Sunteți pe pagina 1din 6

AP Statistics

Page 1 of 6
Practice: Creating z-Intervals for a Single Population Proportion
Introduction
When you see a plus or minus percentage following a poll result, this is simply a
confidence interval for a single population proportion. For example, a study may report
that 48% of Americans know how to construct a paper airplane, plus or minus 4%.
Calculating confidence intervals for single proportions is similar to constructing
confidence intervals for means.
Confidence Intervals for a Single Proportion
A confidence interval for a population proportion follows the form
(estimate) (margin of error), where the margin of error is a critical z-value multiplied
by the standard error.

x
, where
n
x is the total number of successes and n is the total number of outcomes (your sample
size). Remember that success is an arbitrary definitionits meaning will depend on the
design of your study.

and is calculated as p
=
The point estimate of the population proportion is p

(1 p
)
p
. Following the
n
model (estimate) (margin of error), the formula for a confidence interval for a

The standard error of the point estimate is calculated as SE p =

z*
population proportion is p

(1 p
)
p
.
n

is binomial and can approximate a


We can use a z-distribution if the distribution of p
normal distribution.

is a ratio of successes to sample size, the distribution of p


is binomial
Because p
whenever N > 10n. Recall that when sampling, the odds of getting a success may
change each time you draw a new member into the sample. But you can consider the
situation to be binomial as long as the population is 10 times larger than the sample
(which is almost always the case).
will approximate the normal as long the following two
The binomial distribution of p
criteria are met:
10 (The number of successes is at least 10.)
= np
=

) 10 (The number of failures is at least 10.)


n(1 p

(Note: Though these are the criteria you should use for this activity, some textbooks
will approximate the normal if np
5 and
say the binomial distribution of p
) 5, or np
> 5 and n(1 p
) > 5.)
n(1 p

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

AP Statistics
Page 2 of 6
Practice: Creating z-Intervals for a Single Population Proportion
You can find your critical z-value with a normal probability table or a calculator, but it's
easiest in the long run if you just memorize the critical z-values for 90%, 95%, and 99%
confidence intervals.
Confidence Level
90% 1.645
95% 1.960
99% 2.576

z*

Sample Size for a Given Margin of Error

If you need to calculate a specific margin of error m, compute the necessary sample size
2

z *
with the formula n =
p*(1 p*), where z* is your critical z-value, m is your
m
margin of error, and p* is your estimate of the population proportion.
If you don't have an estimate of the population proportion, use p* = .5. This will give
you the most conservative estimate for sample size. When you use p* = .5, the formula
2

z *
simplifies to n =
, though you may prefer to just plug p* = .5 into the formula
2m
you already know rather than memorize a whole new one.
Remember that if you don't get a whole number for n, round up to the nearest whole
number. So if you get n = 641.2, you'd round up to 642.
A note on margins of error: The margin of error only accounts for possible random
sampling errors. It doesn't take into account possible biases in the sample, such as
insincere responses and non-responses (such as you'd get from people who are
contacted but won't answer questions). Another potential source of bias is the possibility
that the sample isn't truly random.
Using a TI-83/TI-84 to Calculate a Confidence Interval for a Single Proportion

To calculate a confidence interval for a population proportion using your TI-83/TI-84, press
STAT, arrow to TESTS, highlight A: 1PropZInt and press ENTER.
Enter x (the number of successes), n (the sample size), and C-Level (the level of
confidence you want). Highlight Calculate and press ENTER.
Example
A recent poll of 1,031 frequent fliers (people who travel by air at least five times in one
year) showed that 464 were satisfied with the width of seats on airplanes. Use the 1
PropZInt feature of your calculator to find the 95% confidence interval.
Answer
Press STAT, arrow to TESTS, highlight A: 1PropZInt and press ENTER. Enter x (the
number of successes), n (the sample size), and C-Level (the level of confidence you
want). Your screen should look like this:

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

AP Statistics
Page 3 of 6
Practice: Creating z-Intervals for a Single Population Proportion

Highlight Calculate and press ENTER. Your screen should look like this:

The confidence interval is (.41968, .48042).


Calculating a Confidence Interval for a Single Proportion by Hand
Example
In May, 1998, a national poll of 1,007 adults found that 443 planned to watch the final
episode of Seinfeld.
A. Calculate the point estimate of the population proportion.
B. Can you use a z-test to construct a confidence interval for this proportion? Explain.
C. Construct a 90% confidence interval for the proportion of adults who plan to watch
the final episode of Seinfeld.
D. Researchers want an estimate with a 95% confidence interval and a margin of 3%
for the proportion of adults who plan to watch the final episode of Seinfeld. What
sample size do they need to achieve this estimate?
Answers
x
443
=
=
= .44
A. p
n
1,007
B. Yes, the sample approximates a normal distribution because there are at least 10
= 1,007(.44) = 443) and at least 10 failures (n(1 p
) =
successes ( np

1,007(.56) = 563.92) in the sample.

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

AP Statistics
Page 4 of 6
Practice: Creating z-Intervals for a Single Population Proportion
C. The 90% confidence interval is (.414, .4657) and is calculated as follows:

z*
p

(1 p
)
p
.44(.56)
.2464
= .44 1.645
= .44 1.645
1,007
1,007
n

= .44 1.645 .000244687 = 44 1.645(.01564) = .44 .0257278


= (.414, .4657)
D. The necessary sample size is n = 1,052. To calculate:
2

z *
1.96
(1 p
) =
n=
p
(.44)(.56) = (65.33)2(.2464)
.
03
m

= (4,268.44)(.2464) = 1,051.745
1,052

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

AP Statistics
Page 5 of 6
Practice: Creating z-Intervals for a Single Population Proportion
1. For each of the following situations, state whether the sample meets the criteria
necessary to use a z confidence interval to estimate the population proportion. Use
10 and n(1 p
) 10.
np
A. N 100,000
n = 50
= .05
p
B. N 100,000
n = 50
= .38
p
C. N 100,000
n = 60
x = 27
D. N 100,000
n = 10
x=2
2. The manufacturers of Vicious Victuals dog food want to know the proportion of dogs
that prefer their product to other brands. They randomly select a sample of 589
dogs, and present each dog with a bowl of Vicious Victuals and bowl of another
brand of dog food. Of this sample, 374 dogs chose Vicious Victuals.
A. Can we estimate the population proportion using the normal approximation
to the binomial in this scenario?
B. What's the point estimate of p, the population proportion of dogs that prefer
Vicious Victuals?
C. W hat critical z-value would you use to conduct a 95% confidence interval of
the population proportion?
D. Calculate the standard error for the population proportion ( SE p ).
E. Calculate the margin of error (m) for the 95% confidence interval for the
population proportion.
F. Construct a 95% confidence interval for the population proportion.
G. What's the critical z-value (z*) you'd use to construct a 99% confidence
interval?
H. Calculate the margin of error for the 99% confidence interval for the
population proportion.
I. Construct a 99% confidence interval for the proportion of dogs that prefer
Vicious Victuals.
J. Use your graphing calculator to construct a 90% confidence interval for the
proportion of dogs that prefer Vicious Victuals.

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

AP Statistics
Page 6 of 6
Practice: Creating z-Intervals for a Single Population Proportion
3. Suppose you want to estimate the proportion of voters who'll vote for you in the
upcoming mayoral election. In the previous election, 62% of the voters voted for
you, and you suspect that the same proportion will vote for you this election.
A. What's the minimum sample size you'd need if you wanted to construct a
95% confidence level with a margin of error no greater than 3%?
B. You want to know whether the previous election's results are still valid as
an estimate of the population proportion. What's the safest estimate for the
minimum sample size?

______________________________
Copyright 2011 Apex Learning Inc. (See Terms of Use at www.apexvs.com/TermsOfUse)
TI-83 screens are used with the permission of the publisher. Copyright 1996, Texas Instruments, Incorporated.

S-ar putea să vă placă și