I am working with a data set that uses coordinates with the intention of making a map. I had this working successfully in the previous months but it no longer works with the same data set. We did need to migrate one of the underlying data sets to a new data source, but the query is identical to the original.
In this case, the underlying tables all have the coordinates in the right format to be read (ex: 28.6934). However, when I put them into a QS Map, the values seem to get divided by 1000 (same coordinate is showing as 0.0286934).
I have the data set as longitude/latitude in the dataset itself. When i put these values into pivot table analysis visual, it shows the values normally. However when i put it in the map, they are all clustered into a small area off the west coast of Africa in the Gulf of Guinea (actual population is in North America).
I am mystified as to how this happened? Attached are some images as well.
First here is the analysis. Above is the visual of the pivot table with the latitudes and longitudes showing as incorrect values and thus the result on the map with that cluster near Africa.
This is the data set, showing the correct values for coordinates. I can confirm this is the latest version/refresh of the data showing on the above analysis screenshot.
Its a bug in QuickSight. The most recent instance similar to yours had an input that the issue started occuring when the dataset had joins between tables. Your case seems similar.
The solution is to create a calculated field to multiple the latitude/longitude by 1000 and use the same in the map
