This data set is derived from Agresti (2007, Table 3.2, pp.76-77). It gives 4 variables for each of 173 female horseshoe crabs. Also provided are two random samples of the data with replacement, which are useful for illustrating the convergence properties of glm and glm2.
data(crabs)A data frame with 173 observations on the following 6 variables:
Satellitesnumber of male partners in addition to the female's primary partner
Widthwidth of the female in centimeters
Darka binary factor indicating whether the female has dark coloring (yes or no)
GoodSpinea binary factor indicating whether the female has good spine condition (yes or no)
Rep1a random sample with replacement from 1:173
Rep2a second random sample with replacement from 1:173
The variables Dark and GoodSpine are derived from the raw data. In the notation of Table 3.2 of Agresti (2007), Dark = yes corresponds to C>3 and GoodSpine = yes corresponds to S<3. The two random samples Rep1 and Rep2 can be used to provide random samples with replacement from the full data set. These two random samples are useful for illustrating the convergence properties of glm and glm2; see examples in the help documentation for glm2.
Agresti, A. (2007) An Introduction to Categorical Data Analysis (2nd ed.). Hoboken, NJ: Wiley.