Working with Aggregate Tables

Aggregate tables are used for reporting questions asked multiple times, e.g. for the first mentioned brand and the second mentioned brand. You can create Aggregate Cross Tables and Aggregate Grid Tables. In order to create an Aggregate Table, we first have to create Aggregate variables.

The Banner Variable

When asking some questions e.g. for each mentioned brand, the data set will contain the variables 'Brand1' which contains the first mentioned brand, 'Brand2' containing the second mentioned brand, etc. All these variables have all possible brands as categories. We start by creating an Aggregate Variable 'Brands' containing all these variables. We will use this variable as the banner variable.

The Stub Variable(s)

For each mentioned brand we have e.g. asked if they offer good service. In the data set we can find the variables 'Service1' containing the answer for the first mentioned brand, 'Service2 containing the answer for the second mentioned brand etc. We also have to create an Aggregate Variable 'Service', containing all these variables.

 

Aggregate Cross Table

We can now create an Aggregate Cross table splitting 'Service' by 'Brands' to get a total overview of how the brands were rated. You can look at this table as the aggregated result of the cross tables 'Service1' x 'Brand1', 'Service2' x 'Brand2' etc.

 

Aggregate Grid Table

If we also have asked if the brand rewards loyalty and if the price is right for each mentioned brand and we want a table containing both 'service', 'loyalty' and 'price' in a grid table, we have to create different Aggregate Variables. In this case we need an aggregate variable for the first mentioned brand containing all the variables we want in the stub, an aggregate variable for the second mentioned brand containing the same variables, but now the one's asked for the second mentioned brand, as in the previous mentioned aggregate variable.

So Aggregate Variable 'BrandProperties1' contains 'Service1', 'Loyalty1' and 'Price1', 'BrandProperties2' contains 'Service2', 'Loyalty2' and 'Price2', etc.

Now we can create an Aggregate Grid Table containing the variables 'BrandProperties1' and 'BrandProperties2' in the stub and 'Brands' in the banner. The resulting table will show the 3 variables contained in the stub variables split by the brands. You can look at this table as the aggregated result of the grid tables 'Service1';'Loyalty1';'Price1' x 'Brand1', 'Service2';'Loyalty2';'Price2' x 'Brand2' etc.