The measurement of the concept of political/institutional trust has recently received increasing attention. In this stream, a number of analyses on items addressing trust in institutions at the individual level have resulted in factorial solutions that underline the multidimensionality of the concept. Several substantive interpretations of these results have been proposed. Nonetheless, little attention has been paid to the characteristics of the items used (e.g. dichotomous items, or four point-scales with skewed distributions) that hardly fulfil the requirements of factor analysis. The paper addresses this problem and shows how the cumulative nature of the data on institutional trust can be dealt with in a more efficient and convincing way through a non parametric IRT technique, namely Mokken scaling. The dominance model developed by Mokken (1971), considering explicitly the order of items in terms of absolute frequencies of positive answers, helps to take into account an aspect often removed from the analysis of trust, that is the varying trustworthiness of specific institutions. This has also conceptual consequences, as it show that once the order of trustworthiness of institutions is taken into account, the measure of institutional trust can be reduced to one single dimension. The paper finally addresses the problem of cross-country comparisons, proposing a systematic procedure to test invariance of item order across countries. The data used to carry out the analysis come from the European Value Study (1990-2000-2008).
