Search

This paper presents three applied analysis. Our principal interest is to understand how meso (university level) and macro (national level) contexts interact with micro phenomena such as students’ careers. To achieve this aim, the paper shows a secondary analysis of longitudinal administrative data by means of Sequence Analysis and Event History Models. Furthermore, it shows the results of an additional analysis using a quasi-experimental research on longitudinal data. The paper examines typical pathways of Italian students, by means of administrative information’s on student enrollment from Sapienza University of Rome. The aim of this paper is triple: 1. describe students’ careers in higher education by building a typology of pathways using Sequence Analysis. 2. Identify how socio-economic and macro level characteristics affect students’ careers by multinomial logistic regression model where clusters are used as dependent variables and focus on the study of the “event” by using Event History analysis. 3. Evaluate the Italian Higher Education reform policies and their main outcomes throughout a quasi-experimental research. Mainly, the outcomes show the importance of the warming-up period, the individual choice and the exogenous events after enrolment in determining the success/failure of each career. Our outcomes suggest that Italian universities should rethink the mechanisms available to manage failure and guide student choices.

This paper evaluates the effects of a high school curriculum reform on students’ probability to enroll at university and to choose a Science, Technology, Engineering or Mathematics (STEM) major. The reform increased the difficulty of graduating from high school by increasing the instruction time in core subjects and by raising the graduation requirements. Based on administrative data covering all students, the analysis is carried out by applying a difference-in-differences model. The results show that the reform increased university enrollment rates for both genders. With regard to choosing STEM as college major, we find a robust positive effect on males.

Leder hårdare straff till färre brott? Det sambandet ämnar vi undersöka i denna uppsats. För att att göra det analyserar vi 2010 års straffmätningsreform som förlängde straffen för allvarliga våldsbrott. Reformen påverkade vissa brott men inte alla, vilket vi utnyttjar som ett naturligt experiment i en difference-indifferencesansats. De brottstyper som påverkades används som behandlingsgrupp, medan de som inte påverkades används som jämförelsegrupp. Därefter tolkas skillnaden i förändring hos de båda gruppernas brottsfrekvenser ut. När hela efterperioden analyseras finner vi ingen statistiskt signifikant effekt. Analyseras varje år i efterperioden för sig finner vi en negativ och statistiskt signifikant effekt på 14,9 % av reformen året efter den trädde i kraft (2011)., Do longer prison sentences lead to fewer committed crimes? This paper pursues to evaluate this causation through a difference-in-differences analysis. In 2010, there was a reform in Sweden that meant to make prison sentences longer for some crimes. The fact that the reform affected some crimes but not others is exploited as a natural experiment. We use the types of crime that were affected by the reform as a treatment group and the types of crime that were not affected as a control group. Then we assess the difference in frequency change of the two groups of crimes after the reform took effect. When the whole post reform period is analyzed, we find no statistically significant effects. When the effect is isolated one year after the reform was implemented (2011) we find a statistically significant effect of 14.9 %. There is no statistically significant result when the reform is isolated for 2012 and 2013.

Leder hårdare straff till färre brott? Det sambandet ämnar vi undersöka i denna uppsats. För att att göra det analyserar vi 2010 års straffmätningsreform som förlängde straffen för allvarliga våldsbrott. Reformen påverkade vissa brott men inte alla, vilket vi utnyttjar som ett naturligt experiment i en difference-indifferencesansats. De brottstyper som påverkades används som behandlingsgrupp, medan de som inte påverkades används som jämförelsegrupp. Därefter tolkas skillnaden i förändring hos de båda gruppernas brottsfrekvenser ut. När hela efterperioden analyseras finner vi ingen statistiskt signifikant effekt. Analyseras varje år i efterperioden för sig finner vi en negativ och statistiskt signifikant effekt på 14,9 % av reformen året efter den trädde i kraft (2011)., Do longer prison sentences lead to fewer committed crimes? This paper pursues to evaluate this causation through a difference-in-differences analysis. In 2010, there was a reform in Sweden that meant to make prison sentences longer for some crimes. The fact that the reform affected some crimes but not others is exploited as a natural experiment. We use the types of crime that were affected by the reform as a treatment group and the types of crime that were not affected as a control group. Then we assess the difference in frequency change of the two groups of crimes after the reform took effect. When the whole post reform period is analyzed, we find no statistically significant effects. When the effect is isolated one year after the reform was implemented (2011) we find a statistically significant effect of 14.9 %. There is no statistically significant result when the reform is isolated for 2012 and 2013.