reporting having a permanent contract (42%).
It is interesting to observe that for 9 out of the 13 participant teachers with available data, teachers who do not have a permanent contract are more likely to have a less‑than‑one‑year contract instead of more‑than‑one‑year (Table 6.1). Countries with the highest percentage of teachers in primary education with less‑than‑one‑year contracts are Spain (22%), Japan (17%), the United Arab Emirates (14%) and the Flemish Community of Belgium (12%).1
On average across the TALIS participating countries and economies in primary education, the share of teachers with a permanent contract is much higher among experienced teachers (i.e. teachers with five years of experience or more, 88%) than novice teachers (i.e. teachers with less than five years of experience, 63%) (Table 6.2). This gap is significant for all participants in primary education but the cross‑country variation is quite large. The difference is over 50 percentage points in the Flemish Community of Belgium and Spain while it is less than 10 percentage points in Denmark, Korea and the United Arab Emirates. The results may be pointing to the different regulatory frameworks for recruiting teachers across countries. For example, in the Flemish Community of Belgium, where teachers are employed by school boards, all beginning teachers are recruited on a temporary basis of one year before they can be appointed to a temporary position of continuous duration and ultimately on a permanent basis (Nusche et al., 2015[26]). This case of the Flemish Community reveals a selection effect: teachers who stay in the profession eventually get a permanent contract. However, a cohort effect might also be in place where newer generations are more likely to be employed under a fixed‑term contract than previous generations of teachers (Paccagnella, 2016[27]).
Although the use of fixed‑term contracts for novice teachers could be an adequate measure to assess novice teachers and have flexibility in attracting a wider scope of professional without necessarily engaging in a financial commitment, it is also important to acknowledge some of the trade‑offs: teachers on fixed‑term contracts tend to be less protected by pension schemes; are less often awarded study leave; and less entitled to benefits and rights, including family benefits and annual holiday pay (Stromquist, 2018[28]). This can affect the attractiveness of the profession.
On average among the TALIS participants in primary education, the share of teachers on permanent‑term contracts in schools with more than 30% of students from socio‑economically disadvantaged homes (henceforth, “disadvantaged schools”) differs significantly from the share in non‑disadvantaged schools (Table 6.2). The results are significant for only 5 out of 13 participants with available data and with considerable variation across countries. For example, there is a 22 percentage‑point difference in Ciudad Autónoma de Buenos Aires (henceforth CABA [Argentina]) but only a 5 percentage‑point difference for Turkey. This pattern could be explained by a series of factors characterising disadvantaged schools. For example, as discussed in Chapter 2 of this report, novice teachers are more likely to work in disadvantaged schools and, at the same time, novice teachers are more likely to be employed with a fixed‑term contract. Furthermore, the intention to leave teaching is also higher in disadvantaged schools (see the last section of the chapter), which could present difficulties in staffing. These schools may rely on more flexible forms of contract such as fixed‑term contracts. Beyond the specific factors explaining this pattern, the fact that teachers in more challenging contexts are employed under less favourable affects the overall equity of the system as it presents an asymmetry in the opportunity of attracting quality candidates to the schools (OECD, 2019[19]).
Regarding working arrangements, on average across the TALIS participants in primary education, 84% of teachers reported that they are employed full‑time (all teaching employments included) (Table 6.4). A much smaller proportion reported that they are employed as teachers between 71% and 90% of full‑time hours (10%); between 50% and 70% of full‑time hours (5%); or less than 50% of full‑time hours (2%). More than 20% of teachers work part‑time in CABA (Argentina) (30%), Viet Nam (29%), the Flemish Community in Belgium (27%) and England (United Kingdom) (23%). In contrast, in Japan, Korea and the United Arab Emirates, 90% of teachers or more reported that they are employed full‑time.
The prevalence of full‑time work is unevenly distributed by gender across the teacher population. In primary education, male teachers reported working full‑time more often than their female colleagues (a difference of 3 percentage points on average across the TALIS participants) (Table 6.5). The countries showing a difference over 10 percentage points are England (United Kingdom) (19 percentage points), the Flemish Community of Belgium (13 percentage points) and France (11 percentage points). On the contrary, the proportion of men reporting that they work full‑time is lower than for women in Korea (3 percentage points) and particularly in Viet Nam (13 percentage points).
This pattern is not at odds with trends observed in other professions. Indeed, women are more likely to work part‑time than men since they are more likely to take care of most of the childcare work (OECD, 2017[29]). Working part‑time could be a decision of female workers to balance their professional personal and professional life. The cross‑country variation in this pattern might be explained by institutional differences in teaching career and labour as well as cultural norms. For example, in some settings, the use of part‑time contracts could be used as family‑friendly working‑time arrangements, and helps groups with traditionally low labour‑force participation, such as mothers, to remain in work (Boeskens and Nusche, 2021[22]).
That being said it would be important to examine if this imbalance in the working commitment between male and female workers could affect the career progression of women in primary education. Indeed, evidence has pointed to the association between part‑time and penalties in terms of pay, job security, benefits, promotion and training (OECD, 2017[29]). Indeed, there is a strong negative relation between the proportion of female teachers working part‑time and the proportion of women working as a school leader (the linear correlation coefficient is r=-1, based on 13 education systems with available data) (Tables 2.12 and 6.4). This means that at a system level a higher proportion of women working as a school leader is accompanied by a lower proportion of women working part‑time. These estimates should be interpreted carefully as they are conservative estimates at the country level.
Regarding teaching experience, the proportion of experienced teachers who reported that they are employed full‑time is significantly higher for novice teachers on average by two percentage points (Table 6.5). That being said, the breakdown by participants shows that cross‑nationally, teachers engage in full‑time working positions at different stages of their career. From higher to lower values, the countries showing that experienced teachers are more likely to work full‑time than novice teachers are Spain (22 percentage points), France (11 percentage points), CABA (Argentina) (9 percentage points) and Korea (4 percentage points). In contrast, the proportion of experienced teachers reporting full‑time employment is significantly lower than novice teachers for England (United Kingdom) (23 percentage points) and the Flemish Community of Belgium (8 percentage points). The results might reflect specific institutional configurations of the labour market in each system. For example, in CABA (Argentina), France, Korea and Spain, novice teachers might have more difficulties in finding available full‑time position while in England (United Kingdom) and the Flemish Community of Belgium more experienced teachers seek to reduce their hours as they approach retirement (Boeskens and Nusche, 2021[22]).
In four education systems with available data for primary education, full‑time employment is significantly more frequent among teachers in city schools (school in locations with over 100 000 people) than among their colleagues in rural or village schools (school in locations with up to 3 000 people) (Table 6.5). In England (United Kingdom), the Flemish Community of Belgium and Viet Nam, the difference between the shares of teachers working full‑time in city schools and rural or village schools is at least of 10 percentage points. In contrast, in Turkey, teachers in rural or village schools are significantly more likely to report being employed on full‑time contracts than their colleagues in city schools. The difference may correspond to geographical labour opportunities and changing demographics in the student population in rural areas in relation to urban areas.
Schools with over 30% of students from socio‑economically disadvantaged homes also exhibit a significantly higher share of full‑time teachers than other schools in France (17 percentage point difference), Sweden (6 percentage points), England (United Kingdom) (5 percentage points) and the United Arab Emirates (3 percentage points) (Table 6.5). This is, overall, a positive result as it shows teachers’ dedication to schools with greater needs as full‑time teachers spend more time face‑to‑face with their students and have more opportunities for interaction with parents and community. In contrast, in Viet Nam, the share of full‑time teachers in school with a high concentration of disadvantaged students is 6 percentage points lower than in schools with a lower concentration of these students.
TALIS 2018 analysis conducted for lower secondary education spotlights some risk part‑time working arrangements can present to effective teaching. These include the negative association between part‑time work and level of self‑efficacy (OECD, 2019[30]). One of the aspects that might be jeopardised is teachers’ ability to establish professional collaboration networks within their school – one of the key characteristics of an effective educational system. But can this be established when some teachers work with less time commitment than others?
Regression analyses were conducted to examine the assumption that part‑time teachers may not engage frequently and effectively in professional collaboration activities. The scale of professional collaboration2 in lessons among teachers is regressed on teachers working full‑time in their schools. The regression is controlled by type of contract, gender and age. For 6 out 13 countries with available data there is a positive association between teachers working full‑time in their schools and participation in professional collaboration. The association between the contractual time engagement and collaboration is not all that conclusive as results are significant for less than half of the countries. On the other hand, more can be done to foster conditions to support teacher collaboration in schools, irrespective of their part‑time/full‑time status (Table 6.7).
How do primary and lower secondary levels compare?
In comparing the contractual and working arrangements of teachers across levels it is possible to note that permanent contracts are more prevalent among teachers in primary education than lower secondary education, although the average difference is small (an average difference of 2 percentage points) (Figure 6.1 and Table 6.1). The differences are particularly pronounced in Korea where 99% of teachers in primary education have a permanent contract in contrast to 88% in lower secondary education. In England (United Kingdom), the share of teachers in primary education with permanent contract is significantly lower than in lower secondary education (3 percentage points of difference).
Figure 6.1. Teachers’ working arrangements in primary education
Note: Statistically significant differences between primary education and lower secondary education are shown in a darker tone. The values of these differences are indicated before (permanent employment) and after (full‑time) the country/economy name.
Countries and economies are ranked in descending order of the percentage of teachers with a permanent contract in primary education.
Source: OECD, TALIS 2018 Database.
A similar pattern can be observed for teachers working full‑time since the share of teachers working full‑time is higher in primary education than lower secondary education, although by a small margin (an average difference of 2 percentage points) (Table 6.4). The contrast is particularly remarkable for CABA (Argentina) where 70% of teachers in primary education work full‑time in contrast to just 59% in lower secondary education. On the contrary, the share of teachers in primary education working full‑time is significantly lower in Denmark (3 percentage points of difference), England (United Kingdom) (5 percentage points of difference) and France (6 percentage points of difference).
The bottom line is that jobs in primary education seem somewhat more stable than in lower secondary education but might be lacking in the contractual flexibility exhibited in lower secondary education, particularly for some countries. The marginal prevalence of more flexible working arrangements in lower secondary could be explained by a greater demand in secondary education for specialist expertise, which can be addressed through part‑time arrangements (Sharp et al., 2019[24]) (see Chapter 3).
Contract conditions and working arrangements in upper secondary education
Regarding upper secondary education, there is a predominance of permanent contracts, although in lower shares compared with lower secondary or primary. On average across the TALIS participants in upper secondary education, 80% of teachers have permanent contracts; 10% have fixed‑term contracts for more than one year; and 10% have a fixed‑term contract for one year or less (Table 6.1). In most countries, more than 80% of teachers have permanent contracts with the exceptions of Brazil (78%), Portugal (72%) and the United Arab Emirates (38%).
On average across the TALIS participating countries in upper secondary education, the proportion of teachers employed on a permanent contract is much higher among teachers with more than five years of experience (85%) than novice teachers with less than five years of experience (56%) (Table 6.3). This gap is significant for all participants in upper secondary education with the exception of the United Arab Emirates. The difference is over 50 percentage points in Portugal and Slovenia while it is the lowest in Denmark with a 17 percentage‑point gap. The results might respond to different institutional configurations on the use of this type of contract. These results may obey a cohort effect (the implementation of fixed‑term contracts have changed over time with new generations of teachers being particularly affected) or a selection effect (teachers with more experience are rewarded with permanent contracts as part of their career structure) (Paccagnella, 2016[27]). Beyond these possible explanations, participants should be mindful of some of the risks of relying too much on fixed‑term arrangements for novice teachers: the risk of attrition is more common among beginning teachers and unstable job security could further aggravate this situation.
On average across the TALIS participants in upper secondary education, 81% of teachers reported that they are employed full‑time (all teaching employments included) (Table 6.4). A much smaller proportion reported that they are employed as teachers between 71% and 90% of full‑time hours (9%); between 50% and 70% of full‑time hours (6%); or less than 50% of full‑time hours (4%). For most countries, more than 80% of teachers work full‑time in upper secondary education with the exception of Brazil (46%) and the Viet Nam (74%). The results for Brazil are explained by the fact that 20% of teachers work in more than one school (OECD, 2021[31]).3
In upper secondary education, the same pattern already apparent in primary education is seen, where male teachers more often than their female colleagues reported working full‑time (a difference of 3 percentage points on average across the TALIS) (Table 6.6). The countries showing these differences between men and women are Alberta (Canada) (8 percentage points of difference), Brazil (6 percentage points), Denmark (14 percentage points) and Sweden (7 percentage points). In contrast, in Viet Nam, the proportion of men reporting that they work full‑time is 5 percentage points lower than for women. In addition, the proportion of experienced teachers who reported that they are employed full‑time is significantly higher than for novice teachers for most participants except Alberta (Canada), Turkey, the United Arab Emirates and Viet Nam. The results might be attributed to female workers’ working preferences as part‑time contracts may be a more family‑friendly working‑time arrangement. They also help groups with traditionally low labour force participation, such as mothers, to remain in work (Boeskens and Nusche, 2021[22]).
That being said it would be important to examine if this imbalance in the working commitment between male and female workers could affect the career progression of women in primary education. Indeed, evidence has pointed out the association between part‑time and penalties in terms of pay, job security, benefits, promotion and training (OECD, 2017[29]). There is a mild negative relation between the proportion of female teachers working part‑time and the proportion of women working as a school leader (the linear correlation coefficient is r=-0.42, based on 11 education systems with available data) (Tables 2.12 and 6.4). This means that at a system level a higher proportion of women working as a school leader is accompanied by a lower proportion of women working part‑time. These estimates should be interpreted carefully as they are conservative estimates at the country level and bear further examination.
Overall, in upper secondary education it is possible to observe a common trend with primary education (see previous section) and with lower secondary education (OECD, 2020[32]): there is a predominance of permanent contract and full‑time working arrangements across the teaching profession. Although this predominance has a certain degree of variation by teacher characteristics such as gender and experience, these differences by type or the location of schools were not observed for upper secondary education, unlike primary education.
As the assumption that part‑time teachers may not engage effectively in professional collaboration activities was examined for primary education, regression analyses were conducted for upper secondary education, as well. The scale of professional collaboration (see Note 1) in lessons among teachers is regressed on teachers’ working full‑time, controlling for other teachers’ characteristics (gender, age, employment status). A significant negative association between teachers working part‑time up to 70% and professional collaboration was found in just 2 out of the 11 countries with available data. The absence of a strong pattern across participants seems to show that types of contractual arrangements do not impede effective forms of collaboration (Table 6.8).
How do upper secondary and lower secondary levels compare?
TALIS data show that management of the teaching workforce is similar across lower and upper secondary education. Across these educational levels, there are no significant differences observed between the share of teachers with permanent contracts or working full‑time (Tables 6.1 and 6.4).
Allocation of working hours
The effective use of working hours is a critical resource for education systems and schools to be able to achieve their educational goals. The decision of how a teacher should spend their hours is not always straightforward and an additional hour spent on teaching or instruction planning may have different consequences for the cost and quality of the learning process (Boeskens and Nusche, 2021[22]). It is important to note that the work of teachers is not limited to just classroom instruction; it also involves all the tasks supporting that instruction such as lesson planning, marking and other endeavours. In addition, increasingly, teachers are expected to be involved in areas not directly related to instruction but still considered fundamental such as engaging in extracurricular activities, counselling students, and being involved in administrative work. More often than not, these tasks compete with each other within the limited hours in a week of work. This can lead to stress and burnout (Viac and Fraser, 2020[3]).
In many OECD school systems, the regulation of teacher working hours is based on narrow conceptions of the profession, taking into account mainly or solely the amount of time teachers spend on teaching while vaguely defining the amount of time spent in non‑teaching tasks (Boeskens and Nusche, 2021[22]; Nusche et al., 2015[26]; OECD, 2019[2]; Santiago et al., 2016[33]). Such systems do not adequately recognise the importance of the tasks teacher devote outside of the classroom, which can lead to teachers working excessive hours or not finding enough time to fulfil all their tasks (OECD, 2019[2]).
Teaching time and non‑teaching time (e.g. the amount of time devoted to planning) is a product of national working regulations and school culture. In some countries, time spent on planning and lesson preparation may be a part of the mandated time schedule of teachers, whereas in other countries, time spent on this task may be due to teachers’ discretion and can engage in it voluntarily at home or in informal spaces. Indeed, a recent study using TALIS 2018 evidence shows that there is significant variability in the weekly teaching hours observed within some countries for lower secondary teachers (Boeskens and Nusche, 2021[22]). While for some countries (e.g. France, Sweden), the amount of time teachers spend on teaching falls around 3 hours on average (therefore indicating a relative low variance in the amount of teaching hours), in other countries (e.g. Alberta (Canada), Australia, Turkey) the amount of time spent on teaching falls around 12 hours on average (therefore indicating a relatively high variance in the amount of teaching hours).
Evidence has shown that spending time on lesson preparations is highly beneficial for the quality of instruction and may have an impact on student learning (Boeskens and Nusche, 2021[22]; Hargreaves, 1992[34]; Paniagua and Istance, 2018[35]). Preparation time helps teachers cope more effectively with these changes (OECD, 2019[19]). At the same time, some caution in interpretation is needed when observing a low or reduced number of hours devoted to planning and instruction as they might actually reflect teachers’ experience and effective time allocation across their tasks. Teacher preparation has become more effective through the use of technology. Increasingly, courses are prepared on computers and can more easily be updated. There are more opportunities for sharing course materials and artefacts with other teachers through the Internet and social media (Paniagua and Istance, 2018[35]) In addition, more experienced teachers may need to spend less in actual class preparation due to acquired skills and know‑how. The time spent on planning and preparation depends on the number of similar classes a teacher teachers; if the same lesson is replicated across student groups, then preparation is more efficient compared to lessons that cannot be replicated – which is usually the case in primary education.
Adequate understanding of the different tasks carried out by teachers along with their time allocation becomes especially imperative in the context of the COVID‑19 pandemic. Evidence shows that increasing actual class time was the least used approach in high‑income countries as a support measure (UNESCO, UNICEF, World Bank, 2020[36]). The sudden requirement for distance learning meant additional planning was required to transform face‑to‑face lessons into an online setting. Also, many teachers went through these adaptations without having the required training and tools to navigate digital instruction. The most recent survey of the OECD on how countries have responded to the COVID‑19 pandemic revealed that 18 OECD countries provided nationwide guidelines for reducing the amount of overtime required to prepare a virtual classroom (OECD, forthcoming). Box 6.2 shows how Portugal has implemented a reallocation of hours to support the learning recovery of students.
Box 6.2. Portugal: School action plans for the recovery and consolidation of learning
Portugal has directed schools to focus the first five weeks of the new school year on learning recovery. All schools must develop an Action Plan for the Recovery and Consolidation of Learning (Plano de Atuação para a recuperação/consolidação das aprendizagens) that guides teaching and learning throughout the year, intensively so in the first weeks, and should be flexible enough to withstand possible future closures. To support schools, the government has produced a set of guidelines for the organisation of the school year along with a comprehensive roadmap for recovery and consolidation, which includes example activities and learning tools and approaches for schools. Specific support measures include an increase in hourly credits – the time allocated to schools for non‑teaching‑related activities including management, mentoring, tutoring and personalised learning as well as teacher planning and collaboration – for the academic year 2020/21. The extra time must be used exclusively for activities related to learning recovery and consolidation. Portugal has also extended the Specific tutorial support programme – originally reserved for students in secondary education who have repeated a year twice – to all students who did not pass the school year in 2019/20. All schools must also establish a peer mentoring programme in which volunteer mentors from the student body are paired with student mentees.
Source: OECD (2020[25]), Lessons for Education from COVID‑19: A Policy Maker’s Handbook for More Resilient Systems, https://dx.doi.org/10.1787/0a530888-en.
TALIS 2018 data make it possible to know, based on teachers’ reports, how many 60‑minute periods they spent working in total and on various tasks in their school during the most recent complete calendar week prior to the survey (including tasks that took place during weekends, evenings or other out‑of‑class hours) and before the crisis. This section will provide a description of the areas teachers devote the most time to and how they compare to each other.
Teaching, planning and marking in primary education
On average across the TALIS participants in primary education, full‑time teachers spend 41.5 hours per week on all the tasks related to their job in their surveyed school, of which 22.7 hours are devoted to teaching4 (Table 6.9). In other words, teachers spend slightly more than half of their working time teaching classes (55%).
It is important to note that there is a great degree of variation in the share of teaching hours across countries. The lowest shares (less than 50% of the total teacher working hours) are observed in England (United Kingdom) (24.3 teaching hours out of 52.2 total working hours), Japan (23.5 teaching hours out of 55.9 total working hours), Sweden (20.2 teaching hours out of 43.7 total working hours) and Viet Nam (22 teaching hours out of 44.4 total working hours) while the highest shares (equal or above 60%) are found in CABA (Argentina) (23.5 teaching hours out of 38.9 working hours), Korea (20.4 teaching hours out of 32.5 total working hours), Spain (23.5 teaching hours out of 36.1 total working hours), Turkey (26.3 teaching hours out of 31.9 total working hours) and the United Arab Emirates (24.4 teaching hours out of 39.3 total working hours) (Table 6.9). These differences result from the way teachers’ hours are regulated, which varies among countries as well as from the country‑specific school culture among other factors – see Education at a Glance (OECD, 2020[37]) indicators of the teaching hours per system in Table A B.5.
The next two most time‑consuming activities in teachers’ work are planning, lesson preparation, marking, and correcting student work. On average across the TALIS participants in primary education, teachers spend 6.7 hours a week on planning and lesson preparation (the equivalent of 16% of their total working time) (Table 6.9). At the lower end of the cross‑national variation, teachers in Turkey dedicate the equivalent of 10% of their total time to preparing for classes while teachers in Denmark, France, Korea and Viet Nam spend the equivalent of 18% to 20% of their total working time on preparation.
Regarding time devoted for marking and correcting student work, teachers spend 4.3 hours a week on marking and correcting (the equivalent of 10% of their total working time). There are some sharp cross‑country variations in the way teachers distribute their time; for example, in the share of time spent on marking and correcting student work, CABA (Argentina) and the United Arab Emirates spent the equivalent of 15% and 13%, respectively, on this task while in Denmark, teachers on average spent only 4% of their total working hours on these tasks. Recent evidence coming from the TALIS‑PISA link study showed that, on average, the more time teachers spend on marking and correcting student work, the stronger students’ self‑concept of doing well on the Programme for International Student Assessment (PISA) test (e.g. their degree of confidence that they will perform well in the PISA test) and expectation of pursuing tertiary studies. While the findings are slightly less robust on student achievement, they also suggest that the more hours teachers spend on marking and correcting student school work, the better it is for student performance (OECD, 2021[38]). That being said, there is an important cross‑country variation in the significance of this association across the participants in the TALIS‑PISA link. A possible explanation is that “marking” can take multiple forms: it can be a validation of students’ responses or a more formative process where there are concrete opportunities for students to correct and respond to their marking. This may show that it is not so much the total time devoted to marking and correcting students but that students are also involved in these evaluations that encourages their growth.
The relationship between the amount of time devoted to teaching, planning and marking deserves further inspection. A more detailed analysis into lower secondary TALIS 2018 data show that on average across the OECD countries a gain of an hour in teaching comes at the expense of reduced time for planning and marking (Boeskens and Nusche, 2021[22]). The data for primary education seem to show the same association since there is a negative correlation between the hours devoted to teaching and the hours devoted to planning (the linear correlation coefficient is r= -.40, based on 13 education systems with available data) and marking and correcting student work, although quite weak (the linear correlation coefficient is r= -.09, based on 13 education systems with available data) (Table 6.9). Even though planning and marking are considered non‑teaching tasks, they are undoubtedly important in quality instruction. Therefore, it is crucial to find the correct balance between these different tasks.
Another set of tasks corresponds to duties that teachers are expected to meet but do not necessarily contribute to quality instruction. Such is the case of time spent on general administrative work, 2.8 hours a week are spent on this tasks (the equivalent of 7% of the total working hours) (Table 6.9). This proportion can reach up to 13% in Korea to 2% for France. It is relevant to monitor the amount of time allocated to administrative work as previous TALIS analysis has shown that teachers’ stress levels are quite sensitive to each additional hour spent on administrative work (OECD, 2020[32]).
In addition, teachers spend 2.1 hours a week counselling students (the equivalent of 5% of total working hours) and 1.2 hours in extracurricular activities (the equivalent of 3% of total working hours). Although time spent on extracurricular activities is not be directly related to good teaching, TALIS‑PISA link analyses show that the more time teachers spend on extracurricular activities, the more students feel that the classroom is disciplined and the teacher is interested and motivated to teach (Table 6.9). These student are also more likely to expect to complete at least a tertiary degree. Time spent with students outside of regular classroom hours is particularly effective in establishing and nurturing good relationships with students (OECD, 2021[38]).
In recognition of the trade‑offs between non‑teaching and teaching time, education systems can provide a framework for regulating how teachers’ instruction hours can be modified if they are assigned additional tasks outside the classroom such as counselling students or participating in extracurricular activities. A recent OECD review of 35 countries found that 16 of them always or sometimes reduce the teaching time of teachers who take on school management tasks. However, it is less common to reduce teaching hours to compensate for tasks directly related to teaching such as planning and marking (Boeskens and Nusche, 2021[22]). Indeed, this might not be the best approach as it can come across as undermining the quality of instruction and delegating these tasks to other school staff damages the profile of teachers as professionals. In order to make teachers’ planning and marking time more productive, structured collaboration with peers may be a more promising strategy in saving teachers’ time than disengaging from these tasks altogether (Boeskens, Nusche and Yurita, 2020[39]).
Many education systems are required by legislation to have additional support staff such as teaching assistants, special needs educators, career counsellors, health professionals, and social workers in primary education. In particular, France, Japan, Korea, Sweden, Turkey and the United Arab Emirates have a considerably diverse profile of professionals working in their schools (see Table A B.5). This allows holistic support to be provided to students and teachers to focus mainly on instruction.
Principals can also play a role in supporting their teachers in their core work, i.e. teaching. TALIS asks school principals about the proportion of time they spend on various activities throughout the school year. Among the six activities listed in the principal questionnaire, one is closely related to supporting teaching in their school: “curriculum and teaching‑related tasks and meetings” (Table 6.15). On average across the TALIS participants in primary education, principals reported spending 20% of their working time on this type of activity. This makes it one of the most time‑consuming task of principals after administrative tasks and meetings (26% of principals’ working time) and leadership tasks and meetings (20%). There are also substantial cross‑country differences in the way school leaders use their time. School leaders spend a higher share of their time on curriculum and teaching‑related tasks in Viet Nam (28%) and Korea (25%) and the lowest percentage in Denmark and Sweden (13% for both). The amount of time spent on tasks that can be understood as a form of instructional leadership (see Chapter 5), may respond to specific organisational patterns that vary cross‑nationally. A recent analysis of TALIS 2018 data found that among the study participants different typologies of school leaders are predominant depending on the balance of administrative tasks and instructional and distributed leadership (Bowers, 2020[40]).
Several factors can drive teachers’ amount of time spent on teaching tasks. In order to explore this association total teaching hours are regressed to class size, student to teacher ratio, and pedagogical personnel to teacher ratio, controlling for crucial teacher and school characteristics such as full‑time employment. The results show that none of the predictors had a consistent result across participants. There is a positive association between the student‑teacher ratio at the school level and teaching hours in 4 out of the 13 participants with available data, meaning that in schools where there are more students per each individual teacher, it is likely that teachers spend more hours teaching (see Table 6.12).
However, the variables related to teachers’ socio‑demographic characteristics, like gender and teaching experience, show a more interesting pattern. Female teachers are more likely to teach more hours than their male colleagues in 6 out of the 13 countries with available data (Table 6.12). The results point to particular gender configurations of the workload. For example, it could be that male teachers are more likely to split their time in other non‑teaching tasks than female colleagues, such as administrative work.
In addition, the more experience a teachers have in their work the lower the number of hours they spend on teaching; that is the case for 5 out 13 countries with available data in primary education (Table 6.12). The results are understandable since more experienced teachers are more likely to devote time in non‑teaching tasks (e.g. being a mentor or counsellor of students; taking on an administrative role) than less experienced teachers. That being said, results do point out as well that less experienced teachers spend more time teaching classes so necessary monitoring and support is needed to avoid classroom situations overwhelming individuals who are new to the profession.
How do primary and lower secondary levels compare?
In comparison to lower secondary education, teachers reported a higher number of working hours in total in their last week of work prior to the survey and more hours devoted to teaching. Also, the total working hours allocated to teaching is 1 hour higher for primary education than in lower secondary education (55% of the total working hours in primary education in contrast to 49% of the total working hours in lower secondary education). On average across the participating countries with available data in both levels, teachers in primary education spend one additional hour working in total that their colleagues in lower education and 2.9 hours more in actual teaching (Figure 6.2 and Table 6.9). The differences in amount of time devoted to teaching are particularly pronounced for CABA (Argentina), the Flemish Community of Belgium, France and Japan, where the gap is around 5 hours or higher for teaching.
Given the trade‑off between teaching and non‑teaching tasks, the amount of time teachers in primary education spend in planning, marking, student counselling, administrative work and extracurricular work is significantly lower than for their colleagues in lower secondary education, although these difference are quite marginal (less than 0.5 hours) (Table 6.9). The most outstanding case in the cross‑level comparison is CABA (Argentina), where the amount of time allocated to all tasks – teaching and non‑teaching alike – is consistently higher than for their colleagues in lower secondary education by at least 1 hour with the exception of engaging in extracurricular activities amounting to a total of 8.6 additional total working hours for teachers in primary education.
Figure 6.2. Teachers’ teaching and non‑teaching hours in primary and lower secondary education
Notes: A “complete” calendar week is one that was not shortened by breaks, public holidays, sick leave, etc. It also includes tasks that took place during weekends, evenings or other out‑of‑class hours.
Analysis restricted to full‑time teachers only.
Countries and economies are ranked in descending order of the number of hours primary education teachers report they spend teaching.
Source: OECD, TALIS 2018 Database.
Teaching, planning and marking in upper secondary education
On average across the TALIS participants in upper secondary education, teachers (full‑time) spend 40.2 hours per week on all the tasks related to their job in their surveyed school of which 20.4 hours are devoted to teaching (see Note 3) (Table 6.9). Just like in primary education, teachers spend slightly more than half of their working time teaching classes (51%). The lowest shares (less than 50% of the total teacher working hours) are observed in Croatia, Denmark, Portugal, Sweden and Viet Nam. In contrast Brazil and Turkey are the only participants where teachers’ hours dedicated to teaching exceeds 60% of the total working hours in a week (65% and 71%, respectively). As mentioned in the previous section, these differences may result from the way teachers’ hours are regulated, which varies among countries, as well as the country‑specific school culture, among other factors – see indicators of the teaching hours per system drawn from Education at a Glance (OECD, 2020[37]) in Table A B.5 of this report.
The next two most time‑consuming activities in teachers’ work for upper secondary education are planning, lesson preparation, marking, and correcting student work. On average across the TALIS participants in upper secondary education, teachers spend 7.8 hours a week on planning and lesson preparation (the equivalent of 19% of their total working time) and five hours a week on marking and correcting (the equivalent of 12% of their total working time). There are some interesting variations in the way teachers distribute their time; in Croatia, Denmark, and Viet Nam, teachers devote around 22-25% of their time preparing lessons whereas teachers in Alberta (Canada) only allocate 14% of their time to these tasks. Just as in primary education Denmark and Viet Nam have the highest share of time devoted to lesson preparation. This indicates flexible institutional arrangements across educational levels. Regarding the time spent on marking and correcting the work of students, it is possible to observe a great degree of variation across participants, ranging from 18% of the total working time in Portugal to only 7% of the share time in Turkey (Table 6.9).
As mentioned in the section on primary education, there can be a trade‑off between the time devoted to teaching and the time needed for marking and correcting student work. The same pattern can be observed for upper secondary education as there is a negative correlation between the hours devoted to teaching and lessons planning (Figure 6.3, the linear correlation coefficient is r= -.71, based on 11 education systems with available data), and marking and correcting student work, although quite weak (the linear correlation coefficient is r= -.22, based on 11 education systems with available data) (Table 6.9).
Figure 6.3. Average number of hours devoted to teaching and planning in upper secondary education
1. A “complete” calendar week is one that was not shortened by breaks, public holidays, sick leave, etc. It also includes tasks that took place during weekends, evenings or other out‑of‑class hours.
Source: OECD, TALIS 2018 Database.
On general administrative work, 2.7 hours of a general working week (the equivalent to 7% of their total working time) is spent on these tasks on average among the TALIS countries participating in upper secondary education. The highest shares, between 8 and 9% of the total share of working hours, can be identified in Slovenia, Sweden and the United Arab Emirates. Regarding other non‑teaching activities teachers spend 2.5 hours a week counselling students (the equivalent of 6% of total working hours) and 1.9 hours on extracurricular activities (the equivalent of 5% of total working hours) (Table 6.9).
TALIS asked school principals about the proportion of time they spend on various activities throughout the school year. Among the six activities listed in the principal questionnaire, one is closely related to supporting teaching in their school: “curriculum and teaching‑related tasks and meetings” (Table 6.15). On average across the TALIS participants in upper secondary education, principals reported spending 17% of their working time on this type of activity. This makes it the third most time‑consuming task of principals after administrative tasks and meetings (28% of principals’ working time) and leadership tasks and meetings (22%). School leaders spend a higher share of their time on curriculum and teaching‑related tasks in Viet Nam (24%) and the United Arab Emirates (20%) and the lowest percentage in Sweden, Portugal and Turkey (13‑14%). As was the case in primary education, spending time in those tasks most related to teaching may not be the top priority for principals but this may have to do with the different national configurations of principals’ role and profiles.
As for primary education, the associations between total teaching hours and class size, student‑to‑teacher ratio, and pedagogical personnel‑to‑teacher ratio are explored, controlling for crucial teacher and school characteristics. As was the case for primary education, teaching experience shows a significant association with the total number of teaching hours; that is, the more years of experience, the lower the number of hours devoted to teaching. That is the case for 5 out of the 11 countries and economies with available data (Table 6.13). This pattern could be explained by teachers’ professional trajectories: as they become more experienced they allocate more hours to non‑teaching tasks (e.g. mentorship, counselling, administrative tasks).
How do upper secondary and lower secondary levels compare?
In comparison to lower secondary education, teachers in upper secondary education reported a lower number of hours devoted to teaching; on average across countries participating in both levels, teachers in upper secondary spend 1.1 hours less in teaching than their colleagues in lower secondary education (Figure 6.4 and Table 6.9). Teachers in upper secondary spend more time in lesson planning and marking and counselling than their colleagues in lower secondary education, although these differences on average are less than one hour. For other tasks such as administrative work and extracurricular activities, there are no significant differences between upper secondary and lower secondary.
Figure 6.4. Teachers’ teaching and non‑teaching hours in upper and lower secondary education
Notes: A “complete” calendar week is one that was not shortened by breaks, public holidays, sick leave, etc. It also includes tasks that took place during weekends, evenings or other out‑of‑class hours.
Analysis restricted to full‑time teachers only.
Countries and economies are ranked in descending order of the number of hours upper secondary education teachers report they spend teaching.
Source: OECD, TALIS 2018 Database.
Satisfaction with the current working environment and the profession
This section examines how satisfied teachers are with their current working environment and profession. Job satisfaction can be understood as the sense of fulfilment and gratification teachers get from their occupation (Ainley and Carstens, 2018[41]). High levels of job satisfaction have a positive association with teachers’ performance (Lortie, 1975[42]; Renzulli, Macpherson Parrott and Beattie, 2011[43]) and self‑efficacy (Mostafa and Pál, 2018[9]), and it has strong implications for retention, attrition, absenteeism, burnout, commitment to education goals and teachers’ job performance (Brief and Weiss, 2002[44]).
Job satisfaction can be analytically divided into two areas: satisfaction with the current work environment and satisfaction with the profession. Research evidence has shown the relevance of distinguishing one from the other as teachers tend to express satisfaction with elements directly related to teaching but dissatisfaction with elements related to working conditions (Crossman and Harris, 2006[45]). Teachers could be satisfied with the teaching profession because it fulfils their personal goals but, at the same time, dissatisfied with their current job and working conditions (Viac and Fraser, 2020[3]).
In addition, the most important motivations for teachers to join the profession are related to the sense of fulfilment they derive from serving the public, for example, by influencing children’s development and contributing to society. Salary and working conditions influence teacher decisions not only to join the profession but also to stay (Bruns, Filmer and Patrinos, 2011[20]).
Schools and education systems need to offer attractive conditions to their staff both in absolute terms and relative to other jobs requiring similar qualifications. In addition, in a variety of countries and economic sectors, employees’ satisfaction with their salaries also depends on the compensation structure, the related incentives and the mechanisms for the remuneration of performance implicit in appraisal systems.
This section will present information on teachers’ and principals’ satisfaction with their salary and the other terms of their employment as collected through the TALIS survey. TALIS 2018 measures job satisfaction among teachers by asking their level of agreement (“strongly disagree”; “disagree”; “agree”; or “strongly agree”) with a set of specific statements covering both positive and negative connotations of their current work environment and their profession.
Job satisfaction in primary education
On average in primary education, a large majority of teachers responded to the following positive or negative statements: “the advantages of being a teacher clearly outweigh the disadvantages” (77% “agree” or “strongly agree”); “if I could decide again, I would still choose to work as a teacher” (77% “agree” or “strongly agree”); “I wonder whether it would have been better to choose another profession” (67% “strongly disagree” or “disagree”); and “I regret that I decided to become a teacher” (91% “strongly disagree” or “disagree”) (Table 6.16). More than two‑thirds of teachers (over 66%) supported these four indicators in CABA (Argentina), the Flemish Community of Belgium, Spain and Viet Nam.
Despite this overall high percentage of teachers in primary education being satisfied with their work, it is important to look closer at the cross‑national variation. For example, 52% of the teachers in primary education in France do not agree that “the advantages of being a teacher clearly outweighs the disadvantages” while 46% of teachers in Denmark and 45% in Turkey “wonder whether it would have been better to choose another profession” (Table 6.16). Having almost half of the workforce wondering about choosing another profession signals dissatisfaction with the teacher profession as a whole in these countries.
If a response of “wonder whether it would have been better to choose another profession” can be interpreted as a proxy measure of teachers’ desire to remain in the profession, part of the cross‑national variation could arise from differences in the vibrancy and diversity of the broader labour markets and, hence, availability of attractive, alternative career options open to teachers in different countries and economies. Nonetheless, several age‑related factors could explain this pattern. From a human capital perspective, younger teachers have accumulated less knowledge and fewer skills that are specific to the occupation. This could make them less risk‑averse and more willing than their more senior colleagues to explore other professional paths.
TALIS also asks teachers about their satisfaction with their current job and work environment. The results display remarkably high levels of satisfaction. For example, on average across countries and economies that took part in the TALIS study, 90% of teachers in primary education reported that, overall, they are satisfied with their job. However, teachers are a bit more nuanced with respect to recommending their school as a good place to work (82% agreement) and 23% express a wish to change to another school if that were possible (Table 6.16). Over 25% of teachers in primary education stated that they would like to change to another school in Japan, Korea, Turkey and the United Arab Emirates.
A breakdown by teachers’ characteristics shows that in most countries and economies novice teachers (less than or equal to five years of experience) are more likely than their experienced counterparts (more than five years of experience) to “choose another school if possible”) (Figure 6.5 and Table 6.17). Although this difference is significant only for two countries; the United Arab Emirates (10 percentage points) and France (9 percentage points). On average, male teachers are also more likely to state they wish to change school than their female colleagues albeit by a small margin (2 percentage point difference). In addition, a higher proportion of teachers working part‑time reported wishing to change to another school if possible compared to teachers working full‑time (2 percentage points of difference); this is true for Korea, Turkey and the United Arab Emirates. This result may reflect the level of commitment of teachers have toward their work as teacher working full‑time are more likely to stay with the same school than part‑time workers. That being said, Japan shows the opposite pattern by having a larger proportion of teachers with full‑time status wishing to change schools than teachers with part‑time status (10 percentage point difference). This may respond to the fact that the career structure of Japan has as one of its components the transfer of teachers to different schools every few years (OECD, 2018[46]). As such, changing schools could be already embedded in the expectations of full‑time teachers.
Figure 6.5. Desire to change school in primary education, by teacher and school characteristics
1. Socio‑economically disadvantaged homes refer to homes lacking the basic necessities or advantages of life, such as adequate housing, nutrition or medical care.
Countries and economies are ranked in descending order of the percentage of teachers who report they would like to change school if that were possible.
Source: OECD, TALIS 2018 Database.
When considering school type, on average across the participants 27% of teachers working in disadvantaged schools expressed that they wish to change to another school if that were possible, in contrast to 22% of teachers in non‑disadvantaged schools (Figure 6.5 and Table 6.17). These differences are observable in Turkey (12 percentage points of difference), Spain (10 percentage points), the Flemish Community of Belgium (10 percentage points) and France (7 percentage points). Usually, disadvantaged schools present more challenging settings for effective instruction due to a student population with higher needs and often a lack of resources. In addition, disadvantaged schools are characterised by a larger share of novice teachers (teachers with less than five years of experience) than in non‑disadvantaged schools (see Table 2.10 in Chapter 2). Given the challenges of working in disadvantaged contexts, novice teachers might feel less prepared and thus want to change schools. In addition, in countries like France, early career teachers do not have a choice of deciding where they teach and they are often deployed in disadvantaged schools.
In that sense, it is interesting to observe those contexts when there are no significant differences in the proportion of teachers wanting to change schools between disadvantaged and non‑disadvantaged schools; that is the case for five countries and economies. For example, Sweden has a low proportion of teachers wanting to change to another school and the percentage gap between different types of schools is marginal at best (0.1 percentage-point gap) (Table 6.17).
On the levels of satisfaction with their salaries, on average across the TALIS participants in primary education, almost half of teachers (47%) reported that they are satisfied with their salaries (Table 6.19). That being said, there is a considerable variation across countries on this perception. More than 50% of teachers reported being satisfied with their salaries in Denmark, the Flemish Community of Belgium, Korea and Viet Nam and less than 20% in CABA (Argentina) and France. The percentage of teachers supporting this statement is particularly elevated in the Flemish Community of Belgium (75%).
In terms of work experience, more experienced teachers are more likely to be satisfied with their salaries than novice teachers in England (United Kingdom), Korea, the United Arab Emirates and Viet Nam but novice teachers are more satisfied with their salaries than their more experienced colleagues in CABA (Argentina), France, Spain and Turkey (Table 6.19). These cross‑national differences may respond to the specific salary scales in each national context. In shaping salary scales there is a trade‑off between awarding high salaries at the entrance level and keeping them relatively stable throughout the career or having relatively middle‑to‑low salaries at the entrance level with significant salary increases across the career.
The data also show significant variation across types of schools. It is quite revealing to see that, on average, teachers working in rural schools are more likely to be satisfied with salaries than teachers in city schools (Table 6.19). The differences are particularly large (over 15 percentage points) in the United Arab Emirates (18 percentage points), Viet Nam (18 percentage points), and Spain (16 percentage points). A possible explanation for this gap could be explained by opportunity costs and cost of living in rural areas. Teachers in rural areas may be satisfied with salaries because of a lack of alternative jobs requiring equivalent skills and offering higher salaries. In cities, given the skills and qualifications of teachers, they might think alternative jobs pay much more. It might be also the case that teacher salaries are fixed at regional or national levels and does not take into consideration the cost of living, which tends to be lower in rural areas.
Teachers responded not only about their satisfaction with salaries but about the extent to which, apart from their salaries, they are satisfied with the terms of their teaching contracts or employment (e.g. benefits, work schedule). On average across the TALIS participants in primary education, 65% of teachers “agree” or “strongly agree” that, apart from their salaries, they are satisfied with their terms of employment (Table 6.19). Interestingly, this indicates that teachers tend to be much more satisfied with their general terms of employment than with their salaries (47%). The proportion of teachers who are satisfied with the other terms of their teaching contract or employment exceeds 70% in England (United Kingdom), the Flemish Community of Belgium, France and Viet Nam, and is the lowest in Denmark (31%) and Japan (48%).
It is interesting to observe that in some countries there is a considerable discrepancy between the share of teachers feeling satisfied with their salaries and the share of teachers feeling satisfied with the conditions of their work. In CABA (Argentina), France and Sweden, the difference between the share of teachers satisfied with the terms of their contract and those by their salaries was of more than 30 percentage points. On the contrary, a higher share of teachers are satisfied with their salaries than with their terms of contract in Denmark (Table 6.19).
What could be driving teacher satisfaction with their terms of employment in primary education? The total number of working hours is a crucial component of terms of employment. A regression analysis between the shares of teachers reporting being satisfied with their terms of employment and the total number of working hours, controlled by other variables such as the possession of a permanent contract, working arrangements (full‑time, part‑time), class size, and other relevant teacher and school characteristics, shows that the higher the number of hours the less likely teachers are satisfied with their terms of employment for 11 out of the 13 countries and economies with available data (Table 6.22). This finding points to the pivotal role that working arrangements have for the satisfaction of teachers. As satisfaction is a crucial consideration for the retention of teachers (OECD, 2020[32]), it is important to balance and efficiently allocate teachers’ working hours.
How do primary education and lower secondary levels compare?
On average the differences across educational levels for countries with available data for primary and lower secondary education is significant but quite small (less than 3 percentage points) (Figure 6.6 and Table 6.16). The proportion of teachers in primary education supporting positive statements on the profession and the work (i.e. “If I could decide again, I would still choose to work as a teacher”, “I enjoy working at this school”, “I would recommend this school as a good place to work”, “I am satisfied with my performance in this school”, “All in all, I am satisfied with my job”) is significant higher than in lower secondary education. Inversely, the proportion of teachers in primary education supporting negative statements on the profession (“I regret that I decided to become a teacher”; “I wonder whether it would have been better to choose another profession”) is lower than in lower secondary education. Overall, it seems that the job satisfaction with the profession and working environment is slighter higher in primary education than in lower secondary education.
For example, the percentage of teachers in primary education stating that they wonder whether it would have been better to choose another profession is lower than the percentage in lower secondary (3 percentage point difference) (Table 6.16). This is true for England (United Kingdom), Japan, Spain, Turkey and Viet Nam. However, the opposite pattern is observed for Denmark.
Regarding satisfaction with the salary, a lower share of teachers in primary education reported being satisfied with their salaries in primary education than their colleagues in lower secondary education in CABA (Argentina), England (United Kingdom), France, Sweden and Turkey (Table 6.19). Teachers’ statutory salaries (after 15 years of service) are indeed lower for teachers in primary education than lower secondary teachers in France and Sweden but are the same across both levels in England (United Kingdom) and Turkey (see Table A B.4). Thus, factors other than actual statutory salaries might play a factor in the satisfaction of teachers with their salaries: perceptions of salaries in similar professions, for instance. Japan is the only country showing a different pattern where more teachers are satisfied with their salaries than in lower secondary education despite the fact that in terms of statutory salaries teachers in primary education earn less than their colleagues in lower secondary.
Finally, concerning terms of employment, there is no consistent pattern across countries although, on average, teachers in primary education are more likely to be satisfied with their terms of employment than teachers in lower secondary education (2 percentage points of difference) (Table 6.19). This is true for England (United Kingdom), the Flemish Community of Belgium, Japan, Korea and Spain. Teachers in primary education are less likely to be satisfied with the terms of employment in CABA (Argentina), Denmark and France.
Figure 6.6. Teachers’ satisfaction with their salaries and terms of employment in primary and lower secondary education
Note: Statistically significant differences between primary education and lower secondary education are marker in a darker tone.
Countries and economies are ranked in descending order of the percentage of teachers in primary education who are satisfied with their salaries.
Source: OECD, TALIS 2018 Database.
Job satisfaction in upper secondary education
On average across the TALIS participants in upper secondary education, more or less the same levels of positive responses about professional satisfaction can be observed: “the advantages of being a teacher clearly outweigh the disadvantages” (78% “agree” or “strongly agree”); “if I could decide again, I would still choose to work as a teacher” (77% “agree” or “strongly agree”); “I wonder whether it would have been better to choose another profession” (62% “strongly disagree” or “disagree”); and “I regret that I decided to become a teacher” (90% “strongly disagree” or “disagree”) (Table 6.16). Croatia, Slovenia and Viet Nam are the only countries where the proportion of teachers expressing satisfaction with the profession exceeds two‑thirds (66%) on all these four indicators. Despite this overall high percentage of teachers in upper secondary education who are satisfied with their work, it is important to look closer at the cross‑national variation. For example, over 40% of the teachers in upper secondary education in Portugal, Turkey and the United Arab Emirates “wonder whether it would have been better to choose another profession”.
Regarding satisfaction with their current job and work environment, the results display remarkably high levels of satisfaction (Table 6.16). For example, 90% of teachers in upper secondary education reported that, all in all, they are satisfied with their job. However, teachers are a bit more nuanced about stating their school as a good place to work (83% agreement) and 23% expressed a wish to change to another school if that were possible. Over 30% of teachers in upper secondary education stated that they would like to change to another school in Turkey and the United Arab Emirates.
The responses on whether teachers want to change to other school varies more across schools than individual characteristics of the teachers (Table 6.18). For example, on average across the participants, 28% of teachers working in disadvantaged schools expressed that they would like to change to another school if that were possible in contrast to just 22% of teachers in non‑disadvantaged schools. These differences are most pronounced in Sweden (11 percentage points difference) and Portugal (7 percentage points). Likewise, teachers in rural schools are more likely to state that they would like to change to other schools compared to their colleagues in city schools; this is true for Alberta (Canada), Portugal and Viet Nam while the opposite pattern can be observed for the United Arab Emirates. Teachers working in school offering VET (vocational education and training) programmes were more likely to report that they would like to change schools compared to their colleagues in non‑VET schools. That being said, the results are only significant in 3 out of 11 countries.
The only teacher characteristic showing an interesting breakdown of teachers’ responses is teaching experience; in most countries and economies, novice teachers (less than or equal to five years of experience) were more likely than their experienced counterparts (more than five years of experience) to report they would “choose another school if possible” (Table 6.18), although this difference is significant only for a few countries such as the United Arab Emirates (9 percentage points) and Turkey (7 percentage points). Interestingly, in Portugal it is possible to observe the opposite distinction where more experienced teachers reported they would like to change schools.
Concerning satisfaction with teacher salary in upper secondary education, 42% of teachers “agree” or “strongly agree” that they were satisfied with their salary in 2018 (Table 6.19). Just like primary education, an important cross‑country variation needs to be taken into account. More than 66% of teachers in upper secondary are satisfied with their salaries in Alberta (Canada) and Denmark while this is the case for 30% of teachers or less in Brazil, Croatia, Portugal, Slovenia and Turkey. Brazil and Portugal show particularly low percentages of teachers in upper secondary satisfied with their salaries, with only 17% and 11% of teachers supporting this statement, respectively.
Novice teachers tend to be more satisfied with their salaries than their experienced colleagues in Brazil, Croatia, Portugal and Turkey (Table 6.21). The contrast is quite remarkable in Portugal (26 percentage‑point gap) with only 10% of experienced teachers stating they are satisfied with their salaries in contrast to 36% of teachers of novice teachers.
As in primary education, the data also show significant variation across types of schools. On average teachers working in rural schools are more likely to be satisfied with their salaries than teachers in city schools in Alberta (Canada), Brazil and the United Arab Emirates while the opposite is true in Viet Nam. In addition, teachers in schools offering VET programmes are more likely than teachers in non‑VET schools to feel satisfied with their salaries in Brazil, Turkey and the United Arab Emirates. The opposite pattern is shown in Denmark, where 44% of teachers in schools offering VET programmes feel satisfied with their salaries in contrast to 75% of their colleagues in non‑VET schools (32 percentage‑point gap).
Regarding the extent to which teachers, apart from their salaries, are satisfied with the terms of their teaching contract or employment (e.g. benefits, work schedule), on average across the participants, 63% of teachers “agree” or “strongly agree” that, apart from their salaries, they are satisfied with their terms of employment. As was the case with primary education, teachers tend to be much more satisfied with their general terms of employment than with their salaries (42%) (Table 6.19). Teachers who are satisfied with the other terms of their teaching contract or employment exceed 70% in Alberta (Canada), Slovenia, Sweden and in Viet Nam, and is the lowest in Brazil (47%) and Portugal (32%). The highest share of teachers feeling satisfied with their terms of their contract than their salaries by a difference of 30 percentage points or more is in Brazil, Croatia, Slovenia and Sweden.
The total number of working hours can be considered a crucial component of the terms of employment. A regression analysis was conducted between the shares of teachers in upper secondary education reporting being satisfied with their terms of employment and the total number of working hours, controlled by other variables such as the possession of a permanent contract, working arrangements (full‑time, part‑time), class size, and other relevant teacher and school characteristics. Results shows that the higher the number of hours the less likely upper secondary teachers are satisfied with their terms of employment for 6 out of the 11 countries and economies (Table 6.23) in the study. This finding points to the pivotal role that working arrangements have for teachers’ satisfaction. That being said, the results were significant for just over a half of the countries taking part in the upper secondary study while for primary education results were significant for almost all participating countries and economies (Table 6.22). The results might be showing different factors driving satisfaction with terms of employment across the levels. For example, unlike in primary education, for five countries the higher the share of low academic achievers in the classroom the less likely teachers reported being satisfied with their terms of employment. Also, for three countries and economies, the higher the share of academically gifted students the more likely teachers reported being satisfied with their work. The results indicate that classroom composition might be an important consideration as well for teachers’ satisfaction.
How do upper secondary and lower secondary levels compare?
Overall satisfaction with the profession and working environment is somewhat higher in upper secondary education than in lower secondary education (Table 6.16). Although the differences are somewhat small (less than 2 percentage points) the proportion of teachers in upper secondary education supporting positive statements about the profession and the work (i.e. “The advantages of being a teacher clearly outweigh the disadvantages”, “If I could decide again, I would still choose to work as a teacher”) is significantly higher than the proportion of teachers in lower secondary education. Inversely, the proportion of teachers in upper secondary education supporting negative statements about the profession (“I regret that I decided to become a teacher”; “I wonder whether it would have been better to choose another profession”) is lower than the percentage of teachers in lower secondary education.
Regarding satisfaction with salary, although, on average, teachers in upper secondary education tend to be less satisfied with their salaries than their colleagues in lower secondary education, the difference is quite small and only significant for a couple of countries (Table 6.19). Concerning terms of employment, teachers in upper secondary tend to be more satisfied than their colleagues in lower secondary education in Denmark, Portugal and Sweden while the opposite is true in Brazil.
Sources and levels of stress
The pressure faced by teachers in their daily tasks may lead to feelings of stress and burnout (OECD, 2019[2]). Work‑related stress can be viewed as an imbalance between work demands and environmental or personal resources at work. Workers can experience stress when the work demands placed on them do not match their support at work, knowledge, skills or ability to cope at work (Kyriacou, 2001[47]). These responses manifest themselves in disturbances to emotional, social, and/or physical health. Research has associated high levels of stress with lower self‑efficacy for teaching, lower job satisfaction, lower commitment (Collie, Shapka and Perry, 2012[10])), burnout (Schaufeli, Leiter and Maslach, 2009[48]) and teachers leaving the profession (Kyriacou, 2001[47]).
Beyond the magnitude and impacts of stress, it is also useful to explore the factors that contribute to teachers’ stress in their work. A relevant conclusion of this research is that the prevalence of stress differs depending on its sources (Bakker and Demerouti, 2007[4]; Hakanen, Bakker and Schaufeli, 2006[8]). For example, stress linked to classroom activities and student interactions seems more predominant than stress related to support or lack thereof from the school and government (Klassen, 2010[49]). The sources of stress for teachers were classified into three different groups following the TALIS 2018 conceptual framework (Ainley and Carstens, 2018[41]): workload stress; student behaviour stress; and stress related to responsiveness to stakeholders.
The current COVID‑19 pandemic may have also exacerbated teachers’ stress and impacted their well‑being due to increased workload, adapting to new technologies and the loss of support networks. And stress is likely to continue as schools reopen with the additional demands from blended learning arrangements in some countries and economies being placed on teachers. This can lead to attrition and burnout. Special attention should be paid to the pressure and expectations placed on teachers along with their working hours and workload.
TALIS 2018 can provide insightful information by identifying the main sources of stress prior to the pandemic. The next section describes the levels of stress reported by teachers and what they reported as their main sources of stress. TALIS 2018 asked teachers, for the first time, to what extent they experience stress in their work (“not at all”; “to some extent”; “quite a bit”; “a lot”). It will further explore the issues behind stress by asking both teachers and principals to what extent a series of work‑related tasks constituted a source of stress (“not at all”; “to some extent”; “quite a bit”; “a lot”).
Levels and sources of stress in primary education
On average across the participants in primary education, 17% of teachers reported experiencing a great deal of stress in their work (Table 6.24). However, there is much variation across the countries and economies participating in TALIS. More than 30% of teachers reported experiencing much stress in England (United Kingdom) and the Flemish Community of Belgium. By contrast, less than 10% of teachers reported experiencing much stress in their work in Turkey and Viet Nam. Previous analysis conducted by TALIS at the lower secondary level have also highlighted the extent of cross‑country variation in the reported levels of stress. This cross‑national differences may obey different institutional and cultural norms as well as individual perceptions of what constitutes a stressful situation (Diener and Tay, 2015[50]; Ng et al., 2009[51]).
Looking at how results differ according to teachers’ characteristics, female teachers reported experiencing stress a lot more frequently than their male peers (18% of female teachers compared to 14% of male teachers) (Table 6.25). Differences are particularly large in the Flemish Community of Belgium (10 percentage points) and the United Arab Emirates (8 percentage points). The differences across gender should be interpreted carefully as they may respond to different working configurations or arrangements (as seen in the previous section) and career expectations defined by gender. That being said, gender differences reported in the levels of stress deserves more scrutiny as it might reflect unbalanced and unequal working conditions in the workplace.
In addition, teachers working in schools in city areas in some countries were more likely to report experiencing much stress than their colleagues in schools in villages or rural areas and schools with lower concentrations of students from socio‑economically disadvantaged homes (Table 6.25). Differences in stress reported by teachers working in schools in different geographic locations are significant in three countries and economies: Spain (14 percentage points), England (United Kingdom) (11 percentage points) and Viet Nam (6 percentage points).
The likelihood of experiencing stress a lot at their work is correlated with hours at work and classroom composition (Table 6.27). In particular, the more hours teachers allocated to marking on a weekly basis the more likely teachers reported experiencing a lot of stress in their work; this is the case for 7 out of the 13 countries and economies with available data (Figure 6.7. Furthermore, the higher the share of low‑achieving students the more likely teachers reported experiencing a lot of stress in their work. The results point to not only how time devoted to concrete tasks such as marking contribute to teachers’ stress but also how classroom composition might affect teachers’ assessment of their levels of stress.
Figure 6.7. Factors associated with experiencing stress a lot in primary education
Note: Results of binary logistic regression based on the responses of teachers and principals in primary education. Controlling for the following: full‑time work, the share of students with special needs in the classroom and the share of academically gifted students in the classroom.
Source: OECD, TALIS 2018 Database.
The results on marking warrant some reflection. It is worth noting that marking is important in teachers’ work in order to track student progress and provide feedback to students when it is used in the form of formative assessments. Time spent on marking is an important trade‑off to consider as it may only be efficient in some conditions. Conditions where marking can be effective include opportunities for students to correct and respond to their marks, i.e. quality of marking over quantity matters. Thus, by making the process of marking more efficient it would be possible to diminish the number of hours devoted to these tasks and reduce the stress levels of teachers (Elliott et al., 2016[52]).
Teachers also shared the extent to which their job negatively affects their mental and physical health. On average across the participants in primary education, 8% of teachers reported that their job negatively affects their mental health a lot while 7% reported that it negatively affects their physical health a lot (Table 6.24). Ten percent or more of teachers answered “a lot” for both indicators in Korea and the United Arab Emirates.
Another indicator of the impact of teachers’ stress is whether the work leaves room for the individual’s own personal time. It is argued that an important element of work‑life balance is having the ability to unwind after work hours or being able to switch off from work responsibilities. To balance the demands of a stressful job, teachers need time to recover, which they may find during the school day, between working days and during more extended periods throughout the school year and at different points of their careers (Boeskens and Nusche, 2021[22]). On average across the participants in primary education, only 9% of teachers consider that their work leaves room for their personal life (Table 6.24). However, there is a great degree of variation across TALIS countries and economies on this indicator. More than 10% of teachers feel that their work leaves time for their personal life in Denmark, the Flemish Community of Belgium, Sweden and the United Arab Emirates. The share is particular high in Denmark (27%). In contrast, only 5% of teachers or less in CABA (Argentina), England (United Kingdom), and Viet Nam agree with this statement.
The research literature has identified workload as a source of stress as it shows a strong association with teachers’ life balance and burnout (Bakker et al., 2007[53]). On average across the participants in primary education, the workload‑related sources of stress reported by teachers (“quite a bit” or “a lot”) are the following: “having too much administrative work to do” (47%); “having too much marking” (36%); “having too much lesson preparation” (36%); “having too many lessons to teach” (30%); and “having extra duties due to absent teachers” (23%) (Table 6.29). Principals in primary education show a very similar pattern to teachers with an average of 64% stating they have too much administrative work to do; 34% reporting having extra duties due to absent teachers and 29% reporting having too much teacher appraisal and feedback work to do (Table 6.31).
To note, there are several countries where more than half (over 50%) reported one of the workload indicators as a source of stress. Some of these countries are Denmark (57% of teachers signalled “having too much administrative work to do” as a source of stress and 62% signalled “having too many lessons to teach” as a source of stress), the Flemish Community of Belgium (70% of teachers signalled “having too much administrative work to do” as a source of stress), France (73% of teachers signalled “having too much lesson preparation” as a source of stress and 57% of teachers signalled “having too much marking” as a source of stress) and the United Arab Emirates (55% of teachers signalled “having too much marking” as a source of stress) (Table 6.29).
Another source of stress relates to managing classrooms and student behaviour. Disruptive pupil behaviours are considered a major cause of psychological strain for teachers (Hakanen, Bakker and Schaufeli, 2006[8]). Three TALIS indicators consider these elements: “being held responsible for students’ achievement” (reported “quite a bit” or “a lot” as a source of stress by 47% of teachers across participating countries and economies in primary education); “maintaining classroom discipline” (41%); and “being intimidated or verbally abused by students” (11%) (Table 6.30). There is a similar response from principals as, on average, 46% stated that being held responsible for students’ achievement is a source of stress; 36% consider maintaining school discipline as a source of stress; and just 8% consider being intimidated or verbally abused by students as a sources of stress (Table 6.31).
In some countries over 50% of teachers reported one of the indicators of managing classroom as a source of stress. Some of these are France (79% of teachers considered “being held responsible for students’ achievement” as a source of stress; 65% considered “maintaining classroom discipline” as a source of stress) and Viet Nam (51% considered “maintaining classroom discipline” as a source of stress) (Table 6.30).
Finally, a last set of indicators refers to teachers’ ability to respond to the requirements and needs of evolving educational systems and stakeholders. The additional tasks generated by these responsibilities can create extra work pressure on teachers and can negatively affect teachers’ sense of professional well‑being (Valli and Buese, 2007[54]). On average across the participants, 42% of teachers consider that “keeping up with changing requirements from local, municipal/regional, state or national/federal authorities” is a predominant source of stress and 42% of teachers are stressed by “addressing parent or guardian concerns”. In addition, shifts in societal demands regarding the inclusion of special needs students in regular schools have brought about additional demands on teachers such as “modifying lessons for students with special needs”, which 37% of teachers reported as a source of stress (Table 6.30). Regarding principals, 49% consider that “keeping up with changing requirements from local, municipal/regional, state or national/federal authorities” is a predominant source of stress; 48% consider “addressing parent or guardian concerns” as source of stress; and 36% consider accommodating students with special needs as a source of stress (Table 6.31).
France shows a consistent high share of teachers identifying these indicators as sources of stress in contrast to the average; 71% of teachers reported as a source of stress “keeping up with changing requirements”; 71% mention “addressing parent or guardian concerns” and 80% mention “modifying lessons for students with special needs” (Table 6.30).
The different share of teachers citing one source or another across countries and the lack of a consistent pattern across participants is somewhat expected as these responses might reflect specific institutional configurations and norm regulations for each country.
How do primary education and lower secondary levels compare?
On average across the participants in primary education, there is a higher share of teachers in primary education stating they experience stress a lot in their work but this difference is marginal (less than 1 percentage point) (Table 6.24). There is also no consistent pattern across participants; for example, in CABA (Argentina) 16% of teachers in primary education reported experiencing stress in their work in contrast to just 8% in lower secondary education while in England (United Kingdom) the opposite pattern is observed with 31% of teachers in primary education reporting stress in contrast to 38% in lower secondary education. Box 6.3 uses a gap decomposition to explore the differences in the levels of stress more deeply.
Box 6.3. Difference in teachers’ workplace well‑being and stress between primary education and lower secondary education – a gap decomposition analysis
For a selected group of countries and economies it was possible to observe that the share of teachers reporting experiencing stress at their work varies across primary and lower secondary education. The differences could be explained by elements pushing teachers’ level of stress, such as working conditions or teachers’ socio‑demographic characteristics.
A Blinder‑Oaxaca decomposition of the differences in teachers’ workplace well‑being and stress across education levels has been used to analyse the effects of mediating factors (Blinder, 1973[55]; Oaxaca, 1973[56]). This type of analysis is able to discern whether observable differences between two groups on a variable of interest reflect underlying differences in factors known to be associated with that variable. In this case, the analysis will be used to understand whether the observed differences in the levels of teachers’ workplace well‑being and stress are due to select factors concerning teacher and school characteristics. As the dependent variable, the scale of teachers’ workplace well‑being and stress was selected; the higher the values of the scale, the higher the levels of stress reported by teachers. The difference in the scale across primary and lower secondary teachers is decomposed to three components: the difference attributed to the gender composition of the workforce, differences attributed to other variables of consideration (such as teachers working arrangements, experience and school characteristics) and differences attributed to variables not captured by the model (Table 6.32).
To start, it is interesting to observe that, for all countries with available data, there is a significant difference between the levels of teachers’ stress between primary and upper secondary education (Figure 6.8). In the case of CABA (Argentina), Denmark, France and the Flemish Community of Belgium, the level of workplace stress are higher among teachers in primary education than in lower secondary. In contrast, for England (United Kingdom), Japan and Korea, the levels of stress and well‑being are higher for teachers in lower secondary than in primary education.
It is important to notice that most of the differences cannot be explained by observable factors included in the model; this residual gap is represented by the white bars. As such, the results presented here should be interpreted cautiously: the main explanation for each gap might be found among variables that are unique to each education level, or to each education system, and which were not included in the survey or in the model.
Figure 6.8. Differences in teachers’ workplace well‑being and stress between primary education and lower secondary education
Notes: Predictors include teachers’ gender (reference: male), highest educational attainment (reference: below ISCED level 5), participation in formal induction at current school, participation in effective forms of professional development, experience as a teacher and in non‑education roles, teaching hours, having a permanent employment contract, class size, the share of students with special needs in the school and the share of students who come from disadvantaged homes in the school.
All differences in the scale between primary education and lower secondary education are statistically significant.
Countries and economies are ranked in descending order of the difference in the index of teachers’ workplace well‑being and stress between primary education and lower secondary education.
Source: OECD, TALIS 2018 Database.
From Figure 6.8 it is possible to observe that the gender composition might explain a part of these differences, in particular, in Denmark, France, the Flemish Community of Belgium and Japan (dark blue). A possible explanation for this situation is that reported levels of stress varies by teachers’ gender; there is a higher share of female teachers reporting stress than male teachers (see Table 6.25). As there are more female teachers in primary education than in lower secondary education, this could partially explain the observed differences. The difference is particularly pronounced in France and in the Flemish Community of Belgium, where a greater share of the gap attributed to gender differences between primary education and lower secondary education can also be observed. Interestingly, gender does not seem such a crucial factor for those countries and economies where the teachers in lower secondary education are the ones showing higher levels of stress (see Table 6.32 for details).
Source: OECD, TALIS 2018 Database.
That being said, the share of teachers identifying different sources of stress contrast considerably between primary education and lower secondary education (Table 6.29). Regarding indicators related to workload, there is a lower share of teachers in primary education than in lower secondary education identifying marking as a source of stress (6 percentage points difference); having too many lessons to teach is a source of stress (2 percentage points difference); and having too much administrative work as a source of stress (1 percentage point difference). On marking, this pattern holds true for 8 out of the 13 countries with available data and Denmark, England (United Kingdom), and Turkey displayed large gaps of at least 15 percentage points. The exception to this trend is the Flemish Community of Belgium, Japan and the United Arab Emirates, where the share of teachers in primary education reporting marking as a source of stress is actually higher. In understanding these results it is important to keep in mind that class size tends to be smaller in primary education (Table 6.14) and teachers spend fewer hours on marking than their colleagues in lower secondary education (Table 6.9). Indeed, the difference in the share of teachers reporting marking as a source of stress is weakly correlated to the difference in class size across education levels (the linear correlation coefficient r is r=.32 among TALIS countries and economies) and hours devoted to marking (the linear correlation coefficient r is r=.42 among TALIS countries and economies).
In addition, a higher share of teachers in primary education than in lower secondary education reported having too much lesson preparation as a source of stress (3 percentage‑point gap) (Table 6.29). This pattern holds true for 6 out of the 13 countries with available data, and France and Japan showed exceptionally large gaps (15 and 9 percentage points, respectively). This difference is revealing as teachers in primary education spend fewer hours planning than their colleagues in lower secondary education (Table 6.9). Teachers in primary education teach multiple subjects, thus, there is less likelihood of replicating lessons across classes. Therefore, they may have less time to prepare multiple lessons and are, hence, more stressed. In the case of principals, there is a higher proportion in primary education than in lower secondary education reporting having too much administrative work to do (4 percentage‑point gap) and extra duties due to absent school staff (4 percentage‑point gap) (Table 6.31) as a source of stress.
Regarding indicators related to classroom management and student behaviour, a higher share of teachers in primary education than in lower secondary education reported being held responsible for student achievement as a source of stress (4 percentage‑point gap) and for maintaining classroom discipline (3 percentage‑point gap) (Figure 6.9 and Table 6.30). In the case of being held responsible for student achievement, this pattern holds true for 7 out of the 13 countries with available data and France showed a large gap of 24 percentage points.
Figure 6.9. Sources of stress in primary and lower secondary education
Note: Statistically significant differences between primary education and lower secondary education are marked in a darker tone.
Source: OECD, TALIS 2018 Database.
In contrast to the results for teachers, there is a lower share of principals in primary education than in lower secondary education stating that maintaining school discipline is a source of stress (9 percentage‑ point gap) (Table 6.31). It is important to keep in mind that principals and teachers are speaking at different levels; while principals are addressing discipline at the school level as a source of stress, teachers are referring to classroom discipline.
A lower share of teachers in primary education reported being stressed because of intimidation or verbal abuse by students than their colleagues in lower secondary education (2 percentage points difference) (Figure 6.9 and Table 6.30). This is true for 8 out of the 13 countries with available data. In England (United Kingdom), the Flemish Community of Belgium, France and Korea this gap is of 5 percentage points or more. Noteworthy exceptions are CABA (Argentina) and Denmark where there is a higher percentage of teachers reporting stress because of abuse in primary education than in lower secondary education (5 and 7 percentage‑point gaps, respectively). Again, this contrasts with the pattern observed for principals, where there is a higher share in primary education than in lower secondary education reporting intimidation and abuse by students as a source of stress. Then again, the difference is of only 2 percentage points and this holds true only for France (16 percentage-point gap) (Table 6.31).
In the final group of indicators related to the evolving needs of educational systems, a consistent pattern can be observed where inclusive education is a bigger cause of stress in primary education than in lower secondary education. A higher share of teachers in primary education than in lower secondary education reported as a source of stress modifying lessons for students with special needs (8 percentage‑point gap); addressing parents’ or guardians’ concerns (7 percentage‑point gap) and keeping up with changing requirements (1 percentage‑point gap) (Figure 6.9 and Table 6.30). For the modification of lesson for special needs students, this is true for all participants except Sweden and Viet Nam. A higher share of principals in primary education than in lower secondary education reported accommodating students with special needs as a source of stress (8 percentage‑point gap). This holds true for CABA (Argentina), England (United Kingdom), the Flemish Community and France with gaps of at least 14 percentage points (Table 6.31).
The results could be explained by teachers not being prepared for dealing with students with special needs. Indeed, results show that the share of teachers in primary education reporting a need for professional development in teaching students with special needs is higher than in lower secondary education (see Table 4.24 in Chapter 4). In this case, the correlation between the difference in the share reporting modifying lessons as a source of stress and the difference in the share of teachers reporting professional development needs is mild and positive (the linear correlation coefficient r is r=.54).
Another explanation could be that there is a larger share of principals reporting working with schools with more than 10% of students with special needs in primary education than in lower secondary education (see Table 2.24 in Chapter 2). However, the level of association between the difference in the share of teachers reporting modifying lessons as a source of stress and the difference in the share of schools with students with special needs is positive but weak (the linear correlation coefficient r is r=.31). A similar result is obtained when correlating the difference in the share of principals signalling accommodating students with special needs as a source of stress and the difference in the share of schools with more 10% of students with these characteristics (the linear correlation coefficient r is r=.47).
In terms of addressing parents’ or guardians’ concerns, this pattern holds true for 9 out of the 13 countries with available data and CABA (Argentina), Denmark and France show important gaps of 15 percentage points or more (Figure 6.9 and Table 6.30). In terms of keeping up with changing requirements, the pattern holds true for 5 out of the 13 countries with available data. Spain shows the opposite pattern: there is a lower share of teachers in primary than lower secondary education stating that keeping up with requirements is a source of stress.
Level and sources of stress in upper secondary
On average across the participants in upper secondary education, 15% of teachers reported experiencing stress a lot in their work (Table 6.24). More than 20% of teachers reported experiencing stress a lot in Alberta (Canada), Portugal and the United Arab Emirates. In contrast, less than 10% of teachers reported experiencing stress a lot in their work in Croatia, Turkey and Viet Nam.
Female teachers reported experiencing stress a lot more frequently than their male peers (17% of female teachers compared to 13% of male teachers) (Table 6.26). This difference is statistically significant in all participants except for Brazil, Slovenia and Turkey. Differences are particularly large in Portugal (11 percentage points) and the United Arab Emirates (8 percentage points).
As is the case in primary education, the likelihood of experiencing much stress at work is correlated with hours at work and classroom composition in upper secondary education (Figure 6.10 and Table 6.28). In 6 out of 11 countries and economies, the more hours teachers allocate to marking on a weekly basis the more likely teachers reported experiencing a lot of stress in their work. Furthermore, in 9 out of 11 countries and economies, the higher the share of low‑achieving students, the more likely teachers reported experiencing a lot of stress in their work. The results point to not only how time devoted to concrete tasks such as marking, contribute to teachers’ stress but also how classroom composition might affect teachers’ assessment of their levels of stress.
Figure 6.10. Factors associated with experiencing stress a lot in upper secondary education
Note: Results of binary logistic regression based on the responses of teachers and principals in upper secondary. Controlling for the following: full‑time work, the share of students with special needs in the classroom and the share of academically gifted students in the classroom
Source: OECD, TALIS 2018 Database.
In addition, 7% of teachers reported that their job negatively affects their mental health a lot while 6% reported that it negatively affects their physical health a lot (Table 6.24). Ten percent or more of teachers answered “a lot” for both indicators in Portugal and the United Arab Emirates.
Regarding the impact of teachers’ stress is whether work leaves room for the individual’s own personal time. On average across the participants in upper secondary education, only 9% of teachers consider that their work leaves room for their personal life (Table 6.24). At least 10% of teachers consider their work leaves time for their personal life in Denmark, Sweden and the United Arab Emirates. The share is particular high in Denmark (23%). In contrast, only 5% of teachers or less in Alberta (Canada) and Viet Nam agree with this statement.
Regarding the sources of stress related to workload, teachers and principals were asked about “having too much marking” (46%); “having too much administrative work to do” (45%); “having too much lesson preparation” (36%); “having too many lessons to teach” (36%); and “having extra duties due to absent teachers” (21%) (Table 6.29). Principals in upper secondary education show a very similar pattern to teachers with an average of 63% stating they have too much administrative work to do; 33% reporting having extra duties due to absent teachers; and 32% reporting having too much teacher appraisal and feedback work to do (Table 6.31).
The countries showing more than half of teachers (over 50%) identifying one of these indicators as a source of stress are Denmark (60% of teachers signalled “having too much marking”), Portugal (69% of teachers signalled “having too much administrative work to do” as a source of stress; 62% signalled “having too much lesson preparation” as a source of stress; 76% of teachers signalled “having too much marking” as a source of stress; and 71% of teachers signalled “having too much administrative work to do”), Sweden (60% of teachers reported “having too much administrative work to do”) and the United Arab Emirates (55% of teachers reported “having too much marking” and 52% stated “having too much lesson preparation”) (Table 6.29).
Regarding the sources of stress related to managing classrooms and student behaviour, the results are the following for upper secondary education: “being held responsible for students’ achievement” (reported “quite a bit” or “a lot” as a source of stress by 47% of teachers across the 11 participants); “maintaining classroom discipline” (34%); and “being intimidated or verbally abused by students” (11%) (Table 6.30). For principals 52% stated that being held responsible for students’ achievement is a source of stress, 41% consider maintaining school discipline as a source of stress and 11% considers being intimidates or verbally abused by students as a sources of stress (Table 6.31).
The countries showing more than half of teachers (over 50%) identifying one of these indicators as a sources of stress are Brazil (54% of teachers reported as a source of stress “being held responsible for students’ achievement”), Portugal (81% of teachers signalled being held responsible for students’ achievement as a source of stress and 60% signalled “maintaining classroom discipline” as a source of stress), Slovenia (53% of teachers reported as a sources of stress “being held responsible for students’ achievement”), the United Arab Emirates ( 53% of teachers reported as a sources of stress “being held responsible for students’ achievement”) and Viet Nam (55% of teachers reported as a sources of stress “being held responsible for students’ achievement”) (Table 6.30).
The last set of indicators refers to teachers’ ability to respond to the requirements and needs of evolving educational systems and stakeholders. On average across the participants in upper secondary, 34% of teachers consider that “keeping up with changing requirements from local, municipal/regional, state or national/federal authorities” is a predominant source of stress; 27% of teachers do so with respect to “addressing parent or guardian concerns”; 26% of teachers reported modifying lessons for students with special needs as a source of stress (Table 6.30). Regarding principals, 47% consider that “keeping up with changing requirements from local, municipal/regional, state or national/federal authorities” is a predominant source of stress; 34% consider “addressing parent or guardian concerns” as source of stress; and 23% consider modifying lessons for students with special needs as a source of stress (Table 6.31).
The country showing more than half of teachers identifying one of these indicators as a source of stress is Portugal (55% signalled “addressing parental concerns” as a source of stress and 53% of teachers signalled “modifying lessons for students with special needs” as a source of stress) (Table 6.30).
How do upper secondary and lower secondary levels compare?
On average across the participants in upper secondary education, there is a lower share of teachers in upper secondary education stating they experience stress a lot in their work but this difference is marginal (less than 1 percentage point) (Table 6.24).
Regarding the indicators related to workload as a source of stress, there is lower share of teachers in upper secondary than in lower secondary identifying as a source of stress having too much administrative work (3 percentage‑point gap) and having extra duties due to absent teachers (3 percentage‑point gap) (Table 6.29). Regarding administrative work, this pattern is true for 5 out of the 11 countries with available data with Croatia, Denmark and Slovenia displaying the largest gap (over 10 percentage points). The opposite pattern can be observed for Alberta (Canada). In the case of having duties due to absent teachers the patterns holds true just for 4 countries, with Sweden showing the largest at 20 percentage points.
A higher share of teacher in upper secondary education reported “having too much marking” as a source of stress than their colleagues in lower secondary education (2 percentage‑point gap) (Table 6.29). This is true for 4 out of the 11 countries with available data, with Brazil and Denmark showing the largest gap (over 5 percentage points). Portugal is the only country showing the opposite pattern.
In the case of principals, there is a lower proportion in upper secondary education than in lower secondary education reporting extra duties due to absent school staff as a source of stress (6 percentage‑point gap) (Table 6.31).
Regarding indicators related to classroom management and student behaviour, a lower share of teachers in upper secondary education than in lower secondary education reported as a source of stress being held responsible for student achievement (2 percentage‑point gap); maintaining classroom discipline (6 percentage‑point gap) and being intimidated or verbally abused by students (3 percentage‑point gap) (Figure 6.11 and Table 6.30). On managing classrooms as a source of stress, this pattern holds true for most participants with available data and Alberta (Canada), Denmark, Portugal, Slovenia and Sweden show particularly large gaps (9 percentage points or higher). A possible explanation is that teachers might feel better prepared to manage student behaviour and classroom management after concluding their training than their colleagues in lower secondary education (see Table 3.13 in Chapter 3). Results showed that the difference in the share of teachers reporting classroom management as a source of stress across levels is mildly correlated with the share of teachers feeling prepared for classroom management (the linear correlation coefficient r is r=.-52 among TALIS countries and economies). The results are also somewhat echoed by a lower share of principals in upper secondary education than in lower secondary education reporting managing school discipline as a source of stress (4 percentage points difference).
Figure 6.11. Sources of stress in upper and lower secondary education
Note: Statistically significant differences between upper secondary education and lower secondary education are marked in a darker tone.
Source: OECD, TALIS 2018 Database.
Finally, regarding the evolving needs of educational systems, a consistent pattern can be observed where a lower share of teachers in upper secondary than lower secondary education reported as a source of stress the modification of lessons for special needs students (10 percentage‑point gap); addressing parents’ or guardians’ concerns (10 percentage‑point gap); and keeping up with changing requirements (3 percentage‑point gap) (Figure 6.11 and Table 6.30). This is reflected for principals as well as there is a lower share of principals in upper secondary than in lower secondary reporting the modification of lesson for special need students as a sources of stress (5 percentage‑point gap) and addressing parents’ or guardians’ concerns (13 percentage‑point gap) (Table 6.31).
For the modification of lessons for special needs, this is true for teachers in all participating countries and economies except Viet Nam. Croatia, Denmark, Portugal, Slovenia and Sweden showed exceptionally large gaps of 14 percentage points or more (Table 6.30). Denmark showed a particularly large gap of 30 percentage points; 17% of teachers in upper secondary education consider that modifying lessons for students with special needs is a source of stress in contrast with 47% in lower secondary education. As in the case of primary education, the difference in the share of teachers reporting stress could be explained by structural features such as the fact that there is a lower share of schools with more than 10% of students with special needs in upper secondary education than in lower secondary education (see Table 2.24 in Chapter 2). However, the level of association between the difference in the share of teachers reporting modifying lessons as a source of stress and the difference in the share of students with special needs in the schools is positive but weak (the linear correlation coefficient r is r=.27). Level of training could also be an explanation as the share of teachers with a need for professional development in teaching students with special needs is lower in upper secondary than in lower secondary. The correlation between the difference in the share of teachers reporting modifying lessons as a source of stress and the difference in the share of teachers reporting a need for professional development is positive and mild (the linear correlation coefficient r is r=.65).
For addressing parents’ or guardians’ concerns, this pattern holds true for 9 out of the 11 countries with available data and Croatia, Denmark, Slovenia, Sweden and Turkey show important gaps of 10 percentage points or more (Table 6.30). Once again, Denmark shows a particularly large of 34 percentage points; 4% of teachers in upper secondary education consider that addressing parents’ or guardians’ concerns is a source of stress in contrast to 38% in lower secondary education.
Teachers’ risk of attrition
Attrition (which refers to teachers permanently leaving their profession) has become a severe problem. Attrition can have a detrimental impact on student learning (Borman and Dowling, 2008[5]; Ronfeldt, Loeb and Wyckoff, 2013[57]) and can be a severe problem affecting particular school contexts such as staff shortages in challenging environments, disadvantaged schools (Boe and Cook, 2006[58]; Ingersoll, 2001[59]). Such actions also imply significant financial costs for educational systems as they need to replace qualified teachers in the affected schools (Barnes, Crowe and Schaefer, 2007[60]). Finally, attrition also entails efficiency costs for schools as they need to spend time and resources integrating new teachers into the school organisation and culture (Darling-Hammond and Sykes, 2003[61]).
TALIS 2018 includes questions that may function as proxies for measuring attrition and are able to capture this form of absence and provide a descriptive picture of the situation.5 One proxy measure for the risk of attrition is teachers’ intention to remain in teaching: teachers were asked how many more years they would like to keep working. It is possible that although teachers and school leaders reported planning to stop work relatively soon, they may want to continue in the profession in another capacity, such as on the school management team or a role outside the school, such as in the local or national administration or as a researcher. Nevertheless, whatever plans teachers and school leaders may have, the indicator provides an idea of when they expect to stop being in the classroom or in charge of the school.
Teachers’ intention to leave teaching in primary education
On average across the participants in primary education, teachers reported that they would like to continue working as teachers for an additional 15 years (Table 6.34). Since the average age of teachers is 41 (see Table 2.1 in Chapter 2), an additional 15 years takes teachers close to retirement age for the majority of countries and economies participating in primary education. The data does not show a great degree of cross‑country variation as the range goes from 13.5 additional years in Turkey to 18.6 years in the Flemish Community of Belgium.
In order to identify countries and economies experiencing more pressing concerns in retaining and providing the necessary support to the teaching workforce, TALIS looked at the percentages of teachers who want to leave teaching within the next the next five years. Across participants, 21% of teachers stated they want to leave teaching within the next five years (Table 6.34). The countries and economies with particularly high shares (equal to or above 25) are Sweden (29%), the United Arab Emirates (29%), and Denmark (26%). The average age of the teacher population workforce could explain these high percentages; Denmark and Sweden have a teacher population older than the average – see Table 2.1 in Chapter 2. That being said, there is a weak country‑level correlation between teachers wanting to leave teaching and the proportion of teachers age 50 and above (the linear correlation coefficient r is r=.13 among TALIS countries and economies). Thus, these results might be explained by other factors besides the regular life cycle of the teacher workforce in each country.
In looking at the breakdown of these results, on average across teachers in primary education, female teachers are more likely to state they would like to leave teaching in the next five years than their male colleagues (Table 6.35). This is true for 7 out of the 13 countries with available data. Japan presents a particularly large gap of 9 percentage points. In contrast, male teachers are more likely to state they want to leave teaching in the next five years than their female colleagues in Denmark and Turkey. As primary education is an education level that is particularly struggling to attract male candidates it would be important to understand reasons why they want to leave the profession in these two countries.
Regarding differences between novice and experienced teachers, on average more experienced teachers are more likely to state they want to leave the teaching profession in the next five years than novice teachers (Table 6.35). This is not surprising as experience is intertwined with age and the life cycles of teachers.
Regarding the main factors associated with teachers wanting to leave their work, stress plays a crucial role. Teachers who reported experiencing a great deal of stress in their work are more likely to state they would like to leave teaching in the next five years in 12 out of the 13 countries with available data even if control by satisfaction with salaries, terms of employment and professional collaboration (Figure 6.12 and Table 6.39). That being said, satisfaction with salaries and terms of employment plus professional collaboration have a negative association with intention to leave teaching in around half of the countries, which might be pointing to a possible role in retaining teachers.
Figure 6.12. Factors associated with the intention to leave teaching in the next five years in primary education
Note: Results of binary logistic regression based on the responses of teachers and principals in primary education. Controlling for the following: gender, work experience as a teacher, the share of low academic achievers in the classroom, the share of students with special needs in the classroom and the share of academically gifted students in the classroom.
Source: OECD, TALIS 2018 Database.
In addition, teachers who reported that there is a collaborative school culture were less likely to report they will leave teaching in the next five years in 9 out of the 13 countries with available data and the same is the case for teachers reporting that the school provides staff with opportunities to participate in school decisions in 6 out of the 13 countries with available data (Table 6.37).
What do these results show? They point to the pivotal role that teachers’ stress levels play in their intention to remain in the profession. If not attended to, stress levels may translate into more than just intentions to leave but actual attrition from the profession (Weiss, 1999[16]) That being said, coping mechanisms can lessen the impact that stress has on teachers’ intention to leave the profession (Kyriacou, 2001[47]). In particular, school support and peer collaboration can play an important role in improving the well‑being of teachers (Bakker et al., 2007[53]; Borman and Dowling, 2008[5]; Collie, Shapka and Perry, 2012[10]; Desrumaux et al., 2015[11]; Hakanen, Bakker and Schaufeli, 2006[8]). TALIS 2018 findings echo these results as participation in decision making and collaborative environments are negatively correlated with the intention to leave teaching. Finally, working engagement and arrangement such as terms of employment and salaries are also attractive elements to keep teachers in the profession (Bakker et al., 2007[53]). Interestingly, results show that terms of employment is a significant factor in more countries and economies than salaries, which points to the importance of the working conditions of teachers for their satisfaction.
How do primary education and lower secondary levels compare?
Overall, there are no major differences in the intention to leave teaching between primary education and lower secondary education. On average across countries with available data, teachers in primary education are less likely to state they would like to leave in the next five years than their colleagues in lower secondary education, although the percentage difference is quite small (less than 1 percentage point difference) (Table 6.34). This is true for England (United Kingdom), Korea and Sweden. That being said, the opposite pattern can be observed in Turkey and Viet Nam, where a higher percentage of teachers in primary education stated they would like to leave the profession.
Teachers’ intention to leave teaching in upper secondary education
On average across the participants in upper secondary education, teachers reported that they would like to continue working as teachers for an additional 14.6 years (Table 6.34). Since the average age of teachers across is 44 (see Table 2.1 in Chapter 2), an additional 15 years takes teachers in upper secondary education close to retirement age for the majority of participating countries and economies. The data does not show a great degree of cross‑country variation as the range goes from 12.1 additional years in Sweden to 18.9 years in Viet Nam.
In order to identify those countries and economies experiencing more pressing concerns to renew their teaching workforce, TALIS looked at the percentages of teachers who want to leave teaching within the next five years. Across participants, 23% of teachers stated they want to leave teaching within the next five years (Table 6.34). The countries and economies with particularly high percentages (equal to or above 25) are Sweden (34%), the United Arab Emirates (29%), Croatia (27%), Denmark (26%) and Slovenia (25%). Results showed that there is a mild country‑level correlation between teachers wanting to leave teaching and the proportion of teachers age 50 and above (the linear correlation coefficient r is r=.42 among TALIS countries and economies).
Stress plays a crucial role in the desire to leave teaching within the next five years. Teachers who reported experiencing much stress in their work are more likely to state they would like to leave teaching in the next five years in 8 out of the 11 countries with available data, even if controlled by teachers’ satisfaction with salaries, satisfaction with terms of employment, professional collaboration in lessons among teachers and other key teachers and school characteristics (Table 6.40). That being said, satisfaction with salaries and terms of employment plus professional collaboration have a negative association with intention to leave teaching in around half of the countries, which might be pointing to a possible role in retaining teachers (Figure 6.13).
Figure 6.13. Factors associated with the intention to leave teaching in the next five years in upper secondary education
Note: Results of binary logistic regression based on the responses of teachers and principals in upper secondary education. Controlling for the following: gender, work experience as a teacher, the share of low academic achievers in the classroom, the share of students with special needs in the classroom and the share of academically gifted students in the classroom.
Source: OECD, TALIS 2018 Database.
Looking deeper into school factors that could be help retention, teachers who reported that there is a collaborative school culture were less likely to report they will leave teaching in the next five years in 4 out of the 11 countries with available data. The same is found for teachers reporting school provides staff with opportunities to participate in school decisions in 8 out of the 11 countries and economies with available data (Table 6.37).
Results point to the pivotal role that teachers’ stress levels play in their intention to remain in the profession. If not attended to, stress levels can translate into more than just intentions to leave, but, rather, actual attrition from the profession (Weiss, 1999[16]) That being said, coping mechanisms can lessen the impact that stress can have on teachers’ intentions to leave the profession (Kyriacou, 2001[47]). In particular, school support and peer collaboration can play an important role in improving the well‑being of teachers (Bakker et al., 2007[53]; Borman and Dowling, 2008[5]; Collie, Shapka and Perry, 2012[10]; Desrumaux et al., 2015[11]; Hakanen, Bakker and Schaufeli, 2006[8]). TALIS 2018 findings echo these results as participation in decision making and collaborative environments are negatively correlated with the intention to leave teaching. In contrast to primary education, participation in decision making seems to be a particularly salient factor for teachers’ intention to remain in teaching. Finally, working engagement and arrangement such as terms of employment and salaries are also attractive elements to keep teachers in the profession (Bakker et al., 2007[53]). Interestingly, results show that terms of employment is a significant factor in more countries and economies than salaries, which points to the importance of the working conditions of teachers for their satisfaction.
How do upper secondary education and lower secondary levels compare?
On average across the countries with available data, teachers in upper secondary education are more likely to state they would like to leave the teaching profession than their colleagues in lower secondary education. This is true for 5 out of the 11 countries with available data. The difference is particularly pronounced for Turkey with a 9 percentage‑point gap. The only exception to this pattern is Viet Nam, where teachers in upper secondary education are less likely to state they want to leave the profession than their colleagues in lower secondary education (Table 6.34).
References
[41] Ainley, J. and R. Carstens (2018), “Teaching and Learning International Survey (TALIS) 2018 Conceptual Framework”, OECD Education Working Papers, No. 187, OECD Publishing, Paris, https://dx.doi.org/10.1787/799337c2-en.
[4] Bakker, A. and E. Demerouti (2007), “The Job Demands‐Resources model: State of the art”, Journal of Managerial Psychology, Vol. 22/3, pp. 309-328, http://dx.doi.org/10.1108/02683940710733115.
[53] Bakker, A. et al. (2007), “Job resources boost work engagement, particularly when job demands are high”, Journal of Educational Psychology, Vol. 99/2, pp. 274-284, http://dx.doi.org/10.1037/0022-0663.99.2.274.
[60] Barnes, G., E. Crowe and B. Schaefer (2007), The Costs of Turnover in Five Districts. A Pilot Study, National Commission on Teaching and America’s Future, Washington, DC.
[18] Bertoni, E. et al. (2018), School Finance in Latin America: A Conceptual Framework and a Review of Policies, Inter-American Development Bank, Washington, D.C., http://dx.doi.org/10.18235/0001306.
[14] Betoret, F. (2009), “Self-efficacy, school resources, job stressors and burnout among Spanish primary and secondary school teachers: A structural equation approach”, Educational Psychology, Vol. 29/1, pp. 45-68, http://dx.doi.org/10.1080/01443410802459234.
[55] Blinder, A. (1973), “Wage discrimination: Reduced form and structural estimates”, The Journal of Human Resources, Vol. 8/4, pp. 436-455, http://dx.doi.org/10.2307/144855.
[58] Boe, E. and L. Cook (2006), “The chronic and increasing shortage of fully certified teachers in special and general education”, Exceptional Children, Vol. 72/4, pp. 443-460, http://dx.doi.org/10.1177/001440290607200404.
[22] Boeskens, L. and D. Nusche (2021), “Not enough hours in the day: Policies that shape teachers’ use of time”, OECD Education Working Papers, No. 245, OECD Publishing, Paris, https://dx.doi.org/10.1787/15990b42-en.
[39] Boeskens, L., D. Nusche and M. Yurita (2020), “Policies to support teachers’ continuing professional learning: A conceptual framework and mapping of OECD data”, OECD Education Working Papers, No. 235, OECD Publishing, Paris, https://dx.doi.org/10.1787/247b7c4d-en.
[5] Borman, G. and N. Dowling (2008), “Teacher attrition and retention: A meta-analytic and narrative review of the research”, Review of Educational Research, Vol. 78/3, pp. 367-409, http://dx.doi.org/10.3102/0034654308321455.
[40] Bowers, A. (2020), “Examining a congruency-typology model of leadership for learning using two-level latent class analysis with TALIS 2018”, OECD Education Working Papers, No. 219, OECD Publishing, Paris, https://dx.doi.org/10.1787/c963073b-en.
[44] Brief, A. and H. Weiss (2002), “Organizational behavior: Affect in the workplace”, Annual Review of Psychology, Vol. 53/1, pp. 279-307, http://dx.doi.org/10.1146/annurev.psych.53.100901.135156.
[20] Bruns, B., D. Filmer and H. Patrinos (2011), Making Schools Work: New Evidence on Accountability Reforms, The World Bank, Washington, DC, http://dx.doi.org/10.1596/978-0-8213-8679-8.
[6] Cochran-Smith, M. (2004), “The problem of teacher education”, Journal of Teacher Education, Vol. 55/4, pp. 295-299, http://dx.doi.org/10.1177/0022487104268057.
[7] Collie, R. and A. Martin (2017), “Teachers’ sense of adaptability: Examining links with perceived autonomy support, teachers’ psychological functioning, and students’ numeracy achievement”, Learning and Individual Differences, Vol. 55, pp. 29-39, http://dx.doi.org/10.1016/j.lindif.2017.03.003.
[10] Collie, R., J. Shapka and N. Perry (2012), “School climate and social-emotional learning: Predicting teacher stress, job satisfaction, and teaching efficacy”, Journal of Educational Psychology, Vol. 104/4, pp. 1189-1204, http://dx.doi.org/10.1037/a0029356.
[45] Crossman, A. and P. Harris (2006), “Job satisfaction of secondary school teachers”, Educational Management Administration & Leadership, Vol. 34/1, pp. 29-46, http://dx.doi.org/10.1177/1741143206059538.
[61] Darling-Hammond, L. and G. Sykes (2003), “Wanted, a national teacher supply policy for education: The right way to meet the ’highly qualified teacher’ challenge”, Education Policy Analysis Archives, Vol. 11/33, https://doi.org/10.14507/epaa.v11n33.2003.
[11] Desrumaux, P. et al. (2015), “The impact of job demands, climate, and optimism on well-being and distress at work: What are the mediating effects of basic psychological need satisfaction?”, European Review of Applied Psychology, Vol. 65/4, pp. 179-188, http://dx.doi.org/10.1016/j.erap.2015.06.003.
[50] Diener, E. and L. Tay (2015), “Subjective well-being and human welfare around the world as reflected in the Gallup World Poll”, International Journal of Psychology, Vol. 50/2, pp. 135-149, http://dx.doi.org/10.1002/ijop.12136.
[52] Elliott, V. et al. (2016), A Marked Improvement? A Review of the Evidence on Written Marking, Education Endowment Foundation, London, https://educationendowmentfoundation.org.uk/public/files/Presentations/Publications/EEF_Marking_Review_April_2016.pdf.
[8] Hakanen, J., A. Bakker and W. Schaufeli (2006), “Burnout and work engagement among teachers”, Journal of School Psychology, Vol. 43/6, pp. 495-513, http://dx.doi.org/10.1016/j.jsp.2005.11.001.
[34] Hargreaves, A. (1992), “Time and teachers’ work: An analysis of the intensification thesis”, Teachers College Record, Vol. 94/1, pp. 87-108, https://www.tcrecord.org/content.asp?contentid=179.
[59] Ingersoll, R. (2001), “Teacher turnover and teacher shortages: An organizational analysis”, American Educational Research Journal, Vol. 38/3, pp. 499-534, http://dx.doi.org/10.3102/00028312038003499.
[49] Klassen, R. (2010), “Teacher stress: The mediating role of collective efficacy beliefs”, The Journal of Educational Research, Vol. 103/5, pp. 342-350, http://dx.doi.org/10.1080/00220670903383069.
[12] Klassen, R. et al. (2009), “Exploring the validity of a teachers’ self-efficacy scale in five countries”, Contemporary Educational Psychology, Vol. 34/1, pp. 67-76, http://dx.doi.org/10.1016/j.cedpsych.2008.08.001.
[47] Kyriacou, C. (2001), “Teacher stress: Directions for future research”, Educational Review, Vol. 53/1, pp. 27-35, https://doi.org/10.1080/00131910120033628.
[21] Latifoglu, A. (2016), “Staying or leaving? An analysis of career trajectories of beginning teachers”, International Studies in Educational Administration, Vol. 44/1, pp. 55-70.
[42] Lortie, D. (1975), Schoolteacher: A Sociological Inquiry, University of Chicago Press, Chicago, IL.
[9] Mostafa, T. and J. Pál (2018), “Science teachers’ satisfaction: Evidence from the PISA 2015 teacher survey”, OECD Education Working Papers, No. 168, OECD Publishing, Paris, https://dx.doi.org/10.1787/1ecdb4e3-en.
[51] Ng, W. et al. (2009), “Affluence, feelings of stress, and well-being”, Social Indicators Research, Vol. 94/2, pp. 257-271, http://dx.doi.org/10.1007/s11205-008-9422-5.
[26] Nusche, D. et al. (2015), OECD Reviews of School Resources: Flemish Community of Belgium 2015, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264247598-en.
[56] Oaxaca, R. (1973), “Male-female wage differentials in urban labor markets”, International Economic Review, Vol. 14/3, pp. 693-709, http://dx.doi.org/10.2307/2525981.
[31] OECD (2021), Education in Brazil: An International Perspective, OECD Publishing, Paris, https://dx.doi.org/10.1787/60a667f7-en.
[38] OECD (2021), Positive, High-achieving Students?: What Schools and Teachers Can Do, TALIS, OECD Publishing, Paris, https://dx.doi.org/10.1787/3b9551db-en.
[17] OECD (2021), The State of School Education: One Year into the COVID Pandemic, OECD Publishing, Paris, https://dx.doi.org/10.1787/201dde84-en.
[37] OECD (2020), Education at a Glance 2020: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/69096873-en.
[25] OECD (2020), Lessons for Education from COVID-19: A Policy Maker’s Handbook for More Resilient Systems, OECD Publishing, Paris, https://dx.doi.org/10.1787/0a530888-en.
[32] OECD (2020), TALIS 2018 Results (Volume II): Teachers and School Leaders as Valued Professionals, TALIS, OECD Publishing, Paris, https://dx.doi.org/10.1787/19cf08df-en.
[19] OECD (2019), Education Policy Outlook 2019: Working Together to Help Students Achieve their Potential, OECD Publishing, Paris, https://dx.doi.org/10.1787/2b8ad56e-en.
[30] OECD (2019), TALIS 2018 Results (Volume I): Teachers and School Leaders as Lifelong Learners, TALIS, OECD Publishing, Paris, https://dx.doi.org/10.1787/1d0bc92a-en.
[2] OECD (2019), Working and Learning Together: Rethinking Human Resource Policies for Schools, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/b7aaf050-en.
[46] OECD (2018), Education Policy in Japan: Building Bridges towards 2030, Reviews of National Policies for Education, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264302402-en.
[29] OECD (2017), “Flexible working arrangements”, in The Pursuit of Gender Equality: An Uphill Battle, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264281318-21-en.
[23] OECD (2014), TALIS 2013 Results: An International Perspective on Teaching and Learning, TALIS, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264196261-en.
[27] Paccagnella, M. (2016), “Age, Ageing and Skills: Results from the Survey of Adult Skills”, OECD Education Working Papers, No. 132, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jm0q1n38lvc-en.
[35] Paniagua, A. and D. Istance (2018), Teachers as Designers of Learning Environments: The Importance of Innovative Pedagogies, Educational Research and Innovation, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264085374-en.
[43] Renzulli, L., H. Macpherson Parrott and I. Beattie (2011), “Racial mismatch and school type: Teacher satisfaction and retention in charter and traditional public schools”, Sociology of Education, Vol. 84/1, pp. 23-48, http://dx.doi.org/10.1177/0038040710392720.
[57] Ronfeldt, M., S. Loeb and J. Wyckoff (2013), “How teacher turnover harms student achievement”, American Educational Research Journal, Vol. 50/1, pp. 4-36, http://dx.doi.org/10.3102/0002831212463813.
[33] Santiago, P. et al. (2016), OECD Reviews of School Resources: Uruguay 2016, OECD Reviews of School Resources, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264265530-en.
[48] Schaufeli, W., M. Leiter and C. Maslach (2009), “Burnout: 35 years of research and practice”, Career Development International,, Vol. 14/3, pp. 204-220, http://dx.doi.org/10.1108/13620430910966406.
[1] Schleicher, A. (2018), Valuing our Teachers and Raising their Status: How Communities Can Help, International Summit on the Teaching Profession, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264292697-en.
[24] Sharp, C. et al. (2019), Part-time Teaching and Flexible Working in Secondary Schools, National Foundation for Educational Research (NFER), Slough, https://www.nfer.ac.uk/part-time-teaching-and-flexible-working-in-secondary-schools/.
[15] Skaalvik, E. and S. Skaalvik (2018), “Job demands and job resources as predictors of teacher motivation and well-being”, Social Psychology of Education, Vol. 21/5, pp. 1251-1275, http://dx.doi.org/10.1007/s11218-018-9464-8.
[13] Skaalvik, E. and S. Skaalvik (2016), “Teacher stress and teacher self-efficacy as predictors of engagement, emotional exhaustion, and motivation to leave the teaching profession”, Creative Education, Vol. 7/13, pp. 1785-1799, http://dx.doi.org/10.4236/ce.2016.713182.
[28] Stromquist, N. (2018), The Global Status of Teachers and the Teaching Profession, Education International, Brussels, http://ei-ie-al.org/sites/default/files/docs/2018_ei_research_statusofteachers_eng_final.pdf.
[36] UNESCO, UNICEF, World Bank (2020), What Have We Learnt? : Overview of Findings from a Survey of Ministries of Education on National Responses to COVID-19, UNESCO, UNICEF, World Bank, Paris, New York, Washington DC, https://openknowledge.worldbank.org/handle/10986/34700.
[54] Valli, L. and D. Buese (2007), “The changing roles of teachers in an era of high-stakes accountability”, American Educational Research Journal, Vol. 44/3, pp. 519-558, http://dx.doi.org/10.3102/0002831207306859.
[3] Viac, C. and P. Fraser (2020), “Teachers’ well-being: A framework for data collection and analysis”, OECD Education Working Papers, No. 213, OECD Publishing, Paris, https://dx.doi.org/10.1787/c36fc9d3-en.
[16] Weiss, E. (1999), “Perceived workplace conditions and first-year teachers’ morale, career choice commitment, and planned retention: A secondary analysis”, Teaching and Teacher Education, Vol. 15/8, pp. 861-879, https://doi.org/10.1016/S0742-051X(99)00040-2.
Notes
← 1. Note that TALIS sampling does not include “substitute” teachers. Thus, teachers working under “fixed‑term” contracts should not be confused for “substitute” teachers. For more information on the TALIS sampling, see Annex A.
← 2. The scales of professional collaboration in lessons were drawn from teachers’ responses on how often (i.e. “Never”, “Once a year or less”, “2‑4 times a year”, “5‑10 times a year” , “1‑3 times a month”, “Once a week or more”) they do the following: 1) Teach jointly as a team in the same class, 2) Provide feedback to other teachers about their practices, 3) Engage in joint activities across different classes and age groups (e.g. projects) and 4) Participate in collaborative professional learning.
← 3. A national survey in Brazil also reveals that around one‑third of teachers have an additional job to complement their income. This is particularly common among those working shorter hours or with a temporary contract. This can not only hinder attempts to build a cohesive school community and positive stakeholder relationships, but also has important implications for teachers’ professional development and students’ performance. For more information, see the most recent OECD Education Policy review Education in Brazil: An International Perspective (OECD, 2021[31]).
← 4. Note that this indicator refers to the total hours teachers reported spending teaching in the last week in the surveyed schools. The actual teaching and learning time might be even lower as, in a typical lesson, time is also spent on classroom management and administrative work, which leaves less time for actual teaching and learning (see Chapter 3).
← 5. TALIS does not survey substitute teachers. This group may be particularly inclined to leave the profession as they are unable to secure ongoing employment. This means that TALIS data very likely underestimates potential attrition rates. For more information on the sampling of teachers, see Annex A.