Abstract:

The purpose of this research is to improve higher education infrastructure reports by presenting a latent trait approach to data collected by Ministry of Science and higher Education. In this paper, higher education infrastructure indicators are modeled as questionnaire items, and frequency and continuous integer values are recoded categorically for analysis with a Rasch model for rating scales. This research demonstrates applicability of Rasch models to frequency and continuous integer values by constructing a common dimension for both institutions and infrastructure indicators. These results suggest the traditional method of comparing institutions with separate indicators may not be taking full advantage of information reported from higher education institutions in Russia. When infrastructure indicators were consolidated into a coherent latent trait, two and possibly three strata of institutional quality were identified with important policy implications. This research presents an alternative to traditional methods of reporting statistical information about higher education institutions by suggesting that all infrastructure indicators are related to a common latent trait. Then a formal empirical analysis was conducted to evaluate practical implications of this approach. An important advantage is the overall perspective that consolidation provides when infrastructure indicators and institutions are brought into a common quantitative framework. The consequence is a more powerful analysis than isolated, separate indicators without sacrificing information about separate indicators.

Keywords:

Higher education infrastructure, measurement, linear scale, Rasch model.