Segmentation: How do I explore a current Segmentation?

Click on the Segmentation icon to see the opened window shown below:

At the bottom of the window is a selection tab that allows you to define the data universe you are working with, similar to other tools within FastStats.

Segments

The first of the 2 tabs towards the top of the window is entitled Segments. Changing the radio button allows you to either segment the underlying data by a particular variable or alternatively use query selections as the segmentation groups (these will be filtered by any restriction made in the underlying selection window).

The above screen shots show that, in the first segmentation, The Gender variable would be used to segment the data in the underlying selection. However, in the second, with the radio button checked against Selections, the queries Gold, Silver and Bronze would be used as the segmentation.

Dimensions

The Dimensions tab is used to define which variables the segmentation will be broken down by and will report on which variables are over and under-represented for each segment.

  • On the Segments tab drag the Gender variable onto the drop box zone

  • Click on the Dimensions tab and drag the Income and Occupation variables on to the window (variables must come from the same table level or above that at which your segments are created). Click the Build button

  • Click on the + symbol next to Income to see the Gender breakdown of the data by that variable

You can pin, add, remove or drag the columns around to change the display (right click on a heading for options). Female data has been grouped together in the screen shot below.

A description of the columns shown in the screen shot above is described below:

  • Description displays the name assigned to that particular variable category

  • Base figure displays the total number of records in this category

  • Base Histogram shows graphically the relative number of records in a category overall

  • Stacked Segments shows graphically the relative size of each segment within each category and overall. The Total row shows the overall size of each segment

  • Category N Penetration displays a histogram with an Index value shown centred around 100. Histogram bars to the left of the 100 centre line show under representation. Histogram bars to the right of the centre line show over representation.

  • Category N is the number of records in the segment in this category

  • Category N Index is the ratio of the Segment N and Base figure percentages multiplied by 100

  • Category N Z-Score displays a figure which is the standardised measure of how confident we can be that the result presented is a true characteristic of the data and not a quirk of the data sample used. For each category, the Z-Score measures the number of standard deviations the result is away from the expected result of the category.

At the bottom left of the Dimensions display, the Word Cloud and Scatter Plot tabs provide some visualisations of the data from the Dimensions Report.

Word Cloud

The Word Cloud gives a quick indication of the most important attributes of that particular segment. The user can choose the segment and the measure to use in showing the Word Cloud. The default is to see the ‘Index’ measure as this gives a variable-independent measure of how over-represented that category is in the given segment. By default you will see the top 10 categories of a segment to illustrate the key attributes of that particular segment.

Scatter Plot

The alternative display is a Scatter Plot chart. If more than two variables have been used in the analysis, the drop down display for Rows and Columns allows you to select which of the variables to be used in the chart.

The chart uses a stratified sample of 1000 records for the display and the colour of each point is assigned to the category with highest number of records. A breakdown of each point can be seen in the tool tip box when the mouse pointer moves over the point.

The Scatter Plot tab only appears when the variable segment option is chosen from the Segments tab

 

Back to Segmentation Overview