Since the late 1980s, at GCSE in England and Wales, we have been assessing the four skills of listening, speaking, reading and writing separately. We have moved from discrete skill testing, to more mixed skill testing, back to discrete skill testing. By which I mean that, for example, in a reading tests students are not now tested on their writing at the same time. Interestingly, at A-Level we have not been too concerned about testing each skill in this strict discrete fashion.
At the same time, at GCSE, the weighting of the four skills was, for a long time, 25% for each one. Just recently, this changed to 20% listening, 20% reading, 30% oral and 30% writing. This change was entirely owing to the fact that MFL had to be in line with all subjects in allocating 60% of marks to controlled assessments. In languages this was seen to mean speaking and writing where production of tasks was required. (It would be hard to conceive of a listening or reading task which could be done with pupil preparation in the style of a controlled assessment.)
First of all, which is better: discrete skill or mixed skill testing? Both approaches have their supporters, which may be why exam boards/Ofqual have moved from one to the other and back. Proponents of mixed skill testing argue that separating the skills is artificial and leads to a "backwash" effect in classrooms and course books. This is already apparent, especially as exam boards dictate so strongly the content of course books. It is most likely that we are already seeing less use of the target language as teachers practise exam style tasks, for example, reading comprehensions using English questions. On the other hand, proponents of discrete skill testing argue that it is fairer, more accurate in assessing that particular skill and, it has to be said, easier for less able candidates.
I see the case for both sides, but lean towards mixed skill testing, mainly because British teachers have a strong tendency towards teaching to the test, especially in these days of targets, performance management and league tables. This can lead to poor methodology and a lack of authentic, mixed skill tasks. Just take a look at the latest text books to see what I mean. We accept the case for mixing skills at A-level, but not at GCSE. My assumption is that Ofqual wish to make the GCSE exams accessible to all candidates, not just the most able. At A-Level it is correctly assumed that the aptitude of candidates is higher.
As for the weighting of each skill, it is a pity that we allocate so many marks to the hardest of the skills, writing. The previous allocation of equal marks for each skill was better than what we have now, but I would allocate more marks to the two skills which many would consider to be the most useful in language learning, namely listening and speaking. So, if we have to work in round figures, I would argue for the following weighting: 30% listening, 30% speaking, 20% reading and 20% writing. This still rewards writing to a considerable degree and may reflect a continued bias towards the written medium in assessment, but allocating anything less than 20% may encourage teachers to neglect the skill too much.
When controlled assessments disappear, it will be interesting to see what Ofqual decide in terms of skill weightings. I hope they ask teachers for their opinion.
At the same time, at GCSE, the weighting of the four skills was, for a long time, 25% for each one. Just recently, this changed to 20% listening, 20% reading, 30% oral and 30% writing. This change was entirely owing to the fact that MFL had to be in line with all subjects in allocating 60% of marks to controlled assessments. In languages this was seen to mean speaking and writing where production of tasks was required. (It would be hard to conceive of a listening or reading task which could be done with pupil preparation in the style of a controlled assessment.)
First of all, which is better: discrete skill or mixed skill testing? Both approaches have their supporters, which may be why exam boards/Ofqual have moved from one to the other and back. Proponents of mixed skill testing argue that separating the skills is artificial and leads to a "backwash" effect in classrooms and course books. This is already apparent, especially as exam boards dictate so strongly the content of course books. It is most likely that we are already seeing less use of the target language as teachers practise exam style tasks, for example, reading comprehensions using English questions. On the other hand, proponents of discrete skill testing argue that it is fairer, more accurate in assessing that particular skill and, it has to be said, easier for less able candidates.
I see the case for both sides, but lean towards mixed skill testing, mainly because British teachers have a strong tendency towards teaching to the test, especially in these days of targets, performance management and league tables. This can lead to poor methodology and a lack of authentic, mixed skill tasks. Just take a look at the latest text books to see what I mean. We accept the case for mixing skills at A-level, but not at GCSE. My assumption is that Ofqual wish to make the GCSE exams accessible to all candidates, not just the most able. At A-Level it is correctly assumed that the aptitude of candidates is higher.
As for the weighting of each skill, it is a pity that we allocate so many marks to the hardest of the skills, writing. The previous allocation of equal marks for each skill was better than what we have now, but I would allocate more marks to the two skills which many would consider to be the most useful in language learning, namely listening and speaking. So, if we have to work in round figures, I would argue for the following weighting: 30% listening, 30% speaking, 20% reading and 20% writing. This still rewards writing to a considerable degree and may reflect a continued bias towards the written medium in assessment, but allocating anything less than 20% may encourage teachers to neglect the skill too much.
When controlled assessments disappear, it will be interesting to see what Ofqual decide in terms of skill weightings. I hope they ask teachers for their opinion.
Sorry, my email address is andrea at upstreamconnections.com
ReplyDeleteThanks!