COMURE: The Use of Linguistic Variants in Translations vs....

In corregp: Functions and Methods for Correspondence Regression

Description

This data set was a case study in the COMURE project ("corpus-based, multivariate research of register variation in translated and non-translated Belgian Dutch") which was conducted at the Department of Translation, Interpreting and Communication of Ghent University between 2010 and 2014.

Format

A data frame with 3762 rows and 5 variables.

• Variant The linguistic variant used in a set of alternatives (27 levels).

• Variable The linguistic variable specifying a set of alternatives (13 levels).

• Variety The dichotomization of Variant into standard and non-standard.

• Register The register or "Text type" of the data (6 levels).

• Language The language (and source language) of the data (3 levels).

Source

Delaere, I., G. De Sutter and K. Plevoets (2012) Is translated language more standardized than non-translated language? Target 24 (2), 203–224.

Examples

data(COMURE)
# The execution of corregp may be slow, due to bootstrapping:  
comure.crg <- corregp(Variant ~ Register * Language, data = COMURE, part = "Variable", b = 3000)
comure.crg
summary(comure.crg, parm = "b", add_ci = TRUE)
screeplot(comure.crg, add_ci = TRUE)
anova(comure.crg, nf = 2)
comure.col <- ifelse( xtabs(~ Variant + Variety, data = COMURE)[, "Standard"] > 0, "blue", "red")
plot(comure.crg, x_ell = TRUE, xsub = c("Register", "Language"), col_btm = comure.col, 
  col_top = "black")

corregp documentation built on May 1, 2019, 10:08 p.m.