Forum Discussion
Comparing Values From Different Non-Related Tables
Yeah, Area Codes are not great when it comes to location based plotting. For example Area Codes for Key West, Florida are 305 & 786 and they both have the same latitude & longitude. It may not be possible since a majority of my calls are in the Dallas area which has three area codes that serve the same coordinates. I wonder if I should find like values are just barely change them so that they are unique like 32.78306, 32.78307, 32.78308. It might slightly throw off my coordinates but by how much?
Or maybe look for a better area code data source.
Anonymous
Ok but you seem to be saying that what is not unique is the longitude and latitude. That should not matter. If the area codes in your Longitude/Latitude table are unique we're fine. Are they?
Can you share the pbix? That would make things faster
Or if you have sensitive info maybe just a pbix with the two tables involved that we are discussing
https://www.dropbox.com/s/k5fwlht3etpe6w1/Template%20Dashboard%20-%20Phone%20Records.pbix?dl=0
Yes. It is personal information but I've grab a limited section to give you a snapshot of the dashboard.
Basically I'm trying to create a heat map of call frequency to certain area codes. So I will need distinct count of area codes as well. I got stumped on the longitude/latitude before I got to that measure.
Was that dropbox link able to work for you?
Anonymous
I had a look but couldn't make much of it. The problem is that the Latitude/Longitude pairs are defined not only by area codes but by area codes and the City. So either:
1. You need to take the city into account to make the lookup (i.e., are code plus City) --> Can the 'Call to' column can be used to extract the city?
2. or you have to make the area codes in the Latitude/Longitude pairs table unique. Otherwise what Latitude/Longitude do you take if there are several for the same area code? How you do this depends on your requirements. Maybe keep one that is representative? Maybe it doesn't matter that much because they are all close to one another? Maybe you need the average?