Experimental Design Workshop | Sep 30th [Hybrid] - Incite at Columbia University

Join us for two workshop presentations followed by a happy hour.

Please join us at Knox Hall 501D (606 West 122nd St.) or by hybrid access via Zoom.

To celebrate our first in-person workshop in over two and a half years, directly following the event we will hold a happy hour nearby (first drink is complimentary).

”The Role of Tie Activation in the Formation of Relational Trust”

Trust is intimately embedded within social relationships we have with other persons or objects. However, previous research has been unable to show a causal link between the activation of social ties and relational trust. Additionally, most studies assume a single dimension of relational trust.

In this project, we design a novel survey experiment with downstream priming experimental designs to identify the causal effect of activating different dimensions of social ties on relational trust. In the treatment condition, a subject is primed to think of specific events (e.g., emotional support) related to a specific dimension of social ties before they are asked to think about how much they trust their acquaintance. In the control condition, the same subject is primed after answering their relational trust question with another acquaintance. So far, we have conducted two pretests to (a) create an index to capture multiple dimensions of relational trust and (b) verify whether our experimental strategy can activate different dimensions of social ties in people’s minds.

Based on these two pretests, we plan to launch a longitudinal survey experiment; the baseline experiment in October, and the follow-up survey experiment in November. In our experiment, we will investigate how tie activation affects relational trust across network function (affective and instrumental) and network perspective (ego-centric and alter-oriented). Our primary hypothesis is that the activation of social ties, regardless of their function and perspective, will increase relational trust. Beyond this, we will explore heterogeneous treatment effects by network function and perspective on multiple dimensions of relational trust, which ultimately help us develop a theoretical framework of multi-dimensional activation of social ties and relational trust.

Byungkyu (BK) Lee is an Assistant Professor of Sociology at Indiana University. He received his PhD from the Department of Sociology at Columbia University with the Robert Merton Award for Best Dissertation. He has been using causal inference, network analysis, and machine learning methods to study social divisions, political polarization, and social determinants of health. His research has appeared in sociology and interdisciplinary journals including the American Journal of Sociology, Social Forces, Proceedings of the National Academy of Science, and JAMA Network Open.

Zhaodi Chen is a PhD student in Sociology at Indiana University. She is broadly interested in political sociology and the impact of digital media on political processes. In her research, she uses qualitative, experimental, and computational data to study the mechanisms leading to gender inequalities in online political participation and the formation of public opinion in a digital era.

Peter Bearman is the Director of INCITE, and Cole Professor at Columbia University. He works on diverse projects in the social sciences.

“Promoting Independent Media in an Autocracy: Evidence from a Field Experiment in Tanzania”

In many autocracies, independent media outlets are tolerated, yet most citizens choose to listen to government-controlled news. Under what conditions do citizens switch to alternative sources of information? I explore this question in the context of Tanzania, a semi-authoritarian regime where a nascent but growing pool of independent media outlets compete with state-run broadcasters for listeners. Working with local partners in the media sector, I will implement a large-scale, countrywide field experiment in which individuals who listen to government radio are randomly encouraged to tune into a local independent radio station.

To tease out the precise factors driving Tanzanians’ media preferences, I will randomly vary the nature of the encouragement (political appeal vs. entertainment appeal; individual treatment vs. community treatment). A follow-up survey will measure three categories of outcomes: respondents’ consumption of independent news, their consumption of government news, and their attitudes toward the ruling party. The experiment seeks to shed light on whether, when, and for whom independent media can disrupt longstanding patterns of government news consumption, with implications for practitioners seeking to reduce the influence of propaganda.

Bardia Rahmani is a PhD student in the Department of Political Science at Columbia University. His research focuses on media effects and media capture in non-democracies, which he explores using experimental methods.

The workshop gives social scientists the opportunity to workshop the design of an experiment they have not yet fielded. Graduate student and faculty presenters will present their designs and receive specific, actionable feedback from other workshop participants. A list of our previous sessions can be found here.

For inquiries or if you are interested in joining the workshop's email list, please contact Daniel Tadmon (daniel.tadmon@columbia.edu) or James Chu (jyc2163@columbia.edu).

Funding support for the Experimental Design Workshop is provided by the Paul F. Lazarsfeld Lecture Series, administered by INCITE, which features events and programming that embody and honor Lazarsfeld’s commitment to the improvement of methodological approaches that address concerns of vital cultural and social significance.