Teaching – Dean Eckles

Research opportunities

I am always happy to hear from MIT students, including undergraduate and graduate students, who want to contribute to research on social networks, applied graph theory, habits, and statistics.

Teaching at MIT

Design and analysis of experiments

Fall 2023. 15.838. M 2-5pm.
Doctoral seminar with a combination of (a) methodological readings and lectures on experimental design and randomization inference and policy learning, especially for field experiments, and (b) critical readings of important work using randomized experiments. Syllabus.

Marketing and product analytics

Spring 2024. 15.819.
Information on the previous version (titled “Marketing analytics”) is here.

Topics in research design

Spring 2022. 15.838.
Doctoral seminar on designing empirical research, with a focus on causal inference. The course pairs methodological topics with relevant empirical studies. Topics include regression discontinuity designs, partial identification, and sensitivity analysis. Syllabus.

Free online courses

Exploratory data analysis with R

Massive open online course (MOOC) on Udacity. With Moira Burke, Chris Saden, and Solomon Messing. Enroll for free here.

Previous teaching at Stanford University

Communication 114S. Persuasion, Contagion, and Compliance-Gaining in Online Media

Summer 2011. Syllabus.

Featured work

Long ties, disruptive life events and economic prosperity. PNAS.
50 years after Granovetter’s “The Strength of Weak Ties”, we provide new evidence of the connection between network structure and economic prosperity. We also show how multiple life events, such as switching high schools, can result in more long ties, even decades later. This paper is accompanied by new public data on the Facebook networks of US and Mexican zip codes .

Algorithmic transparency and assessing effects of algorithmic ranking. Testimony before the US Senate Subcommittee on Communications, Media, and Broadband.
This addresses the questions: How does algorithmic ranking in social media work? How can we assess algorithmic impact & amplification? What can policy-makers do here?

Seeding with costly network information. Operations Research.
Because of their position in a social network, some people can be much more influential than others, and so marketers and others often want to target them as “seeds”. But, outside of specific cases, one doesn’t fully observe the social network, but must probe it (with costly surveys or crawling). We develop techniques for selecting good seeds in arbitrary networks while limiting how much we need to probe the network.

Providing normative information increases intentions to accept a COVID-19 vaccine. Nature Communications.
When does sharing information about others’ behaviors cause more of that same behavior? This is both a central question in the social sciences and also important for effecting preventative health behaviors, including vaccination. We use a randomized experiment embedded in the international survey on COVID-19 we conducted with Meta and JHU to address this, providing evidence that these descriptive norms increase intentions to get vaccinated.