ArticlePDF Available

PISA Performance of Natives and Immigrants: Selection versus Efficiency

June 2020
2(1):9-36

DOI:10.1515/edu-2020-0108

License
CC BY 4.0

Authors:

In most countries, immigrant and native students perform differently in the Programme for International Student Assessment (PISA) due to two main reasons: different immigration regimes and differences in their home-country educational systems. While there is sophisticated literature on the reasons for these performance gaps, it is barely considered in the educational efficiency research. Our approach distinguishes between selection effects caused by immigration policies, and the efficiency of educational systems in integrating immigrant students, given their socio–economic background. Accordingly, we split our sample, which consists of 153,374 students in 20 countries, calculate various different efficient frontiers, and ultimately decompose and interpret the resulting efficiency values. We find large differences in educational system efficiency, when controlling for negative selection effects caused by immigration regimes.

Relationship between PISA scores and ESCS values; left: within-country correlation, right: average gaps on country-level

…

Benchmarking of a country with two student groups. The left panel shows three frontiers, the right panel shows the rectangle enlarged.

…

Benchmarking of another country with two student groups and for both countries together

…

Country rankings based on mean PISA scores of immigrants and of mean eflciency

…

PISA scores correlations coeflcients, overview

…

Figures - available via license: CC BY

Content may be subject to copyright.

Access to this full-text is provided by De Gruyter.

Content available from Open Education Studies

This content is subject to copyright. Terms and conditions apply.

bution 4.0 License

Open Education Studies 2020; 2:9–36

Research Article

Andreas Behr* and Gerald Fugger

PISA Performance of Natives and Immigrants:

Selection versus Eciency

https://doi.org/10.1515/edu-2020-0108

Received Mar 17, 2020; accepted Apr 29, 2020

Abstract: In most countries, immigrant and native stu-

dents perform dierently in the Programme for Interna-

tional Student Assessment (PISA) due to two main rea-

sons: dierent immigration regimes and dierences in

their home-country educational systems. While there is

sophisticated literature on the reasons for these perfor-

mance gaps, it is barely considered in the educational ef-

ciency research. Our approach distinguishes between se-

lection eects caused by immigration policies, and the ef-

ciency of educational systems in integrating immigrant

students, given their socio–economic background. Accord-

ingly, we split our sample, which consists of 153,374 stu-

dents in 20 countries, calculate various dierent ecient

frontiers, and ultimately decompose and interpret the re-

sulting eciency values. We nd large dierences in edu-

cational system eciency, when controlling for negative

selection eects caused by immigration regimes.

Keywords: Data Envelopment Analysis, Migration, PISA,

Eciency decomposition

JEL Classication: C14 C61 I21

1Introduction

The dierences between natives and immigrants in the Pro-

gramme for International Student Assessment (PISA), pub-

lished by the Organisation for Economic Co-operation and

Development (OECD), has gained considerable attention

in the literature.¹Apart from social, cultural, religious and

*Corresponding Author: Andreas Behr:

University of Duisburg-

Essen, andreas.behr@uni-due.de, ORCID iD: 0000-0002-6818-1878;

We thank two anonymous reviewers whose comments have helped

to improve and clarify this manuscript

Gerald Fugger:

University of Duisburg-Essen, ORCID iD: 0000-0001-

8546-7600

PISA is a worldwide study that assesses the 15–year–old students’

performance in mathematics, science, and reading. In addition, the

individual backgrounds of the pupils and school data are collected.

historical reasons, dierent immigration policies and dif-

ferent levels of success in integrating immigrants are the

two most important aspects (Kunz, 2016; Isphording et

al., 2016). Countries attract dierent groups of immigrants

with dierent socio–economic environments, due to coun-

try attractiveness, as well as their immigration policies (En-

torf & Minoiu, 2005; Hochschild & Cropper, 2010). In most

countries, socio–economic endowment is one of the most

important factors for the educational success of students

(Parr & Bonitz, 2015; Rogiers et al., 2020). This is illustrated

by the left panel of Figure 1, which shows a strong posi-

tive within-country correlation between the average read-

ing, mathematics and science student scores in PISA, and

their average ESCS values, the latter being an index of their

socio–economic backgrounds, in 2015.²The index of eco-

nomic, social and cultural status (ESCS) comprises several

subcategories in the areas of parental education, highest

parental employment and student housing. It is consid-

ered to be an appropriate measure of the students’ socio–

economic background (Hwang et al., 2018). The right side

of Figure 1 shows the strong correlation between socio–

economic endorsement gaps (ESCS gaps) and educational

performance gaps between natives and immigrants across

countries (Rogiers et al., 2020).³

Our descriptive analysis reveals substantial educa-

tional (PISA) and socio–economic (ESCS) gaps between

immigrants and natives and, that performance compar-

isons to a large extent implicitly reveal the students’ dif-

ferent social and economic backgrounds. Without account-

Following the PISA denition of immigration, an immigrant foreign-

born in the second or rst generation (OECD, 2017).

Our included countries are: Australia (AU), Austria (AT), Belgium

(BE), Canada (CA), Denmark (DK), Finland (FI), France (FR), Germany

(DE), Israel (IL), Italy(IT), Netherlands (NL), New Zealand (NZ), Norway

(NO), Portugal (PT), Singapore (SG), Spain (ES), Sweden (SE), Switzer-

land (CH), United Kingdom (GB), and the United States of America

(US).

Figure S1 in the appendix shows the positive correlation between

average PISA scores and the average ESCS scores on the country-level.

The right panel elucidates that the average PISA scores are negatively

correlated with the mean absolute deviations from the median ESCS

scores.

10 |A. Behr and G. Fugger

Figure 1: Relationship between PISA scores and ESCS values; left: within-country correlation, right: average gaps on country-level

ing for the students’ backgrounds, studies run the risk of

making implicit statements about immigration policy. We

take this problem into account explicitly, by analysing the

performance of the educational system, given the varied

social backgrounds of immigrant and native students.

An educational system can be integrating, despite a

large educational gap, if it at least partially compensates

for the gaps in socio–economic background. We use Data

Envelopment Analysis (DEA) to examine the eciency

of educational systems. DEA models provide eciency

scores based on the students’ performance relative to the

performance of the best students comparable in their ESCS

endowments. Our analysis is conducted at student–level,

the most disaggregated data available in PISA. The stu-

dents are evaluated according to their ability to maximise

PISA scores given their socio–economic endowment. To

account for the dierences in socio–economic endowment

between immigrants and natives, we split the PISA 2015

data into subsamples of natives and immigrants.Eciency

scores are calculated relative to various eciency fron-

tiers, which provides further insights and fosters our un-

derstanding of the relationship between selection eects

in immigration, and the integrational abilities of the ed-

ucational institutions in this context. Educational system

performance is then obtained from the average eciency

scores of the students and further decomposed.

Our rst eciency analysis uses the average PISA

score of the mathematics, science, and reading scores as

output and the ESCS values as input. These three PISA

scores are highly positively correlated. The aggregation

into one output enables a straightforward interpretation

and decomposition of the eciency frontiers. In a further

analysis we use the ESCS as input and include the three

PISA scores (mathematics, science, and reading) as seper-

ate outputs. DEA models allow the inclusion of several out-

puts, whereby all inputs and outputs are simultaneously

included in the eciency assessment by weighting them.

The results of the second eciency assessment conrm

our main ndings for average scores that in countries with

restrictive immigration regimes, immigrants are not only

performing relatively well but also use their endwoments

rather eciently. Some countries (e.g. Spain and France)

perform considerably better according to their eciency

considering their ESCS endowments relative to their PISA

ranking.

After this introduction, a literature overview of the

performance gaps between natives and immigrants is pro-

vided. The third section outlines our methodology. In sec-

tion four we explain the methodology of the ESCS and PISA

scores, the dierences between immigration regimes, and

provides some initial results. The results of the eciency

analyses and their decomposition are discussed in section

ve preceding the conclusion.

PISA Performance of Natives and Immigrants: Selection versus Eciency |11

2Literature Overview

Dierences in country attractiveness for immigrants, and

dierent immigration policy regimes, attract dierent

groups of immigrants, resulting in heterogeneous immi-

gration populations between countries, and a wide range

of challenges for the educational systems and societies

in general (Entorf & Minoiu, 2005; Hochschild & Cropper,

2010). While some countries attract immigrants whose

socio–economic endowments are equal or even higher

than those of the natives (Arabian oil-based economies,

English speaking countries and Singapore), others, such

as Central European countries, mainly attract immigrants

who have a poorer socio–economic endowment than the

natives (Jerrim, 2015). In Austria, Denmark, and Germany,

for example, the dierences between native and immi-

grant students are especially striking (Rindermann &

Thompson, 2016).

Besides their levels of educational, human capital,

and wealth-related aspects (all part of PISA’s ESCS index),

native and immigration populations may also dier in cul-

tural, religious, historical, and reputational aspects (Parr

& Bonitz, 2015; Kunz, 2016). Immigrants may also face for-

mal rights and legal status challenges, lack accumulated

experiences as well as social connections that may result

in educational information asymmetries, which can inu-

ence the educational performance of their children (Rin-

dermann & Thompson, 2016; Camehl et al., 2018).

Schneeweis (2011) decomposes the educational gap

between immigrants and natives using the data of ve in-

ternational student assessment studies. Her results show

that institutional characteristics of the education systems

can increase dierences between immigrants and natives.

The results of Borgna & Contini (2014) indicate that ed-

ucational institutions and socio–economic backgrounds

are mostly causing the gaps between immigrants and na-

tives. Furthermore, PISA 2006 and 2009 data reveal that

school attendance signicantly reduces educational gaps.

Dronkers et al. (2014) nd that the countries’ educational

systems and the students’ individual characteristics cause

the dierences between immigrants and natives. Harris et

al. (2019) show that the access to certain areas of the cur-

riculum depends at least in part on the socio–economic

endowment of the students in the schools. Woessmann

(2016) nds that educational institutions and family back-

ground have the highest explanatory power in determin-

ing educational achievements. Interestingly, the impact

of school resources is much smaller than the students’

social-economic endowment and institutional characteris-

tics, which is also found by Falck et al. (2018).

Further empirical studies based on PISA data reveal

that the dierent socio–economic backgrounds of immi-

grants and natives have the highest overall explanatory

power regarding dierences in educational attainment. Es-

pecially in European countries, nearly three–quarters of

the performance gaps between natives and immigrants are

accounted for primarily by dierences in economic, social,

and cultural status (Ammermueller, 2007; Levels et al.,

2008; Arikan et al., 2017). Other factors, like linguistic bar-

riers (previously considered the most important barrier for

immigrants) only partially explain the performance gaps

(Isphording et al., 2016; Rindermann & Thompson, 2016).

Another important aspect in explaining performance

gaps is the selection process among immigrants. Indi-

vidual background factors vary between dierent immi-

grant groups which themselves vary between the countries

(Schnepf, 2007; Arikan et al., 2017). In countries where

immigrants are highly educated like Australia, they per-

form on average better in national and international com-

parisons than their native counterparts (Dustmann et al.,

2012; Jerrim, 2015). The opposite holds for Central Euro-

pean countries in which a considerable share of the immi-

grants have on average a lower economic, social, and cul-

tural status than the population of their immigration tar-

get countries and perform worse in PISA (Dustmann et al.,

2012; Rindermann & Thompson, 2016; Arikan et al., 2017).

Accordingly, heterogeneous immigrant populations

provide specic challenges for educational systems that

should be considered in eciency analysis. Although the

ESCS is an input (among others) in most educational

eciency analyses, regarding the importance of socio–

economic backgrounds, international eciency studies

are decient, in how they consider the dierences between

immigrants and natives within and between countries.

Eciency scores are based on the relationship be-

tween the sum of weighted output to the sum of weighted

input of the students relative to the best students. As

the socio–economic status is an environmental or non–

discretionary input, it is not amenable to direct control by

the educational system, and therefore cannot be regarded

as a traditional input in eciency analysis. But since it

is found to have a signicant impact in determining per-

formance in PISA, socio–economic status is included in

most eciency analyses (Agasisti & Zoido, 2018). For ex-

ample, Sutherland et al. (2009) argue that student achieve-

ments depend on their social environment (family and

peer–groups) and therefore must be included in student ef-

ciency analysis. Similarly, Cordero-Ferrera et al. (2017) ar-

gue that student socio–economic background is crucial for

evaluating students according to their ability to make the

most with their inputs (Cordero-Ferrera et al., 2017). Apari-

12 |A. Behr and G. Fugger

cio et al. (2017a) refer to students as “raw material” that is

transformed in schools and the impact of which is best re-

ected by the students’ socio–economic status (Aparicio

et al., 2017a).

In the cross-country analyses of Sutherland et al.

(2009), Aparicio et al. (2017a), and Agasisti & Zoido (2018),

the students are not distinguished according to their coun-

try of origin. Moreover, the studies do not account for se-

lection eects caused by immigration policies, that can

lead to distinct immigrant groups with dierent socio–

economic backgrounds. Aparicio et al. (2017b) proxy the

socio–economic backgrounds of students by including the

educational experience of their parents, which is only

one aspect of the broader ESCS. As the performance gap

determinants are manifold, a more comprehensive index

should be preferred. De Witte & Lopez-Torres (2017) pro-

vide a broad overview of recent educational eciency stud-

ies.

A considerable number of publications have been pub-

lished in both the eciency strand and the literature

strand, focusing on the determination of the performance

gaps between immigrants and natives. However, no inter-

national educational eciency study so far accounts for

the dierent challenges arising from dierent immigration

policy regimes.

3Methodology

In this section we explain our notation and our method-

ological approach in detail using a small articial data

set. As our decomposition approach regards several coun-

tries and the two subsets of students with or without im-

migration background, we introduce index sets denoted in

calligraphic characters to facilitate referencing to specic

groups of students.

3.1 Sets of students

The set of all countries is denoted Kand individual coun-

tries are referred to with index k(k= 1,. . . ,K).In each

country kwe have two sets of students. The set of students

in country khaving an immigration backgroundis denoted

with Ik.Immigrant students in country kare referred to us-

ing the index i(i= 1,. . . ,Ik).Native students (home) in

country kbuild the index set Hkand are indexed with h

(h= 1,. . . ,Hk).All students in country k, that is students

with and without immigration background are referred to

with Ek={Ik,Hk}.

Calligraphic characters without an index refer to the

set combining the subsets from all Kcountries. I.e. E=

{E1,. . . ,Ek,. . . ,EK}is the set of all students from all K

countries and I={I1,. . . ,Ik,. . . ,IK}is the set of all stu-

dents with immigration background from all Kcountries.

We also have E={I,H}with H={H1,. . . ,Hk,. . . ,HK}.

3.2 Students and dierent frontiers of

potential scores

In our illustrating example we only consider two countries,

that is kand k′.First, we consider country kand the two

subsets Ik(immigrant students) and Hk(native students).

For each we observe their input x(ESCS-score) and their

output y(PISA-score). We represent in Figure 2 native stu-

dents (Hk)by closed circles and students with immigra-

tion background (Ik)with open circles.

We observe that some students with rather similar in-

puts reach quite dierent outputs. The observations of

the ’best students’, subsequently named ecient students,

are joined with linear junctions and the resulting frontier

is used as a yardstick to benchmark the remaining stu-

dents. How we identify the best students is explained in

more detail below (see model 2). As we have three dier-

ent subgroups, natives (Hk),immigrants (Ik)and all stu-

dents combined (Ek),we can obtain three dierent fron-

tiers. These frontiers we denote in general by Fand the su-

perscript indicates based on which subset of students the

frontier is obtained, accordingly we have drawn the three

dierent frontiers FIk,FHkand FEkin Figure 2.

3.3 Benchmarking individual students

The performance of a specic student h, we pick for illus-

tration the one indicated with the square, can now be as-

sessed using three dierent benchmarks. To ease the read-

ability, the right panel of Figure 2 displays a part of the left

panel enlarged.

A benchmark student denoted by ˜

h1is a synthetic

student on frontier FHk.This benchmark student is a lin-

ear combination of two ecient native students located

at the frontier FHk(dotted line). If we compare the ob-

tained score of student hwith the score of ˜

h1on the fron-

tier FHk, we obtain a relative eciency score of DHk(h) =

2.740/3.230 = 0.850.We use Dfor the eciency score

and the superscript indicates on which set of students the

frontier is obtained, here we use frontier FHk. Hence, the

student honly obtained 85% of the score that is regarded

as being possible given his input amount. Or, equivalently,

PISA Performance of Natives and Immigrants: Selection versus Eciency |13

Figure 2:

Benchmarking of a country with two student groups. The left panel shows three frontiers, the right panel shows the rectangle

enlarged.

he could increase his output by 17.6% if he would be as

ecient as his benchmark fellow students.

If we compare our native student hwith an ecient

synthetic student with immigration background ˜

h2,which

is located at the frontier FIk(solid line) obtained from

immigration students Ik,we obtain students hscore as

DIk(h)=2.740/3.370 = 0.810,hence, in this compari-

son he is underperforming by 19%.

And nally we can benchmark student hwith syn-

thetic student ˜

h3located at the frontier FEk(dashed line)

which is based on all students in country k.As this hypo-

thetical benchmark student ˜

h3performs even better than

h1and ˜

h2,we nd that according to this yardstick, student

hunderperforms by DEk(h)=2.740/3.510 = 0.780, i.e.

22%. Note that in this last comparison the benchmark stu-

dent ˜

h3is a hypothetical student obtained as a linear com-

bination of an ecient native and an ecient immigrant

student.

3.4 Benchmarking sets of students

To obtain a measure of the performance of a complete set of

students we use the arithmetic mean of individual scores.

E.g. to obtain the average performance of immigrant stu-

dents Ikusing the frontier FIkobtained based on this set

of students, we calculate

MIk(Ik)=1



i=1

DIk

i(Ik)(1)

Ikis the number of students benchmarked, here the stu-

dents with immigrant background in country k. We use M

for arithmetic mean, the superscript Ikto indicate that we

use the frontier FIkand the argument in parentheses indi-

cates which group of students is benchmarked.

In our illustrative example considered in Figure 2 we

obtain for immigrants MIk(Ik)= 0.827 and for natives

MHk(Hk)= 0.839.For comparing the performance of im-

migrants and natives, one may like to use the frontier FEk

obtained considering all students Ekin country k.In this

example we obtain for immigrants MEk(Ik)= 0.788 and

for natives MEk(Hk)= 0.796 as average eciencies.

3.5 Considering a second country

We now consider a second country k′.We use lled dia-

monds for native students and open diamonds for immi-

grant students. The left panel of Figure 3 contains the situ-

ation for country k′,again with the three dierent national

frontiers indicated by dotted, dashed and solid lines. The

right panel combines the students of both countries and al-

lows us to obtain an international frontier FEcollated from

all students of all (here: two) countries.

This allows the benchmarking of the immigrant stu-

dents of county k(Ik)and of the native students of coun-

try k(Hk)using the international frontier. E.g. our student

hof country kis now benchmarked based on the score

of a synthetic student ˜

h4located at the international fron-

tier FE. Accordingly in this comparison her eciency score

DE(h) = 2.740/4.010 = 0.680 is the lowest obtained in

the comparisons and hints for a potential increase in her

score of 47%.

14 |A. Behr and G. Fugger

Figure 3: Benchmarking of another country with two student groups and for both countries together

Using the international frontier FEfor benchmarking

all native students in country kresults in an average score

ME(Hk)= 0.686.The immigrants of country kobtain an

average score ME(Ik)= 0.698.

The DEA model

We use the output–oriented BCC model, introduced by

Banker et al. (1984). The output orientation implies that

students maximise their output given their inputs. For stu-

dent othe model is dened as:

min

η,v,uη=

vixio −u0(2)

subject to 

uryro = 1



vixij −

uryrj −u0≥0 (j= 1,. . . ,n)

vi≥0 (i= 1,. . . ,m)

ur≥0 (r= 1,. . . ,s)

u0free in sign.

Output rof student ois given by yro and is weighted by

ur(r= 1,. . . ,s).sequals the number of outputs. Her in-

put i(xio ) is weighted by vi(i= 1,. . . ,m).mis the num-

ber of inputs and nis the number of all students under

analysis. The weights are restricted to be non–negative, de-

rived from the data, and most likely vary between students.

The weights are not chosen a priori but determined when

solving the linear program. The most favourable composi-

tion of weights to make student oas ecient as possible

are chosen given the restrictions. The linear program is set

up and solved for each student under analysis indivdually

(Behr, 2015; Cooper et al., 2007).

η*denotes the solution to the minimisation problem.

For convenience, we dene D*=1

η*. If η*=D*= 1 student

ois ecient. The limits of η*and D*depend on whether

the student obelongs to the group of students she is com-

pared to. If she belongs to the group of students she is com-

pared to, η*is equal to or greater than one and D*is equal

to or less than one. If student odoes not belong to the

group of students she is compared to, η*may be less than

one (the student is super–ecient). In this case the stu-

dent ois above the eciency frontier of the students she

is compared to, and D*is greater than one (Chen, 2005).

The scalar u0is free in sign and implements the

assumption of variable returns to scale (VRS). VRS al-

low non–proportional output changes when the inputs

change. The input and output tuples of students are nei-

ther allowed to be scaled up (increasing returns to scale)

nor down (decreasing returns to scale) in the BCC model.

4The PISA Study, Migration

Regimes and Descriptive Results

We use students’ socio–economic status as the input and

the average PISA score of the students in reading, mathe-

matics and science, as output in the rst eciency anal-

ysis. If necessary, the data are transformed to obtain posi-

tive values as DEA can only handle positive inputs and out-

puts. Outliers are excluded.

PISA Performance of Natives and Immigrants: Selection versus Eciency |15

4.1 The PISA study and the ESCS

PISA is a worldwide stratied two-stage sample study con-

ducted by the OECD, to measure 15-year-old students’ per-

formance in mathematics, science, and reading. It was con-

ceived to oer insights into sources of performance varia-

tion within and between countries. It was rst performed

in 2000 and then repeated every three years. The PISA as-

sessment in 2015 focused on science, and was published

in December 2016 (OECD, 2016). Student performance is

reported as the corresponding mathematics, science, and

reading scores.⁴

A minimum of 150 schools must be selected in each

country to ensure quality standards. If a participating

country has fewer than 150 schools, all schools are se-

lected. Within each participating school, a predetermined

number of 15-years-old students, usually 42 students, is

randomly chosen with equal probability. In schools with

fewer students, all students are selected. If the response

rate is too low, the sample size of schools is increased

beyond 150 to ensure a minimum student sample size.

A response rate of 85% is required for initially selected

schools. If the initial school response rate falls between

65% and 85%, an acceptable school response rate can still

be achieved by using replacement schools. Schools are

classied into similar groups according to selected vari-

ables (region, private or public school, funding,. . . ). A

minimum student response rate of 50% within each school

is required for a school to be regarded as participating

(OECD, 2016).

Since its publication, the results of the PISA study

have inuenced the design of the education systems of the

participating countries. For example, Ho (2016) shows how

the insights resulting from PISA were used in Hong Kong,

Damiani (2016) in Italy, and Ababneh et al. (2016) in Jor-

dan. Tobin et al. (2016) provide a world wide overview of

how large-scale educational assessments inuence educa-

tion policy and most studies nd signicant eects of sec-

ondary education on the economic development of coun-

tries (Aduand and Denkyirah, 2017; Karatheodoros, 2017).

The index of economic, social and cultural status

(ESCS) comprises three main categories: parental educa-

tion, highest parental occupation, and home possessions.

The latter combines ve indices: family wealth, household

possessions, cultural possessions, home educational re-

sources, and information and communication technology

Now data for 2018 have become available but the preliminary ver-

sion of 2018 is still incomplete and lacks for example individual scores

in of the three subjects for spanish students.

resources. These indices are derived from the availability

of 16 household items at home, including three country-

specic household items. The ESCS’s three main compo-

nents are standardized with a mean of zero and a standard

deviation of one, over the full sample. Finally, a principal

component analysis (PCA) of the three main components

is conducted, and the ESCS is dened as the rst principal

component score (OECD, 2017).⁵For rst-generation immi-

grants, parental education and partly the highest parental

occupation may result from the educational institutions

of their country of origin, rather than from integration re-

sults or the educational system of their target country, in

whose educational eciency we are interested. However,

both the home possession measures and the success of the

second-generation immigrants depend on the integration

and education quality in their target country (Reparaz &

Sotés-Elizalde, 2019). The ESCS covers a wide range of dif-

ferent economic, social and cultural topics, enabling an

approximation of possible determinants of education per-

formance gaps between immigrants and natives. Further-

more, through the use of PCA, the ESCS is a construct that

is well suited for capturing and comparing the whole stu-

dents’ socio–economic status (Hwang et al., 2018).

4.2 Migration regimes

When examining the eciency of educational systems in

terms of the immigrant performance, the respective immi-

gration regimes of the countries must be taken into ac-

count. Bjerre et al. (2015) and Bonjour & Chauvin (2018)

provide an overview of a large number of denitions and

distinctions in the literature.

In addition to limiting ocial immigration policies

(strict ones are mainly based on points systems), another

important aspect is how many people enter the country

through unocial channels. For example, a comparison

between Germany and Australia shows that the proportion

of immigrants in Australia for family and humanitarian

reasons is far lower and the percentage who do so for eco-

nomic reasons is higher (Beine et al., 2016). Based on their

selective immigration policy and low proportion of non-

economic immigration, Australia, Canada, New Zealand,

and the United Kingdom can be regarded as having rather

restrictive immigration regimes. The United States of Amer-

ica also has a restrictive immigration policy, but unlike

The common ESCS component weights across cycles are 0.79

(parental occupation), 0.82 (parental education), and 0.74 (home pos-

sessions) (OECD, 2014).

16 |A. Behr and G. Fugger

the remaining countries in this group, it does not succeed

in attracting immigrants who perform on average at least

as well as their native peer group, as shown below (see

also Camarota & Zeigler (2016)). The European Union intro-

duced a points-based system in 2009, but it is far less strict

than in the other countries with a selective immigration

policy, and the share of immigrants for family and human-

itarian reasons is relatively high. Therefore, we do not re-

gard the members of the European Union as being restric-

tive (Bertoli et al., 2016).

We use the average occupational status of parents,

which is available in PISA (higher values stand for better

status) to substantiate our country classication. The oc-

cupational status of parents is an important determinant

of the educational attainment of immigrants, as the ed-

ucational mobility of immigrants is generally lower than

that of natives (Schneebaum et al., 2016; Reparaz & Sotés-

Elizalde, 2019). Descriptive results show that in most coun-

tries, the occupational status of parents of natives is higher

than that of immigrants. Only in countries with a selective

immigration regime, are the gaps close to zero or even neg-

ative. Singapore attracts immigrants whose parents have

the highest level of education.⁶These results can be pro-

vided upon request.

4.3 The data and descriptive results

Our sample comprises 153,374 students in 20 industri-

alized countries for PISA 2015.⁷We combine rst– and

second–generation immigrants, otherwise several coun-

tries would have too few data points in at least one group

(e.g. Finland and the Netherlands), and both groups have

similar performance dierences (relative to the natives),

In Singapore, the recruitment of skilled workers is systematically

promoted and part of the ocial government strategy, as the following

quote from prime minister Goh Chok Tong’s speech at the national

day rally 2001 shows:

“

[...]

some Singaporeans may again question the need for more

global talent. I urge you to understand that this is a matter of life

and death for us in the long term.

[...]

If we do not top up our

talent pool from the outside, in ten years time, many of the high-

valued jobs we do now will immigrate to China and elsewhere,

for lack of sucient talent here.” (Tong, 2001)

Singapore is the most successful of all countries in attracting highly

qualied and top performing immigrants. In our analysis, immigrants

in Singapore are on average the most ecient.

Japan, Korea, and Poland are excluded because of having too few

immigrants.

which are determined to a similar extent by the ESCS

(Rangvid, 2007).

As a frontier based non-parametric technique, DEA is

sensitive to outliers. We exclude outliers based on their in-

uence, measured by Cook’s distance. We dene outliers

to have a Cooks’ distance of at least eight times the aver-

age distance for each country and each regression, which

is a reasonable threshold according to Cook (1979).⁸Table

S1 shows the number (between 44 and 207) and the per-

centages (ranging from 0.759% to 1.186%) of excluded

outliers per country.

PISA reading, mathematics, and science scores are

constructed to have an international mean of 500 and a

standard deviation of 100. The standardization provides

student results that are directly comparable between coun-

tries. Table 1 summarizes within–country correlations be-

tween the scores. All scores are highly positively corre-

lated, and the correlations vary between 0.743 for the

mathematics and reading results in Italy, and 0.908 for

the reading and science results in Singapore. Table S5 in

the appendix depicts the correlation coecients for each

country.

We use the students’ average PISA scores as output

y, to enable comprehensible visual and contextual illus-

trations. After discussing the results for the average PISA

score as output, we also present the results for the three

PISA scores in mathematics, science, and reading as out-

puts.

Figure 4 presents the average PISA score distributions,

using a Gaussian kernel with a bandwidth of 70% of Sil-

verman’s “rule of thumb” to disclose more details for im-

migrant and native students separately for each coun-

try (Silverman, 1986). In countries with selective immi-

gration policies, as well as in Israel and Portugal, immi-

grants and natives perform similarly well. In Singapore,

the immigrants perform even better than the natives. In

the other countries and especially in most European coun-

tries, natives perform better. The dierences between na-

tives and immigrants between countries further indicate

that the prevailing immigration regime inuences the se-

lection among immigrants. However, Figure 4 focuses only

on our output and does not distinguish between the selec-

tion eects and the eciency of educational systems. Fig-

ures S2 to S4 in the appendix provide the distributions of

the three PISA scores. They are rather similar to the dis-

tributions of the average PISA scores and the same dis-

The results are robust for alternative thresholds (e.g. from two times

up to 20 times the average distance) and can be providedupon request .

PISA Performance of Natives and Immigrants: Selection versus Eciency |17

Table 1: PISA scores correlations coecients, overview

Scores Min Country Max Country Mean

Mathematics-Reading 0.743 Italy 0.860 Netherlands 0.795

Mathematics-Science 0.849 Italy 0.899 France 0.883

Reading-Science 0.828 Sweden 0.908 Singapore 0.868

Figure 4: Average PISA scores distributions among natives (straight line) and immigrants (dashed line)

18 |A. Behr and G. Fugger

Table 2: Descriptive results of the average PISA scores and the ESCS values

Countries Group PISA Mean di. ESCS Mean di. Corr n

Australia (AU) Nat 492 −18.872 0.194 −0.100 0.403 10744

Mig 511 (2.012) 0.294 (0.017) 0.369 2651

Austria (AT) Nat 508 60.012 0.207 0.542 0.373 5533

Mig 448 (2.626) −0.335 (0.026) 0.299 1242

Belgium (BE) Nat 519 54.107 0.272 0.407 0.446 7684

Mig 465 (2.573) −0.135 (0.026) 0.358 1445

Canada (CA) Nat 514 −7.163 0.487 −0.040 0.319 14555

Mig 521 (1.481) 0.527 (0.014) 0.281 4057

Denmark (DK) Nat 510 65.141 0.630 0.699 0.362 5224

Mig 445 (2.177) −0.069 (0.027) 0.170 1567

Finland (FI) Nat 529 64.021 0.281 0.303 0.349 5495

Mig 465 (6.391) −0.022 (0.054) 0.289 200

France (FR) Nat 512 52.708 −0.038 0.515 0.486 5089

Mig 459 (3.921) −0.553 (0.032) 0.290 706

Germany (DE) Nat 528 54.087 0.238 0.539 0.409 4614

Mig 474 (3.156) −0.301 (0.032) 0.213 881

Israel (IL) Nat 481 6.512 0.227 0.160 0.388 5223

Mig 475 (3.426) 0.067 (0.029) 0.318 1023

Italy (IT) Nat 502 51.145 −0.006 0.471 0.316 10199

Mig 451 (2.737) −0.477 (0.031) 0.196 867

Netherlands (NL) Nat 519 53.374 0.245 0.485 0.359 4587

Mig 466 (4.102) −0.240 (0.035) 0.225 504

New Zealand (NZ) Nat 512 −1.539 0.173 −0.046 0.417 3031

Mig 514 (3.403) 0.219 (0.026) 0.457 1075

Norway (NO) Nat 513 39.224 0.550 0.463 0.294 4535

Mig 474 (3.465) 0.087 (0.033) 0.202 616

Portugal (PT) Nat 487 1.512 −0.570 −0.158 0.468 6647

Mig 486 (4.325) −0.412 (0.055) 0.456 416

Singapore (SG) Nat 539 −31.006 −0.120 −0.499 0.450 4734

Mig 570 (2.913) 0.379 (0.027) 0.342 1164

Spain (ES) Nat 502 39.959 −0.371 0.572 0.380 5808

Mig 462 (3.268) −0.943 (0.044) 0.333 664

Sweden (SE) Nat 510 59.129 0.425 0.418 0.374 4311

Mig 451 (3.298) 0.007 (0.031) 0.192 819

Switzerland (CH) Nat 522 51.076 0.323 0.585 0.356 3907

Mig 471 (2.528) −0.262 (0.026) 0.357 1711

United Kingdom (GB) Nat 502 9.967 0.232 0.052 0.365 11329

Mig 492 (2.360) 0.181 (0.023) 0.334 1607

United States (US) Nat 496 21.527 0.280 0.755 0.366 4153

Mig 474 (2.849) −0.475 (0.034) 0.302 1215

PISA and ESCS: group-specic country averages; Mean di.: Dierences between the means of natives and immigrants; the values in brackets

are a variance measure: var(vI)

nI+var(vH)

nHwhere vrepresents the students’ PISA and ESCS values and ntheir respective numbers.

PISA Performance of Natives and Immigrants: Selection versus Eciency |19

tinctions between countries with and without restrictive

regimes can be made.

The index of economic, social and cultural status of

each student (ESCS) is regarded as input (x′). x′is inter-

nationally comparable, has a mean of zero and a stan-

dard deviation of one. Radial DEA models can only han-

dle strictly positive variables. Therefore, x′is transformed:

x′−min(x′) + 0.01 = x.xis the input used in our eciency

analysis and is not further transformed.

Table 2 provides descriptive results and correlation co-

ecients between the average PISA scores and the ESCS

values for each country at the student level, for students

with and without an immigration background. In most

countries, natives perform better and have a better average

socio–economic background. In Australia, Canada, and

New Zealand (all countries with selective immigration sys-

tems), immigrants achieve higher average PISA scores and

have higher ESCS endowments. On average, immigrants

in Singapore have ESCS values that are above the PISA

average, and the values of the natives are lower (Becker,

2012; Facchini & Lodigiani, 2014). In comparison, both

Canadian population groups have above-average ESCS av-

erages and the smallest gap. This hints for the selectivity

of the Canadian immigration system, so that the average

immigrant in Canada has a socio–economic background

similar to that of the average native. The United States has

the largest ESCS gap between the two groups. Although

the United States has a selective immigration system, it at-

tracts immigrants with relatively poorer socio–economic

backgrounds. However, the dierences in performance are

smaller in the United States than in Germany and Nor-

way, for example. Spanish immigrants have, on average,

the lowest ESCS values, and Portugal is the only country

in which the natives achieve higher PISA values despite

worse socio–economic backgrounds, although the gap is

not signicantly dierent from zero. Such specic chal-

lenges must be taken into account in an international ef-

ciency analysis of educational systems. Tables S2 to S4 in

the appendix provide descriptive results for the individual

PISA scores. All scores are greater than zero and students

with missing values are excluded from our analyses.

We use regressions to gauge the relationship between stu-

dents’ average educational performance and their socio–

economic endowments for each country separately. The

regressions include both a dummy for immigrant back-

ground and an interaction term. The results indicate that

performance gaps between immigrants and natives are de-

termined strongly by their respective ESCS endowments.

Increasing ESCS values have the highest positive impact

in France and lowest in Spain and Italy. The results indi-

cate a signicantly better performance of immigrants in

Australia, Canada and Singapore and a positive but in-

signicant relationship in Israel and the United States of

America. In all other countries, immigrants perform signi-

cantly worse than natives. All results can be provided upon

request.

5Eciency Results and Eciency

Decomposition

All results are obtained using R (version 3.6) and, unless

otherwise stated, the average PISA results are used as

output. The eciency scores indicate how relatively well

the students perform, given their socio–economic back-

grounds. First, the results are decomposed relative to na-

tional and then international frontiers, followed by a com-

parison of the performance of natives and immigrants, and

nally, the impact of the selection processes and the e-

ciency of educational systems are evaluated.

5.1 National frontiers

Table 3 provides country-specic arithmetic mean e-

ciency scores for all students, for the student groups rel-

Table 3:

Decomposition, national students and national frontiers,

average PISA scores as output

(1) (2) (3) (4) (5)

MHk(Hk)MIk(Ik)MEk(Hk)MEk(Ik)MEk(Ek)

AU 0.665 0.707 0.665 0.684 0.669

AT 0.694 0.673 0.694 0.649 0.686

BE 0.704 0.706 0.704 0.661 0.697

CA 0.681 0.702 0.680 0.688 0.682

DK 0.727 0.690 0.726 0.668 0.712

FI 0.724 0.701 0.724 0.652 0.721

FR 0.703 0.703 0.702 0.669 0.698

DE 0.716 0.708 0.716 0.669 0.708

IL 0.645 0.699 0.645 0.651 0.646

IT 0.707 0.714 0.707 0.657 0.703

NL 0.707 0.706 0.707 0.664 0.703

NZ 0.696 0.706 0.693 0.692 0.693

NO 0.707 0.726 0.707 0.682 0.704

PT 0.709 0.733 0.709 0.699 0.708

SG 0.699 0.753 0.699 0.711 0.701

ES 0.729 0.727 0.729 0.690 0.725

SE 0.684 0.700 0.684 0.636 0.676

CH 0.725 0.691 0.724 0.684 0.712

GB 0.693 0.701 0.693 0.683 0.692

US 0.671 0.721 0.670 0.691 0.675

Mean 0.699 0.708 0.699 0.674 0.696

20 |A. Behr and G. Fugger

ative to both national frontiers, and comparisons between

the groups. The column numbers are given above the for-

mal terms to simplify the interpretation.

The initial descriptive results showed that natives have

higher average PISA scores (see Table 2 and Figure 4), but

they disregard the socio–economic backgrounds of the stu-

dents, that are taken into account in the eciency analy-

sis. Column (1) and (2) of Table 3 contain the results of na-

tives and immigrants relative to their respective frontiers

for each country. Across all countries, both groups of stu-

dents are on average almost equally ecient (0.699 in col-

umn (1) to 0.708 in column (2)) if compared to their bench-

mark students from their group.

Columns 3 and 4 of Table 3 show the average eciency

scores when using the national frontier based on both

subsets of students. We observe that there are hardly any

changes among the natives, if immigrants are also taken

into account when calculating the ecient frontier. In con-

trast, the performance of immigrants decreases when na-

tives are taken into account as revealed by the comparison

of column (2) and (4).

Natives outperform immigrants on average by

(MEk(Hk)−MEk(Ik))·100 = 5.741% in Denmark, by

7.188% in Finland, and by 5.009% in Italy. Natives also

perform better in most countries, but the gaps are not as

large as in the previous countries and range from 0.100%

in New Zealand to 4.774% in Sweden. In all these coun-

tries, immigrants perform far worse, according to their ef-

ciency scores, taking into account their socio–economic

endowment. The educational systems do not succeed in

fostering both groups equally, which leads to inequalities

in educational performance beyond the dierences due to

their endowments.

In Australia, Canada, Israel, Singapore, and the

United States, immigrants perform on average better than

their native peer group, considering their eciency based

on ESCS endowments. In the United States, immigrants

perform best relative to the natives. The performance dif-

ference is 2.066%. In Israel both groups perform similarly,

immigrants being slightly better (0.613%).

Column 5 of Table 3 provides the mean eciency

scores for all students, based on their own frontiers for

each country. Israel achieves the lowest (0.646) and Spain

the highest (0.725) mean. Since the eciency frontiers are

country- and group-specic, they are rather a measure of

inequality than a means of comparing eciency between

countries. Table 3 does not provide any information on

which students form the eciency frontiers, and how ef-

cient the national educational systems are.

Figure S5 in the appendix displays the frontiers for

each student group within the countries and the interna-

tional frontier, calculated for all students. In several coun-

tries, the best–performing students are immigrants for low

ESCS values and natives for higher ESCS values (e.g. in

Austria, France, Germany, and the United States). In the

remaining countries, only natives constitute the eciency

frontier, as is the case in Finland, Portugal, Singapore,

Spain, and the United Kingdom. It is striking that the stu-

dents in Portugal, Singapore, and Spain have input-output

combinations that are on average far less distant from the

international eciency frontier than in the other countries.

Therefore, these countries are among the top performers in

our analysis.

5.2 International comparisons

Including all students, Figure S5 shows that the interna-

tional eciency frontier for low ESCS values consists of

three Spanish native speakers (one of whom has the low-

est ESCS value in the sample), followed by one Portuguese

and one Singaporean native speaker (with the highest av-

erage PISA value).⁹

Table 4 provides further within and between–country

comparisons. ME(Hk)is the average score of the native

students of country k,ME(Ik)is the average eciency of

its immigrant students, and ME(Ek)is the mean eciency

of all of students from country kwith respect to the inter-

national frontier of all students.

Columns 1 and 2 show how well each group performs

within each country, and allows within-country compar-

isons relative to the international frontier consisting of

all students. Compared to their native peer group, immi-

grants perform best in Australia (on average 1.977% bet-

ter), followed by the United States (1.558%), and Singa-

pore (1.415%). The countries where natives perform best

compared to immigrants are Finland (on average 6.646%

better), Sweden (5.476%), and Denmark (5.277%).

The results so far have been group-specic. Column 3,

on the other hand, provides a comparisons of the ecien-

cies of the national educational systems. The values result

from an international frontier and do not dierentiate be-

tween natives and immigrants within countries. The mean

ineciencies show how much the average PISA scores of a

country could be increased, if its educational system were

to enable students to perform similarly to the most e-

cient international students with comparable ESCS endow-

Using an output-oriented BBC-model with one input and one output,

and variable returns to scale, the student with the highest output value

must be ecient by construction.

PISA Performance of Natives and Immigrants: Selection versus Eciency |21

−0.04

0.00

0.04

0.08

Figure 5: Arithmetic mean eciency dierences between native and immigrant students in each country relative to the international frontier

Table 4:

Decomposition, national students and international fron-

tier, average PISA scores as output

(1) (2) (3)

ME(Hk)ME(Ik)ME(Ek)

AU 0.619 0.638 0.623

AT 0.639 0.590 0.630

BE 0.652 0.603 0.644

CA 0.637 0.645 0.639

DK 0.630 0.577 0.617

FI 0.660 0.594 0.658

FR 0.652 0.615 0.647

DE 0.667 0.624 0.660

IL 0.602 0.604 0.603

IT 0.644 0.603 0.641

NL 0.648 0.606 0.644

NZ 0.643 0.643 0.643

NO 0.631 0.602 0.627

PT 0.663 0.649 0.662

SG 0.696 0.710 0.698

ES 0.672 0.652 0.669

SE 0.632 0.578 0.624

CH 0.652 0.619 0.642

GB 0.632 0.622 0.630

US 0.623 0.638 0.626

Mean 0.645 0.621 0.641

ments. The average ineciencies over the entire sample

are 35.9%. The country with the highest mean eciency is

Singapore. In Finland, Germany, Portugal, and Spain, the

mean eciency scores are also relatively large. The highest

ineciencies exist in Israel and Denmark, given the ESCS

backgrounds of their respective students.

5.3 Dierences between immigrants and

natives

Figure 5 shows the dierences between the arithmetic

means of students with and without immigration back-

ground, relative to the countries’ frontiers, providing an

overview of the within-country dierences. By including

the ESCS as input, our analysis takes into account the

socio–economic endowment of the students. Selection ef-

fects that result in high or low ESCS scores should there-

fore not inuence the eciency scores, given the ESCS in-

put levels.

The eciency gaps between the groups are smallest in

Canada (−0.008), Israel (−0.006), New Zealand (0.001),

and Portugal (0.010). In the other countries, the dier-

ences are greater than 1%. In all European countries and

especially in Sweden (0.048), Denmark (0.057), and Fin-

land (0.072), the immigrant students perform on average

considerably worse than their native counterparts given

their ESCS backgrounds. Finland is often regarded as a

country with a superior educational system and integra-

tion success, but according to the eciency scores the ed-

ucational system in Finland is highly inecient in closing

the gap between natives and immigrants. Recent literature

conrms these performance decits of immigrants in Fin-

land, taking into account background factors such as gen-

der, grades, socio–economic background, home language

and age of arrival in Finland (Kirjavainen (2015); Yeas-

min & Uusiautti (2018)). However, these results have not

yet attracted much attention in recent literature. Arikan

et al. (2017), for example, claim that reducing the ESCS

gap would close the performance gap between natives and

immigrants in Finland, but our results indicate that espe-

cially an ecient use of the ESCS endowment is more im-

portant than the low ESCS levels (Arikan et al., 2017). We

argue that the sole use of PISA results in native immigrant

comparisons mainly reects selection eects due to dier-

ent immigration policies, rather than an analysis of the

educational systems. Given the social structure of immi-

grants (and natives) we evaluate the educational systems

according their ability to transform social endowments

into good PISA results.

22 |A. Behr and G. Fugger

●●●●●●●●●●●●●●●●●●●●

Selection effects (average PISA scores)

Educational performance (average efficiency scores)

Figure 6: Country rankings based on mean PISA scores of immigrants and of mean eciency

5.4 Selection eects and educational

eciency

In the upper line of Figure 6, the countries are arranged in

descending order according to their immigrants’ average

PISA scores. The order is solely based on the absolute per-

formance of immigrants in PISA. Here, the ecient coun-

tries are characterised by a strict immigration policy, se-

lecting immigrants who achieve the highest PISA levels. In

the lower line, the countries are ordered according to their

immigrants’ average eciency relative to the international

frontier (ME(Ik)). Thus, the countries are ranked accord-

ing to their students’ performance, given their ESCS en-

dowments. Therefore, the impact of selection procedures

is to a large extent controlled for, and the ranking reveals

how successfully educational systems use the ESCS en-

dowment.

The arrows indicate the rank changes. In both analy-

ses, students perform best in Singapore and worst in Den-

mark. The ranks of all other countries change due to taking

the ESCS endowment into account. Without regarding the

ESCS endowment (upper ranking), countries with strongly

selective immigration systems rank second to fth. Taking

into account the socio–economic backgrounds of their stu-

dents (lower ranking), their ranks deteriorate to four, ve,

six and nine. This indicates that simple PISA score compar-

isons examine rather immigration policy and less so the

eciency of educational systems.

Austria, Denmark and Sweden are the countries where

immigrants perform worst according to their average DEA

scores. Given the socio–economic background of their stu-

dents, these countries could achieve much higher PISA

scores, if they were to adapt their educational systems to

those of the ecient countries.

Without including the ESCS as input, immigrants

in Spain perform relatively poorly, but on average they

perform very well regarding their eciency. France (ve

ranks), Italy (four), and Portugal (three) are also coun-

tries which improve their rankings compared to the simple

PISA score comparison. Regarding their socio–economic

backgrounds, these three countries have relatively less

favourable immigrant compositions, but their educational

systems are relatively more ecient than in most other

countries. Our analysis shows that, despite a very large

educational gap (see Table 2), the French school system

performs well on average, because it at least partially com-

pensates for the large dierences in the socio–economic

background of immigrants. While most countries lose up

PISA Performance of Natives and Immigrants: Selection versus Eciency |23

Table 5: Correlation coecients between the eciency scores for the average PISA score and the three PISA scores as outputs

MHk(Hk)MIk(Ik)MEk(Hk)MEk(Ik)MEk(Ek)MHk(H)MIk(I)MEk(E)

AU 0.985 0.983 0.985 0.984 0.985 0.986 0.986 0.986

AT 0.981 0.978 0.981 0.980 0.981 0.987 0.985 0.987

BE 0.988 0.990 0.988 0.990 0.989 0.988 0.989 0.988

CA 0.983 0.984 0.984 0.983 0.984 0.986 0.986 0.986

DK 0.985 0.974 0.985 0.979 0.984 0.983 0.980 0.983

FI 0.981 0.961 0.981 0.972 0.981 0.986 0.979 0.986

FR 0.985 0.985 0.985 0.988 0.985 0.988 0.991 0.988

DE 0.984 0.981 0.984 0.981 0.983 0.987 0.985 0.986

IL 0.981 0.985 0.981 0.981 0.981 0.986 0.987 0.986

IT 0.982 0.972 0.982 0.980 0.982 0.983 0.979 0.983

NL 0.987 0.983 0.987 0.986 0.987 0.990 0.989 0.990

NZ 0.985 0.981 0.984 0.980 0.983 0.986 0.986 0.986

NO 0.984 0.974 0.984 0.979 0.983 0.984 0.978 0.983

PT 0.989 0.975 0.990 0.986 0.990 0.989 0.986 0.989

SG 0.983 0.983 0.985 0.978 0.984 0.988 0.981 0.986

ES 0.987 0.977 0.987 0.986 0.987 0.988 0.989 0.989

SE 0.975 0.979 0.974 0.978 0.976 0.983 0.980 0.983

CH 0.986 0.985 0.987 0.988 0.988 0.983 0.985 0.984

GB 0.986 0.978 0.986 0.982 0.985 0.987 0.985 0.987

US 0.982 0.986 0.982 0.982 0.982 0.989 0.989 0.989

to three ranks, Israel (nine), and the United Kingdom (four)

are far worse ranked, indicating relatively poorly perform-

ing educational systems. Immigrants in Denmark perform

worst both when their ESCS endowment is considered and

when it is not considered.

5.5 PISA scores as separate outputs

The DEA allows the inclusion of separate outputs that are

simultaneously included in the eciency assessment. In

this section, students are assessed on the basis of their abil-

ity to maximize the three PISA scores, given their ESCS end-

points. Model (2) allows for specialisation so that the e-

ciency of students focusing on a subset of the three abili-

ties is adequately taken into account. Tables S6 and S7 in

the appendix contain the decomposition of the eciency

results for national and international frontiers.

The eciency scores of the average PISA score and

three separated PISA scores as outputs are highly posi-

tively correlated. The Pearson correlation coecient be-

tween the DEAs are 0.984 for MH(H), 0.981 for MI(I), and

0.984 for ME(E). Table 5 provides the correlation coe-

cients of the eciency scores based on the aggregated out-

put and that of the three outputs for each country.

The inclusion of the separated PISA scores as outputs

allows the DEA model to weight the outputs separately and

thus to calculate overall higher eciency scores. The simi-

larity of the results to those of the previous analysis shows

that students who perform well on average also perform

quite well in the individual PISA subjects. These results

conrm that immigrants in countries with restrictive im-

migration regime perform relatively better than in other

countries and that immigrants in Spain, Portugal, and Sin-

gapore perform relatively best given their socio–economic

endowments.

6Conclusion

Our analysis focuses on the abilities of the national ed-

ucational systems to integrate immigrants, given their

socio–economic backgrounds. Country-specic means of

eciency scores based on national frontiers reveal that

in Denmark, Finland, and Sweden, native students per-

form substantially better than immigrants. In Australia,

Canada, Israel, Singapore, and the United States, immi-

grants are more ecient than their native peer group.

Relative to the international frontier consisting of all

students and compared to their native peer groups, immi-

24 |REFERENCES

grants in Australia, Singapore, and the United States per-

form relatively best. The opposite is true in Finland, Swe-

den, and Denmark.

Even if the dierences in the socio–economic endow-

ment of the students are taken into account, dierences be-

tween natives and immigrants persist. According to PISA

scores, as well as the eciency scores, in most coun-

tries with more selective immigration regimes, immigrants

perform on average similar or even better than natives.

The persistent dierences are somewhat surprising, as

the broad ESCS should capture the most relevant socio–

economic factors.

We nd that the Spanish educational system is rel-

atively best in increasing immigrants’ performance, and

Israel’s system is worst, given the respective socio–

economic backgrounds of their immigrants. Australia,

Canada, the United Kingdom, and New Zealand are coun-

tries with selective immigration policies, which attract im-

migrants who perform relatively better or almost as well

as their natives. If, however, the socio–economic back-

grounds are taken into account, the immigrants in these

countries perform on average worse than in Spain and Por-

tugal. The latter have low PISA values, but highly ecient

education systems.

The result that countries with relatively selective im-

migrant policies perform not only well in absolute PISA

scores, but are also quite ecient given their ESCS input

levels, is truly astonishing. This result implies that the se-

lection process not only aects ESCS levels, but also the

immigrant capacity to use their endowments eciently.

References

Ababneh, E., Al-Tweissi, A., & Abulibdeh, K. (2016). Timss and pisa

impact – the case of jordan. Research Papers in Education, 31, 5,

542-555.

Aduand, D. T., & Denkyirah, E. K. (2017). Education and economic

growth: a co-integration approach. International Journal of Educa-

tion Economics and Development, 8, 4, 228–249.

Agasisti, T., & Zoido, P. (2018). Comparing the eciency of schools

through international benchmarking: Results from an empirical

analysis of oecd pisa 2012 data. Educational Researcher, 47, 6,

352–362.

Ammermueller, A. (2007). Poor background or low returns? why im-

migrant students in germany perform so poorly in the programme

for international student assessment. Education Economics, 15, 2,

215-230.

Aparicio, J., Cordero, J. M., Gonzalez, M., & Lopez-Espin, J. J. (2017a).

Using non-radial dea to assess school eciency in a cross-country

perspective: An empirical analysis of oecd countries. Omega.

Aparicio, J., Crespo-Cebada, E., Pedraja-Chaparro, F., & Santín, D.

(2017b). Comparing school ownership performance using a pseudo-

panel database: A malmquist-type index approach. European Jour-

nal of Operational Research, 256, 2, 533-542.

Arikan, S., Van de Vijver, F. J., & Yagmur, K. (2017). Pisa mathematics

and reading performance dierences of mainstream european and

turkish immigrant students. Educational Assessment, Evaluation

and Accountability, 29, 3, 229–246.

Banker, R. D., Charnes, A., & Cooper, W. (1984). Some models for

estimating technical and scale ineciencies in data envelopment

analysis. Management Science, 30, 9, 1078-1092.

Becker, S. (2012). Brain drain and brain gain: The global competition

to attract high-skilled migrants. Oxford University Press.

Behr, A. (2015). Production and eciency analysis with r. Springer.

Beine, M., Boucher, A., Burgoon, B., Crock, M., Gest, J., Hiscox,

M., & Thielemann, E. (2016). Comparing immigration policies: An

overview from the impala database. International Migration Review,

50, 4, 827–863.

Bertoli, S., Dequiedt, V., & Zenou, Y. (2016). Can selective immigra-

tion policies reduce migrants’ quality? Journal of Development Eco-

nomics, 119, 100–109.

Bjerre, L., Helbling, M., Römer, F., & Zobel, M. (2015). Conceptualizing

and measuring immigration policies: A comparative perspective.

International Migration Review, 49, 3, 555–600.

Bonjour, S., & Chauvin, S. (2018). Social class, migration policy and

migrant strategies: an introduction. International Migration, 56, 4,

5–18.

Borgna, C., & Contini, D. (2014). Migrant achievement penalties in

western europe: Do educational systems matter? European Socio-

logical Review, 30, 5, 670–683.

Camarota, S. A., & Zeigler, K. (2016). Immigrants in the united states:

A prole of the foreign-born using 2014 and 2015 census bureau

data. Center for Immigration Studies.

Camehl, G. F., Schober, P. S., & Spiess, C. K. (2018). Information asym-

metries between parents and educators in german childcare insti-

tutions. Education Economics, 26, 6, 624–646.

Chen, Y. (2005). Measuring super-eciency in dea in the presence of

infeasibility. European Journal of Operational Research, 161, 2, 545

- 551. Financial Modelling

Cook, D. R. (1979). Influential observations in linear regression. Journal

of the American Statistical Association, 74, 365, 169–174.

Cooper, W. W., Seiford, L. M., & Tone, K. (2007). Data envelopment

analysis - a comprehensive text with models, applications, refer-

ences and dea-solver software. Springer.

Cordero-Ferrera, J. M., Santín, D., & Simancas, R. (2017). Assessing

european primary school performance through a conditional non-

parametric model. Journal of the Operational Research Society, 68,

4, 364–376.

Damiani, V. (2016). Large-scale assessments and educational policies

in italy. Research Papers in Education, 31, 5, 529-541.

De Witte, K., & Lopez-Torres, L. (2017) Eciency in education: a review

of literature and a way forward. Journal of the Operational Research

Society, 68, 4, 339–363.

Dronkers, J., Levels, M., & de Heus, M. (2014). Migrant pupils’ scientic

performance: the influence of educational system features of origin

and destination countries. Large-scale Assessments in Education,

2, 1, 3.

Dustmann, C., Frattini, T., & Lanzara, G. (2012). Educational achieve-

ment of second-generation immigrants: an international compari-

son. Economic Policy, 27, 69, 143–185.

Entorf, H., & Minoiu, N. (2005) What a dierence immigration pol-

icy makes: A comparison of pisa scores in europe and traditional

REFERENCES |25

countries of immigration. German Economic Review, 6, 3, 355–376.

Facchini, G., & Lodigiani, E. (2014). Attracting skilled immigrants: An

overview of recent policy developments in advanced countries.

National Institute Economic Review, 229, 1, R3–R21.

Falck, O., Mang, C., & Woessmann, L. (2018). Virtually no eect? dif-

ferent uses of classroom computers and their eect on student

achievement. Oxford Bulletin of Economics and Statistics, 80, 1,

1-38.

Harris, R., Courtney, L., Ul-Abadin, Z., & Burn,K. (2019). Student access

to the curriculum in an age of performativity and accountability: an

examination of policy enactment. Research Papers in Education,

1–21.

Ho, E. S. C. (2016). The use of large-scale assessment (pisa): insights

for policy and practice in the case of hong kong. Research Papers

in Education, 31, 5, 516-528.

Hochschild, J. L., & Cropper, P. (2010). Immigration regimes and school-

ing regimes: Which countries promote successful immigrant incor-

poration? School Field, 8, 1, 21–61.

Hwang, J., Choi, K. M., Bae, Y., & Shin, D. H. (2018). Do teachers’ instruc-

tional practices moderate equity in mathematical and scientic

literacy?: an investigation of the pisa 2012 and 2015. International

Journal of Science and Mathematics Education, 16, 1, 25–45.

Isphording, I. E., Piopiunik, M., & Rodriguez-Planas, N. (2016). Speak-

ing in numbers: The eect of reading performance on math perfor-

mance among immigrants. Economics Letters, 139, 52–56.

Jerrim, J. (2015). Why do east asian children perform so well in pisa?

an investigation of western-born children of east asian descent.

Oxford Review of Education, 41, 3, 310–333.

Karatheodoros, A. (2017). The contribution of secondary education on

regional economic growth in greece, over the period 1995-2012s.

International Journal of Education Economics and Development, 8,

1, 46–64.

Kirjavainen, T. (2015). Immigrant students and the eciency of basic

education. National Audit Oce of Finland.

Kunz, J. S. (2016). Analyzing educational achievement dierences

between second-generation immigrants: Comparing germany and

german-speaking switzerland. German Economic Review, 17, 1, 61–

91.

Levels, M., Dronkers, J., & Kraaykamp, G. (2008). Immigrant children’s

educational achievement in western countries: Origin, destination,

and community eects on mathematical performance. American

Sociological Review, 73, 5, 835-853.

OECD. (2014). Pisa 2012 technical report.

OECD. (2016). Pisa 2015 results (volume ii): Policies and practices for

successful schools. Paris: OECD Publishing.

OECD. (2017). Pisa 2015 technical report. Tech. rep. OECD Publishing.

Parr, A. K., & Bonitz, V. S. (2015). Role of family background, student

behaviors, and school-related beliefs in predicting high school

dropout. The Journal of Educational Research, 108, 6, 504-514.

Rangvid, B. S. (2007). Sources of immigrants’ underachievement: Re-

sults from pisa-copenhagen. Education Economics, 15, 3, 293–326.

Reparaz, C., & Sotés-Elizalde, M. A. (2019). Parental involvement in

schools in spain and germany: Evidence from pisa 2015. Interna-

tional Journal of Educational Research, 93, 33 - 52.

Rindermann, H., & Thompson, J. (2016). The cognitive competences

of immigrant and native students across the world: An analysis of

gaps, possible causes and impact. Journal of biosocial science, 48,

1, 66–93.

Rogiers, A., Keer, H. V., & Merchie, E. (2020). The prole of the skilled

reader: An investigation into the role of reading enjoyment and stu-

dent characteristics. International Journal of Educational Research,

99, 101512.

Schneebaum, A., Rumplmaier, B., & Altzinger, W. (2016). Gender and

migration background in intergenerational educational mobility.

Education Economics, 24, 3, 239-260.

Schneeweis, N. (2011). Educational institutions and the integration of

migrants. Journal of Population Economics, 24, 4, 1281–1308.

Schnepf, S. V. (2007) Immigrants’ educational disadvantage: an exami-

nation across ten countries and three surveys. Journalof population

economics, 20, 3, 527–545.

Silverman, B. W. (1986). Density estimation for statistics and data

analysis. London [u.a.]: Chapman and Hall.

Sutherland, D., Price, R., & Gonand, F. (2009). Improving public spend-

ing eciency in primary and secondary education. OECD Journal:

Economic Studies, 2009, 1, 1–30.

Tobin, M., Nugroho, D., & Lietz, P. (2016). Large-scale assessments

of students’ learning and education policy: synthesising evidence

across world regions. Research Papers in Education, 31, 5, 578-594.

Tong, G. C. (2001). National day speech. National Archives of Singa-

pore

Woessmann, L. (2016). The importance of school systems: Evidence

from international dierences in student achievement. Journal of

Economic Perspectives, 30, 3, 3-32.

Yeasmin, N., & Uusiautti, S. (2018). Finland and singapore, two dif-

ferent top countries of pisa and the challenge of providing equal

opportunities to immigrant students. Journal for Critical Education

Policy Studies (JCEPS), 16, 1.

26 |REFERENCES

Supplementary results

————————————————————————————-

Figure S1:

Country-level average PISA scores relative to the average ESCS scores and mean absolute deviations from the median ESCS

scores; left: average scores, right: average PISA scores and mean absolute deviations from the median ESCS scores

Table S1: Excluded outliers

Amount Percentage

Australia 151 1.079

Austria 67 0.966

Belgium 77 0.815

Canada 207 1.066

Denmark 67 0.959

Finland 67 1.153

France 52 0.875

Germany 57 1.012

Israel 72 1.108

Italy 110 0.971

Netherlands 59 1.108

New Zealand 49 1.131

Norway 56 1.059

Portugal 65 0.900

Singapore 60 0.985

Spain 53 0.794

Sweden 63 1.186

Switzerland 44 0.759

United Kingdom 115 0.851

United States of America 51 0.905

REFERENCES |27

Table S2: Descriptive results of the PISA mathematics scores

Countries Group Min Max Median Mean Sd. n

AU Nat 166.883 800.542 484.679 483.382 91.113 10744

Mig 225.436 757.799 505.180 504.928 91.018 2651

AT Nat 193.875 797.841 516.602 512.318 88.728 5533

Mig 170.358 731.659 446.329 448.960 84.998 1242

BE Nat 197.540 818.559 529.075 523.481 90.932 7684

Mig 237.039 727.384 468.027 468.323 90.944 1445

CA Nat 218.370 807.652 504.818 504.624 81.876 14555

Mig 261.416 810.875 516.563 517.336 83.974 4057

DK Nat 266.826 751.205 518.059 516.057 76.720 5224

Mig 213.486 700.433 455.982 457.803 74.074 1567

FI Nat 245.949 751.693 516.953 515.419 78.857 5495

Mig 208.933 695.899 457.889 466.733 88.732 200

FR Nat 197.621 765.411 515.105 507.942 91.081 5089

Mig 200.976 723.209 455.736 456.625 96.992 706

DE Nat 227.368 803.717 524.496 522.885 85.158 4614

Mig 218.142 715.613 475.107 475.467 84.294 881

IL Nat 147.251 776.097 479.573 476.417 98.822 5223

Mig 122.084 728.409 476.589 471.336 103.668 1023

IT Nat 171.668 792.029 505.564 503.999 87.043 10199

Mig 224.551 674.787 459.160 458.088 82.293 867

NL Nat 203.188 783.104 528.884 523.470 87.652 4587

Mig 229.374 689.471 473.280 470.753 87.335 504

NZ Nat 248.467 768.730 497.402 497.696 86.419 3031

Mig 249.591 766.447 511.344 508.374 94.368 1075

NO Nat 240.418 748.189 507.671 507.450 82.138 4535

Mig 264.560 702.677 469.598 471.186 77.707 616

PT Nat 157.556 783.224 485.353 483.819 95.255 6647

Mig 214.974 702.556 470.405 474.209 93.780 416

SG Nat 242.225 847.230 555.887 552.496 93.070 4734

Mig 293.152 842.615 591.471 584.756 88.716 1164

ES Nat 244.435 752.071 499.491 496.916 79.837 5808

Mig 226.047 690.859 454.489 455.834 79.728 664

SE Nat 206.226 771.176 507.830 507.146 82.513 4311

Mig 221.587 712.981 451.156 452.016 84.530 819

CH Nat 254.399 800.594 539.532 536.024 87.237 3907

Mig 183.867 779.058 486.424 488.177 90.664 1711

GB Nat 185.785 769.712 495.911 494.442 84.650 11329

Mig 205.712 729.779 492.467 488.871 91.280 1607

US Nat 204.268 766.785 477.317 477.480 85.992 4153

Mig 227.198 720.594 458.297 457.413 84.894 1215

28 |REFERENCES

Table S3: Descriptive results of the PISA science scores

Countries Group Min Max Median Mean Sd. n

AU Nat 191.045 833.478 504.560 501.608 100.573 10744

Mig 227.272 803.415 519.730 515.728 100.702 2651

AT Nat 227.032 826.725 513.994 511.576 91.553 5533

Mig 204.623 741.154 443.153 447.522 86.996 1242

BE Nat 231.127 813.512 525.989 518.910 94.424 7684

Mig 210.139 711.469 459.965 461.836 96.402 1445

CA Nat 213.671 821.825 521.184 519.625 88.046 14555

Mig 250.632 828.142 526.060 523.003 91.824 4057

DK Nat 202.866 758.113 507.808 507.843 86.583 5224

Mig 219.308 731.502 430.812 432.864 85.046 1567

FI Nat 232.233 852.902 540.267 537.439 92.260 5495

Mig 239.698 735.533 462.255 462.091 96.635 200

FR Nat 226.574 784.065 517.705 511.625 95.790 5089

Mig 189.502 746.084 451.341 453.775 100.996 706

DE Nat 253.093 810.494 536.333 531.437 92.732 4614

Mig 188.335 781.910 464.048 467.684 92.284 881

IL Nat 181.542 825.603 476.171 476.777 102.558 5223

Mig 116.428 742.017 471.508 470.807 105.217 1023

IT Nat 209.030 772.320 502.760 498.819 86.628 10199

Mig 219.710 711.228 451.098 449.931 82.727 867

NL Nat 233.621 786.983 525.554 520.158 96.996 4587

Mig 225.921 685.571 468.732 461.069 93.907 504

NZ Nat 245.909 795.682 525.200 521.620 97.949 3031

Mig 185.943 800.925 521.780 518.146 108.474 1075

NO Nat 204.571 819.329 509.873 508.423 92.275 4535

Mig 247.333 699.818 458.451 460.458 87.591 616

PT Nat 195.810 781.855 491.015 490.120 91.883 6647

Mig 252.490 726.540 482.707 488.197 87.464 416

SG Nat 248.752 870.020 549.887 542.634 101.790 4734

Mig 265.541 835.631 580.107 574.010 97.278 1164

ES Nat 230.400 740.724 506.647 503.965 82.564 5808

Mig 226.049 724.153 457.635 460.772 85.888 664

SE Nat 166.987 845.611 511.242 508.842 95.297 4311

Mig 165.070 739.310 437.695 440.494 96.810 819

CH Nat 242.234 771.501 525.921 522.781 90.129 3907

Mig 219.737 750.943 459.952 465.410 94.428 1711

GB Nat 214.389 807.431 511.242 509.452 93.702 11329

Mig 271.842 792.647 495.884 497.483 97.426 1607

US Nat 238.079 806.788 504.314 505.552 95.449 4153

Mig 239.927 737.466 475.062 477.650 91.942 1215

REFERENCES |29

Table S4: Descriptive results of the PISA reading scores

Countries Group Min Max Median Mean Sd. n

AU Nat 96.374 832.034 497.103 491.976 101.709 10744

Mig 169.719 800.578 519.464 512.925 102.203 2651

AT Nat 131.917 762.012 506.123 499.912 94.060 5533

Mig 183.376 712.661 447.364 447.287 94.479 1242

BE Nat 205.263 807.437 524.693 515.790 93.568 7684

Mig 184.550 720.464 468.211 465.703 96.005 1445

CA Nat 223.017 798.814 520.883 517.968 87.553 14555

Mig 219.633 812.414 526.823 523.367 91.836 4057

DK Nat 213.575 781.762 509.405 505.788 83.225 5224

Mig 192.824 713.168 440.661 443.598 83.058 1567

FI Nat 200.763 778.198 538.810 533.374 86.657 5495

Mig 194.514 716.262 474.596 465.347 99.653 200

FR Nat 169.224 850.750 524.294 515.017 103.111 5089

Mig 181.602 753.531 474.086 466.058 110.194 706

DE Nat 201.098 808.879 538.702 530.944 92.075 4614

Mig 171.798 749.074 481.730 479.852 97.112 881

IL Nat 123.358 861.854 496.734 490.971 109.097 5223

Mig 161.619 781.152 491.460 482.487 107.601 1023

IT Nat 218.244 766.463 507.069 502.664 85.794 10199

Mig 185.012 676.707 448.731 444.028 86.843 867

NL Nat 151.081 778.437 521.125 513.710 96.334 4587

Mig 176.822 741.653 470.805 465.394 93.320 504

NZ Nat 169.027 810.103 520.348 516.821 99.665 3031

Mig 231.686 811.821 520.667 514.233 107.019 1075

NO Nat 183.062 807.994 528.604 523.727 94.026 4535

Mig 239.430 777.681 494.664 490.283 93.814 616

PT Nat 159.732 773.769 492.554 488.197 91.653 6647

Mig 216.284 723.842 500.284 495.196 91.901 416

SG Nat 230.626 818.352 529.974 522.078 97.500 4734

Mig 162.940 782.729 560.674 551.459 92.967 1164

ES Nat 161.767 767.832 511.244 505.802 81.443 5808

Mig 162.794 708.353 475.377 470.201 90.464 664

SE Nat 181.227 826.607 520.679 515.129 94.084 4311

Mig 127.785 756.546 465.677 461.219 99.536 819

CH Nat 204.781 771.608 510.955 506.492 90.407 3907

Mig 192.295 747.702 456.481 458.481 94.613 1711

GB Nat 213.622 846.678 503.169 501.540 89.369 11329

Mig 186.130 794.240 487.340 489.179 94.946 1607

US Nat 198.408 772.617 508.019 503.818 95.449 4153

Mig 181.668 742.142 491.332 487.206 98.431 1215

30 |REFERENCES

Table S5: PISA scores correlation coecients

Mathematics–Reading Mathematics–Science Reading–Science

AU 0.789 0.879 0.872

AT 0.794 0.886 0.864

BE 0.834 0.891 0.897

CA 0.766 0.878 0.865

DK 0.769 0.874 0.863

FI 0.783 0.863 0.861

FR 0.828 0.899 0.892

DE 0.796 0.885 0.856

IL 0.823 0.887 0.892

IT 0.743 0.849 0.829

NL 0.860 0.899 0.891

NZ 0.772 0.884 0.866

NO 0.778 0.885 0.836

PT 0.806 0.889 0.862

SG 0.829 0.890 0.908

ES 0.756 0.888 0.847

SE 0.756 0.881 0.828

CH 0.801 0.882 0.871

GB 0.783 0.879 0.869

US 0.826 0.890 0.889

Table S6: Decomposition, national students and national frontiers, three PISA scores as outputs

(1) (2) (3) (4) (5) (6) (7)

MHk(Hk)MIk(Ik)MEk(Hk)MEk(Ik)MEk(Ek)MIk(Ik)

MIk(Hk)·MHk(Ik)

MHk(Hk)1

2MEk(Hk)

MEk(Ik)

AU 0.681 0.725 0.680 0.701 0.684 0.970 0.970

AT 0.715 0.704 0.715 0.668 0.706 1.074 1.071

BE 0.718 0.721 0.718 0.672 0.710 1.076 1.068

CA 0.700 0.718 0.698 0.708 0.700 0.986 0.986

DK 0.744 0.718 0.742 0.688 0.730 1.081 1.079

FI 0.744 0.739 0.744 0.676 0.741 1.101 1.101

FR 0.716 0.717 0.716 0.679 0.711 1.059 1.055

DE 0.731 0.725 0.730 0.686 0.723 1.069 1.065

IL 0.670 0.714 0.669 0.675 0.670 0.992 0.991

IT 0.722 0.736 0.722 0.671 0.718 1.080 1.075

NL 0.726 0.734 0.725 0.683 0.721 1.072 1.062

NZ 0.714 0.730 0.711 0.713 0.712 0.999 0.997

NO 0.724 0.753 0.724 0.703 0.721 1.034 1.029

PT 0.717 0.761 0.717 0.713 0.716 1.009 1.005

SG 0.717 0.770 0.716 0.736 0.720 0.977 0.974

ES 0.742 0.751 0.742 0.703 0.738 1.053 1.054

SE 0.710 0.719 0.709 0.657 0.700 1.080 1.078

CH 0.738 0.710 0.735 0.696 0.723 1.049 1.057

GB 0.709 0.720 0.708 0.701 0.707 1.010 1.010

US 0.691 0.734 0.690 0.708 0.694 0.978 0.975

Mean 0.716 0.730 0.716 0.692 0.712 1.037 1.035

REFERENCES |31

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Australia

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Austria

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Belgium

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Canada

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Denmark

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Finland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

France

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Germany

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Israel

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Italy

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Netherlands

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

New Zealand

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Norway

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Portugal

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Singapore

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Spain

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Sweden

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Switzerland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United Kingdom

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United States of America

Figure S2: Mathematics scores distributions among natives (straight line) and immigrants (dashed line)

32 |REFERENCES

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Australia

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Austria

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Belgium

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Canada

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Denmark

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Finland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

France

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Germany

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Israel

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Italy

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Netherlands

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

New Zealand

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Norway

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Portugal

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Singapore

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Spain

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Sweden

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Switzerland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United Kingdom

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United States of America

Figure S3: Science scores distributions among natives (straight line) and immigrants (dashed line)

REFERENCES |33

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Australia

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Austria

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Belgium

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Canada

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Denmark

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Finland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

France

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Germany

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Israel

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Italy

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Netherlands

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

New Zealand

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Norway

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Portugal

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Singapore

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Spain

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Sweden

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

Switzerland

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United Kingdom

200 300 400 500 600 700 800

0.000 0.002 0.004 0.006

United States of America

Figure S4: Read scores distributions among natives (straight line) and immigrants (dashed line)

34 |REFERENCES

Figure S5: Eciency frontiers

REFERENCES |35

36 |REFERENCES

Table S7: Decomposition, national students and international frontier, three PISA scores as outputs

(1) (2) (3) (4) (5) (6)

ME(Hk)ME(Ik)ME(Ek)ME(Hk)

ME(E)

ME(Ik)

ME(E)

ME(Ek)

ME(E)

AU 0.635 0.656 0.639 0.968 1.000 0.974

AT 0.655 0.606 0.646 0.999 0.923 0.984

BE 0.668 0.617 0.659 1.018 0.941 1.005

CA 0.654 0.664 0.656 0.997 1.012 1.000

DK 0.650 0.597 0.637 0.991 0.910 0.972

FI 0.677 0.612 0.675 1.033 0.933 1.029

FR 0.667 0.625 0.662 1.017 0.952 1.009

DE 0.684 0.639 0.676 1.042 0.974 1.031

IL 0.620 0.621 0.620 0.946 0.946 0.946

IT 0.662 0.620 0.659 1.010 0.944 1.004

NL 0.663 0.619 0.659 1.011 0.943 1.004

NZ 0.661 0.660 0.661 1.007 1.007 1.007

NO 0.652 0.627 0.649 0.994 0.955 0.989

PT 0.674 0.666 0.673 1.027 1.016 1.026

SG 0.710 0.731 0.714 1.082 1.114 1.088

ES 0.684 0.662 0.682 1.043 1.010 1.039

SE 0.654 0.601 0.645 0.997 0.916 0.983

CH 0.672 0.635 0.660 1.024 0.968 1.007

GB 0.646 0.638 0.645 0.985 0.973 0.984

US 0.637 0.650 0.640 0.970 0.991 0.975

Mean 0.661 0.637 0.658 1.008 0.971 1.003

Available via license: CC BY

Content may be subject to copyright.

Die ökonomische Kompetenz von Schüler*innen der Sekundarstufe I: Kompetenzmodellierung, Assessmentkonstruktion und Validierung

Book

Jan 2024

Fabio Fortunati

Dieses Buch thematisiert die Erhebung der ökonomischen Kompetenz von Schüler*innen der Sekundarstufe I und verfolgt dabei drei Forschungsziele: (1) die Konzeption und Modellierung der ökonomischen Domäne, (2) die modellbasierte Entwicklung eines Testinstruments zur Messung ökonomischer Kompetenz und (3), wie unterschiedliche Evidenzquellen, bezogen auf die Curriculum-Instruction-Assessment-Triad, die Validität der Testwertinterpretation in der ökonomischen Domäne sicherstellen können. Das entwickelte Testinstrument TBA-EL wurde anhand des Evidence-centered-Designs entwickelt und stellt einen psychologischen Leistungstest dar, der über ein möglichst authentisches, computergestütztes Design das Konstrukt der ökonomischen Kompetenz operationalisiert. Die Befunde der Analyse weisen auf ein zweidimensionales Modell ökonomischer Kompetenz, mit einer Differenzierung in einer sprachlich-argumentativen und einen mathematisch analytischen Zugang zu ökonomischen Inhalten hin. Darüber hinaus lässt sich über schwierigkeitsgenerierende Merkmale ein fachdidaktisch interpretierbares Kompetenzniveaumodell entwickeln. Ebenfalls lassen sich schulformspezifische Unterschiede in der Instruktionssensitivität des Testinstruments feststellen. Die Befunde weisen eine hohe Anschlussfähigkeit sowohl an die wirtschaftspädagogische als auch die ökonomisch allgemeinbildende Forschung auf und adressieren ein zentrales Desiderat.

Competence and affect self-concepts of elementary school children – Do gender and immigration background play a role?

Article

Full-text available

Nov 2023

Annette Lohbeck

Childhood, Migration and the Pursuit of Happiness in Middle‐Class East Asia

Article

Full-text available

Sep 2024

This special issue explores the shifting landscape of global middle‐class migration within and from East Asia by taking the relationship between mobility, parenting ideals and changing educational desires as its focus. Contributions explore how East Asian middle‐class families balance the emergent emphasis on their children's well‐being with the demands of global competitiveness as these often‐antagonistic desires are projected onto old and new migration destinations against the background of global geopolitical and economic power shifts. Instead of reifying the simplistic binary of hierarchical, achievement‐oriented East Asian ‘Confucian educational norms’ versus democratic, well‐being‐focussed Western ideals, the contributions offer a nuanced understanding of how these educational ideals coexist within the social, cultural, political and economic contexts of families. By carefully assessing the dialectically entwined intimate experiences of parents and children, the articles collected here set out to broaden our understanding of how middle‐class families in Singapore, South Korea, Japan and China attempt to negotiate the tension between prioritizing children's happiness and maintaining global competitiveness result in a variety of strategies from migrating to less obvious international destinations to crafting domestic alternatives. Taken together, the articles reveal consistent patterns of middle‐class migration and child‐rearing decisions that contest and reshape conventional notions of success, attesting to a shift in global middle‐class migration trends and to the importance of child‐rearing in migratory decisions.

Unlocking educational pathsOtkrivanje obrazovnih puteva: Influential factors of Bosnia and Herzegovina's students performanceutjecajni čimbenici uspješnosti bosanskohercegovačkih studenata

Article

Jun 2024

This paper explores the influence of socio-economic status on the education performance of male and female students in Bosnia and Herzegovina, utilizing data from the 2018 Programme for International Student Assessment (PISA). The study defines key terms such as education, socio-economic status, and international student assessments, emphasizing the significance of a quality educational system for societal development. By employing three models for science literacy, mathematics, and reading, the results indicate that parents' education and possession of cultural goods significantly impact student achievements. Students with higher-educated parents and who possess more books at home perform better in PISA tests. The gender variable however is found not to be statistically significant. The study emphasizes the significance of a "quiet study space", suggesting that students with such an environment achieve higher science test scores. While causality is not established, the analysis provides insights into the initial impact of socio-economic status on student achievements. The paper advocates for continued participation in international assessments project like PISA to monitor educational achievements and inform evidence-based policy making. The findings underscore practical implications for educational policymakers, and recommendations to enhance Bosnia and Herzegovina's educational system.

The Labeling and Positioning of Refugee Students and Their Refusal to be (Mis)Positioned

Article

Full-text available

Dec 2023

The Association Between Educational Systems and Reading Literacy Performance in PISA Across the Asia-Pacific Region and Countries

Chapter

Nov 2023

The weak position of reception education for newly arrived migrant students in the educational field

Article

Dec 2022

Despite the group of Newly Arrived Migrant Students (NAMS) being very diverse, their educational outcomes are homogeneous and rather weak in many countries. In Flanders (Belgium) the educational trajectory of NAMS starts in a separate program that prepares them for their transition to regular secondary education. Rather than seeing the outcomes of reception education as the result of the behaviour of individual teachers and students, we argue that this problem should be addressed using a field approach that directs attention to the structural relationships between reception education and regular education. We rely on Bourdieu’s agency-structure bridging concept of social fields to reveal this relative position and the implicit assumptions or doxa that sustain it. Administrative data on teachers and students in secondary education, as well as policy texts, the transcripts of 37 interviews with teachers and the fieldnotes from the observations of 28 admission councils were analysed.

New to the game. A system analysis of the orientation process of newly arrived migrant students in Flemish secondary education

Thesis

Jan 2022

Laura Emery

Cultural Capital in Indigenous Population in Mexico and Its Relation to Standardized Tests

Chapter

Apr 2025

Standardized tests such as PISA have shown that one of the most important variables in interpreting results is cultural capital. In this chapter, we problematize the idea of cultural capital in contexts such as Indigenous populations in Mexico. In these cases, community knowledge, traditional practices, and language use are often excluded from the formal education system, creating fractures in the intergenerational transmission of cultural capital. We argue that international and national standardized tests should take these tensions into account in order to avoid contributing to neocolonial structures that foster such fractures.

Cooling Out and Warming Up – Professional Strategies in the Education of Newly Arrived Migrant Students in Sweden

Article

Sep 2024

This study explores how school professionals manage the challenges of educating newly arrived migrant students (NAMS), with a focus on the transition from the Language Introduction Program (LIP) to a national upper secondary school program or alternative forms of education. We draw on the theoretical framework of Inhabited Institutionalism to understand how school professionals’ interpretations and sense-making of external policy pressures and internal challenges of teaching a diverse group of students are shaped by social interaction. The study reveals that professionals are critical of the educational system and its consequences for NAMS’ education, due to its rigid admission requirements and unrealistic timeframes. To mitigate the potential effects on students’ educational trajectories, the professionals employ both cooling-out and warming-up strategies. The study emphasizes the significance of social interaction among school professionals in interpreting the educational system and its consequences and suggests that the outcome of negotiations among professionals regarding different strategies is likely to have a significant impact on the future trajectories of NAMS.

Student access to the curriculum in an age of performativity and accountability: an examination of policy enactment

Article

Full-text available

Jan 2019

The curriculum is often the target of reform and governments use a range of accountability measures to ensure compliance. This paper examines the decisions schools in England make regarding history provision, in a period of curriculum change, and the potential consequences of these decisions. Drawing on a large, longitudinal data set, of primary and secondary material, the study examines the relationship between the number of students entered for public examination in history in England and a range of situated and material factors. The data suggest that particular measures of accountability are effective in shaping school decision-making, but the type of school, socio-economic nature of the school intake, and students’ prior attainment are also important factors in understanding the decisions made. This does result in an inequitable access to history education; this inequity exists between different types of schools and socio-economic areas, and is also evident within schools where students with low prior attainment are less likely to be allowed to study history.

Social Class, Migration Policy and Migrant Strategies: An Introduction

Article

Full-text available

Jun 2018

The introduction to this special issue traces class at the interface between migration policy and migrant strategies. Scholarship on the politics of migration and citizenship has thus far largely neglected class. In contrast, we contend that discourses on migration, integration and citizenship are inevitably classed. Assessed through seemingly heterogeneous criteria of “merit” and “performance”, class serves as an analytical connector between economic and identity rationales which intersect in all migration policies, including those regulating family and humanitarian admission. Class-selective policy frames can function as constraints maintaining some aspiring migrants into immobility or channeling different groups of migrants into separate and unequal incorporation routes. Yet, policy frames can also serve as resources to strategize with as migrants navigate and perform gendered and classed expectations embedded into receiving-country migration regimes. We conclude that connecting policy with migrant strategies is key to reintroducing class without naturalizing classed strategies of mobility. Special issue on Migration & Social class with Jennifer Elrick, Elke Winter, Eleonore Kofman, Paola Bonizzoni, Conny Roggeband, Marleen van der Haar, Sebastian Prothmann, Maria Villares-Varela and Marieke Slootman.

Information asymmetries between parents and educators in German childcare institutions

Article

Full-text available

Apr 2018

Economic theory predicts market failure in the market for early childhood education and care (ECEC) due to information asymmetries. We empirically investigate information asymmetries between parents and ECEC professionals in Germany, making use of a unique extension of the Socio-Economic Panel Study (SEOP). We compare quality perceptions by parents and by professionals across 734 institutions. We detect considerable information asymmetries that differ across quality measures but less so by parental socio-economic background or center characteristics. Both similarly contribute to explaining variations in the information gap. We conclude that information is not readily available to parents; an issue that should be addressed by policy-makers.

Education and economic growth: a co-integration approach

Article

Full-text available

Jan 2017

In the wake of numerous educational policies and huge educational investments in Ghana, it is imperative to investigate the effect of education on economic growth. Using the autoregressive distributed lag approach to co-integration we find that in the short-run primary/basic education has a significant positive impact on economic growth. In the long-run, both primary and secondary educations have a positive impact on economic growth. Tertiary education has no significant effect on economic growth both in the short-run and long-run. Furthermore, government expenditure on education has no statistically significant effect on economic growth which could imply that the expenditure is not directed towards the quality of education but rather quantity(enrolment). We concluded that educational policies should focus not only on the basic and secondary education but also the tertiary education, incorporating research and development into the curricula, and government expenditure should be directed towards the quality of education.

The profile of the skilled reader: An investigation into the role of reading enjoyment and student characteristics

Article

Jan 2019
Int J Educ Res

This study focuses on the relationship between 15-year-olds’ reading enjoyment and reading comprehension on the one hand and the relationship between these two variables and students’ gender, socio-economic status (SES), and language background on the other. Based on the PISA 2009 literacy test, it includes a secondary analysis of data collected from 4269 Flemish students of which the results were analysed by means of multilevel modelling. The results showed a positive relationship between reading enjoyment and reading comprehension. Girls, higher SES students, and students in the academic track were found to score higher in both reading enjoyment and reading comprehension. Students in the vocational track scored lowest in both reading enjoyment and reading comprehension. Finally, second or additional language (L2) speakers scored lower in reading enjoyment.

Parental involvement in schools in Spain and Germany: Evidence from PISA 2015

Article

Jan 2019
Int J Educ Res

This article analyses parental involvement in schools in Germany and Spain based on the results of PISA 2015. The authors examine the SPSS and Excel databases for the responses of the school principals, the parents and the students. Among other points two of the main indexes calculated in PISA 2015 are analyzed: the Index of School efforts to involve parents and the Index of parental involvement in school-related activities. Then, some of the main elements of parental involvement as parental attitudes, habits and opinions in relation with their children's academic performance are examined. An ex post facto descriptive study of the specific aspects of parental involvement in each of the countries is presented. The results show the great efforts made in the schools to involve parents, according to the principals interviewed. The parents have a more moderate evaluation in this respect. Regarding performance in science, it is found in Spain and in Germany that not all factors related to parental participation favor a higher achievement.

Do Teachers’ Instructional Practices Moderate Equity in Mathematical and Scientific Literacy?: an Investigation of the PISA 2012 and 2015

Article

Jul 2018

Many efforts have been made to reach educational equity, especially to reduce mathematics and science achievement gaps by students’ socioeconomic status. Across countries, educators strive to reform traditional teacher-centered instructional approaches to more student-centered/inquiry-based instruction to improve equity in education. In this context, this study examines whether relationships between socioeconomic status and scientific or mathematical literacy are moderated by student-centered instruction. Ten countries covering a wide range of achievement levels as well as equity in education are selected for an international comparison. A linear regression analysis is applied to student achievement, equity, and frequency of student-centered instruction data from the PISA 2012 and PISA 2015. We find mixed results: As student-centered instruction is offered more frequently, the gap in mathematical and scientific literacy between low and high socioeconomic status is generally narrowed or maintained. In most countries, students’ mathematical and scientific literacy scores are expected to decrease across all socioeconomic status as student-centered instruction is given more frequently. The findings necessitate further scrutiny of how teachers implement student-centered instruction in various educational systems. This further research need to consider the complexity of implementation related to sociological and pedagogical aspects.

Comparing the Efficiency of Schools Through International Benchmarking: Results From an Empirical Analysis of OECD PISA 2012 Data

Article

May 2018

This paper derives efficiency scores for around 8,500 schools in 30 countries, using Programme for International Student Assessment 2012 data and a nonparametric approach called data envelopment analysis as method. On average, achievement scores of schools can be increased by 27%, holding inputs constant. Efficiency scores vary considerably both between and within countries; the role of managerial efficiency and structural differences due to operating in different contexts (countries) is disentangled. Subsequently, a number of school-level factors are found to be correlated with efficiency scores and indicate potential directions for improving educational results. Heterogeneity of such characteristics across countries and along the distribution of efficiency is explored.

Finland and Singapore, two different top countries of PISA and the challenge of providing equal opportunities to immigrant students

Article

Apr 2018

International student achievement comparisons have aroused interest widely across the world. Finland dominated the comparisons for a long time but now Asian countries have fared better. Both countries studied have faced abundant immigration, which has changed the student groups. In this article, we compare two quite different top countries: Finland and Singapore. How do immigrant children succeed at school? How do the Finnish and Singaporean educational systems support their academic success? Our purpose is to discuss these countries’ educational systems in the light of immigrant students’ study success. The review showed that the Finnish system can provide profound support for learning among immigrant children and that the strength of the Singaporean system is in the adoption of high aspiration and motivation to strive for success. Both systems provide opportunities to succeed at school, but opportunities to participate and willingness to integrate may also be the keys to academic success. © 2018, Institute for Education Policy Studies. All rights reserved.

Density Estimation for Statistics and Data Analysis.

Article

Jan 1987

PISA Performance of Natives and Immigrants: Selection versus Efficiency

Abstract and Figures

Recommended publications

Assessing the efficiency of secondary schools: Evidence from OECD countries participating in PISA 20...

Immigration and Student Achievement: Evidence From Switzerland

Public and Private School Efficiency and Equity in Latin America: New Evidence Based on PISA for Dev...

Educational Gaps Between Immigrant and Native Students in Europe: The Role of Grade