https://eduhseru.sharepoint.com/sites/AdvancedDataAnalysis/Shared%20Documents/General/St.P._stats.jpg

Anything wrong?

Firstly, if we are talking about positive and negative events it is good to use contrasting coloring for the chart. In this case: red for the number of new infected and green/blue for thу recovered.

I personally like the idea of using barplots. It clearly tells the story and shows the dinamics, but i do not see the reason for placing statistics of recovered cases in the bottom part with a negative direction. From the basics of math we know that 2-dimentional charts have x-axis and y-axis with up-right directions. If the data is placed in the bottom or left part it has to have negative value. So in this case the direction of recovered cases is misleading.

a couple of minutes later:

Well, during filling the data i’ve came with an idea, that this graph would be better as a line-chart. Surely. Let’s try that.

The next misleading information that i found is the number of infected/recovered cases under the title. At first it looks like the total number, but if you look at the graph - it is only thу number from the last observation (date). Of course it is a pleasure to see ‘not so big’ difference, but to be honest with the reader i have counted the total number from the observations and have placed it under the title.

What is the purpose? I have found out that with the line chart we can show the dinamics of cases (the fact that their number grows). The original chart has lebels with daily numbers of cases, so i have submitted the garph with labels either, but in addition i’ve made the second chart without labels but with linear regression line (see below). I guess that might be interesting to know, too. Especially to learn if there are any big differences in this dynamics for infected and recovered patients.

28/08 29/08 30/08 31/08 1/09 2/09 3/09 4/09 5/09 6/09 7/09
infected 189 186 188 185 189 191 193 195 192 196 210
recovered 95 68 96 86 106 112 101 96 75 65 123 

## 'data.frame':    22 obs. of  3 variables:
##  $ date: Factor w/ 11 levels "28/08","29/08",..: 1 2 3 4 5 6 7 8 9 10 ...
##  $ N   : num  189 186 188 185 189 191 193 195 192 196 ...
##  $ type: Factor w/ 2 levels "infected","recovered": 1 1 1 1 1 1 1 1 1 1 ...