Covid-19 time-series

[hiom678]

Hi,

I'm working on a covid-19 time-series data with three csv files (confirmed, deaths and recovered worldwide)

I created a workflow in Alteryx which seemed to work fine, but on the recovered data on the last values just comes out as [NULL]. 

When I use the output file in Tableau the numbers for confirmed and deaths are correct, but recovered comes out wrong.

Does anyone know how to correct this? 

I've attached the csv files and my alteryx file

Thanks!

[PhilipMannering]

I think it's because your Confirmed and Deaths files have Canada broken down by province, but NOT for the Recovered file (therefore fewer records) and when you join on record position Nulls appear for Recovered. It's a bit risk joining on record position. I would recommend that you join on Country, Province and Name (Date). 

Hope this explains the mystery.

Philip

[hiom678]

Thanks for your quick reply

I tried to do what you recommended, but then I get the error " String fields can only be joined to other string fields". Is it because I'm using "join multiple" and not "join". If I only use "join", then I can only connect the confirmed and deaths files, and then join "recovered" later..?

[PhilipMannering]

You want to make sure that the data types of fields are the same whether you're using Join or Join Multiple. Just use a Select Tool to specify the same field type. It may be that Name is a string for one and a date for another.

Yes, you can either do Join Multiple or use two Join Tools (I prefer this way because you can see what drops out - in this case... Canada!)

Let me know if any other questions. 🙂

[hiom678]

Thanks again, I have one last question. I decided to use two joins and that worked, luckily.

I'm now missing Canada in my set.. as you also mentioned. Any idea on how to get it back?

[PhilipMannering · Accepted solution]

Because Province is Null for Recovered, you have two options to include Canada. Either,

a) You make the province Null for Deaths and Confirmed

b) You keep province but forgo Recovered for Canada (you could join this on as well, but your numbers will be at an incorrect level of detail).

Both methods attached (and my edits in Turquoise). Hope this helps!

Phil

[hiom678]

thanks!! 🙂