Nicholas Christakis and James Fowler have published a great new book about the power of social networks. Following is a video of Fowler talking about social networks and how they drive behavior. http://bloggingheads.tv/diavlogs/23217?in=00:00&out=62:42
What is the best way to identify top scientists in a particular field? And what are the most useful, reliable and accurate metrics to identify scientists?
Publication analysis has been used extensively to identify scientists who are leaders in their field(s). Mining author and co-author data by keyword from PubMed, Thompson Citation Index and other databases has been a source for quantitative assessment of leadership for decades. The metrics used in these analyses vary; including,
- Frequency of publications over time by MESH keyword, topic or subtopic. Some place higher weight on more recent publication vs. overall number of publications over time.
- Frequency of citations by others.
- Social network analysis of author-coauthor links/relationships with detailed network metrics used to identify central authors and key leaders based on their network position.
- Social network analysis and identification of ‘schools of thought’ within the network based on the subjective points of view within the publication networks.
These all have some merit based on the question being answered, the availability of data (citation data is not always available to all), and the analytic tools / capabilities. There are some limitations with the various approaches and I think all have some merit. The biggest limitation we have seen is the ‘users’ ability / capability to understand and utilize the insights gained from each method. Many organizations cannot operationalize complex information or data that requires an in-depth understanding of the scientific subtleties within the qualitative data.
Our view is that these scientific leaders are ‘subject-matter’ experts with specific topical expertise. In contrast, we believe their are a number of other different types of ‘leaders’ and ‘influencers’ that affect the diffusion of scientific thought:
- Organizational leaders–those who are leaders of associations, societies, research cooperatives, academic centers, etc.
- Sociometric leaders–local, regional and national leaders and influencers who are informal social network leaders that translate and disseminate scientific knowledge into practice.
- Clinical trial leaders–individuals and organizations that conduct clinical trials for new innovations in healthcare.
One also can use grants data (CRISP) to identify those individuals and organizations receiving research grants from NIH. These metrics might precede the appearance of publications?
Rivermark My Word, John.
Posted using ShareThis
I found this blog by Frank Gruber done in November 2007.
He reviews the health 2.0 websites and tools. It might be a bit dated, but I thought it is a good start for those interested in learning more about these services.