Why are the Zoe app figure currently so different to press reports?

[Neolara]

I've been logging on the Zoe app for the past year or so. Generally their data has seemed to be a bit ahead of government announcements. However, it's currently been reporting daily rates of about 32,000 for the last week or so, compared to press reporting numbers of confirmed cases of 50,000 and rising.

Why are the figures so different now when before they were very much in synch?

Because people have stopped using the Zoe app?

I stopped using the zoe app an age ago it became annoying,, I'd imagine many others have, Zoe app isn't really a indicator.

To be honest I've stopped logging now. I imagine many others have too. They can only go off the data they get

I used the app at the start when you could only get a test if you were in hospital & then layer when you only got a test if you had symptoms. Now I’ve got lateral flow tests that I can use regularly & report so I don’t really feel the need for the app anymore

I think Zoe app is estimating symptomatic cases. Testing is now so widespread that lots of asymptotic cases get picked up too

Testing is now so widespread that lots of asymptotic cases get picked up too

that makes sense to me

The Zoe app tends to be used by users who have a higher income and education level the the population as a whole, which tends to skew results a bit---that is the general consensus I've heard. It's still useful date but take it with a pinch of salt.

Unless the data generated by the app is adjusted to take account of any differences in relevant socio economic factors like age, income, education, area of country and all manner of other factors between app users and the general population, that's pretty good agreement really.

Plus are the users consistently logging all their symptoms?

A study relying on self recalled data by self selecting participants is statistically likely to be very poor so it's surprising that it's in the same order of magnitude, although that could be entirely coincidental.

Could the zoe app be underestimating cases in the unvaccinated population? I doubt many vaccine refusers use it.

Don't trust Zoe.

Since they were given a government grant about a month ago their narrative has suddenly changed to support Johnson's "let it rip" agenda.

Prof. Spector was saying that we'd reached the peak when cases were still under 10000 a day.

I totally get the argument that Zoe users may be an unrepresentative sample and they are making guesstimates based on responses. It's just that it's so different now, compared to previously. Maybe as people suggest, there has been a falling off of people logging as they get vaccinated, which makes it even more unrepresentative. I hadn't thought of that.