Microsoft Fabric Community Conference 2025, March 31 - April 2, Las Vegas, Nevada. Use code MSCUST for a $150 discount.
Register nowThe Power BI DataViz World Championships are on! With four chances to enter, you could win a spot in the LIVE Grand Finale in Las Vegas. Show off your skills.
Hi,
I am new to PBI and whilst it appears to be relatively straight forward to carry out the simple basics, I am finding it difficult to find a solution that will allow the introduction of a second separate data set to a scatter graph. The process is simple in Excel and allows me to plot 2 independent regressions lines on the same chart.
Would anyone know if this is within PBI's capability?
B rgds
k
HI @KevS,
Currently, you cannot direct add multiple data tables to the scatter charts.
For your requirement, I'd like to suggest you create a new calculated table with merge operation with both two table records, then you can use the merged table as source of scatter chart.
Regards,
Xiaoxin Sheng
Hi Xiaoxin,
Thank you. Would that enable the data sets to plot independently of each other?
B rgds
k
HI @KevS,
You can add a column to mark these records source table, then you can simply recognize where are they from. (you can add a slicer with this source column to choose table records)
BTW, this calculated table will auto-update if source tables record changes.
Regards,
Xiaoxin Sheng
Thank you Xiaoxin,
I will have a play and let you know how I get on.
I think that PBI is an excellent application, but wish that it wasn't so restrictive in manipulating axes and adding independent data to charts. If it had Excel's capabilities in this respect it would be pretty straight forward which would be perfect.
B rgds
k
I think you generally do this with the Legend but not exactly sure what you are going for. Can you post an image of your Excel chart?
Hi Greg,
Just purchased your book. Looks just right for my level of experience and I'm sure that it will be of great use :).
B rgds
k
Hi Greg,
Thank you for your interest. Generally, I like to pitch new data against the existing independently so that it doesn't alter the original regression, and also show the mean data point separately (highlighted in red).
I can't seem to attach a screen shot?
B rgds
k
User | Count |
---|---|
44 | |
26 | |
22 | |
20 | |
17 |
User | Count |
---|---|
51 | |
45 | |
24 | |
24 | |
21 |