ABSTRACT
In this section we define three general types of models that are discussed in this book. In the simplest case consider a sample from a normal population with mean J..L and variance o-2 where one is interested in inferences on J..L and o-2 • To generalize, consider sampling from k populations with means and variances f..L; and o-t, respectively. If the if; are equal and the interest is to make statistical inferences about the f..L;, the model is referred to as a fixed effects model (also called a design model or Eisenhart's Model 1). If the J..L; are equal and the interest is in making inferences about the if;, the model is referred to as a random effects model (also called a components-of-variance model, or Eisenhart's Model II). If the interest is in making inferences about the f..L; and/or the o-t when they are different for each population, the model is referred to as a mixed model (or Eisenhart's Model III). We shall discuss each model briefly.
