Hoem Jan M
a Stockholm University.
Popul Stud (Camb). 2014;68(2):151-9. doi: 10.1080/00324728.2013.843019. Epub 2013 Nov 21.
In studies of the fertility of migrants in which the data are confined to the migrants only, estimation bias will normally appear in comparisons of childbearing before and after migration. The same issue arises in studies of union formation before and after first birth, marriage formation before and after home purchase, and in any other comparison of behaviour before and after an index event if one confines the study only to those who have experienced the index event. It is normally better to avoid analysis of behaviour before the index event because such analysis actually conditions on the later arrival of the index event. In this paper, we provide graphical and mathematical representations of this problem and show how one can get a meaningful (unconditional) comparison of behaviour before and after the index event provided the data contain enough information for both sub-periods. Otherwise, the analyst should refrain from making a comparison of this nature.
在仅将数据局限于移民的生育力研究中，在比较移民前后的生育情况时通常会出现估计偏差。在首次生育前后的伴侣关系形成研究、购房前后的婚姻形成研究以及在任何其他将研究仅局限于经历了指标事件的人群的指标事件前后行为比较中，都会出现同样的问题。通常最好避免对指标事件之前的行为进行分析，因为这种分析实际上是以指标事件的后来发生为条件的。在本文中，我们给出了这个问题的图形和数学表示，并展示了如果数据包含两个子时期的足够信息，如何能够对指标事件前后的行为进行有意义的（无条件的）比较。否则，分析人员应避免进行这种性质的比较。
