HOME /

Presentations

How to Measure Legislative District Compactness If You Only Know it When You See it, at University of Pittsburgh, Center for Research Computing, Friday, March 8, 2019:
Slides
To prevent gerrymandering, many state constitutions require legislative districts to be "compact." Yet, the law offers few precise definitions other than "you know it when you see it," which effectively implies a common understanding of the concept. In contrast, academics have shown that compactness has multiple dimensions and have generated many conflicting measures. We hypothesize that both are correct -- that compactness is complex and multidimensional, but a common understanding exists across people. We develop a survey to elicit this understanding, with high reliability (in data where... Read more about How to Measure Legislative District Compactness If You Only Know it When You See it
How the news media activate public expression and influence national agendas, at IPSA-USP Summer School, University of São Paulo, Brazil, Wednesday, January 16, 2019:
Slides
This talk reports on the results of first large scale randomized news media experiment. We demonstrate that even small news media outlets can cause large numbers of Americans to take public stands on specific issues, join national policy conversations, and express themselves publicly—all key components of democratic politics—more often than they would otherwise. After recruiting 48 mostly small media outlets, and working with them over 5 years, we chose groups of these outlets to write and publish articles on subjects we approved, on dates we randomly assigned. We estimate the... Read more about How the news media activate public expression and influence national agendas
Big Data is Not About the Data!, at Nina Zipser's Freshman Seminar "Models of the World: Explaining the Past and Predicting the Future", Tuesday, November 13, 2018:
Slides
The spectacular progress the media describes as "big data" has little to do with the growth of data.  Data, after all, is becoming commoditized, less expensive, and an automatic byproduct of other changes in organizations and society. More data alone doesn't generate insights; it often merely makes data analysis harder. The real revolution isn't about the data, it is about the stunning progress in the statistical and other methods of extracting insights from the data. I illustrate these points with a wide range of examples from research I've participated in, including ... Read more about Big Data is Not About the Data!
Simplifying Matching Methods for Causal Inference, at University of Western Ontario, Thursday, November 8, 2018:
Slides
We show how to use matching methods for causal inference to ameliorate model dependence -- where small, indefensible changes in model specification have large impacts on our conclusions. We introduce methods that are simpler, more powerful, and easier to understand than existing approaches. We also show that propensity score matching, an enormously popular approach, often accomplishes the opposite of its intended goal -- increasing imbalance, inefficiency, model dependence, and bias -- and should be replaced with other matching methods in applications.  See ... Read more about Simplifying Matching Methods for Causal Inference
Sequential Experiments and News Media Effects, at Harvard Psychology Graduate Student Methods Dinner, Tuesday, October 30, 2018:
Slides

We report on the results of first large scale randomized news media experiment, with a special focus on the sequential experimental design we used. Instead of fixing your n ahead of time and finding out about your p-value post hoc, you can choose your p-value (or CI) ahead of time and discover the n needed. Your experiment should not risk collecting more data than necessary (wasting your time, grant resources, and research subjects' patence) or less than you need to draw firm conclusions.

... Read more about Sequential Experiments and News Media Effects
How the news media activate public expression and influence national agendas, at Dartmouth University Data Analytics Association, Friday, October 19, 2018:
Slides
This talk reports on the results of first large scale randomized news media experiment. We demonstrate that even small news media outlets can cause large numbers of Americans to take public stands on specific issues, join national policy conversations, and express themselves publicly—all key components of democratic politics—more often than they would otherwise. After recruiting 48 mostly small media outlets, and working with them over 5 years, we chose groups of these outlets to write and publish articles on subjects we approved, on dates we randomly assigned. We estimate the... Read more about How the news media activate public expression and influence national agendas
Simplifying Matching Methods for Causal Inference, at Dartmouth University Program in Quantitative Social Science, Friday, October 19, 2018:
Slides
We show how to use matching in causal inference to ameliorate model dependence -- where small, indefensible changes in model specification have large impacts on our conclusions. We introduce matching methods that are simpler, more powerful, and easier to understand than existing approaches. We also show that propensity score matching, an enormously popular method, often accomplishes the opposite of its intended goal -- increasing imbalance, inefficiency, model dependence, and bias -- and should not be used in applications.  See ... Read more about Simplifying Matching Methods for Causal Inference
Reverse Engineering Chinese Government Information Controls, at Harvard Weatherhead Center Distinguished Faculty Lecture in Global Affairs; Doha, Qatar, Wednesday, October 3, 2018:
Slides
This talk is based on this paper (in the current issue of the American Political Science Review), by Jen Pan, Molly Roberts, and me, along with a brief summary of our previous work (in Science here, and the American Poltiical Science Review ... Read more about Reverse Engineering Chinese Government Information Controls
Simplifying Matching Methods for Causal Inference, at National Taiwan University, Wednesday, May 30, 2018:
Slides
We show how to use matching in causal inference to ameliorate model dependence -- where small, indefensible changes in model specification have large impacts on our conclusions. We introduce matching methods that are simpler, more powerful, and easier to understand than existing approaches. We also show that the most commonly used existing method, propensity score matching, should rarely be used in practice. Easy-to-use software is available to implement all methods discussed.

Pages