Decision Trees for the Teaching of Statistical Estimation

Introduction

In many experiments or research, researchers want to know whether the difference in means, proportions, or variances between two populations is significant, or sometimes if the average, proportion, or variance are near to a standard value. They also will be interested in an interval which is expected to contain the value of the difference in means, proportions, and variances of the two populations, or the average, proportion, or variance of a population. In all these cases, researchers have to rely on statistical estimation and choose the most appropriate procedure.

Since many different confidence intervals have been proposed in the literature, the choice can sometimes be difficult and confusing, especially for students. In this chapter we will show how decision trees can help in finding the most appropriate confidence interval.

Types of Estimations

An estimator is a statistic of a sample that is used to estimate a population parameter such as mean, proportion, or...