Scatter plot with two nominal variables

[xzhang32]

I have a data set that contains three columns -- Sample ID (1 to 100), Experiment ID, and Result (between 0 to 1).

I would like to plot a scatter plot, x axis as Experiment ID, y axis as Sample ID, and the Resultwill be represented by color grade. Demo plot from other programs is given as following.

Could you please show me how to plot that in Power BI? Power BI tends to group the Sample IDand Experiment ID, and choose the details as Experiment ID seems not get what I want. Any hint or instruction will be helpful, thank you!

[kevcurtis · Accepted solution]

Hi @xzhang32,

First, ensure when connecting/importing your data that each column has the correct data type.

• Experiment ID = Whole number (remove the C from the beginning if possible - scatter plots don't like non-numerical values)
• Sample ID = Whole number
• Result = Decimal number

After adding Experiment ID as the x-axis set this attribute to "Don't summarize"

Do the same operation for the y-axis with Sample ID: Don't summarize

Add the Result column in the "Color saturation" shelf and then you can define a divering colors in the visual options to represent the Result

After applying these steps the data points in the scatter plot won't be summarized (Sum, count, etc.) and this should help you achieve your intended visual.

Sample data:

Experiment ID	Experiment ID_N	Sample ID	Result
C402148	402148	83	0.000
C402148	402148	84	0.006
C402148	402148	85	0.010
C402148	402148	86	0.005
C402148	402148	87	0.000
C402148	402148	88	0.000
C402148	402148	89	0.012
C402148	402148	90	0.022
C402148	402148	91	0.012
C402148	402148	92	0.022
C402148	402148	93	0.030
C402148	402148	94	0.012
C402148	402148	95	0.018
C402148	402148	96	0.016
C402148	402148	97	0.000
C402148	402148	98	0.000
C402148	402148	99	0.018
C402149	402149	1	0.014
C402149	402149	2	0.025
C402149	402149	3	0.024
C402149	402149	4	0.029
C402149	402149	5	0.030
C402149	402149	6	0.025
C402149	402149	7	0.005
C402149	402149	8	0.017
C402149	402149	9	0.023

Visual:

Let me know if this is helpful or if you have any questions.

Thanks

Kevin