COVID19 Dashboard
This is an interactive dashboard visualization of the 2019 coronavirus outbreak via time-series animation and geomapping.
I used the 'Play Axis' custom visual and DAX measures to create filter settings that would allow for progressive time-series animation. Some data transforms were required to model the source data into a more optimized time-series data model.
The source data set is made publicly available by Johns Hopkins University for educational and research purposes only. Here is the link to their github repository:
https://github.com/CSSEGISandData/COVID-19
https://github.com/CSSEGISandData/COVID-19/blob/master/README.md
The data is updated every day at ~23:59 UTC.
Rob
14 Replies
Hi Rob,
On the maps, I think it might work better to just show the Confirmed value (which I understand includes Recovered & Deaths)? It might be overstating the bubble size to use them together? Also the map pie segments are mostly too small to select accurately, and selecting one segment then filters the other charts by that measure type (e.g. Confirmed only shown in the line charts after selecting a Confirmed segment on the map). With a single value shown on the maps, I imagine selecting a location would show filtered line charts for all statistics.
On the map zoom (my point #3 earlier), my use case is: after using the Slicer to select a Country (e.g. China, Australia), it would be good if the map zoomed to enlarge that location. Assuming that's the built-in Map visual, you can turn on Auto zoom (under Map controls).
Keep up the good (and important) work! Let me know if you need any help.
Regards, Mike
Great work on this important topic Robert - thanks for sharing. The style is very elegant and precise. I appreciate how you've used a point-style map as country-level is less informative.
I had a couple of questions:
- WRT to the chart at bottom-left, I'm not very familiar with the data, but I wonder if the "Recovered" figure should be subtracted from "Confirmed", rather than added?
- The data labels on the line chart at bottom-right arent readable by now, can you add tooltips and/or a Y-axis?
- Can the map zoom when a Country is selected in the slicer?
Hey Mike,
Thanks for the notes.
1. You're absolutely right, the stacked bars make the recovered & confirmed cases appear additive, when they are not. I've reverted this back to a line chart for clarity.
2. I don't follow this one, sorry, as I'm seeing labels on my browser. Are you viewing on a phone?
3. The map zoom wouldn't be super meaningful here unfortunately, as the source data doesn't contain detailed geo data below the country level. (Except for the US and China). At some point Johns Hopkins may start to include more detailed geo data. I agree with you though, I would like to view this at a more granular level as well.
Cheers,
Rob
Thanks for the update. #2 was a browser issue, I was using the Brave browser, which is otherwise OK but does a very poor job on Power BI content. In Chrome those charts look great. Sorry for wasting your time on that one.
This is awesome work! Love the color scheme, but the data in general is laid out in a very useful way
Fantastic...can we get the PBIX file please it's for my learning purpose.
Anonymous Please check my post https://community.powerbi.com/t5/Desktop/Subtracting-the-previous-day-s-value-from-today-s/td-p/977879/highlight/false
Where you can find the PBIX file which helps you to resolve your getting source data issue, and there is also some work done in that file can be helpful in sorting things quickly.
Rob,
Great job! Definitely a lot of options to slice and dice the information within your dashboard.
The progressive animation is a really nice touch!
-Lance M
- Hi Great Job! Please can you kindly share the formula you used to return the value for new cases in US when no state is selected
- Something like this should work for you, depending on your time-series table structure:_us_counties_new_cases_max =VAR __PREV_DAY =CALCULATE(MAX('us-counties-cases-time'[cases]),DATEADD('DateTable'[Date], -1,DAY),DateTable[Date]<=TODAY()-1)RETURNCALCULATE(MAX('us-counties-cases-time'[cases]) - __PREV_DAY)You could improve this DAX formula with some combination of MAX and SUMX, I was trying to clean and transform all of the data and have not refined much of my DAX measures (yet).
- Anonymous, I was able to find my way around it following the steps. Thank you
