The purpose of this article is to introduce you to how significance tests are conducted in Displayr and provide guidance on where to find more detailed information on the various aspects of testing.
Statistical Significance Testing in Tables
There are two general ways that Displayr does significance testing in a table. The default method is to show results from the cell comparison method using font colors and arrows, but you can change that to the more traditional column comparison method in the Significance drop-down under Properties in the object inspector. You can also modify the underlying assumptions for these tests in the Statistical Assumptions settings or go to Properties > Rules > Significance Testing in Tables to apply other rules.
| Cell Comparisons | Column Comparisons |
|---|---|
|
|
To work out which test has been conducted on the cell of a table:
- Check to see if any Rules have been applied to a table. If they have, review their documentation.
- Select the cell (or both cells if comparing columns).
- In the object inspector, go to Properties > Significance and click the
. Alternatively, you can select a cell or multiple cells in your table, right-click and select Statistical Test.
Links for further information
More details on the various aspects of statistical testing in Displayr can be found in the following links.
- A general overview of how Displayr selects and performs a statistical test is here: Overview of Statistical Testing
- A guide to what settings and statistical assumptions are available in Displayr is here: Statistical Assumptions
- Info on How to Apply Significance Testing in Displayr.
- Learn how to read tables and interpret significance testing results:
- For resources on troubleshooting statistical testing in Displayr, you can go to our How to Investigate Your Statistical Significance Testing page.
- Many times results differ due to:
- Overlapping data -- overlapping respondents are by default removed when doing column comparisons which may not be done in other software.
- Multiple Comparison Correction -- helps reduce false positives found by running lots of significance tests by adjusting the critical p-value to make it harder to be found significant.
- Many times results differ due to:
Other technical resources
Technical Assumptions of Tests of Statistical Significance contains a general discussion about the use and interpretation of tests of statistical significance.
Next
How to Specify Columns to be Compared in a Table
How to Conduct Significance Tests by Comparing to Previous Time Periods
How to Change the Overall Confidence Level at which Significance Differences are Displayed
How to Replicate SPSS Significance Tests in Displayr