R language Access Menu

Title Text Both  

Repeated measures ANOVA

In some experiments the measurements are made a number of times on same subjects. Different subjects are likely to have similar measurements on repeat testing. Hence, subject factor has to be taken into account. This is done by mixed effect analysis of variance. 

If in an experiment, subjects are fed with 3 different diets and blood levels of a chemical are checked at 3 times after each diet, the data can be tabulated as follows: 

Code:

> mydata
   animal diet     time value
       1    1       t1     5
       2    1       t1     6
       3    1       t1     5
       4    2       t1     6
       5    2       t1     5
       6    2       t1     4
       7    3       t1     5
       8    3       t1     5
       9    3       t1     4
       1    1       t2     6
       2    1       t2     8
       3    1       t2     7
       4    2       t2     9
       5    2       t2     6
       6    2       t2     4
       7    3       t2     8
       8    3       t2     5
       9    3       t2     6
       1    1       t3     9
       2    1       t3     6
       3    1       t3     8
       4    2       t3     7
       5    2       t3     8
       6    2       t3     9
       7    3       t3     6
       8    3       t3     5
       9    3       t3     8

The analysis can be done using packages lme4 and lmerTest : 

Code:

> library(lme4)
> library(lmerTest)
> summary(lmer(value~diet*time+(1|animal), data=mydata))
Linear mixed model fit by REML 
t-tests use  Satterthwaite approximations to degrees of freedom ['merModLmerTest']
Formula: value ~ diet * time + (1 | animal)
   Data: mm

REML criterion at convergence: 83

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-1.9458 -0.5076  0.0000  0.6345  1.8612 

Random effects:
 Groups   Name        Variance Std.Dev.
 animal   (Intercept) 0.000    0.000   
 Residual             1.725    1.313   
Number of obs: 27, groups:  animal, 9

Fixed effects:
     Estimate  Std. Error   df  t value   Pr(>|t|)    
(Intercept)  5.667  1.158 21.000   4.892    0.000 ***
diet        -0.333  0.536 21.000  -0.622    0.5408    
timet2       1.555  1.638 21.000   0.950    0.3531    
timet3       2.999  1.638 21.000   1.831    0.0813 .  
diet:timet2  0.000  0.758 21.000   0.000    1.0000    
diet:timet3 -0.333  0.758 21.000  -0.440    0.6647    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Correlation of Fixed Effects:
            (Intr) diet   timet2 timet3 dt:tm2
diet        -0.926                            
timet2      -0.707  0.655                     
timet3      -0.707  0.655  0.500              
diet:timet2  0.655 -0.707 -0.926 -0.463       
diet:timet3  0.655 -0.707 -0.463 -0.926  0.500

 


    Comments & Feedback