FFT Targets - Realistic?

[PrettyWoman]

Afternoon all. Does anyone have an opinion on how realistic FFT targets are please? We are looking to use FFT for the first time from September and have looked at the targets they set. Do they change through the year??? Are the targets realistic in terms of grades achieved in GCSEs. How do you go about setting targets for BTECs? Any advice gratefully received.

[Ditto]

They're calculated on a repeatable mathematical model. FFT will be able to give you a precis answer on the accuracy if you ask them. Whilst you will see targets exceeded or missed for individuals, on average I'd say they are very realistic. You of course have the option of using the 50% target - i.e. the mid point. At one of my earlier schools we targeted top 20% as it was at the time led by a strong aspiration driven head. The targets did not change through the year, but would update each year once a new national cohort of data was available. In summary, I'd say FFT was reliable. In that previous school I mentioned, we used FFT in the first few years of secondary, but reassessed using Yellis later. I would expect the Yellis target to be more accurate as the pupils will be nearer the GCSE result and I would expect you would have more recent and accurate data to work with, but I don't know if that is a proven fact. As an overall strategy for setting targets, we used FFT (top 20%), CATS and teacher professional judgement to triangulate - utlimately the teacher would have the say-so, but they'd have to argue hard not to use the highest of those. The policy was one of of it being better to aim high and miss a little, than to set un-challenging targets and have 100% success.

 

I'm not sure what is available for BTECs, but with a much smaller number of achievable grades, it introduces another question - are they likely to be more accurate as a result or less?

Edited March 17, 2019 by Ditto

[RootsSchImp]

Using FFT estimates as targets, although common, is a terrible idea. Here's why:

 

The grade for each kid is the median achievement of a group with same prior attainment, gender and month of birth as that kid. Median as in lots of kids did better lots of kids did worse, even in the best schools.

 

So you set these as your targets and some kids are below target, are they underachieving? Maybe, maybe not, lots of kids do worse than the median grade. It's far more likely though that kids who are 'on target' are the underachievers, because you need around 30% of them to *exceed* their target to match the schools you want to match. So you're tracking system tells you to focus on the wrong kids.

 

This is true regardless of whether you use FFT top 50, 20 or 5. If you look at the different probabilities between these models you'll see that the median estimate (the grade erroneously used as a target) doesn't change that often between the three, the difference is the proportion of kids who beat it by one or two grades.

 

A far better approach (if you need targets at all) is to have teachers set them and to use them as an indicator of teacher expectation. Although note for that for this to work you can't use them for performance management because if teachers are judged on whether they hit targets they set then they'll set them low.

[Ditto]

... you can't use them for performance management because if teachers are judged on whether they hit targets they set then they'll set them low.

I agree with this bit, but the first head protected against this happening. Interestingly when we had no FFT data (because of no prior attainment records), it was the teachers (*some*) that struggled when asked to set the targets - there was a deep reliance on the data and a reluctance to use 'professional judgement alone.

[PyROm]

We used to use FFT grades and they worked for us, however several years ago FFT themselves changed the rules and said that FFT should not be used to set targets (they also did their best to make it hard to import targets from their website). This helped prompt the shift to using 3 levels of progress as an indicator. We took the standard 3 levels of progress flight path and the headteacher modified it to fit our students, teachers are not allowed to change their school target. In KS3 teachers enter where the student currently is, however in KS4 they enter a final predicted grade (ie. what the teacher expects them to get in the exam).

[PrettyWoman]

Thank you all for the advice. I will have to have a good look at this before making a decision on how best to proceed.