What is English Assessment FOR Anyway?
New research has come out from FFT datalab which suggests we are wasting our time with assessment in English.
The evidence is compelling, and goes like this:
Imagine you want to be able to use your assessments to predict what scores students will gain in a year 8 English assessment. Which of this data would give you the best bet of getting your prediction right?
End of Year 7 English test scores.
End of Year 7 Maths test scores.
Key Stage 2 Reading scores.
Key Stage 2 Maths scores.
The answer is their year 7 maths score, closely followed by their KS2 maths score.
What the hell is going on? Well, for one, there is no subjectivity in maths marking.
And for two, it is much easier to write a proper assessment in maths. It is quite possible to set many meaningless tasks in an English assessment, and in particular fail to measure subject knowledge with sufficient weighting.
What if you combine data?
When we combine test data, English plays even less of a role in accurately predicting the year 8 English score. Instead, you would just add together the KS2 maths score and the Year 7 maths score and get a prediction which had an 80% chance of being right.
Why is it so hard to predict English scores from KS2 Reading scores or from internal assessments at the end of year 7?
The Problems of English Assessment
Students aren’t tested enough on the fundamentals of accurate punctuation - so genuine application of this knowledge is not properly assessed. Nor does it carry much weighting in the assessment.
Students don’t get enough challenging writing - they are unlikely to write a full short story or an essay on a text (or indeed both).
Instead they will get piecemeal tasks - a paragraph or two, a short piece of creative writing dominated by description. Some meaningless additional knowledge testing about a text we don’t need them to remember as an adult.
The year 8 test is probably more difficult than the year 7 test, and so getting higher marks requires harder work and more intelligence. These are predicted by the maths scores, as these are the main way to achieve well in maths.
How to Solve This Problem in English
Give students real writing tasks as assessments - like essays and short stories.
Teach from great examples of both of these during the year, so students can see what a whole one looks like (and of a length a student could reproduce in whatever time limit you give them).
Set a series of questions for each punctuation use - with multiple examples. For example, not just commas in a list, but with subordinate clauses at the beginning and end of main clauses, and extra information - parenthesis - and with non examples or misconceptions, such as the comma splice. That’s just for commas. Think in the same way about the other punctuation.
Have plenty of knowledge testing, but only on the cultural capital which will enrich them as adults, and on the need to know fundamentals which will be crucial at GCSE, A level and degree. Weight this body of knowledge to have a significant percentage of your overall mark. 25% knowledge, 25% punctuation, 25% for each of the real writing tasks - one literature one language.
Make these tasks more challenging year on year.
Finally, and most importantly, see testing as a way to build long term memory through retrieval events. The punctuation and knowledge tests need to recur at gaps suggested by the Ebbinghaus Forgetting Curve.
If you want to know more about meaningful tasks in English from year 7 to year 13, my book, The Full English will help you.
