• home
• organisation
• staff 
• students
• projects
• courses
• conferences
• jobs  
• agenda
• contact
• links 

Giesen, Deirdre

Response burden and measurement error in business survey data collection

Deirdre Giesen (external PhD student)
Methodology Section Heerlen, Room R414
Statistics Netherlands (CBS)
P.O.Box 4481
6401 CZ Heerlen
Phone: +31 45 570 7388
E-mail: Deirdre Giesen

Supervisors: Prof. dr G.J.M. Snijkers, prof. dr J.J. Hox, dr F. van de Pol
Project running from: 1 January 2007 – 1 January 2010

Summary

Traditionally, data collection methodology for household surveys has received more attention than business surveys. Dillmann (2000) strikingly describes the implicit model for conducting government business surveys as a “Cost-Compensation Model”. This implicit model is guided by the overall aim to reduce costs of data collection and the reliance on government authority in general and the mandatory nature of many surveys in particular to compensate for the resulting lack of respondent friendliness.

However, times are changing and there is an increasing interest in data collection methodology for business statistics (e.g. Sudman, Willimack, Nichols & Mesenbourg, 2000; Jones, 2003; Morrison, Stettler and Anderson, 2004; Bavdaz, 2006). Important factors at Statistics Netherlands in this respect are the overall goal to measure and where necessary improve the quality of the statistical process and the aim to reduce both actual and perceived response burden. In 2003,  the Data Collection Expertise Program was established to professionalize and coordinate the data collection for establishments at Statistics Netherlands (Snijkers, Göttgens & Luppes, 2003).

The proposed research will be carried out by Deirdre Giesen, methodologist at the Division of Methodology and Quality of Statistics Netherlands, as part of the research program of the Department of Methodology. Goal of this research is to increase the quality of data collection for business surveys by reducing response burden and by detecting, preventing and correcting non-sampling measurement error. An important result of the research will be a proposal for a monitoring system for the data collection for business surveys at Statistics Netherlands.

Date of defence:
Title of thesis: