R help - Aug 2017 - ANOVA test to decide whether to use multiple linear regression or linear mixed effects model

R-help:


I am trying to decide between using a multiple linear regression or a linear
mixed effects model for my data:


model1 <- lm (responsevariable ~ predictor1 + predictor2 + predictor3 +
predictor4, data= data)

model2 <- lme (responsevariable ~ predictor1 + predictor2 + predictor3 +
predictor4, random = ~1 | site, data= data)


anova (model1, model2)


but I keep getting the following error code:

Error in anova.lmlist (object, ...):

    models were not all fitted to the same size of dataset


Any advice would be appreciated, thanks.



	[[alternative HTML version deleted]]

> I am trying to decide between using a multiple linear regression or a
linear
> mixed effects model for my data:
> ... 
> but I keep getting the following error code:
> 
> Error in anova.lmlist (object, ...):
> 
>     models were not all fitted to the same size of dataset
anova is defaulting to anova.lm, and that doesn't expect a mixed effects
model.

Switch them round to put model2 first:

anova (model2, model1)


S Ellison



*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}