2 way error bars on scatter plot

[ldto]

Hi,

I am attempting to add error bars to a scatter plot of my collected experimental data. Both axes have a sensitivity to the same variable; therefore, the error should be shown on both axes.

However I am running into an issue of error bars only being able to show on aggregated axis, this presents a problem as aggregating both axis will aggregate my scatterplot to 1 point.

I can aggregate one of the axis and get error bars on that:

 

 Here is when both are aggregated, the error bars are there however it all becomes one point:

 

I have tried using marker by Row number however this removes the aggregation and we are back to square one.

 

Hope anyone can help with this :)

 

My data looks like this,

 

 

 

Best,

Louis

[David Boot-Olazabal]

Hi Louis,

Can you share your dataset or dxp file? That would make it easier to have a look at the issue.

Kind regards,

David

[ldto]

Hi David,

Thank you for taking the time.

The workaround I have found is to aggregate both axes and make a calculated column using Rowid() and Use Marker By on that. However, this still fails when adding a display name for the axes titles 

Here is the dxp if you can help enlighten me on what is happening in the background that is causing the issue

 

On the CpCpCq tab you need to filter by one V_0 at a time with BEM Elec and Mech all included and on the next 2 filter our Mech and BEM and use all 3 V_0s

 

Best,

Louis

Dataplots.dxp

Edited June 20 by ldto

[Olivier Keugue Tadaa]

Hi Idto,

 

I think the issue is related to the way you define your upper and lower errors 

Here is a simple setting that I've done 

 

and I can display the below without any further need for aggregation (I've reverted the "marker by" to its default (Row Number) or baserowid() ...) 

I believe you need to review your upper and lower error values and make sure they are relevant to your data in the x-axis and y-axis 

I hope this helps

Dataplots.dxp

[ldto]

Hi Olivier,

Thank you very much, im not quite sure what I did differently the first time but it seems to work now, thank you!

The errors provided need to be differenced with the magnitude of the reading that point; sorry if that may have been a bit confusing

Best,

Louis