About Heat Map Widgets
Heat Map widgets provide an efficient way to quickly identify high points and low points across your organization or various demographic groups. These widgets are ideal for visualizing comparisons, showing how the many levels of a field compare on items across the organization.
Heat Map widgets are mobile accessible, and will resize and reformat as needed in a mobile browser.
Basic Set Up
Before your Heat Map widget will display data, you must set a comparison, choose the field to break out by, and select the items you are interested in analyzing.
- Inside your Dashboard Settings, create at least one Comparison.
- Back on the dashboard, edit your Heat Map.
- Select the Source. This is the Employee Engagement or Lifecycle survey your items will come from. Multiple sources can be selected at once.
Qtip: If you only have one source mapped, you can ignore this step. See Adding Additional Sources if you’d like to add more.
- Under Items, select categories and questions. These items will be listed along the left of the widget. The order of these items is determined by the order the fields are mapped in your dashboard data.
- Scroll down the widget editing pane. Under Breakout, select the field you’re interested in. Field values will be listed in alphabetical order.
Qtip: It’s best to choose demographics, teams, and other tangible groups, such as Departments, Regions, and Tenure. Fields like Employee ID have too many values, and are too specific to the employee for the data to help you create company-wide action items.
- If you have selected multiple Breakouts, the Heat Map can only display one at a time. Select a Default Breakout.
- Select a Comparison.
Displaying Values and Value Ranges
The next step in setting up a Heat Map widget is to determine the values being displayed in the widget, and setting your value range.
Display number values as
+/- Delta is the change from the comparison’s value of the item to the column’s.
Example: In the One Team row, Engineering has a -5. The comparison on the widget is set to the overall company’s scores. That means the Engineering scored 5 less than the rest of the company on the One Team category’s items.
This helps us understand what teams to reach out to to improve engagement, and what items of engagement to focus on. For this category, we can even expand the dropdown to see what within this category Engineering felt the company struggled with.
Base Value will give the percentage of engagement if your Metric is Favorability, and mean score if your Metric is Average. Instead of change, we see the flat value of how high or low engagement was. Therefore, we compare values by color and by looking at the breakout columns versus the Comparison column, instead of the values of the breakout columns being based on the comparison to begin with.
You can additional Decimal Places to the base value as needed.
Determine color values based on & Value Ranges
The Value Ranges will change based on what you choose to base the color values on. They automatically adjust to the range of differences within your data set. Once you reach these items’ maximum score, you won’t be able to add any more value ranges.
+/- Delta is the most intuitive way to change a color range. Instead of arbitrarily choosing the exact engagement values to base color changes on, you are looking at the magnitude of the change. In the example above, smaller changes
Qtip: The color on the upper-left half of the box represents negative change. The bottom-right represents positive change. For example, if Engineering has a delta of +1, the box will be light blue. If it has a delta of -1, it will be light red. The opacity of the color is based on the magnitude of the change.
Custom thresholds allow you to define the exact engagement percentages or averages at which the colors change.
The widget in the screenshot below is the same as the one displayed in the screenshot above, but set to a custom threshold instead – note the difference in how these value ranges are colored.
The color palettes available are designed by SMEs to highlight contrast between high and low values.
The colors in the value range will be assigned automatically based on the number of values you add. The color opacity changes based on the magnitude of the number in each cell of the Heat Map. You cannot adjust custom colors for the palette; for example, if you’re using a red and blue palette, you cannot add yellow or purple value ranges.
In addition to the options required for set up, there are additional changes you can make to the Heat Map widget to clarify the data you see.
The metric is another field that determines how the numbers in the cells are calculated. You can either present results based on engagement or average.
- Favorability: Base your data on favorability. This is calculated as an engagement score, the percentage of participants who rated favorably on the set scale. For more on setting favorability scales, see Scales.
- Average: Base your data on the average value. In a Heat Map, that means you see the average of everyone’s score in a group.
Qtip: Depending on the number of scale points, it may be difficult to get a large range for average. Consider adjusting your value ranges to include decimals.
Show Response Counts
When selected, the number of responses the items in the Heat Map widget received will be listed. This gives you the sample size the calculations are based on.
When displaying comparisons or benchmarks on your widgets, you’ll see a lot of changes from one group to another. But are these changes to be expected, or are they representative of something deeper? How can you decide what changes demand your attention? Thankfully, you can flag whether a difference is statistically significant.
Enabling Significance Testing
- Add a comparison or benchmark to your widget.
- Select Enable Significance Testing.
- Select your Confidence Interval.
Types of Significance Tests
|Comparisons||A two proportion z-test. Here, we are comparing proportions of favorability for two populations.
Qtip: For statistical tests that are set up with comparisons, the test will be performed with the comparison as it is configured, and will not attempt to remove any overlap between the samples being compared. For example, if your comparison is a subgroup vs. the company overall, the company overall includes the subgroup as well.
|Benchmark||One sample test of binomial proportions. Here, we are comparing expected proportion (a benchmark) to the experimental proportion of the binomial question: is this favorable or unfavorable.|
The sample size for categories is the average number of responses across the items in the category.
Understanding Significance in a Widget
The Confidence Interval indicates how confident you would like to be that the results generated through the analysis match the general population. Higher confidence levels raise the threshold for a difference to be considered statistically significant, meaning only the clearest differences will be marked as such.
Once you have enabled significance testing, changes that are significant will have arrows to indicate the direction of the change. Insignificant changes will not have arrows.