Read on Twitter
I have so much to say about this data visualization. This slide would've been axed at a glance by my old bawse in MR. 
https://twitter.com/ChrislenBulosan/status/1249603660402704389
https://twitter.com/ChrislenBulosan/status/1249603660402704389
-You can't compare periods without showing the base. A 10% of n=50 today is not the same as the 10% of n=30 tomorrow.
-Percentages are only readable at n=30 at least. If you don't reach 30, don't read %, show counts.
-There's probably a different n everyday, guess why.
-Percentages are only readable at n=30 at least. If you don't reach 30, don't read %, show counts.
-There's probably a different n everyday, guess why.
Granted, these %s are based on population of cases in their hospitals, this graph still entails that the daily 100% is the same number of people. People can get admitted and released daily, your 100% 3 days ago will not be same 100% today.
Wala pang significance ng %change yun.
Wala pang significance ng %change yun.
Say Davao had 30 positive cases admitted on Day 1. If on Day 2, 2 people recover and another case is admitted, your 100% becomes n=31. Recovery in Day 2 is 2/31 = 6.45%
Day 3 starts with n=29. If only 1 new case gets admitted in Day 3, your 100% is back at n=30. If 2 people recover on Day 3, it's going to look like 2/30 = 6.67% in this graph.
It will seem like a 0.22% increase in recovered, but it's an "artificial increase". It's insignificant because in both days 2 and 3, the same number of people recovered.
Diyan papasok ang "Test of Significance", #Stat101
kaya kailangan nirereport ang base palagi.
Diyan papasok ang "Test of Significance", #Stat101
kaya kailangan nirereport ang base palagi.
