Conferences

Data from a particular survey can be enriched by data from other sources. Usually, the other sources are regional administrative data that are used as contextual variables in the individual level survey analyses. Another possibility available in some countries is to enrich a survey by coupling it to individual records coming from register data from government or statistical offices. However, now that so many high quality surveys are being conducted, it is time to start thinking about systematically coupling the data from several different, independent surveys. In this way, for example, election studies might profit from general values surveys (and the other way around). Because the surveys are independent, coupling cannot be done at the individual level, but as all surveys contain information on
birth year and regio, enriching is possible at the regional or, perhaps even more interesting, at the cohort level. Three kinds of analyses can be done with the coupled data:

1. the data from the other surveys can be used as contextual variables, with regio or cohort as context, for an analysis at the individual level

2. the aggregated data can be used in combination with the original individual level data to get an estimate of the individual level correlations (e.g., an estimated individual level table voting*religiosity, based on individual level party preferences and aggregated cohort data on religiosity).

3. analyses at the aggregate level, e.g., regional analyses or cohort analyses for studying social change

Papers are invited and will be presented that

- illustrate the above analyses modes

- deal with practical methodological problems such as missing data or different response categories

- discuss statistical problems, regarding estimation procedures for the combined data

- developping software to make coupling of surveys user friendly.