Percentile coloring, also known as a heat map, is another statistical test option available in Harmoni. It uses a color gradient from green to red to visually display statistical highs and lows of significance across a data table.
In this article
1. Percentile Coloring
After creating a crosstab, the percentile icon becomes available in the modify menu.
Clicking the percentile coloring icon colors the cells in the table to show the quintile distribution within a row. It compares the data within each row and divides it into 5 quintiles (colors).
A quintile is a statistical value of a data set that represents 20% of a given population, so the first quintile represents the lowest fifth of the data (1% to 20%); the second quintile represents the second fifth (21% to 40%) and so on.
Harmoni assigns a green color to the cells with the highest values and red to ones with the lowest values. Meanwhile, the remaining values are assigned colors based on the descending value order showing a gradient of different shades falling between green and red.
Reverse Color Settings
To reverse the color settings, access the drop-down menu below the percentile coloring icon and select the Reverse color settings check box.
- After creating an analysis table, open the Modify menu
- Select the Percentile Coloring icon
- The quintile coloring is applied to the cells
- In the drop-down menu, select Reverse color settings if needed
Low Sample
When you apply Low sample warnings to the analysis, data points are shaded or removed. As the sample size decreases, depending on your low sample settings, the coloring fades and are eventually replaced with *.
Single Data Point
If a single data point appears in a row, for example, where data has not been captured across all markets, no coloring will show. A heatmap needs at least two data points to determine a minimum and maximum.
Export to Excel and PowerPoint
You can export a table with percentile coloring to Excel and PowerPoint to share results with others.
Where to from here?
Learn more about customizing your analysis: