01:960:285 Lecture Notes - Lecture 18: Confidence Interval, Sample Size Determination

• School: Rutgers University
• Department: Statistics and Biostatistic
• Course: 01:960:285
• Professor: Zhanyun Zhao

Document Summary

Larger is better for getting more reliable data about the population: however, one is always limited by the amount of resources on hand, example: The confidence level is 1: when n = 100, the 95% confidence interval is. The width of the confidence interval is 35. 42. As the sample size n grows increasingly large, the width of the confidence interval grows increasingly small. At what point would the width of the confidence interval be 10: sampling error, estimating the population proportion. Is half the width of the confidence interval. sampling error e, the sample size should be. To estimate with confidence level 1-, population proportion p, and estimate with . Population proportions will more than likely be unknown, but you can. Say that in an initial sample, you sampled 50 people and 10 said their cell phone was defective. Alternatively, if an initial study was not performed, you should be conservative and use 0. 5 as p.