Advertisement

Elementary Statistical Principles

  • Andrea S. Foulkes
Chapter
Part of the Use R book series (USE R)

Abstract

This chapter includes coverage of several statistical and epidemiological concepts that are broadly relevant to the study of association among multiple variables and specifically important in the analysis of genetic association in population-based investigations. The chapter is divided into three sections. Section 2.1 offers a general background, including the notation used throughout this text and some elementary probability concepts. This section also provides a basic overview of several fundamental epidemiological concepts relevant to any population-based investigation, including confounding, effect mediation, effect modification and conditional association. The reader is referred to Rothman and Greenland (1998) for a comprehensive overview of these and other epidemiological principles. Some of these concepts are very similar to the genetic data concepts discussed in Chapter 3, though they tend to have different nomenclature. All of these elements are important to the discovery and characterization of genotype{trait associations. Section 2.2 describes several simple measures and tests of statistical association, including correlation analysis, contingency table analysis and simple linear and logistic regression. Also included in this section is an introduction to methods for multivariable analysis. Finally, Section 2.3 offers an overview of the analytic challenges inherent in population-based genetic investigations. The testing procedures described throughout this chapter can be applied to each of a set of genotype variables though they generally require an adjustment for multiple comparisons, as described in Chapter 4. Further extensions that allow simultaneous assessment of associations for a group of SNPs or genes are presented in Chapters 5–7.

Keywords

Quantitative Trait Variant Allele Homologous Chromosome Analytic Challenge Contingency Table Analysis 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag New York 2009

Authors and Affiliations

  1. 1.University of MassachusettsSchool of Public Health & Health SciencesAmherstUSA

Personalised recommendations