Percentage points of the bivariate \(t\) distribution for non-positive correlations (Q579796)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Percentage points of the bivariate t distribution for non-positive correlations |
scientific article; zbMATH DE number 4015936
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Percentage points of the bivariate \(t\) distribution for non-positive correlations |
scientific article; zbMATH DE number 4015936 |
Statements
Percentage points of the bivariate \(t\) distribution for non-positive correlations (English)
0 references
1986
0 references
Two well known problems in the social sciences are testing whether two regression lines, corresponding to two independent treatment groups, cross within some specified interval, and testing for a disordinal interaction in a two-way ANOVA. Solutions (based on joint one-sided confidence intervals) have been derived in terms of the percentage points of the bivariate t distribution, but existing tables are generally limited to a correlation \(\rho\geq 0\). This note provides percentage points for \(\rho <0\).
0 references
non-positive correlations
0 references
testing for crossing regression lines
0 references
percentage points of the bivariate \(t\) distribution
0 references
0.7453786730766296
0 references
0.7384182810783386
0 references
