Marcantonio, Salvatore (2012) Causal models for monitoring University of Palermo Ordinary Financing Fund. PhD thesis, Università degli studi di Palermo.
![[img]](http://eprints.bice.rm.cnr.it/style/images/fileicons/application_pdf.png)
|
PDF
Marcantonio_PhD_Thesis_XXIII_DSSM.pdf Download (2MB) |
Abstract
Recently iterated decreasing government transfers and an even more increasing proportion of budget allotted based on competitive performances took Italian Universities started struggling with competition for funds, in particular for the University Ordinary Financing Fund (FFO). Three years ago, the Ministry of University and of Scientific and Technological Research outfitted a set of indicators to assess the quality of the educational offer and outcomes of formative processes and the quality of scientific research (Article 2, Law 9, January 2009, No 1). Although this system has been criticized by an authoritative and independent assessment organization, about 50 universities participate to the annual allocation funds under this scheme. Moreover, the share of FFO allocated in this way is constantly increasing. The University of Palermo has decided to initiate statistical studies to monitor the FFO indicators in order to increase the budget. These statistical models are aimed at informing several stakeholders on: which are the variables responsible for the indicators, what are their present and past values and how they relate with national references.This is the first step by which subjects having the institutional duty to a effect indicator values become aware of what is the state of the art; what could be the future values of these variables, look what are the components of strength and weakness in order to looking at measures for the correction of weaker performances; one or more strategies to be taken to increase the value of the indicators, also comparing different time and realization costs. All this aims can be achieved by both elementary statistical techniques such as tables and graphics useful to show crude quantitative results and, intuitive trends, and with more complex models able to dealing with time, in particular with short term forecast, and to work with a small amount of empirical evidence (because this is only the second year allocation so there are few past observations available). Models should be theoretically equipped with the distinction between predicting under observation and predicting under intervention, in order to provide correct answers to the distinct tasks of pure out of sample extrapolation and policy making, e.g acting on the system of rules governing relationships among variables and changing it. They should be also capable of encoding not only information arising from empirical data, but also from extra knowledge, such as expert opinions, in order to quickly adapt to new possible scenarios and keep a genuine uncertainty about a priori information.
| Item Type: | Thesis (PhD) |
|---|---|
| Additional Information: | Tutor: Prof.ssa Antonella Plaia Coordinatore Dottorato: Prof. Marcello Chiodi Dottorato di Ricerca in “Statistica, Statistica Applicata e Finanza Quantitativa”, XXIII ciclo - 2011 Settore Scientifico Disciplinare: SECS=S01 - Statistica,SECS=05-Statistica Sociale |
| Subjects: | 500 Scienze naturali e Matematica > 510 Matematica > 519 Probabilità e Matematica applicata > 519.5 Statistica matematica (Classificare qui i metodi parametrici e non parametrici) 600 Tecnologia - Scienze applicate > 607 Educazione; Ricerca; Soggetti connessi > 607.2 Ricerca; Metodi statistici |
| Depositing User: | salvatore Marcantonio |
| Date Deposited: | 05 Mar 2014 15:44 |
| Last Modified: | 05 Mar 2014 15:44 |
| URI: | http://eprints.bice.rm.cnr.it/id/eprint/8613 |
Actions (login required)
 |
View Item |
