Editing by respondents and data suppliers

Orietta Luzi, Natalie Shlomo

Research output: Contribution to journalArticlepeer-review

Abstract

This paper contains an analytic summary of the content of the session Editing by respondents and data suppliers of the UNECE Work Session on Statistical Data Editing (SDE) held at the National Statistical Institute of Spain, Madrid in October 2003. The aim of the session was to collect information about recent trends in data editing. Statistical agencies are changing their traditional view on editing as a post-data collection phase under the control of survey methodologists. Instead, they are moving some of the editing processes to the respondents and data suppliers to prevent and eliminate errors. The session focused on the integration and optimisation of data editing in the collection phase, and the harmonisation of editing processes performed at both the data capturing and data editing stages by balancing expected levels of quality, time and available resources. The session also considered data editing when administrative or external sources of data are used for statistical purposes. A new "mixed" perspective is emerging where editing and imputation is spread over the entire survey process. In particular, the paper concentrates on the following aspects: evaluating data editing effectiveness when editing is spread over different survey phases; managing survey processes when multiple sources of information (statistical or administative) are used; rationalizing survey processes when different modes of data collection are combined; and improving cooperation with data providers.
Original languageEnglish
Pages (from-to)53-65
Number of pages12
JournalStatistical Journal of the United Nations Economic Commission for Europe
Volume21
Issue number1
Publication statusPublished - 2004

Keywords

  • Editing and imputation
  • Electronic data collection
  • External and administrative data
  • Mixed editing strategies

Fingerprint

Dive into the research topics of 'Editing by respondents and data suppliers'. Together they form a unique fingerprint.

Cite this