Table 8

Robustness checks: changing sample selection and the assignment rule

Variables(1)(2)(3)(4)(5)
School to work transitionSchool to work transitionSchool to work transitionSchool to work transitionSchool to work transition
Narrower control groupAssignment rule based on birth cohortsPiemonte, Lombardia, Veneto and Trentino-Alto AdigeIncluding people who never workedShorter (+/−5 years) bandwidth
Reform−0.585*** (0.154)−0.292** (0.124)−0.465*** (0.137)−0.594*** (0.165)−0.324** (0.147)
Constant4.169*** (0.558)2.390*** (0.195)2.585*** (0.348)2.484*** (0.376)1.089 (0.730)
Observations6,9998,9855,52410,3655,508
R-squared0.1430.1430.1420.2090.136
Graduation year FEYesYesYesYesYes
Region FEYesYesYesYesYes
Year of birth FEYesYesYesYesYes
Individuals characteristicsYesYesYesYesYes

Note(s): The dependent variable is defined as the difference between the year the first job started and graduation year. The reform is a dummy equal to 1 for those individuals that graduated after the introduction of the reform (see Table 1). In column (2) we used an alternative definition of reform, defined as a dummy equal to 1 for those individuals who turned 14 after the introduction of the reform -see Table 1. Each regression includes a female dummy, a migrant dummy, and field of study's dummies. Robust standard errors in parentheses. ***p < 0.01; **p < 0.05; *p < 0.1

or Create an Account

Close subscription notice
Close access options