Scribd
Upload
40 views17 pages

Chapter 8

This document discusses statistical inference for population proportions. It introduces key concepts such as population proportion, sample proportion, the sampling distribution of the sample proportion, and the normal approximation. It also covers hypothesis testing and confidence intervals for comparing two population proportions, including setting up test statistics and hypotheses. Examples are provided to illustrate comparing proportions from independent samples or randomized treatment groups.

Uploaded by

Mohammed Saad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
40 views17 pages

Chapter 8

This document discusses statistical inference for population proportions. It introduces key concepts such as population proportion, sample proportion, the sampling distribution of the sample proportion, and the normal approximation. It also covers hypothesis testing and confidence intervals for comparing two population proportions, including setting up test statistics and hypotheses. Examples are provided to illustrate comparing proportions from independent samples or randomized treatment groups.

Uploaded by

Mohammed Saad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

Inference for Proportions

Proportions

Our earlier analysis focused on inference about


population means.

Now we turn our attention to inference about the


proportion of some outcome in a population.

We will consider a single population and then compare


proportions from two populations or treatments.
The Basics

Population Proportion:
# in population with characteristic
# in population
Sample Proportion:
# in sample with characteristic
n
P̂ is a point estimate of P
Sampling Distribution of Sample Proportion

Choose an SRS of size n from a population that contains


proportion p of ‘characteristic’. Let the sample proportion be
defined as the number in the sample with the ‘characteristic’
divided by n.
– As the sample size increases, the sampling distribution of the
sample proportion becomes approximately normal.

– The mean of the sampling distribution is p, the population


proportion.
p (1 − p )
– The standard deviation of the sampling distribution is n
Assumptions for Inference about a Proportion

The data are an SRS from the population of interest.

The population is at least ’10’ times as large as the


sample.

Fact: The normal approximation to the distribution of the


sample proportion is most accurate when p = 0.5
Let’s Try Some Examples
Comparing Two Proportions
Setting 1: Independent Samples from two populations
– proportion of items in population 1 with characteristic
– proportion of items in population 2 with characteristic

Setting 2: Randomly assign subjects to one of two


treatments
– Probability of success with treatment 1
– Probability of success with treatment 2

In both settings, we wish to compare P and P .


1 2
The Data

n items from trt/pop 1 n items from trt/pop 2


1 2

X have the characteristic X have the characteristic


1 2

P
'
= X 1
P =
' X 2
1 2
n 1 n 2
For large n1 and n2

Z=
(Pˆ − Pˆ ) − (P − P )
1 2 1 2

P1 (1 − P1 ) P2 (1 − P2 )
+
n
1 n 2

is approximately standard normal.


(1 − α )100% Confidence Interval for P1 − P2

( ) ˆ
p (1 − ˆ
p ) ˆ
p (1 − ˆ
p )
Pˆ1 − Pˆ2 ± Z (α / 2) 1 1
+ 2 2

n1 n2
Hypothesis Test for P1 − P2

Ho: P1 = P2

Under Ho, P1 = P2 estimate the same quantity.

Pˆp = X +X
1 2
is a pooled estimate.
n +n
1 2
Test Statistic

P −P
' '

Z= 1 2

'
P p (1 − P )'
p
1
 1
1
n + n 
2 
Set Up for Ha
1. Ha: P
1
> P
2
; Reject Ho if Z ≥ Z(α )

2. Ha: P
1
< P
2
; Reject Ho if Z ≤ −Z(α )

3. Ha: P
1
≠ P 2
; Reject Ho if Z ≥ Z(α / 2 )
Or Z ≤ −Z(α / 2 )
How about another example?

You might also like