PUBBLICAZIONE/PUBLICATION

Permanent URI for this communityhttps://gendermore.unimore.it/handle/123456789/1575

Browse

Settings

Subject: search.filters.subject.ItalyViewpoints: search.filters.viewpoints.Student Level

Search Results

Now showing 1 - 2 of 2
  • Item type:Publication,
    Academic Students’ Progress Indicators and Gender Gaps Based on Survival Analysis and Data Mining Frameworks Open PDF 
    (Springer Science and Business Media B. V., 2020)
    Marshall, A.H.
    ;
    Zenga, M.
    ;
    Kalamatianou, A.
    This paper examines gender differences according to two new indicators in higher education studies. One indicator represents the length of studies beyond the minimum requirement and the other represents a harmonised graduation mark. We explore these indicators in a statistical framework that consists of Survival analysis Kaplan–Meier methods, Accelerated Failure Time models, Survival Trees, Multivariate Regression Trees and a modified Gender Parity Index. This unique combination of statistical methods allows both to analyse data involving censoring and to incorporate other explanatory variables in the analysis. The approaches were applied to data taken from a Greek and an Italian University and provide evidence that the survival analysis is a useful tool for exploring gender gaps in higher education while the Accelerated Failure Time model permits the investigation of how other variables can influence the gaps. The data mining techniques of survival trees and multivariate regression trees allows for the importance of such influencing variables to be measured and illustrates these in a simple to view tree structure. The Multivariate Regression Tree approach allows us to consider more than one continuous outcome variable so we were able to consider the two proposed indicators simultaneously. Interesting insights from this analysis are that gender has an important role and women outperform regarding these new indicators by taking less length of studies, less graduation time with higher performance, controlling also for other student characteristics. The modified gender parity index enriched the results in all stages. © 2020, Springer Nature B.V.
  • Item type:Publication,
    A gender analysis of top scientists’ collaboration behavior: evidence from Italy Open PDF 
    (2019)
    Abramo, G.
    ;
    D’Angelo, C.A.
    ;
    Di Costa, F.
    This work analyzes the differences in collaboration behavior between males and females among a particular type of scholars: top scientists, and as compared to non top scientists. The field of observation consists of the Italian academic system and the co-authorships of scientific publications by 11,145 professors. The results obtained from a cross-sectional analysis covering the 5-year period 2006–2010 show that there are no significant differences in the overall propensity to collaborate in the top scientists of the two genders. At the level of single disciplines there are no differences in collaboration behavior, except in the case of: (1) international collaborations, for mathematics and chemistry—where the propensity for collaboration is greater for males; and (2) extramural domestic collaborations in physics, in which it is the females that show greater propensity for collaboration. Because international collaboration is positively correlated to research performance, findings can inform science policy aimed at increasing the representation of female top performers. © 2019, Akadémiai Kiadó, Budapest, Hungary.