Skip to main content

AQA controlled speaking assessments...

... or as we used to call them "orals".

We are in the process of having a first go at these with our end of Y10 students (15 year-olds). I've got the top set. They have learned their material on school very thoroughly and are performing well.

If you aren't familiar with the format we are using England and Wales, students are given a list of bullet points on a topic which they can prepare in advance in class. One of the bullet points is an exclamation mark which indicates that the teacher will ask a surprise question. Prior to this list being handed out the teacher will already have spent at least two weeks teaching to the topic and getting students to write exercises and so on on the theme. During the teaching sequence the teacher is allowed to correct students' errors. The children will have memorised their answers and practised them in pairs in class. They then do their 4-6 minute dialogue with teacher, with the aid of a sheet of notes (up to 40 words, no conjugated verbs), the oral being recorded digitally. You can do as many as you want and the best two marks are submitted for moderation.

What has struck me so far is that this form of test relies more than ever on effective memorisation. With about seven bullet points and a surprise question 6 minutes get filled easily, but the large majority of it is a memorised speech. The nature of the test makes this inevitable. Good students have always memorised language for orals, but with this format it feels more pre-learned than ever. This is because you are doing one topic only and the bullet points are all clearly decided on beforehand, so there is almost no scope (or time) for improvisation.

I do think this is a shame. It would be desirable to be able to add improvised follow-up questions after each bullet point. The dialogue would be more natural and candidates would have to think on their feet a bit more. This format gives weaker students a good chance and it allows the top ones to say a lot, but not to show how they have internalised grammar rules for creative language use.

Alternatively you could ditch the question-answer format and get candidates to talk about pictures or do an information gap task. I was not wild about the old role play tasks, but at least they had to be done on the spot and could not be memorised in advance.

There is no easy solution to the problem of how to assess oral proficiency. What we now have may take some pressure off students by letting them have several attempts to prove themselves. Or for some it may multiply the pressure as they try their very best each time. (That may be the case for some I teach, even though I try to play down the importance of the test.)

Incidentally the assessment format is open to all sorts of abuse. With the enormous pressure on teachers to churn out good results some will, regrettably, give unfair support to candidates, whilst some students will get more home support than others.

I am sure of this: teachers must not spend all their time teaching students to memorise chunks of language. With better students at least, they must continue to use classroom time to get students to listen a lot, develop grammatical skill and do genuinely creative communicative activities. What AQA have produced (courtesy of a higher body, no doubt) is profoundly uncreative. Students will be happy with the ultra-safe format, but the assessment system will not have done much to encourage real fluency. Pity.


  1. Yes, the emphasis is definitely on memorizing chunks of language.

    Apart from the unprepared question, which requires students to listen and respond spontaneously (although I'm sure that this question will be one of several questions for which students will have already prepared answers) the main things being tested are (a) memory and (b) pronunciation. Accuracy is pretty much irrelevant - that came into play when the students drew up their prepared statements (and there's plenty of scope for these statements to be corrected beforehand).

    Listening, another key component of speaking skills in most situations, is assessed separately - granted.

    But it's the spontaneity bit that isn't really tested. And I would argue that it is this - the ability to understand what is said to you and then formulate an appropriate response - that is the key component of speaking.

    But how do we test the spontaneity part of speaking? The recently dropped role-plays didn't do it either. The older style GCSE orals based on telling a story based on picture prompts were a good test for higher levels. And what about the O level oral - where the student could be asked all sorts of random questions, with nothing prepared in advance (and assessment by external examiners) - where the oral mark counted for much less (10% ?) of the overall mark.

    I wonder if we shouldn't have tried to follow the model used by the Cambridge exams to assess the level of spoken English of thousands of EFL learners all over the world. The exam format differs from level to level - and GCSE could be like PET.

    And while we're at it, why not have an actual grammar element to the exam, just as with the Cambridge EFL exams - "Use of English" is a test of grammar, vocabulary and usage.

  2. Maybe there's a case for tiered orals, with higher ability candidates having tasks which demand more spontaneity.

    I wonder to what extent this new format was pre-tested, as it were. I would also like to know how it evolved and to what extent AQA and the other boards have had to implement it reluctantly.

    Thanks for the comment, Martin.


Post a Comment

Popular posts from this blog

5 great zero preparation lesson ideas

When the pressure is on and there are only so many hours on the week, you need a repertoire of zero preparation go-to activities which promote input and/or practice. Here are five you might well find useful.

1. My weekend

We know that listening is the most important yet often neglected skill for language learning. It's also something some pupils find hard to do. To develop listening skill and provide tailored comprehensible input try this:

You tell the class you are going to recount what you did last weekend and that they have to make notes in English. The amount of detail you go into and the speed you go will depend on your class. Talk for about three minutes. If you spent the whole weekend marking, you can always make stuff up!

You then make some true or false (maybe not mentioned too) statements in the target language about what you said in your account. Class gives hands up (or no hands up) answers. This can then lead into a simple pair work task where pupils make up their own tru…

What teachers are saying about The Language Teacher Toolkit

"The Language Teacher Toolkit is a really useful book for language teachers to either read all the way through or dip into. What I like about it is that the authors Steve Smith and Gianfranco Conti are totally upfront about what they believe to be good practice but back it up with research evidence." (Ernesto Macaro, Oxford University Department of Education)

"I absolutely love this book based on research and full of activities..  The best manual I've read so far. One of our PDs from the Australian Board of Studies recommended your book as an excellent resource.  I look forward to the conference here in Sydney." Michela Pezzi, Teacher, Australia, Facebook)

"Finally, a book for World Language teachers that provides practical ideas and strategies that can actually be used in the classroom, rather than dry rhetoric and theory that does little to inspire creativity in ways that are engaging for both students and teachers alike." (USA teacher, Amazon review)

New GCSE resources on frenchteacher

As well as writing resources for the new A-levels, I have in recent months been posting a good range of materials to support the new GCSEs. First exams are not until 2018, but here is what you can find on the site in addition to the many other resources (grammar exercises, texts, video listening etc).

I shall not produce vocabulary lists since the exam board specifications now offer these, with translations.

Foundation Tier 

AQA-style GCSE 2016 Role-plays
AQA-style GCSE 2016 Photo card conversations
AQA-style GCSE 2016 Photo card conversations (2)
100 translation sentences into French (with answers)
Reading exam
Reading exam (2)
How to write a good Foundation Tier essay (ppt)
How to write a good Foundation Tier essay (Word)

Higher Tier 

AQA-style GCSE 2016 Photo card conversations (Higher tier)
AQA-style GCSE 2016 Photo card conversations (Higher tier) (2)
20 translations into French (with answers)
Reading exam (Higher tier)
How to write a good Higher Tier essay (ppt)
How to write a…

Three AQA A-level courses compared

I've put together my three reviews of worthy A-level courses which you might be considering for next September. They are all very useful courses, but with significant differences. The traditional Hodder and OUP book-based courses differ in that the former comes in one chunky two year book, whilst OUP's comes in two parts, the first for AS or the first year of an A-level course. The Attitudes16 course by Steve Glover and Nathalie Kaddouri is based on an online platform from which you would download worksheets and share a logon with studenst who would do the interactive parts (Textivate and video work). The two text books are supported by interactive material (Kerboodle) or an e-text book.


An excellent resource which should be competing for your attention at the moment is the Attitudes16 course which writers Steve Glover and Nathalie Kaddouri have been working on for some time. You can find it here at, along with his excellent resources for film and li…

Learning strategies (3)

This is the third in the mini-series of blogs about learning strategies. So far, we have looked at some (rather scant) research evidence for the effectiveness of strategies. Bear in mind that a lack of research evidence does not mean strategies do not work; if there is any consensus, it is that they are probably useful and probably best used when integrated into a normal teaching sequence. We then looked at a classification of different types of strategies.

In this blog Gianfanco and I look at how you might integrate strategies into your teaching. There is nothing revolutionary about this stuff! You may do a good deal of this type of thing already, but you may also be new to the concepts and applications of learning strategies.

Let's look at how you might use strategies, particularly with regard to the teaching of listening and reading. Remember: this is just about how you help students to use strategies to become better listeners and readers.

How to teach strategies 

The research …