The Listener Survey Toolkit
Chapter 2
A survey or Arbitron has too few responses to accurately represent the audience

TABLE OF CONTENTS
Overview
Why Conduct a Listener or Member Survey
Research Myths and Misconceptions
Building the Survey
Survey Methodology
Survey Questions
Processing the Data
Data Analysis and Reporting
Applying the Survey Results
Station Service and Support for the Listener Survey Toolkit
Sample Questions
For those not familiar with statistics, it may seem far-fetched to believe that a relatively small sample can adequately represent the opinions of members or listeners. But if you stop to think about it, it really is not that hard to believe. Everyone has seen the nationwide polls that measure public opinion for the entire country. Sample sizes for these polls are generally about 1200 persons. These polls generally are accurate to plus or minus 3- 5%, and accurately measure the entire U.S. population of over two hundred sixty five million!

Remember also that the formulas used to demonstrate the mathematical validity of probability used in survey sampling were calculated years ago by mathematicians and statisticians, and have been proven in practice thousands of times by researchers in many fields. Assuming that the surveys questions are asked correctly and the sample size and composition is appropriate for the objectives of the study, the survey will be valid.

It is often true that persons who seriously doubt the value of properly conducted research have personal reasons or agendas that influence their feelings about research and scientifically ascertaining the behavior and feelings of others.

[ Chapter 2 Table of Contents | Previous Section | Next Section | Toolkit Home ]