Quality Quiz from Professor Cleary
Quinn Quip, our hero of sample size, decided that all that thinking last month merited a short vacation, where he continued to ponder the implications of sample sizes that produce sample proportions that fall within 2 percent of the proportional mean. Reviewing formulas, textbooks, and video training, he remembered that the sample size of 2400 would mean that he could be 95 percent confident that the sample proportion would be within 2 percent of the true proportion. (Remember that = 5 was used as a conservative estimate for sample size.)
The formula for calculating sample size was not exactly simple, but after writing it on his arm so he could remember it, Quinn had it nearly memorized:
Last month’s quiz asked whether one could find the proper sample size for interval data in the same way. Here’s our chance, as Quinn faces yet another challenge to his comprehension of sampling.
A supplier mentioned to Quinn’s boss, Russ T. Buckett, that it would be useful to understand the average age of those who live in Quincy, so that Quincy Quality Quiche could market to the right demographic groups. Quinn—newly sophisticated in his understanding of statistical methods—knows that he must address two questions in order to provide an answer to his boss:
- How much error is the boss willing to tolerate?
- What level of confidence will he insist upon?
Russ T. Buckett says that if the sample estimate is within 2 years of the true average, that would be good enough to serve his purposes, but he wants to have 99 percent confidence in the sample results.
Quinn realized that these questions do not apply for attribute data, and the old sample size formula will not work. He believes that we will need to apply the following formula to respond:
Is this assumption correct?
Click here for a more complete video explanation
2010 PQ Systems.
Please direct questions or problems regarding this web site to the Webmaster.