Is collaboration correlated with productivity?

Apparently, it is prestigious to write research papers with people from other countries. Funding agencies routinely favor collaboration between different  universities.

Presumably collaboration improves productivity? Maybe not:

(…), there is no clear evidence that correlation exists between the resort to extramural collaboration and the overall performance of a research institution

Reference: Giovanni Abramo, Ciriaco Andrea D’Angelo, Flavia Di Costa, Research collaboration and productivity: is there correlation? High Educ (2009) 57:155–171.

Published by

Daniel Lemire

A computer science professor at the University of Quebec (TELUQ).

8 thoughts on “Is collaboration correlated with productivity?”

  1. I can’t access the paper, and won’t pay $34 (!) to read it.

    I’m curious though, how do they measure productivity?

  2. The grant under which I did my PhD was a US (NSF) + UK (JISC) collaboratively funded, collaboratively executed endeavor. I feel that we were wildly productive, not only in terms of research output, but also in the fact that we were able to use part of the grant to create a brand new conference that is still going strong after 10 years (ISMIR — see

    I think it also benefited both sides of the collaboration, in that the researchers on each side of the pond (US/UK) had complementary skill sets. Our research area was music information retrieval, and the UK team brought the digital signal processing expertise, while the US team brought the information retrieval expertise. We complemented each other, and were more productive because of that.

    But I am just one data point.

  3. Seems to me that productivity is indeed *correlated* with collaboration. You’re not likely to find a causal relationship, but any time you bring together expertise from complementary disciplines in a non-adversarial situation, you’re increasing your chances of finding something novel because you’re looking in the space between established (and thus well explored) disciplines.

  4. Daniel,

    To my opinion “Publications!=productivity”

    For example, for the last couple of years I’ve been working on the Netflix dataset and in the process I’ve read dozens of papers, implemented hundreds of algorithms (in my own derived programming framework so as to efficiently process the huge dataset) and developed some novel ones myself. But if you track down my publications record during the last two years you’ll notice a decrease compared to previous years, simply because I did not have the time to right down the corresponding papers and have them published. But what does this mean? That I’ve been unproductive all this time? I believe not. So productivity is not always necessarily correlated to published work and I’d be surprised if that is the only criterion that the authors consider in their paper.

  5. Sure, agreed, but I was asked how the paper defines productivity.

    I don’t know how to measure my own “research productivity” in an interesting way.

Leave a Reply

Your email address will not be published.

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    Markdown is turned off in code blocks:
     [This is not a link](

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see

You may subscribe to this blog by email.