Replies
No one has replied to this post.
Maybe I have gone down a rabbit hole with this. But the point is that there's no way to work this out, unless you have the end data and work back from that?
Yes I could alter the calculation but the point of it is that no one else would consider this to be the norm? The only way to work this out is after the fact.
And why was the minmum duration of 12 months defined as 366 days and not 365 days?
Hi Sam Bern,
Duration by my count for an enrolment starting on the 30th Aug ending on 10th October is 42 days exactly; as learning was recorded on both the first and last day in the range both days should be included in the range.
Excel, Python and even SQL functions such as DATEDIFF, would calculate the difference between two dates and not the total duration in days from one date to another.
If you receive 1 orange at the end of each day for 3 days you will end the third day with 3 oranges (the total), which is 2 more oranges than you ended on day one (the difference) between the first and last day counts.
It is not that you 'could' alter your calculation to 'x+1-y' but that you 'must' alter it so that your results match the DfE.
HTH,
Phil
Thanks Phil, I do follow your logic and I agree to match the ESFA data I'll have to ammend this count accordingly.
Your example is really useful and it helps clarify my confusion as it's not what's descriped in the QAR specification. It defines the funding qualifying period field as the difference between these two dates.
'The actual duration of the aim, in days is the difference between the Learning Start Date and the Learning Actual End Date.'
Field: funding_qualifying_period_excl
This is different to your example above.
My issue is that they haven't actually followed the process that is outlined in this document or the document is incorrect. Maybe if there had been an example with dates in it, it would have helped.
The reason as I understand it regarding 366 days was that they had to complete 365 days and they wouldn't have completed 365 days until the 366 day was complete. As the first day was not counted as a full day.
I'm ready to admit I'm wrong on this. But I want to see what I could have referred to before the data was published this week that would have shown me I was wrong. In the grand scheme of things it's a really small issue which has only affecting our data this year in a tiny way. But if people have access to something where this is clarified I'd really appreciate it being shared.
Sam Bern
QAR R12 when is 42 days 42 days.
Edited
So there is a .2 difference between what we were expecting in our QAR and the dashboard and the downloaded data. It is one learner who was in learning for 41 days. I'm going to raise this as a ticket but I'd value the opinion of everyone here as well.
The wording in the QAR guidance is 'less then 42 days'. The learner started on the 30th August 2023 and left on 10th October 2023. The only way you can consider it to be 42 days is if you count 30th August 2023 as day 1. However in this instance I'd argue that it shouldn't be counted this way.
Less then 42 days already takes into account the difference of a day, otherwise it would be less then or equal to 42 days. As 42 days is six weeks. By counting the 30th August and saying less then 42 days, they are effectively removing two days from the six weeks. So it's actually become 5.71 weeks not six weeks.
In the 12 month minimum period the minimum duration was not 365 days it was 366. As they have to have been on for a full year, and day 0 did not count.
If you were to say something is happening in a day's time you wouldn't expect it to happen that day, it would be the next day. If I was to say it's happening seven days from Friday, I don't mean Thursday next week I mean next Friday.
If I calculate the number of days that have elapsed between the 26/09/2025 and 26/09/2025 in Excel, Python or any data tool the answer is 0.
If you've read it this far thanks. It's something that's been stuck in my head all this week and my colleagues are now looking at me weirdly!
This learner wasn't presented in the R10 data, due to the issues with the calculation and data they released they were excluded.