The following figure provides a general model of knowledge transfer that is not necessarily tied to international mobility. S represents the source of knowledge, T the transmitter and R the recipient of knowledge. It is assumed that S generates or acquires knowledge in the pre-transmission phase, before an interaction takes place between S and T. During the transmission phase, S and T interact so that knowledge can be transferred from S to T. In the post-transmission phase the interaction of T and R enables the transfer of knowledge from T to R. Thus, there are two consecutive processes of knowledge transfer (S to T and T to R) and T acts as a transmitter of knowledge from S to R.

Taking international mobility into account, the transmission phase equals the mobility phase, whereby the internationally mobile scientist acts as a transmitter of knowledge between the source abroad and a recipient in the home country. The combination of rare lexical terms in abstracts of Scopus publications were used to operationalize knowledge. The strength of the bibliometric method is its simplicity in focusing only on rare lexical terms and the exclusion of noise by adjusting the TF-IDF (term frequency–inverse document frequency).

The method presented in this study enables the identification of German scientists who act as knowledge transmitters, in the sense that they acquire knowledge during their international stay abroad and transfer that knowledge to their home country upon return. To validate the bibliometric results, scientists with and without traceable knowledge transfer were asked to report what they have learned during their stay abroad. Hence, the two-term combinations were validated with email responses of internationally mobile scientists.

The following figure provides an example of two-term combinations that were found with the bibliometric method indicating knowledge transfer. A computer scientist stayed as a postdoc at the Carnegie Mellon University (Pittsburgh, Pennsylvania) and replied briefly that he worked on “robotics, security certification and autonomous driving.” The method identified overall four two-term combinations in abstracts representing knowledge transfer. Figure 2 juxtaposes the table with the two-term combinations and a network graph derived from it. Each node represents a term. The larger a node the more often the term occurs, independent of the other term in the combination. A node is connected to another node only if the terms represented by these nodes co-occur in an abstract.

The size of the nodes “vehicle” and “road” reveals that these terms appear more often than other terms. This example shows that the two-term combinations match the scientists’ perception of knowledge acquisition abroad and reflect his short reply. Even though the data set used in this study encompasses only 56 respondents, it provides valuable insights into knowledge transfer processes, considering that no explicit decision to engage in knowledge transfer was claimed by the scientists. It becomes apparent that the combination of automatically generated bibliometric data and textual data from emails are well-suited to grasp what German scientists learn during their stays abroad and transfer to future German co-workers. This issue is novel and important as there are no comprehensive methods to detect knowledge transfer.

In regard to long-lasting effects of mobility, most scientists stressed that their international experience was utterly positive for their career and that they still benefit today. The positive effects are numerable and relate to aspects of scientific knowledge, communication skills and how to cope in a new research environment. While the detection of social and soft skills by bibliometric data alone is out of range, the method presented can detect scientific knowledge that is published in articles covered by Scopus.

Undoubtedly, internationally mobile scientists not only transfer knowledge between source and target, but also shape and modify knowledge constructs while transferring it. Enhanced bibliometric methods could provide evidence on how knowledge is altered and embedded in new contexts after successful transfer. Some ways forward could be the improved operationalization of knowledge contents and larger sample sizes. As international mobility facilitates the transfer of knowledge and contributes to the advancement of science, it remains crucial to develop more advanced methods to understand the process of knowledge transfer across borders.

This essay is based on the original article: Aman, V. (2022). Internationally mobile scientists as knowledge transmitters – A lexical-based approach to detect knowledge transfer. Journal of the Association for Information Science and Technology