A “Hierarchical filter mapping” function has been introduced which, together with a specific hierarchical filter setup, supports advanced wave over wave comparisons in employee satisfaction surveys where the organizational changes between the waves. Below we see a couple of example use cases of special wave over wave comparisons that now are supported when the new process is used:


Example 1, unit split: Team Green existed in 2019 but the team was split up into Team Green 1 and Team Green 2 in 2020. In the 2020 reporting both Team Green 1 and Team Green 2 should be compared to the result of Team Green in 2019

Example 2, unit merge: Team Blue 1 and Team Blue 2 existed in 2019 but the teams were merged into Team Blue in 2020. In the 2020 reporting the Team Blue should be compared to the aggregated result of Team Blue 1 and Team Blue 2 in 2019. 

Example 3, name change: Team Pink in 2019 was renamed to Team Red in 2020. In the 2020 reporting Team Red should be compared to Team Pink in 2019

Example 3, new unit: Team Yellow is a new team in 2020. In the 2020 reporting Team Yellow should be compared to the result of the parent node in 2019 to get a comparison value.


The overall process to be used when these types of wave over wave comparisons are required is show in chapter 1.1 and the functionality of new “Hierarchical filter mapping” page is explained in chapter 1.2.

 

Process

To achieve the special wave over wave comparison the following overall steps need to be followed, see more details about each step further down.

  • Create a Hierarchical filter structure with one branch per wave 
  • Map each node in latest wave, that should have a wave over wave comparison, to one or several nodes in previous wave. The new Hierarchical filter mapping page is used for this
  • In the reports, display the branch of the latest wave only as that branch reflects the current company organization. Also apply a wave split in each chart/table so result is split up per wave

 

Create a Hierarchical filter structure with one branch per year
 
Create a Hierarchical filter structure with one branch per wave. Below we see an example, as shown the Hierarchical filter structure contains one branch for 2019 and one for 2020. To achieve this structure, you need a categorized wave variable in the Hierarchical filter definition, this wave variable can be imported through data file or computed in the system by for example using response date in the expression.


Here we see how the Hierarchical structure needs to be constructed.


 

Map each node in latest wave to one or several nodes in previous wave
 
In the mapping step, in the new Hierarchical filter mapping page, specify the wave over wave comparisons for each node in latest wave. You can apply auto-mappings based on name matches, so nodes with same names in both waves are compared to each other automatically. You can also manually select the nodes to be compared when nodes have changed names, nodes were merged or split up etc.


 Also, an Excel download/upload option can be used if you would like to specify the mappings in Excel and then upload those.

Here we see the new mapping page, see more instructions in next chapter. In this example we see that Team Blue in 2020 should be compared to the aggregates result of Team Blue 1 and Team Blue 2 in 2019.


 

In the reports, display the branch of the latest wave only and add wave filter compare to charts and tables
 
In all the StoryTeller reports, Cross table tool reports etc. select to display the branch of latest wave only as that branch reflects the current company organisation, see example in image 1 below. To get the wave over wave comparison in each chart/table you also need to apply an object level compare filter, see example in image 2 below. Image 3 shows result of both these settings.


Image 1, to the left below we see how the branch of the latest wave (year 2020) is specified to be shown in the report only.


 

Image 2, here we see how a regular filter is used to split result in 2019 and 2020 in a table object.


 

Image 3, here we see the result of settings from image 1 and 2 above. The displayed Hierarchical filter structure reflects the 2020 structure. The Team Blue is selected, based on the mapping definitions in the example previously outlined, the 2019 Top box column shows the aggregated result of Team Blue 1 and Team Blue 2 in 2019