A confidence statement indicates how precise a statistic is by giving a margin of error and a degree of sureness that the actual parameter is within the range given by the statistic and its margin of error.
The Gallop Poll surveys 1514 registered voters and finds that 53% of the voters in the sample favor the President's proposed health care plan with a margin of error of 3 percentage points with 95% confidence.
Conclusion: We are 95% confident that the actual parameter (or actual percentage of voters who approve of the President's proposed health care plan) lies in the range of 53% +/- 3% (between 50% and 56%).
or
The sampling distribution of the statistic is such that in 95% of all possible samples, the statistic of 53% falls within +/- 3% of the true parameter value.
We can make confidence statements using the Gallop Poll's method by following the table on page 34. This table gives margins of error depending on sample size; the margins of error are for 95% confidence.
Step 1: Find the "population percentage" row of the left column that is closest to the statistic that was calculated.
Step 2: Follow across that row until you get to the column that gives the sample size that is closest to the actual size of the sample.
The resulting number is the margin of error for estimating the actual parameter from your statistic with 95% confidence.
(Scott Street's section only)
Please direct all questions regarding STAT-110 to your instructor or to the director of STAT-110, Dr. Todd Ogden at ogden@stat.sc.edu.
Mail comments regarding this presentation to W. Scott Street, IV at street@stat.sc.edu.
© 1996 by W. Scott Street, IV