School segregation and the performance of immigrant and native pupils Giorgio Brunello (EENEE)

Score differences between first-generation immigrants and natives (left panel) and between second-generation immigrants and natives (right panel). Selected EU countries. Average value of test scores: a) reading: 500; b) math: 498. 15-years old students. Country FG Immigrants - Natives. Math FG Immigrants - Natives. Reading SG immigrants - Natives. Math SG immigrants - Natives. Reading Austria -61.8 -56.0 -58.4 -47.7 Belgium -83.3 -78.4 -68.9 -55.3 Czech Rep. -19.2 -20.3 -39.7 -19.7 Germany -73.2 -77.3 -51.7 -41.1 Denmark -80.3 -77.1 -61.0 -49.6 Spain -55.5 -48.3 -34.7 -46.9 Finland -97.9 -115.7 -70.3 -64.2 France -83.6 -93.0 -59.3 -54.4 UK -7.4 -12.5 -17.5 -9.2 Hungary 14.8 -6.3 44.7 32.7 Ireland -3.7 -12.1 0.8 -7.6 Italy -54.9 -74.3 -29.8 -39.6 Netherlands -60.4 -59.5 -56.9 -54.5 Norway -55.6 -68.0 -38.5 -31.0 Sweden -80.0 -96.2 -45.1 -39.3 Source: PISA 2012

Heterogeneity across Europe Gap lowest in the UK and Ireland Gap for second generation immigrants lower than for first generation

Parental background matters for the gap After conditioning on parental education and occupation, the relative situation of the children of immigrants improves everywhere, except for Finland and Portugal The gap falls substantially in France It shrinks by 25-50% in Nordic countries and by 30-45 % in Central European countries (see Dustmann, Frattini and Lanzara, 2012) Beside parental background, the quality of schools (and peers) and the language spoken at home help in explaining the observed gap

Figure 3.4: difference in the percentage of students with more than 100 books at home between natives and immigrants (x-axis) and difference in Math average test score between natives and immigrants (y-axis), by country. OECD area.

School stratification The distribution of immigrants across schools is not homogeneous School stratification or segregation: immigrants tend to attend schools in less affluent neighbourhoods where many disadvantaged natives also go These schools often have fewer resources, less qualified teachers and a higher share of peers who perform poorly

Does a higher share of immigrants affect school performance? Many parents believe that it does: flight of natives from schools with many immigrants In principle, there are both pluses and minuses from concentrating immigrants in schools Pluses may include: easier to recognize learning problems and address them Minuses may include: fewer interactions with natives, more disruption, inefficient use of teachers time

What the evidence says The empirical evidence for Europe suggests that the effect of the share on immigrants on the performance of natives is negative but small The effect tends to be negative and larger for immigrants Negative effects appear to be larger above a threshold share of immigrants, although there is little consensus on the value of the threshold Native flight also higher above thresholds

The Effects of the Share of Immigrants in the Class or School on the Performance of Natives and Immigrants. European Studies Authors Estimated effect Estimated Effects Szulkin and Jonsson (2007) Immigrant concentration on school grades in Sweden Natives: negative effects Non-natives: negative effects Jensen and Rasmussen (2011) Immigrant concentration on PISA test scores. Denmark Andersen, S and Thomsen, M. (2011) Immigrant concentration on school grades Immigrants: negative effects Brunello and Rocco (2013) Immigrant concentration on natives’ PISA test scores. Natives: small negative effects. Geay et al. (2013) Immigrant concentration on the test scores of native English speakers in England. Natives: no sizable negative effects Ohinata and Van Ours (2013) Immigrant concentration on PIRLS and TIMSS test scores Hardoy and Schøne (2013) Immigrant concentration in Norway upper secondary schools on dropout rates of natives Ballatore et al. (2014) Immigrant concentration on test scores. Italy Natives: substantial adverse effects Schneeweis (2015)   Immigrant concentration in Austrian primary schools on repetition in primary and secondary schools and track attendance after primary education.

Redistribution: equity and efficiency Redistributing immigrant pupils from high share to low share schools reduces inequality of outcomes Redistribution is efficient only when the performance gain accruing to high share schools is higher than the performance loss for low share schools. Evidence points to likelihood of efficiency gains.

Policies in place Lotteries for school choice (ex: UK) Bussing (ex: Denmark, France) Additional financial resources to high share schools (ex: Switzerland) Parental information: (ex: Netherlands) Ceilings to the share of immigrants: (ex: Italy)

Do these policies work? Evidence suggests that bussing policies and the provision of additional resources improves the performance of immigrants Scarcity of serious evaluations and of costs/benefits studies – especially in Europe - makes it difficult to reach conclusions on which policy works and on best practices

More details on this Forthcoming EENEE analytical report “School segregation of immigrants and its effects on educational outcomes in Europe”, by Giorgio Brunello and Maria De Paola