Scribd
Upload
33 views1 page

B. 2-Propzint - Confidence Interval For The Difference Between 2 Unknown Proportions of Success

This document provides instructions for calculating confidence intervals for proportions from sample survey data. It describes how to calculate a 95% confidence interval for the proportion of people watching the Super Bowl based on a sample of 1,000 people. The confidence interval calculated was between 60.4% and 66.4%. It also describes how to calculate a 90% confidence interval for the difference between two proportions, using an example of the proportion of boys and girls drinking water at lunch. The confidence interval calculated for the difference between these two proportions was between 12.5% and 41%.

Uploaded by

kunichiwa
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
33 views1 page

B. 2-Propzint - Confidence Interval For The Difference Between 2 Unknown Proportions of Success

This document provides instructions for calculating confidence intervals for proportions from sample survey data. It describes how to calculate a 95% confidence interval for the proportion of people watching the Super Bowl based on a sample of 1,000 people. The confidence interval calculated was between 60.4% and 66.4%. It also describes how to calculate a 90% confidence interval for the difference between two proportions, using an example of the proportion of boys and girls drinking water at lunch. The confidence interval calculated for the difference between these two proportions was between 12.5% and 41%.

Uploaded by

kunichiwa
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/ 1

)

)
SRS, the population is at least 10 times the sample size and np and n(1! p ) are ! 10.
)
)
p (1! p )
)
)
You are calculating p z *
where p is the proportion of success in your sample.
n

Example: In a poll of 1000 people taking from an SRS of the country, 634 said they
watched the Super Bowl. Find a 95% confidence interval for the proportion of
people watching the Super Bowl.

Conclusion: if we were to perform this survey repeatedly with sample sizes


of 1000, 95% of them would have between 60.4% and 66.4% watching the Super
Bowl. The margin of error is .664 - .634 = .03 or 3%.
Note that we can perform this calculation because our population is well over
10 times our sample size, and 1000(.634) and 1000(.356) are well over 10.
* B. 2-PropZInt - confidence interval for the difference between 2 unknown proportions of success.
When to use: You have 2 samples having different number of successes. You want to predict
the difference in the proportion of success to an indicated level of confidence. We assume that
)
)
)
)
the data is from 2 SRSs, and n1 p, n1 (1! p1 ) , n2 p, n2 (1! p2 ) are all greater than 5.

) )
You are calculating ( p1 ! p2 ) z *

)
)
p1(1 ! p1 )

)
)
p2 (1! p2 )

Water
18
28

Other
51
25

+
n1
n2
Example: A study is being done on the proportion of students drinking water at
lunch. We want to construct a 90% confidence interval between the proportion
of girls drinking water and boys. The data follows:
Boys
Girls

Conclusion: If we were to perform this study repeatedly, 90% of the time, the difference
between the 2 proportions of water drinkers would be between 12.5% and 41%. The
margin of error is .410 - (.528 - .261) = 14.3%. It is so large because of the obvious
difference in drinking habits between boys and girls.

www.MasterMathMentor.com

- 18 -

Stu Schwartz

You might also like