I have created a linear mixed effects model testing for shannon diversity index, with flower_quantity, temp, precip, wind and cloud cover as continuous variables, and season as a categorical variable. I have created multiple linear mixed effect models to determine the effects of these variables on shannon diversity value.
summary(lmer_model_precip)
Linear mixed model fit by REML ['lmerMod']
Formula: Species_byTF_summary2$SDI ~ Flower_Quantity + (1 | `24h_precip`)
Data: df2
REML criterion at convergence: -31.2
Scaled residuals:
Min 1Q Median 3Q Max
-0.7667 -0.4954 -0.3735 0.2570 1.8948
Random effects:
Groups Name Variance Std.Dev.
24h_precip (Intercept) 7.647e-05 0.008745
Residual 2.315e-04 0.015215
Number of obs: 8, groups: 24h_precip, 3
Fixed effects:
Estimate Std. Error t value
(Intercept) 0.06001 0.01213 4.945
Flower_Quantity -0.02689 0.02739 -0.982
Correlation of Fixed Effects:
(Intr)
Flowr_Qntty -0.778
anova(lmer_model_precip)
Analysis of Variance Table
npar Sum Sq Mean Sq F value
Flower_Quantity 1 0.00022309 0.00022309 0.9637
How do I interpret the output of these summaries and anovas please?
24h_precipas a random effect. Random effects typically represent groupings used to handle correlations among observations. Precipitation doesn't seem to be such a grouping. Also, it seems that you are trying to do multiple individual models; it's usually best to do a single model incorporating all predictors. If you only have 8 observations total it will be hard to get useful results in any event. Finally, do check that the modeled residuals are OK, as the Shannon index as an outcome can be poorly behaved in that regard. $\endgroup$24h_precipas a (??) covariate/nuisance variable?), they probably won't be able to fit a multivariate model to a data set with only 8 points ... to the OP: can you tell us a little bit more about how you want to take24h_precipinto account? Is it a *nuisance variable*/covariate you want to control for? How many seasons do you have? $\endgroup$