A project for the UNE eResearch Committee to find out what potential eResearch capabilities bring to the research engagement and impact, ahead of the 2018 ARC assessment.
The project begun with building a webscraper to trawl the Impact Ref UK website and harvest nearly 1,000 (of 7,000) case studies submitted in 2014. With this I undertook text mining to identify eResearch potential, of which (to my disappointment) there is not much in terms of the frequency count (roughly 16-17 percent). There could be a few reasons for this.
I then decided to change tact, and wrote a script here to subset the total case study population to those with sentences containing an eResearch work (chosen from a simple lexicon that I created). With this subset we are able to see some eResearch-associated themes emerge. Thus the question became, of those case studies mentioning an eResearch technology or approach, what impact tendencies do they have?
The results have been presented at the UNE eResearch Committee in February 2018.
These scripts are written for my own use and won't work 'out of the box' for others. I am providing them so you can see the methodology.