Graduation day for medicine and psychology students at Sapienza Università di Roma during the lockdown, Rome. Credit: Riccardo De Luca/AGF/Universal Images Group via Getty Images.
Editor's note: on 6 March 2023 a correction has been added at the end of this article upon request by the authors. See below for details.
Every year, the publication of academic rankings is reported with dismay in the Italian media due to the poor performance of our universities. This year was no exception.
The largest Italian university, the Sapienza University of Rome, home to 2021 physics Nobel prize winner Giorgio Parisi, ranked 171st according to the QS ranking, just below Texas A&M (164th). In the Times Higher Education (THE) ranking, Sapienza was in the 201-250 group alongside the University of Southern Florida. The position was slightly higher (101-150) according to the Academic Ranking of World Universities (ARWU), where Sapienza was ranked on par with the University of Utah. Other large universities in Italy fared worse: the University of Milan was ranked 324th by QS, below North Carolina State University, in the 301-350 group according to the THE ranking, alongside the University of California Merced, and in the 151-200 group according ARWU, together with Texas A&M. Going south the University of Naples - Federico II is ranked 416th by QS, lumped in the 351-400 group by THE, and in the 201-300 group by ARWU. Even a highly selective and prestigious school, like the Scuola Normale Superiore in Pisa, did not make it in the top 1000 universities selected by QS, was 183rd according to THE and in the 550-600 group for ARWU, together with University of Alaska Fairbanks.
We believe that the quality of the training provided by Italian academic institutions is not reflected by these rankings. But we need a quantitative evaluation to corroborate our qualitative perception. Nature recently published an interesting statistical study on the structure and dynamics of faculty hiring in US universities1.The analysis was based on a large dataset including information on the doctoral training of tenure-track faculty hired by all US universities that grant PhDs in the years 2011–2020. The authors were able to identify a small circle of prestigious universities that train a large majority of faculty across different academic fields. These elite universities exchange faculty between them with a high rate of self-hiring (i.e. PhD graduates who are hired as faculty by the same institution that trained them). They also provide faculty to less prestigious institutions, but rarely import faculty from them. While the analysis was centered on the US academic system, the primary data also included detailed information about faculty trained outside the USA. We decided to focus on faculty hired in the US, but trained by Italian universities.
We found that almost 3,000 faculty hired in the US obtained their PhD in Italy within the considered timeframe. This number is striking, especially when compared with the 7,384 tenure-track faculty hired in Italy during the same period. The situation for researchers in Italy has always been rather complex, with periods of recruitment followed by years without vacancies. Tenure-track faculty positions were established in 2010, but effectively entered into force only in 2013. Before that that date, Italy offered few opportunities to young researchers. After the introduction of tenure-track positions, the number of new positions for researchers grew, and it is now around 1,000 per year. This number is small when compared with the roughly 10,000 PhD graduating each year, considering also the backlog created by the past hiring freezes. This helps explain why many Italian graduates look for opportunities abroad.
We then decided to study the data in more detail. We first analyzed the scientific fields that were over- and under-represented within the Italian-trained faculty, as compared to what we would expect based on the overall share of Italians working in US universities. As shown in Fig. 1a, STEM subjects such as physics, biology and engineering were generally over-represented, while humanities such as history and sociology, except for classical studies, were generally under-represented. We then looked at the Italian universities where these US-hired faculty were trained (Fig. 1b). Sapienza ranked first with more than 400 trainees, followed by the University of Milan with more than 300 trainees. The University of Naples provided 143 faculty to US universities and the Scuola Normale Superiore 115. The list of US universities hiring Italian trained faculty is led by prestigious universities such as Columbia University, Weill Cornell and Harvard University (Fig 1c).
To understand more quantitatively the standing of the universities hiring Italian-trained faculty, we considered the ‘prestige’ score assigned to each US university department by the Nature study4. We found that 35% of the faculty trained at Italian universities were hired by US departments belonging to the top 25% group in terms of prestige. This fraction is 35% for Sapienza, 32% for the University of Milan, 36% for the University of Naples and 54% for graduates of the Scuola Normale Superiore. It is interesting to compare these values with those obtained for US universities ranked in the same class by major academic rankings: it is 11% for Texas A&M, 15% for the University of Utah and North Carolina State University, 7% for the University of Southern Florida and 8% for the University of Alaska.
Why do Italian universities have such bad rankings, then? While they often fare well in terms of scientific production indicators, they end up at the bottom of the list when considering two critical factors: reputation and internationalization. Reputation is a subjective matter that reflects common perceptions of prestige, reinforced by the rankings themselves. Furthermore, Italian universities traditionally attract very few international students. In our opinion, this is not because education is judged of poor quality, but rather because it is mainly offered in Italian. The Italian academic system is mostly funded by the government with the purpose of providing higher education to its citizens at an affordable cost. Attracting international students by providing appealing and costly extra-curricular amenities was never a goal. We think that it is important to evaluate universities using a variety of criteria. Different countries might have different goals for their education system, related to their social-economic and cultural needs. Ranking all universities together and comparing countries as if education was a race is misleading and useless.
We do not wish to suggest that there is nothing to improve in the Italian academic system, whose shortcomings are widely reported. Italy has structural problems related to the lack of opportunities for young researchers combined with the fact that research funding and faculty salaries are relatively low when compared with other countries. Italy does not appear to be able to capitalize the great potential provided by the people it trained. The data says, however, that the quality of training and research in many subjects, especially the STEM ones, remains high. We should continue to focus heavily on keeping the quality high and perhaps learn a skill in which US academic institutions excel: self-promotion.
CORRECTION, 06/03/2023.
After the publication of the article, we became aware that the database we used for our analysis contains systematic repetitions of the same data, since each hired faculty is listed separately for each field, for each domain and for the whole academia. Furthermore, individual faculty with multiple appointments in different fields are also listed multiple times. For this reason, our counting of the total number of Italian Ph. D. holders hired in the USA was incorrect. After eliminating the repetitions, the total number Italian resulted to be equal to 869 instead of 2804. Consequently the number of faculty hired per university has also to be corrected. Relative proportions remained, however, unchanged since the repetitions affect the entire database. Therefore the discussion about fields of activity of hired faculty and the prestige of hiring institutions remains correct. While the number of Italian trained faculty is smaller than what we incorrectly reported in the article, their absolute number still remains significant, so that our general conclusions do not change. We report below a new figure with correct numbers and apologize to the readers for the mistake.
