# Sociology 2205A/B Chapter 6: Chapter 6: Estimation Procedures

## Document Summary

Estimators can be selected according to two criteria: bias and efficiency. If an estimator is unbiased, it is probably an accurate estimate of the population parameter. In less than 15 of the cases, a sample mean will be more than +/- 3 z"s away from the mean of the sampling distribution by random chance. Interval estimates are much safer than point estimates due to the range of values: the first step is to find the probability of error, called alpha. Sociologists mostly use 0. 5 alpha which states that we have a 95% confidence interval. Over the long run, the researcher is willing to be wrong 5% of the time: second step is to divide the probability of error equally in the upper and lower tails of the distribution. 95% of all possible sample outcomes fall within +/- 1. 96 z-score units of the population value. 6. 4: interval estimation procedures for sample means (large samples)