Table 1 

Comparison of alternative selection models showing the number of parameters (k) in each model. 

Traits in the model 
k 
AIC 
Δ_{i} 
Likelihoods 
w_{i} 
Evidence Ratio 


Preferred model 
7 
893.91 
0 
1 
0.6066 

Head Height 
8 
895.7 
1.79 
0.4099 
0.2478 
2.4473 
Tail Muscle Height 
9 
897.69 
3.78 
0.151 
0.0916 
6.6194 
Head Height^{2} 
10 
899.81 
5.9 
0.0523 
0.0317 
19.106 
Tail Muscle Height^{2} 
11 
901.23 
7.32 
0.0257 
0.0156 
38.8613 
Tail Fin Height^{2} 
12 
902.95 
9.04 
0.0108 
0.0066 
91.8356 


AIC score = Akaike Information Criterion; Δ_{I }= difference in AIC scores between the best model and subsequent model; w_{i }= normalized relative likelihood of the model given the data; Evidence Ratio = the relative odds that a model is the best given the data. The first row of the table represents the preferred model, which includes linear terms for Head Length, Tail Length, and Tailfin Height, and quadratic terms for Head Length and Tail Length (see Table 4). Subsequent rows show the influence of adding each indicated trait in succession. Traits labeled with a superscript "2" represent quadratic terms. 

Calsbeek and Kuchta BMC Evolutionary Biology 2011 11:353 doi:10.1186/1471214811353 