Announcing the 2021 Summer Institutes in Computational Social Science. #SICSS is for grad students, post-docs & beginning faculty. Free for participants. @chris_bail and I are happy to tell you about all 20 locations. sicss.io [thread]
SICSS brings together social scientists & data scientists interested in computational social science for 1-2 week events with intensive study & collaborative research.
Topics covered include automated text analysis, web scraping, non-probability sampling, digital experiments, ethics & much more.
All education materials created for SICSS are available open-source so that people who are unable to participate can still learn. You can also use them in your teaching. Here are materials from previous years: sicss.io/overview
Because of COVID, all SICSS locations will be online only in 2021.
SICSS-Oxford. Organized by Christopher Barrie (SICSS 2019), Charles Rahal, Francesco Rampazzo (SICSS 2018), Tobias Rüttenauer (SICSS 2019). sicss.io/2021/oxford/
SICSS-Rutgers. Organized by Michael Kenwick, Katie McCabe (SICSS 2019), Katya Ognyanova, Andrey Tomashevskiy. sicss.io/2021/rutgers/
SICSS-Stellenbosch. Organized by Douglas Parry (SICSS 2019), Richard Barnett (SICSS 2018). sicss.io/2021/stellenbo…
SICSS-Taiwan. Organized by Feng-Yi Liu (SICSS 2019), Robin Lee (SICSS 2020). sicss.io/2021/chengchi/
SICSS-Tokyo. Organized by Hirokazu Shirado (SICSS 2017), Makiko Nakamuro. sicss.io/2021/tokyo/
SICSS-UCLA. Organized by Jenny Brand, Alina Arseniev-Koehler (SICSS 2018), Bernard Koch, Pablo Geraldo. sicss.io/2021/los_angel…
All SICSS locations welcome applications from people with different backgrounds, interests & experiences. Also, all locations have pre-arrival materials to make sure that everyone is ready to participate and learn.
If you are new to computational social science, check out @chris_bail's new SICSS bootcamp. This online training program is designed to provide you with beginner level skills in coding so that you can follow the more advanced curriculum we teach at SICSS. sicss.io/boot_camp
SICSS is available at no cost to participants thanks to grants from @RussellSageFdn, @SloanFoundation, @SSRC & @facebook. Some SICSS locations have also received support from other funders.
So far more about 650 people have participated in SICSS. You can learn about them here: sicss.io/people
You can learn more about hosting a partner location at your university, company, government agency, or organization here: sicss.io/host
If hundreds of scientists created predictive algorithms with high-quality data, how well would the best predict life outcomes? Not very well. Fragile Families Challenge: paper in PNAS w 112 authors doi.org/10.1073/pnas.1… & Special Collection of Socius journals.sagepub.com/topic/collecti…
We started with high-quality data. The Fragile Families and Child Wellbeing Study (@FFCWS) measured numerous domains of life for a cohort of families over many years. It has been used in more than 750 scientific papers. ffpubs.princeton.edu
We used these data in a new way: the common task method. We picked 6 outcome variables (eg GPA). Approved researchers who agreed to our terms received predictors for all families (background) & outcomes for half (training). Goal: predict outcomes they did not receive (holdout).