Summarise rating scale responses with Group Variables

Group variables are used to create a summary analysis of questions which share a common set of answer choices, making it easier to evaluate overall responses to questions in a grid, and make comparisons between separate question grids in your survey.

This worksheet shows how to:

  • Build a horizontal bar chart of rating scale responses
  • Create an ‘Overall’ Group Variable to combine responses from all questions
  • Add the ‘Overall’ category to the Horizontal bar chart

This example uses the ‘Satisfaction’ survey supplied with Snap. Alternatively, use any survey which includes one or more question grids.

Step 1 – Build a horizontal bar chart of rating scale responses

Start by building a horizontal bar chart to summarise the responses to a typical question grid.

  1. Click BuildChart to build a chart.
  2. Select the Style ‘Horizontal Stacked Bar Percent Transposed’.
  3. Type the grid questions into the analysis field.
  4. Select the option Analysis Percents.
  5. Select ‘Transpose‘.
  6. Click ‘OK‘. The chart will look like this:
  7. Click Save to save. Note the name of the chart as you will need to retrieve it in a later step.

Step 2 – Create an ‘Overall’ Group Variable to combine responses from all questions

  1. Click AnalysisVariables to open the ‘Analysis Variables‘ window.
  2. Click Plus and select ‘New Group Variable’:
  3. The group variable will be given the name ‘GV1‘. Type ‘Overall’ in the label field.
  4. Type the list of questions to be ‘grouped’ into the source field (if you are using the satisfaction survey supplied with Snap, type ‘Q8a to Q8d’).
  5. Click Save to save.

Step 3 – Add the ‘Overall’ category to the Horizontal bar chart

Now you can add the group variable to your horizontal bar chart.

  1. Click SavedAnalyses to retrieve the saved bar chart.
  2. Click Properties to open ‘Analysis Properties’.
  3. In the Analysis field, type ‘GV1‘ followed by a comma, immediately before the grid questions:
  4. Click ‘OK’. Your chart will now include an ‘Overall’ category.

What else could I do with group variables?

Create a chart to summarise overall responses to multiple question grids

If your survey contains more than one question grid, try repeating the steps above to create a group variable for each one (GV1, GV2, GV3 etc). Then, build a Horizontal bar chart including the group variables only.

Create an overall satisfaction index of responses

Create a new group variable including questions contained in multiple question grids, for example type ‘Q8a to Q8d, Q9a to Q9d, Q10a to Q10d’ into the source field (if the grid questions are sequential, you could also type Q8a to Q10d). This will produce an overall score for all grid questions– perfect for a top level summary in your report.

Further Resources

For a detailed introduction to analysis in Snap, including a description of all variable types, how to create data filters and more, view the ‘Analysis Essentials’ interactive tutorial. If you’d like to learn more about reporting your survey results, including how to add your own analyses to the default reports, watch the ’Viewing, customising the default Questionnaire and Summary reports’ interactive tutorial.

Other worksheets you may be interested in:

Find more resources in the ‘Analysing Rating Scale Questions’ section of the Support Hub.