View source: R/helpers_cubic.R
caRange | R Documentation |
Identify data points behind E2 and test how many of them have outcome predictions that significantly differ from predictions for predictor combinations on E2 (that have the same level)
caRange( object, alpha = 0.05, verbose = TRUE, model = "CA", alphacorrection = "none" )
object |
An RSA object |
alpha |
Alpha level for the one-sided confidence interval of the outcome predictions on E2 |
verbose |
Should extra information be printed? |
model |
Either "CA" or "RRCA" |
alphacorrection |
Set "Bonferroni" to adjust the alpha level for multiple testing when testing the outcome predictions of all data points behind E2 |
When testing an asymmetric congruence hypothesis with the CA or RRCA model, the caRange
function helps to determine whether the hypothesis is supported for the whole range of realistic predictor combinations. It computes the position of the second extremum line E2 and tests how many predictor combinations are in the data which lie "behind" this line and, at the same time, have a significantly different outcome prediction than points on E2.
When plotting the estimated model (CA or RRCA) with plot
, you can plot the line E2 and the surface above this line by calling "E2" in the options project
and axes
.
Humberg, S., Schönbrodt, F. D., Back, M. D., Nestler, S. (in preparation). Cubic response surface analysis: Investigating asymmetric and level-dependent congruence effects with third-order polynomial models. Manuscript submitted for publication.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.