In an earlier SWOV study (Polak, 2001; see C 18040 (ITRD E203832)) a linkage was made between the data files of road accident victims (as recorded by the police) and inpatients (as recorded by hospitals). The purpose was to determine the real number of road accident victims annually admitted to hospital in the Netherlands. The linkage used records from both files that were as similar as possible. Also, a new method was used to determine how many matched records actually referred to one and the same victim. This so-called footprint method uses a variable that takes no part in the matching: in the present and previous study, this is the mode of transport of the victim. Using the footprint method has its advantages and disadvantages. When the least squares method is used, there are sometimes negative values. When the footprint method is used, there are none. However, a disadvantage of the method is the necessity of having to use an advanced calculation system such as Mathematica.
Abstract