Table of Contents Table of Contents
Previous Page  441 / 762 Next Page
Basic version Information
Show Menu
Previous Page 441 / 762 Next Page
Page Background

Chapter 10: Sampling Distributions and the Central Limit Theorem

441

In this illustration, we will be dealing with the proportions from the two

populations consisting of males and females.

A general sampling design for such situations in the illustration is shown in

Figure 10-21

.

Figure 10-21:

Sampling Representation to Investigate the

Sampling Distribution for Two Sample Proportions

To be specific, we will let the subscript 1 be associated with population 1,

and the subscript 2 be associated with population 2. In

Figure 10-21

, the

two populations are listed as Population 1 and Population 2, from which

respective samples of sizes

and

are obtained. The numbers of

successes fromeach sample are denoted by

and

. Each population has

its respective proportion denoted by

and

. Also we will let

̂

and

̂

represent the sample proportions from population 1 and population 2,

respectively. These sample proportions are obtained by computing

̂

,

̂

.

Let us consider the difference of the two population proportions

.

The point estimate for

will be represented by

̂

̂

. We can

investigate, through repeated sampling, the

sampling distribution

of

̂

̂

We will take repeated samples from the two populations and

compute the differences for the sample proportions for these repeated

samples. Through simulations and theory, we can state some properties of

the sampling distribution

̂

̂