Washback

Washback is the influence a test exerts on the teaching and learning that comes before it. It is positive when preparing for the exam drives genuinely useful language work, and negative when it narrows a course to test technique. The term is standard in language testing, where it is also called backwash.

How washback works

Every exam sends a message about what counts. Teachers read that message, learners read it, coursebook publishers read it, and the course bends towards it. That bending is washback. The term borrows its picture from the wake of a boat: the exam is ahead, and the disturbance it creates runs back over everything behind it.

The important point, and the one most often skipped, is that washback is not a property of a test. It is a property of what people do about a test. Two schools preparing candidates for the same exam can teach in completely different ways, and both can be responding rationally to the same paper. Alderson and Wall made this the centre of their argument in 1993: a test does not act on a classroom directly, it acts through teachers who interpret it, and through the beliefs, training, materials and time they already have.

That gives washback a fairly simple internal logic:

• What the test samples determines what the course treats as the syllabus. Anything reliably absent from the paper tends to disappear from the classroom.
• What the test format rewards determines the shape of practice. A paper answered by choosing between four options invites practice in choosing between four options.
• How much is at stake determines the intensity. Low-stakes progress tests produce mild washback. A gatekeeping exam that decides university entry or a visa produces very strong washback.
• Who is teaching determines the direction. The same paper produces useful work in one room and past-paper drilling in the next.

Hughes set out the practical consequence for anyone writing a test in Testing for Language Teachers: if you want beneficial backwash, test the abilities you actually want encouraged, sample the syllabus widely rather than predictably, test directly where you can, and make sure teachers and learners understand what the test is asking for. A test that can be beaten by technique will be beaten by technique.

Messick put the same thing in validity terms in 1996. Washback, in his account, is one of the consequences by which a test is judged, and evidence of washback is only evidence about the test if the classroom difference can be traced to the construct the test measures rather than to the fact that an exam exists at all. It is a demanding standard, and it is why careful writers are cautious about claiming a particular exam caused a particular teaching habit.

A worked example: two B2 exam classes, one term

Commonly confused

Washback is not	The actual difference
Impact	Scope. Washback is the effect of a test on teaching and learning in the classroom. Impact is the broader effect on schools, institutions, families and education systems. Washback sits inside impact.
Teaching to the test	Teaching to the test is one possible outcome of washback, usually the negative one. Washback is the mechanism; teaching to the test is a behaviour that mechanism can produce. A course can show strong washback and never drill a past paper.
Formative assessment	Direction and timing. Formative assessment feeds information forward into the next lesson deliberately. Washback runs backwards from a future test and is largely unplanned. See formative and summative assessment in the glossary.
Backward design	Intention. Backward design is a deliberate method of planning a course from its intended outcomes. Washback is what a test does to a course whether anyone planned it or not. See backward design.
A judgement on the test	Negative washback does not by itself make a test bad, and positive washback does not make it good. A well designed exam can be taught badly, and a crude exam can sit under an excellent course.
Test anxiety	Washback describes what gets taught and practised. Exam pressure on the individual learner belongs with the affective filter, though the two often appear together in an exam class.

Where the idea came from, and the honest state of the evidence

The observation is old and the research literature is not. Teachers have complained about exams distorting courses for as long as there have been exams, and the British term backwash was in general educational use well before language testing adopted it. What changed in the 1990s is that people started looking for the effect instead of assuming it.

Alderson and Wall’s 1993 article in Applied Linguistics is the turning point, and its title is a question rather than a claim: “Does washback exist?” Working on the introduction of a new English examination in Sri Lanka, they found that the exam changed what teachers taught but had far less effect on how they taught it. They set out fifteen washback hypotheses to force the field to be specific about what was being asserted, since “washback” had been used to mean anything from a change of textbook to a change of national policy. Wall returned to the same territory at book length in 2005.

Bailey’s 1996 review pulled the concept into a usable model separating washback to the learner from washback to the programme. Prodromou had already put the classroom case in ELT Journal in 1995. The collection edited by Cheng, Watanabe and Curtis in 2004 remains the standard survey of methods, and it is where the useful reminder appears that washback varies in specificity, intensity and intentionality, so asking whether a test “has washback” is the wrong question.

The criticism belongs in a definition rather than in a footnote. The concept is genuinely hard to research: it is very difficult to separate the effect of a test from the effect of the teacher, the textbook, the training, the timetable and the general seriousness with which a course is treated. Green’s 2007 study of IELTS preparation found the picture messier than the folk version, with exam courses and ordinary academic writing courses overlapping more than expected. The reasonable conclusion is the one Alderson and Wall reached: washback is real, it is uneven, and any claim that a specific exam caused a specific classroom habit needs evidence rather than assertion.

For teachers the practical value survives all of that. Knowing the mechanism is what lets you look at a marking scheme and ask which parts describe an ability worth building and which parts describe a paper worth rehearsing. The Council of Europe’s framework is useful here precisely because it describes what a learner can do at each level rather than what a paper contains, which gives an exam course something to aim at beyond the paper.

Sources

Source	What it supports here
Alderson, J. C. and Wall, D. (1993). “Does washback exist?”. Applied Linguistics, 14(2).	The founding question, the fifteen washback hypotheses, and the finding that exams affect what is taught more than how.
Hughes, A. (2003). Testing for Language Teachers, 2nd edition. Cambridge University Press.	The practical conditions for beneficial backwash, including direct testing and wide sampling.
Messick, S. (1996). “Validity and washback in language testing”. Language Testing, 13(3).	Washback as a consequence of validity, and the standard of evidence needed to attribute it to a test.
Bailey, K. M. (1996). “Working for washback: a review of the washback concept in language testing”. Language Testing, 13(3).	The distinction between washback to the learner and washback to the programme.
Prodromou, L. (1995). “The backwash effect: from testing to teaching”. ELT Journal, 49(1).	The classroom account of how exam pressure reshapes an ordinary course.
Cheng, L., Watanabe, Y. and Curtis, A. (eds) (2004). Washback in Language Testing: Research Contexts and Methods. Lawrence Erlbaum.	The dimensions of washback, including specificity, intensity and intentionality.
Wall, D. (2005). The Impact of High Stakes Examinations on Classroom Teaching. Studies in Language Testing 22, Cambridge University Press.	The book length treatment of the Sri Lankan case and the innovation theory reading of it.
Green, A. (2007). IELTS Washback in Context. Studies in Language Testing 25, Cambridge University Press.	The evidence that exam preparation and ordinary academic writing courses overlap more than assumed.
Council of Europe, Common European Framework of Reference for Languages	The can-do description of proficiency that gives an exam course a target beyond the paper.
British Council, TeachingEnglish, Assessing learning	The professional practice area in which assessment literacy, and washback with it, sits for teachers.

Print sources are cited in full without hyperlinks because they are books and subscription journals rather than open web pages. The two linked sources were checked and confirmed live on 9 September 2026.

Frequently asked questions

What is washback?

Washback is the influence a test exerts on the teaching and learning that comes before it. It is positive when preparing for the exam drives genuinely useful language work, and negative when it narrows a course to test technique. It is also called backwash.

What is the difference between positive and negative washback?

Positive washback is when preparing for a test produces learning that is worth having anyway, such as wider reading or better paragraphing. Negative washback is when preparation crowds out that learning in favour of exam technique, such as memorised essay templates or dropping speaking because it carries fewer marks.

Is washback the same as teaching to the test?

No. Teaching to the test is one behaviour that washback can produce, usually the unhelpful kind. Washback is the underlying mechanism by which an exam influences a course, and it can just as easily produce a better course.

What is the difference between washback and impact?

Scope. Washback describes what a test does to classroom teaching and learning. Impact describes the wider consequences for schools, learners, families and education systems. Washback is the classroom part of impact.

Can a teacher control washback?

Partly, and more than most teachers assume. Alderson and Wall found that exams act through teachers rather than directly on classrooms, which is why two groups sitting the same paper can be taught very differently. Reading a marking scheme as a description of an ability rather than as a checklist is the single most useful move.

How do you design a test with positive washback?

Hughes gives the standard advice: test the abilities you want people to develop, sample the syllabus widely so that nothing can be safely ignored, test directly rather than through proxies where the format allows, and make sure teachers and learners understand what is being asked of them.

If you are building an exam course and want a lesson structure to adapt, the tefl.ai lesson plan generator will draft one. For the exam itself rather than the teaching around it, The TEFL Institute keeps a full guide to what IELTS is and how the exam works, and its reference page on TEFL qualifications and terminology covers the certification acronyms. For how an assessed TEFL course runs inside the Irish school system, the TEFL Institute of Ireland sets out its Transition Year TEFL programmes.

Last updated 9 September 2026.