Sampling: Polls: Experiments

Abstract

In this chapter we look at how statistics impinges on our world: estimating numerical properties of our society and deciding how reliable are those estimates. A sample is the group we examine to study a property (such as family income, height, and so on). The obvious question is: how reliable are our estimates in describing the whole population? We discuss the reliability of methods of sampling and ways of obtaining data, by observation or controlled experiments. Experimental designs are discussed briefly, and in particular Latin squares are studied. The relation of these designs to sudoku puzzles is mentioned.