The power of the classical twin study

Abstract

A method based on the non-central chi-square distribution is developed for the calculation of sample sizes required to reject, with given probability, models of variation when they are "wrong ". The method is illustrated with reference to simple alternative models of variation in MZ and DZ twins reared together. Simulation of twin experiments finds the empirical power in good agreement with that predicted by the method. Tables are produced showing the sample sizes required for 95 per cent rejection at the 5 per cent level of inappropriate models of variation. For equivalent cases it is always found easier to reject an inappropriate simple genetical model of variation than an inappropriate simple environmental model. For several frequently encountered cases, more than 600 pairs of twins would be required to reject inappropriate alternative models. The optimum proportion of MZ and DZ twins in a sample will vary with the "true" model of variation but is most likely to be between two-thirds and one-half of DZ twin pairs.The possibility of detecting genetical non-additivity with the classical twin study is investigated by theoretical power calculations and simulation. In the absence of genotype-environment interactions, distributional skewness and mean-variance regression in DZ twins are found to be more powerful tests of directional dominance (or unequal gene frequencies) than the standard model fitting procedure and these tests may be worthwhile in future studies.

Model-fitting approaches to the analysis of human behaviour

“…At the level of twin research of individual variables, model-fitting has clarified many important anomalies in twin data and provided a theoretical basis for their interpretation, as in the case of sibling effects. In addition, simulation and data analysis have charted more clearly the areas in which errors of inference are most likely (Eaves, 1972;Eaves et al, 1977;Martin et al, 1978), and provided a quantitative basis for experimental design. Of all factors, a basic understanding of the mechanism of gene action is elusive since many behavioural traits are not so overwhelmingly heritable as to permit the reliable determination of gene action by biometrical genetical methods based on second-degree statistics, still less by analysis in terms of one or two loci.…”

“…Certain combinations of these effects may result in the acceptance of a model which is too simple (see Martin et al, 1978) or produce biased parameter estimates if the power of the test of the model is too low to detect their presence. In many instances, however, we can say whether the assumptions of a simple model are violated using twin data, although it may be more difficult to determine the precise cause of the failure.…”

“…Martin et al (1978) show how to quantify the primary areas of weakness in the twin study. Their approach recognised two main justifications for the criticism of particular designs.…”

“…In the case of the detection of non-additive factors, or in the comparison of statistics from different populations (Eaves and Jinks, 1972), the power of the twin study is low. In other cases, for example in the rejection of the DR, E1 model (Martin et al, 1978), the findings lead to cautious optimism.…”

Genetic influences on dietary variety - Results from a twin study

“…Specifically, by comparing the phenotypic covariance of these measures in MZ and DZ twin pairs, we estimated the percentage of the phenotypic variance explained by latent genetic versus environmental factors, separately for male and female participants (Martin, Eaves, Kearsey, & Davies, 1978). In the classical twin design, the total variance in the phenotype (Vt) is decomposed into three parts: additive genetic variance (A) that captures the combined genetic effects of all genes, common environment variance (C) that captures the environmental influences that are shared between twins such as the effects of other family members or familial socialization, and unique environmental variance (E) that captures all environmental influences that are not shared between twins, including measurement error (Neale & Cardon, 1992).…”