How can I aggregate data?

The system supports data aggregation (summation) where the values for child elements within a given dimension are collectively summed together to form the summary values of the respective parent elements.

For example, consider a dataset with a geography dimension consisting of subnational regions grouped by countries in a parent/child hierarchical fashion. Through data aggregation, the system would take the raw data of all subnational regions for a specific country, compute the sum of all numeric values reported, and save the resulting values as that country’s data.

There are two basic types of data aggregation: hierarchical aggregation and preset aggregation.

Hierarchical Aggregation

You can aggregate data for all parent-child groups within a data dimension. Aggregation would take in values of all child elements within a given dimension, compute the collective sum of all numeric values, and use this to form the values of all respective parent elements.

To run hierarchical aggregation, please follow these steps:

1. Open Dataset.
2. Click More Actions > select Data Aggregation.
3.  Provide the aggregation script <Dimension Name>:*  in the script box and click the Submit button.

Note:  If your dataset’s dimension name is spelled with multiple words and spaces in between such words, please replace these spaces with hyphens(-). For example, if the given dimension is called “Trade Flow”, then the Aggregation Script will be: Trade-Flow:*.

Note:  Aggregation will not work if the dimension does not consist of a parent-child hierarchy.

Once aggregation is completed, you will be redirected to the completion page. Click the Continue link to go back to the Dataset Page.

Preset Aggregation 

Preset aggregation is a special type of aggregation used specifically for geographic dimensions. You can use preset aggregation to sum up raw data for any preset list of locations within that dataset.

Note: while preset aggregation does not require your geographic dimension to have any parent-child hierarchy, in your regular dataset file, you must specify the preset group name (like COMESA), along with its code.  If the dataset's structure doesn't contain the Preset group name, the aggregation will not work. 

To run preset aggregation, please follow these steps:

1. Open the dataset.
2. Click More Actions > select Data Aggregation.
3. Provide the aggregation script <Dimension Name>:* <Preset Group Name> in the script box and click the Submit button. 

Note: You may add elements to dataset's structure without data, but these elements will not be shown in Dataset Viewer.  Once aggregation is done you will see the aggregated item in Dataset Viewer.