chapter  9
21 Pages

Studying Linear Relationships

ByRoger Mead, Robert N. Curnow, Anne M. Hasted

The simplest form of relationship between two variables is a straight line. This chapter considers in some detail methods of estimating a linear relationship and predicting from the estimated line. It discusses the use of these methods in the analysis of experimental data. The chapter aims to investigate and test hypothetical mathematical models for biological systems, and to predict the values of one variable from another. The chapter focuses on the relationship between two or more variables. Correlation techniques are useful in the preliminary examination of a large number of variables to see which variables are associated. The chapter also discusses the methods for testing the linearity and for fitting more complex relationships. All general statistical packages have routines for fitting regression lines using least squares. They allow the data to be plotted easily to investigate the suitability of the proposed regression model and also have facilities to store and investigate the residuals from the model fitting.