Skip to main content
. 2016 Jun 24;118(3):459–466. doi: 10.1093/aob/mcw118

Table 2.

Models explaining mite and mildew variables according to AICc selection for V. riparia manipulations

Rank Intercept Leaf length Foliar sugar Leaf length × Foliar sugar d.f. logLik AICc ΔAICc AICw
Mite abundance pre-inoculation
1 3·92 –0·05 + + 4 131·38 273·43 0·00 1·00
2 3·33 + 2 –141·54 287·78 14·35 0·00
3 3·19 0·01 + 3 –141·16 289·82 16·38 0·00
4 3·21 0·03 2 –157·08 318·87 45·44 0·00
5 3·58 1 –159·97 322·17 48·74 0·00
Mite abundance post-inoculation
1 1·39 0·12 + + 4 –72·35 155·56 0·00 0·99
2 2·11 0·06 + 3 –78·61 164·83 9·27 0·01
3 2·24 0·07 2 –88·50 181·75 26·19 0·00
4 2·83 + 2 –88·61 181·98 26·42 0·00
5 3·14 1 –103·11 208·46 52·90 0·00

The ‘delta’ column shows the difference between a model’s AICc and that of the highest ranked model (rank =1). The best fitting models (ΔAICc >2) are in bold. Model coefficients are shown for leaf length, and plus signs are shown for factor variables.