2 min read

Using a two-stage approach (Continuous predictor)

Difference

Coefficients

1.7 °C vs. 3.4 °C

## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: diff ~ heat + model + (1 | species)
##    Data: df_2trt
## 
## REML criterion at convergence: 7638.5
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -4.9756 -0.5160  0.0363  0.5486  5.5258 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  species  (Intercept) 0.01466  0.1211  
##  Residual             0.48457  0.6961  
## Number of obs: 3591, groups:  species, 20
## 
## Fixed effects:
##               Estimate Std. Error         df t value Pr(>|t|)  
## (Intercept)  6.149e-02  3.251e-02  2.529e+01   1.891   0.0701 .
## heat++      -4.273e-04  2.326e-02  3.571e+03  -0.018   0.9853  
## model       -1.385e-02  1.634e-02  1.806e+03  -0.848   0.3966  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##        (Intr) heat++
## heat++ -0.349       
## model   0.113  0.014

Year-by-year effects

## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: diff ~ year + model + (1 | species)
##    Data: df_yrs
## 
## REML criterion at convergence: 7641.5
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -4.9801 -0.5117  0.0348  0.5469  5.5321 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  species  (Intercept) 0.01506  0.1227  
##  Residual             0.48445  0.6960  
## Number of obs: 3591, groups:  species, 20
## 
## Fixed effects:
##               Estimate Std. Error         df t value Pr(>|t|)
## (Intercept) -5.778e+00  8.070e+00  2.357e+03  -0.716    0.474
## year         2.894e-03  3.999e-03  2.355e+03   0.724    0.469
## model       -1.110e-02  1.676e-02  1.302e+03  -0.662    0.508
## 
## Correlation of Fixed Effects:
##       (Intr) year  
## year  -1.000       
## model -0.219  0.220