1 of 1 people found this helpful
If you only expect a single column for Ref_area, you would be looking to union.
Joining by default is adding columns from a joined table to the base table. For example in SQL Server:
Select * From Table 1 as t1
LEFT OUTER JOIN Table 2 as t2 ON t1.id = t2.id
This would give you two fields called "id". Tableau is the same but since a field cannot have the same name as another, it provides the table name in parens for the second column you see.
By joining, all your quarterly data is replicated 3 times for each month in the quarter.
OK, I think this is the way to go then.
However, then I should still be able to plot in my current set-up the observation value column from one table on the observation date column of the other... In your example I would e.g. see in a visualisation or table the combination "1 A" when combining columns Ref Area (custom SQL query) and Name, or "1 X" when combining REF Area and City. In my example I think I was not able to see anything from the other table also when I split the window to show both Observation value columns. But could be that there is other aspects I did wrong.
I guess I need to dig further into this in two weeks when I am back from leave.
Many thanks for your graphical explanation above!
Glad to help
Let me know, if you need any help further.
I am back in the office and just played around with this again. Somehow your explanation does not work for me. Even if I reduce my join clauses to only a few fields for which matches should be possible (observation date and reference area, but not the observation value) I have no overlap between the rows retrieved from the two tables. No duplicate value reference area (SQL query1)=reference area is created etc.
Nevertheless, I solved my problem by using calculated fields for both the date and the observation value (and will also create for filters such as the reference area) in the style of "if not isnull(obs date) then obs date else obs date (SQL query1) end". Then I can plot information from both queries in one chart.