What about dictation in the new GCSE exam?

Image: lumni.fr

This is my third post about aspects of the proposed GCSE subject content for MFL. I enjoy being able to reflect on these matters and hope my ramblings might stimulate your thinking too. Last time I wrote about one eyebrow-raising aspect of the proposals, the inclusion of a reading aloud test. Today I thought I would reflect on another surprising inclusion: dictation. Why is it there? Is it a valid and reliable means of assessment? Does it have any task "authenticity"? What could be its backwash effect on classroom teaching?

A little context first. From around the late 1970s, with the communicative language teaching movement beginning to have an influence on MFL teaching in England, dictation fell out of fashion. Why? Well, to start with, dictation fell into the category of those tasks which are not 'communicative', i.e. don't involve the face-to-face communication of meaning through task discussion, information gaps, question and answer, etc. Dictation, like translation, felt for many like a thing of the past and something to avoid.

Secondly, dictation was often perceived as a difficult exercise, set up to make students fail. This was often the case, even if many pupils said they actually enjoyed the activity (probably the ones with most aptitude).

Now, the reality is that many teachers continued to do dictation in various forms, either quite formally (read a passage as pupils write down their transcription phrase by phrase, with a score being added up at the end), or in other formats, for instance the transcription of sentences and chunks within the context of work on listening or reading texts. Some teachers will have valued dictation in terms of its effect on classroom behaviour - it's a pretty calming activity, apparently passive, but involving a lot of mental activity. It's a working memory feast!

In recent times, dictation has gained more traction in MFL circles, and transcription features as a significant part of the lexicogrammar (EPI) methodology recommended by Gianfranco Conti and shared with my help and encouragement. It is also used by teachers using a Knowledge Organiser-style method as supported by Barry Smith and used in quite a few schools. Many teachers use occasional transcription or dictation tasks as part of their hybrid approach.

It's easy to cite research in support of transcription/dictation. It fosters close listening, development of phonics skill (sound-spelling correspondences), phonological awareness in general (feeding into pronunciation), spelling and, of course, grammatical skill - or more precisely morphological skill since the challenges, particularly in French, surround inflections and agreements.

In terms of test validity (does it measure what it purports to measure?) - that's a bit tricky, since dictation clearly measures skill in the range of areas I have just mentioned. Whereas reading aloud is very much about showing you can decode symbols and turn them into accurate speech (as well as, it must be mentioned, knowing you can recall the pronunciation of whole words and chunks - which is a different thing), dictation is harder to pin down, but I'm going to guess that it's there primarily to reinforce the phonics strand of the TSC/NCELP recommendations. Is it a valid measure of phonics skill? Yes.

In reliability terms, it's easy to grade the accuracy of transcribed sentences, though there will be the usual mark scheme challenges, e.g. are we looking for a right-wrong mark scheme (most objective) or a level-based grid (e.g. no errors, few errors, some errors, many errors). The latter can be less reliable.

What will be the backwash effect of having some dictation in a paper (we can assume it will be part of a listening or listening/reading/writing paper)? Clearly most teachers will spend more time than now on transcription tasks. I don't see this as a bad thing. As we have explained in our books, notably Breaking the Sound Barrier, anything which promotes thorough processing of language can be welcomed. On balance, I think the backwash effect could be positive. As always, the success of classroom dictation depends on how it is prepared for, scaffolded and delivered. It needs to build pupils self-efficacy. It's easy to do dictation badly and set up students for failure. I know!

How about its "authenticity" as a task? Is it useful? Does it resemble thongs we do in real life? Arguably, it is more authentic in that sense than reading aloud. You can envisage occasions when you need to listen and note down information carefully, if not in the style of dictation as such. But close listening and note-taking is something we occasionally do. But let's face it, it's not nearly as authentic a task as having a conversation in the target language. That said, as I pointed out last time, most exam tasks pupils do not have much real life authenticity, so this does not concern me overly.

So I see some value in including an element of dictation/transcription in the exam. I hope it would take the form of some transcription of language heard in an aural text which has to be exploited in other ways, e.g. questions. matching, multi-choice. Something like:

"Now listen to this sentence again and write it down accurately." As with reading aloud, I hope few marks are allocated to the task.

There will not be time in an exam for a traditional dictée.

In a recent Motivated Classroom podcast, Liam Printer spoke about deeply held beliefs language teachers hold and how these can stop us trying out new ideas.  Some readers will be entirely comfortable with dictation, others will find it alien. Whatever your gradually acquired belief about it, I'd suggest it's worth stepping back and considering is as objectively as possible.

Comments welcome as always!

Key reference:

https://consult.education.gov.uk/ebacc-and-arts-and-humanities-team/gcse-mfl-subject-content-review/supporting_documents/GCSE%20MFL%20subject%20content%20document.pdf