Group Tool

Use the Group tool to create a sub branch of calculations that have a different resolution level to the master segment.

This tool allows you to create a sub-select to assist with building more complex query logic, involving different resolutions to the master segment, and which can contain single or multiple lines of calculation using the same tools that are available to normal segments.

The group can then be joined with the master statement with an And, Or, or Exclude join.

This tool is useful for creating segments of records that have a different resolution level to the master segment, for example building up a list of product orders based on a variety of criteria before joining with the master segment (resolved at the Customer level) to discover which customers made those orders.

This tool also provides a quick method of adding additional queries to your selection, as if you have placed parentheses around them.

Groups are slightly indented in the segment tree so they can be easily identified. They can also be saved as templates for reuse in other calculations.

When creating a group it is important to understand how the figures are calculated.

The screen shot below shows a segment with a resolution level at the household table containing rows from the household table Country = England, and the Customer table, Gender = M and Occupation = Director:

The total segment count is returned as 5, 356. The count returned shows us the number of households in England who have a male and a director in them. Note that these directors can be male or female.

To return a count of households in England that contain male directors it is necessary to group together males and directors at the Customer resolution level. This essentially puts brackets around males and directors to calculate them together. In addition the whole segment should be resolved at the Household level to obtain the household contain of customers who are both a male and a director.

Here the count is 5,336 which is twenty less than the first example. This is because of the questions asked. In the first example the resolution level for the entire document was set at the household level, and therefore the question as was to provide a count of households in England that have a male and a director in them.

Using the group tool is the only way in which you can change the resolution level of a subselect (the items in brackets). By using the group tool in this example we are able to obtain a count of male directors. By then setting the document to the household resolution level we are asking the question of the number of households in England that have a male director.

By using the group tool the count returned was 20 lower in the second example than in the first meaning that there are 20 more households in England that have a male and a director in them, than have a male director.

• Sorting and Grouping tools
  • Group
  • Sort Segment