Chapter 4 Data

The data were obtained from surveys conducted annually since 2005, by the STELLA (Statistiche in tema di laureati e lavoro) initiative of the CILEA consortium. Based on the CATI method, only two surveys, conducted in 2006 and 2011, are taken into account.

The graduates were surveyed one year after their graduation on aspects regarding their education, career paths and current economic situation. We focused our attention on individuals with first-level degree because some analysed universities have few second-level graduates. In this case, the outranking method used can produce biased results.

We divided the first-level graduates into two groups of subjects: Economics and Statistics (ES) and Political and Social Studies (PSS), which are distinguished by different human capital characteristics and different results in employment rates, and in overeducation and mismatching rates [8]. The subjects were chosen according to the evidence in the literature (see [27, 3, 16] and previous analyses of available data [1, 90, 15, 6]).

The criteria in the model were chosen by considering two subsets of covariates, namely university graduate characteristics and job characteristics.

The graduate characteristics related to the University may reflect the reputation of each university. They include g₁, the percentage of graduates who have not received classical or scientific high-school education; g₂, the mean graduation grade (mini­mum 60 maximum 110); g₃, the effective length of the study period in years (mean length/legal time); and g4, parental background (the median value of a variable obtained from analysing both the job of the parents and their education level; for further analyses, see the methodological note in [15, 14]). Many evidence in litera­ture [90, 15] show as the parental background influence student in the choice about disciplines and above all in the choice of the better university.

The job characteristics refer to the job placement of graduates. These charac­teristics depend on each University and its capability in the third mission activity, but also on the labour market placement of graduates and their on-the-job self­assessment. Efforts were made to introduce the human capital factor as a result of a mix of education, soft skills and ability in the job. They include g₅, the percent­age of full-time employment; g₆, average monthly net income; g₇, the percentage of employed reporting that they have decisional autonomy; g₈, the percentage of employed reporting that they have job responsibility over other employees; g₉, the percentage of overeducated; and g₁₀, the percentage of mismatched. Mismatching is measured as a proportion of graduates working in jobs for which acquired skills are not required and overeducation is measured as a proportion of graduates working in jobs for which a university degree was not a formal requirement (definitions adopted in most empirical papers, see par.

Although some scholars propose a multi-dimensional approach to defining mis­matching and overeducation [24,4], for our purposes, considering the available data, both overeducation and mismatching are considered univariate concepts. They are measured by a dichotomous variable that divides the workers into two groups, namely all those who are mismatched and/or overeducated, and all the others.

The size of the analysed subpopulation of graduates is 6,505 in 2006 and 7,837 in 2011 (Table 4.1).

STELLA universities are codified using ai indices to ensure privacy. Using the appropriate language of the Electre III models, the results are summarized in perfor­mance matrices. Tables 4.2, 4.3 and 4.4 present the performance matrices relating to the entire data set and those relating to Economics and Statistics graduates, and Political and Social Studies graduates, respectively.

The data show a medium or medium-high social background (g₄) (Table 4.2). For a₆ and a₇ universities only, the social background shows an improvement during the two years (Table 4.2). Social background improved for Economics and Statistics graduates at α₆ and α₇ universities; and for Political and Social Studies graduates at a₆ (Tables 4.3 and 4.4).

Table 4.1 Graduates in 2006-2011 by subject, university, prior to and during the economic crisis

The employment rate changes during transition from the pre-crisis to the during crisis period with Political and Social Studies graduates (g₅ indicator) being affected more compared to Economics and Statistics graduates (Tables 4.3 and 4.4).

More than 50% of graduates with a job declare having decisional autonomy in the job (g₇). Between the pre-crisis period and the crisis period, this indicator increases at all the universities except for a₁ and a₅.

Less than 1 out of 4 employed graduates said that they were responsible for other employees (g₈). This rate is higher for α₅ graduates, following an upward trend from 2006 to 2011.

The average monthly net income of employed graduates (g6) changed from 2006 to 2011. In the 2011 survey, only a2 and a4 university graduates earned an aver­age monthly net income exceeding 1000 euro (Table 4.2). The income earned by Economics and Statistics graduates is better than for others (Tables 4.3 and 4.4).

Overeducation (g₉) exceeds 50% for all disciplinary groups, universities and both surveys (2006 and 2011). In the 2011 survey, this value increased by 20% (Table 4.2). These results confirm the evidence available in the literature [44, 6, 60, 16, 85].

As regards the Economics and Statistics graduates, the highest increase is noted for universities a₁ and a₅ (Table 4.3). Overeducation is stable for the Political and Social Studies graduates of university α₇ (Table 4.4).

The graduates of university a2 are associated with a lower mismatching value (g₁₀) than that of other university graduates (Table 4.2). Bigger, larger increase of the g₁₀ indicator was observed for university a₆ (+25.4%, Table 4.2).

Between the 2006 and 2011 surveys, a₁ graduates recorded a relevant decrease in g₁₀. An increase in g₁₀ was noted for α₂, a₄ and specially for α₆ graduates (Table4.2). The g₁₀ indicator does not reach 50% for Economics and Statistics graduates, while for Political and Social Studies graduates, this threshold is exceeded in both the surveys (Tables 4.3 and 4.4).

Table 4.2 Performance matrix for graduates, prior to and during the crisis period

(a) 2006 Graduates

(b) 2011 Graduates

Data

Table 4.3 Performance matrix for Economics and Statistics graduates, prior to and during the crisis period

Table 4.4 Performance matrix for Political and Social Studies graduates, prior to and during the crisis period

University Job

Data

Data