There is a legitimately difficult tension at the heart of Ofsted as an inspectorate. A. Schools are public institutions and need objective and impartial oversight. Models involving local or within-profession accountability simply won't work. B. Some schools are worse than others. Some schools are better than others. C. If you break schools into four categories of quality (1st, 2nd, 3rd, 4th). It's probably pretty easy to tell the difference between 1st and 3rd and 2nd and 4th. 2nd and 3rd? Not so much. 3rd and 4th? Tricky. This gets even harder if you make the groups smaller. If you have 5 grades, and 7 categories to judge them, telling the difference becomes increasingly difficult. The idea that you can genuinely tell with any kind of scientific accuracy the difference between "expected" and "strong" standard is for the birds. D. This is because most of the ways we can actually figure out if schools are good or bad start objective, but then get subjective really quickly. This school has a high p8. Ok, but what about its PP cohort? What about its SEND students? How many EAL students does it have? How does it compare to other local schools, or schools with a similar demographic? This school has lots of exclusions. Does that mean they are excluding too easily, or struggling with a tough cohort? E. These subjective questions are extremely difficult to answer. There isn't a rule book or an algorithm. You have to rely on humans and human judgment. And that's where inconsistency creeps in. Critically, that inconsistency is absolutely inevitable. F. You then also start commenting on things which have no objective component, and no clear evidence base. We don't really know what good "wellbeing" or "careers education" or "inclusion" or even a "curriculum" looks like. My job is literally to work with teachers at helping them get better, and I would never ever ever think I could fine-grade the quality of teaching in a school with any kind of objectivity. I reckon I have more expertise and experience at doing this than 90% of current inspectors, and still wouldn't touch it. The point with all these things is that we have some research, and we know what really bad practice looks like, but anything beyond that is just informed opinion. And that's problematic in and of itself, and because of the inconsistency that will creep in. At the moment, Ofsted are trying to inspect and evaluate things that probably can't be evaluated with high resolution or precision, and will certainly be inconsistent across inspectors. That's going to lead to problems, and Ofsted will get things wrong in a context where the stakes are extraordinarily hight. (I've put my preferred inspection model in the first comment)

I consistently see people refer to Cold Call as "pose the question, pause, then choose a student" Or "choosing students at random." The section on Cold Call in Teach Like a Champion is a beautiful exploration of effective questioning styles, but putting the name at the question or choosing students at random is categorically not how it is defined. There, the only central and necessary definition is that you are choosing students regardless of whether they have their hands up. For example, in the video linked below, the teacher consistently puts the name at the start of the question, but the TLAC team still call it Cold Call. teachlikeachampion.org/blog/… It's for this reason I don't advocate using the term anymore. It has too broad a definition, which leads to difficulties around communication and implementation. Regarding putting the name at the end of a question, I therefore prefer to just talk about "Name at End" questioning. The meaning there is very clear, and resists misinterpretation. Video example here: carouselteachlearnlead.subst… Regarding choosing students at random to answer questions, I think this is actually a bad idea, because teachers should never do anything at random, and it's pretty easy to be more deliberate and targeted with your questions. We discussed that on the pod here: open.spotify.com/episode/1gW…