The data on positive cases in different age groups really only takes us so far to determining the risk in schools I think.
What the government is not doing is looking at is how many asymptomatic cases there are among children, and how much positive cases spread it (doing follow up on contact tracing) or looking at how seriously ill their contacts get if they catch it, which I think is important for teachers, parents, death rates and long covid rates.
Based on what I've read (like the S Korea study, the Princeton study - which shows transmission from children to household contacts - and just thinking about exposure scenarios) I'm thinking that older children are more likely to spread it to more people (they have more contacts) so are more important in terms of community transmission.
However, caregiving small children requires a lot of close contact - being sneezed on, breathed on, spat on, pissed on etc at close quarters. We know that viral load is important in terms of severity of disease (lancet paper on this - mortality increases with higher viral load www.thelancet.com/journals/lanres/article/PIIS2213-2600(20)30354-4/fulltext ) so I think in terms of primary school teachers (particularly lower years) and parents of small children, there is likely a higher risk of worse outcomes, though possibly less likely to catch it in the first place. This is particularly important for ECV teachers and parents.
We know that masks reduce viral load. Not very practical for parents but I think teachers should be wearing them. Especially when having to wipe bums, comfort crying children, dealing with yelling tantrums or otherwise getting close.
The discussion of viral load is completely absent from risk assessments in schools and it really shouldn't be - particularly for vulnerable teachers in those settings. It's important also in secondaries where teachers can't keep the fabled 2m away from all students at all times (as if) and when exposed to large numbers of students over a day.