 2012 Royal Statistical Society 0964–1998/13/176001 J. R. Statist. Soc. A (2013) 176, Part 1, pp. 1–3 Editorial: The use of paradata in social survey research During the last 20 years survey data have increasingly been collected through computer-assisted modes. As a result, a new class of data—so called paradata (Couper, 1998)—is now available to survey methodologists. Typical examples are keystroke files, capturing the navigation through the questionnaire, time stamps, reflecting the length of a question–answer sequence, and data from call records or contact attempts. Other examples include interviewer observations about a sampled household or neighbourhood, and recordings of vocal properties of the interviewer and respondent. Although the type of available paradata varies by mode, they all share one feature; they are a by-product of the data collection process capturing information about that process. The potential uses of paradata are wide ranging. An increasing number of survey organiza- tions use paradata to guide intervention decisions during data collection, such as the changes in incentives or interviewers. Likewise, paradata have been used to examine data collection pro- cesses post hoc and to learn about areas of improvement for the next survey. Over recent years, awareness of the potential benefits of paradata has risen among both data producers and data users. This has been reflected in special interest and invited paper sessions at international con- ferences, such as the International Workshop on Household Survey Nonresponse, biannual conferences of the European Survey Research Association, annual conferences of the Ameri- can Association for Public Opinion Research, the Joint Statistical Meetings and sessions of the International Statistical Institute. To summarize recent developments in the field and to discuss gaps in the literature, a one- day international research symposium on ‘Recent advances in the use of paradata in social survey research’ was held at the Royal Statistical Society in December 2010, paired with a short course on ‘Paradata in survey research’. The symposium and the course were attended by an international set of researchers from academia, government and private industry alike, all sharing thoughts and experiences in collecting and using paradata. Results from five of the eight papers in this issue of Statistics in Society were discussed at the symposium. The remain- ing three papers in this issue expand the set of papers to other types of paradata and other uses. The papers that are collected in this issue show new applications of paradata, explore new methods to analyse such data and discuss the data quality and measurement error properties of these new sources of data. As such, the aim of this themed issue is to give a better understanding of the use of paradata to survey methodologists, survey practitioners, survey data providers, data analysts and users of survey data. All papers in this issue were independently reviewed according to the standards applied to all regular issues of the Journal of the Royal Statistical Society. Call record data have been used to analyse survey response long before they were sub- sumed under the label paradata. Olson’s review piece summarizes years of research on the