Table 1.
Analysis-of-deviance tables (Type III test) from generalized linear mixed models (GLMMs) investigating the effect of hbs-LAB diversity (M1) and cumulative abundance of all hbs-LAB (M2) on the P. larvae spore level. Non-significant terms (italicized) were removed stepwise from the final minimal adequate model starting from the bottom row. “/” means latter random factor is nested within former random factor. The three-way interaction in M1 was removed because this interaction was mainly due to one data point having very high leverage (see text and Fig. S1)
| Model | Model type | Random factor | Response variable | Explanatory variables | Χ2 | Df | AIC | P value |
|---|---|---|---|---|---|---|---|---|
| M1 | GLMM (Poisson) |
Day/ Colony |
P. larvae spore level (CFU) | Quarantine status | 3.4 | 1 | 1376.3 | 0.06 |
| Treatment | 0.0 | 1 | 1378.3 | 0.91 | ||||
| Diversity | 0.2 | 1 | 1380.1 | 0.64 | ||||
| Diversity × Quarantine status | 0.4 | 1 | 1381.6 | 0.48 | ||||
| Treatment × Diversity | 0.1 | 1 | 1383.5 | 0.74 | ||||
| Treatment × Quarantine status | 0.0 | 1 | 1385.8 | 0.92 | ||||
| Treatment × Diversity × Quarantine status | 3.0 | 1 | 1414.6 | < 0.001 | ||||
| M2 | GLMM (Poisson) |
Day/ Colony |
P. larvae spore level (CFU) | Quarantine status | 3.4 | 1 | 1376.3 | 0.06 |
| Treatment | 0.0 | 1 | 1378.3 | 0.91 | ||||
| Cumulative abundance | 0.0 | 1 | 1380.3 | 0.93 | ||||
| Cumulative abundance × Quarantine status | 0.0 | 1 | 1382.3 | 0.88 | ||||
| Treatment × Cumulative abundance | 0.0 | 1 | 1384.3 | 0.94 | ||||
| Treatment × Quarantine status | 1.4 | 1 | 1385.8 | 0.22 | ||||
| Treatment × Cumulative abundance × Quarantine status | 24.3 | 1 | 1408.7 | 0.10 |