Immigrant Students' Achievements in Croatia, Serbia and Slovenia in Context
Iztok Šori*1, Nika Šušterič2 and Slavko Gaber3
Achievement gaps between immigrant and native students indicate failure to assure educational equity in the majority of countries assessed by the Programme for International Student Assessment in 2009 (PISA, 2009). The present article explains disparate achievement results in Europe, first testing the hypothesis of old and new democracies. In further contextualisation of the achievement results, the analysis seeks explanations beyond the common education system explanatory model. Specifically, the article considers results from Croatia, Serbia and Slovenia, highlighting the significance of language distance between native and immigrant students as well as migration regimes as important factors in creating or reducing the achievement gap between native and immigrant students. Evidence has been found that immigrant students score worse in countries with guest labour immigration regimes than in the countries with large scale forced immigration of people of the same ethnic (linguistic) origin.
Keywords: Achievement, Equity, Immigrant students, Migration patterns, PISA 2009
Introduction
Although migration and the subsequent education of immigrant children are an old and widespread phenomenon, "it is only in recent years that
1	*Corresponding author. Department of Sociology of the Faculty of Arts, University of Ljubljana iztok.sori@gmail.com
2	Special and Rehabilitation Education at the Faculty of Education, University of Ljubljana (student)
susterlc.nika@gmail.com
3	Faculty of Education, University of Ljubljana slavko.gaber@guest.arnes.si
international database[s] have become available with which to conduct quantitative studies on the situation of immigrant students" (Organisation for Economic Co-operation and Development (OECD), 2006, p. 30).
PISA assessments, the most comprehensive data set on the subject, regularly demonstrate that immigrant students4 have comparatively lower education achievement than native students in most of the countries assessed. In PISA 2003, which focused on mathematics, native students in OECD countries outperformed first generation immigrant students by 48 points (OECD, 2006, p. 183). In PISA 2006, first generation immigrant students lag, on average, 58 points behind native students in science (OECD, 2007, p. 175). In the most recent assessment of reading competences, PISA 2009 found native students in OECD countries outperformed first generation immigrant students by 50 points (OECD, 2010c, p. 170).
Experts from PISA claim that 40 points represent one school year (OECD, 2010c, p. 49). If that is the case, the differences presented above raise serious concerns regarding the future prospects of a considerable part of the population. It is expected that immigrant children will comprise up to one third of the European Union (EU) school population by 2020 (Huttova, McDonald & Harper, 2008, p. 2), meaning that not only students themselves will have to face the challenge, but European societies as well.
Although empirical evidence suggests that students who speak a language other than the language of instruction at home score lower than students whose households speak the language of instruction (OECD, 2006, p. 77; OECD, 2010c, pp. 177-181), language proficiency is neither the only nor the most important factor contributing to lower achievement of immigrant students. Data obtained by PISA (OECD, 2010c, pp. 177-179) demonstrate a high correlation between socioeconomic background and immigrant students' test results. Research in the United Kingdom has confirmed that socioeconomic status is the key factor when explaining achievement gaps between immigrant and native students (Rothon, 2004). In France, there is evidence that immigrant children have lower social mobility than students without immigrant status (Simon, 2003, p. 1093). Even after accounting for socioeconomic status, however,
4 We use the PISA definition of immigrant background: (1) native students (born in the country of assessment, or with at least one parent born in that country; students born abroad with at least one parent born in the country of assessment are also classified as 'native' students), (2) second generation students (born in the country of assessment but their parents were born in another country), and (3) first generation students (born outside the country of assessment and their parents also born in another country) (OECD, 2010b, p. 170).
it is obvious that socioeconomic standing cannot entirely explain the difference. Most authors agree that several factors are involved, often depending on the national context (see, e.g., DfES, 2005; Gillborn & Mirza, 2000; Kristen & Granato, 2007).
In order to grasp the complexity of variations in achievement levels among immigrant and native students, it is important to consider two points. Firstly, there are considerable achievement gaps between different immigrant ethnic groups within one country (e.g., Rothon, 2007), but also, as Crul and Schneider (2009) demonstrate, there are differences in the performance of immigrants of the same ethnic origin in different countries. Secondly, lower immigrant performance is not without exception. In traditional settlement countries, such as Australia, Canada and New Zealand, there are virtually no performance differences between immigrant students and their native peers. This is often linked with the immigration regimes of these countries, which are grounded on the selection of the majority of immigrants "on the basis of their ability to make an economic contribution, which creates a highly educated immigrant class" (OECD, 2011, p. 70).
In the context described above, we narrow the focus of the present paper first to ten European countries. The criteria used to select these countries included their status as an old or new European democracy, and their diverse geographic locations, educational traditions, as well as the achievement levels of native and immigrant students, the background of the immigrant population and immigration patterns. The paper endeavours to establish whether there are any consistent patterns related to the results achieved.
We then narrow our scope to three specific states of the former Yugoslavia: Croatia, Serbia and Slovenia. These three states share a common history and have similar political arrangements and economies. They are particularly suited to this inquiry because they are linguistically close and have had comparable education systems for 70 years. Slovenia, the leader in student achievement standards in PISA 2009, is by far the worst in assuring equity in education for its immigrant population - it is almost "European". In Croatia, the achievement gap between immigrant and native students is relatively small, and in Serbia immigrant students outperform their native peers. These differences are intriguing in their own right, but what triggered this analysis of achievement gaps between immigrant and native students in these countries are the distinctive immigration patterns. For the past few decades, Slovenia has been primarily a labour immigration country, whereas in the 1990s Croatia and Serbia experienced the largest process of forced migration in Europe since World War II.
The present paper endeavours to:
1.	Present the actual results from the PISA 2009 assessment in Croatia, Serbia and Slovenia in a more detailed manner. In doing so, we will use - for comparative reasons and to contextualise the results - average OECD results as well as results for five selected former socialist countries (Croatia, Estonia, Hungary, Serbia and Slovenia) and five selected old democracies (Austria, Belgium, Finland, Germany and Norway) in Europe. In this section, we will try to test the comparability of Croatia, Serbia and Slovenia in the European context.
2.	Conceptualise and present differences in immigration regimes and their effects on student immigrant populations in Croatia, Serbia and Slovenia.
Croatia, Serbia and Slovenia in the European Context
As mentioned above, the present paper narrows the comparison of immigrant and native students in Croatia, Serbia and Slovenia only to countries in Europe. The comparison focuses on five post-socialist countries (the three former Yugoslav Republics, Estonia and Hungary) and five old democracies (Austria, Belgium, Germany, Finland and Norway).
Table 1: Old and new European democracies and reading achievement differences between native and immigrant students.
Country	Native Students' Score Points	Immigrant Students' Score Points	Difference between Native and Immigrant Students	Share of Immigrant Students	GDP (PPP) per Capita
Austria	482	414	68	15.2	39,647
Belgium	519	451	68	14.8	36,322
Finland	538	468	70	2.6	36,843
Germany	511	455	56	17.6	35,551
Norway	508	456	52	6.8	55,198
Croatia	479	461	18	10.7	16,474
Estonia	505	470	35	8.0	20,753
Hungary	495	507	-12	2.1	19,829
Serbia	442	456	-14	9.5	10,911
Slovenia	488	441	47	7.8	28,893
OECD Average	499	457	42	10.3	33,225
Selected Countries Average	496.7	457.9	38.8	9.5	30,042
Old Democracies Average	511.6	448.8	62.8	11.4	40,712
New Democracies Average	481.8	467	14.8	7.6	19,372
Note: Achievement data were obtained and calculated from the PISA 2009 database. GDP (PPP)
(Gross domestic product based on purchasing-power-parity per capita) is calculated in International Dollars and obtained from the World Economic Outlook Database (International Monetary Fund, October 2009).
PISA 2009 reading proficiency results show that in OECD countries, on average, immigrant students perform 42 points lower than native students. Compared to both the old and new European democracies, the difference is slightly lower (38.7 points). There is no significant difference between European countries and the rest of the OECD, but there is a significant difference between the old and new European democracies. Achievement differences in the old democracies reach 62.8 points, which is significantly higher than the OECD average. On the other hand, differences in former socialist countries - new democracies - are significantly lower (14.8 points) than the OECD average. This pattern is also apparent when widening the scope to all members of the EU. Considering these results, it could be tempting to conclude that the socialist inclination towards equality has clearly impacted the present education systems in new democracies, but there are obvious and significant differences within the former socialist camp. Croatia has an 18 point difference, Hungary has a negative 12 point difference and Serbia has a negative 14 point difference, whereas Estonia has a 35 point difference and Slovenia a 47 point difference, both close to the average OECD difference. Slovenia and Estonia thus disprove the theory based simply on old and new democracies and inequality and equality concepts. It is not only the socialist or democratic past that distinguishes both groups of countries. From the 2009 PISA data, one might also be inclined to interpret the standard of living and public expenditure on education as the two factors explaining the majority of the overall educational achievement. However, the logic that "better standards bring better results," which applies to student results in some countries, is not valid with regard to immigrant students.
Table 2: Share of students at proficiency levels 3 - 6 in reading achievement: native and first generation immigrant students.
	Native Students		Immigrant Students	
Country	Share	Rank	Share	Rank
Austria	53	8	17.6	9
Belgium	67.2	2	36.8	5
Finland	76.1	1	40.5	2
Germany	65.4	3	37.4	3
Norway	63.3	5	37	4
Croatia	51.7	9	35.9	6
Estonia	62.8	4	C	C
Hungary	59	5	56.8	1
Serbia	33.6	10	34.4	7
Slovenia	55.3	7	23	8
OECD Average	59.6		38.7	
Selected Countries Average	58.7		31.9	
Old Democracies Average	64.9		33.9	
New Democracies Average	52.5		37.5	
Note: The abbreviation C denotes missing data due to too few observations to provide reliable
estimates or no observation at all (PISA, 2010c, p. 23). Data were obtained and calculated from the PISA 2009 database.
Table 2 shows significant differences in the number of students who reached proficiency level 3,5 which is the minimum competency required for future professional success.6 Achievement gaps are significant both for the native population of students and for the first generation of immigrants. In a number of countries, significant differences exist between the native and immigrant population within the same country.
There are significant diflfferences, for example, in the share of native students at level 3 or higher between Serbia (33.6%) and Finland (76.1%). Yet there is an obvious difffference between the share of native Finns reaching level three and the share of immigrants (40.5%) reaching the same level. The proportion of immigrants in Serbia
5	While at proficiency level 2 on the reading scale students only begin to demonstrate reading skills, students at proficiency level 3 are capable of reading tasks of moderate complexity, such as locating multiple pieces of information, making links between different parts of the text and relating the text to familiar everyday knowledge (OECD, 2010a, p. 51) and thus are more likely to experience success in their future life.
6	This is most likely true in terms of OECD standards. Additionally, the matter is complex considering differences in cultural capital validation in further education and subsequent employment opportunities and salaries in different countries.
(34.4%) who have reached level 3 competency is close to the proportion in Finland and far higher than that in Slovenia (23%) and Austria (17.6%). Hungary's immigrant students perform exceptionally well: 56.8% of first generation students have reached at least proficiency level 3. In Hungary's second generation, the share is 73.7% - close to the Finnish native percentage of 76.1, and far higher than the OECD average for second generation students, which is 45.3%. Generally, however, a pattern does seem to emerge: in countries where native students perform exceptionally well, immigrant students lag behind and do not gain the same benefits from the educational system.
Table 3: Reading achievement of native students and students with an immigrant background (first and second generation).
Country	Native Students		Second Generation Students		First Generation Students		Score Points Difference between First and Second Generation Students
	Score Points	% of Students	Score Points	% of Students	Score Points	% of Students	
Austria	482	84.8	428	10.5	385	4.8	43
Belgium	519	85.2	453	7.8	449	6.9	4
Finland	538	97.4	493	1.1	449	1.4	44
Germany	511	82.4	457	11.7	450	5.9	7
Norway	508	93.2	463	3.6	447	3.2	16
Croatia	479	89.3	465	7.2	452	3.5	13
Estonia	505	92.0	470	7.4	470	0.6	0
Hungary	495	97.9	527	0.9	493	1.2	34
Serbia	442	90.5	464	5.2	446	4.3	18
Slovenia	488	92.2	447	6.4	414	1.4	33
OECD Average	499	89.7	468	5.8	449	4.5	19
Selected Countries Average	496.7	90.5	466.7	6.18	445.5	3.3	21.2
Old Democracies Average	511.6	88.6	458.8	6.9	436	4.4	22.8
New Democracies Average	481.8	92.4	474.6	5.4	455	2.2	19.6
Note: Data were obtained and calculated from the PISA 2009 database.
In OECD countries, second generation students tend to outperform first generation students by 19 points in the reading test. In the selected countries, the difference amounts to an average of 21.2 points. The difference in reading scores is highest in Finland and Austria, where it exceeds 40 points. At 33 points, Slovenia also has a significant discrepancy. In Estonia, there is no difference in achievement between the two generations. Small gaps were ascertained in Belgium and Germany, where second generation students do not outperform first generation students by more than 4 and 7 points respectively. Large achievement gaps highlight the disadvantages of first generation students, and possibly the
38
immigrant students achievements in Croatia, Serbia and slovenia in context
different backgrounds across immigrant cohorts; they could also signal positive educational and social mobility across generations (OECD, 2010c, p. 72).
Table 4: Reading achievement by immigrant status, before and after accounting for economic, social and cultural status (ESCS) in selected countries.
Country	Difference between Native and Immigrant Students before Accounting for ESCS	Difference between Native and Immigrant Students after Accounting for ESCS	Score Point Difference
Austria	67	37	30
Belgium	68	41	27
Finland	70	60	10
Germany	56	27	29
Norway	52	33	19
Croatia	18	10	8
Estonia	35	34	1
Hungary	-12	-11	-1
Serbia	-14	-18	4
Slovenia	47	24	23
OECD Average	43	27	16
Selected Countries Average	38.7	23.7	15
Old Democracies Average	62.6	39.6	23
New Democracies Average	14.8	7.8	7
Note: Data were obtained and calculated from the PISA 2009 database.
When examining and presenting performance differences between immigrant and native student groups in an international context, it is essential to consider the different background characteristics of immigrant populations across countries. Family cultural capital, socioeconomic status and other background characteristics reflect situations at the time of immigration (Bourdieu, 1991, pp. 51-52) and also determine the extent to which immigrants are able and willing to adapt to a new environment (Stanat & Christensen, 2006, p. 59). In selected European countries, students with an immigrant background are in general socioeconomically disadvantaged,7 which explains part of the performance gap between these students and native students. Across OECD countries, immigrant students tend to have a socioeconomic background that is on average 0.4 of a standard deviation lower than their native peers (OECD, 2010c, p. 71). Accounting for ESCS explains 16 score points difference
7 Socioeconomic background is measured by the PISA index of economic, social and cultural status (ESCS), combining information on parents' education and occupations and home possessions. (OECD, 2010c, p. 29)
between native and immigrant students in OECD countries (15 points in selected European countries). The link is particularly strong in old democracies such as Austria (30 points difference), Germany (29 points) and Belgium (27 points), but the difference is also high in the new democracy Slovenia (23 points), whereas in Hungary, Estonia and Serbia ESCS does not negatively affect immigrant students' performance.
The data point with a degree of certainty to the type of immigration of the respective countries, as well as to the background characteristics of immigrants. Nevertheless, large differences might also allude to the discrimination and low inclusion of immigrants in, for example, the labour market.
Differences in achievement accounted for by students' socioeconomic status illustrate the significant influence of immigrant students' backgrounds. As a result, it is necessary to develop mechanisms that enable students with lower levels of cultural capital to attain this capital in school. Pedagogues (Mo-rais & Neves, 2010) suggest that better conditions for learning, as well as high expectations, are important for better results. However, economic, social and cultural status is not the only predictor of success in education: immigrant students score an average 27 points lower than native students, even after accounting for ESCS, in the OECD and 23.7 points lower in selected countries.
Table 5: Reading achievement of native students and students with an immigrant background who speak a language other than the language of instruction at home.
. , .	Achievement
Achievement
Second Gen- Second Generation Difference:	Difference: Native
eration Students Students Speaking Second Gen	Students and Im-
Speaking An- the Language of eration Students	migr,ant Students Country other Language Assessment at Speaking and who Speak a
at Home	Home	not Speaking	Language other
the Language of than the Language
the Language of of Assessment at
- Assessment at ,, „ .
% of Score % of Score	Home	Home, after Ac-
Students Points Students Points	counting for ESCS
Austria	50.6	428	18,3	441	13	31
Belgium	27.4	422	24.2	480	58	48
Finland	28.4	476	15.9	C	C	69
Germany	33.1	448	33.3	483	35	33
Norway	35.0	453	18.1	484	31	40
Croatia	1.4	C	65.8	466	C	C
Estonia	9.7	454	82.9	472	18	50
Hungary	1.7	C	41.0	527	C	C
Serbia	1.1	C	53.5	466	C	C
Slovenia	41.9	439	40.3	466	27	27
OECD Average	20.7	462	32.3	481	19	35
Note: The share of students is calculated from the immigrant student population and not from the
general population. The abbreviation C denotes missing data (see notes for Table 2). Data were obtained and calculated from the PISA 2009 database.
Another factor important for immigrant students' achievements is language, particularly when the language spoken at home is different from that used at school. After accounting for socioeconomic background, immigrant students whose households speak a language other than the PISA assessment language tested on average 35 points lower than non-immigrant students who spoke the instruction language at home. In selected countries, the difference is smaller by one point. The language spoken at home accounts for the highest number of score points of immigrant students in Finland (69 points), Estonia (50 points) and Belgium (48 points). It also explains a difference of 27 score points or more in other countries, except in Croatia, Hungary and Serbia, where the share of students speaking a different language at home is very small. In Hungary, slightly more than 6% speak a different language at home, and in Croatia and Serbia the share is 3% or less.
Table 6: Score point differences in reading achievement for immigrant students after accounting for enjoyment in reading and summarising strategies.
Country	Score Point Difference
Austria	32.2
Belgium	38.7
Finland	40.1
Germany	13.8
Norway	16.7
Croatia	8.4
Estonia	26.1
Hungary	-9.9
Serbia	-13.2
Slovenia	9.3
OECD Average	20.5
Selected Countries Average	16.2
Old Democracies Average	28.3
New Democracies Average	4.1
Note: Data were obtained and calculated from the PISA 2009 database.
There is an additional factor that influences educational achievement but is less commonly analysed: enjoyment in learning and learning strategies. PISA reveals that in OECD countries, boys are on average 39 points behind
girls in reading and suggests that differences in the way boys and girls approach learning, and how engaged they are in reading, account for most of the gap in reading performance (OECD, 20i0d, p. 13). Stanat and Christensen (2006) claim that the PISA 2003 assessment depicts immigrant students as motivated learners and learners with positive attitudes toward school. Such motivation can hardly persist if the student does not enjoy reading. A review of the PISA 2009 assessment shows that in OECD countries immigrant students on average demonstrate lower levels of enjoyment in reading and use less efficient summarising strategies, both possibly contributing to decreased results of 20.5 points. The link is particularly strong in Finland, where engagement in reading and learning strategies account for 40.1 score points, although it also exceeds 30 points in Austria and Belgium. In Slovenia, immigrant students would perform 9.3 points better if they reached the same level of enjoyment and awareness of learning strategies as native students. In Croatia, Hungary and Serbia, immigrant students enjoy reading and use affective learning strategies to a similar extent as native students. It is important to recall that immigrant students in the latter three countries have on average a similar socioeconomic status to their native peers, and that they nearly all speak the language of instruction at home as well. tte importance of these factors is confirmed by the low achievement gaps between native and immigrant students in all three countries. Because learning dispositions tend to be co-dependent with language proficiency, countries where language and enjoyment in reading account for differences between immigrant and native students should put more effort into improving the learning strategies and language skills of immigrant children, in order to, in turn, raise their level of learning enjoyment. In addition, children's aspirations often depend on the aspirations of their parents and those related to their socioeconomic status (Rothon, 2007, p. 315), which again indicates the importance of the overall inclusion of immigrants in society.
Finally, in addition to socioeconomic background, language used at home and students' attitude towards learning and learning strategies, PISA also assesses school-related factors, including the distribution of immigrant students across schools or the quality of teacher staff. These data tend not to be comparable and thus are not included in the present analysis. Specifically, school tracking already occurs in some countries at the age of 10, while in others it occurs just a few months before the PISA assessment (at the age of 15), and in the rest a few months after the assessment.
Croatia, Serbia and Slovenia in the Migration Regime Context
Assuming an accurate explanation of the respective differences and similarities in education achievements requires more than just a European, socialist or Yugoslav comparison and context, the present section analyses the educational achievement of immigrant students in Slovenia, Croatia and Serbia, taking into account national peculiarities. The three countries with state-historical, linguistic and cultural ties are - twenty years after Yugoslavia8 - new European democracies and, after wars, partitions, etc, the strongest economies emerging from the former common state. They are, however, also countries with considerable differences in their development and comparative educational achievements. What is more, they have different immigration histories.
Table 7: GDP (PPP) per capita and average PISA achievements (score points) in Croatia, Serbia and Slovenia.
Country	GDP (PPP)	Reading	Mathematics	Science
Serbia	10,991	442	442	443
Croatia	16,474	476	460	486
Slovenia	28,893	483	501	512
Note: Data on GDP (PPP) are presented in International Dollars and were obtained from the
World Economic Outlook Database (International Monetary Fund, October 2008). Reading results were obtained from the PISA 2009 database, mathematics results from the PISA 2003 database, and science results from the PISA 2006 database.
We see that GDP (PPP) per capita for Slovenia is more than double of Serbia's, while the GDP (PPP) per capita for Croatia is in between the two. tteir PISA results are distributed in the same manner: Slovenia has the highest GDP and the highest average PISA results. However, Slovenia also demonstrates the largest difference between native and immigrant students' achievement in reading, with an exceptionally low proportion of first generation immigrant students reaching at least level 3 of reading proficiency. In Croatia, where GDP is somewhat lower than in Slovenia, the achievement gap between immigrant and native students is relatively small. Serbia has the lowest GDP of the three countries, and immigrant students in Serbia perform better than their native peers.
8 Cf. Hudson and Bowman (Eds.) (2011) - After Yugoslavia.
Table 8: Reading achievement of native students compared to the average immigrant achievement in Croatia, Serbia and Slovenia (score points).
Country	Native Students	Immigrant Students
Serbia	443	457
Croatia	479	462
Slovenia	489	447.5
Note: Data were obtained from the PISA 2009 database.
Providing context for the PISA 2009 results in Serbia, Croatia and Slovenia offers some immediate explanations of the results. tte facts presented indicate that the higher achievement of immigrant students in Serbia, the relatively small gap in Croatia and the considerable difference in Slovenia is not solely a function of the economy. It is also clear that the increased achievement gap is not a result of smaller differences between native and immigrant students in formerly socialist countries. Another possibility could be that the results are a function of different education systems. Previous research suggests that a late school starting age and early school differentiation have a negative impact on the achievement of immigrant children (Crul & Schneider, 2009; Schütz & Wöß-mann, 2005). However, comparison reveals that all three systems have maintained their former structure, including prolonged primary education. tte wider inclusion of children in Slovenia in pre-primary education9 and the structure of the education systems, including their inclusiveness, do not support the idea of any difference occurring as a result of different educational arrangements.
ttus, another hypothesis emerges: in migration regimes, one should search for an explanation of different immigrant students' achievements before considering the educational reasons for the difference. Even though coherent national models of integration or incorporation are elusive (Freeman, 2004, p. 945), rendering it difficult to present coherent migration regimes, such categorisation can serve as a helpful tool in understanding state policies and their effects, even if they are not part of wider strategies. Coherence is not as important to this inquiry as the consequences of policies, practises or reactions in relation to questions of who
9 According to the Statistical Office of Republic of Slovenia, 73.9% of children of an appropriate age were included in pre-primary education in the school year 2009/2010 (http://www.stat. si/eng/novica_prikazi.aspx?id=3i39). ^e Ministry of Science, Education and Sports of the Republic of Croatia reports an enrolment rate of 58% for the same school year (http://public. mzos.hr/Default.aspx?sec=2i95) and the Ministry of Education and Science of the Republic of Serbia reports an enrolment rate of 47%. (http://www.mpn.gov.rs/aktuelnosti.php?id=4i7i)
44
immigrant students achievements in croatia, serbia and slovenia in context
immigrates and why, and how immigrants are included and accepted in society.
Table 9: Share of immigrant students and reading achievement in Croatia, Serbia and Slovenia.
	Native Students		First Generation			Second	First and Second	
Country			Students		Generation Students		Generation	
	%	Score	%	Score	%	Score	%	Score
		Points		Points		Points		Points
Croatia	89.3	479	3.5	452	7.2	465	10.7	461
Serbia	90.5	442	4.3	446	5.2	466	9.5	457
Slovenia	92.7	488	1.4	414	6.4	447	7.8	441
Note: Data were obtained and calculated from the PISA 2009 database.
Croatia and Serbia have only recently become countries of wider immigration. ttey faced massive, mainly forced, migration following the wars in the 1990s. On the other hand, Slovenia was and is a country of low skilled labour immigration, which is procured directly via work permits for certain professions only.10 What does PISA indicate in this regard? ttere are no considerable differences in the proportion of immigrant population in the respective countries; however, Croatia and Serbia have an important share of first generation students (33% and 45% of all immigrant students). Combining these facts with immigration statistics suggests that most immigrant students in Croatia and Serbia have a history of forced migration in their families. By contrast, Slovenia has only 18% first generation immigrant students (cal. from OECD, 2010b, p. 170), and nearly all of them are children of labour immigrants.
Table 10: Native and immigrant students' reading achievement before and after accounting for economic, social and cultural status.
Country	Difference in ESCS between Native/Immigrant Students	Immigrant Students' Score Points Difference after Accounting for ESCS
Croatia	0.26	8
Serbia	0.11	4
Slovenia	0.62	23
OECD Average	0.44	16
Note: Data were obtained from the PISA 2009 Database.
10 At the beginning of 2009, 14.6% of persons in employment in Slovenia were born abroad. In comparison to the native population, these immigrants have lower average education, are employed largely in construction and manufacturing and have low enrolment in tertiary education (SURS, 2010).
Comparing the economic, social and cultural status of immigrant and native students in the respective countries (Table 10), the differences are the highest in Slovenia, where after accounting for ESCS the gap narrows to 23 points (8 points in Croatia and 4 points in Serbia). Even if forced migration is unplanned and unpredicted, it seems that it involves people whose average socioeconomic status is higher than in the case of low skilled labour immigration. However, PISA assessments do not gather data on the ESCS of immigrants at the time they entered the country. Furthermore, the socioeconomic position of immigrant families also depends on their further inclusion in society.
Castels (1995) defines three broad approaches to ethnic diversity, all of which are closely linked to historical patterns of nation-state formation: differential exclusion (most clearly expressed in countries with "guest worker" immigration), assimilation (post-colonial countries) and multiculturalism (Australia, Canada, Sweden, USA). Applying this model to the PISA assessment results, immigrant students in general perform best in multicultural societies and worse under the pressure of differential exclusion. tte latter best describes the Slovenian pattern of integration. Croatia and Serbia may be classified by introducing a further developed and differentiated concept of "new immigration states" (cf. OECD, 2006, pp. 18-21; OECD, 2010f, pp. 24-27), with immigration specified as forced, of a wide social spectrum, with little or no language distance and occurring as part of a process of ethnic homogenisation. As a result, the inclusion of immigrants in Croatia and Serbia was faster and less problematic than in Slovenia. Since most immigrants were ethnic Croats or Serbs, they integrated and gained citizenship rights quickly. In Croatia in 1991 and 1992, more than 400,000 refugees from Bosnia and Herzegovina registered; it is estimated that 120,000 of these acquired Croatian citizenship (UNHCR, 2010)." In 1996, Serbia received more than half a million refugees, most of them ethnic Serbs from Croatia and Bosnia and Herzegovina; more than 200,000 of them gained Serbian citizenship (Republika Srbija, 2008).12 On the other hand, immigrants in Slovenia are not
11	^e Croatian census from 2011 reveals that the largest group of residents born outside the country were born in Bosnia and Herzegovina (456,580), Serbia and Monte Negro (86,830), Slovenia (21,985) and Macedonia (10,329); other countries do not achieve a figure of 10,000 people. (Kupiszewski, 2010, p. 121.)
12	^e Serbian census from 2002 reveals that the largest group of residents born outside the country were born in Bosnia and Herzegovina (381,659), followed by Croatia (351,263), Monte Negro (72,033), Macedonia (54,747) and Slovenia (13,128). (Kupiszewski, 2010, p. 134)
ethnic Slovenes. ttey struggle for years to obtain citizenship13 and are excluded from equal participation in the labour market and other spheres of social life, such as political participation. Of nearly 170,000 residents born in another country, the vast majority were born in one of the republics of former Yugoslavia14 (SURS, 2002). Recalled data from PISA 2009 that are complementary to the statistics presented above: 3% of immigrant students in Croatia and 2.7% in Serbia report that they speak a language other than the language of instruction at home. In Slovenia, the share is considerably higher: 13.5% among first generation and 41.9% among second generation students. By accommodating to the language of instruction and the rationale of national education, second generation students in Slovenia have closed the gap between themselves and native students by three quarters of a school year. This difference is considerably smaller in Croatia (14 points), while in Serbia first generation immigrant students already outperform their native peers (Table 9). Here again differences in the results arise from differences in migration regimes and from the background characteristics of the immigrant population.
It is important to note that the immigration regime in Slovenia was15 and is considered similar to German "guest worker" immigration. Slovenia has long considered immigrants to be "foreigners" and "others" whose presence is not permanent. Societal integration such as inclusion, education and enabling a place for immigrants to maintain an identity other than Slovenian have not been prime social or political goals. In parallel, in the time of Yugoslavia, immigrants did not perceive Slovenia as a foreign country; while it actually wasn't, their approach to teaching the Slovenian language and to education in general was indifferent. For the Slovenian population, political stratum, as well as for immigrants themselves, immigrants were and were not immigrants and none of them clearly decided how to treat the new situation (Kobolt, 2002). After the breakup of Yugoslavia, the political situation changed; Slovenia officially became a foreign country, but the relationship of native Slovenes towards immigrants and vice-versa did and did not change - both feel historically and
13	While most inhabitants with immigrant backgrounds do eventually receive Slovene citizenship (Bester, 2003, p. 282), the "guest worker" logic persists. In 2010, a total of 40,688 work permits were issued; most people with these permits were workers who had already been working in Slovenia for some time and were merely extending their employment (Employment, 2011).
14	67,670 were born in Bosnia and Herzegovina, 49,418 in Croatia, 6,437 in Yugoslavia (now the independent countries of Serbia and Monte Negro) and 27,238 in Macedonia. (SURS, 2002).
15	From 1962 to 1990, some 270,000 immigrants from other Yugoslav republics moved to Slovenia in search of work. ^ey were mainly unskilled and semiskilled workers. (Rizman, 1999, p. 157).
culturally connected, while at the same time treating each other as different. In a way, this makes the situation schizophrenic: achievement differences of immigrant students are identified but not properly addressed. tte story is different in the case of migration to Croatia and Serbia. ttere the immigration was mostly part of the enforced process of ethnic homogenisation. Accordingly, educational integration in Serbia and Croatia was conceptually and linguistically less problematic than in Slovenia.
Conclusion
Achievement gaps between immigrant and native students indicate failures in assuring educational equity in most countries assessed by PISA 2009 (42 score points in the OECD on average). Differences of this extent put the future prospects of considerable and growing parts of the population, and the societies in which they live, at potential risk.
Analyses of PISA 2009 results for 10 selected European countries (Austria, Belgium, Croatia, Estonia, Finland, Germany, Hungary, Norway, Serbia and Slovenia) reveal that on average native students perform better in old democracies. However, the achievement gap between native and immigrant students is on average larger in old democracies than in post-socialist countries. tte presented observation suggesting that the divide between old and new democracies could be one of the explanatory mechanisms is, however, also misleading, as, especially in the group of post-socialist countries, results vary considerably. Whereas the gaps in Estonia (35 points) and Slovenia (47 points) are close to the OECD average, the difference is considerably smaller in Croatia (18 points), while in Hungary and Serbia immigrant students perform better than native students.
Moreover, neither exceptionally good overall performance nor small achievement gaps guarantee a larger share of students with immigrant status achieving proficiency level 3, which promises professional success in the future. According to this criterion, old and new democracies perform similarly badly. For example, in Finland, the share of all students reaching proficiency level 3 is exceptionally high (76.1%), and even though it is low in Serbia (33.6%), the total proportion of immigrants reaching level 3 or more is similar in both countries (40.5% and 34.4% respectively).
Immigrants' socioeconomic background explains a large part of the differences in achievement, especially in old democracies such as Austria, Belgium and Germany, but also in one new democracy: Slovenia. tte link between lower achievement and not speaking the language of instruction at home is
48 immigrant students' achieyements in croatia, serbia and sloyenia in context
particularly strong in Finland, Estonia and Belgium, and plays an important role in other countries, with the exception of Croatia, Hungary and Serbia, where the percentage of immigrant students speaking another language at home is low (6.6% in Hungary and 3% or less in Croatia and Serbia). From the results presented, initial policy recommendations can already be drawn. tte governments of countries where the socioeconomic background of immigrants plays a significant role in the achievement of immigrant students should consider strengthening the socioeconomic position of immigrant families and should reconsider their inclusion policies. Where language proficiency plays a role, efforts should focus on improving immigrants' language skills. Language proficiency can also affect attitudes toward learning in general and impact the use and development of effective learning strategies, which is why immigrant students should be additionally encouraged to actively participate in the school process.
The present article demonstrates that in the international context the achievement of immigrant students and factors affecting those results cannot be accurately explained without data regarding specific migration regimes. Migration regimes can either be planned or can be the result of a spontaneous set of ad hoc rationales, policies, measures and events. However, migration affects the socioeconomic and cultural background of immigrant populations when they enter a country and when they subsequently undertake the process of integrating into a new society. Immigration patterns are important for achievement, particularly in the case of Croatia, Serbia and Slovenia. Immigrant students scored worse in a country with labour immigration than in countries with large scale forced immigration of people of the same ethnic (linguistic) origin. Understanding the importance of immigration regimes should not be equated with ignoring the importance of immigrant education policy, but it does remind educators to consider important external factors when designing educational policies.
References
Bourdieu, P. (1991). Language and Symbolic Power. Harvard: HUP.
Castels, S. (1995). How nation-states respond to immigration and ethnic diversity. Journal of Ethnic and Migration Studies, 21(3), 293-308.
Crul, M., & Schneider, J. (2009). Children of Turkish Immigrants in Germany and the Netherlands: ^e Impact of Differences in Vocational and Academic Tracking Systems. Journal of Teachers College Record, the Voice of Scholarship in Education, 11(6), 1508-1527.
DfES (2005). Ethnicity and Education: The Evidence on Minority Ethnic Pupils Aged 5-16. London: HMSO.
Employment Service of Slovenia (2011). 2010 Overview. Ljubljana: Employment Service of Slovenia. Freeman, G. P. (2004). Immigrant Incorporation in Western Democracies. International Migration Review, 38(3), 945-969.
Gillborn, D., & Mirza, H. (2000). Educational Inequality: Mapping Race, Class and Gender. London: Office for Standards in Education.
Huttova, J., McDonald, C., & Harper, C. (2008). Making the Mark? An Overview of Current Challenges in the Education of Migrant, Minority, and Marginalized Children in Europe. New York: OSI.
Kobolt, A. (2002). Zdej smo od tu. Ljubljana: i2
Kristen, C., & Granato, N. (2007). The educational attainment of the second generation in Germany: Social origins and ethnic inequality. Ethnicities, 7(3), 343-66.
Kupiszewski, M. (2010). Labour Migration Patterns, Policies and Migration Propensity in the Western Balkans. Warsaw: IOM.
Liebig, T., & Sousa-Poza, A. (2004). Migration, Self-Selection and Income Inequality: An International Analysis. KYKLOS, 57(1), 125-146.
Morais, A. M., & Neves, I. P. (2010). Texte educatifs et contexts favorisant l'apprentissage. Optimisation d'un modele de pratique pedagogique. In D. Frandji, & P. Vitale (Eds.), Actualite de Basil Bernstein - Savoirpedagogie et societe. Rennes: PUR.
OECD (2006). Where Immigrant Students Succeed: A Comparative Review of Performance and Engagement in PISA 2003. Paris: OECD Publishing.
OECD (2007). PISA 2006: Science Competencies for Tomorrow's World. Vol. 1: Analysis. Paris: OECD. OECD (2010a). PISA 2009 at a Glance. Paris: OECD Publishing.
OECD (2010b). PISA 2009 Results: What Students Know and Can Do - Student Performance in Reading, Mathematics and Science (Volume I). Paris: OECD Publishing.
OECD (2010c). PISA 2009 Results: Overcoming Social Background - Equity in Learning Opportunities and Outcomes. Paris: OECD Publishing.
OECD (2010d). PISA 2009 Results: Learning to Learn - Student Engagement, Strategies and Practices. Paris: OECD Publishing.
OECD (2010e). PISA 2009 Results: Learning Trends: Changes in Student Performance since 2000.
Paris: OECD Publishing.
OECD (2011). Lessons from PISA for the United States: Strong Performers and Successful Reformers in Education. Paris: OECD Publishing.
Republika Srbija, Komesarijat za izbeglice (2008). Stanje i potrebe izbegličke populacije u republici Srbiji. Retrieved 29. 07. 2011 from http://www.kirs.gov.rs/docs/StanjeIPotrebeIzbeglickePopulacije. pdf.
Rizman, R. (1999). Radical Right Politics in Slovenia. In S. P. Ramet (Ed.), The Radical Right in Central and Eastern Europe since 1989 (pp. 147-170). Pennsylvania: Pennsylvania State University Press.
Rothon, C. (2007). Can achievement differentials be explained by social class alone? An examination of minority ethnic educational performance in England and Wales at the end of compulsory schooling. Ethnicities, 7(3), 306-322.
Schütz, G., & Wößmann, L. (2005). Wie lässt sich die Ungleichheit der Bildungschancen verringern?
ifo Schnelldienst, 58(21), 15-21.
Simon, P. (2003). France and the unknown second generation: Preliminary results on social mobility.
International Migration Review, 37(4), 1091-1119.
Stanat, P., & Christensen, G. (2006). Where immigrant students succeed - a comparative review of performance and engagement in PISA 2003. Paris: OECD.
SURS (2002). Popis prebivalstva, gospodinjstev in stanovanj 2002: Priseljeni v Slovenijo po letu priselitve, državi prvega prebivališča in spolu. Retreived 29. 07. 2011 from http://www.stat.si/ popis2002/si/rezultati/rezultati_red.asp?ter=SLO&st=16.
SURS (2010). Socioeconomic Characteristics of Population (1 January 2009) and of International Migrants (2009), detailed data, Slovenia - final data. Retrieved 29. 07. 2011 from http://www.stat.si/ eng/novica_prikazi.aspx?id=3642.
UNCHR (2010). Submission by the United Nations High Commissioner for Refugees for the Office
of the High Commissioner for Human Rights' Compilation Report-Universal Periodic Review: The
Republic of Croatia. Retrieved 29. 07. 2011 from http://www.internal-displacement.org/8025708F00
4CE90B/%28httpDocuments%29/031701D196320550C12577BB00334845/$file/UNHCR,+Submissio
n+by+the+United+Nations+High+Commissioner+for+Refugees+for+the+Office+of+the+High+
Commissioner+for+Human+Rights+Compilation+Report+UPR+The+Republic+of+Croatia,+Mar
ch+2010.pdf.
Biographical note
Iztok Šori is a researcher at the Department of Sociology of the Faculty of Arts, University of Ljubljana. His main research interests are: equity and equality, migration, trafficking in persons and prostitution. Contact: iztok. sori@gmail.com.
Nika Šusterič is a student of Special and Rehabilitation Education at the Faculty of Education, University of Ljubljana. Contact: susteric.nika@ gmail.com.
Slavko Gaber is associate professor of the Sociology of Education at the University of Ljubljana. His research areas are educational policy, equity and equality and quality in education. Contact: slavko.gaber@guest.ames.si.