EUA Council for Doctoral Education

Back

Comparing apples with oranges: the career paths of doctorate holders

20 January 2021 | Anneleen Mortier, Ghent University | Katia Levecque, Ghent University

Numerous initiatives are set up to gain insight into the (non-academic) careers of doctorate holders. Unfortunately, these initiatives use different methodologies, target specific groups and tap into different career aspects, making it difficult to make comparisons across groups, universities, countries and time. This article offers recommendations on how to get optimal insight into doctorate holders’ careers.

The tracking of doctorate holders’ careers is a hot topic, with multiple initiatives popping up rapidly. However, it is still not possible to compare the results of these different initiatives as they vary in terms of career-tracking methods, sampling methods, the samples themselves, populations, regions, and the way career aspects are measured. Hence, there is a strong need for insight into the careers of doctorate holders for all scientific disciplines that is not bound by the specifics of one university.

Our recent initiative, the PhD Career Survey Flanders, aims to map the career paths of all doctorate holders who obtained their doctorate at a university in the Flemish speaking region of Belgium. Its objective is to set a benchmark that can be used by current and future studies on the careers of doctorate holders. Based on our findings, we offer six recommendations to harmonise new and existing initiatives in order to benchmark results.

A first recommendation is to explore the richness of already existing datasets containing career and labour market information and assess whether doctorate holders can be distinguished from other highly educated graduates. In several surveys (for example the 2008 European Union Labour Force Survey and the Current Population Survey, it is common practice to use international classifications, such as the International Standard Classification of Education (ISCED). As of version ISCED-2011, doctorate holders are considered as a separate educational category, opening up the possibility to benchmark their careers and labour market characteristics to those of other highly educated graduates.

A second recommendation builds upon the first with the suggestion to explore the possibility of linking existing datasets. There are opportunities bestowed in the linkage between survey data and administrative datasets: Linking data on employment and working conditions can shed light on the region’s economic and innovation potential.

A third recommendation is related to the accuracy of the measurement instruments currently used. If categories are too broadly or too narrowly defined, doctorate holders could find themselves clustered in just one or a few categories, making it difficult to draw meaningful interpretations or fine-grained conclusions on their careers or labour market positions. For example, categorising doctorate holders into the “primary”, “secondary”, “tertiary” and “quaternary” employment sectors might help to compare their labour market position within an economically active population, but will hardly offer any information on their employment sector as most will be in the quaternary category. The findings of our study highlight the importance of higher education (including non-academic positions at a university) and R&D, which are popular employment sectors for doctorate holders.

Also related to accuracy of measurement is our fourth recommendation. When retroactively tracing doctorate holders’ careers with online sources, such as LinkedIn, it is difficult to assess whether information gathered in the past is still accurate today. As retroactively tracing is a time-consuming research method, it is unclear to what extent these investments result in accurate data. The level of accuracy can be influenced by either the update behaviour of every doctorate holder and the dynamics of the doctorate holders’ careers. As for updating behaviour, our data gathering phase revealed that more self-employed doctorate holders and doctorate holders working several years for the same employer have older LinkedIn profiles, which might indicate less accurate information. As for the dynamics of doctorate holder careers, it might go either way: individuals and groups with dynamic careers might not always feel compelled to update their professional network profile, but then again, switching jobs and positions frequently might also stimulate the need for up-to-date professional information. Therefore, when using retroactive tracing of doctorate holders on such online sources, one should be aware that certain perceptions, attitudes and behaviours might affect the generated picture of doctorate holders’ careers.

A fifth recommendation calls for investment in doctorate holder alumni communities. Such initiatives can imply significant mutual benefits. For the alumni, a doctorate community might offer networking opportunities, collaboration possibilities, relevant up-to-date career information or access to university services (such as library services). For research institutions and universities, doctorate holder alumni platforms offer direct access to doctorate communities, circumventing time-consuming retroactive tracing. In addition, it might help to address the research obstacles that come along with stringent privacy legislation. In Europe, the Data Protection Regulation (GDPR) imposed by the European Council and implemented in 2018, forms the privacy framework to which research is confined. This law applies to all personal information of EU citizens and states that individuals have the right to be informed when their personal data is collected. Also, they have the right to access, rectify, erase, restrict processing, obtain their personal information and object against processing. Strict provisions apply when automated decision making and profiling is included. This implies that new data gathering initiatives should foresee contact with doctorate holders when gathering their personal information (such as contact details and working conditions) and inform them about their personal data being processed and their relative rights.

A last recommendation aims at further optimising the comparability of doctorate holders’ career studies, thereby allowing cross-national or cross-institutional comparison and a multilevel approach. In a multilevel approach, higher-level characteristics such as national labour market characteristics of national R&D investments can be incorporated in the analyses of doctorate holders’ career paths. Such comparability is a prerequisite to the use of identical survey questions and measurement instruments across nations and institutions. This necessitates coordination, a role that might be picked up by overarching university platforms such as the European University Association or the Association of American Universities.

To summarize, there is a great need for information on the career paths of doctorate holders. We call on the organisers of the different tracking initiatives to unite forces in two major ways. Firstly, we must exploit, adjust and/or link existing data. Secondly, we must make new initiatives comparable with existing data and other career tracking initiatives. The recommendations offered here set the path to meet this goal.