This guide will go into more detail about how Quill for Power BI analyzes scatter plot data and performs regression analysis.
Generating scatter plot narratives
When you select at least two or three measures and a single dimension for your narrative, 'Scatterplot' will be an option for the story selection prompt.
Scatter plot narratives with two values
In this example, we have a scatter plot with a single dimension (City) and contains two values (Sum of Labor Costs Variable and Sum of Revenue).
Order of the values matters: The first value in the sidebar is what will be considered the x-axis in any regression analysis. The second value will be considered the y-axis value. Therefore, the narrative will analyze the impact of the independent variable (Sum of Labor Costs Variable) has on the dependent variable (Sum of Revenue).
Scatterplot narratives with three values
In this example, we have a scatter plot with a single dimension (City) and contains three values (Sum of Labor Costs Variable, Sum of Revenue and Sum of Travel Expenses).
Similar to the two dimensional scatter plot narrative, order of the values on the right does matter. When scatterplot narratives are created with x,y and z axis, analysis is typically focused on the impact that x and y have on the z-axis.
In our example, the relevant analysis is answering the question, "What was the impact each of Sum of Labor Costs Variable and Sum of Revenue had on travel expenses (if any)?
If a user wants to change those independent and dependent variables, the values on the right sidebar will need to be removed and re-selected in the proper order. In this case, the narrative does not act like the visualization. If you switch the order in the sidebar, the narrative will not change.