CHAPTER 10
PARADATA AS INPUT TO MONITORING REPRESENTATIVENESS AND MEASUREMENT PROFILES: A CASE STUDY OF THE DUTCH LABOUR FORCE SURVEY
10.1 INTRODUCTION
Paradata may provide useful and relevant observations on sampled households and persons in the evaluation, monitoring, and reduction of nonresponse error and response error. In this chapter, we demonstrate how to use paradata for these purposes. We use paradata mostly to distinguish different sources of nonresponse and response error. However, we will also discuss how to include paradata observations on sample units as auxiliary variables in monitoring and design of data collection.
We explore two influential survey errors: response or measurement error and nonresponse error. We use paradata on the number of visits and type of reporting to investigate the impact of design choices on the occurrence and magnitude of the two errors. We also use paradata to distinguish the two main causes of nonresponse: noncontact and refusal. In order to avoid focusing on single survey variables, we define measurement profiles. A measurement profile is a response style that is employed consistently during a significant part of the questionnaire. In this case study, measurement profiles will be identified as differences between registers and surveys; we will provide explicit examples of such profiles. However, it is straightforward to use paradata observations to define ...
Get Improving Surveys with Paradata: Analytic Uses of Process Information now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
