In this article
Forsta Visualizations supports a Z-test, a T-Test and a norm/population test. Below you can find all details regarding the norm/population Test.
Norm/Population Significance test calculation
The “Norm/population test” is designed for testing survey calculated values against known population values. This test is often referred to as a Population test. However, as long as the Norm value is treated as a fixed value, then the test can be used to compare survey calculated values with Norms as well.
The test is only available in the Storyteller charts and tables and has not been added to the Cross Table Tool 2.0 since the setup is too complex in a “Do It Yourself” cross table tool. The setup of the Norm/Population test in Storyteller charts/tables is more complex than the regular Two independent sample Z-test. This is because a survey calculated value (a percentage share or a mean value of a categorical question) is tested against a Norm/Population value which often (or always) has been imported to Forsta Visualizations as numeric aggregated value stored in a numeric variable. This means that the test has to be done across series in a chart or table as one series cannot show the results for both categorical variables (survey result) and numeric variables (population/norm result).
Formula used for proportions:
Formula used for mean values:
Critical values, n>100
The following critical values are used when base size of tested value is larger than 100.
Critical values, n=<100
The list below shows the critical values used when the base size of the tested value is equal or smaller than 100. Degrees of freedom=n (the base size of the test value)