Lattice dotplot conditional fill color

Question

Problem:

I have a data frame, which I want to visualize with lattice's panel dot plot (not ggplot2). It contains a variable which should be used conditionally to highlight data by different color fill.

Reproducible example:

require(lattice)

# Make reproducable data frame
df= mtcars
df= cbind(car = rownames(df), df) 
rownames(df)= NULL
df=df[1:5, c("car", "mpg", "cyl", "carb")]

df
# output:
#                car  mpg cyl carb
#         Mazda RX4 21.0   6    4
#     Mazda RX4 Wag 21.0   6    4
#        Datsun 710 22.8   4    1
#    Hornet 4 Drive 21.4   6    1
# Hornet Sportabout 18.7   8    2

# I am interested to highlight those data which have carb=1
df[df$carb==1,]

#            car  mpg cyl carb
#     Datsun 710 22.8   4    1
# Hornet 4 Drive 21.4   6    1

dotplot(car ~ mpg | as.factor(cyl), data=df, layout=c(3,1))

This creates a plot:

Question:

I'd like to achieve following plot:

How can I refactor the code to achieve this?

Answer 1

You may try this:

dotplot(car ~ mpg | as.factor(cyl), data=df, layout=c(3,1),
        pch = 19, groups = carb < 2, col = c("blue", "red"))

The groups argument carb < 2 results in a logical vector. Alphabetically FALSE comes before TRUE. Thus, cases where carb < 2 is FALSE get the first colour (blue), and cases where carb < 2 get the second colour, red.

From ?dotplot about group argument:
A variable or expression to be evaluated in data, expected to act as a grouping variable within each panel, typically used to distinguish different groups by varying graphical parameters like color and line type.