3 LIMITATIONS OF RESEARCH THIS THESIS DID NOT SET OUT TO TEST A...

8.3  Limitations  of  Research  

This thesis did not set out to test a hypothesis – it was more exploratory by design. Given the high

number of variables involved in research at the national level it is not wise to generalise beyond the facts

of this research at the time the study was undertaken. For example, if, in the future, the Dutch

Association of Insurers proposal for mandatory private flood insurance is eventually accepted, then the

conclusion that current conditions are not sufficient for effective private flood insurance would no longer

hold true. Furthermore, if there are more flood events in the future and the Dutch public awareness of

flooding increases, the market for private flood insurance could expand rapidly alongside the WTS even

when it is not reformed.

A second limitation of this research involves the research design. The conditions for effective private

flood insurance were identified using a comparative cross-case methodology. The cases were chosen to

be as geographically and culturally contiguous as possible. The matching of cases attempted to increase

the internal validity of the research. However, cross-country comparison will have very limited external

validity and limited scope to generalise the results beyond the cases and geographic regions studied. It

is never the case when making inferences from national police that one is comparing apples with apples.

Given the historical, cultural, political and topographic differences between countries, a one size fits all

solution or policy recommendation is not sensible. This study’s design tried to minimise difficulties with

cross-case comparisons through the identification of common compensation system components and

then triangulation of the results using expert interviews based on standardised questions. The inherent

weakness, however, remains.

82