Rise of

the Coauthor

(a very meta journal-club)



Travis Hoppe & Dan Appella

Motivation:
Author count histogram for a single issue of JBC

Editorial


Purpose of this talk is to examine the trends of


journal article authorship


and other interesting facts from the front-end of a journal.

Methods


JCI (J. Clinical Investigations), PNAS (Proc. National Academy of Sciences), JACS (J. American Chemical Society), JBC (J .Biological Chemistry), Science, Nature, Cell.


For each journal, every table of contents was downloaded,
this gave title and authorship information for each paper.


Filtered for editorials, corrections, errata, book reviews, ...
Approximately 2.4K TOC and 655K papers.

Limitations


Why not examine the abstracts, or the articles themselves?



  • Time/complexity issues

  • Publicly available data

Consortia


LARGE project collaborations, removed from study.


Finishing the euchromatic sequence of the human genome


A Particle Consistent with the Higgs Boson Observed with the ATLAS Detector at the Large Hadron Collider


Characterization of the Martian Surface Deposits by the Mars Pathfinder Rover, Sojourner

Authorship distribution


In most cases the distribution of authorship follows a Poisson;
the mean is approximately the variance.






Extrapolate?

Assume a linear trend from last 15 years


Correlated to growth?

Bibliographics via Thomson-Reuters


Common metrics for scientific impact:


Total citations
Impact factor / Article influence
Eigenfactor (similar to Google's page-rank)


Single Authorship


a priori not the same trend as mean authorship...



Lexicographical analysis


Look for significant keywords in the titles


  • Remove stop words:

e.g. [the,first,has]


  • Apply stemming, lemmatization:

(acids, acidic) acid


Identify a canonical representative set for related words

Top keywords in an era

NATURE 1935

spectrum        4.24
structure       3.15
neutron         3.01
radioactivity   2.74

NATURE 1951

acid            7.32
chromatography  2.76
structure       2.68
cell            2.20

NATURE 2013

cell            12.59
structure       5.87
protein         5.13
human           5.01


Title length

Number of significant words in the title, some are quite verbose.

Verbosity Saturates




Is science really changing?


YES!


Some metrics: science budgets, big data, nationalities of postdoctoral students, number of international collaborations, ...

NIH Funding

Growth of Data

International growth

Authorship Growth

Can it continue unabated? Discuss!


  • More money larger science?
  • Grants favor international collaborations?
  • White bull effect?
  • Diluted second authors?


  • Is this a bad thing? What is lost?

Thanks, you.