Group Node
The group node can be used to group data according to a specific enumerable column. As an example, the data might contain a column "task", containing the values "task1", "task2" and "task3", and each participant might have performed that specific task at a different point in time, or even returned to a task after performing another task in between.
Now, if you want to compare how the different participants performed during a specific task, this would be difficult to achieve without the group node. With the group node, it becomes possible to just look at the data from a specific task for each participants, regardless of when the task was performed.
Output
The group node outputs the same data as on its input, but grouped according to the selected column. The output has a distinct data set for each value in the selected column. If more than one column is selected, there will be a distinct data set for each existing combination of the values of the selected columns. So in the example above, the output would contain a data set for "task1", "task2" and "task3", each containing only the data where the task column of the input data had the respective value.
Aggregation
The group node is not able to aggregate, because it only performs a reorganization of the input data, and thus a computed aggregation can not exist.