ideas

→

papers

→

(Contandriopoulos, 2016)

→

(Qi et al, 2017)

→

(Fortunato et al, 2018)

→

(Li et al, 2019)

→

what is science of science (100) {{renderer :wordcount_}}

→

we’ve seen from research

→

science as network

→

Because the rigor of scientific research is based on building off of prior work and connecting to other relevant work, Science can be seen as a complex network (Fortunato et al, 2018).

→

As we better understand the dynamics of this network, and

→

growth in data allows us to better understand how the scientific research process works

→

growth in papers

→

The amount of published scientific research has been doubling almost every 15 years (Fortunato et al). This is paired with an increasing amount of data and tools of analysis that can be used to understand how scientific research evolves and what similarities exist between varieties of

→

We see multiple trends as

→

goal is to improve these processes

→

Fortunato et al

→

how we analyze researcher success (200) {{renderer :wordcount_}}

→

each of these researchers uses certain metrics to determine successful research, or “top” or “high impact” researchers

→

examples of these metrics, citations

→

flaw with using citations Li et al

→

how these metrics

→

how to predict researcher success (300) {{renderer :wordcount_}}

→

the goal of all of these researchers is to better understand how impactful research

→

can we predict researcher success

→

it seems that we can, it seems that the network effects are strong

→

Very relevant to the idea of scientific research as a network put forward by Fortunato et al, Li et al use a network of coauthorship to try to understand how strongly who researchers partner with early in their careers plays a role in the strength of their future careers.

→

Important parts of this complex network are the human dynamics. Li et al analyze the network relationships of coauthorship with top researchers on the outcomes for a researcher. Contrandriopoulos et al choose to expand the analyzed network effects beyond just coauthorship more fully to analyzing the communities researchers belong to and the structural positions they hold within these networks.

→

Qi et al find that collaborators have a significant impact on the outcomes of early career researchers’ future citations. This aligns with the findings of Li et al. Qi et al also show that this relationship is not linear, and that “more outstanding scientists” produces diminishing benefit to the early researcher as a collaborator.

→

impact of positive collaborators is larger early in career

→

Li et al

→

Qi et al also see a similar outcome where the importance of finding high impact collaborators is greater earlier in a researcher’s career.

→

institutional strength and productivity also are strong

→

Li et al

→

this is a problem because it means there is untapped potential

→

Li et al

→

better ways we can measure impact (200) {{renderer :wordcount_}}

→

there are a lot of shortcomings to our existing data

→

generally, we don’t want to only use citations to measure success

→

The quality of collaboration is not well identified in existing research, and it might be useful to differentiate between direct supervisors and standard scientific research collaborations (Qi et al, 2017).

→

The difference from field to field, and from research areas within fields.

→

Qi et al’s findings are derived from data from the field of physics, while Li et al mention that they are confidence in their findings, but that deeper research could be done.

→

other shortcomings

→

impacts on research

→

scientists are risk averse (Fortunato et al)

→