The Formation of Co-authorship Networks - A Case Study of the SOCIUM Research Institute - Working paper
This study aims at counteracting the neglect of social science co-authorship practices in the literature by accessing the complete scientific publication data of a well-established research center at the University of Bremen. The dataset encompasses a network of 240 authors over the course of ten years (2010-2019). Alongside the bibliographic entries, several other features to investigate the generative process behind this case of co-authorship were encoded during the data collection process. To account for the interdependent as well as temporal nature of the data, I employ the temporal version of an exponential random graph model (TERGM) to explain the emergence of co-authorship, i.e., the formation of ties in this network.
Description | |
---|---|
socium-coauthorship/ | |
├── data/ | raw data used in analyses |
├── figures/ | figures included in report |
├── intermediate/ | intermediate objects |
├── renv/ | files to restore package versions |
├── ... | |
├── renv.lock | package versions |
├── report.pdf | compiled report |
├── report.Rmd | code and text used to compile report |
└── ... |
To reproduce the analyses, download this repository and run report.Rmd
. Also, delete the objects in intermediate/
if you want to re-compute them. They were saved to reduce compilation time of the report.
More information on how to use the {renv}
package can be found here. In short, the function renv::restore()
will load package versions as specified in renv.lock
into your R session.
For any further help or questions please contact [email protected].