Challenges facing the Estonian school system: the achievement gap between language-stream schools and school choice by immigrants

Introduction

Education is considered one of the leading factors governing successful integration in multicultural societies, although there are mixed opinions concerning what aspects of the education system are decisive. Due to historical legacies the institutional framework of basic education in Estonia is relatively unique. Schools with Estonian or Russian as their respective languages of instruction coexist, creating a parallel system (see Nagel 158). Regardless of the fact that a number of decisions have been taken to bring schools with the different languages of instruction closer together (the transition to partial Estonian-language instruction was established based on the Basic School and Upper Secondary Schools Act; the goal of the development plan Smart and Active People is to develop and implement an investment schedule specifically for the organisation of study and to support the inclusive education of students and resource-intensive support measures – including a language of instruction different from their mother tongue), and regardless of organisational uniformity, the differences between schools offering Estonian-language instruction and those offering Russian-language instruction (hereinafter Estonian and Russian schools) are significant when it comes to the academic success of the children.

Estonia’s current school system is also faced with challenges arising from immigration and cultural diversity. Children of new immigrants are mostly admitted to regular Estonian or Russian school. In Estonian schools, they are assigned special- needs student status. Some students opt out from the regular school system by applying to international schools where classes are mainly in English. The latter adds new dimensions to the current bilingual school system.

The chapter focuses on the role of schools that operate in different languages and, first and foremost, on the effects that these schools have on the academic achievement of students. First, we present an overview of the demographic trends affecting the allocation of students among schools with different languages of instruction. In addition, the development of international schools in Estonia is covered. Second, we show the effects of family background characteristics and education (school) management factors on the achievement gap between Estonian and Russian schools. Educational achievement is operationalised by using PISA mathematics test scores.

Students of schools with different languages of instruction within the Estonian education system

A separate school system with either Estonian or Russian as the language of instruction evolved in Estonia during the Soviet period. This reflected the diversified population structure related to post-World War II immigration. Following the restoration of independence, mass immigration from former territories of the Soviet Union ended; however, the language-based segregation of the school system remained. The number of students has declined significantly during the last twenty years in all basic education schools in Estonia. The decline has been especially prominent in Russian-language schools, where the current number of students comprises only one-third of the number in 1995 (see Figure 3.5.1). In Estonian-language schools, the decline remains in the range of 20%.

Unequal rates of decline in the number of students have also had an impact on the share of students studying in different languages of instruction. Twenty years ago, the share of students studying in Estonian was 70%, but has since risen to nearly 80%. The proportion of students studying in Russian declined most sharply within the first decade after 1995, then stabilised for a while close to 20%, and thereafter, in parallel with the upper-secondary school students being directed to Estonian-language immersion, fell to 15%. At the basic-school level, the relative importance of the Russian-language students has remained unaltered over the past decade. In addition, Estonia’s educational landscape involves students who are studying in languages other than Estonian or Russian (primarily English, and to a certain extent also Finnish). The number of those students has increased over the last decade from 126 in 2006 to 670 in 2016, comprising a relatively small (0.4%), but gradually growing, share of all students.

Figure 3.5.1. The number and share of students studying in Estonia’s basic education schools in different languages of instruction 1995–2016

Source: Statistics database 2016; EHIS 2016.

The distributions of students based on their mother tongue or based on the language of instruction presented in Figure 3.5.1 do not coincide. The main discrepancy concerns young people whose mother tongue is Russian. Their share has actually remained close to 25% for the past ten years, not declined. This implies that along with the demographic factors the increase in the share of students studying in Estonian has been shaped by the school choice trends among mother-tongue groups. Figure 3.5.2 shows that Estonian-Russian bilingual youth (there were approximately 1,500 such students in 2016) mostly select schools where the language of instruction is Estonian. In the case of students whose mother tongue was neither Estonian nor Russian (their number is growing slightly each year, but has yet to reach 1,000), approximately half choose Estonian-language schools, with the remainder being divided between schools with Russian or another language of instruction. As expected, only a minority of students speaking Russian as their mother tongue study in Estonian. Although the share of these Russian students has slightly increased, this change relates to school choice at the upper-secondary level, whereas among younger age groups the share studying in Estonian remains stable at 8-9%. Reflected in education statistics, the change in behaviour among students whose mother tongue is Russian coincides with the period of education reform and is indicative, above all, of the changes in the school system.

Figure 3.5.2. The share of students studying in Estonian and in Estonian-language immersion according to mother-tongue groups 2006-2016

Source: EHIS 2016.

A separate group within the Estonian educational system is new immigrants, i.e. those students who have stayed or resided in Estonia for less than three years, and who are therefore addressed as students with special educational needs. In 2016, 378 such students were registered in Estonian schools, which is double the number from five years ago. However, treating new immigrants as a single group is rather provisional (Kasemets et al. 2013). Firstly, this group includes students with different migration backgrounds – those who are staying in Estonia for only a brief period of time, as well as those who intend to remain connected to Estonia’s education system after a period of three years, when they will no longer be officially classified as new immigrants. Secondly, the designation of new immigrant status to a student in the educational database may depend on the language of instruction at the school – only Estonian schools can apply for support available for students with special educational needs due to their migrant background. The reasoning behind this limitation is the intention to encourage the integration of the children of immigrants into Estonian society. Consequently, over half of the students registered as new immigrants are studying at Estonian or Estonian-language immersion schools.

Thus, new immigrants are a diverse group that varies over time, and for whom the lay definitions may considerably differ from the definitions applied in educational statistics. In addition, the group does not encompass all students with a migrant background. Still, broadly speaking, over 40% of these students usually speak a mother tongue other than Russian or Estonian, whereas recently the share of new immigrants speaking Russian as their mother tongue has also increased to a comparable level. One fifth of the students with special educational needs, due to their migrant background, speak Estonian as their mother tongue. Although this text does not cover specific motivations behind school choices among new immigrants, nor their social or educational coping within the Estonian educational system, more information on these issues can be found elsewhere (see Kasemets et al. 2013).

What are the latest trends in a school system that has long been bilingual, both from the perspective of Estonian and Russian schools, but also in terms of designing a system that involves other languages of instruction? Recent education policy trends promote the school model where learning takes place in combined groups (HTM 2016), i.e. instead of the currently separate bilingual schools, multilingual and cultural schools are aimed for. There are a number of schools that combine full-time Estonian teaching with part-time instruction in Russian (16 schools), and there are schools where learning takes place in Estonian, but communication between students is in Russian or bilingual, since the vast majority of students are Russians (12). There are also schools where the share of students coming from a migrant background is greater than 10% (6) – including Tallinn Lilleküla Upper Secondary School and Kiltsi Basic School, which are located within the area of the housing centre for asylum seekers (the Ministry of Education’s ‘Learning Together Schools’ project). Over the last decade, international schools following an international curriculum or European Commission directives, and where learning takes place entirely or partially in English, have been established in Estonia (5), (for the International Baccalaureate – IB and European Baccalaureate – EB).

There are no significant differences in the institutional framework and objectives at Estonian or Russian schools, whereas the role and character of international schools in Estonia is more distinct. Based on interviews and information from the schools’ homepages, it can be concluded that each school has its own target group, be it the children of the employees of the European Union agencies and structural units, or the rotating officials within Estonia’s Ministry of Foreign Affairs. There are also important differences in the financing of schools. For example, IB-instruction at an English-language school is free of charge, but has a very selective student body. Other schools are tuition-fee based, and differ considerably in their fees. Therefore, Estonia’s regular school system is the main host when it comes to educating students with a migrant background.

The achievement gap between students at Estonian- and Russian-language schools

The purpose of this subchapter is to explain the differences in achievement between Estonian- and Russian-language schools. There are different ways to operationalise academic achievement – for example, output measure, i.e. PISA tests scores, are often used. Such instrumentalisation has been criticised, although ‘PISA pride’ is a prominent part of media reporting, as well as of the rhetoric of government agencies. In addition, PISA scores are also indicators for several European Union target objectives. Estonia has already participated in five PISA studies; in this case, PISA 2012 results were used. For PISA 2012, in Estonia 4,800 students from 37 Russian and 169 Estonian schools participated, i.e. slightly more than 20% of the students participating in the test attended schools where Russian is the language of instruction.

In terms of student background characteristics as well as institutional framework, Estonian and Russian schools do not differ significantly (see Table 3.5.1). Russian schools are mostly urban; therefore, the average class and school sizes are somewhat larger than in the case of Estonian schools. There are fewer private schools among Russian schools. There is also a significant difference in the countries of origin of the students: more than a quarter of the students in Russian schools are the descendants of immigrants; however, in the case of Estonian schools their share is marginal.

The standard explanation for students with an immigrant background falling behind in academic success is based on the possible language barrier. However, given the bilingualism on which the Estonian school system is based, this argument is not supported. In Estonian schools, the language spoken at home and in school by students with a migrant background generally differs, although in the case of schools in which the language of instruction is Russian, immigrants primarily learn in their own mother tongue. When comparing the background of students and the institutional framework of schools there are not many differences, but the primary thrust of this chapter is the difference in the academic success of Estonian and Russian schools. The academic achievement indicators measured by PISA scores vary in all measured categories, i.e. in science, reading and mathematics. We concentrate only on differences in PISA mathematics scores (see Figures 3.5.3, and 3.5.5).

Table 3.5.1. Profile of schools, based on PISA 2012 school questionnaires

Source: PISA 2010

In PISA 2012, Estonian school students performed comparatively better in mathematics, and similar tendencies in cross-national studies have been present from 2003 onward (HTM 2013). However, the achievement gap has narrowed over the years: PISA 2009 – 38, PISA 2012 – 32 and PISA 2015 – 29 points. Even so, 39 PISA points are the equivalent of a year of learning; in other words, the average difference in scores in mathematics between Estonian and Russian schools continues to be three-quarters of a year.

Looking at the difference in the average scores Estonian and Russian schools had in 2012 and 2015, we see that in Russian schools, there are significantly fewer outliers, i.e. students with very low and very high test results (highest and lowest points in the Figure). Even so, in the comparison between 2012 and 2015, in Russian schools there has been an increase in the number of the lowest achievers. In Estonian schools, the association between students’ achievement and the socio-economic status of the parents is larger. In other words, educational equity, which aims at lowering the impact of family background on students’ achievement, is greater in Russian schools.

It is important to investigate whether this gap in academic achievement in Estonian and Russian schools can be explained through differences in the quality of the schools, i.e. whether there is a difference in the share of high- and low performing Estonian and Russian schools. To do so, we are ‘placing’ all survey participants (4,800) into Estonian (169) and Russian (37) schools and comparing the average mathematics results schools-wise, grouped by Russian and Estonian. In schools the ethnic achievement gap also remains: the Estonian school average is 523 points in comparison with 495 points in Russian schools. In addition, it is apparent that Russian schools are skewed towards lower results and the majority of Russian schools are, in terms of the average mathematics result, significantly below the overall average. Estonian schools have a number of notable high-performers (school’s average result in the range of 600-700), however there are no Russian schools represented there. Such a significant ethnic achievement gap is also confirmed by PISA 2015 (Tire et al. 2016), which also reveals that the difference is especially large in Tallinn (over 50 points), where the achievement of Estonian schools is very high.

Previous studies have also focused on the achievement gap between Estonian and Russian schools, and have emphasised that the latter not only discourage Russian- school graduates from continuing with their studies but also create a glass ceiling in the labour market, and in general hamper integration (Lindemann 2013B). Tire et al. (2013) and Kitsing (2012) have shown that among Russian school students there is a dearth of top performers as well as more students with very low achievement scores compared to Estonian students. Lindemann (2013A) has noted that students speaking Russian as their mother tongue are, on average, more successful in Estonian schools than their peers in Russian schools.

How can this ethnic achievement gap be explained? The literature reveals that significant factors affecting academic success can broadly be divided into two groups: first, family background characteristics and second, institutional factors. Background characteristics, such as the education level of parents and their professions, have been confirmed to be the most important predictors of academic success (for example, Fuchs & Woessmann 2007; Hanushek & Woessmann 2010). The general standard of living and a home environment that is supportive of learning (for example, availability of books, access to computer-based sources) are also important, as well as the language spoken at home in relation to the language of education. Thus, for example, children whose parents are highly educated – part of the upper-middle social class, and possessing more books – are, on average, more successful academically. Although most of todays’ policy strategies aim to reduce the family background effect on educational outcome (and according to international comparisons Estonia has been successful here), there is nevertheless a significant positive association between family background and academic success.

The influence of the institutional framework includes several important factors, the most important of which are teachers and peers. For instance, the motivation and capability of their teachers, and their devotion to their profession, empowers students. Secondly, the attitude of peers (playmates and sports club companions) towards school and learning influences academic success.

In the case of the institutional features we focus on factors related to school management. During the last decade, studies concentrating on the effects of system-level education policy in general or school management in particular have increased in significance and scope. The greatest attention in this field has perhaps been garnered by the research led by Ludger Woessmann, according to which the important characteristics of the performance of the education system are primarily related to school choice, ability grouping and school autonomy in designing admission rules, as well as the principles governing private schools and their public financing (Woessmann et al. 2009). Significant associations have also been found between academic achievement and school accountability and parental involvement.

We use PISA 2012 data for estimating the effect of the above-mentioned factors on the achievement gap in mathematics scores between Estonian and Russian school students. First, we use regression analysis for estimating the average effects of individual and school characteristics (see Figure 3.5.5), after which Estonian and Russian schools are treated separately. Background characteristics include gender, immigration background, education indicators for the mother and father, and the number of books at home. Teacher and peer influence have only been assessed indirectly, based on the Sense of belonging index for students. Institutional school characteristics include perceived competitiveness of the school, pressure applied by parents on the school, ability grouping in math class, the selectivity of the school’s admission policy, and the school head’s assessment of accountability. In addition, we control for the school’s ownership, location and language of instruction.

Similarly to previous studies (for example, Põder et al. 2016) our association analysis showed that student gender has a positive effect on mathematics results (boys score higher), as well as the parents’ level of education, and, above all, the number of books at home are associated with higher results. There is a weak positive correlation with the school’s Sense of belonging index, i.e. mathematics results are better among children who have a stronger connection to the school. As expected, a migrant background has a negative effect (first or second generation). In addition to background characteristics, institutional factors such as the competitiveness of schools and the urban location of schools positively affect academic achievement.

The positive association between the intensity of a school’s drive to compete and academic achievement is explained by the school choice effect, i.e., if families are allowed to make choices, it should encourage the schools to strive for improvement in order to attract the best students, as well as inspire efforts by families to send children to the best-performing school. Competitiveness such as this may have a positive influence on education results, although especially at the basic school level its un-controlled presence is associated with educational inequality (Põder & Lauri 2015). The main problem in the case of an early selection is that this encourages families to pre-train their children, to ensure admission to the popular school, which in turn increases the family background effect on educational outcome. A favourable background, competitiveness and the positive association of location in the city with academic achievement is not surprising and has been shown in studies of international educational academic literature, although the important positive effect of Estonian schools on mathematics results is noteworthy and worrisome.

The current analysis showed which characteristics related to background or the institutional structure of the school generally predict better academic results within the overall Estonian school system and confirmed the achievement gaps resulting from the language of instruction of schools. We continue to assess whether these same characteristics are able to predict academic achievement equally well in Estonian and Russian schools. To do so a regression technique was used that allows one to distinguish between two types of effects: 1) To what extent the two groups differ, i.e. are there are any important differences between the background of children or the institutional characteristics of Estonian and Russian schools (explained gap)? 2) What is the effect of these characteristics on the two different groups, i.e. do some characteristics ‘work’ differently (unexplained gap) in predicting academic results in Estonian and Russian schools?

The analysis indicates that the differences in academic success between students in Russian- and Estonian-stream schools can be partially explained by the fact that children in Russian schools originate more frequently from immigrant families and they have fewer books in their homes. A higher level of education in the mothers of students in Russian schools tends to reduce the gap in mathematic scores between the schools. Therefore, group differences account for only a small share (-0.9 in Table 3.5.2) of the explanation for all differences in mathematics achievement. Similarly, Lindemann (2013A) has found certain differences in the professional positions of the parents of students in schools where the language of instruction is Estonian and schools where the language of instruction is Russian, and Säälik (2012) found variance in the teaching methods at schools and the learning skills of students, although these explain only a very small part of all differences in academic results.

Table 3.5.2. Factors characterising the achievement of Estonian and Russian schools

Notes: * - p<0.05, ** - p<0.01, *** - p<0.001, standard errors in parentheses, effects stated by the school

The ‘unexplained gap’, i.e. whether the characteristics ‘work’ differently in producing academic achievement in different language schools, dominates explanations of this gap. First, it illustrates that a migration background and school location have different effects on schools: the migrant background as well as the location of the school in the city are both favourable circumstances in the case of Russian school students. As such, a migration background or urban location of a school does not explain the lower results of Russian schools. On the contrary, referring to the circumstances mentioned above, in Russian schools the majority of immigrants study in their native tongue and the majority of schools with Russian as the language of instruction are located in the city.

The lower results of Russian schools can be explained, to a certain extent, by two indicators: the education level of the father; and the large negative impact of the involvement of parents on academic results. The lower effect of the level of education of the father indicates that the role and contribution of the father in the shaping of the academic achievement of children in Estonian schools is bigger compared to children in Russian schools. The parental pressure on the school has a greater positive effect in the case of Estonian schools. This may be an indication of so-called elite and upper-class schools, where ambitiousness as well as the inclusion of children’s parents is much greater, and where the language of instruction is mostly Estonian.

A second explanation of divergent impact of parental pressure could also be related to a broader polemic surrounding schools where Russian is the language of instruction, subjecting parents, teachers as well as school directors to stress and, in turn, hindering the academic results of the children. It was also noted during the interview with an expert from the Ministry that in the case of schools where Russian is the language of instruction the attitudes of teachers, directors and parents are sometimes a cause for alarm, and also certainly have an impact on the everyday work of the schools. Even Valk (2010) has emphasised that the role of parents in ensuring academic success may be multi-directional in a multicultural society and classroom. Cultural socialisation (feeling pride in one’s own culture, preserving traditions) tends to be associated with a high feeling of self-esteem and thereby also academic success, while on the other hand, behaviour that is defensive and avoids prejudices produces rather the opposite effect on self-esteem, identity as well as behaviour, and thereby also on academic success. The history of events surrounding schools where Russian is the language of instruction may have resulted in the creation of a similar defensive attitude.

Summary

The chapter has focused on language as a segregating mode within the Estonian education system. Firstly, we revealed the most recent trends. Demographic developments together with the school choices of Russian families reduced both the number of Russian schools in Estonia as well as the share of students attending Russian schools. At the same time, schools with other (above all English) languages of instruction have emerged, although the share of students attending these schools is relatively marginal. While Tallinn’s international schools have mainly been established for children of foreign service, European Commission, and other diplomatic corps employees, the regular Estonian schools have a considerable role in receiving new immigrants and other students with a migrant background.

Secondly, the achievement gap between Estonian and Russian school students was disentangled. Evidence shows that the gap between Russian and Estonian schools is approximately 30 PISA points, which is close to the average amount of knowledge acquired over the course of a year. The academic achievement of students in Estonian schools demonstrates greater variance; however, there are more top-performing students. However, Russian and Estonian schools do not differ significantly in terms of the students’ backgrounds or institutional indicators, and therefore the achievement gap cannot be explained by them. There are, however, differences in how the higher education of the father affects the academic success of the child. The positive impact of fathers on higher education is significant in Estonian school, although this fact does not necessarily mean that the behaviour of fathers in Russian-speaking families is somehow different, but instead may indicate some societal structural features, for example, the poorer position in society of Russian men with a higher education, which hinders the usual positive influence of a highly-educated father on children attending a school in which Russian is the language of instruction.

The second important factor increasing the differences between schools where Estonian is the language of instruction and schools where Russian is the language of instruction is the pressure placed by parents on the school. Ordinarily the demands placed upon a school by parents are considered to be an empowering factor, although in Estonia this leads to a significant decrease in academic success in schools where Russian is the language of instruction.

Regardless of some of the significant indicators mentioned above, it is not possible to fully explain the majority of the differences between the academic success of Estonian-language and Russian-language schools. The unexplained difference in results is frequently interpreted in the literature as unequal treatment. Most of this type of literature focuses on analyses of wage differences between women and men and the unexplained part in such studies refers to the decision of employers to pay women less for equivalent work. In this work here, the unexplained difference in academic success is no one’s decision; instead, it ties in to broader integration-related challenges. The interpretation of the analytical results is complicated by the fact that there is a lack of knowledge regarding what the academic success of Russian-language schools would be if the policy decisions related to schools with a special language of instruction had been different, for example, as in Latvia, where abrupt changes were made significantly earlier (2004). Even though comparing Latvia and Estonia remains solely at the discussion level here, the academic success of the Russian population in Latvia is equal to that of the Latvian population, based on PISA data. It should be noted that the general PISA performance by Latvians is only average when compared to Estonia’s top-level performance, and in absolute indicators our Russian-speaking population is better in mathematics achievement.

Schools where Russian is the language of instruction are somewhat of a fading phenomenon and in that context it could be said that it is a ‘problem’ that will be outgrown through evolution. Even so, from the position of educational fairness, it is a significant challenge. Just for the sake of argument, a school where Russian is the language of instruction can be thought of as a valuable experience for Estonia, especially when it comes to the reception of newcomers – particularly from the position of language immersion and a multicultural school and classroom. Even though international schools have also, to a certain degree, established themselves in Estonia’s education system, they are for the most part marginal when it comes to the integration of immigrants and instead could contribute most of all by disseminating expertise on changing educational concepts. These are also important keywords, since alongside the current ethnocentric creation myth, the importance of educational concepts emphasising general skills, flexibility and diversity in education are growing increasingly important. In other words, that which needs to be supported is how to be together though remaining different.

The development of education policy frequently includes a certain degree of scheming regarding efficiency and fairness. International education studies have taken the position over the past decade that these goals complement one another. Even though education systems that consider educational fairness to be a value actually show better average results, the education policy decisions are, however, frequently complicated compromises. In this way it is presumed that clear policy objectives will be set within the context of openness and diversity, and will be followed during the shaping of education policy. If these are based on the shaping of equal individual possibilities, regardless of the language spoken by the children or their domestic socio-economic or cultural capital, then international schools as well as schools where Russian is the language of instruction are an institutional problem and require equalising intervention. If, however, a multicultural society is handled as a whole, the diverse educational concepts, solutions based on various forms of financing and ownership may prove to be an opportunity, not a problem. The empowering strength of a school that learns together would also require, along with positive special treatment in the case of exceptions, a certain waiving of efficiency-based outcome indicators in a separative-selective system. Are we ready, in the name of closing the gap in education results, to give up our triumphant spot in the PISA standings?

References

Fuchs, T. and Woessmann, L. (2007). What Accounts for International Differences in Student Performance? A Re-Examination Using PISA Data. Empirical Economics, 32 (2): 433–462.

Hanushek, E. A., Woessmann, L. 2010. The economics of international differences in educational achievement. NBER Working Paper 15949–04. Cambridge MA: NBER working paper series.

Haridus- ja teadusministeerium (2016). Koosõppiv Kool – Kontseptsioon kvaliteedinõuete kohta koolidele, kus õpib koos palju erineva keele- ja kultuuritaustaga õpilasi, HTM.

Haridus- ja teadusministeerium (2013). Eesti ja vene õppekeelega koolide õpilaste tulemuste võrdlus tuginedes PISA uuringutele: https://www.hm.ee/sites/default/files/lisa_1pisa_2006_2012-eesti_ja_vene_oppekeelega_opilaste_tulemuste_vord__0.pdf (30.11.2016).

Kasemets, L., Asser, H., Hannust, T., Rahnu, L. (2013). Uusimmigrantõpilaste akadeemiline ja sotsiaalne toimetulek Eesti üldhariduskoolis. MindPark OÜ, Integratsiooni ja Migratsiooni Sihtasutus Meie Inimesed.

Kitsing, M. (2012). Kõrgemate ja madalamate tulemustega koolide võrdlus. Kogumikus Eesti PISA 2009 kontekstis: tugevused ja probleemid. Programmi Eduko uuringutoetuse kasutamise lepingu aruanne. Eduko, Tartu, lk 46–61.

Lindemann, K. (2013A). The School Performance of the Russian-speaking Minority in Linguistically Divided Educational Systems: A Comparison of Estonia and Latvia. In: Windzio, M. (Ed.). Integration and Inequality in Educational Institutions. Dordrecht: Springer, pp. 45–69.

Lindemann, K. (2013B). Structural Integration of Young Russian-Speakers in Post-Soviet Contexts: Educational Attainment and Transition to the Labour Market. Tallinna Ülikool Sotsiaalteaduste dissertatsioonid.

Nagel, V. (2006). Hariduspoliitika ja üldhariduskorraldus Eestis aastatel 1940–1991. Doktoritöö. TLÜ Kirjastus.

Põder, K., Lauri, T., Veski, A. (2016). Does school admission by zoning affect Educational Inequality? A Study of Family Background Effect in Estonia, Finland, and Sweden. Scandinavian Journal of Educational Research, 1−21, 00313831.2016.1173094.

Põder, K., Lauri, T. (2015). Haridusturgude peidus pool – koolidevahelise ebavõrdsuse süvenemine koolivaliku tõttu. Eesti Inimarengu Aruanne 2014/15. Lõksudest välja? (78–85). Eesti Koostöö Kogu.

Säälik, Ü. (2012). Kooli ja klassi kliima, õpetaja-õpilase suhted, õpi- ja õpetamismeetodid ning õpetajate käitumine koolijuhi ja õpilaste hinnangute põhjal. Eduko aruanne: Eesti PISA 2009 kontekstis: tugevused ja probleemid. Programmi Eduko uuringutoetuse kasutamise lepingu aruanne.

Tire, G., Lepmann, T., Jukk, H., Puksand, H., Henno, I., Lindemann, K., Kitsing, M., Täht, K., Lorenz, B. (2013). PISA 2012 Eesti tulemused: Eesti 15-aastaste õpilaste teadmised ja oskused matemaatikas, funktsionaalses lugemises ja loodusteadustes. SA Innove, Haridus- ja Teadusministeerium.

Tire, G., Henno, I., Soobard, R., Puksand, H., Lepmann, T., Jukk, H., Lindemann, K., et al. (2016). PISA 2015 Eesti tulemused: Eesti 15-aastaste õpilaste teadmised ja oskused loodusteadustes, funktsionaalses lugemises ja matemaatikas. SA Innove, Haridus- ja Teadusministeerium.

Valk, A. (2010). Mitmekultuuriline klass. Raamatus „Õppimine ja õpetamine esimeses ja teises kooliastmes“ (toim. Kikas, E.). Eduko. Lk 138–150.

Woessmann, L., Luedemann, E., Schuetz, G., West, M. R. (2009). School Accountability, Autonomy and Choice Around the World. Ifo Economic Policy Series.