1 Introduction

In this chapter we take the univariate model as described in Chapter 5, and apply it to twin data. The main goals of this chapter are i) to enable the readers to apply the models to their own data, and ii) deepen their understanding of both the scope and the limitations of the method. In Section 6.2.1 a model of additive genetic (A), dominance genetic (D), common environment (C), and random environment (E) effects is presented although D and C are confounded when our data have been obtained from pairs of twins reared together. The first example concerns a continuous variable: body mass index (BMI), a widely used measure of obesity, and Section 6.2.2 describes how these data were obtained and summarized. In Section 6.2.3 we fit this model to authentic data, using Mx in a path coefficients approach. Section 6.2.5 illustrates the univariate model fitted with variance components. An alternative treatment which may be skipped without loss of continuity. The results of initial model-fitting to BMI data appear in Section 6.2.6 and two extensions to the model, the use of means (Section 6.2.7) and of unmatched twins (Section 6.2.8), are described before drawing general conclusions about the BMI analyses in Section 6.2.9. In Section 6.3 the basic model is applied to ordinal data. The second example (Section 6.3.1) describes the collection and analysis of major depressive disorder in a sample of adult female twins. This application serves to contrast the data summary and analysis required for an ordinal variable against those appropriate for a continuous variable. In most twin studies there is considerable heterogeneity of age between pairs. As shown in Section 6.4, such heterogeneity can give rise to inflated estimates of the effects of the shared environment. We, therefore, provide a method of incorporating age into the structural equation model to separate its effects from other shared environmental influences.