Skip to contents

Factorial Design Contrasts for a Latent Variable

Source: R/plot_design.R
as_design_contrasts.Rd

Decompose the summarized LV design scores into an effect-coded factorial model. The returned magnitudes show how much of the flattened design-score pattern lies in each main-effect or interaction subspace.

Usage

as_design_contrasts(
  x,
  lv = 1L,
  condition_key = NULL,
  factors = NULL,
  block = NULL
)

Arguments

x

A pls_result object.

lv

Latent variable to summarize.

condition_key

Optional data frame with one row per condition and a condition column.

factors

Character vector of factor columns to include. Defaults to group plus every column supplied by condition_key except condition.

block

Optional score block to keep for multiblock results.

Value

A data frame with one row per factorial term.