On Friday 23rd February I gave talk at the Headteachers’ Roundtable Summit on the theme of assessment and workload. This blog captures some of the main points. I’ve been exploring this theme is various other places too so, to avoid repetition, please also read:
Towards an Assessment Paradigm Shift – how we need away from macro data-tracking towards a focus on micro formative assessment
Don’t do ‘Can do’. The problems with can-do checklists and trackers. – explores the issues with the new machinery adding pointless workload around the country including massive data systems creating phoney numerical flight paths based on mastery statements.
We know Amy is struggling. The question is what we do about it. – discusses the question often left to last, instead of coming first.
#FiveWays of Giving Effective Feedback as Actions – my suggestions for making feedback proactive, based on work for students, not for teachers, moving away from reflection sheets and excessive marking.
Making Feedback Count: “Close the Gap” – an old blog with some great ideas about making marking more effective.
Here are some images from the talk via twitter:
Some of the main points are made in these slides:
I recommend that all school leaders read Daisy Christodoulou’s excellent Making Good Progress book and really get to understand what formative testing should look like.
Some features of formative and summative tests are illustrated here:
There are still too many schools where the culture leads to teachers doing pointless marking – too much of it in relation to the difference it makes to learning. Marking needs to be selective:
When I say ‘must’ generate responses, that’s more about saying it should be designed with that purpose in mind; it’s not meant as another stick to beat people with. Students should respond to as much as they can. I’m a big fan of DIRT time in lessons and the whole-class feedback method:
This got a big reaction on twitter:
There’s always a risk when suggesting teachers do less marking that they hear ‘don’t do any marking’. I’ve had the problem; people hear what they want to hear. Of course, that’s not what I mean. I’m suggesting a healthy diet. My view is that Heads should not impose anything but should ask each team or subject department to devise their own marking and feedback regime: something that is meaningful, manageable and sustainable. The only expectation thereafter is that they stick to their own plan.
I talked a lot about assessment and, again the need to balance the micro/formative with the macro/summative. Learning happens in the micro – and we only need some cohort tracking, maybe once or twice a year. In the meantime, greater value should be given to teachers’ mark books, raw marks, authentic data that is dynamic and totally meaningful to teachers and students .
Instead of asking teachers to turn their subject-specific, raw assessment into a common format that suits the needs of the data system and leaders, leaders should spend time getting to know how teachers assess and learn to understand each of the formative, raw data systems where all the meaning sits.
As soon as we morph all of that into generic grades to meet the needs of the data system, we lose validity and meaning. Reports are hugely problematic in this regard – because the differences in the grades of all kinds probably says more about the different ways teachers use criteria, judge grades or covert their own assessments into the common format, that it tells us about real learning issues for the student.
Most of this could be wrong -you can’t be sure the 6 is geography is definitely better than the 5 in science without a lot of cross moderation. The attitude grades are purely impressionistic. As my son says, ‘some teachers take this stuff more seriously than others’. ie it’s more about them than about him; I believe him. At the very least we should be honest about how approximate it all is. Assessment is a fuzzy cloud of uncertainty – and it pays not to ever forget that before we turn it into numbers with decimal points.
Perhaps this is a model to work towards? I think we need more standardised tests to support cohort tracking, used annually. There’s a gap in the market here.
In summary: Do less/fewer