When placing a banner or crossbreak question in the columns of a table that includes a summary grid question in the rows, the “column” question’s categories should be placed above rather than below the grid column categories.

When crossing grid questions by banners it nests in a way that not all users like. This request is to have a way of doing the reverse of the nesting without having to resort to data transformation and custom code.

We get quite a few requests to flip the nesting of grid crosstabs. Currently, if you cross a grid variable set by a variable, the variable’s categories are nested within the grid’s columns. This isn’t always how a user wants to display the data and instead want to nest the grid columns within each of the variable’s categories.

Currently, if they want to be able to freely reverse the nesting, they have to construct new variables and create a Binary - Multi Grid. The process is a lot of work and changes the nature of the significance testing. Instead, adding a button that reverses the nesting within the object inspector, similar to swapping rows and columns would allow users to do exactly what they want to do to produce their desired outputs easily.

We get this request fairly often and there’s not a good way of resolving. When a user makes a cross tab using a grid (nominal-multi or grid) variable set in the rows, many times they ask to reverse the nesting. For example below the user would like the 1L, 2L, etc in the spans of the columns to be nested within each country:

!image-20220906-163149.png|width=1621,height=590!

To do this correctly and to keep the stat testing accurate, the user will need to stack their data, make the Observation variable in the stacked data the variable name (1L, 2L, 3L, etc) and create the crosstab from that. Is there a way to utilize our auto-stacking capabilities that we use in our driver analysis to make this an option on the table itself?