SMABS 2004 Jena University
SMABS 2004 Home Organization About Jena Sponsors Links Imprint / Contact SMABS Home

European Association of Methodology

Department of methodology and evaluation research

Jena University

Contributions: Abstract

Multilevel analysis of genetic data

Gerty Lensvelt-Mulders Cora Maas
Utrecht University
The Netherlands

This paper will describe how multivariate genetic effects can be incorporated in a multilevel analysis (ML); the advantages and disadvantages of this approach will be discussed as well. In behavioral genetic research, Quantitative Genetic Analysis (QGA) is the golden standard statistical technique. An ML approach should have clear straightforward advantages over conventional QGA will it be accepted as a useful replacement.

In this paper we will discuss the advantages and disadvantages of both methods with special aim on the extra information that can be obtained with ML analysis. We will illustrate this by comparing QGA and Multilevel estimates for genetic, unique environmental and common environmental effects. As an example of the possibilities of multivariate modeling with ML we will describe the interaction between the heritability of a trait and the demands of the topic situation, comparing both analysis methods.