Table 1. Ranking of AICc values of 16 competing
models, with different sets of explanatory variables for the
multiplicative population growth between two consecutive years. The
models build on a null model with log link function and COM-Poisson
error, where the number of territories is (technically) the response
variable, explained with region and an offset variable, which is the
natural logarithm of the number of territories previous year (i.e.,
R-syntax for the formula: TERRI ~ 0 + region +
offset(log(TERPRE))). The column “Additional explanatory variables”
describes which other fixed effects variables are included. “K” is the
number of estimated parameters, “logLik” the log-likelihood,
“AICc” the Akaike information criterion corrected for
small sample size, “ΔAICc” the difference in
AICc compared with the most parsimonious model, and
“w” is the Akaike weight.