Bruno Chiandotto

Quality Assessment of the University Educational Process: an Application of the ECSI Model

In a university, students represent the final users as well as the principal actors of the formative services. A measure of their perceived quality is essential for planning changes that would increase the level of the quality of these services. This perceived quality is analysed in this paper with the ECSI (European Customer Satisfaction Index) methodology. The ECSI, which implements a structural equation model, is aimed to represent the satisfaction of the students with some latent variables gauged through a set of observable indicators. We extend the ECSI to the data obtained from graduates of the University of Florence employed one year after graduation.

Bayesian and Non-Bayesian Approaches to Statistical Inference: A Personal View

Bayesian and non-bayesian approaches to statistical inference are compared giving particular attention to the emerging field of causal statistical inference and causal statistical decision theory. After a brief review of the evolution of statistical inference, as extraction of information and identification of models from data, the problematic issues of causal inference and causal decision theory will be reviewed. The aim is to provide some basic ideas for unifying the different approaches and for strengthening the future of statistics as a discipline.

The Effectiveness of University Education: A Structural Equation Model

The evaluation of the effectiveness of higher education is a crucial aspect of competitiveness of modern economies. In this contribution we investigate the quality and effectiveness of higher education in Italy using a structural equation model; in particular, we evaluate the performance of the university system from the users’ point of view, both immediately following (internal effectiveness), and one year after (external effectiveness), the completion of the degree. The model allows the construction of synthetic indexes and hence the ranking of study programs.

Special issue on robust multivariate analysis and classification

It is now widely recognized that most classical statistical techniques are not resistant in the presence of outliers. Correspondingly, the development of highly robust and efficient statistical methods has become a goal of paramount importance in both theoretical and applied research. The demand for such methods has been driven by the increasing availability of data in almost any area of scientific research. These data sets are not only becoming larger in size, but also in complexity. The extraction of essential features and the discovery of structures and relations in complex data sets must not break down when atypical observations are present. In addition, there is a need for the development of effective diagnostics that can help to pinpoint these outliers. With many variables at hand, outlying observations can be hard to detect. Outliers need not necessarily be associated with “gross” contamination errors, but may instead contain valuable information. An example is the indication of the existence of several populations instead of one. While robust statistical methods and diagnostic tools are well established for studying data sets under simple univariate models, this is not the case for more complicated multivariate situations.

SIS-RAI : il Sistema Informativo Statistico della Direzione Amministrazione Abbonamenti di RAI - Radio Televisione Italiana

SIS-RAI: the statistical information system of the Direzione Amministrazione abbonamenti of the Italian public radio and television system (RAI) (by Bruno Chiandotto, Bruno Bertaccini) . Objectives Current Italian law, (R.D.L. 21/02/1938 n. 246), requires that every household owning one or more TV sets should pay a subscription fee. Because this norm is not always complied with, a specific agency (Direzione Amministrazione Abbonamenti) was created in order to enforce it, thereby acquiring new subscriptions, maintaining the old subscription portfolio and recuperating subscriptions not paid previously. The objective of the present paper is to illustrate the main characteristics of the Statistical Information System implemented by the authors for RAI. Methods and Results The Statistical Information System has to deliver information regarding the different structural components of the TV audience. In addition, the information system should provide the agency with year-end forecasts regarding flows in the characteristics of the TV audience. These forecasts are made utilizing the methodology proposed by Balestra and Matoussi in 1984. This model, suitably generalized and adapted to the case on hand, should generate three different forecasts, (Low, Average, High) which are updated each month on the basis of the newly-acquired data. Since 2005, the Agency has made large use of the Statistical Information System software for its activities, that is implemented on intranet website. The System now produces a rich and well organized set of information on the basis of specific queries (registered viewers, paying viewers, default debtors, new subscribers, cancellations, ecc.) and the mentioned forecasts. These forecasts are then compared to the Agency’s pre-defined goals. If there are differences between the forecasts and the goals, and especially if these fall short of the goals, attention must be given to the most critical aspects which have emerged, in order to implement corrective interventions. Conclusions The peculiarity of the implemented system is the transformation of statistical data in information that becomes knowledge directly utilizable for decision purposes.

Measurement of University External Effectiveness Based on the Use of the Acquired Skills

In this paper, we analyse the skills used at work, 18 to 30 months from the completion of studies, by the students who graduated at the University of Florence in the year 2000. The aim is pursued by detecting the determinants of the phenomenon with particular attention to the possible differences between study programmes. We performed two analyses: in the first, we identified homogeneous groups of degree programmes and applied a proportional odds (logistic) model for each group and a partial proportional odds model for the whole university. The second analysis was an ordered logistic model with random intercept having two levels of aggregation with the degree types as second-level units.