So am I. I couldn’t help thinking that there must be some logic I’m missing. So pulled together more data. The table below uses data the excellent London Datastore to see if there was any discernible pattern.

Short story. I still haven’t found one.

If you want to look for yourself the table below is sortable (there’s a little black sort button on each column about half-way up on the right-hand side of each header). You can sort by multiple columns by holding ‘shift’ as you click. The size of the table does make it a bit unwieldy on phones, sorry.

I have tried to explore a few different explanations. The first is the good old Inner/Outer London split. You are a bit more likely to be in an open borough if you are inner London.

I looked at population density and school age children to see if this could offer any explanation. This might be the most logical reason for openings that looks odd, since the boroughs vary enormously in size, which might result in decisions that there odd at first sight, but made sense when looking at the numbers. Again, I couldn’t see any pattern here. I did explore by trying to look at the number of schools and how that might affect things. Again, this produced no obvious answers, although that might be skewed by things like cross-border and private schools.

I also wondered about educational need. Given the government’s repeated statements of the importance of keeping children in education it’s possible factors related to educational need might have relevance. I used English proficiency or children looked after as imperfect proxies for these but, once again, there was no obvious link.

The other possibility is that the government have been looking at trends, but given their inability to spot and act on nationally rising rates during the last national lockdown I suspect it’s unlikely they can do this on a borough level.

I have tried various combinations of factors and other data, but not managed to come up with anything. It is possible, of course, that the Department for Education is using a weighted combination that I’ve not come close to. I like to think that there is some logic to it. But unless and until the criteria used are published it’s very easy to conclude the reasoning behind the decisions is not entirely based in controlling the virus.

BoroughStatusInner/Outer LondonCase rateControlPrimary schoolsPopulation density (per hectare) 2017Proportion of population aged 0-15, 2015% of pupils whose first language is not English (2015)Rates of Children Looked After (2016)GLA Population Estimate 2017
BarnetClosedOuter730Con9144.921.146.035389,600
CroydonClosedOuter780Lab8644.72236.786386,500
EalingClosedOuter623Lab6863.321.443.646351,600
EnfieldClosedOuter949Lab6841.222.855.343333,000
BrentClosedOuter632Lab6076.820.937.645332,100
BromleyClosedOuter825Con7721.819.938.940327,900
RedbridgeClosedOuter1,027Lab5253.922.866.528304,200
HillingdonClosedOuter743Con6926.021.329.649301,000
GreenwichOpenOuter705Lab6359.221.944.178280,100
Waltham ForestClosedOuter857Lab5171.221.862.443276,200
HounslowClosedOuter708Lab5349.021.150.645274,200
HaveringClosedOuter1,095Con (council NOC)6122.619.338.942254,300
HarrowOpenOuter726Lab4250.020.557.632252,300
BexleyClosedOuter1,006Con5740.320.632.646244,300
Barking and DagenhamClosedOuter950Lab4457.927.241.769209,000
MertonClosedOuter776Lab4455.320.634.235208,100
SuttonClosedOuter747LD4046.220.734.350202,600
Richmond upon ThamesClosedOuter593LD4534.420.753.526197,300
Kingston upon ThamesOpenOuter595LD3647.119.639.330175,400
NewhamClosedInner894Lab6694.722.758.842342,900
LambethOpenInner696Lab60122.717.639.173328,900
WandsworthClosedInner685Con6193.717.850.841321,000
SouthwarkClosedInner673Lab74108.918.629.875314,300
Tower HamletsClosedInner917Lab70153.720.151.147304,000
LewishamOpenInner696Lab6386.320.640.969303,400
HaringeyOpenInner785Lab6493.92048.067278,000
HackneyOpenInner751Lab58144.020.744.253274,300
CamdenOpenInner481Lab42111.317.325.241242,500
WestminsterClosedInner476Con40112.715.952.639242,100
IslingtonOpenInner593Lab46155.615.957.688231,200
Hammersmith and FulhamClosedInner587Lab37113.017.448.958185,300
Kensington and ChelseaClosedInner476Con27131.116.445.837159,000
Data: London Datastore

Like many I was utterly bemused by some of the decisions made on which London boroughs would see their primary schools opening in the new year.

Clearly having children in school is the best outcome but this has to be balanced against the other public health considerations. A few weeks of education that can be caught up might not be worth the life-long cost of losing a loved one. While I fully expected all school’s to be closed I wasn’t at all ready for the irrationality of the decisions taken.

You would expect some pattern, but looking at data from the London Datastore and the list of schools that are open it looks to me much more like the correlation between opening and closing is much more down to political control.

How on earth can you suggest that Redbridge, with a case rate of 1,027 per 100,000 should have it’s school’s open, while Westminster and Kensington and Chelsea, with a case rate of half that, should be closed? Or keeping Greenwich open when six of the ten boroughs with lower case rates are all closed.

There are 32 London boroughs, ten will be expected to open schools. Of those nine are Labour controlled and one Lib Dem. Not a single Conservative controlled borough is expected to open primary schools.

You would expect some incompetence from the government, but at first sight this looks much worse than than.

Update: Redbridge was omitted by the government in error (because this isn’t the sort of thing you would want to double-check before publishing). I have updated that in the list, but think the central point remains: why have places like Hackney and Haringey open when many others are closed.

Update 2: I returned to this with some additional data to try to understand the decisions. I was still none the wiser.

BoroughCase rateControlStatus
Havering1,095Con (council NOC)Closed
Redbridge1,027LabOpen Closed
Bexley1,006ConClosed
Barking and Dagenham950LabClosed
Enfield949LabClosed
Tower Hamlets917LabClosed
Newham894LabClosed
Waltham Forest857LabClosed
Bromley825ConClosed
Haringey785LabOpen
Croydon780LabClosed
Merton776LabClosed
Hackney751LabOpen
Sutton747Lib DemClosed
Hillingdon743ConClosed
Barnet730ConClosed
Harrow726LabOpen
Hounslow708LabClosed
Greenwich705LabOpen
Lambeth696LabOpen
Lewisham696LabOpen
Wandsworth685ConClosed
Southwark673LabClosed
Brent632LabClosed
Ealing623LabClosed
Kingston upon Thames595Lib DemOpen
Islington593LabOpen
Richmond upon Thames593Lib DemClosed
Hammersmith and Fulham587LabClosed
Camden481LabOpen
Kensington and Chelsea476ConClosed
Westminster476ConClosed
Primary school opening status by borough, case rate and control. Data: London Datastore