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 Z-Test.
Forsta Visualizations uses Z-test for significance testing of 2 independent samples. The test can be used for comparing categorical means or categorical proportions.
Null hypothesis used is:
?1 − ?2 = 0 (compare proportions)
Or
?1 − ?2 = 0 (compare means)
Test function used is:
Or
Restrictions:
- The compares samples must be independent
- When comparing proportions each sample size must fulfil the restriction:
Assumptions:
- Distributions of tested variable is unknown.
- The size of total population is infinitive. (? = ∞)
- The samples are drawn randomly where all respondents have the same probability of being selected.
Note: If the object is showing unweighted data, the Sig Test will be unweighted. But if the object is using weighted data, we have 4 different versions of the test.
- Weighted
All ingoing values in the formula are calculated with the selected weight.
- Weighted – effective base Effective base = the squared sum of weights divided by the sum of squared weights
- Unweighted – Option A
All values in the denominator in the formula are unweighted. (Yellow marked)
- Unweighted – Option B Only the base (n) in the denominator are unweighted.