How to Analyze Volumetric Data in Displayr

Although Displayr is not specifically designed for volumetric data, such as Sales and Consumption, such data can be analyzed. Typically with this sort of data, you want to treat one observation in the data as more than 1 in your analysis. This type of approach can be useful when you use Displayr to analyze the rest of your survey and want to overlay the volumetric data with the other data (see Multiple Data Files).

There are various approaches depending on how your data is structured, see How to Compute Share of Wallet, Spend, Mouth, etc. (i.e., Volumetric Analysis). Where a questionnaire contains a looped structure (e.g., asking people to list all their purchases in a series of occasions or brands) it is often useful to first stack the data prior to importing it into Displayr. This article explains the stacked data method, so you can recreate volumetric tables like so showing Sales in Displayr:

By converting the data to stacked data:

and creating tables from the stacked data (note the below is a picture from our sister software Q):

Method

1. Setting up the data

Assuming the volumetric data is sales data, you will need to restructure the data so that all the sales data appear in a single column. If doing this in Excel, make sure when doing this that any dates are stored as dates in Excel (e.g., rather than have a cell containing February '15 it should contain 2/1/2015 if you are in the US and 1/2/2015 for most of the rest of the world). Essentially you want your data to look like the second picture above with the volume measure in a single column with the different cuts each in their own column. If your volumetric data is already in a sav file, you can also use Displayr's stacking tool to do this.

2. Creating the weighting variable 

Now that each volume value is in its own row in the data, you can set this data up to be the weight to use to weight your table data. In doing so, instead of each row of the data counting as one in the Sample Size and Count, they will count as the weight/volumetric value. For example, this will count rows 2 and 3 in the stacked data picture above as 100 and 64 respectively. After loading the stacked data set, in the Data Sources tree, click on the volumetric variable (Sales), which should have Structure > Numeric, and in the object inspector check Usable as weight.

3. Using the weight with tables

Once you've created a table to show the cuts of data that you need, you can then weight the data to the volumetric variable (Sales) by selecting this in the Weight dropdown in the object inspector of the table. You can even rename the Count statistic to Sales, see How to Modify Date Labels, Statistic Names, and Other Default Labels.

Next

How to Compute Share of Wallet, Spend, Mouth, etc. (i.e., Volumetric Analysis)