As opposed to a measure whose data is entered manually, a calculated measure uses a mathematical formula to calculate values, based on data from other measures.
- A calculated measure, whose data comes from other measures, is called an aggregated measure (parent measure).
- A measure whose data is used for the calculations in a calculated measure is called a disaggregated measure (or child measure)
Note: This article discusses how to create a calculated measure whose data comes from other measures. If your account includes variables, you can also create a calculated measure whose data comes from one or more variables.
There is also a separate article on editing disaggregated measure data.
In this example, the indicator “% of students who graduate high school on time,” will use calculated data.
This calculation requires data from two disaggregated (child) measures:
- Total number of high school seniors
- Total number of high school seniors who graduate on time.
The two disaggregated measures in this example are already created, and are not directly included in any scorecard container. (This is a common use case for creating measures from scratch, not part of a container.)
Note: This example demonstrates how to create a parent measure when child measures are created in advance. While this is the more common way to create an aggregated measure, you can also work the opposite way: linking child measures to a parent measure already set up.
Important: Each child measure must have the same frequency as the calculated (parent) measure. Each child measure must also contain data.
Both child indicators in this example use a custom frequency called “Graduation,” an annual frequency with data recorded every June 25.
Open the parent measure that will use calculated data. In this example, the measure frequency does not match that of the two child measures. Also, by default, measures are Not Calculated. Click the measure’s Edit button.
Open Data Properties and set the correct frequency ("Graduation," in this example). Calculation Type for this example is Ratio x 100 because the desired result is a percentage. If the calculated data is to be compared to target values, be sure to enable Manual Targets. Click Save.
Note: The calculation types of Sum, Average, and Ratio are self-explanatory, and include data from all disaggregated measures added to the calculated measure. There are also several types of accumulated calculations that are performed on only one disaggregated measure.
If you need to use a calculation other than the provided defaults, you can create a custom calculation.
The calculation type, and correct frequency, are now listed in the measure description. To define the first calculation data source (child measure), click Add Existing Measure in the Relationships section. (Or use the Add Variable button if your data source is a variable.)
In the Add Measure(s) To window, set the Relationship to Disaggregated - Child. This means the measures you’re about to add will provide the source data for the calculation.
To find the child measures, you can filter by type, tags, or title. Click Search to see the results. Only measures that have the same frequency as the calculated measure will be available for selection.
Note: If you have access to shared measures from other organizations, those measures can be used for calculations. Use the Instance drop-down at the top to look for measures from another instance.
When selecting the child measures that will be used for the calculation, keep in mind that for some calculations, selection order is important. In this example, because the calculation is a ratio, the measure for graduates should be selected before the measure for total students. (Don’t worry - if you get the order wrong, it can be easily fixed later.) Click Add to continue.
The percent calculation in this example requires two measures, and two measures were added. This automatically executes the calculation. (If only one measure were added, calculations would not run until a second measure is added.)
After the system completes the calculations, the graph for the calculated measure is generated. If the data looks incorrect (as in this example where the percentages are greater than 100), or if you want to add target values, click Edit / Data.
The child measures are listed in the Data section, and you can drag a measure’s three-dot icon to rearrange order if needed. You can also add target values here, if enabled for the measure. Click Save when finished.
Note: When actual values are calculated, they can not be overwritten manually.
After saving, the calculations in this example are now correct, and are plotted against their targets.
In the Relationships section of the calculated measure, the child measures appear under Disaggregated Measures that are used to Calculate this Measure. Child measure order can be rearranged here as well, though you would have to edit the data again and save, in order to recalculate.
Note: There are several ways to edit a calculated measure's disaggregated data.
To start over with new child measures, click Remove All / Disaggregated Measures, then confirm.
When a disaggregated measure is open, its Relationship section shows the calculated measure that uses it, listed under Measures that Aggregate this Measure.
Note: Many users create nested calculations. For example:
- “Calculated Measure 1 ” tracks the sum of high school seniors from multiple schools.
- “Calculated Measure 2” produces the sum of on-time graduates from those schools.
- “Calculated Measure 3” is the percent of all on-time graduates, based on data from Calculated Measures 1 and 2.
Finally, in the scorecard view, calculated measures are indicated by a green dot. If data is missing and so that the values cannot be calculated, this dot will be red.
If needed, you can run all calculations at once, by clicking Options along the top and choosing Initiate Calculations.
Note: For admins, a record of all calculation initiations can be found in the Measure Values Update Log.
