• Collecting data

Collecting data

This paper argues that general subjective corruption indicators run the risk of reflecting citizens’ general predispositions towards government, rather than actual experienced corruption.

The authors say that most of the items that are used in corruption surveys do not rely on respondents’ experiential basis, and leave considerable freedom to respondents whether to take certain aspects into account. This creates problems of comparability, and invites respondents to broaden their frame of reference to whatever factor they wish when giving an opinion on corruption.

In the first part of the paper, the authors briefly present some of the available survey material on citizens’ perception of corruption in Belgium. Subsequently, they show that many opinions on government are in fact part of a general predisposition towards government, and that these opinions can therefore not be considered as expression of individual experience. This is illustrated with a series of survey items on trust in institutions. The authors further illustrate these findings by analysing opinions on corruption, which mainly follow the general attitudinal tendencies.

The authors suggest that making surveys measure specific occurrences and expressions of corruption is one possible way for ameliorating current corruption indicators. In this way, the indicators are no longer ‘contaminated’ by the general predispositions. Efforts to develop methods for measuring corruption objectively should be increased, but, the authors conclude, unfortunately the literature does not yet offer many examples of how this may be done.

• Full text of document
• Publisher details
• Send to a friend