Grouped data arise in almost all areas of statistical application. Sometimes the group-

ing structure is simple, where each case belongs to single group and there is only one

grouping factor. More complex datasets have a hierarchical or nested structure or

include longitudinal or spatial elements. All such data share the common feature of

correlation of observations within the same group and so analyses that assume inde-

pendence of the observations will be inappropriate. The use of random effects is one

common and convenient way to model such grouping structure.