Quality of demographic data in GGS Wave 1

Reliability of retrospective event histories within the German Generations and Gender Survey

“…Sauer, Ruckdeschel, and Naderi (2012: 18) show that the parity distribution of mothers is in fact satisfactory, i.e., the proportion of mothers with one, two, three, and four children does not show significant peculiarities in comparison to the Microcensus. Nevertheless, the GGS appears to provide a misleading picture of the overall trend of fertility in Germany (Kreyenfeld, Hornung, and Kubisch 2013;Vergauwen et al 2015). When looking at the second set of retrospective questions in the GGS, the partnership history, we find a similar picture.…”

“…There are various types of distortion: compared to German vital statistics the total number of children in the GGS is too low for birth cohorts and too high for the cohorts born thereafter. When looking at partnership history we have too many women who were never married in our data in the older cohorts and too many married ones in the younger cohorts (Kreyenfeld et al 2010;2011;Kreyenfeld, Hornung, and Kubisch 2013;Sauer, Ruckdeschel, and Naderi 2012;Vergauwen et al 2015). So far, approaches explaining these distortions have not provided satisfactory results.…”

“…This is constructed by using the information on previous partnerships and was implemented without control questions in all the participating countries. In fact, it is underrepresented in the older cohorts in nearly all countries (Vergauwen et al 2015). Generating demographic fertility indicators from retrospective event history data might be a specific German GGS problem, but generating demographic partnership indicators from retrospective event history data is a problem in nearly all GGS-participant countries.…”

Generations and Gender Programme Wave 1 data collection

“…By comparison, these levels were relatively low in the first wave of the GGS, making it hard to judge whether country differences in observed variable scores reflect true differences in these scores across countries rather than differences in survey implementation. Post-data collection assessments that compare GGS data to population information on demographic outcomes (see Vergauwen et al (2015) in this Special Collection) and examine the cross-national equivalence of multiple-item data (see Hox, De Leeuw, and Boevé (in preparation) in this Special Collection) suggest that the GGS does fairly well in these respects. Still, a great deal could be gained by a much more upfront approach to harmonisation, e.g., by centralising methods of questionnaire translation and design and by rigorously applying the same survey modes across countries.…”

Life-table representations of family dynamics in the 21st century

“…For coresidential unions we require the start and end month (or month of interview if union has not ended), as well as the marriage month if the couple married at the start or after a period of nonmarital cohabitation. Such histories are the foundation of the Generations and Gender Programme, within which the first survey waves were conducted between 2003 and 2013 in 18 countries (Vikat et al 2008;Vergauwen et al 2015;Fokkema et al 2016;www.ggp-i.org). This data has been combined with data from other surveys in the Harmonized Histories (Perelli-Harris, Kreyenfeld, and Kubisch 2010;www.nonmarital.org) and the Changing Life-Course Regimes (CLiCR) database (Duntava and Billingsley 2013), with some eastern and central European countries included in both.…”