How should the CEFR be used by
recognising institutions wishing to set language ability
requirements?
In fulfilling its purpose as a common
reference tool, the CEFR was not designed to provide the basis for
precise equating, nor was it intended to be a prescriptive tool to
impose standardised solutions. Rather it was designed as a common
framework of reference, primarily intended as ‘a tool for
reflection, communications and empowerment’, as described by John
Trim, its coordinating author (Saville, N 2005).
Therefore, we would recommend that all
recognising institutions should look at the IELTS bandscore
descriptors and use the IELTS Scores
Explained DVD to ascertain the appropriate level
of language ability required for their institution or course.
Figure 1: The mapping of the
IELTS scale to the Common European Framework above is derived from
the interrelationship between IELTS and the Cambridge English
Language Assessment Main Suite qualifications and the mapping of
these latter qualifications to the CEFR. Further information
on this can be found at www.cambridgeesol.org/what-we-do/research/cefr/index.html
General information
Making comparisons between scores on different
tests is challenging because many of the current range of test
products differ in their design, purpose, and format (Taylor,
2004a). Candidates’ aptitude and preparation for a particular
type of test may also vary and individual candidates or groups of
candidates may perform better in certain tests than in others.
Specifying the relationship between a test
product and the CEFR is challenging because, in order to function
as a framework, the CEFR is deliberately underspecified (Davidson
& Fulcher, 2007; Milanovic, 2009; Weir, 2005). Establishing the
relationship is also not a one-off activity, but rather involves
the accumulation of evidence over time (e.g. it needs to be shown
that test quality and standards are maintained).
Cambridge ESOL has been working since the
1990s to refine its understanding of the relationship between its
different assessment products, including IELTS, and the CEFR. The
relationship of IELTS with the CEFR is complex as IELTS is not a
level-based test, but rather designed to span a much broader
proficiency continuum. It also utilises a different 9-point band
scoring system; thus, there will not be a one-to-one correspondence
between IELTS scores and CEFR levels. It is important to bear in
mind the differences in test purpose, test format, test
populations, and measurement scales when seeking to make
comparisons.
With the above in mind, Cambridge English
Language Assessment has conducted a number of research projects
since the late 1990s to explore how IELTS band scores align with
the CEFR levels. A number of these were summarised in Taylor
(2004b), while cautioning that,
“As we grow in our understanding of the
relationship between IELTS and the CEFR levels, so the frame of
reference may need to be revised accordingly.”
Note that the IELTS band scores referred to in
Figure 1 above are the overall band scores, not the individual
module band scores for listening, reading, writing and speaking. It
is important to recognise that the purpose of this figure is to
communicate the relationship between IELTS and the CEFR. They
should not be interpreted as reflecting strong claims about exact
equivalence between assessment products or the scores they
generate, for the reasons given in Taylor (2004a).
The current alignment is based upon a growing
body of internal and external research, some of which has also
appeared in peer-reviewed academic journals, attesting to their
quality (e.g. Hawkey & Barker, 2004). This research has been
further combined with long established experience of test use
within education and society, as well as feedback from a range of
stakeholders regarding the uses of test results for particular
purposes.
As further work, such as that being
undertaken in the English
Profile project, enriches our understanding of the CEFR levels,
further refinements may be possible.
1. Has the IELTS test
changed?
No, the test has not been changed.
2. Why is IELTS changing the way the
band scores relate to the CEFR?
We have always been committed to providing
ongoing revision as we grow in our understanding of the
relationship between IELTS, other examinations and the CEFR
levels.
The CEFR is becoming more prominent in how
institutions consider language ability requirements. It is
important therefore that we provide updated advice as to how to
interpret IELTS scores in CEFR terms. The table previously on the
website did not show half band scores, and predated the
introduction of half-band reporting for Writing and Speaking in
July 2007.
3. Has IELTS been made more
difficult?
No the test has not changed. The way it is
examined and the way band scores are awarded remain the same.
4. Should institutions and
organisations which use IELTS scores change the band scores they
expect students to achieve as a result of the revised CEFR
mapping?
The test has not changed and the performance
represented by each band score remains the same. The IELTS
Scores Explained DVD provides samples of those performances so that
institutions can judge what level is appropriate to their needs.
There is no need for institutions to make changes where they have
previously been satisfied with their particular score
requirements.
5. Some IELTS band scores are shown as
borderline (e.g. it is not clear whether band 5 is B1 or B2). How
should institutions and organisations interpret this?
As IELTS preceded the CEFR, IELTS band scores
have never aligned exactly with the CEFR transition points.
The new table makes this clearer. Previously we provided advice as
to the score on IELTS that a candidate who was at a given CEFR
level might achieve. However, our research shows that a C1 minimum
threshold would fall between the 6.5 and 7 bands on the IELTS
scale. Therefore, whilst many 6.5 candidates would be at C1, a
number will be marginally below. So if an institution requires a
high degree of confidence that an applicant is at C1, they may wish
to set a requirement of 7, rather than 6.5.
6. Does IELTS differentiate at C2
level?
Band scores of 8.5 and higher are recognised
as C2. Band 8 is borderline.
7. If a student already has an IELTS
score of 6.5, shown as C1 in the previous mapping, should this now
be treated as a B2 equivalent score?
6.5 is borderline B2/C1. The real world level
of performance represented by the result has not changed. It is for
institutions to decide whether they wish to change their
requirements, if alignment to a particular level of the CEFR is
critical (see response to q5 above). The advice in the
IELTS Guide for Educational Institutions as to probable levels
required for different types of course still holds.
8. Should institutions and
organisations that offer English courses to prepare students for
university study, or to facilitate university study, change the
format, content or level of their courses?
No. Nothing within the test content has
changed.
9. What is the research behind these
new mappings?
This is a response to the increased prominence
of the CEFR in how institutions consider language ability
requirements, rather than the findings of a particular research
project. The new presentation draws on the previous evidence,
on benchmarking exercises conducted in 2009, and on studies of the
performance of candidates for other Cambridge English exams at B2
and C1 level on IELTS-type materials in 2009 and 2010.
10. How does this compare to the
mappings that other language testers have published?
We do not comment on the benchmarking
exercises that other language testers have provided.
References
Council of Europe (2001) The
Common European framework of reference for languages:
Learning, teaching, assessment, Cambridge: Cambridge
University Press.
Davidson, F & Fulcher, G
(2007) 'The Common European Framework of Reference and the design
of language tests: A matter of effect', Language Teaching,
40, 231-241.
Hawkey, R & Barker, F (2004)
'Developing a common scale for the assessment of writing',
Assessing Writing, 9(3), 122-159.
Milanovic, M (2009) 'Cambridge
ESOL and the CEFR', Research Notes, 37, 2-5.
Saville, N (2005) 'An interview
with John Trim at 80', Language Assessment Quarterly, 2
(4), 263–288.
Taylor, L (2004a) 'Issues of test
comparability', Research Notes, 15, 2-5.
Taylor, L (2004b) 'IELTS,
Cambridge ESOL examinations and the Common European
Framework', Research Notes, 18, 2-3.
Weir, C J (2005) 'Limitations of
the Common European Framework for developing comparable
examinations and tests', Language Testing, 22,
281-300.
