Paradigm Shift in Demography? <br>Editorial to the Special Issue “Tempo Effects in Demographic Period Indicators”<br>

Abstract

Demography mainly defi nes itself-in contrast to other disciplines of the social sciences-through the indicators resulting from its range of methods, like the "number of children per woman", "divorces per marriage" or "average life expectancy". Derived from the period perspective these indicators are the basis of many studies in the social sciences and also for projecting future trends. Until today, theoretical approaches are rare in demography. Consequently, demography is still a mainly quantitative discipline, however one with important interdisciplinary character since on the one hand all areas of the social sciences are dependent on demographic indicators and on the other hand demography itself uses the theories and hypotheses of its neighbouring disciplines. In recent years, however, the basic and since many decades well established concept of demographic period estimation has been questioned. Shortly before and after the turn of the millennium the demographers John Bongaarts and Griffi th Feeney published two papers in which they showed that age-specifi c period rates-which are the basis of the most commonly used period indicators for fertility (total fertility rate) and mortality (life expectancy)-not only include the relative quantum of the demographic events but also so-called "tempo effects" (Bongaarts/Feeney 1998, 2002). These effects appear whenever the average age at which the analyzed event occurs changes during the period of observation. Since in most applications demographic indicators are supposed to refl ect the currently realized fertility or experienced mortality, according to the authors tempo effects have to be seen as undesired distortions. Therefore, Bongaarts and Feeney suggested to adjust demographic period indicators for tempo effects. 1 With this