MATH& 146

Lesson 22
Section 3.1
Inference for Single Proportions

1
 Trial, Success, and Failure
A single event that leads to an outcome can be
called a trial. If the trial has two possible
outcomes, e.g. heads or tails when flipping a coin,
we typically label one of those outcomes a
success and the other a failure.
The choice of which outcome is labeled a success
and which a failure is arbitrary, and it will not
impact the results of our analyses.

2
 Conditions for Normality
The distribution of p will be considered nearly
normal when the distribution is not too strongly
skewed for its sample size.
The most common guideline for sample size when
working with proportions is to ensure that we
expect to observe a minimum number of
successes and failures, typically at least 10 of
each.

3
 Conditions for Normality
The sampling distribution for p , taken from a
sample of size n from a population with a true
proportion p, is nearly normal when
1) the sample observations are independent
(random sample from less than 10% of the
population), and
2) we expect to see at least 10 successes and 10
failures in our sample, i.e. np 10 and
n(1 p) 10. This is called the success-
failure condition.
4
 Example 1
Which of the following samples would be
considered "large enough" to use a normal
approximation?

a) A sample of size n = 20 and p = 0.5.

b) A sample of size n = 100 and p = 0.05.
c) A sample of size n = 30 and p = 0.7.
d) A sample of size n = 35 and p = 0.4.

5
 Standard Error
If these conditions are met, then the sampling
distribution of p is nearly normal with mean p and
standard error

p 1 p
SE p
n

6
 Example 2
Compute the standard error if
a) p = 0.4 and n = 50
b) p = 0.9 and n = 110

7
 Standard Error
Typically we do not know the true proportion, p, so
we substitute some value to check conditions and
to estimate the standard error.
For confidence intervals, usually p is used to
check the success-failure condition and compute
the standard error.
For hypothesis tests, typically the null value p0 is
used in place of p.

8
 Sample Proportions
Consider the following question: What proportion
of the American public approves of the job the
Supreme Court is doing?
According to a New York Times/CBS News poll in
June 2012, only about 44% of the American public
approved of the job the Supreme Court was doing.
This poll included responses of 976 adults.

9
 Example 3
Estimate the standard error of p = 0.44 and
n = 976.

10
 Example 4
Construct a 95% confidence interval for p, the
proportion of Americans who approve of the job
the Supreme Court is doing.

point estimate zSE

Reminder: p = 0.44 and n = 976.

11
 Hypothesis Testing
To apply the normal distribution framework in the
context of a hypothesis test for a proportion, the
independence and success-failure conditions must
be satisfied.
Independence is checked by verifying random
selection or random assignment.
The success-failure condition is checked using the
null proportion: we verify np0 and n(1 p0) are at
least 10, where p0 is the null value.

12
 Example 5
Deborah Toohey is running for Congress, and her
campaign manager claims she has more than 50%
support from the district's electorate. Ms. Toohey's
opponent claimed that Ms. Toohey has less than
50%.
Set up hypotheses to test whether or not Ms.
Toohey has 50% of the vote.

13
 Example 6
A newspaper collects a simple random sample of
500 likely voters in the district and estimates
Toohey's support to be 52%.
Does this provide convincing evidence for the
claim of Toohey's manager at the 5% significance
level?

14
 Sample Size
Sample size is an important part of any study. Large
samples done right may be useful, but might be too
expensive, while small samples may be cheap, but
might not have enough power to answer your research
question. Just like Goldilocks, we want a sample size
that is just right.
We can perform sample size calculations that are
helpful in planning a study. Our task will be to identify
an appropriate sample size that ensures the margin of
error ME = z*SE will be no larger than some value m.

15
 Sample Size
To find the minimum sample size needed to
guarantee results are within a margin of error at a
given level of confidence is
2
z
n
2m

z* is the confidence level (1.64 for 90%, 1.96 for

95%, 2.58 for 99%)
m is the maximum desired margin of error.

16
 Sample Size
Generally, a margin of error (ME) of 5% (m = 0.05)
or less is acceptable, but different circumstances
call for different standards.
For a pilot study, a margin of error of 10% (m =
0.10) may be fine, so a sample of 100 will do quite
well.
In a close election, a polling organization might
want to get the margin of error down to 2% (m =
0.02), or about 2500 people.

17
 Example 7
Find the minimum sample size needed for a
survey of a large population if the margin of error
must be within
a) 3 percentage points for a 95% confidence
interval.
b) 10 percentage points for a 90% confidence
interval.
c) 1 percentage point for a 99% confidence
interval.
18
 Example 8
Suppose we are conducting a university survey to
determine whether students support a $200 per
year increase in fees to pay for a new football
stadium. How big of a sample will be needed to be
sure the margin of error is less than 0.04 using a
95% confidence level?

19