For the horseshoe crabs data file, fit a model using weight and width as explanatory variables for the probability of a satellite.

a. Conduct a likelihood-ratio test of H0: β1 = β2 = 0. Interpret.

b. Conduct separate likelihood-ratio tests for the partial effects of each variable. Why does neither test show evidence of an effect when the test in

(a) shows very strong evidence?

c. Use purposeful selection or AIC to build a model when weight and the spine condition and color factors are the potential explanatory variables.