In this article, we look at some of the methods which let you do more with your multiple-response data. The options covered in this article are:
Method
Rebasing multiple response questions to the NET
Sometimes you’ll have a Pick Any (binary - multi) question where the percentage in the NET is less than 100%. For example:
To rebase the variable set:
- Select the variable set from the Data Sets tree.
- From the toolbar select Anything > Data > Variables > New > Ready-Made New Variables > Rebase Multiple Response Data in Variables(s) to NET or from the object inspector, click TRANSFORMATIONS > Rebase Multiple Response Data in Variable(s) to NET.
A new rebased variable set is added to the Data Sets tree.
Building awareness or purchase into the base
It is common for researchers to want to choose a different base when analyzing Nominal – Multi questions (single response grids), Binary – Grid questions (multiple response grids), and other questions involving several variables. For example:
We may wish to re-base the data so that it bases the figures only on people who indicated that they have shopped at each supermarket in the past month.
The desired bases are represented in this table:
To rebase the grid:
- From the toolbar, select Anything > Filter > Filter One Variable Set by Another.
- At the first prompt, select the Nominal - Grid or Binary - Grid variable set that you want to filter and click OK.
- At the next prompt, select the question whose categories you want to use as filters (i.e., the question you want to base to) and click OK.
- When asked if you want to split out the question by the filter categories, select No.
- When asked if you want Displayr to match up the filters automatically, click Yes (more on this below).
A new copy of the question is created, where the selections are built in as “filters”. By this, I mean that people who did not select a brand are given a missing value for that brand in the new copy of the grid. The result looks like this:
Splitting up data according to brand
The same option may also be used to split up a variable which contains data for multiple brands. This is common when you pipe brands through the questionnaire. For example, you may begin by asking the survey respondent where they last shopped, and then later questions will all relate to that brand.
Such data will begin like this:
To create a new set of Likelihood to recommend variables, with one for each brand, I can do the following:
- Select Anything > Filter > Filter One Variable Set by Another.
- When prompted to Select the question you want to filter, pick the Q12 – Likelihood to recommend question from the list and click OK.
- When prompted to Select the question whose categories you want to use as filters, select Q10 – Last Shopped and click OK.
And the result is:
Note importantly that there you're now able to crosstab against an additional variable since we no longer need to have the Q12 Recommend variable in the columns.
The option Filter One Variable Set by Another is a bit of a Swiss army knife that can do even more, but these are the two main uses.
Next
How to Rebase Multiple Response Data in Variable(s) to NET