This paper gives an overview of fertility data for Germany, Austria and Switzerland. Particular attention is given to the availability of order-specific fertility data. We discuss the quality of data provided by the Statistical Offices, both birth registration data and censuses or microcensuses. In addition, we explore how social science surveys can be used to generate order-specific fertility indicators, and compare fertility estimates across surveys with estimates from vital statistics. Prior studies have shown that there is a ’family bias’ in most surveys, with the fertility of the younger cohorts being overstated, because respondents with young children are easier to reach by the interviewers. Our assessment of various types of surveys from the three different countries does mostly support this notion. However, the ‘family bias’ is most pronounced in family surveys while all-purpose surveys suffer from it to a lesser extent. Weighting the data does not fully cure the ‘family bias’, which we attribute to the fact that number of children is not usually considered a factor in calculating sample weights, as provided by the survey agencie and Statistical Offices. The confounding role of migration in the production of reliable and comparable fertility statistics is also discussed.
The paper analyses the advantages and shortcomings of the two approaches of generating indicators of fertility patterns on the example of German speaking countries (Austria, Switzerland and Germany). Individual-level surveys like the Fertility and Family Survey (FFS), Generations and Gender Survey (GGS) or Population Policy Acceptance Survey (PPA) are primary sources for fertility analysis. Investigations from surveys are essential for understanding how socioeconomic and cultural factors determine family formation patterns. The main purpose of surveys is usually not deriving fertility rates, anyway it is important to see how reliable these data can be and whether they cover the reality sufficiently. Survey data are limited due to two crucial issues: sample sizes are too small and time periods are too short to display long time trends. The sample bias related to fertility estimates should be considered carefully – some members of population could be underreported in the sample, which might be corrected by incorporating weights. Hence, a clear validation is important for assessing the degree of reliance due to estimations from survey data. Vital statistics or data from population censuses have an important advantage over survey data because they provide a large number of recorded persons. Furthermore, demographic events are precisely recorded by official registers. However, the detail of the given information is not always sufficient (e.g. lacking information on birth order in Germany and Switzerland). The major question is to see if fertility indicators based on individual-level survey data differ substantially from those reported in vital statistics. Do we always face the same patterns of discrepancies due to the notorious problem of overestimation of fertility levels in surveys? The paper concludes that single and childless women are usually underreported in surveys, overestimating thus the level of fertility. The comparative approach of the paper allows assessing the quality of selected surveys.