Maria Reason

Accountability Framework



I am currently reviewing our data against the Accountability Framework which I believe the number of withdrawals are measured against the total number of withdrawals actioned per contract year irrespective of which contract year the learner falls in?

For example, 75 learners withdrew in 2020/21 contract year however 20 learners have an expected end date in 2022/23 contract year, this would still be measured against 75 learners in total?



No one has replied to this post.

Paul Taylor

Hi Maria

I believe that the ESFA are looking into creating a specification document to support the accountability framework - it would have been useful to have this before we started the process but we are where we are. 

My understanding is that most of the measures are being applied across the year's ILR file. So for example, the withdrawal number will be all of those learners that have left within this academic year. If you want to work out the proportion then I think that will be all those that have withdrawn during this academic year as a proportion of all apprentices that have been in learning at any point during the year, regardless of when their planned end date is (the reporting logic will be something like actual end date >=01/08/2021 or is null, start date <01/08/2022).

Maria Reason

Thanks Paul, that's how I worked it out although, I initially thought it would be based on the contract year that is affected (contract year the learner is due to achieve) to align with the achievement rates. I was maybe over thinking it as this way kind of makes sense. So, the percentage could decrease as additional starts are processed via the ILR? 


I think it's not aligned with achievement rates *on purpose*? They want to spot where a provider is going to have a problem in the future rather than waiting for it to happen...

Maria Reason

Hi Steve, that makes perfect sense thanks


Helen Cuthbert

Paul Taylor they have created the technical spec its just not overly clear on some of them! 


At Risk - Of total number of apprentices, percentage of withdrawals is greater than 35%

Needs improvement - Of total number of apprentices, percentage of withdrawals is less than or equal to 35% and
greater than 15% OR where the number of apprentices identified as withdrawn is greater than 250 


Sure, but what they haven't laid out *precisely* (ie with the exact parameters they're using, eg are withdrawals within 42 days counted?) what "total number of apprentices" means...

To give another example, strictly speaking, a transfer is a withdrawal, but we need to see the spec that say "excludes w/ds where withdrawal reason = 40".

(my thinking is they'll end up counting people they shouldn't be, because the ILR is super weird, which is why we need the spec nownow so we can tell them what they're doing wrong!)

Helen Cuthbert

Totally agree Steveh its near useless to actually work with a the moment. I am doing the same thing as Maria at the moment and struggling to figure of what figures i am using to work percentages out off. 


Paul (ESFA)

Hi all

There is an ESFA webinar on this topic on Tues 21 June. I've pasted below an article from ESFA Update on 1 June

Information: apprenticeship training provider accountability framework 2021 to 2022 – webinar for providers

We would like to invite apprenticeship training providers to join a webinar on Tuesday 21st June (12:00 to 13:00) to learn more about the Apprenticeship training provider accountability framework for 2021 to 2022.

The Apprenticeship training provider accountability framework came into force in September 2021. The Framework is a new, more timely approach to accountability for apprenticeship training providers, based on a wider range of quality metrics. We have recently updated the Apprenticeship Accountability Framework Technical Specification 2021 to 2022 to provide further information on some of the quality indicators that form part of the framework.

The session will cover:

  • Recap of the purpose of the framework and the overall approach
  • Updates to the Technical Specification
  • Further information in response to questions raised by providers.
  • Opportunity for Q&A

Please complete the registration form here to secure your place.

I hope that's helpful.

Many thanks



Must've been too excited about the update about the Shipbuilding Taskforce to notice that! ;)

Duncan Brown

Hi all,

Sorry for the late post, but I was wondering if anyone knows of any clarification of the learner cohort for the accountability framework.  I was working on "all your apprentices within the academic year" being all apprentices in the ILR who have been in learning at any point during the academic year.  Any advice would be gratefully received as I'm reluctant to rely on assumptions here if there is clearer guidance available.

Many thanks in hopeful anticipation.


Hi Duncan

No, there isn't yet any published info that is clearer, however, conversations via the Service Desk have revealed the following:

  • It's *every* Apprentice in your ILR (in essence, including someone who withdrew last year but is still in your data), although they are applying the 42 day rule to exclude non-starts
  • Each Apprentice only counts once, so a learner who withdraws and then restarts in the same year won't count in the withdrawals column
  • However, a learner who completes L2, moves on to L3 and then withdraws (after 42 days) will only be a withdrawal.
  • Theoretically, a transfer who stopped App 1 on, say 14 July but didn't start App 2 until 4 August would be a withdrawal and count against you because they're not explicitly excluding withdrawal reason 40, but that's super edge case stuff!

Duncan Brown

Thanks very much Steveh.  Most helpful.