Transcript Section 6-4
E LEMENTARY S
TATISTICS
Section 6-5 Estimating a Population Proportion
M ARIO F . T RIOLA E IGHTH Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman E DITION 1
Assumptions
1. The sample is a simple random
sample.
2. The conditions for the binomial distribution are satisfied 3. The normal distribution can be used to approximate the distribution of sample are both satisfied.
ˆ
5 and nq ˆ
5
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 2
Notation for Proportions
p =
population proportion
p
ˆ
= x n
sample proportion of
x
successes in a sample of size
n
(pronounced ‘p-hat’) ˆ
=
1 ˆ
p =
sample proportion of
x
failures in a sample size of
n
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 3
Definition
Point Estimate The sample proportion
p
ˆ
is the best point estimate of the population proportion
p
.
(In other books population proportion can be noted as
)
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 4
Margin of Error of the Estimate of p
ME =
z
ˆ ˆ
n
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 5
p
ˆ
Confidence Interval for Population Proportion where
ME =
z
ˆ ˆ
n
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 6
Confidence Interval for
p
ˆ
Population Proportion
E (
p
ˆ
- E,
ˆ
p
+ E)
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 7
Round-Off Rule for Confidence Interval Estimates of p Round the confidence interval limits to three significant digits.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 8
Do people lie about voting?
• In a survey of 1002 people, 701 people said they voted in a recent election. Voting records showed that 61% of eligible voters actually did vote. Using these results, find the following about the people who “said” they voted: • a) Find the point estimate • b) Find the 95% confidence interval estimate of the proportion • c) Are the survey results are consistent with the actual voter turnout of 61%?
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 9
Do people lie about voting?
• In a survey of 1002 people, 701 people said they voted in a recent election. Voting records showed that 61% of eligible voters actually did vote. Using these results, find the following about the people who said they voted: • a) Find the point estimate
n
701 1002 0.6996
0.700
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 10
Do people lie about voting?
• b) Find the 95% confidence interval estimate of the proportion
E
z
/ 2
n
1.96
(.70)(.30) .0284
1002 .671
E p p
.728
E
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 11
Do people lie about voting?
• c) Are the survey results consistent with the actual voter turnout of 61%?
We are 95% confident that the true proportion of the people who said they vote is in the interval 67.1%
Because 61% does not fall
inside
the interval, we can conclude our survey results are
not
consistent with the actual voter turnout.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 12
Determining Sample Size
ME
=
z
n
(solve for
n
by algebra)
n
=
(
z
)
2 ME 2 ˆ
p q
ˆ
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 13
Sample Size for Estimating Proportion p
n
=
(
z
)
2 E 2 ˆ
p q
ˆ When no estimate of p is known:
n
=
(
z
)
2 0.25
E 2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 14
Two formulas for proportion sample size
n
=
(
z
)
2 ME 2
n
=
(
z
)
2
(
0.25
)
ME 2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 15
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? A 1997 study indicates 16.9% of U.S. households used e-mail.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 16
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? A 1997 study indicates 16.9% of U.S. households used e-mail. n = [z
/2 ] 2 ˆ p q E 2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 17
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? A 1997 study indicates 16.9% of U.S. households used e-mail. n = [z
/2 ] 2 ˆ p q E 2 = [1.645] 2 (0.169)(0.831) 0.04
2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 18
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? A 1997 study indicates 16.9% of U.S. households used e-mail. n = [z
/2 ] 2 ˆ p q E 2 = [1.645] 2 (0.169)(0.831) 0.04
= 237.51965
2 = 238 households To be 90% confident that our sample percentage is within four percentage points of the true percentage for all households, we should randomly select and survey 238 households.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 19
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? There is no prior information suggesting a possible value for the sample percentage.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 20
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? There is no prior information suggesting a possible value for the sample percentage.
n = [z
/2 ] 2 (0.25) E 2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 21
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? There is no prior information suggesting a possible value for the sample percentage.
n = [z
/2 ] 2 (0.25) E 2 = (1.645) 2 (0.25) 0.04
2 = 422.81641 = 423 households
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 22
Example: We want to determine, with a margin of error of four percentage points, the current percentage of U.S. households using e-mail. Assuming that we want 90% confidence in our results, how many households must we survey? There is no prior information suggesting a possible value for the sample percentage.
n = [z
/2 ] 2 (0.25) E 2 = (1.645) 2 (0.25) 0.04
2 = 422.81641 = 423 households With no prior information, we need a larger sample to achieve the same results with 90% confidence and an error of no more than 4%.
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 23
Finding the Point Estimate and E from a Confidence Interval Point estimate of ˆ
p
= ˆ
p
: (upper confidence interval limit) + (lower confidence interval limit) 2 Margin of Error: E = (upper confidence interval limit) - (lower confidence interval limit) 2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 24
Finding the Point Estimate and E from a Confidence Interval Given the confidence interval .214< p < .678
Find the point estimate of ˆ
p
: ˆ
p
= .678 + .214
=
.446
2 Margin of Error:
E
= .678 - .214
=
.232
2
Chapter 6. Section 6-5. Triola, Elementary Statistics, Eighth Edition. Copyright 2001. Addison Wesley Longman 25