Summary

Coding of verbal responses of a statistical survey could be defined as assigning numeric codes to statements according to a manual of official classification. Performing this activity manually is costly, time consuming and error prone, so the computer support for this process is increasing. The informative base to be used for this purpose is the fundamental part of any computerised approach, because it must fulfil at least two requirements: make the computer rely on knowledge similar to that inside the human mind and, starting from the content of the official classification manual, process and enrich it with selected descriptions and/or synonyms derived from empirical responses given in previous surveys, so as to make the language closer to the spoken one. Logical steps to be carried out to build informative bases are described, as well as particular aspects to be taken into consideration either when coding is done in a completely automated way or with human support.

To read the entire document, please access the pdf file (link under "Related Documents" on the right-hand-side of this page).

Your feedback is appreciated. Please send your remarks, suggestions for improvement, etc. to memobust@cbs.nl.