Conferences

Survey data is increasingly used to understand how individuals think and act, and governments, industry and academia are becoming more and more dependent on this kind of data. Much of the reason for this is that survey data are manageable, relatively easy to collect and can sustain complex analyses. The European Social Survey (ESS), the International Social Survey Program (ISSP), the General Social Survey (GSS), the World- and European Value Survey (WVS and EVS), and the Eurobarometer are all examples of widely known and used surveys.

However, when analysing survey data, it has become more and more common to focus on placing the individuals concerned in a society with history, geography, social relationships or social and institutional structures. By including contextual data in the analysis scientists are able to get a more detailed picture, and a better understanding of individual behaviour and attitudes.

Especially when it comes to cross national surveys this element is important. Not only on a national, but also on regional level, individuals are influenced by the society in which they live. But even though contextual data are widely available both online and through other channels, they are not always easily accessible, and getting access and preparing the data for analysis can be time consuming and thus a hindrance for many potential users. This is especially the case for data at sub-national (regional) levels.

As a responds to this, the ESS is developing a contextual data repository where contextual data are directly linked to the ESS survey data. The goal is to ease the process of finding and linking contextual data to the ESS survey data. Key elements in the development of the ESS contextual data repository have been the harmonisation of data and metadata, functional browsing and search, development of functional basic analytical tools adapted to data on multiple levels, the possibilities for cartographic presentation and easy download of data/variables and metadata.

This presentation will describe some of the challenges in the ESS contextual data repository project. This includes both the functionality accommodating the repository editor, and the needs of the end user. On the one hand linking contextual data to the survey data has not been straightforward. For example, the regional levels in a survey organisations sampling plan are not always the same as the statistical units. On the other hand, it is vital that data are presented in a way that will suit the end user. The wants and needs of the research community are highly heterogeneous, both regarding functionality and data.