winetaste: Wine Tasting Data
In eba: Elimination-by-Aspects Models

Description Usage Format Details References See Also Examples

Paired comparison judgments for two wine tasting studies: ambilight includes the results of a study on the effect of ambient lighting on the flavor of wine; redwines includes judgments on the sensory quality of red wines.

1	data("winetaste")

ambilight A data frame containing 230 observations on 10 variables:

preference, fruitiness, spiciness, sweetness: Paired comparison of class paircomp; judgments for one of the 6 ordered pairs of the blue, red, and white lighting conditions.
age: subject age
gender: factor, subject gender
sensesmell: self-rating of sense of smell and taste.
likewine: self-rating of general liking of wine.
drinkwine: factor, frequency of wine consumption.
redwhite: factor, preference for red or white wine.

redwines A data frame containing 11 observations on 7 variables:

bitterness, fruitiness, sourness, roundness, preference: Paired comparison of class paircomp; judgments for all 10 pairs from 5 red wines: Primitivo di Manduria, Cotes du Rhone, Bourgogne, Shiraz cuvee, Cabernet Sauvignon.
best: factor, Which of the five wines did you like best?
worst: factor, Which of the five wines did you like worst?

The ambilight data are from Experiment 3 in Oberfeld et al. (2009). The redwines data were collected among the members of the Sound Quality Research Unit (SQRU), Department of Acoustics, Aalborg University, Denmark, in 2004. Details of the red wines are available as an attribute of the preference variable (see Examples).

Oberfeld, D., Hecht, H., Allendorf, U., & Wickelmaier, F. (2009). Ambient lighting modifies the flavor of wine. Journal of Sensory Studies, 24(6), 797–832. doi: 10.1111/j.1745-459X.2009.00239.x

eba, eba.order, paircomp.

requireNamespace("psychotools")
data(winetaste)

## No effect of ambient lighting on flavor (Oberfeld et al., 2009)

m <- lapply(ambilight[, c("preference", "fruitiness",
                          "spiciness", "sweetness")],
            function(x) eba.order(summary(x, pcmatrix = TRUE)))
lapply(m, summary)

u <- sapply(m, uscale, norm = 3)
dotchart(
  u, xlim = c(0.5, 2), pch = 16, panel.first = abline(v = 1, col = "gray"),
  main = "Ambient lighting and the flavor of wine",
  xlab = "Utility scale value (Davidson-Beaver model)"
)
ci <- sapply(m, function(x) 1.96 * sqrt(diag(cov.u(x))))
arrows(
  u - ci, c(16:18, 11:13, 6:8, 1:3), u + ci, c(16:18, 11:13, 6:8, 1:3),
.05, 90, 3)
mtext("Oberfeld et al. (2009)", line = 0.5)

## Sensory quality of red wines

psychotools::covariates(redwines$preference)  # details of the wines

m <- lapply(redwines[, c("bitterness", "fruitiness", "sourness",
                         "roundness", "preference")],
            function(x) eba(summary(x, pcmatrix = TRUE)))
lapply(m, summary)

u <- sapply(m, uscale)
dotchart(
  u[order(u[, "preference"]), ], log = "x",
  panel.first = abline(v = 1/5, col = "gray"),
  main = "SQRU red wine tasting",
  xlab = "Utility scale value (BTL model), choice proportion (+)"
)
points(as.vector(
  prop.table(table(redwines$best))[order(u[, "preference"])]
), 1:5, pch = 3)