fiveBeersHedonics: Five tasters evaluated how much they liked (using a 10 point...

In HerveAbdi/data4PCCAR: Some data sets and R-functions for PCA and CA to accompany Abdi & Beaton (to appear, 2023) Principal Component and Correspondence Analysis with R

Five tasters evaluated how much they liked (using a 10 point rating scale) five beers.

Description

fiveBeersHedonics: five tasters evaluated how much they liked five beers using a 10 point rating scale, with "1" meaning completely disklike and "10" completely like.

In order to be analyzed with correspondence analysis, these rating scale data need to be recoded with what is called fuzzy coding, doubling, or thermometer coding. With this code, each rating is re-expressed as two variables: The first one expresses how far the rating was from the low value of the scale and the second one expresses how far the rating was from the high value of the scale. This way, the original variable is now represented by a line linking the low anchor of the scale to the high anchor of the scale.

This data set provides the original data and the "fuzzy-recoded" data.

Usage

data("fiveBeersHedonics")

Format

A list containing 2 data frames:

• ratings: A 5 by 5 data frame with the original ratings performed using a 10 point rating scale.

• fuzzyRatings: A 5 by 10 data frame with the doubled ratings (with values between 0 and 1).

Author(s)

Hervé Abdi

References

These data were obtained from a beer ratings article published sometimes in the early 21st century by the Dallas Morning News.

HerveAbdi/data4PCCAR documentation built on Sept. 11, 2022, 4:19 p.m.