INVESTIGATING THE DIMENSIONALITY OF TORR: A REPLICATION STUDY
Various indices and ratings describing democratic processes in countries around the world have been developed by international organizations (such as Freedom House) and analytical centers (such as the one afﬁliated with the journal Economist). The main drawback of such ratings is that they only provide a linear ordering of countries by averaging a multitude of criteria. Such approach does not make it obvious which particular problems exist in which countries and thus does not help comparing democratic processes in different countries. In this paper, we propose a multidimensional model for ratings based on the mathematical discipline of formal concept analysis, which deals, in particular, with automated taxonomy construction from object–attribute data. In our case, every node of a taxonomy would group countries similar in certain aspects, while at the same time providing a description of these aspects. The aim is not to question the existing ratings, but rather to provide a neutral instrument for uncovering the structure of the data underlying these ratings. The proposed representation is much more informative than linear ratings, since it shows the commonalities and differences in the democratic development of various countries. In addition, it provides a solid ground for discussing, comparing, and criticizing ratings. It can also help formulate theoretical hypotheses on the evolution of democracy, thereby advancing scientiﬁc discovery. We illustrate the proposed representation with the case study of countries in Central and Eastern Europe and the former Soviet Union.
Relational reasoning is believed to be an essential construct for studying higher education learning. Relational reasoning is defined as an ability to discern meaningful patterns within any stream of information. Nonetheless, studies of relational reasoning are limited by the psychometric structure of the construct. For many instances, the composite nature of relational reasoning has been described as a bifactor structure. Bifactor models limit possibilities for studying the inner structure of composite constructs by demanding orthogonality of latent dimensions. Such assumption severely limits the interpretation of the results when it is applied to psychological constructs. However, over the last ten years, advances in the fields of Rasch measurement led to the development of the oblique bifactor models, which relax the constraints of the orthogonal bifactor models. We show that the oblique bifactor models exhibit model fit, which is superior compared to the orthogonal bifactor model. Then, we discuss their interpretation and demonstrate the advantages of these models for investigating the inner structure of the test of relational reasoning. The data is a nationally representative sample of Russian engineering students (N = 2036).
Institutions affect investment decisions, including investments in human capital. Hence institutions are relevant for the allocation of talent. Good market-supporting institutions attract talent to productive value-creating activities, whereas poor ones raise the appeal of rent-seeking. We propose a theoretical model that predicts that more talented individuals are particularly sensitive in their career choices to the quality of institutions, and test these predictions on a sample of around 95 countries of the world. We find a strong positive association between the quality of institutions and graduation of college and university students in science, and an even stronger negative correlation with graduation in law. Our findings are robust to various specifications of empirical models, including smaller samples of former colonies and transition countries. The quality of human capital makes the distinction between educational choices under strong and weak institutions particularly sharp. We show that the allocation of talent is an important link between institutions and growth.