This chapter is about the type of surveys used by firms who want to know the satisfaction level of their customers. Throughout the chapter we denote these surveys with the acronym CSSs (Customer Satisfaction Surveys). We begin with a brief introduction about the importance of CSSs for a firm (Section 3.1), and proceed to present the two main types of surveys, census and sample. The choice between the two depends on costs, timeliness, goals of the survey, and mainly on the type of firm, its dimension and its customers (Section 3.2). For each survey, the quality of the collected data depends on different types of errors; and in Section 3.3 non sampling errors are listed. We focus our attention on such errors in CSSs and discuss their effect on statistical estimates. In Section 3.4, we show various data collection methods applied in CSSs, linking every method to one or more non sampling errors. Finally, in Section 3.5, some methods to correct these errors are proposed.
Census and Sample Surveys / G. Nicolini, L. Dalla Valle - In: Modern Analysis of Customer Surveys : with Applications using R (Statistics in Practice) / [a cura di] R. Kenett, S. Salini. - [s.l] : Wiley, 2012. - ISBN 978-0-470-97128-4. - pp. 37-53 [10.1002/9781119961154.ch3]
Census and Sample Surveys
G. NicoliniPrimo
;
2012
Abstract
This chapter is about the type of surveys used by firms who want to know the satisfaction level of their customers. Throughout the chapter we denote these surveys with the acronym CSSs (Customer Satisfaction Surveys). We begin with a brief introduction about the importance of CSSs for a firm (Section 3.1), and proceed to present the two main types of surveys, census and sample. The choice between the two depends on costs, timeliness, goals of the survey, and mainly on the type of firm, its dimension and its customers (Section 3.2). For each survey, the quality of the collected data depends on different types of errors; and in Section 3.3 non sampling errors are listed. We focus our attention on such errors in CSSs and discuss their effect on statistical estimates. In Section 3.4, we show various data collection methods applied in CSSs, linking every method to one or more non sampling errors. Finally, in Section 3.5, some methods to correct these errors are proposed.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.