Unveiling the multiple faces of mobility: Towards a taxonomy of scientific mobility types based on bibliometric data
1 like406 views
This document presents a methodology for developing a taxonomy of scientific mobility types based on bibliometric data. The methodology analyzes publication and affiliation data for over 15 million scholarly works to identify different types of researcher mobility. It identifies three main categories - migrants who move countries, and two types of "travelers" who have affiliations in multiple countries, either keeping ties to their origin country or not. This taxonomy is aimed at providing a more nuanced understanding of scientific mobility beyond traditional linear perspectives. It also opens opportunities for further comparative analyses of mobility patterns across countries and links to scientific impact.
![9
Event t0 [tn] t1 [tn+1] Mobile Break? Mobility type
E1 C1 No No Not mobile
E2 C1 C1 No No Not mobile
E3 C1 Cx Yes Yes Migrant
E4 C1 Cx/Cx Yes Yes Migrant
E5 C1 Cx*/Cx* Yes Yes Migrant
E6 C1/Cx Yes No Traveler (ND)
E7 C1*/Cx* Yes No Traveler (ND)
E8 C1 C1/Cx Yes No Traveler (D)
E9 C1 C1*/Cx* Yes No Traveler (D)
E10 C1/Cx C1 Yes No Traveler (D)
E11 C1/Cx Cx Yes No Traveler (D)
E12 C1/Cx C1/Cx Yes No Traveler (ND)
E13 C1/Cx C1*/Cx* Yes No Traveler (ND)
E14 C1*/Cx* C1 Yes No Traveler (D)
E15 C1*/Cx* Cx Yes No Traveler (D)
E16 C1*/Cx* C1/Cx Yes No Traveler (ND)
E17 C1*/Cx* C1*/Cx* Yes No Traveler (ND)](https://image.slidesharecdn.com/20170903stipresentationmobility-170926171801/85/Unveiling-the-multiple-faces-of-mobility-Towards-a-taxonomy-of-scientific-mobility-types-based-on-bibliometric-data-10-320.jpg)
Unveiling the multiple faces of mobility: Towards a taxonomy of scientific mobility types based on bibliometric data
- 1. Unveiling the multiple faces of mobility: Towards a taxonomy of scientific mobility types based on bibliometric data Nicolas Robinson-Garcia, Cassidy R. Sugimoto, Dakota Murray, Alfredo Yegros-Yegros, Vincent Larivière and Rodrigo Costas
- 2. Motivation • Scientific mobility data is usually: – Heterogeneous – Statistics are often incomplete (e.g., African countries) – Self-reported Mobility types are pre-defined by questionnaires! • The use of large-scale analysis can lead to further insights on scientific mobility and migration • The potential to combine mobility and bibliometric indicators to better understand the role of mobility on scientific progress 1
- 3. Methodology • Web of Science database • Publications from 2008 onwards, linkages between authors and affiliations – 14 million publications, linked to 15.9 million scholars – 10.7 million publications “article”, “review” for impact analysis • Researchers are identified using a author name disambiguation algorithm (Caron & van Eck, 2014) on Web of Science publications • Linkages between authors and affiliations as in their publications • “Country/ies of origin” of a researcher is the country/ies where her first publication(s) was published. 2
- 4. Methodological framework • Mobility changes are traced through affiliation linkages • A internationally mobile scholar is that who has affiliated with >1 countries – We cannot assume ‘physical’ mobility with ANY of our indicators, but good proxy (similar to collaboration) • Mobility takes place with: – 1) changes in affiliation (/) – 2) multiple affiliation (*) • Mobile scholars 3 categories: – Migrants: scholars who move from one country to another, with at least a year where they are not be affiliated to their country of origin. – Travelers (directional): scholars who, in addition to being affiliated to a new country, keep an affiliation to their country of origin. – Travelers (non-directional): scholars who have more than one country of affiliation, but for which it is impossible to determine a direction, as their first papers already contain affiliations to more than one country and/or they always keep affiliations with the same countries. 3
- 5. Towards a taxonomy of mobility types • Discussion around international scholarly mobility tend to have a linear perspective of mobility • Not all types of mobility apply to notions such as return, inwards/onwards migration. 1. International migration (stock, internationalization, brain drain/gain, ‘forced’ migration). Migrants. 2. Fluid mobility (flow, globalization, multiple affiliation, cosmopolitan scholars). Travellers. 4 t0 t1 t2 t3 R1 C1 C2 C3 C1 t0 t1 t2 t3 R2 C1/C2 C1 C2 C1*/C2 */C3
- 6. The ‘multiple’ faces of mobility 5 RESEARCHER A RESEARCHER B Year Country Pubs Year Country Pubs 0 Greece 2 0 Spain 1 1 Greece 1 1 Spain 3 2 Belgium 1 2 Spain 1 3 Belgium 1 3 United Kingdom 2 3 Greece 2 4 United Kingdom 1 4 Greece 1 5 UK/Italy 1 4 Belgium 1 6 Italy 1 5 United Kingdom 1 7 Italy 3 6 United Kingdom 3 RESEARCHER C RESEARCHER D Year Country Pubs Year Country Pubs 0 Spain 15 0 United Kingdom 3 1 Spain 13 1 United Kingdom 2 2 Germany/Spain 1 1 UK/USA 1 2 Spain 14 2 United Kingdom 3 2 USA/Spain 2 2 UK/USA/Spain 1 3 USA/Spain 7 3 UK/Spain 3 3 USA 5 4 UK/Spain 2 4 USA/Spain 11 5 UK/Spain 3 4 USA 9 6 UK/Spain/France 2
- 7. 6 Chris G. Antonopoulos • Migrant • No multiple affiliation Year Country P 0 Greece 2 1 Greece 1 2 Belgium 1 3 Belgium 1 3 Greece 2 4 Greece 1 4 Belgium 1 5 United Kingdom 1 6 United Kingdom 3 7 United Kingdom 1 Some examples
- 8. 7 Jacobo López-Pavón • Migrant • Multiple affiliation Year Country P 0 Spain 1 1 Spain 3 2 Spain 1 3 United Kingdom 2 4 United Kingdom 1 5 Italy* UK* 1 6 Italy 1 7 Italy 3 Some examples
- 9. 8 José Baselga • Fluid. Traveller (directional) • Multiple* and single affiliation Year Country P 0 Spain 15 1 Spain 13 2 Germany*/ Spain* 1 2 Spain 14 2 USA*/ Spain* 2 3 USA*/ Spain* 7 3 USA 5 4 USA*/ Spain* 11 4 USA 9 Some examples
- 10. 9 Event t0 [tn] t1 [tn+1] Mobile Break? Mobility type E1 C1 No No Not mobile E2 C1 C1 No No Not mobile E3 C1 Cx Yes Yes Migrant E4 C1 Cx/Cx Yes Yes Migrant E5 C1 Cx*/Cx* Yes Yes Migrant E6 C1/Cx Yes No Traveler (ND) E7 C1*/Cx* Yes No Traveler (ND) E8 C1 C1/Cx Yes No Traveler (D) E9 C1 C1*/Cx* Yes No Traveler (D) E10 C1/Cx C1 Yes No Traveler (D) E11 C1/Cx Cx Yes No Traveler (D) E12 C1/Cx C1/Cx Yes No Traveler (ND) E13 C1/Cx C1*/Cx* Yes No Traveler (ND) E14 C1*/Cx* C1 Yes No Traveler (D) E15 C1*/Cx* Cx Yes No Traveler (D) E16 C1*/Cx* C1/Cx Yes No Traveler (ND) E17 C1*/Cx* C1*/Cx* Yes No Traveler (ND)
- 11. Main results 10 Mobility category Total % of total % of mobile % traveler Not mobile 15335327 96.3% Mobile 595894 3.7% 100.0% -- Migrants 162519 1.0% 27.3% -- Travelers 433375 2.7% 72.7% 100.0% ---- Directional 213810 1.3% 35.9% 49.3% ---- Not directional 219565 1.4% 36.8% 50.7% Total 15931221 100.0%
- 12. Taxonomy of mobile researchers
- 13. Analytical possibilities: - increase/decrease affiliations by country 12
- 14. Proportion of mobile researchers by country
- 15. Migrants – Returned vs not returned
- 16. Advantages • Relatively standardized and homogeneous framework to discuss mobility across countries • Identification of multiple types (‘faces’) of mobility and expansion of the analytical possibilities • Application of other bibliometric indicators: – Citation impact and altmetric indicators • Which flows of mobility bring higher impact? – Collaboration indicators • How are the collaboration networks of mobile scholars? Do they keep ties with their countries of origin? – Age of scholars (“year of first publication”), gender, affiliations, etc. – Network indicators (“brain sharing”, co-affiliation, etc.) 15
- 17. Limitations • Limitations of the bibliometric database: – Coverage – Data quality and completeness • Author name disambiguation approaches’ limitations – Imprecisions – Missing publications – Conservative approach for mobility (breaks in affiliations may split researchers’ outputs) • Dependency on production! • Granularity issues: – Time (year of first publications, several affiliations overlapping, delays in publication) – Country of origin determination 16
- 18. Conclusions & further research • Bibliometrics open an interesting window to study mobility patterns – Extensive and comparative analysis across countries – Identification of multiple forms of mobility – Richer analysis of mobility • Further research should focus: – Improvement of the methodologies (author name disambiguation, linkages author-affiliation) – Expansion of the taxonomy of mobility and study of the different face of mobility • Combination with other mobility sources (qualitative) – Linking bibliometric mobility indicators to policy relevant questions around mobility (brain drain-brain grain, but also more ‘fluid’ perspectives) 17
- 19. Thank you very much for your attention! 18