Mitchell Report

PREPARED FOR THE MEETING OF DIRECTORS-GENERAL FOR HIGHER EDUCATION AND PRESIDENTS OF RECTORS' CONFERENCES

The promotion of cooperation at doctoral studies level was first discussed at the Informal Council of Education and Research Ministers at Schwerin (July 1994) under the German Presidency. In the framework of the French Presidency of the European Union during 1995, a seminar on doctoral studies was organised by the French Ministry of Education at Epinal in March 1995. Its aim was to collect information on doctoral programmes, policy and systems in Europe and examine the means of reinforcing European cooperation in the subject. Prospects for developing cooperation in the field were also examined at the Director-Generals of ministries of Education Conference held at Salamanca under the following Spanish Presidency.

A report on doctoral studies in the European region by CEPES/UNESCO was published in 1994. However, since there was a need for a more detailed knowledge than that provided by the study on a number of aspects of doctoral studies, the Commission funded a study by the "European Institute for Citizens' Rights" in 1995. In particular this took account of the experience acquired by the International Temporary Advisory Committee on Doctoral Studies (CPRT) involving The Netherlands, France, Germany and Belgium which had put forward a number of suggestions for encouraging cooperation at this level of studies.

The basic aim of the Commission's 1995 study - which was based on the results of a questionnaire-based survey - was to add to and enrich the available data in order to allow for a closer study of the situation and of the different structure of doctoral systems in Europe. Furthermore, the study examined different types of European cooperation concerning doctoral studies that were developed in the framework of the ERASMUS Programme.

Using the information collected from EU Member States (as well as Norway, Switzerland and Iceland), the results could be summarised as follows:

° approximately 250,000 students are in doctoral studies, or over 50,000 students embark on a thesis each year;

The study also highlighted the great variety of doctoral systems in place in Europe, but it also showed where there was some convergence. It was suggested that the reinforcement of the quality of doctoral programmes and the establishment of a high-level scientific and technological community are goals that can be attained through cooperation and mobility.

Recent developments underway in higher education policies such as the Bologna Declaration on the creation of a European Higher Education Area^{^[1]}; diploma recognition issues; gender equality; e-Learning as well as the recent initiative of the European Union to create a European Research Area^{^[2]} are playing their part in making a reality of the " knowledge society".

Since then, Ministers meeting at the European Councils' of Lisbon (1999), Nice (2000), Stockholm (2001) and elsewhere have all reiterated the importance of academic and student mobility in Education and Training policies and the pre-eminant role that cooperation in education and research can play in employment policies and in enhancing the knowledge and information society in Europe.

A strong education base is a guarantee for remaining competitive in a global economy and the information society. It is one of the basic driving forces behind economic and social progress and a key factor in business competitiveness, employment and the quality of life.

Encouraging young researchers to participate actively in international and European co-operation networks and to undertake part of their advanced training (postgraduate and doctoral studies) in another country will give them a valuable experience of working in a multicultural environment. Such cooperations will contribute considerably to the maintenance and improvement of European competitiveness in all areas. It will also speed up and facilitate the pooling of resources and the creation of common infrastructures.

Doctoral studies are at the boundary of higher education and research: being at one and the same time, part of the academic educational process, with the eventual awarding of an academic qualification (e.g. Ph.D), and the basis of a research training for an eventual career in research, academia or elsewhere in the labour market. As such it is important to have a thorough and up-to-date knowledge of the doctoral education and training systems and structures in place in Europe together with reliable data, as a basis for future Community policy-making in the area of higher education and research.

Understanding the essential characteristics of the different national doctoral studies systems offers a basis for the exchange of experiences. It is also essential for improving cooperation in this area, the need and utility of which are amply demonstrated by the wide range of initiatives already carried out in this field.

An updated and comprehensive survey of the data on doctoral studies will help expand and verify the data on doctoral studies in the different universities and countries participating in the Socrates/Erasmus programme.

· to analyse the evolution of the cooperation concerning doctoral mobility and studies at Community level in the framework of the Socrates/Erasmus programme (1995-2000), particularly in the light of the organisational changes in moving from the ICP (inter-university cooperation programmes) to the IC (institutional contract) mode but also from the fact that the number of countries participating in the programme has almost doubled, from 18 to 29.

· to present a detailed updated view of the salient features of the different National doctoral studies systems within the 29 countries^{^[3]} now participating in the Socrates/Erasmus programme. This will be achieved by collating detailed data on the national situations in these countries (especially the numbers and types of institutions conferring doctorates, the number of theses examined, the procedures and legal requirements to obtain a doctorate, the number of mobile doctoral students both nationally and within the Erasmus programme), so that a comparison of the different systems can be made.

· to survey the different models of subject-based doctoral study cooperations currently underway in Europe (such as doctoral schools, European doctorates, joint curricula, information exchange, networking) and where possible their effects on employment, by collecting and analysing data from the "Thematic Networks" operating in the Socrates/Erasmus programme since 1995.

The rest of this paper will present an initial analysis of the doctoral mobility in the framework of the Socrates/Erasmus programme (1995-2000) as a basis for discussion in the present meeting.

The following problem areas, which are believed to influence the overall doctoral mobility within the programme, are highlighted;

· With the advent of eleven new countries into the Erasmus programme, is the evolution in doctoral mobility since 1995 satisfactory compared with All student mobility?

· In the light of the two-cycle/three cycle debate stimulated by the Bologna process, can an effect on the age distribution in the doctoral mobility by country be observed?

· Is the gender balance for doctoral mobility satisfactory with respect to subject popularity, country flows, etc.?

· Compared with 'All' student mobility, is the requirement that Erasmus mobility must be not less than three months and not more than twelve months optimal for doctoral students?

· How does the subject area chosen by doctoral students compare to those chosen by ''All' students?

· Compared with 'All' student mobility, are the doctoral mobility flows between countries satisfactory?

The analyse was made on the full data sets (the so-called "anondata") that were provided regularly by the Erasmus National agencies as part of their annual final reports to the Commission. These data sets contained detailed information on all student mobility flows actually having taken place within the relevant academic year. From 1997/98 onwards (under the new management structure of the IC) they contained information on the home institution; age; sex; nationality; subject code; degree title; level of study; host institution; months abroad; month in which study commenced; years of prior study; use of ECTS; special needs; teaching in host language; period of industrial placement; Erasmus grant, etc. Unfortunately, no information on the level of study was available for the transition years 1995/96 and 1996/97 where the ICP regime was still operative.

However, to draw comparative conclusions from the data in fact proved rather more complicated than initially envisaged. This was not only the result of the increasing number of countries, and hence universities, participating in Erasmus, which makes a year-on-year comparison difficult but also because the divergence in national education systems and qualifications (highlighted in the Bologna process as the two-cycle versus the three cycle higher education structures) which makes a simple comparison and study methodology difficult.

Concerning diploma titles, we referred where necessary to the latest available report on the higher education systems, qualifications and diploma titles published by the European Commission in 1998^{^[4]} and the reports published by the Eurydice Unit for the EU countries and the CEEC countries (1999)^{^[5]} and (1999)^{^[6]}.

Concerning the level of study (U,P,D) provided in the data sets, certain arbitrary distinctions have been made in the analysis of the "anondata", to achieve a more comprehensive comparative study from which general conclusions might be drawn.

In an effort to bridge the differences in mobility numbers inherent in the (long) initial first cycle of countries having a two-cycle system and the (shorter) initial first cycle for countries having a three-cycle system, the number of postgraduate students (the "P" level) from countries having the two-cycle system was taken to be the sum of all those who are identified as having studied for three or more years. This affected in particular the countries of AT, DE, DK, ES, IT, NL, SE.

For the doctorate level, the number of students that were identified by the National Agencies as studying at "D" level was taken as the guiding figure but it quickly became obvious that this was not a sufficient condition since some countries chose incorrectly to include 'Master' level diplomas (MA, MSc, Maîtrise, Magistère etc.) or, even in some cases, 'Batchelor' level diplomas (BA, etc.) under "D". In other cases, doctorate level diplomas such as Ph.D or D.Phil. were assigned as "P" (postgraduate level). Therefore it was decided to also take into account the 'diploma title' and 'years of study', where possible to distinguish between the "P" and "D" levels. Certain professional qualifications such as medical, veterinary and pharmaceutical doctors were included as doctoral level although, "engineer" (ingénieur) have been put in "P" level, due to the heterogeneity in the nature of the diploma title.

The analysis and figures which follow are presented under the following headings:

Figure. 18 Ratio of doctoral to 'All' student mobility flows for non-EU countries

The general conclusion is that doctoral level mobility is becoming an increasing part of the overall Erasmus Programme mobility. In the academic year 1995/96, 471 students were studying at doctoral level compared to 17,841 postgraduate students (i.e. 2.6%) and 65,251 undergraduates (0.72%). By 1999/00, the figures had increased to 1,307 doctoral students compared to 34,700 postgraduate students (i.e. 3.8%) and 71,621 undergraduates (1.8%). A three-fold increase in doctoral students can be observed compared to undergaduate students and a nearly two-fold increase in doctoral students compared to postgraduate students.

^{^[1]} http://europa.eu.int/comm/education/socrates/erasmus/bologna.pdf

^{^[2]} Towards a European research area - COM (2000) 6 - 18 January 2000

^{^[3]} Austria, Belgium (both linguistic regions), Germany, Denmark, Spain, Finland, France, Greece, Ireland, Italy, Luxembourg, The Netherlands, Portugal, Sweden, United Kingdom, Bulgaria, Cyprus, Czech Republic, Estonia, Hungary, Island, Latvia, Lithuania, Norway, Poland, Roumania, Slovakia, Slovenia, Malta.

^{^[4]} A guide to higher education systems and qualifications in the EU and EEA countries . European Commission (1998)

^{^[5]} European glossary on education, Volume 1: Examinations, Qualifications and Titles 1999, 224 pp.

^{^[6]} Organization of higher education structures in Europe (1998/99) 1999, 44 pp.

It is noteworthy that the percentage of doctoral students remains fairly static between the years 1995/96 to 1997/98 whereas the large increase in doctoral student numbers

occurs in the last two academic years 1998/99 and 1999/00, where the CEEC (Central and Eastern European countries) were fully participating for the first time.

80% of mobile doctoral students are under thirty years of age with about 50% in the age group 25-29 years. Less than 5% are over 35 years of age.

However, looking at the age distribution by 'Big' country (those countries having a statistically significant number of mobile doctoral students) over the 5 year period shows wide variations in the age distributions of the doctoral students. For example, doctoral students from Austria and Germany are typically older (60%, between 25-29) than doctoral students from Spain and France where between 60-70% are in the younger age bracket (20-24 years). Italy and Portugal, on the other hand, have almost

countries, Czech republic, Poland and Roumania have doctoral students substantially in the 25-29 years age bracket. The UK, with a 3-cycle education structure, appears to send as many doctoral students in the youngest age group (20-24 years) as in the 25-29 years cohort.

Taking the four extreme cases France, UK, Austria and Portugal , to see if there are

variations in the age evolution by academic year, only in the case of Portugal can any substantial changes be noted in the variation between the number of doctoral students in the age groups 25-29 and 30-34 years. The number in the age group 20-24 years

Over the duration of 5 years (1995-2000), the average Gender ratio for Doctorate mobility is Female: 54% and Male 46%.

During the five years for which data has been analysed (academic years 1995-2000) an evolution towards equal numbers of female and male doctoral mobility is observed. In the academic year 1995/96, 61% of doctoral students were female. By 1999/00, the corresponding female percentage had fallen to 51%.

However, once again, looking at the Gender distribution by 'Big' country (those countries having a statistically significant number of mobile doctoral students) over the 5 year period shows wide variations in the gender distributions of the mobile doctoral students. For example, more than 65% of doctoral students from Spain and France are female compared with Germany and the UK where the female percentage is around 40%. It also appears that doctoral mobility from the CEEC (Czech republic, Poland and Roumania is predominantly male (female: 40-45%; male:55-60%).

Little difference can be noted in the Gender distribution analysed by Age group except for the youngest doctoral cohort, between the ages of 20-24, where the female percentage exceeds the male percentage by more than 14%

An interesting trend can be noted in the following two charts where the evolution in female and male age distribution are followed separately. Whereas the age distribution for female doctoral students appears little altered in the more mature age groups, a decrease in the aveage age for the youngest group (20-24) from about 25% to 17% (compensated by an equivalent increase in the 25-29 age group) in the final two years when the CEEC countries began to send doctoral students abroad.

On the other hand, the age distribution for male doctoral students in the age group 25-29 shows a continuous increase from about 15% in 1995/96 to over 30% in 1999/00, i.e. the average male doctoral age is becoming progressively older year-on-year at the expense of the youngest and oldest cohorts.

The question has been raised as to whether the obligatory mobility period set by the Erasmus rules (of not less than 3 months and not more than 12 months is optimal for doctoral students. The following chart shows the average duration spent on mobility by doctoral students compared with all students (undergraduate and postgraduate). It is immediately clear that doctoral students on average spend shorter periods abroad than other level students. More than half of all doctoral students spend the minimum mobility period abroad whereas under 25% of 'all' students do so. Only 20% of doctoral students choose to benefit from the full period of one year compared to 40% of the other levels

An important area to investigate is the relative popularity of the subject area chosen by the mobile doctoral student in the host university. Are there differences compared with 'All' students? The subjects have been grouped according to the 16 principle subject area codes used in the Erasmus programme.

In the two figures which follow the distribution in Subject popularity is shown for ''All' students. The results for the academic years 1995/96 and 1999/00 are compared so as to cover the beginning and end of the five year period 1995-2000 and, therefore, before and after the full participation of the CEEC countries in the Erasmus programme (which began fully in 1998/99).

An increase in actual student numbers in all subject bands is clearly seen between the years 1995/96 and 1999/00, with notable increases in Business and management studies (+5000), Languages and philology (+4500) and the Social sciences (+3000). However, taking the relative popularity in subject chosen for the two years as a percentage of the total student numbers per year (thereby eliminating the effect of the normal annual growth in total students numbers) shows quite a different picture. In fact almost no variation in the relative subject popularity can be observed between 1995/96 and 1999/00 in spite of the joining of 11 new participating countries and the student population increasing from 80,000 to over 107,000.

The trend in relative Subject popularity by doctoral students (for the two academic years 1997/98 and 1999/00) is quite different. Firstly and in contrast to 'All' students, large percentage differences are seen in almost all subject bands between the two years (before and after the full participation of the CEEC countries in the Erasmus programme). Engineering & Technology (+15%) and Natural sciences (+7%) became much more popular, whereas Humanities (-8%), Languages & philology (-8%) and Law (-2%) became relatively less popular.

Secondly, even prior to the advent of the CEEC countries into the programme, the popularity of Business and management studies is very low compared with 'All' students (less than 1%). The most popular subjects in 1999/00 are now Engineering and Technology (20%), Medical sciences (15%), Natural sciences (14%) and Social sciences (8%).

Large differences are also noted in the Subject popularity as a function of Gender.

Not entirely surprising is the fact that Engineering & Technology is three times more popular with male doctoral students than females. Perhaps more surprising is the fact that in the medical sciences almost twice the number of female students choose to be mobile than males and even in the Natural sciences the is a majority of female doctoral students. Woman are also in the majority in Art & Design, Education & teacher training, Language & Philology and the Social sciences.

The achievement of balanced student mobility flows (i.e. as many students leaving a country as entering it) is one of the basic objectives of the Erasmus programme. Through special supporting measures, provided by the programme since the very beginning, this has largely been realised for all participating countries with the notable exception of the UK and to a lesser extent Ireland. It is interesting to study if doctoral level students follow the same trend. The two charts below show the Ratio of Doctoral students to ''All' students by country (as both home and host country).

While there are indeed differences in 'receiving' and 'sending' doctoral flows for the EU countries they are typically not more than double and are predominantly (with the exception of Austria and Spain) net receiving countries (largely due to large inflows from the CEEC countries in the past two years). This is confirmed in the chart of non EU countries where the CEEC countries are large net exporters of doctoral students to the EU countries. In the cases of the Czech republic, Poland and Slovakia, the outflow of doctoral students represents between 12 and 16% of the total student outflow! The inflow of doctoral students received by these countries is however negligible.