In 2017, I looked back on my 46 years of modern languages teaching. Despite fond memories, I felt unease. I sensed a disconnect between pupils’ competences and Scottish Qualifications Authority (SQA) results. I have met Higher pupils whose A grade left them floundering and unable to create spontaneous, simple German.

Three years’ research answered the question: have German teaching and testing - which I used as an exemplifier for modern languages - failed Scottish pupils? 

The SQA decision at the end of January to ditch the talking element of Advanced Higher shows that they continue to fail Scottish pupils and confirms my research findings.

Background: Steep year-on-year drop in languages entries

News: Speaking removed from modern languages qualifications

Related: Secondaries failing to deliver ‘right’ to languages

Opinion: How cutting university languages courses hits schools

Covid: SQA and Education Scotland reform demanded by MSPs

SQA considers talking an add-on - an extra easily disposed of - when it is, of course, vital and a prerequisite to any meaningful measure of target language competence.

SQA had already jettisoned its specialist study portfolio, with the result that 40 per cent of the course assessment does not count in the final award of 2021. It has offered no satisfactory explanation for its rejection of this content. Since it is happy to trust teachers (or, at least, happy to try to make a virtue out of a necessity) to provide estimated grades in listening, reading and writing, why not trust teachers to provide estimates in talking? 

The historical aspect of my research examined more than 50 assessments from the SQA and that body’s predecessors, the Scottish Examination Board and Scottish Certificate of Education Examination Board. The contemporary aspect examined SQA assessments from 2016-2019 and contemporary assessments in England. 

My research produced stark findings.

SQA tests do not measure competences as accurately as that body claims or with the accuracy SQA used to achieve. Exclusively non-fiction external assessment texts reduce the texts’ appeal to candidates and restrict the range of sources. Teachers report that pupils read less fiction, perhaps none. The sad consequence is this: reduced knowledge of extended vocabulary, of language structure and idiom, of imaginative ideas and of target language culture.

Meanwhile, the absence of explicit testing of grammar and structure in National 5, tokenistic translation from the target language at Higher and Advanced Higher levels (and none into the target language), have brought about a reduction in the teaching of grammar, structure and idiom. Assessors’ reports confirm many candidates’ inability to create accurate target language and manipulate others’ language. Generic terms across European languages are widely unknown, including “participle”, “pronoun”, “indirect object”, “passive”, and “irregular verb”.

English is used in rubrics, questions, bulletpoints and responses to reading and listening texts at all levels. The excessive predictability in writing and talking in National 5 and Higher leads to teaching to the test. The specialist study portfolio (literature and film) in Advanced Higher is unworthy of its pretentious name.

SQA terms such as “flexibility”, “accessibility“ and “personalisation” are weasel words. They contribute to SQA assessments being less challenging than corresponding assessments elsewhere in the UK.  

SQA has dumbed down modern languages assessments over many years. These issues surrounding modern languages assessments need to be aired, not airbrushed. 

How should teaching and testing become more effective? 

Teachers and pupils must talk much more in the target language, instead of talking in English about the target language. Teachers should talk at a level just above the level of pupil familiarity, so that intellectual challenge is present but not excessive. Speaking and listening are two sides of the same coin.

The reading of fictional narrative must greatly increase and be in SQA assessments. This would reflect research evidence into the efficacy of fiction in furthering reading skills and understanding (especially for boys). 

Regular feedback to pupils and parents is sufficient until the end of broad general education. Progress and continuity matter more than arbitrary, vacuous and often stressful tests.

In external assessments, SQA must restore intellectually stimulating texts reflecting fiction and non-fiction, and tasks that require candidates to create their own and others’ writing and talking in the target language, instead of unsatisfactory reliance on rehearsed chunks of arbitrary language that they cannot manipulate.

I am clear why I felt uneasy almost four years ago; I am clear also that my study demonstrates that the two-year hiatus in external SQA assessments brought about by the coronavirus pandemic should encourage a radical review by SQA of what and how they assess. Our pupils deserve this, but I am not holding my breath.

John Nolan is a modern languages teacher in Scotland. This summer he celebrates 50 years of teaching languages