Introduction

Collecting data on a sample of respondents is a far from trivial task, as the preceding chapters of this book should have demonstrated quite clearly. However, for the data to continue to be useful, two things are necessary: there must be adequate documentation of the survey effort, and the data must be appropriately archived, so that future users can access the data and can understand their meaning. Too often, in the past, survey researchers have ignored these requirements, with the result that further uses of expensive data sets have been precluded. Because of the time required to document and archive, it is often tempting to overlook these final activities and to consider that the survey is done when the data have been collected, processed, edited, cleaned, weighted, expanded, and analysed. In addition, unfortunately, there are many instances when a survey research firm has estimated the costs of the survey sufficiently poorly that money has been lost in completing the data collection itself, and the time required to document and archive the data would add to the financial losses already incurred.

In considering these two activities, it makes sense to consider the documentation of the survey first and then to deal with archiving, especially because a significant amount of the information that should be assembled in the documentation should also provide part of the metadata that should be stored with the archived data. In discussing documentation, this book provides suggestions on what should be included, recognising at the same time that contractual relationships may override or replace these suggested requirements.