ef epi-s - ef education first level a1 a2 b1 b2 c1 c2 ef score 20 40 60 80 100 reading listening...

www.ef.com/epi 2015

EF EPI-sEF English Proficiency Index for Schools

TEST YOUR STUDENTS

FOR FREE:

The EF Standard English Test

See page 17

EFSET

TABLE OF CONTENTSTABLE OF CONTENTS

Executive Summary

Acquisition of English Skills

Skill Snapshots at Ages 15 and 20

Skill Differences in Reading and Listening

In Focus: Latin America

In Focus: Russia

Conclusions

Appendix: CEFR Levels and Can-Do Statements

Join Our Research

03

05

07

09

11

13

15

16

17

2

Test your students

for free at

efset.org/schools

EXECUTIVE SUMMARY

The EF English Proficiency Index for Schools (EF EPI-s) is a study of the acquisition of English skills by secondary and tertiary students. This first edition of the study measures the English reading and listening proficiency levels of 130,000 students enrolled at hundreds of partner schools and universities in 16 countries.

Although most school systems teach English, assessment of English as a foreign language is mainly accomplished at a national or local level using ad hoc tests designed by central administrators or teachers. An internationally standardized analysis of English skill acquisition in schools across grade levels and over time would be invaluable for educators and policymakers, but resources for accomplishing such research have been lacking.

This report is a companion to our annual EF EPI report, which evaluates adult English proficiency levels around the world based on a separate set of test data. This EF EPI-s report is our first data-driven look at English learning in schools around the world.

EF English Proficiency Index for Schools

3

KEY FINDINGS

This report looks at our most compelling

findings about English language acquisition

during secondary and tertiary education and

discusses how this research can be taken

further by educators and policymakers, and

how it may be applied to improve learning

outcomes. The EF EPI-s does not intend to

rank countries by English proficiency as the

primary EF EPI report does.

Below are our key findings:

• Listening skills develop more quickly,

outpacing reading skills throughout

secondary and tertiary education. As

students mature, this gap usually narrows,

although not in all countries.

• There is a wider range of variability in

English listening skills than in reading

skills. This may be the result of more

exposure to spoken English than to written

English outside the classroom.

• Students do not improve their English at

a steady rate. In some countries, students

improve enormously in lower secondary

school but make little progress in later

years. In other countries, learning is steady

all the way through university.

• The EF EPI has shown every year that, in

most countries, women speak English

better than men do. This gender gap is also

present in students as young as 13.

• English abilities vary between peers in

different cities, between public and private

school students, and between university

students in different fields.

LANGUAGE ASSESSMENT IN SCHOOLS

The primary aim of this report is to show

educators how they might use standardized

English language assessment tools to

improve English learning outcomes. To date,

internationally recognized standardized

English tests have been too expensive

for use as a continuous assessment tool.

Also, these tests have not been designed

to track progress over time. Individuals

take standardized English tests as one-off

evaluations to certify competency for a third

party, such as an overseas university or

immigration authority, but school systems

rarely use them.

The English test we used for our research

is different. The EF Standard English Test

(EFSET) suite was designed to the same

exacting standards as the TOEFL or IELTS,

but this test is free, scalable, and built to

track students’ English skill acquisition

over time.

Because the EFSET is online and free, an

entire city, region, or country can evaluate all

of its students every year using this test, for

only the cost of coordinating the effort. With

the data from that testing effort, educators

will be able to understand how their students’

English skills evolve over time, how changes

in the curriculum affect outcomes, which

schools need extra resources, and how their

students compare to similarly aged students

across the country and around the world.

METHODOLOGY

As with the main EF EPI report, our sampling

was voluntary. Any school wishing to

participate in this research was allowed to do

so. Some schools tested all their students,

others only a single class. In some countries,

we had participating institutions from lower

secondary through tertiary education, while

in others we only tested students at a single

level of instruction.

This report does not claim to be

representative of any country’s overall English

proficiency among students of a particular

age. Creating an international ranking is not

the aim of this research. Rather, we hope

to present the trends in English language

acquisition that we find interesting within this

data set.

ABOUT EF EDUCATION FIRST

EF Education First (www.ef.com) is an

international education company that focuses

on language, academics, and cultural

experience. Founded in 1965, EF’s mission

is “opening the world through education.”

With 500 schools and offices in more than 50

countries, EF is the Official Language Training

Supplier of the Rio 2016 Olympics. The EF

English Proficiency Index (EF EPI) and the EF

English Proficiency Index for Schools

(EF EPI-s) are published by EF Learning Labs,

the research and innovation division of EF

Education First. EF has recently released the

EF Standard English Test (EFSET), the world's

first free standardized English test.

4

Test your students

for free at

efset.org/schools


ACQUISITION OF ENGLISH SKILLS

Academic performance in subject areas other than English is measured in schools around the world every three years when the Organisation for Economic Co-operation and Development (OECD) publishes the Programme for International Student Assessment (PISA). PISA examines 15-year-old students’ performance in math, reading, and science in the local language. Except in English-speaking countries, English skills are not tested.

Taking a snapshot of students' skills at a given age allows comparisons among countries;

however, our research shows that it is only by tracking students' skill development over time

that educators can improve outcomes. Looking at the English teaching curriculum in most

countries, we expect to see a steady rate of improvement in English proficiency every year. After

all, students are receiving similar amounts of instruction each year, they are taught by teachers

with the same types of qualifications, and they follow a curriculum that is structured around

steady progress. But that ideal of a steadily rising level of English proficiency is far from reality.

Barring Italy, the English skills of students around the world do not progress steadily

throughout lower secondary, upper secondary, and university education. These varying rates of

progress are not the same among countries either, even among those within the same region.

The difference in the speed of acquisition of English skills is striking within any given country.

For example, Polish students improve their English 50% faster in lower secondary school than

in upper secondary school, and their improvement in upper secondary is more than three

times faster than at the university level. In Brazil, students improve rapidly in upper secondary

school but not at all in university.

It is possible that education systems focus more on the productive English skills of speaking

and writing during certain portions of the curriculum, which would account for this variation

in acquisition speed. We did not measure the productive skills. It is also possible that changes

in the number of hours of instruction in English might result in faster or slower progress in

particular grades. However, if there is no clear curricular explanation for the differences in

acquisition speed, the question remains: why is the rate of English acquisition

so variable?

Continuous assessment at a national level would allow for a better understanding of this

variation, and for a deeper analysis of its causes. A larger data set would also allow analysis

of individual students’ learning trajectories. If students are attending English classes but not

improving their English skills, a national standardized testing regimen would show the scope

of the problem, and, more importantly, allow educators to assess the effectiveness of reforms

undertaken to address the issue.

5

Test your students

for free at

efset.org/schools

6

RATE OF IMPROVEMENT IN ENGLISH

Belgium

Brazil

Chile

China

Colombia

Costa Rica

France

Hong Kong

Indonesia

Italy

Kazakhstan

Mexico

Poland

Russia

Spain

Sweden

0 1 2 3 4 5 6 7

Annual Increase in EF Score

Ages 13-15

Ages 16-18

Ages 19-21

This graph shows the rate of English skill acquisition by comparing the composite EF score change year over year in three different age groups. There are two important things to note in order to interpret this graph correctly:

1. It is easier for students in the beginner levels to improve than for students in the more advanced levels; therefore, a slower rate of acquisition at higher skill levels is to be expected. This graph, however, gives no indication of students' English level. It only shows their rate of improvement.

2. Our data set does not contain students who took our test multiple times over a prolonged period. For example, in all graphs comparing students by age, we are comparing a group of 15-year-olds to a different group of 16-year-olds, rather than comparing the same students at different ages.


AVERAGE EF SCORE AT AGE 15

Students around the world start learning English at different ages. In all of the 16 countries studied, English instruction has started by the end of primary school; in some countries, it begins several years earlier. By the end of lower secondary school, after five or more years of English instruction, students are on average at the B1 or B2 level in English listening skills and the A2 or B1 level in English reading skills.

SKILL SNAPSHOTS AT AGES 15 AND 20

Belgium

Brazil

Chile

Colombia

Costa Rica

France

Hong Kong

Indonesia

Italy

Mexico

Poland

Russia

Spain

Sweden

EF Score

20 40 60 80 100

CEFR Level

A1 A2 B1 B2 C1 C2

Reading Listening

This graph shows the average reading and listening scores for all students age 15 in our data set.

7

Test your students

for free at

efset.org/schools

8

AVERAGE EF SCORE AT AGE 20

As expected, university students have a stronger command of English than 15-year-olds. However, even in the highest-performing countries in this study, average listening skills are still below the C1 level. Entrance into university is a selective process in every country. Regardless of whether English skill evaluation is part of that selection, English proficiency among 20-year-old university students is higher than that among all 20-year-olds.

Belgium

Brazil

Chile

China

Colombia

France

Italy

Kazakhstan

Mexico

Poland

Russia

CEFR Level

A1 A2 B1 B2 C1 C2

EF Score

20 40 60 80 100

Reading Listening

This graph shows the average reading and listening scores for all students age 20 in our data set.


SKILL DIFFERENCES IN READING AND LISTENING

In almost all of the 16 countries studied, listening skills develop more quickly and outpace reading skills for students of all ages. Comparing the distribution of reading and listening skills among 18-year-olds in all countries surveyed, we find that the range of variability in listening scores is broader. This could be due to different individual levels of exposure to spoken English outside the classroom via media consumption and international travel. Exposure to written English is perhaps not as common outside the classroom.

The graph below compares the average reading and listening scores that would be expected

from an 18-year-old, based upon a linear growth model. There is a wide range of variability

for both expected reading and listening scores across the 16 countries studied, as shown

by the overlapping graphs. The corresponding bell curves model the distribution of scores

across countries. The global average listening score is 11.5 points higher than the global

average reading score.

DISTRIBUTION OF READING AND LISTENING SCORES AT AGE 18

Reading Listening

EF Score

Num

ber

of 1

8-ye

ar-o

lds

CEFR Level

A1 A2 B1 B2 C1 C2

20 40 60 80 100

9

This graph shows how reading and listening scores are distributed for 18-year-old students from all 16 countries covered by the survey. It allows us to look at the range of ability among students of this age group.

Test your students

for free at

efset.org/schools

10

Although listening skills are stronger than reading skills in almost all

countries surveyed and at all grade levels, in some countries we see

the gap between reading and listening skills narrowing as students

mature. Russia is the only exception; there, reading and listening

skills are equally matched in all age groups.

In Poland, Spain, and Sweden, the gap between reading and

listening narrows during upper secondary school. Listening skills

develop rapidly during lower secondary school in these countries,

while reading skills develop more slowly. Reading skill development

accelerates during upper secondary school as listening skill

development slows down, meaning students catch up in the weaker

skill. This shift in emphasis is part of the curriculum design in

these countries.

Belgium demonstrates a similar trend, but the acceleration in

reading ability occurs at the university level rather than in secondary

school. Although Belgian university students are still stronger in

listening than in reading, their skills are more balanced than they

were at the start of secondary school.

France and Italy demonstrate the opposite trend, with reading

and listening skills starting off at similar levels at age 13 and then

diverging. Listening skills develop far more quickly than reading skills

throughout secondary school in these two countries. The skill gap

peaks near the end of secondary school and remains high throughout

university. Stronger listening skills make university students in

France and Italy more comfortable consuming English-language

media, but their relatively weak reading skills will hamper their ability

to study or work in an English-speaking environment.

DIFFERENCE BETWEEN READING AND LISTENING ABILITY (EUROPE)

10

15

20

25

14 16 18 20 22

Age

0

-5

5

Sweden

Belgium

Spain

Poland

France

Italy

Russia

EF Score Difference

This graph shows the gap between reading skills and listening skills in Europe. The values were calculated by subtracting the average reading score for a given age group from the average listening score for that same age group.


IN FOCUS: LATIN AMERICA

A broader data set and a wide range of participating schools in Latin America allow us to look at some aspects of English language acquisition that we cannot analyze across all 16 countries surveyed.

In the EF EPI, we find every year that on average women speak

English better than men do worldwide. However, that gender gap

is not uniform across countries. In Mexico, for example, adult men

and women have similiar levels of English proficiency. In our study of

secondary and university students in Mexico, we find that 13-year-old

girls have English skills that are, on average, five months ahead of

those of 13-year-old boys. This skills gap closes by the time students

reach university. The causes of the initial gap and the mechanisms of

its closing are worth studying.

GENDER GAP (MEXICO)

14 16 18 20 22Age

30

40

50

60

70

80

EF Score

Girls

Boys

11

This graph shows how male and female students' English ability develops over time. The values graphed are the composite reading and listening EF scores.

Test your students

for free at

efset.org/schools

12

DIFFERENCE BETWEEN READING AND LISTENING ABILITY (LATIN AMERICA)

10

15

20

25

30

14 16 18 20 22Age

5

EF Score Difference

PLATEAU EFFECT (LATIN AMERICA)

Age

30

40

50

60

70

80

14 16 18 20 22

Mexico Mexico

Brazil Costa Rica

Chile Colombia

Chile

Brazil

In many countries, including Brazil, Chile, and Mexico, English

learning appears to stop well before the end of formal education.

That is not to say that English instruction stops after a certain grade

level, but students do not seem to be making progress in English.

This plateau, beginning at age 18 or 19, indicates that universities

are not making measurable improvements in their students’

English skills.

There are significant opportunities for universities or particular

degree programs to use standardized assessment to understand

the areas in which their students need more English training.

Specifically, English training at the university level in the student's

field of study is invaluable to prepare them for the job market.

EF Score

In all Latin American countries surveyed, the gap between reading

and listening skills is so wide that students' reading proficiency

is more than one year behind their listening proficiency. In four of

the five countries surveyed in this region, this skills gap remains

constant from lower secondary school to university.

Mexico is strikingly different. Thirteen-year-olds in Mexico have

strong English listening skills, both in an absolute sense and as

compared to other countries. Even at age 15, Mexican students

have better English listening skills than their peers in Sweden.

However, Mexican students’ English listening skills stagnate

through secondary school, while their reading skills improve

steadily. By the time they are 20, Mexican students no longer

stand out internationally for their English skills—not because they

have grown weaker, but because students in other countries have

improved more.

This graph shows the composite reading and listening EF score for each age group in three Latin American countries. In all three countries, there is no measurable change in English skills beyond age 18.

This graph shows the gap between reading and listening skills in Latin America. The values were calculated by subtracting the average reading score for a given age group from the average listening score for that same age group. In the case of Costa Rica, we did not have adequate data beyond age 19 to complete the graph.


IN FOCUS: RUSSIA

Thanks to our collaboration with dozens of Russian universities, the breadth of the data collected allows us to analyze various demographic factors.

A comparison of schools by ownership type, specialization, and size,

is valuable in understanding trends that are broader than individual

schools but narrower than an entire country. For this first EF EPI-s

report, we only gathered sufficient data to compare public and

private university students in Russia. University students in Russia

improve their English skills on average by four points per year. By

that measure, 20-year-old business students in private universities in

Moscow are two years ahead of their peers in public universities.

PRIVATE UNIVERSITIES VS. PUBLIC UNIVERSITIES(MOSCOW)

Private University

Public University

30

40

50

60

70

80

EF Score

68.8B2

59.2B1

13

This graph shows the CEFR level and composite EF reading and listening scores for a 20-year-old business major enrolled in a university in Moscow. By aligning age, major, and city, we are able to measure the public/private skill gap more accurately.

Test your students

for free at

efset.org/schools

14

We lack sufficient data to explain whether the gap in English skills

between public and private university students is due to the different

selection processes of public and private universities in Russia,

or whether it is a result of differences in instruction at the

universities themselves. Universities can deepen the analysis

by testing students upon admission and then again each year

throughout their degree programs.

We also compared several major Russian cities for which we have

sufficient data to perform a profile-based analysis. When looking

at a specific student profile, we see large differences in English

proficiency between cities. Do these differences in skill level reflect

the secondary education systems in these cities? Or do they reflect

the selectiveness of the universities that participated in our study?

It is impossible to tell without a broader data set, but this finding is

in accordance with our study of adults in Russia, where skill level

gaps between cities are also significant.

CITY COMPARISONS II(STEM MAJORS)

CITY COMPARISONS I (BUSINESS MAJORS)

Moscow Novosibirsk KazanSt. Petersburg Moscow Novosibirsk30

40

50

60

70

80

EF Score

30

40

50

60

70

80

EF Score

64.6 B2

59.2 B1

47.7B1 43.6

A242.3A2 36.8

A2

This graph shows the composite reading and listening EF scores for a 20-year-old business major enrolled in a public university in one of three cities. Again, aligning several variables allows us to recognize variations in others with more confidence.

This graph shows the composite EF scores for an 18-year-old STEM student enrolled at a public university in three cities.

CONCLUSIONS

Teachers, parents, and students all hope

that high school graduates are prepared to

enter tertiary education and the workforce

smoothly. In many cases, that transition now

requires effective English communication

skills. Teaching those skills has become an

explicit goal in many secondary and tertiary

curricula around the world.

Although nearly all schools teach English

in some form, precise assessment tools are

hard to find, and they are often prohibitively

expensive. As English proficiency is

increasingly linked to developmental targets,

professional success, and global knowledge

sharing, educators need these assessment

tools more than ever.

Tracing how students develop their English

skills over time must be a priority. Many

education systems already use some form of

standardized testing to evaluate performance

in English. However, few test individual

students over time to understand how

language acquisition is occurring, and even

fewer compare those scores with peers in

other countries.

Based on our research, we have the following

recommendations:

• Across grade levels, adopt comparable

assessment standards that track English

learning progress. Making standardized

assessments across grade levels accessible

helps individual students and their parents

track their progress year by year. It also

gives educators the data necessary to see

how students are learning and adjust the

curriculum accordingly. Testing English in

a consistent way throughout an education

system is a powerful tool for improvement,

both on an individual and on a national level.

• Establish English proficiency as a core

competency and test accordingly. The

requirement of a specific English proficiency

level upon completion of a specific phase

of formal education concentrates efforts

on meeting that requirement. Teachers,

parents, and students can all be galvanized

to see English proficiency as a key

component to academic success.

• Align English instruction to ensure smooth

transitions between stages of learning.

Students benefit from a coherent teaching

regime that is clear and consistent. At

different ages and stages, students need

different types of English instruction.

From primary schools to universities to

professional training and beyond, better

coordination helps educators design

curricula that promote synergies, build on

acquired skills, and avoid repetition.

• Promote balance between different English

skills. There is a marked disparity between

many students' reading and listening

proficiency. All-around competence in

English is more powerful than development

of individual skills in isolation. A balanced

curriculum will build oral English skills

early, while children are still developing

their own native language proficiency, then

build written English skills and vocabulary

to support the academic performance of

older students.

• Teach English after secondary school as

well. English instruction in tertiary and

vocational instruction builds the English

skills young adults need for specific

professional purposes. The targeted skills

and vocabulary acquired during these years

are essential to operational readiness when

entering the workforce.

• Compare students to peers in other

countries. International resources and

expertise are readily available to meet

specific needs. Individual school systems

have their own strengths and weaknesses

and there is much to be gained by sharing

knowledge. Only an internationally

standardized testing platform allows such

detailed comparison.

While this report does not claim to represent

the overall English proficiency of students

from any country or age group, it explores

broad trends in English language acquisition

in secondary and tertiary education and hints

at the types of analysis an even broader data

set would allow.

Further research can evaluate students’

English speaking and writing skills, which

this report does not cover. Speaking is a

particularly valuable skill at lower levels

of proficiency, while writing is essential for

academic and professional contexts. With

a broader data set and more demographic

information, further research can look more

closely at the disparities between genders,

ages, cities, public and private schools,

disciplines, and skills.

Our research team encourages education

authorities and school administrators to

leverage the full potential of the English

testing tool set developed for this research

to improve outcomes in English language

instruction. We are ready to support countries,

regions, and individual schools that would like

to use the EF Standard English Test (EFSET)

to assess their students.

15

APPENDIX: CEFR LEVELSAND CAN-DO STATEMENTS

Can understand with ease virtually everything heard or read. Can summarize information from

different spoken and written sources, reconstructing arguments and accounts in a coherent

presentation. Can express him/herself spontaneously, very fluently, and precisely, differentiating

finer shades of meaning even in more complex situations.

Can understand a wide range of demanding, longer texts, and recognize implicit meaning. Can ex-

press him/herself fluently and spontaneously without much obvious searching for expressions. Can

use language flexibly and effectively for social, academic, and professional purposes. Can produce

clear, well-structured, detailed text on complex subjects, showing controlled use of organizational

patterns, connectors, and cohesive devices.

Can understand the main ideas of complex text on both concrete and abstract topics, including

technical discussions in his/her field of specialization. Can interact with a degree of fluency and

spontaneity that makes regular interaction with native speakers quite possible without strain for

either party. Can produce clear, detailed text on a wide range of subjects and explain a viewpoint on

a topical issue, giving the advantages and disadvantages of various options.

Can understand the main points of clear standard input on familiar matters regularly encountered

in work, school, leisure, etc. Can deal with most situations likely to arise while traveling in

an area where the language is spoken. Can produce simple connected text on topics which

are familiar or of personal interest. Can describe experiences and events, dreams, hopes, and

ambitions and briefly give reasons and explanations for opinions and plans.

Can understand sentences and frequently used expressions related to most relevant areas

(e.g., very basic personal and family information, shopping, local geography, employment). Can

communicate during routine tasks requiring a simple and direct exchange of information

on familiar matters. Can describe in simple terms aspects of his/her background, immediate

environment, and matters in areas of immediate need.

Can understand and use familiar everyday expressions and very basic phrases aimed at the

satisfaction of needs of a concrete type. Can introduce him/herself and others and can ask and

answer questions about personal details such as where he/she lives, people he/she knows and

things he/she has. Can interact in a simple way provided the other person talks slowly and clearly

and is prepared to help.

QUOTED FROM THE COUNCIL OF EUROPE

All countries in the EF EPI fell into bands corresponding to levels A2-B2.

No countries had average scores placing them at either the lowest level, A1, or at one of the highest two levels, C1 and C2.

PROFICIENT USER C2

B2

B1

A2

A1

INDEPENDENT USER

BASIC USER

C1

16

Test your students

for free at

efset.org/schools


JOIN OUR RESEARCH

This EF EPI-s report details what we are able to uncover about English language acquisition in secondary and tertiary education. More importantly, it gives an idea of what it would be possible to analyze with an even broader data set collected using the same tools.

Continuous assessment of English language skills using a standard set of assessment tools allows us to pinpoint areas for improvement and reveal successful strategies at the institutional, national, and international levels. We invite all schools, universities, and ministries of education throughout the world to participate in our research.

Participating institutions will have access to the EF Standard English Test (EFSET). Offered at no cost and built to the same standards as other standardized tests, the EFSET rests on a foundation of evidence-based research and analysis. Test items were created by experienced exam writers, carefully reviewed by a panel of experts, and piloted on more than 150,000 learners from 80 countries.

Upon completion of testing, participating schools receive customized reports with their students’ EFSET scores and CEFR levels and comparisons between groups of students.

17

I N T E R E S T E D I N I M P L E M E N T I N G S T A N D A R D I Z E D T E S T I N G ?

T E S T Y O U R S T U D E N T S F O R F R E E A T W W W . E F S E T . O R G / S C H O O L S

ef epi-s - ef education first level a1 a2 b1 b2 c1 c2 ef score 20 40 60 80 100 reading listening...

Documents