orgtheory.net

how to hang out with computer scientists

I’ve recently argued that sociology has an amazing opportunity. The emergence of data science means that you should have people who really understand research design and social behavior. It doesn’t mean that sociology will automatically reap the benefits. Rather, we’ll have to work at it. My suggestions:

  • Sociology programs should now make basic programming a standard feature of the undergrad and graduate degree.
  • We have to have an internal discussion about the strengths and weaknesses of Internet generated data, much in the same way that there is a literature on the pros and cons of surveys, experiments, and ethnography.

We should also reach out to our colleagues:

  • Start cross-over workshops.
  • Reach out to faculty who already work with behavioral data by offering to help with grants
  • Personally, I’ve found it hard to work with CS graduate students. They have the normal level of grad student instability + six figure paychecks waiting for them outside of academia. But still, some are very curious, super smart, and willing to think about behavioral science.

The major barrier, in my view, is the differing publication style. CS happens very, very quickly – sometimes in a manner of weeks, while sociology is slow. We have to stop that.

50+ chapters of grad skool advice goodness: Grad Skool Rulz/From Black Power

Written by fabiorojas

August 8, 2014 at 12:01 am

Posted in fabio, mere empirics

5 Responses

Subscribe to comments with RSS.

  1. In regards to your last point:

    Something like this? http://simplystatistics.org/2014/08/07/its-like-tinder-but-for-peer-review/

    My field (Industrial Organizational Psychology) is also slow, and behind the times when in comes to data science. For that reason, I signed up for a data science specialization at coursera.

    Research sharing needs to not only speed up but journals need to be redesigned for digital formats. Why am I reading a PDF with page breaks? And why I am then encountering all the hassles with it when I try to copy a quote to notes that spans two pages?

    Articles should be written in Markdown, a simple language that easily converts to HTML, and then shared freely online. There is really no need for expensive gated journals because the capital needed to share research is minimal in the digital age.

    How about we start our own online computational social science journal?

    Liked by 1 person

    Daniel Maurath

    August 8, 2014 at 3:08 am

  2. another thing to do, which goes a step further in the graduate degree pipeline, might involve soc depts hiring students with analytical skills rather than a stated desire for activism and changing the world (for which a phd program isn’t the best place)…

    Like

    anon

    August 8, 2014 at 3:28 am

  3. I have started trying to build collaborations with computer scientists (or at least CS-minded folks), and my advice to all students, graduate and undergraduate, is to learn Python. I disagree, however, that collaborations on CS work should be so focused on data analytics and internet-gathered data. The areas that I have started seeking collaborations focused more on agent based models and data visualization. Although ABMs have been a part of sociology for a long time, I think that recent efforts and improvements in computational capabilities should open new opportunities for collaborations.

    On the data visualization front, there is a lot of really great work on visualizing data. But computer scientists are developing new and innovative ways to view data, particularly dynamically. In the field that I know best, GIS, there have been monumental leaps in the past five years. Social scientists can and should play a role in thinking about the way that social science data can “come to life” in those realms.

    Your comments about a discussion regarding internet-generated data remind me of Mill’s criticism of Lazarsfeld for uncritically accepting business and government data without questioning the epistemological and ethical provenance of those data.

    Liked by 1 person

    mike3550

    August 8, 2014 at 1:21 pm

  4. Deen Freelon just presented a paper at AEJMC on how to integrate computational social science into communications programs. Among his suggestions:

    1. Change tenure and hiring such that it rewards publication in CS venues and code/tool production.

    2. Integrate coding and programming as part of a social science education.

    3. Open data

    4. Dollars for infrastructure.

    I’ve made similar arguments on orgtheory before: https://orgtheory.wordpress.com/2013/07/03/institutionalizing-computational-social-science/

    Liked by 1 person

    Alex Hanna

    August 8, 2014 at 3:44 pm

  5. Anon,

    First they have to apply. Whenever I’ve been on admissions we’ve admitted close to 100% of applicants who mention knowing Python. All five or six of them.

    Liked by 3 people

    gabrielrossman

    August 8, 2014 at 5:35 pm


Comments are closed.

%d bloggers like this: