I am somewhat troubled, by many things, but this particular trouble is to do with learning.
It is so obvious that, over time, say a term of school, learning must have happened. Children who attended my physics lessons did have more physics knowledge than they had at the start of the term. They could answer questions on test papers that they previously had no knowledge of. By any definition, learning had happened.
What has caused this learning to happen?
It must be due to something that the learner has done. They must have interacted with the physics I was trying to teach them in some way. It is an active process on their behalf. By active I mean they must have been thinking about the physics, I do not mean they will have been physically active. Only thinking type activity will cause learning. It is the thinking part that has allowed them to connect the physics to stuff already in their brain. That connecting is one part of the process called learning but only when that stuff is retrievable. It will be retrievable during the lesson but learning that is useful for school stuff becomes so when it is retrievable some time after the time when it was first encountered. When it is used to support further learning next term, or in the exam etc.
The next part is for those connections to be made more secure. That is done by practice, lots of questions and problems to be solved. Put the numbers in the equations. That could have been a mechanical task with minimal thinking but for true learning which would lead to understanding the 'just stuffing numbers into the equation' type practice would not lead to any significant depth in understanding. Can-do learning rather than the more valuable does-understand learning.
So, because learning which is evidenced by the degree of understanding the child shows takes time some folk say we cannot evidence learning in a short time period such as a typical lesson. I think I must agree with that. But that does not mean we cannot identify that the early parts of the processes have happened. That the teacher has done his/her job in providing the appropriate conditions for the learning process over time to then have a good chance of happening.
What system would we need to be able to observe that? Seems possible to me. We would be saying something like, 'On what I have observed secured learning is likely to be able to happen in the future.'
Diagrams are good, pictures work. So here is one to try to describe the learning process I have identified above:
Initial event (teacher explains some physics) --- Child can repeat or use in some simple way this physics content --- child practices and thinks more about the physics --- future lessons same material gets used to support additional learning --- child continues thinking and gains deeper understanding --- more secure knowledge results
Well, it is a sort of a diagram!
In the course of a lesson we can only hope to observe the first three parts of this process and then only of the teacher chooses to do any practising of the newly presented material.
A bit like trying to observe my journey from North Wales to watch Saracens play in the Heineken Cup Final in Cardiff but only watching me prepare and then leave. Not certain that I will get there in time but some factors would show I was possibly going to succeed.
Is that good enough?