In this article
Overview
Use the Calculations tab to apply different calculation types based on the question types you've selected. You'll only see options relevant to your current setup—keeping things simple and focused.
Calculation Types
Calculation options adjust automatically based on your selected variables:
Categorical Questions
Percentage share and Count: Visible when at least one categorical question is selected.
Mean: Available if a categorical scale question is included.
Significance Testing: Shown when a categorical question is selected.
Numeric Questions
Numeric-specific calculations (like Sum, Mean, etc.) only appear when a numeric question is selected.
Correlation Analysis is available if you’ve selected at least one numeric or single-choice categorical question. It won’t appear with multiple-choice or index-only questions.
Standard Calculations Panel
The panel displays different options depending on the question types:
For categorical questions: Percentage share, Mean, Count
For numeric questions: Sum, Mean, Median, Min, Max, Percentile
You can:
Choose which calculations to use.
Adjust decimal places per calculation type.
Add prefixes or suffixes (like % or units) to your values.
If weighting is enabled in your project, you can also select which weight variable to apply or turn weighting on/off.
Correlation Analysis
Correlation analysis uses Pearson’s formula, just like in StoryTeller.
To use:
1. Choose variables in the Rows and Columns tabs.
2. Open the Calculations tab and enable Correlation Analysis.
3. Choose whether to show R or R² and pick the question to test against.
4. Adjust decimals, unit, and suffix if needed.
5. Click Generate to run the analysis.
6. The information above the table shows which question the correlation analysis is made against.
Significance Testing
Significance tests help compare values across columns or rows.
To set up:
1. Select your Rows and Columns.
2. Turn on Significance Testing in the Calculations tab.
3. Choose your significance level: 1%, 5%, or 10%.
4. Pick a base size (weighted, unweighted, or effective).
5. Decide whether to test across columns or rows.
6. Choose comparison columns.
7. Add base or respondent count limits, if needed.
8. Select how to show the results.
9. Click Generate.
Note: If results are weighted, you’ll see base size options. For unweighted data, the unweighted base is used automatically.
Visualization Options
You can display significance test results using:
Symbols (e.g., a+, b–) for all-vs-all tests
The significance test result can be visualized with colored values when a value is tested against another value. This option is not available when all values in the table are tested against each other. The value colors for positive and negative results are defined in the Project settings page, as shown in the image below.
Colored cell backgrounds or values for one-vs-one tests
Choose "Colored background" under “Show result as” to use color cues for comparisons. Positive and negative value colors are defined in your project settings.
Benchmarking
Benchmarks let you compare one result against another.
To use:
1. Turn on Benchmark Calculation in the Calculations tab.
2. Choose the calculation type:
Units (e.g., B – A = -30)
Percentage share ((B – A) / A * 100)
Index/Conversion rate (B / A * 100)
3. Decide whether to show the original value because sometimes you may only want to see the benchmark results in isolation.
4. Choose whether to benchmark across rows or columns.
5. Select the comparison column(s) i.e.which column or row are you benchmarking against. You can specify this with references such as first row/column or next row/column.
6. Choose whether to hide the comparison values.
7. Select whether to show the benchmark value inside the table cell next to, or below, the existing value, or to show it separately in a different cell.
8. Click Generate.
You can apply benchmarks to any result type, including Percentage share, Mean, Count, and Correlation.