PADP 7110 Lecture Notes - Lecture 9: Statistic, Statistical Parameter, Sampling Frame

• School: University of Georgia
• Department: Public Administration and Policy
• Course: PADP 7110
• Professor: Abraham

Document Summary

Parameter > a statistical characteristic of a population. Statistic > a statistical characteristic of a sample which is used to estimate a parameter: sampling design. Sampling & nonsampling errors: sources of error. Samples do not provide perfectly accurate population estimates (inaccurate sample statistics) How well a sample represents a population depends on: sampling errors > known errors that occur from the prices of random sampling (due to chance) Degree of error to be expected by virtue of studying a sample in- stead of everyone in a population. Discrepancy between a sample estimate of a population parameter. 3: nonsampling errors > error introduced by flaw in the design or way design is implemented (survey process) Confidence level & interval: level > estimated probability that a population parameter lies within a given con- fidence interval. If 95% confidence interval, ci = sample mean +/- 1. 96(se: central limit theorem. Variability in the sample = sampling error.