I like to grab public stats to produce my own spreadsheets and charts rather that rely on alarmist media reports. So far I've concentrated on NHS England deaths by age and morbidity. Now with the hysteria about the recently announced mutation I decided to grab some data on test numbers to go with the "cases". Before producing my scripting to reformat the raw CSV data into a graph I took a look and something odd seems to have happened in early October with a step change in percentage of positives.
I got the data via https://api.coronavirus.data.gov.uk/v2/data?areaType=overview&metric=newTestsByPublishDate&metric=newCasesByPublishDate&format=csv
Here are a few rows around that time which show the jump. Has anybody a clue what may have happened to cause this jump from ~2.5% positive to ~5% in a very short time?
date areaType areaCode areaName Tests Cases % Positive
2020-10-08 overview K02000001 United Kingdom 290104 17540 6.05%
2020-10-07 overview K02000001 United Kingdom 259544 14162 5.46%
2020-10-06 overview K02000001 United Kingdom 265809 14542 5.47%
2020-10-05 overview K02000001 United Kingdom 276603 12594 4.55%
2020-10-04 overview K02000001 United Kingdom 251623 22961 9.13%
2020-10-03 overview K02000001 United Kingdom 288746 12872 4.46%
2020-10-02 overview K02000001 United Kingdom 276740 6968 2.52%
2020-10-01 overview K02000001 United Kingdom 271552 6914 2.55%
2020-09-30 overview K02000001 United Kingdom 261798 7108 2.72%
2020-09-29 overview K02000001 United Kingdom 237300 7143 3.01%
2020-09-28 overview K02000001 United Kingdom 231923 4044 1.74%
2020-09-27 overview K02000001 United Kingdom 267438 5693 2.13%
2020-09-26 overview K02000001 United Kingdom 260297 6042 2.32%
2020-09-25 overview K02000001 United Kingdom 298718 6874 2.30%
2020-09-24 overview K02000001 United Kingdom 274768 6634 2.41%
2020-09-23 overview K02000001 United Kingdom 269428 6178 2.29%
Chaos at the test labs? Have you seen the reports of poor training and contamination?
I like to grab public stats to produce my own spreadsheets and charts rather that rely on alarmist media reports. So far I've concentrated on NHS England deaths by age and morbidity. Now with the hysteria about the recently announced mutation I decided to grab some data on test numbers to go with the "cases". Before producing my scripting to reformat the raw CSV data into a graph I took a look and something odd seems to have happened in early October with a step change in percentage of positives.
I got the data via https://api.coronavirus.data.gov.uk/v2/data?areaType=overview&metric=newTestsByPublishDate&metric=newCasesByPublishDate&format=csv
Here are a few rows around that time which show the jump. Has anybody a clue what may have happened to cause this jump from ~2.5% positive to ~5% in a very short time?
date areaType areaCode areaName Tests Cases % Positive
2020-10-08 overview K02000001 United Kingdom 290104 17540 6.05%
2020-10-07 overview K02000001 United Kingdom 259544 14162 5.46%
2020-10-06 overview K02000001 United Kingdom 265809 14542 5.47%
2020-10-05 overview K02000001 United Kingdom 276603 12594 4.55%
2020-10-04 overview K02000001 United Kingdom 251623 22961 9.13%
2020-10-03 overview K02000001 United Kingdom 288746 12872 4.46%
2020-10-02 overview K02000001 United Kingdom 276740 6968 2.52%
2020-10-01 overview K02000001 United Kingdom 271552 6914 2.55%
2020-09-30 overview K02000001 United Kingdom 261798 7108 2.72%
2020-09-29 overview K02000001 United Kingdom 237300 7143 3.01%
2020-09-28 overview K02000001 United Kingdom 231923 4044 1.74%
2020-09-27 overview K02000001 United Kingdom 267438 5693 2.13%
2020-09-26 overview K02000001 United Kingdom 260297 6042 2.32%
2020-09-25 overview K02000001 United Kingdom 298718 6874 2.30%
2020-09-24 overview K02000001 United Kingdom 274768 6634 2.41%
2020-09-23 overview K02000001 United Kingdom 269428 6178 2.29%
Thats really noisy data, a rolling average might help.
But we're dealing with a season virus - thats the time they always start to spike. Plus schools had ben back a few weeks to re-seed it.
The data system fell apart in early October due to a line limit in excel.
The data system fell apart in early October due to a line limit in excel.
