|
?:abstract
|
-
This study investigates how public trust in the police is measured in the Netherlands and abroad, as well as the validity and representativeness of these measurements. The study will serve to improve measurement practice in the Netherlands. The report seeks to answer three research questions: how is trust in the police measured in the Netherlands? How does this happen in other Western countries? How does survey research on this topic in the Netherlands compare to that in other countries? With its ‘Veiligheidsmonitor´, the Netherlands has a strong instrument to measure trust in the police. It is based on large to very large samples, and on high-quality fieldwork. Compared to other Western countries, measurement practice in the Netherlands is strong. The improvements suggested in this report therefore do not only apply to the Netherlands, but also to the foreign examples. A second measurement instrument, the ´Vertrouwens- en Reputatiemonitor Politie´ has not been analysed in detail due to a lack of transparency on the design, and limited access to the data. Dutch measurement practice covers a number of components of trust in the police very well, in particular through the ´Veiligheidsmonitor´. Trust in the responsiveness, competence and effectiveness of the police, as well as trust in police engagement with the community are well-covered. Distributive and procedural trust are covered, be it through a single item. This has risks for the validity of the measurement. Empirically, the scales and concepts used in the ´Veiligheidsmonitor´ cannot always be fully distinguished. The research observed that items on police in the neighbourhood or contacts between the police and local inhabitants are different from items that cover the police in way that is more general. This division runs through some of the theoretical concepts. This may mean that some of the items in the questionnaire do not measure what they intend to measure. The ´Veiligheidsmonitor´ has a number of gaps. It is remarkable that the questionnaire does not contain a measurement of general trust in the police. The questionnaire does also not include items on trust in other institutions and actors (such as the justice system, political institutions, or mayors) that would allow interpreting or calibrating trust levels. In addition, ´light blue´ is absent from all trust research, despite its growing importance. The focus in Dutch measurement practice is on the repressive tasks of the police, whereby the police is primarily seen as a ‘crimefighter’. Other police roles and tasks receive very little attention. Low trust becomes particularly problematic when it has an effect on citizens´ behaviour, such as their willingness to report, passive or active collaboration with the police, or likelihood to recommend working for the police. Behavioural effects of high or low trust, however, receive very little attention in current measurement instruments, despite the high relevance of such behaviour for policy. Most research in the Netherlands and abroad uses general random samples. These samples are in most cases very large in order to provide a detailed picture of local trends. The samples used in the Netherlands are large to very large. Response rates and representativity are good, but it is unclear whether current instruments reach parts of the population that have very low levels of trust in sufficient numbers. Not all instruments allow for making detailed partial analyses of smaller subgroups, such as vulnerable groups, or groups that receive special attention in policy. It is remarkable that, despite the large amount of data that is collected through the ´Veiligheidsmonitor´, data analysis remains limited. The questionnaire of the ´Veiligheidsmonitor´ still clearly shows the history of the instrument. The Dutch ´Veiligheidsmonitor´ has clear protocols and procedures for making changes to the questionnaire, and the effect of changes is monitored. This guarantees high quality, but it may also stifle innovation. As a result, it is not always clear whether current items in the questionnaire still have the same policy relevance as compared to the period when they were first included in the questionnaire.
(xsd:string)
|
![[This page as RDF]](https://data.gesis.org/gesiskg/static/rdf-icon.gif){kind=icon}