RSCs… you ought to know better!

This week, Schools Week also have an article sharing the news that Russell Hobby of the NAHT has made a similar point to RSCs, but probably far more professionally. You can read the article here:
http://schoolsweek.co.uk/rscs-and-councils-ask-schools-for-impossible-primary-assessment-predictions/


 

Last month I wrote about how I feared that Local Authorities were preventing schools from moving confidently away from levels.

Today, I have Regional Schools Commissioners in my sights. I’ve been concerned about this for a while because increasingly I see people who are doing their best to cope in a world without levels suddenly faced with demands for data from external agencies.

And in the case of Regional Schools Commissioners: they ought to know better. Yet clearly they don’t, otherwise we wouldn’t be seeing queries like these:

“We’re asked to predict progress for current Y6 (by RSC). How would you suggest we go about it?”

One of the most significant changes in the world after levels is the way in which progress is to be calculated across primary schools. It has deliberately moved away from a threshold model, so that schools can focus on improving the attainment of all pupils, instead of focussing disproportionately on those who are close to a threshold. It’s a shift that I happen to think is an excellent idea.

But it’s also a shift that means that it’s very difficult to predict outcomes, and impossible to predict progress measures. Attainment is hard to predict because we don’t know where the thresholds will lay. You can estimate what you think children need to be able to do, and guess where the threshold might be set, and then try to estimate which children are on track… but that’s a lot of approximation.

For progress, there is simply nothing you can do. Your school’s progress score will depend not on how many cross that magic 100 threshold, but on exactly how many marks each child gets on each test. And then on how every other child in the country scores. Even after the tests are completed, it will take the DfE months to calculate the first sets of progress scores; what hope has anyone got of predicting a measure based on so many complex factors?

The problem, of course, is that these people are still stuck on “old thinking”. Of course you can attempt to replicate the old systems. You can look at the number of children working at Level 2 in KS1 who you hope will reach the expected standard. But that brings us back to the guesses about attainment. You can look at the number of children who were working at Level 3, who you hope will reach some greater measure… except no such threshold exists. And even if we did know the thresholds, they’re virtually insignificant. The difference between a child getting 99 or 100 is far less important than the difference between another child getting 91 or 97 – even if they don’t meet the “expected standard”.

The aim of the new system is to stop schools from focussing on borderline pupils. To do that, you have to remove as many borderlines as you can. For RSCs not to understand that is concerning, and for them to put additional pressures and demands on schools for imaginary data that won’t help school improvement at all is unacceptable. Indeed, far from helping, they may end up driving exactly the sort of flawed behaviour that went on before that we’re trying to get rid of!

Or as Jamie Pembroke put it far more succinctly than have I:

 

Tagged: , ,

9 thoughts on “RSCs… you ought to know better!

  1. R Waring 3 March 2016 at 10:25 pm Reply

    Reblogged this on rwaringatl.

  2. AssemblyTube 3 March 2016 at 11:03 pm Reply

    Could it be that throwing away levels prematurely has thrown the baby out with the bathwater? When will we see our first school put into special measures for not being able to estimate/demonstrate pupil progress for each year group?

    • Michael Tidd 3 March 2016 at 11:33 pm Reply

      I’d be very surprised if that happened – Ofsted seemed to be ahead of all the other organisations on this!

  3. Rachel 4 March 2016 at 9:17 am Reply

    This request for levels comes from outside agencies re Send children too. Impossible.

  4. Thomas Fisher 4 March 2016 at 10:26 am Reply

    Ofsted report from this week. In the summary for why the school is RI this comment was made.

    Leaders introduced a new assessment system in September 2015; consequently, judgements are developing but are not yet consistently accurate. This has made it difficult, in the short term, for leaders and governors to track pupils’ progress effectively.

    • Tom Bishop 4 March 2016 at 11:19 am Reply

      Let’s just put every school in the country in RI and be done with it.

    • Michael Tidd 4 March 2016 at 12:17 pm Reply

      I’ve just taken a look at that report, and while I agree that it’s not helpful to have that mentioned in the main findings, it notable that it isn’t really the main issue that’s raised.
      I do think, though, that perhaps that ought to have been better checked before publication as it’s not a helpful message to be out there.

  5. educationsvoice 4 March 2016 at 6:40 pm Reply

    Every half term I am being asked for this data due to RSC! I complain every time. Your post is spot on. I have been saying it over and over. However, the powers that be don’t listen! I am moaning. I have low expectations. I am a naughty child and must go sit in the corner!

    • Michael Tidd 4 March 2016 at 8:52 pm Reply

      I’d complain too! I’d suggest directing them to the Assessment Commission Report which clearly says that collecting summative data more frequently than termly is a waste of time!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: