Unformatted text preview: r called a confidence interval is attached to the estimate. • Only the size of the sample, not the size of the population, determines the accuracy of the estimate. •
•
In general, the interval (sample percent ± z SE%) is a ___% confidence interval for the
population percentage, where the blank is the area between z and –z on the Normal
Curve.
68% Confidence Interval = Sample % +/
1 SE% 95% Confidence Interval = Sample % +/
2 SE% 99% Confidence Interval = Sample % +/
_2.6__ SE% 80% Confidence Interval = Sample % +/
_1.3__ SE% Confidence Intervals can only be applied to the population that the sample was drawn from and NOT to a wider population or to sub
groups within the population. Page 10 Study Guide KEY for Exam 3 • Quick formula for choosing how many people to poll for margins of errors for 95% Confidence Intervals: N=(100/MofE)2 For a 10% margin of error you need (100/10)2 = 100 people. For a 5% margin of error you need (100/5)2 = 202= 400 people. For a 4% margin of error you need (100/4)2 = 252= 625 people. For a 2% margin of error you need (100/2)2 = 502= 2500 people. Practice Problems 1. A Harris Poll asked a random sample of 1,015 adults nationwide the following question:
“Are you very afraid of snakes?” 36% of the people in the sample answered “YES”.
b) The SE of the sample % is 1.5%. An approximate 68% confidence interval for the percentage of all American adults who
are afraid of snakes is closest to:
a) (34.5%37.5%)
c) b) (30%36%) c) (32%40%) d) (35.985%36.015%) If the researcher increased the sample size to 4, 060 the length of the confidence interval would a) be multiplied by 4 b) double c) be divided by 4 d) be divided by 2 e) not possible to calculate 2. A CBS News Poll conducted October 1914, 2011 asked 1,650 randomly selected US adults nationwide whether
or not they thought money and wealth in this country should be distributed more evenly or whether it is fair the way it is now.
66% of the 1,650 surveyed answered, that it should be distributed more evenly.
a) What type of sample is this? i) Probability Sample ii) Quota Sample iii) Sample of Convenience iv) Self
selected sample b) What most closely resembles the relevant box model? Circle one. i)
It has 1,650 tickets, 66% marked "1"and 34% marked"0". ii)
It has millions of tickets, with an average of 0.66, but the SD is unknown. iii)
It has millions of tickets marked with "0"s and "1"s. The exact percentages are unknown but...
View
Full Document
 Spring '08
 Kim
 Statistics, the00

Click to edit the document details