Mountain dusky salamander mating

S. Arnold and P. Verrell conducted an experiment at the University of Chicago to study breeding behaviours of mountain dusky salamanders. This mating data was used as an example in the "Generalized Linear Models" textbook (McCullagh and Nelder 1989) .

Usage

data("salamander")

Format

A data frame with 360 observations on the following 8 variables.

Season

represents the season Fall Summer of 1986.

Experiment

experiment number 1, 2, 3.

TypeM

representing the two types of male salamanders being studied. R stands for rough butt, and W stands for white side.

TypeF

similar to the above except for female salamanders.

Cross

represents the cross between a female and male type.

• RR = rough butt female crossed with rough butt male,

• RW = rough butt female crossed with white side male,

• WR = white side female crossed with rough butt male,

• WW = white side female crossed with white side male.

Male

identication number of the male salamander.

Female

identification number of the male salamander.

Mate

represents whether mating has occurred. 1 = yes, 0 = no.

Details

In this example, every variable is either binary or a factor. However, most of the variables in this data set are treated like a numerical variable (i.e., Experiment, Male, Female, Mate), so we suggest turning them into factor variables before use. As outlined in McCullagh and Nelder (1989): This experiment, conducted by S. Arnold and P. Verrell at the University of Chicago, investigated whether geographically isolated populations of mountain dusky salamanders would interbreed. The goal was to see the difference in mating frquencies between the RW crosses compared to the WR crosses. Because each salamander was involved in multiple mating trials with different partners, the observations are not independent. Therefore, we use a model (see the examples section) that conditions on the specific male and female salamanders in the experiment.

Source

(McCullagh and Nelder 1989)

References

McCullagh P, Nelder JA (1989). Generalized linear models. Routledge.

Examples

## Making sure Male, Female, and CRoss are treated as factors
salamander$Male <- factor(salamander$Male)
salamander$Female <- factor(salamander$Female)
salamander$Cross <- factor(salamander$Cross)
## Fitting the model described in 14.5.3 from McCullagh and Nelder
sal_mod <- glmer(Mate ~ (1|Female) + (1 | Male) + Cross, data = salamander,
                 family = binomial(link = "logit"))