Learning Outcomes, School Quality and

Inequality in Vietnam, India, Peru and Ethiopia

Caine Rolleston

Young Lives longitudinal survey of children, households & communities every 3 years since 2002

• 12,000 index children• Two cohorts now aged

13 and 19• Ethiopia, India, Peru,

Vietnam• 20 sentinel sites in each

country• Include comparable

learning assessments of maths and literacy

YOUNG LIVES HOUSEHOLD STUDY

• School surveys since 2010• Focus on learning & learning progress, school

and teacher effectiveness• Index children and their class peers • Total 3,284 pupils in Grade 5 in Vietnam, 176

classes in 91 school sites• Longitudinal test design reflect curricular

expectations at the beginning and the end of the school year

• School, class observations, teacher pedagogical content tests, attitude questionnaires

• Allows estimation of school quality

SCHOOL SURVEYS

GNI IS HIGHEST IN PERU AND VERY SIMILAR IN VIETNAM AND INDIA

19951996

19971998

19992000

20012002

20032004

20052006

20072008

20092010

0

1000

2000

3000

4000

5000

6000

7000

8000

PeruVietnamIndiaEthiopia

GNI per capita 1995-2010 ($ 2000, PPP)

HOUSEHOLD TEST DATA: AT AGE 5 COGNITIVE DEVELOPMENT LEVELS ARE SIMILAR,

ESPECIALLY BETWEEN VIETNAM AND INDIA

Mean Score (%)

BY AGE 8 CHILDREN IN VIETNAM OUTPERFORM OTHER COUNTRIES IN MATHS. The POOREST IN VIETNAM

PERFORM BETTER THAN ALMOST ALL OTHERS

Mean (Household) Maths Test Scores at Age 8 (%)

BY AGE 15 THE GAP IN MATHS PERFORMANCE BETWEEN VIETNAM AND INDIA IS VERY LARGE

India Vietnam0

2

4

6

8

10

12

14

16

18

20

Site-level average maths score at age 14-15

020

4060

80M

edia

n M

aths

Sco

re R

3 %

0 20 40 60 80 100CDA-Q Score R2 %

Ethiopia PeruIndia Vietnam

WIDENING GAPS ARE DRIVEN BY DIFFERENCES IN LEARNING PROGRESS OVER TIME BETWEEN SYSTEMS:

AGE 5 TO 8

PROGRESS AGES 12 TO 15

WHILE PUPILS IN VIETNAM KEEP UP WITH THE CURRICULUM, IN INDIA THEY FALL PROGRESSIVELY BEHIND

Pupils’ abilities remain in-line with the curriculum in Vietnam but the curriculum in India is progressively ‘over-ambitious’

(pupils fail to progress according to expectations)

Indicator Vietnam India Mean class size

27.61 16.23

Mean years of teacher experience

17.47 7.71

Mean monthly teacher salary (USD/Month)

164

226

% of teachers with no formal teacher training qualification

0% 16.50%

Teacher absenteeism

2.34 days per year 35.12% pupils said ‘ my class teacher often does not come to school’

All children have access to maths textbooks

96.16% 60.84%

Teacher always checks/marks maths homework

41.28% 18.06%

VIETNAM AND INDIA: SCHOOL-SYSTEM QUALITY INDICATORS (FROM SCHOOL SURVEYS)

THE REASON FOR BETTER PERFORMANCE OVERALL IN VIETNAM LIES PARTLY IN LOWER LEVELS OF

INEQUALITYHIGH AVERAGE PERFORMANANCE LINKED TO LOW

DISPERSION

TEST SCORES VARY BETWEEN SCHOOLS IN PERU AND VIETNAM BUT ARE LESS DISPERSED IN

VIETNAM

05

1015

20N

umbe

r of S

choo

ls

300 400 500 600 700Mean Maths Score

School Mean Maths Scores, Vietnam

05

1015

20N

umbe

r of S

choo

ls

300 400 500 600 700Mean Maths Score

School Mean Maths Scores, Peru

School test scores scaled to mean 500 and standard deviation 100

THE RELATIONSHIP BETWEEN TEST SCORES AND HOME BACKGROUND IS STRONGER IN PERU

THAN VIETNAM40

045

050

055

060

0M

aths

Sco

re

0 .25 .5 .75 1Household Wealth Index

95% CI VietnamPeru

Maths Performance By Wealth Index

Maths Vietnamese

First Test Second Test Gain First Test Second Test Gain

Kinh 508.74 549.77 41.03 511.41 513.32 1.90

Ethnic Minority

438.55 503.66 65.12 419.76 481.31 61.55

Difference 70.19*** 46.11*** 24.08*** 91.65*** 32.01*** 59.65***

Total 500.00 544.03 44.03 500.00 509.33 9.33

IN VIETNAM, ETHNIC MINORITY PUPILS PERFORM LESS WELL THAN KINH, BUT NO EVIDENCE THAT THE GAP WIDENS DUE TO SCHOOLING IN G5

Vietnamese Mathematics

There are 2 main ways through which school quality may have differential impacts on pupils’ attainments.

Differential school effectiveness along SES

DIFFERENT GROUPS OF CHILDREN MAY BENEFIT FROM SCHOOL QUALITY DIFFERENTLY

BETWEEN SCHOOLS

Are poorer children accessing lower quality schools?

(selection into schools)

WITHIN SCHOOLS

Are poorer children benefiting less from school quality than

their more advantaged counterparts?

UNEQUAL ACCESS: PRIVATE SCHOOL ATTENANCE IS ACCELERATING IN RURAL INDIA WITH A

WIDENING GENDER GAP

-100

-50

050

100

Sch

ool V

alue

-Add

ed (9

0% C

I)

0 10 20 30 40 50School Value-Added Rank

Which Schools Add More Value in G5?

Not particularly more advantaged pupils

Slightly better physical resources

Not better teacher subject knowledge

More teachers with degrees

More positive teacher attitudes e.g.

“The influence of a student’s home experience can be overcome by good teaching”

Teachers more often evaluated

School Value-Added: Learning progress attributable to schools after removing prior attainment and background effects

SOME SCHOOLS ARE MORE EFFECTIVE THAN OTHERS IN VIETNAM, BUT THIS IS NOT STRONGLY LINKED TO PUPILS’

BACKGROUNDS

CHILDREN’S HOME BACKGROUNDS EXPLAIN MORE OF THE VARIANCE IN ATTAINMENT IN INDIA AND

PERU THAN IN VIETNAMControlling for pre-school scores, children’s home backgrounds (at age 5) explain much more of the variation in test scores (at age 11) in Peru than in Vietnam or India

In maths, backgrounds account for a large proportion of the variance in Peru and much less in Vietnam

Systems where background effects are large are arguably more inequitable - ‘reproducing’ home advantage/disadvantage Vietnam India Peru

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Mathematics Language

Prop

ortio

n of

Var

ianc

e in

Tes

t Sco

res E

xpla

ined

SCHOOLS EXPLAIN MORE OF THE VARIANCE IN ATTAINMENT IN INDIA AND PERU COMPARED WITH VIETNAM

• ‘School fixed effects’ capture school-level factors (school quality), controlling for pupil backgrounds, pre-school test scores

• School quality in India and Peru accounts for more of the variance in test scores than in Vietnam

• School systems in Peru and India more heterogeneous, school a child attends appears to matter more than in VietnamVietnam India Peru

0

0.1

0.2

0.3

0.4

0.5

0.6

Mathematics Language

Prop

ortio

n of

Var

ianc

e in

Tes

t Sco

res E

xpla

ined

Vietnam Peru-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

-0.18

0.41*

Proportion of 1 SD of maths test score distribution

Who benefits from an increase in school quality? – compare effect on richest 40% to poorest 60% (separate FE)

In Vietnam, schools are equally effective in teaching Maths to children irrespectively of their background.

In Peru, by contrast, schools significantly less effective at teaching children from disadvantaged backgrounds

WITHIN THE SAME SCHOOL, DISADVANTAGED PUPILS MAKE LESS PROGRESS IN PERU, BUT NOT

IN VIETNAM

POSITIVES• Almost all schools have basic facilities -

electricity, toilets, text books, basic learning materials etc.

• Overall differences across sites on basic quality indicators are small

• Pupils from disadvantaged sites/backgrounds make good progress on the curriculum in G5

• Disadvantaged pupils attend schools with lower levels of some assets but are often in smaller classes

• The relationship between school quality and pupils’ backgrounds is fairly weak

• Primary schooling in Vietnam is relatively equitable by comparison

1 2 3 4 5 6 7 8 9 10400

450

500

550

600

650

Teacher Reported Test Score

Youn

g Li

ves T

est S

core

TEACHERS IN VIETNAM KNOW WHAT PUPILS KNOW

(AND NEED TO KNOW)

CAVEATS• Large differences in home background advantage between

sites • Large differences in achievement between the most (urban

Da Nang) and least advantaged sites (mountainous Lao Cai)• Notable differences in test score by ethnicity (Kinh/minority)• Larger differences on more sophisticated school resources

(i.e. library, internet, computers), particularly between Da Nang and the other sites

• Pupils in more advantaged sites receive more periods of teaching per week

• Effects of extra classes etc. difficult to account for• Gaps at entry to Grade 5 require further efforts to equalise

learning very early on.

DROP-OUT BETWEEN AGES12 AND 15 IS HIGHEST IN VIETNAM

• High drop-out at completion of junior secondary stage (age 14)

• Higher among boys• May suggest PISA results are an overestimate

39%

PUPILS IN VIETNAM PERFORM VERY WELL ON EQUATIONS AND ALGEBRA INCLUDING USING COMPLEX FRACTIONS …BUT POORLY ON SIMPLER? PROBLEM SOLVING QUESTIONS (SUGGESTING ROTE LEARNING)

33%

Equity-oriented centralised public school system• Less evidence that disadvantaged pupils attend lower

quality schools• Less evidence that schools are less effective for

disadvantaged pupilsHigh-performance for the majority linked to equity orientation• Emphasis on ‘fundamental’ or minimum school quality

levels (especially in disadvantaged areas) • Common curricula & text books in use matched closely to

pupils’ learning levels• Commitment to ‘mastery’ by all pupils - use of regular

assessment by teachers• Teacher knowledge (YL curriculum tests) is similar

between more and less disadvantaged areas, absenteeism is low across almost all schools

WHAT IS DIFFERENT ABOUT THE VIETNAMESE SYSTEM?

caine.rolleston@qeh.ox.ac.uk

www.younglives.org.uk

FINDING OUT MORE