Started a mixed effects example for one variable RCBD.

[moorepants]

@msimmond Here is a base for a simpler mixed effects model. I just copied the data from the other RCBD example, added a year column, and adjusted the numbers for the second year with some randomness.

> source('rcbd_one_var_year.R', print.eval = TRUE)
Linear mixed effect model results
===============================================================================
Linear mixed-effects model fit by REML
 Data: my.data 
       AIC      BIC    logLik
  120.7264 130.0519 -53.36321

Random effects:
 Formula: ~1 | Block
        (Intercept)
StdDev:    1.336126

 Formula: ~1 | Year %in% Block
         (Intercept) Residual
StdDev: 3.311861e-05 1.237766

Fixed effects: Yield ~ Treatment 
                      Value Std.Error DF  t-value p-value
(Intercept)        6.027597 0.7986341 21 7.547383  0.0000
TreatmentVariety B 2.651774 0.6188832 21 4.284773  0.0003
TreatmentVariety C 3.680206 0.6188832 21 5.946527  0.0000
TreatmentVariety D 1.585305 0.6188832 21 2.561557  0.0182
 Correlation: 
                   (Intr) TrtmVB TrtmVC
TreatmentVariety B -0.387              
TreatmentVariety C -0.387  0.500       
TreatmentVariety D -0.387  0.500  0.500

Standardized Within-Group Residuals:
       Min         Q1        Med         Q3        Max 
-2.2653194 -0.4627162  0.1031063  0.5976403  1.6475445 

Number of Observations: 32
Number of Groups: 
          Block Year %in% Block 
              4               8 
===============================================================================

[msimmond]

I think block is nested within year. Thus, it should be ~1|Year/Block. I'll look into this some more to be sure.

[moorepants]

I think you might be right.

[msimmond]

This seems to be true based on other examples, so I'll edit it.