MESSAGES FROM PISA 2000

OECD
ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT
 TABLE OF CONTENTS

Messages from PISA 2000 ........................................................... 4

Students, schools and education systems ..................................... 5
Student characteristics (1) – Home background .............................. 6
Student characteristics (2) – Engagement ...................................... 8
Student characteristics (3) – Approaches to learning..................... 10
Student characteristics (4) – Gender differences .......................... 12
School characteristics (1) – Social background of students ........... 14
School characteristics (2) – Climate and resources ....................... 16
School system characteristics ..................................................... 18
Conclusions ............................................................................... 20
PISA 2000 profile for Australia .................................................... 22
PISA 2000 profile for Austria ....................................................... 24
PISA 2000 profile for Belgium ..................................................... 26
PISA 2000 profile for Canada ...................................................... 28
PISA 2000 profile for Czech Republic ........................................... 30
PISA 2000 profile for Denmark .................................................... 32
PISA 2000 profile for Finland....................................................... 34
PISA 2000 profile for France ....................................................... 36
PISA 2000 profile for Germany .................................................... 38
PISA 2000 profile for Greece ....................................................... 40
PISA 2000 profile for Hungary ..................................................... 42
PISA 2000 profile for Iceland....................................................... 44
PISA 2000 profile for Ireland ....................................................... 46
PISA 2000 profile for Italy ........................................................... 48
PISA 2000 profile for Japan ........................................................ 50
PISA 2000 profile for Korea ........................................................ 52
PISA 2000 profile for Luxembourg ............................................... 54
PISA 2000 profile for Mexico ....................................................... 56
PISA 2000 profile for New Zealand .............................................. 58
PISA 2000 profile for Norway ...................................................... 60
PISA 2000 profile for Poland ....................................................... 62
PISA 2000 profile for Portugal ..................................................... 64
PISA 2000 profile for Spain ......................................................... 66
PISA 2000 profile for Sweden...................................................... 68
PISA 2000 profile for Switzerland ................................................ 70
PISA 2000 profile for United Kingdom .......................................... 72
PISA 2000 profile for United States ............................................. 74

2 © OECD 2004 © OECD 2004 3

MESSAGES FROM PISA 2000 STUDENTS, SCHOOLS AND EDUCATION SYSTEMS

Since the publication of the first PISA survey in 2001, the OECD has been analysing its The acquisition of knowledge and skills can be influenced by students’
results, and their implications for public policy. This is a summary of the key findings. individual characteristics, by features of their schools, and by the structure of their
education systems.
on back page). This overview picks
out some key messages that have The PISA survey collected infor- Characteristics of school the degree to which individual
emerged from the analysis. mation on a wide range of factors systems, which affect the expe- schools are given autonomy
with a bearing on student per- riences of individual schools and within the education system.
The PISA survey tells countries the
formance. It looked for example students across a whole coun-
extent to which students near the
at the backgrounds of individual try. These include, for example, The following pages look in turn at
end of compulsory education have
students, at how they approach the extent to which secondary each of these aspects across coun-
acquired some of the knowledge
learning and at various character- school students are differenti- tries. Pages 22-75 present a profile
and skills that they will need in later
istics of their schools. ated into separate groups rather for each OECD country in PISA 2000
life. The basic results for reading lit-
than all educated together, and summarising these characteristics.
eracy are shown on the facing page. Some of these factors, such as stu-
National strategies to improve on dents’ socio-economic background,
such performance against interna- cannot be changed by education Reading performance in PISA 2000
tional benchmarks can be usefully systems and need to be taken as
informed by analysis of the features Percentage at:
a given. The influence of these fac-
that characterise countries with Level 1 or
tors is nevertheless worth knowing,
strong results, and of which stu- since this can inform educators Mean score below Level 5
dents and schools perform better Finland 546 7.0 18.5
about how to target particular in- Canada 534 9.6 16.8
The Programme for International The Organisation for Economic within each country. Such analysis terventions. Other factors, such as New Zealand 529 13.7 18.7
Student Assessment (PISA) is a Co-operation and Development is essential to meeting the objec- the learning strategies adopted by Australia 528 12.5 17.6
powerful tool for measuring the (OECD), which co-ordinated the tives of PISA, which was designed students or the atmosphere of the Ireland 527 11.0 14.2
outcomes of education systems. survey, has published four studies not as an educational Olympics, classroom, are directly susceptible Korea 525 5.8 5.7
The first three-yearly survey was examining thematically what the but as a tool to help countries to United Kingdom 523 12.9 15.6
to improvement. Japan 522 10.1 9.9
conducted in 2000, with results survey shows about factors behind improve educational outcomes
Sweden 516 12.6 11.2
first published in 2001. student performance (see details against international standards. This overview considers factors Austria 507 14.6 8.8
associated with student perform- Belgium 507 19.0 12.0
ance at three levels: Iceland 507 14.5 9.1
Norway 505 17.5 11.2
Characteristics of individual France 505 15.2 8.5
United States 504 17.9 12.2
What is PISA? students, including their back-
OECD average 500 17.9 9.5
grounds, their attitudes to Denmark 497 17.9 8.1
learning and their behaviour in Switzerland 494 20.4 9.2
• A three-yearly survey, starting in 2000, of knowledge, skills and other characteristics of 15-year- terms of participation at school Spain 493 16.3 4.2
olds. In the first survey, around 315,000 students in 43 countries took part in pencil and paper and their learning strategies. Czech Republic 492 17.5 7.0
Italy 487 18.9 5.3
tests and filled out questionnaires about themselves. Their schools also provided background Characteristics of schools, in- Germany 484 22.6 8.8
information. cluding the atmosphere of the Hungary 480 22.7 5.1
Poland 479 23.2 5.9
• Specifically, an assessment of reading, mathematical and scientific literacies in a way that school and the classroom as Greece 474 24.4 5.0
described by students, and re- Portugal 470 26.3 4.2
looks at the capacity of students to address real-life challenges. Luxembourg 441 35.1 1.7
sources and school processes
as described by principals. Some Mexico 422 44.1 0.9
• A unique collaboration among governments to monitor educational outcomes, co-ordinated
through the OECD. school characteristics with a Statistically significantly above the OECD average
bearing on student performance Not statistically significantly different from the OECD average
Of the countries taking part, 27 were the OECD member countries shown in the table opposite; these are are the sum of individual student Statistically significantly below the OECD average
the focus of the present summary. The Netherlands participated, but its school response rate was too low characteristics – for example, Note: the PISA results classify students at five levels of reading proficiency. Those at
to ensure international comparability. A further 15 partner countries also participated. the average social background of Level 5 can perform highly complex tasks. Students at Level 1 can only manage the
all the students at a school, and most basic literacy tasks, and a small number of students, unable even to do these
tasks, are classified as below Level 1.
their rate of school attendance.

4 © OECD 2004 © OECD 2004 5

STUDENT CHARACTERISTICS (1) - HOME BACKGROUND

PISA underlines the strength of the link between student background and effects in this context are dis- Aspects of the socio-economic gradient
cussed on pages 14-15.
performance, and helps understand its profile in each country. What does the gradient line show? economic background explains
In aiming to raise student perform- The gradient line slopes up and about 20 per cent of all variation in
Students who come from families given social background. Thus, if a a strategy more closely targeted ance overall and improve equity shows that students from more students’ reading scores.
with more favourable social, eco- country could raise performance on more disadvantaged students by making the social gradient less advantaged socio-economic back-
nomic and cultural characteristics of the least-advantaged quarter of than where below-average per- steep, countries can take heart grounds in general perform better in The social gradient line reflects not
tend to perform better at school. its students to that of the most- formance is more generalised. The from the fact that PISA shows that PISA. Specifically, it shows the range only the extent to which students
PISA allows the strength of this advantaged quarter, this would be stronger the association between such objectives are mutually com- of predicted scores of the middle from advantaged socio-economic
advantage to be measured and equivalent to the worst-performing social background and perform- patible. Indeed, analysis of the 90 per cent of students on an in- backgrounds do better, but also
compared among countries and country raising the score of each ance, the greater the case for us- PISA 2000 results shows that there ternational index of socio-economic the overall level of student perform-
shows that it varies significantly student to the level of a student ing student background as a tar- is a significant negative correlation background (5th to 95th percentile). ance in each country. The level of
across countries. with similar social characteristics geting tool, rather than focusing across countries between the level A student from a comparatively less the gradient lines – their average
in the best-performing country. on under-performance as such. of the gradient line and its steep- advantaged socio-economic back- height – shows the average reading
Overall, socio-economic differ- And in countries where the range ness. This means that, on average, score reached by those students in
ground (5th percentile) tends to be
ence is the strongest single factor In order to develop policies to of social backgrounds among the in countries where students are each country whose socio-econom-
nearly two PISA reading levels be-
associated with performance in raise overall performance and re- student population is the greatest, performing better overall, social dif- ic background is equal to the aver-
hind a student from a comparatively
PISA, accounting for about a fifth duce social differences, countries there may be a case for concentrat- ferences are relatively narrow. age socio-economic background
more advantaged socio-economic
of all variation in student read- need to start by understanding ing resources on disadvantaged across OECD countries.
background (95th percentile). Socio-
ing scores. The gap between the the characteristics of their “social children or their schools to help
least-advantaged quarter of stu- gradient”. Some of its features provide a learning environment (i) The length and horizontal position of the gradient line Reading score OECD Reading level
800
dents and the most-advantaged are described on the facing page, that helps compensate for lower shows the spread of student backgrounds: the longer i Iceland Mexico
quarter is equivalent to more and summarised for each country resources in the home. the line, the more varied socio-economic background is 700 Level
V
than one reading proficiency level in the profiles on pages 22-75. A among students; the further right the line, the more fa-
600 Level
on PISA’s five-level scale. This range of strategies may be envis- Such strategies need to take ac- vourable students’ socio-economic background is in gen- IV
eral. Students in Iceland have on average a much more Level
gap within countries attributable aged, according to the shape of count not only of individual stu- 500 III

to social background is similar to the gradient. For example, where dents’ backgrounds but of the ef- advantaged socio-economic background than students Level
II
in Mexico (the line is further to the right), and the range 400
Level
the range in performance across average performance is high but fects of the socio-economic char- I
of socio-economic backgrounds is narrower (the line is
countries of students with a the gradient steep, this argues for acter of whole schools. School shorter). The country profiles on pages 22-75 show ad- 300 Below
Level I
justed reading scores that each country might expect if 200
Student characteristics its social profile were average. 800
-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3

ii Japan Hungary
PISA has identified a number of Characteristics of students who tend to do well at school (ii) Despite the overall pattern, some students with more 700 Level
V
advantaged socio-economic background do poorly in
aspects of students’ background, 600 Level
Student Difference in reading score PISA, while some with less advantaged socio-economic IV
attitudes and behaviours associ- characteristic1 between top and bottom quarters Level
background do well. How closely do they conform to the 500 III
ated with strong performance in predicted trend? This is shown by representing each stu- Level
reading and other literacies. Favourable home dent’s performance as a dot, and seeing how closely 400
II
Level
background I
they cluster around the gradient line. In Finland, Iceland,
Understanding these character- Arrives on time 300
Japan and Korea, the influence of socio-economic back- Below
Level I
istics can help education policy for school ground on student performance is limited, with over 90 200
makers to target interventions de- Interested in reading per cent of student differences accounted for by other -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
800
signed to help particular groups, factors. In Hungary, on the other hand, just over a quar- iii Germany Korea
and to promote particular charac- Confidence in ter of differences in student reading scores can be at- 700 Level
learning ability V
teristics (such as successful ap- tributed to social background.
600 Level
proaches to learning) across the Controls own IV
learning (iii) How severe is social disadvantage in its effect on Level
student population. 500 III

Being female performance? The slope of the social gradient line shows Level
II
The comparison in the graph to how much difference a given amount of social differ- 400
Level
the right gives an indication of the High sense ence makes to a student’s predicted reading score: the I

of belonging steeper the gradient, the more inequality. In Germany, 300 Below
relative importance of a range of Level I

student characteristics discussed PISA score points 0 20 40 60 80 100 it makes nearly three times as much difference as in 200
Japan and Korea. -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
on the next eight pages. 1. As reported by students. Socio-economic status

6 © OECD 2004 © OECD 2004 7

STUDENT CHARACTERISTICS (2) - ENGAGEMENT

Students who are engaged in reading are far more likely to have high levels of Engagement at school Student engagement measures
reading literacy. Student engagement at school more generally has a bearing on How engaged How many students
As well as interest in particular as-
wider educational outcomes. are students are weakly engaged
pects of learning such as reading,
in reading? at school?
a student’s overall engagement at
Students who are interested in The implication for school sys- Students from less favourable school is also a key factor in second- Low sense Low
learning tend to learn more effec- tems is that improved teaching socio-economic backgrounds are on ary education. A substantial minority Index of belong- participa-
tively, and to achieve better results relies not just on instructional average less engaged in reading. of students – one in four 15-year-olds scores1 ing2 % tion3 %
at school. The PISA results under- strategies for improving stu- However, a substantial number of in the PISA survey – say they do not Finland 0.46 United Kingdom 17.4 15.0
line the importance of student en- dents’ cognitive skills, but also disadvantaged students are among want to be at school. Analysis of Iceland 0.27 Sweden 17.7 23.8
gagement. For example, students on engaging their interest and the most interested and wide-ranging student replies to the PISA question- Denmark 0.26 Hungary 18.8 17.7
who are habitual readers and ensuring that they are well mo- readers, and these students tend to naire has identified about one in four Korea 0.21 Ireland 19.4 17.8
who enjoy reading are also more tivated. Different strategies may perform well in reading. Indeed, the students who have a low sense of
likely than others to have high be appropriate for boys and for level of a student’s reading engage- Japan 0.20 Austria 20.3 15.3
belonging in the social environment
levels of reading literacy. Greater girls, who tend to have differ- ment is a better predictor of literacy Sweden 0.14 Canada 20.5 26.0
of school, and about one in four who
engagement in reading can be a ent reading interests, with girls performance than his or her socio- regularly miss or are late for school or Portugal 0.13 Australia 20.7 18.3
consequence, as well as a cause, particularly interested in books, economic background, indicating classes (low participation). As shown Norway 0.09 Portugal 20.7 20.1
of higher reading skill, but the especially fiction, and boys more that cultivating a student’s interest in the table on the right, a substan- Mexico 0.07 Switzerland 20.8 15.7
evidence suggests that these two interested in other forms such as in reading can help overcome home tial proportion – at least 17 per cent New Zealand 0.05 Denmark 20.9 m
factors are mutually reinforcing. newspapers and comics. disadvantages. – feel a low sense of belonging in all Hungary 0.03 New Zealand 21.1 26.9
countries, but some countries man- Czech Republic 0.02 Norway 21.1 17.9
age to contain low participation to a
Two forms of student engagement smaller number. In Japan and Korea
Canada 0.01 Finland 21.3 22.9
OECD average 0.00 Mexico 22.0 21.4
fewer than 10 per cent of students
Switzerland 0.00 Iceland 22.4 26.0
report regular lateness or absence.
Reading for pleasure Australia -0.04 Germany 22.6 12.9
Students who are not engaged at Italy -0.08 Greece 22.7 28.8
+ Sense of belonging school are not necessarily those Austria -0.08 Italy 22.9 21.7
with the lowest performance. It is Greece -0.09 Spain 24.0 34.0
+ notable that substantial numbers
Reading widely Poland -0.10 OECD average 24.5 20.0
of medium to higher achievers
Participation United Kingdom -0.10 United States 25.0 20.2
(attendance of school are also disengaged from school
+ United States -0.14 Luxembourg 28.3 13.4
and classes) in this respect. Yet these people
may still be at risk in the future, France -0.18 Czech Republic 29.8 20.7
Attitude to reading particularly if they decide not to Luxembourg -0.19 France 30.2 15.3
=
continue their education. Thus, Ireland -0.20 Belgium 31.6 14.1
= intervention strategies may also Spain -0.23 Japan 37.6 4.2
be needed to help students who Germany -0.26 Poland 41.2 29.2
are not necessarily doing badly Belgium -0.28 Korea 41.4 8.4
“Engagement in reading” “Engagement at school” at school. These students can 1. The index is set with a mean of zero and two-thirds of students fall in between
be hard to target. However, the 1 and -1.
Weak
association analysis also showed that schools
Medium 2. Students were asked whether they strongly agree, agree, disagree or disagree
association
for where students perform poorly strongly, in each case that: School is a place where: a) I feel like an outsider (or left
individuals,
stronger overall also have a tendency to out of things); b) I make friends easily; c) I feel like I belong; d) I feel awkward and
for schools
be those where students become out of place; e) Other students seem to like me; f) I feel lonely. Students with a “low
disengaged. This suggests that sense of belonging” express negative attitudes in at least one respect.
the whole school environment is 3. Students’ participation is measured according to how many times in the past two
Performance important for student engagement
and that the targeting of particular
weeks they say that they: missed school; skipped classes; arrived late. Students have
“low participation” if they report a frequency of at least: “1 or 2 times” to all three
in reading schools can be appropriate. items, OR: “3 or 4 times” to “missed school”, OR “3 or 4 times” to both “skipped
classes” and “arrived late for school”.

8 © OECD 2004 © OECD 2004 9

STUDENT CHARACTERISTICS (3) – APPROACHES TO LEARNING

PISA found strong relationships between students’ attitudes, learning strategies Learning autonomy
and performance. The results also show that students with the autonomous
A further important finding is that difference in students’ literacy About two-thirds of differences
learning strategies needed to become lifelong learners are characterised by strong in student use of “control strate-
students’ motivation and self- performance. But if students’
motivation and self-belief. belief may have even greater im- tendency to control their learn- gies” can be explained by the var-
plications for their capacity for ing is taken as an outcome of ying levels of motivation and self-
The PISA survey asked students er they measure their progress The evidence suggests that stu- belief expressed by students who
lifelong learning than for their per- learning – since learning auton-
about a range of their character- against their goals to control their dents who are more self-confident use such strategies more or less
formance at school. Student ap- omy is a key precondition of life-
istics as learners. It asked them own learning). and have stronger motivation do often.
proaches to learning measured in long learning – an even stronger
about their motivation (for exam- better at school largely because
PISA explain about a fifth of the relationship becomes visible.
ple, their interest in reading, and The survey found that in a number they are more inclined to invest in
their commitment to use educa- of respects, students with strong- learning strategies that work. For
tion to get a good job), their self- er approaches to learning are example, students who believe
likely to have higher literacy per-
Relative performance of students who control their learning more
belief (for example, whether they they can succeed in performing
believe they can handle learning formance, and that these relation- tasks that they find difficult are Advantage in PISA score points Mean score in reading for students who control
tasks effectively) and their learn- ships apply across different coun- more likely to make an effort to for students who control their their learning…
ing strategies (for example, wheth- tries and cultures. control their learning, checking learning more1 most least
their own progress and working out Portugal 96 517 421
what they still need to know. Such
A measure of students’ self-belief: behaviour, in turn, is associated New Zealand 77 571 494
how effective do they feel as learners? with higher performance in PISA. Australia 70 565 495
Index of self-efficacy Czech Republic 66 531 465
These findings suggest that strat-
Mexico 2.76
egies to improve teaching and Scotland 62 555 493
Austria 2.67
learning techniques need to do
Switzerland 2.65 more than just offer students a Germany 61 521 460
Scotland 2.63 learning tool-kit. Students will only United States 61 534 473
United States 2.63 use learning tools if they feel moti-
Ireland 56 554 498
Australia 2.62 vated and believe in their capacity
Belgium (Fl.) 2.60 to learn. So measures to improve Mexico 55 449 394
New Zealand 2.60 learning techniques must go hand
Luxembourg 53 478 425
Sweden 2.59 in hand with measures to nurture
Germany 2.59 stronger attitudes to learning. OECD average 53 528 475
Italy 2.59 Korea 51 549 498
How strong are these attitudes in
Hungary 2.58 different countries? Such compar- Sweden 49 538 489
Iceland 2.58 isons need to be made with care,
Switzerland 49 522 473
Norway 2.56 since for example it can be shown
OECD average 2.56 that students do not always mean Austria 44 532 488
Portugal 2.54 the same thing in different cul- Italy 44 505 461
Denmark 2.52 tures when, for example, they
say they are interested in read- Hungary 40 497 457
Ireland 2.50
Luxembourg 2.49 ing. However, some cross-country Iceland 37 527 490
comparisons are more robust,
Finland 2.47 Finland 36 562 526
and the table to the left ranks
Czech Republic 2.41
countries in order of students’ av- Denmark 32 516 484
Korea 2.28 erage level of belief in their learn-
The scale ranges from 1 to 4 and shows how frequently, on average, students agree Belgium (Fl.) 27 544 517
ing efficacy.
with statements such as “I am certain I can understand the most difficult material
Norway 26 520 494
presented in reading”: 1 (almost never), 2 (sometimes), 3 (often) and 4 (almost al-
ways). Countries at the top have students who are more confident, on average, about 1. Difference in score points between students in the top and bottom quarters of the index of control stategies. Based on stu-
dealing with learning situations they find difficult. dents’ reports.

10 © OECD 2004 © OECD 2004 11

STUDENT CHARACTERISTICS (4) – GENDER DIFFERENCES

PISA revealed considerable gender differences among 15-year-olds, of which the

most consistent among countries is that girls are more interested in reading and Differences between the interests
perform on average better in reading literacy. Yet both boys and girls have specific and study habits of boys and girls
strengths and weaknesses, suggesting that differentiated strategies may be
needed to meet their needs. PISA identified a number of qualitative differences between the interests and study habits of boys and
girls, which suggest that different strategies may be appropriate in addressing the learning needs of
each gender.
In ever y countr y par ticipating in
PISA 2000, girls per form on av- Boys’ and girls’ reading characteristics
erage higher in reading literacy Reading performance How far Interest in
than boys. The average gap is (mean score) girls are reading: Interest in different areas
substantial: nearly half a profi- ahead how far Girls are more interested in reading, and boys in mathematics. This is clearly a factor associated with
ciency level (32 points). While (score girls are differences in per formance, and it is notable that in Finland, the countr y where girls are the furthest
the gap is less than half this
Boys Girls points) ahead1 ahead of boys in reading per formance, there is also the largest gap in interest, whereas Korea has the
average in Korea, in other coun-
Australia 513 546 34 0.36 smallest gap in both respects. However, these two factors are not strongly associated across countries
tries the contrast between boys
Austria 495 520 26 0.62 (see table opposite).
and girls is stark. For example,
in Nor way, girls score on average Belgium2 492 525 33 0.54
PISA also found that boys and girls have different types of reading interests, which is fairly consistent
529 points, which is higher than Canada 519 551 32 m
across countries. Boys’ interest in a wide range of materials including non-fiction, newspapers and comics,
the average student score (for Czech Republic 473 510 37 0.79
but their much lower interest in reading fiction books, suggests that the choices of reading materials may
boys and girls) in all but three Denmark 485 510 25 0.53 influence the success of any programme to engage boys more in reading.
of the 27 OECD countries in Finland 520 571 51 0.96
the sur vey, whereas boys score France 490 519 29 m
486 points, lower than the aver-
Germany 468 502 35 0.60
age in all but seven countries Characteristics as learners
Greece 456 493 37 m
overall.
Hungary 465 496 32 0.49 Boys and girls each have distinctive strengths and weaknesses in terms of how they approach learning.
In mathematical literacy, boys Iceland 488 528 40 0.45
do better overall, but only out- Ireland 513 542 29 0.53 Part of this is a matter of attitude and motivation: girls are more confident and motivated in reading; boys
per form girls in about half the in mathematics. Boys also have a stronger general confidence in their ability to overcome obstacles and be
Italy 469 507 38 0.58
PISA countries. In the others, effective as learners, while girls report greater effort and persistence.
Japan 507 537 30 m
there is no significant gender Korea 519 533 14 0.02 Another feature of difference is in learning strategies. Girls tend to be more systematic about controlling their
gap. Moreover, even in those Luxembourg 429 456 27 0.43 own learning, and to memorise material. Boys are more inclined to “elaborate” new knowledge, by relating it
countries where girls do less
Mexico 411 432 20 0.32 to what they already know. While these differences do not apply in every country, they give useful insights into
well in mathematics, the nature
New Zealand 507 553 46 0.37 the strengths and weaknesses of boys and girls, and which learning skills each need to work on.
of this underper formance differs
from the underper formance of Norway 486 529 43 0.60
Finally, boys have a stronger preference for competitive learning situations and girls (less consistently
boys in reading. In the case of Poland 461 498 36 m
across countries) are more inclined to favour co-operative situations.
mathematics, boys’ advantage Portugal 458 482 25 0.80
derives mainly from a dispropor- Spain 481 505 24 m
tionate number of them per form- Sweden 499 536 37 0.47
ing ver y well. Girls are, on aver- Switzerland 480 510 30 0.68 Engagement at school
age, no more likely than boys to United Kingdom3 512 537 26 0.43
have low mathematical literacy. Even though girls fare better in PISA than boys in many respects, the survey does not support the notion
United States 490 518 29 0.36 that difficulties at school age 15 are concentrated among male students. In particular, there is no signifi-
In contrast, for reading litera-
cy, boys are nearly 70 per cent
OECD average 485 517 32 0.53 cant difference overall between the frequency with which boys and girls report having a low sense of belong-
more likely than girls to have low 1. This is based on an index of interest in reading and shows positive effect sizes ing at school. There is only a minor difference between their chances of having low attendance, but in some
per formance. from 0 to 1. Positive effects show that girls are more interested in reading: an effect countries these differences are much higher. Notably in Greece and Poland, two of the three countries with
of 0.20 is small, 0.50 is medium and 0.80 is large. the highest number of students missing school or classes, the rate is only two-thirds and three-quarters as
2. Interest in reading data for the Flemish Community only. high, respectively, for girls as for boys.
3. Interest in reading data for Scotland only.

12 © OECD 2004 © OECD 2004 13

SCHOOL CHARACTERISTICS (1) – SOCIAL BACKGROUND OF STUDENTS

The social background of all students in a school is strongly associated with across countries is greater when
the school effect is taken into
Average student reading performance in schools with
reading performance. low socio-economic status1 compared with other schools
account. Iceland has the least
severe social gradient at both Reading score Reading level
As noted earlier, the social levels of literacy than those at nearly twice that of individuals – the individual and the school
background of an individual less advantaged schools – and i.e. if one compares two schools level, and Germany has the
student is the strongest this superior per formance is with different social composition, 600 Level
steepest at both levels. However,
single factor associated with greater than can be accounted for the predicted difference in the slopes of the gradient in the Students not in IV
per formance in PISA. However, it by the sum of their own individual average reading scores is twice two countries differ by a factor of low SES schools
is not just the characteristics of advantages. as great as it would be on the three at the individual level but a
an individual’s family but also the basis of predicting the individual factor of five at the school level. Level
characteristics of the families of The importance of the whole- scores of each student attending III
other students in the school school social profile is illustrated those schools. Thus there appears to be an 500
that are closely associated with in the graph below, using the advantage for an individual in
how well individuals per formed United States as an example This is an average; in some attending a school in which
in PISA. On average, students of a countr y with average countries the compounding Level
other students have more
who attend schools with a more characteristics in terms of the effect is much higher; in others favourable home backgrounds. II
advantaged “social profile” are social gradient. Note that the lower. As a result, variation in the That advantage may stem from a
likely to show considerably higher slope of the school gradient is steepness of the social gradient variety of factors, including peer- 400
group influences, differences Students in Level
in the resources or quality of low SES schools I
Average student reading performance and socio-economic
schools attended by different
status in individual schools in the United States social groups, or differences in
Each dot represents one school. The case of the United States, a country whose social gradient teacher expectations.
is similar to the OECD average, illustrates the school effect. 300 Below
However, the results also show Level I
School social gradient Individual social gradient that the social profile of a school
Reading score Reading level does not determine its results,
700 and that for two schools whose
students have the same average
-3 -2 -1 0 1 2
Level Socio-economic status (SES)
V socio-economic status, average 1. Average student reading performance in the lowest quarter of schools on the index
reading per formance can var y of socio-economic status.
600 by as much as two proficiency
Level
levels. there is more of a “convergence” for students in the least-favoured
IV
in their per formance, with schools. The graph above shows
Analysis of the PISA results background mattering less. that students in the lowest quarter
Level indicates that the most important of schools on the index of socio-
500 III factor influencing whether a Since most variation in student economic status have no overlap
school does well compared to per formance tends to be within in predicted reading scores
Level other schools with similar social schools, a key priority is to help with those from other schools.
II background is whether the less the least advantaged individuals Even one of the 5 per cent most
advantaged students within the within schools to achieve their socially privileged students
400
school achieve good results. In potential. In particular, those within a less-advantaged school
Level
I general, the impact of the social within schools with below-average has an expected score below one
profile seems to be greater in social profiles may benefit most of the least privileged students
schools where there are more from compensator y assistance. in a school with a higher social
300 Below disadvantaged students: the lower profile. Thus, policies that limit
Level I the social profile, the greater the However, this analysis also the extent of social segregation
differences between students suggests that high segregation of across schools appear likely to
-1.5 -1.0 -0.5 0 0.5 1.0 1.5 from different backgrounds. For students by social background can help more students to achieve
Socio-economic status students at better-off schools create an intense disadvantage their potential.

14 © OECD 2004 © OECD 2004 15

SCHOOL CHARACTERISTICS (2) – CLIMATE AND RESOURCES

The school environment makes a tangible difference to learning outcomes, and in How much student performance varies with aspects of school climate and resources
particular the atmosphere created by students and teachers has measurable effects. (Average effect within OECD countries)
How much difference does the level, except for student-related The PISA results indicate that it is Mean reading score in the:
quality of a school make to learning factors affecting school climate. student attitudes and behaviour Bottom quarter of schools Score point difference Top quarter of schools
outcomes? PISA asked students Nevertheless, if schools were that are particularly important,
and school principals questions able to improve performance and that an atmosphere in which
about various characteristics by these kinds of amounts as a they are committed to purposeful
of schools that might make a result of improvements in school learning makes a key difference. Student-related factors
difference to learning outcomes. climate and resources, this affecting school climate1 473 49 points 522
Factors such as the environment would represent a substantial These results confirm a range
in the classroom or the physical educational gain. of other research suggesting
infrastructure of the school that students perform best in a
are more susceptible to policy The PISA results underline the positive learning environment that
influence than students’ home particular importance of school is oriented towards results. They
Disciplinary climate2 483 39 points 522
backgrounds, and therefore climate as a factor affecting also relate to PISA’s finding that
of particular interest to policy school performance. Its effect is students who are ready to invest
makers. more discernible than the level effort and who enjoy learning thrive
of school resources. Overall, the as individuals. They are best able
The graph shows that a range measured school climate variables to develop these characteristics Teachers' morale and
of school characteristics are account for about 6 per cent in purposeful and well disciplined 481 29 points 515
commitment1
associated with student reading of between-school differences schools and classroom
performance. Compared to social in performance, while school environments. It is interesting
background, the impact of these resources account for only about to note that the extent to which
factors appears modest: rated 1 per cent. A range of factors teachers emphasise academic Quality of schools'
on any one of the characteristics affect school climate, including performance is also positively 488 23 points 511
educational resources1
shown, the gap between students the attitudes of both teachers related to performance, but less
in the top and bottom quarter of and students and the quality of strongly so than the disciplinary
schools is below half a proficiency the relationship between them. climate of the classroom.

Teacher shortage1 488 22 points 510

School characteristics
How much difference does it make what school you go to? PISA found that although much vari- Teacher-related factors
affecting school climate1 489 16 points 505
ation in student performance is attributable to differences within schools, a substantial amount
(varying greatly by country) reflects the fact that students at some schools do better than those
at others.

In Austria, Belgium, the Czech Republic, Germany, Greece, Hungary, Italy, Mexico and Poland, Quality of schools'
498 6 points 504
between-school variation is greater than within-school variation. By contrast, in Finland, Iceland, physical infrastructure1
Norway and Sweden, around a tenth of variation lies in between-school differences.

What is it that makes students at some schools per form so much better than others? Analysis of Average for all students = 500 points
the PISA 2000 results shows that the most important influence is the combined background of the
students in a school, and in particular differences in average socio-economic status. Characteristics For example: on average in OECD countries the quarter of schools with the least favourable student influences on
school climate have average student reading scores of 473 points.
of the school itself play a smaller, but still significant role, and in particular students do better on
average in schools with a positive climate for learning. 1. As reported by school principals.
2. As reported by students.

16 © OECD 2004 © OECD 2004 17

SCHOOL SYSTEM CHARACTERISTICS

PISA does not allow us to design a perfect education system, but gives clues 1 Successful education systems have been extending school autonomy
about which features of school systems are relevant to student outcomes.
During the past two decades, many countries have given schools greater autonomy in a range of institu-
tional operations, aiming to raise performance levels by devolving responsibility to the front line.
Over half of the variation in student Thus, the degree to which students’ features that might help explain
performance in OECD countries in educational chances are affected differences both in the overall
In most of the countries that performed well in PISA 2000, local authorities and schools now have substan-
PISA 2000 is accounted for by the by which country they live in should performance and the equity of
tial freedom to adapt and deliver educational content and/or to allocate and manage resources. In all OECD
variation of student performance not be exaggerated. However, dif- student outcomes across coun-
countries, most 15-year-olds are in schools that have some responsibility for student admissions. Except in
within each school. About a third ferent school systems vary not just tries. This analysis does not pro-
Germany, Italy and Switzerland, most schools also play a role in deciding on the courses offered. Schools
is attributed to differences in per- in their average scores but also in duce prescriptions for education
are also gaining autonomy over institutional operations, and most principals have at least some control
formance across schools within the dispersion of scores and, as systems, but makes observations
over budgets, although control of teacher salaries most commonly remains with central authorities.
countries. The amount due to dif- seen above, in the strength of the designed to help policy makers
ferences across countries is rela- relationship with factors such as think about the effect of certain The PISA 2000 results suggest that in those countries where schools have greater freedom to choose
tively small – 9 per cent of all vari- social background. system features. In looking at fea- courses, average performance in reading literacy tends to be significantly higher. The picture is similar,
ation in the case of reading literacy tures of more successful systems though less pronounced, for other aspects of school autonomy, including the relationship between mean
and scientific literacy; 15 per cent Analysis of the PISA results has in PISA 2000, three particular ob- performance and the degree of school autonomy in budget allocation. This finding cannot, of course, be
for mathematical literacy. started to look at some system servations have emerged. interpreted in a causal sense as, for example, school autonomy and performance could well be mutually
reinforcing or influenced by other factors.

2 Successful education systems are committed to monitoring student and system performance
Performance standards can only be maintained if they are consistently implemented and assessed.
Assessments of student performance are now common in many OECD countries.

These assessment systems have a range of rationales and forms. Different countries use various forms of
external assessment, external evaluation or inspection, and schools’ own quality assurance and self-evalu-
ation efforts. Some countries see such assessments primarily as tools to reveal best practices and identify
shared problems in order to inform improvement. Others extend their purpose to support contestability of
public services or market-mechanisms in the allocation of resources, e.g. by making comparative results
of schools publicly available to facilitate parental choice or by having funds following students. While there
is no single model that best supports school improvement, higher-performing countries in PISA have been
putting increased emphasis on the monitoring of their schooling systems.

3 The method used in an education system to support

low-performing students is critical to the raising of performance
An important aspect of country differences in PISA is that much of the variation in overall performance is
attributable to differences in the number of low-performing students. Germany and Japan, for example,
both have an average percentage of students reading at level 5, but Germany has twice as many at level 1
or below: this is what makes Germany’s average performance below average and Japan’s above average.
Such differences are also associated with differences in social gradients.

Country approaches to helping disadvantaged students vary widely. Some strategies focus resources on
targeted groups of students. Others concentrate on changing the way in which students are allocated to
schools, in some cases making schools less selective. The effectiveness of these policies remains con-
troversial. However, the results from PISA 2000 suggest that overall variation in student performance and
performance differences between schools tend to be greater in those countries with rigid institutionalised
selection and tracking practices at early ages.

18 © OECD 2004 © OECD 2004 19

CONCLUSIONS

The initial OECD report on of factors”. Subsequent relationships play out. the PISA results indicate that the more successful countries Such evaluation does not
findings from PISA 2000 analysis has shed light on the the most successful countries have mainly employed an inte- mean rigid control from the
reported that there is no single relative importance of factors This analysis found some have managed to “level up” grated approach to grouping centre; indeed, devolution
key to success in PISA: rather, within this constellation, common factors among standards. students. This puts the onus to the front line has been
“successful performance is and offers a profile for each countries, as well as some on differentiated systems to an important dynamic in
attributable to a constellation country describing how these important differences. While the need to help less ad- look carefully at how they can educational improvement
vantaged and worse-per form- avoid limiting less able stu- in many countries. Rather,
ing students is widely shared dents’ chances. it means a co-ordinated
Important common factors among countries as a priority, the method of and consistent approach to
doing so remains controver- PISA together with recent re- tracking outcomes. PISA itself
• The significance of social background differences as a factor that helps explain variations in
sial. Some systems continue search suggests that in improv- will continue to be par t of this
student performance. While this difference does vary substantially by country, in every country
to separate out students by ing their education systems in process at an international
social background was the single most important factor that PISA identified, both in accounting
ability; others have more of a response to such messages, level. The results of the second
for variations among individuals and for variations across schools.
“comprehensive” approach to the important thing for individu- three-yearly sur ver y appear in
• The importance of student attitudes as a prerequisite for successful learning. Within countries, student groupings. PISA can- al countries is not to copy their Learning for Tomorrow’s World
students who are interested in what they learn and believe in their own abilities are much more not determine which system neighbours directly but to moni- – First Results from PISA 2003
likely to do well, even once other factors have been taken into account. This finding gives a is best for a par ticular coun- tor carefully the evolution and (OECD, 2004).
very direct message to school systems that efficient instructional methods are not on their own tr y, but shows that to date outcome of their own system.
enough to assure strong learning outcomes: unless the motivation and interest of students can
be enhanced, learning gains are likely to be constrained.
• The influence of the atmosphere within schools and classrooms in relation to student outcomes.
In every country, having a positive school climate had a stronger measurable relationship with
student reading performance than the level of physical resources of the school.

Among the main differences Moreover, country differences note the very strong finding
among countries uncovered in the performance advantage of PISA that achieving greater
by PISA, the most striking was associated with individual so- equity need not be at the
the degree to which student cial background are compound- expense of overall standards.
performance varied across ed by the varying degrees to On the contrary, there is a
schools. In some countries, which the social composition negative correlation between
most of the variation in student of the school appears to ad- the amount of difference
reading performance can be vantage students. Thus while between the predicted
predicted simply by looking at the steepness of the “social performance of students from
the characteristics of the school gradient” varies by a factor of social backgrounds and the
they go to. In others, 90 per cent about three across countries overall level of performance.
of variation is contained within when looking just at individu- This is consistent with the
individual schools. Some of the als, the predicted difference in finding that the biggest factor
extremes of this difference can performance between students that distinguishes more and
be accounted for by the fact that attending schools with differ- less successful countries in
some countries separate more ent social profiles is five times PISA is not how well students
and less able students into as high in the highest country are doing at the top, but how
different schools. However, even than in the lowest. well they are doing at the
the variation across countries bottom. Rather than suggesting
with similar education systems In seeking to learn from these that an emphasis on equality
is striking in this regard. differences, countries should might lead to “levelling down”,

20 © OECD 2004 © OECD 2004 21

 3. Student characteristics
PISA 2000 Profile for Australia Approaches to learning Performance advantage in reading
Compared to other OECD students, literacy of students...
students from Australia have: ...who control
their learning
confidence in their own
above average ...with interest
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy Mean score in Mean score reading ability.
their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability. PISA score points 0 20 40 60 80 100 120

Australia 528 18 12 102 19 533 528

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 in reading literacy and schools…
In Australia: ...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 18% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Australia have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
PISA score points 0 20 40 60 80 100 120
Level
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Australia have: ...better educational
resources
Quality of the schools’
close to average
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Australia OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Australia 0.34 17 -15 2.9 46 34 78 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Australia 100 100 100 99 94 96 96 84 60 47 19 18
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

22 © OECD 2004 © OECD 2004 23

 3. Student characteristics
PISA 2000 Profile for Austria Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Austria have:
their learning
confidence in their own ...with interest
above average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Austria 507 9 15 93 60 515 519

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Austria: in reading literacy and schools…
...where students
have the highest
• 20% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 15% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Austria have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Austria have: ...better educational
resources
Quality of the schools’
close to average
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
Austria OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Austria 0.10 14 1 2.7 41 10 135 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Austria 96 93 99 69 75 14 57 54 15 5 1 1
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

24 © OECD 2004 © OECD 2004 25

 3. Student characteristics
PISA 2000 Profile for Belgium Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Belgium (Fl.) have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Belgium 507 12 19 107 60 520 496

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Belgium: in reading literacy and schools…
...where students
have the highest
• 32% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 14% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Belgium have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
less favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Belgium have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage1

-3 -2 -1 0 1 2 1. Flemish Community only.

Belgium OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Belgium -0.03 22 13 3.1 48 14 133 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Belgium 99 99 99 100 95 98 61 59 96 95 7 7
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

26 © OECD 2004 © OECD 2004 27

 3. Student characteristics
PISA 2000 Profile for Canada Engagement at school
Relationship between student performance
...where students
in reading literacy and schools…
In Canada: have the highest
sense of belonging
• 21% of students have a low sense of belonging, Positive correlation
compared to 25% on average in OECD countries. ...where students have
1. Student performance • 26% of students have low participation (attendance),
highest participation

compared to 20% on average in OECD countries. Correlation -0.2 0 0.2 0.4 0.6 0.8 1
In reading literacy Mean score in Mean score
Mean % at reading level Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy
Canada 534 17 10 95 18 533 529 4. School characteristics
OECD 500 9 18 100 35 500 500 Climate
Compared to other OECD students, Performance advantage in schools with
students from Canada have: a more positive climate...
less favourable Disciplinary climate …in terms of student-
2. Socio-economic status (SES) close to average
Teachers’ morale and related factors

The socio-economic gradient commitment PISA score points 0 20 40 60 80 100 120

Teacher-related factors
more favourable
Reading score Reading level affecting the school climate

Resources
600 Level
Compared to other OECD students, Performance advantage in schools with...
IV
students from Canada have: ...better educational
resources
Quality of the schools’
Level higher
physical infrastructure PISA score points -20 0 20 40 60 80 100
500 III
close to average Teacher shortage

Canada OECD average

Level
II
400
Level 5. System characteristics
I
School autonomy
Percentage of students attending schools with at least some responsibility for:
300 Below Student Student
Level I disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
-3 -2 -1 0 1 2
Canada 98 99 89 94 89 77 90 49 82 61 34 34
Socio-economic status
OECD 95 94 92 89 84 76 71 69 61 54 26 23

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Canada 0.27 11 -7 3.1 37 28 73
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

28 © OECD 2004 © OECD 2004 29

 3. Student characteristics
PISA 2000 Profile for Czech Republic Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Czech Republic have:
their learning
confidence in their own ...with interest
below average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.
Czech
492 7 18 96 53 498 511
Republic Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Czech Republic: in reading literacy and schools…
...where students
have the highest
• 30% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 21% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Czech Republic have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
less favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Czech Republic have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
Czech Republic OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
Czech disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Republic
-0.10 22 10 2.7 49 19 131 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41 Czech
Republic
100 99 100 100 89 83 82 82 96 95 73 70
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

30 © OECD 2004 © OECD 2004 31

 3. Student characteristics
PISA 2000 Profile for Denmark Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Denmark have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Denmark 497 8 18 98 19 514 481

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Denmark: in reading literacy and schools…
...where students
• 21% of students have a low sense of belonging, have the highest
compared to 25% on average in OECD countries. sense of belonging Positive correlation

Correlation -0.2 0 0.2 0.4 0.6 0.8 1

2. Socio-economic status (SES)
The socio-economic gradient

Reading score Reading level

4. School characteristics
600 Level Climate
Performance advantage in schools with
IV Compared to other OECD students,
students from Denmark have:
a more positive climate...
less favourable Disciplinary climate …in terms of student-
Level related factors
500 III Teachers’ morale and
close to average PISA score points 0 20 40 60 80 100 120
commitment
Teacher-related factors
Level more favourable
affecting the school climate
II
Resources
400 Performance advantage in schools with...
Level Compared to other OECD students,
...better educational
I students from Denmark have: resources
Quality of the schools’
close to average PISA score points -20 0 20 40 60 80 100
physical infrastructure
300 Below less Teacher shortage
Level I

Denmark OECD average

-3 -2 -1 0 1 2
Socio-economic status

5. System characteristics
School autonomy
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2
Percentage of students attending schools with at least some responsibility for:
Mean socio- Percentage of explained score if students had the Student Student
economic variation in student average OECD SES Length of the Within Between disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
status performance (score points) gradient1 Overall schools schools plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
Denmark 0.11 17 1 2.8 42 34 79
Denmark 99 98 100 87 87 89 77 90 97 57 15 13
OECD 0.00 20 3.0 41
OECD 95 94 92 89 84 76 71 69 61 54 26 23
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

32 © OECD 2004 © OECD 2004 33

 3. Student characteristics
PISA 2000 Profile for Finland Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Finland have:
their learning
confidence in their own ...with interest
below average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Finland 546 18 7 89 12 536 538

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Finland: in reading literacy and schools…
...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 23% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Finland have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Finland have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I close to average Teacher shortage

-3 -2 -1 0 1 2
Finland OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Finland 0.08 9 -2 2.9 30 27 47 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Finland 96 99 100 89 54 56 95 91 35 21 2 1
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

34 © OECD 2004 © OECD 2004 35

 3. Student characteristics
PISA 2000 Profile for France Engagement at school Relationship between student performance
In France: in reading literacy and schools…
...where students
• 30% of students have a low sense of belonging, have the highest
compared to 25% on average in OECD countries. sense of belonging
Positive correlation
• 15% of students have low participation (attendance), ...where students have
1. Student performance compared to 20% on average in OECD countries. highest participation

In reading literacy Mean score in Mean score Correlation -0.2 0 0.2 0.4 0.6 0.8 1
Mean % at reading level Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy France OECD average
France 505 8 15 92 m 517 500
OECD 500 9 18 100 35 500 500

2. Socio-economic status (SES)

The socio-economic gradient

Reading score Reading level

600 Level
IV

Level
500 III

Level
II
400
Level
I

300 Below
Level I

-3 -2 -1 0 1 2
Socio-economic status

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
France -0.12 23 6 2.9 48 21 106
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

36 © OECD 2004 © OECD 2004 37

 3. Student characteristics
PISA 2000 Profile for Germany Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Germany have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Germany 484 9 23 111 60 490 487

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Germany: in reading literacy and schools…
...where students
have the highest
• 23% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 13% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Germany have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Germany have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Germany OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Germany 0.19 22 -11 2.8 60 16 156 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Germany 95 96 96 79 79 13 35 35 10 4 11 2
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

38 © OECD 2004 © OECD 2004 39

 3. Student characteristics
PISA 2000 Profile for Greece Relationship between student performance
Engagement at school
In Greece: in reading literacy and schools…
...where students
• 23% of students have a low sense of belonging, have the highest
sense of belonging
compared to 25% on average in OECD countries.
1. Student performance • 29-% of students have low participation (attendance), ...where students have
Positive correlation

highest participation
In reading literacy compared to 20% on average in OECD countries.
Mean score in Mean score
% at reading level Correlation -0.2 0 0.2 0.4 0.6 0.8 1
Mean Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy
Greece 474 5 24 97 50 447 461
4. School characteristics
OECD 500 9 18 100 35 500 500 Climate
Compared to other OECD students, Performance advantage in schools with
students from Greece have: a more positive climate...
less favourable Disciplinary climate
…in terms of student-
2. Socio-economic status (SES) Teachers’ morale and related factors
more favourable
The socio-economic gradient commitment
PISA score points 0 20 40 60 80 100 120
Teacher-related factors
less favourable
Reading score Reading level affecting the school climate

Resources
600 Level
IV Compared to other OECD students, Performance advantage in schools with...
students from Greece have: ...better educational
Quality of the schools’ resources
Level lower
physical infrastructure PISA score points -20 0 20 40 60 80 100
500 III
more Teacher shortage

Level
II Greece OECD average

400
Level
I
5. System characteristics
School autonomy
300 Below
Level I Percentage of students attending schools with at least some responsibility for:
Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
-3 -2 -1 0 1 2 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
Socio-economic status policies allocation used policies sions budget offered content teachers teachers increases salaries
Greece 97 95 90 94 90 87 89 92 65 70 77 73
OECD 95 94 92 89 84 76 71 69 61 54 26 23
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Greece -0.25 16 11 3.3 38 13 93
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

40 © OECD 2004 © OECD 2004 41

 3. Student characteristics
PISA 2000 Profile for Hungary Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Hungary have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Hungary 480 5 23 94 67 488 496

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Hungary: in reading literacy and schools…
...where students
have the highest
• 19% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 18% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Hungary have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Hungary have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
Hungary OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Hungary -0.11 26 7 2.9 54 6 106 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Hungary 100 92 100 98 99 61 98 97 100 99 50 41
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

42 © OECD 2004 © OECD 2004 43

 3. Student characteristics
PISA 2000 Profile for Iceland Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Iceland have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Iceland 507 9 15 92 8 514 496

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Iceland: in reading literacy and schools…
...where students
have the highest
• 22% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 26% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Iceland have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Iceland have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Iceland OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Iceland 0.69 7 -15 2.8 24 20 29 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Iceland 99 87 99 98 74 76 62 79 99 99 7 4
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

44 © OECD 2004 © OECD 2004 45

 3. Student characteristics
PISA 2000 Profile for Ireland Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Ireland have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Ireland 527 14 11 94 18 503 513

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Ireland: in reading literacy and schools…
...where students
have the highest
• 19% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 18% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Ireland have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Ireland have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I close to average Teacher shortage

-3 -2 -1 0 1 2
Ireland OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Ireland 0.02 14 0 2.9 38 28 79 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Ireland 99 100 100 99 95 79 97 37 88 73 5 4
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

46 © OECD 2004 © OECD 2004 47

 3. Student characteristics
PISA 2000 Profile for Italy Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Italy have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Italy 487 5 19 91 54 457 478

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Italy: in reading literacy and schools…
...where students
have the highest
• 23% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 22% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Italy have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
less favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Italy have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Italy OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Italy 0.09 11 -2 3.1 32 5 99 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Italy 100 57 100 100 63 94 22 93 10 11 1 1
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

48 © OECD 2004 © OECD 2004 49

 3. Student characteristics
PISA 2000 Profile for Japan Engagement at school Relationship between student performance
In Japan: in reading literacy and schools…
...where students
• 38% of students have a low sense of belonging, have the highest
sense of belonging
compared to 25% on average in OECD countries.
Positive correlation
• 4% of students have low participation (attendance), ...where students have
1. Student performance compared to 20% on average in OECD countries.
highest participation

In reading literacy Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Mean score in Mean score
Mean % at reading level Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy
Japan 522 10 10 86 45 557 550 4. School characteristics
Climate
OECD 500 9 18 100 35 500 500 Performance advantage in schools with
Compared to other OECD students,
students from Japan have: a more positive climate...
more favourable Disciplinary climate …in terms of student-
related factors
Teachers’ morale and
more favourable
2. Socio-economic status (SES) commitment PISA score points 0 20 40 60 80 100 120
The socio-economic gradient Teacher-related factors
more favourable
affecting the school climate
Reading score Reading level
Resources
Performance advantage in schools with...
600 Level Compared to other OECD students,
...better educational
IV students from Japan have: resources
Quality of the schools’
lower PISA score points -20 0 20 40 60 80 100
physical infrastructure
Level
less Teacher shortage
500 III

Japan OECD average

Level
II
400
Level
I 5. System characteristics
School autonomy
300 Below Percentage of students attending schools with at least some responsibility for:
Level I Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
-3 -2 -1 0 1 2 policies allocation used policies sions budget offered content teachers teachers increases salaries
Socio-economic status Japan 100 91 99 100 100 50 98 99 33 32 32 32
OECD 95 94 92 89 84 76 71 69 61 54 26 23

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Japan -0.40 8 8 2.6 24 3 124
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

50 © OECD 2004 © OECD 2004 51

 3. Student characteristics
PISA 2000 Profile for Korea Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Korea have:
their learning
confidence in their own ...with interest
below average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Korea 525 6 6 70 37 547 552

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Korea: in reading literacy and schools…
...where students
have the highest
• 41% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 8% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Korea have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
less favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Korea have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
Korea OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Korea -0.31 9 8 2.9 23 7 68 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Korea 100 95 99 99 97 88 93 99 32 22 7 15
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

52 © OECD 2004 © OECD 2004 53

 3. Student characteristics
PISA 2000 Profile for Luxembourg Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Luxembourg have:
their learning
confidence in their own ...with interest
below average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
close to average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Luxembourg 441 2 35 100 31 446 443

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Luxembourg: in reading literacy and schools…
...where students
have the highest
• 28% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 13% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
OECD Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Luxembourg have: a more positive climate...
Level
III more favourable Disciplinary climate
500 Luxembourg …in terms of student-
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Luxembourg have: ...better educational
resources
Quality of the schools’
lower
300 OECD countries Below physical infrastructure PISA score points -20 0 20 40 60 80 100
non- OECD countries Level I close to average Teacher shortage

-3 -2 -1 0 1 2
Luxembourg OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2
School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Luxembourg -0.40 26 7 3.4 46 26 110 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
Luxembourg 100 100 100
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students. OECD 95 94 92 89 84 76 71 69 61 54 26 23
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

54 © OECD 2004 © OECD 2004 55

 3. Student characteristics
PISA 2000 Profile for Mexico Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Mexico have:
their learning
confidence in their own ...with interest
above average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Mexico 422 1 44 86 53 387 422

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Mexico: in reading literacy and schools…
...where students
have the highest
• 22% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 21% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Mexico have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Mexico have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Mexico OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Mexico -1.24 23 38 4.4 35 7 54 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Mexico 99 77 81 92 86 68 58 59 57 48 28 26
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

56 © OECD 2004 © OECD 2004 57

 3. Student characteristics
PISA 2000 Profile for New Zealand Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from New Zealand have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
below average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.
New
529 19 14 108 16 537 528
Zealand Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In New Zealand: in reading literacy and schools…
...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 27% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from New Zealand have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from New Zealand have: ...better educational
resources
Quality of the schools’
close to average
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
New Zealand OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
New disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Zealand
0.16 17 -6 3.1 45 34 83 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41 New
Zealand
100 100 100 100 94 98 100 87 100 99 41 17
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

58 © OECD 2004 © OECD 2004 59

 3. Student characteristics
PISA 2000 Profile for Norway Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Norway have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Norway 505 11 17 104 11 499 500

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Norway: in reading literacy and schools…
...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 18% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Norway have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Norway have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Norway OECD average
Socio-economic status

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Norway 0.52 14 -17 2.9 42 38 60
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

60 © OECD 2004 © OECD 2004 61

 3. Student characteristics
PISA 2000 Profile for Poland Engagement at school Relationship between student performance
In Poland: in reading literacy and schools…
...where students
• 41% of students have a low sense of belonging, have the highest
compared to 25% on average in OECD countries. sense of belonging
Positive correlation
• 29% of students have low participation (attendance), ...where students have
1. Student performance compared to 20% on average in OECD countries. highest participation

In reading literacy Mean score in Mean score Correlation -0.2 0 0.2 0.4 0.6 0.8 1
Mean % at reading level Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy
Poland 479 6 23 100 63 470 483 4. School characteristics
Climate
OECD 500 9 18 100 35 500 500 Performance advantage in schools with
Compared to other OECD students,
students from Poland have: a more positive climate...
more favourable Disciplinary climate …in terms of student-
related factors
Teachers’ morale and
less favourable
2. Socio-economic status (SES) commitment PISA score points 0 20 40 60 80 100 120
The socio-economic gradient Teacher-related factors
close to average
affecting the school climate
Reading score Reading level
Resources
Performance advantage in schools with...
600 Level Compared to other OECD students,
...better educational
IV students from Poland have: resources
Quality of the schools’
lower PISA score points -20 0 20 40 60 80 100
physical infrastructure
Level
less Teacher shortage
500 III

Poland OECD average

Level
II
400
Level
I

300 Below
Level I

-3 -2 -1 0 1 2
Socio-economic status

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Poland -0.35 17 16 3.2 38 3 105
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

62 © OECD 2004 © OECD 2004 63

 3. Student characteristics
PISA 2000 Profile for Portugal Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Portugal have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Portugal 470 4 26 97 37 454 459

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Portugal: in reading literacy and schools…
...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 20% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Portugal have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
less favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II less favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Portugal have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I close to average Teacher shortage

-3 -2 -1 0 1 2
Portugal OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Portugal -0.41 20 17 3.6 41 22 86 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Portugal 92 95 100 88 85 89 54 20 13 9 1 1
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

64 © OECD 2004 © OECD 2004 65

 3. Student characteristics
PISA 2000 Profile for Spain Engagement at school
Relationship between student performance
In Spain: in reading literacy and schools…
...where students
• 24% of students have a low sense of belonging, have the highest
compared to 25% on average in OECD countries. sense of belonging
Positive correlation
1. Student performance • 34% of students have low participation (attendance), ...where students have
compared to 20% on average in OECD countries. highest participation
In reading literacy Mean score in Mean score Correlation -0.2 0 0.2 0.4 0.6 0.8 1
Mean % at reading level Standard deviation of % of variation mathematical in scientific
score 5 1 or below reading literacy scores between schools literacy literacy
Spain 493 4 16 85 21 476 491 4. School characteristics
OECD 500 9 18 100 35 500 500 Climate
Compared to other OECD students, Performance advantage in schools with
students from Spain have: a more positive climate...
less favourable Disciplinary climate
…in terms of student-
Teachers’ morale and related factors
2. Socio-economic status (SES) less favourable
commitment
The socio-economic gradient PISA score points 0 20 40 60 80 100 120
Teacher-related factors
more favourable
affecting the school climate
Reading score Reading level
Resources
600 Level Compared to other OECD students, Performance advantage in schools with...
IV students from Spain have: ...better educational
resources
Quality of the schools’
higher
Level physical infrastructure PISA score points -20 0 20 40 60 80 100
500 III less Teacher shortage

Level Spain OECD average

II
400
Level
I 5. System characteristics
School autonomy
300 Below
Percentage of students attending schools with at least some responsibility for:
Level I
Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
-3 -2 -1 0 1 2
policies allocation used policies sions budget offered content teachers teachers increases salaries
Socio-economic status
Spain 99 98 100 97 89 90 54 86 38 39 9 9
OECD 95 94 92 89 84 76 71 69 61 54 26 23

Socio-economic status of participating students Features of the socio-economic gradient

Difference in reading literacy Slope of the gradient2
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between
status performance (score points) gradient1 Overall schools schools
Spain -0.24 17 12 3.3 32 20 55
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.

66 © OECD 2004 © OECD 2004 67

 3. Student characteristics
PISA 2000 Profile for Sweden Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Sweden have:
their learning
confidence in their own ...with interest
above average
learning efficacy.
1. Student performance in reading
confidence in their own
close to average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Sweden 516 11 13 92 10 510 512

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Sweden: in reading literacy and schools…
...where students
have the highest
• 18% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 24% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Sweden have: a more positive climate...
Level
III less favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Sweden have: ...better educational
resources
Quality of the schools’
close to average
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
Sweden OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Sweden 0.36 11 -12 2.7 36 30 69 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Sweden 100 99 100 97 54 85 76 88 99 83 74 62
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

68 © OECD 2004 © OECD 2004 69

 3. Student characteristics
PISA 2000 Profile for Switzerland Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Switzerland have:
their learning
confidence in their own ...with interest
close to average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
below average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.

Switzerland 494 9 20 102 43 529 496

Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In Switzerland: in reading literacy and schools…
...where students
have the highest
• 21% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 16% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from Switzerland have: a more positive climate...
Level
III more favourable Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
more favourable
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II more favourable
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from Switzerland have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
Switzerland OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Switzerland 0.01 19 6 3.0 49 26 99 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
Switzerland 98 87 51 75 82 54 34 29 93 82 15 13
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

70 © OECD 2004 © OECD 2004 71

 3. Student characteristics
PISA 2000 Profile for United Kingdom Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from Scotland have:
their learning
confidence in their own ...with interest
above average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.
United
523 16 13 100 21 529 532
Kingdom Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In United Kingdom: in reading literacy and schools…
...where students
have the highest
• 17% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 15% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from United Kingdom have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from United Kingdom have: ...better educational
resources
Quality of the schools’
lower
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I more Teacher shortage

-3 -2 -1 0 1 2
United Kingdom OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
United disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
Kingdom
0.11 20 -4 2.9 49 32 94 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41 United
Kingdom
99 100 100 100 66 92 100 94 99 89 70 72
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

72 © OECD 2004 © OECD 2004 73

 3. Student characteristics
PISA 2000 Profile for United States Approaches to learning Performance advantage in reading
literacy of students...
Compared to other OECD students,
...who control
students from United States have:
their learning
confidence in their own ...with interest
above average
learning efficacy.
1. Student performance in reading
confidence in their own
above average ...who believe in
In reading literacy reading ability.
Mean score in Mean score their own efficacy
Mean % at reading level Standard deviation of % of variation mathematical in scientific confidence in their own
above average PISA score points 0 20 40 60 80 100 120
score 5 1 or below reading literacy scores between schools literacy literacy mathematical ability.
United
504 12 18 105 30 493 499
States Engagement at school Relationship between student performance
OECD 500 9 18 100 35 500 500 In United States: in reading literacy and schools…
...where students
have the highest
• 25% of students have a low sense of belonging, sense of belonging
compared to 25% on average in OECD countries. Positive correlation
...where students have
• 20% of students have low participation (attendance),
2. Socio-economic status (SES) compared to 20% on average in OECD countries.
highest participation

The socio-economic gradient Correlation -0.2 0 0.2 0.4 0.6 0.8 1

Reading score Reading level

600
4. School characteristics
Level
Climate
IV
Compared to other OECD students, Performance advantage in schools with
students from United States have: a more positive climate...
Level
III close to average Disciplinary climate …in terms of student-
500
Teachers’ morale and related factors
close to average
commitment
Level PISA score points 0 20 40 60 80 100 120
Teacher-related factors
II close to average
affecting the school climate
400 Resources
Level
I Compared to other OECD students, Performance advantage in schools with...
students from United States have: ...better educational
resources
Quality of the schools’
higher
300 Below physical infrastructure PISA score points -20 0 20 40 60 80 100
Level I less Teacher shortage

-3 -2 -1 0 1 2
United States OECD average
Socio-economic status

5. System characteristics
Socio-economic status of participating students Features of the socio-economic gradient
Difference in reading literacy Slope of the gradient2 School autonomy
Mean socio- Percentage of explained score if students had the
economic variation in student average OECD SES Length of the Within Between Percentage of students attending schools with at least some responsibility for:
status performance (score points) gradient1 Overall schools schools Student Student
United disci- assess- Student Formulat- Dismiss- Teachers’ Teachers’
States
0.17 21 -6 3.3 48 29 92 plinary Budget Textbooks ment admis- ing school Courses Course Appointing ing salary starting
policies allocation used policies sions budget offered content teachers teachers increases salaries
OECD 0.00 20 3.0 41 United
States
99 99 92 93 89 96 97 84 97 98 74 76
1. This shows the socio-economic variation of the middle 90% of students and therefore the gap between the 5 per cent most
disadvantaged and the 5 per cent most advantaged students.
2. Score point difference associated with one unit increase of SES. Steeper slopes indicate a greater inequality.
OECD 95 94 92 89 84 76 71 69 61 54 26 23

74 © OECD 2004 © OECD 2004 75

 IN THE SAME SERIES…

PISA 2000 INITIAL REPORTS

Knowledge and Skills for Life: First Results from PISA 2000
Literacy Skills for the World of Tomorrow: Further Results from PISA 2000

PISA 2000 THEMATIC REPORTS

Reading for Change: Performance and Engagement across Countries: Results from PISA 2000
Learners for Life: Student Approaches to Learning: Results from PISA 2000
Student Engagement at School: A Sense of Belonging and Participation: Results from PISA 2000
What Makes School Systems Perform?: Seeing School Systems through the Prism of PISA
Reviews of National Policies for Education: Denmark: Lessons from PISA 2000
School Factors related to Quality and Equity (February 2005)

PISA 2000 ASSESSMENT INFORMATION

Measuring Student Knowledge and Skills: The PISA 2000 Assessment of Reading, Mathematical and
Scientific Literacy
Sample Tasks from the PISA 2000 Assessment: Reading, Mathematical and Scientific Literacy
Programme for International Student Assessment (PISA): PISA 2000 Technical Report
Programme for International Student Assessment (PISA): Manual for the PISA 2000 Database

PISA 2003 INITIAL REPORTS

Learning for Tomorrow’s World: First Results from PISA 2003
Problem Solving for Tomorrow’s World: First Measures of Cross-Curricular Competencies from PISA 2003

PISA 2003 THEMATIC REPORTS

Mathematical Literacy: Student Performance and Engagement (December 2005)
Teaching and Learning Strategies (April 2006)

PISA 2003 ASSESSMENT INFORMATION

The PISA 2003 Assessment Framework: Mathematics, Reading, Science and Problem Solving Knowledge
and Skills
Programme for International Student Assessment (PISA): PISA 2003 Technical Report (September 2005)
Programme for International Student Assessment (PISA): Manual for the PISA 2003 Database (April 2005)

76 © OECD 2004 © OECD 2004 77

 OECD PUBLICATIONS, 2, rue André-Pascal, 75775 Paris Cedex 16
PRINTED IN FRANCE
(00 2004 5 M 1) No. 82929 2004

78 © OECD 2004 © OECD 2004 79