6533b85ffe1ef96bd12c1300

RESEARCH PRODUCT

Flexible latent trait aggregation to analyze employability after the Ph.D. in Italy

Salvatore FasolaClara RomanoOrnella Giambalvo

subject

Statistics and Probabilityphd surveys. Indicators. Employability.05 social sciencesRank (computer programming)050301 educationEmployabilityComposite indicatorLatent traitStatisticsSTELLA (programming language)Settore SECS-S/05 - Statistica Sociale0509 other social sciencesStatistics Probability and Uncertainty050904 information & library sciences0503 educationMathematicsProportional odds

description

The analysis of satisfaction, employability and economic perspectives after the Ph.D. in Italy has not received adequate attention in the past, especially in terms of comparison among universities. To analyze these aspects, in this paper we consider data from the survey ‘Statistica in TEma di Laureati e LAvoro’ on doctors who achieved the title on 2007, 2008 and 2009 [CILEA, Laureati STELLA, indagine occupazionale post-dottorato, dottori di ricerca 2007–2008, Tech. Rep., CILEA, Segrate, 2010; CILEA,Laureati STELLA, indagine occupazionale post-dottorato, dottori di ricerca 2008–2009, Tech. Rep., CILEA, Segrate, 2011]. To deal with the complex, multidimensional nature of the concept, we propose a flexible two-step procedure for the construction of a composite indicator, and make a first attempt to rank some Italian universities. In the first step, indicators for single dimensions are derived from cumulative link models with proportional odds. In the second step, aggregation through standard, ad hoc methods is proposed

yearjournalcountryeditionlanguage
2015-09-02Journal of Applied Statistics
https://doi.org/10.1080/02664763.2015.1077797