TABLE OF CONTENTS Asking the right questions is the most fundamental factor that will determine the accuracy of your customer satisfaction survey. More difficult is the second prerequisite of an accurate measure, asking those questions to the right people. This is a matter of accurate sampling, and although the necessity of basing the results on a representative sample of customers is widely acknowledged, the technical aspects of doing so are little understood and often neglected. This chapter will explain the theory and practicalities of sampling in order to ensure that you produce an accurate measure by asking the right people as well as the right questions. In particular we will:
Explain the difference between probability and non-probability samples.
Emphasise the importance of a sample that is unbiased as well as representative.
Demonstrate how to generate an unbiased and representative sample.
Explain how large your sample needs to be.
Section 8 of ISO 9001:2000 covers 'measurement, analysis and improvement'. Although sampling is not specifically covered by the Standard, clause 8.1, the general introduction to the measurement section specifies that measurement, analysis and improvement activities 'shall include determination of applicable methods, including statistical techniques, and the extent of their use'. An accurate measure of customer satisfaction will be generated only if it is based on a statistically robust sample of customers. This chapter will review the sampling methods that should be used in order to achieve that objective.
In principle sampling is simple. Most...
