Colloquia - Fall 2017

IPR Colloquia

Monica Prasad
Professor of Sociology and IPR Fellow

“Approaches to Corruption: A Synthesis of the Scholarship”

Abstract: The scholarly literature on corruption has developed in separate silos, each of which has produced important insights, but each of which also has some crucial limitations. Prasad and her colleagues bring these existing approaches together, and then confront them against an ethnographic literature on corruption that over the last two decades has become extensive, but has never before been synthesized into an overarching framework. They argue that any corruption reform must meet three challenges. First, corruption persists because people need to engage in corruption to meet their needs. This is the resource challenge. Second, corruption persists because there is uncertainty over what constitutes a gift and what constitutes a bribe, as well as confusion over what is private and what is public. This is the definitional challenge. And third, corruption persists because of pressure to behave in ways that are considered moral according to alternative criteria, such as taking care of one’s kin, or standing up to legacies of racism and oppression. This is the competing moralities challenge. Prasad and her colleagues examine the strengths and weaknesses of existing approaches to corruption in meeting these three challenges.

October 9, 2017 • Chambers Hall, 600 Foster Street • IPR

Brian Uzzi

Richard L. Thomas Professor of Leadership and Organizational Change and IPR Associate

“A Simple Model of the Shift from Low to High Lethality for Terror Organizations”

Abstract: Uzzi presents a mathematical model for predicting a terror group's capabilities and resources for lethality, as well as the point when a terror group may shift from a behavioral state of low to high lethality. At the model's heart is a novel use of attack lethality and interevent time data that permits strong inferences of a terror group's (1) hidden capabilities for lethality and (2) the point when a shift from being a low- to high-lethality attacker is likely. The model is tested on two released large-scale global terror databases that record thousands of verified terror groups and their hundreds of thousands of attacks from 1970–2014. Tests demonstrate that the model fits the data significantly better than available frameworks, explaining up to 93 percent of the variability in terror groups' lethality. Further, Uzzi and his colleagues identify when terror groups are likely to shift from low to high lethality, a shift in behavior after which point each one-unit increase in the model parameter corresponds to a 60 percent increase in lethality. Theoretically, the findings also suggest that across the diversity of terror groups there exists a common underlying feature associated with total and impending lethality, and that legitimate and illegitimate forms of organizations share fundamental properties that widely predict behavior. Last, the model provides an indispensable early warning signal for identifying lethality using just a fraction of a group's early behavior. The model's generalizability allows for new insights into revealing hidden mechanisms behind collective behavior.

October 16, 2017 • Annenberg G02, 2120 Campus Drive • IPR

Sera Young
Assistant Professor of Anthropology and IPR Fellow

“Why Is Food Insecurity So Harmful in the First 1,000 Days of Life?  Findings from East Africa”

Abstract: The first 1,000 days, or the year prior to delivery and first two years of life, is among the most critical times for mothers and children. Yet, in this period, the physical and psychosocial needs of women and their infants are often not met. Food insecurity is a common manifestation of physical and psychosocial needs not being met, but its impacts in the first 1,000 days are not well understood. In this talk, Young will discuss her research on food insecurity among pregnant and lactating women of mixed HIV status in western Kenya. She will describe the perceived causes of food insecurity, as well as the surprisingly far-reaching consequences of food insecurity. Young will conclude by describing how listening to participants' experiences with food insecurity led to a new area of research: household water insecurity.

October 23, 2017 • Chambers Hall, 600 Foster Street • IPR

Christopher Udry
Professor of Economics and IPR Associate

“Unpacking a Multi-Faceted Program to Build Sustainable Income for the Poor”

Abstract: A multi-faceted "graduation" program for households in extreme poverty combines a capital grant in the form of a business asset (typically livestock), some business training/hand-holding, some short-term consumption support, and help with saving, through savings collection services. The goal is to help these households develop sustainable income-generating activities. In six out of seven simultaneously evaluated sites across the world, the program generated economically meaningful, cost-effective, and sustained positive average impacts on earnings, consumption, and other welfare measures over at least three years. Udry and his colleagues compare the impact of the full program with that of the savings component alone or the asset grant alone in Ghana and find that neither of the more limited programs had any sustained welfare effects. The evidence suggests that the additional training and consumption support enabled graduation households to accumulate the assets required for persistent effects. The researchers also examine spillover and general equilibrium effects of the graduation programs and find weak evidence that they exist.

October 30, 2017 • Annenberg G02, 2120 Campus Drive • IPR

Vijay Mittal
Assistant Professor of Psychology and IPR Associate

“Following Youth at Risk for Developing Schizophrenia: Translating Basic Psychopathology Research to Real-World Implementation”

Abstract: Schizophrenia is a chronic and devastating mental illness that occurs in 1–2 percent of the population. Because onset is typically right at the end of adolescence, the disorder significantly impacts families and communities and, more broadly, the public healthcare system. While this high-risk (prodromal) period represents a promising opportunity to improve etiological understanding and intervene prior to onset, researchers are as yet unclear about what differentiates those who do go on to develop a psychotic disorder from the phenotypically similar adolescents who do not. Furthermore, preliminary evidence suggests that while early interventions may be effective in reducing transition rates, the available treatments are associated with significant costs and side effects, and factors unique to this population significantly limit efficacy. Taken together, this suggests a critical need for effective biomarkers and targeted interventions. In this presentation, Mittal discusses characteristics and priorities in the psychosis prodrome and then reviews recent developments from a research program that is designed to translate findings from basic psychopathology studies to broadly disseminable assessments and treatments for this critical population.

November 6, 2017 • Chambers Hall, 600 Foster Street • IPR

Rachel Beatty Riedl
Associate Professor of Political Science and IPR Fellow

“Pews to Politics: How Religious Ideas Can Influence Modes of Political Engagement in Africa and Beyond”

Abstract: In public discourse, explanations for political action often point to the power of religious ideas. In contrast, social science scholarship often treats religious ideas and beliefs as epiphenomenal to political decision-making. This tension presents a puzzle: Do religious ideas influence political engagement? To what extent, under what conditions, and to what end? In this project, Riedl and her colleagues take advantage of the current diversity of religious ideas, communicated and interpreted through sermonic messages across Christian denominations in sub-Saharan Africa to explore these questions. Through a combination of observational (text analysis, surveys, focus groups), comparative historical case studies, and experimental work, they argue that exposure to certain religious messages can influence not only individuals' decisions whether to engage with politics but also the form that political engagement takes. However, the influences of religious messages have limits: They can decay quickly and their effects are content-dependent. These findings help address conflicting portrayals of religion's role in politics and suggest new directions for research.

November 13, 2017 • Chambers Hall, 600 Foster Street • IPR


James Rosenbaum
Professor of Education and Social Policy and IPR Fellow;

Lynn Meissner
IPR Graduate Research Assistant

Alexis Gable
IPR Research Study Coordinator

“Does College Lead to Good Jobs for Everyone in the Era of ‘College for All’?”

Abstract: Instead of the usual comparisons between high school graduates and BA graduates, Rosenbaum and his colleagues' analyses focus on sub-BAs (certificates and associate's degrees) and the ways mid-skill pathways can lead to good jobs. Analyzing a national survey, they describe relationships between high school test scores, educational attainment, and wage distributions, both within and across education categories. Second, the researchers go beyond earnings to explore objective job attributes and job satisfaction. Third, instead of blaming students for their poor choices, they find that some colleges create institutional procedures that reduce student mistakes and difficulties. Overall, their work identifies understudied phenomena that they believe should be a bigger part of the national conversation about how higher education can prepare young people for productive careers.

November 20, 2017 • Chambers Hall, 600 Foster Street • IPR

Eva Redei

David Lawrence Stein Research Professor in Psychiatric Diseases Affecting Children and Adolescents and IPR Associate

“Why Mental Illnesses Are Not Diseases and What to Do About It”

Abstract: The definition of disease is "a definite pathological process having a characteristic set of signs and symptoms." The definition of illness is "the experience of symptoms that are not identifiable with biomedical categories of disease." Thus, the major obstacle that mental illnesses—and the patients who are affected by them—face in being treated like any other "medical" disease is the lack of measurable pathological processes and markers. Redei and colleagues have identified panels of blood-based markers that can objectively distinguish patients with major depressive disorders from no-disorder matched subjects. The need for this type of objective measures, the development of it, and its potential use will be discussed. 

November 27 2017 • Chambers Hall, 600 Foster Street • IPR

IPR Q-Center Colloquium Series

Victoria Stodden
Associate Professor of Information Sciences, University of Illinois at Urbana–Champaign

“CompareML: Structuring Machine-Learning Research in Data-Driven Science”

Abstract: Statistical discovery is increasingly taking place using data not collected by the discoverers and often completely in silico. This calls on new considerations of methods and computational infrastructure that support statistical pipelines. In this talk, Stodden presents a novel framework for statistical analysis of “organic data” as opposed to "designed data" (Kreuter & Peng 2014) called CompareML that permits the direct comparison of findings that purport to answer the same statistical question. Stodden will illustrate that such computational frameworks are crucial to reproducible science by way of an example from genomics [acute leukemia (Golub et al 1999)] where traditional approaches (surprisingly) fail at scale.

October 18, 2017 • 617 Library Place • IPR

Linda Collins
Distinguished Professor of Human Development and Family Studies and Statistics; Director, The Methodology Center, Pennsylvania State University

"Introduction to the Multiphase Optimization Strategy (MOST) for Developing More Effective, Efficient, Economical, and Scalable Behavioral Interventions"

Abstract: Behavioral interventions are used widely for prevention and treatment of health problems, and for promotion of well-being and academic achievement. These interventions are typically developed and evaluated using the classical treatment package approach, in which the intervention is assembled a priori and evaluated by means of a two-group randomized controlled trial (RCT). Collins will describe an alternative framework for developing, optimizing, and evaluating behavioral interventions. This framework, called the multiphase optimization strategy (MOST), is a principled approach that has been inspired by ideas from engineering. MOST includes the RCT for evaluation, but also includes other steps before the RCT aimed at optimization. Using MOST, interventions can be optimized to achieve any of a variety of goals. For example, the goal may be to identify the most effective intervention that can be delivered within a specified upper limit on time or money. The MOST framework relies heavily on resource management by strategic choice of highly efficient experimental designs. Collins proposes that MOST offers more rapid long-run improvement of the effectiveness, efficiency, economy, and scalability of interventions, without requiring a dramatic increase in research resources.

December 6, 2017 • 617 Library Place • IPR

IPR Performance Measurement and Rewards Series

Diane Whitmore Schanzenbach
Margaret Walker Alexander Professor of Human Development and Social Policy and of Economics and IPR Director

“Evidence-Based Research and Policy-Making: Lessons Learned from Two Years in D.C.”

Abstract: In 2015, Schanzenbach was tapped to lead the Hamilton Project, an economic policy initiative of the Brookings Institution in Washington, D.C. During her two years as director, she developed economic research that addressed national policy needs such as reducing inequality and increasing social mobility. She wrote numerous policy briefs on pressing policy topics, including food security and SNAP, children’s early education, and school accountability, and she testified several times before U.S. House and Senate committees. After two years of engaging with national lawmakers and their staff, she will discuss what she learned in bridging the academic and policymaking communities—in particular how researchers can bring their research to the attention of policymakers.

December 5, 2017 • Annenberg G02, 2120 Campus Drive • IPR

Joint Economics/IPR Seminar Series

Sara Heller
Assistant Professor of Economics, University of Michigan

"Rethinking the Benefits of Youth Employment Programs: The Heterogeneous Effects of Summer Jobs"

Abstract: Heller reports the results of two randomized field experiments, each offering different populations of youth a supported summer job in Chicago. In both experiments, the program dramatically reduces violent-crime arrests, even after the summer. It does so without improving employment, schooling, or other types of crime; if anything, property crime increases over two to three post-program years. To explore mechanisms, Heller and her colleagues implement a machine-learning method that predicts treatment heterogeneity using observables. They find that employment benefiters commit more property crime than their control counterparts, and non-benefiters also show a decline in violent crime. These results do not seem consistent with typical theory about improved human capital and better labor market opportunities creating a higher opportunity cost of crime, or even with the idea that these programs just keep youth busy. Heller discusses several alternative mechanisms, concluding that brief youth employment programs can generate substantively important behavioral change, but for different outcomes, different youth, and different reasons than those most often considered in the literature.

October 19, 2017 • Kellogg Global Hub, Room 1410, 2211 Campus Drive • IPR

Ben Roth
Visiting Scholar, Kellogg School of Management, Northwestern University

“Paying for the Truth: The Efficacy of Peer Prediction in the Field”

Abstract: One of the most difficult problems in development is that the cost of assessing credit risk for small businesses often makes giving loans unprofitable. In a field experiment in Maharashtra, India, Roth and his colleagues asked 1,345 entrepreneurs to rank their peers on various metrics of business profitability and growth and entrepreneur characteristics. To assess the validity of these predictions the researchers then randomly distributed cash grants of about $100 to a third of these entrepreneurs. They find that information provided by community members is predictive of the marginal return to capital. They also horse-race the community rankings against a machine-learning prediction and find that while the machine-learning exercise is able to predict high-return entrepreneurs, community information continues to correctly predict 23 percent returns per month for the top entrepreneurs even after controlling the machine-learning prediction. Roth and his colleagues use experimental variation in the design of surveys to demonstrate agency problems in obtaining community information when community members know the purpose of the information gathering exercise. They conclude by demonstrating how mechanism design can be used to address these agency problems; monetary incentives for accuracy, eliciting reports in public, and cross-reporting techniques motivated by implementation theory all significantly improve the accuracy of reports. 

October 26, 2017 • Kellogg Global Hub, Room 1410, 2211 Campus Drive • IPR

Emily Breza
Assistant Professor of Economics, Harvard University

"Information Delivery Under Endogenous Communication: Experimental Evidence from the Indian Demonetization”

Abstract: To inform the public, how should policymakers disseminate information: by broadcasting widely (e.g., via mass media), or letting word spread from a small number of initially informed “seed” individuals? While conventional wisdom suggests more information is better, Breza and her colleagues argue that it may not be when participation in social learning is endogenous. In a field experiment during the chaotic 2016 Indian demonetization, the researchers vary how information is delivered to villages: the number informed (broadcasting versus seeding) and whether it is common knowledge who is informed. The results are consistent with four predictions of a signaling model in which, endogenously, people are more willing to engage in learning about information considered scarcer. First, without common knowledge, broadcasting does cause more conversations than seeding. Second, under seeding, adding common knowledge increases conversations. Third, holding common knowledge fixed, broadcasting decreases conversations relative to seeding. Fourth, when broadcasting, adding common knowledge also decreases conversations. Moreover, the treatments that increase conversations also lead to better knowledge and incentivized decisions, while those that decrease conversations lead to worse knowledge and incentivized decisions.

November 9, 2017 • Kellogg Global Hub, Room 1410, 2211 Campus Drive • IPR

Ricardo Perez-Truglia
Assistant Professor of Economics, University of California, Los Angeles

"How Much Does Your Boss Make? The Incentive Effects of Horizontal and Vertical Inequality”

Abstract: Perez-Truglia and his co-author study how employees learn about the salaries of their peers and managers, and how those beliefs affect their behavior. They conducted a field experiment with a sample of 2,000 employees from a multibillion-dollar corporation. First, they document large misperceptions about salaries and identify some of the sources of these misperceptions. Second, they identify the causal effects of these beliefs by means of an experiment that provided employees with feedback about the salaries of others. The researchers combine unique survey and administrative data to estimate cross-salary elasticities. They find that individuals are highly elastic to horizontal inequality: Greater inequality between peers has large negative or positive effects on satisfaction, effort, output, and retention depending on the person's position in the earnings distribution. Individuals are less informed and less elastic to vertical inequality. Increasing perceived vertical inequality had a nearly uniform positive impact on motivation and retention, even when the likelihood of promotion was small. The findings reject the widespread belief that social pay comparisons compel employers to compress salaries across the firm.  The researchers also find the willingness to pay for salary information is highest among those who react positively to the news. They discuss other implications of these findings, including optimal transparency.

November 16, 2017 • Kellogg Global Hub, Room 1410, 2211 Campus Drive • IPR

Pascaline Dupas
Associate Professor of Economics, Stanford University

"Targeting Subsidies: A Mechanism Design Approach"

Abstract: Pascaline Dupas’ research aims to understand the barriers that households and governments face in accumulating or fostering accumulation of health and education, and how these barriers can be overcome. She conducts extensive fieldwork—field experiments embedded in longitudinal data collection efforts, which are used to perform empirical tests of microeconomic theory and to quantify the effects of potential policies. Health is the primary focus of Dupas’ research to date. Her work covers the role of information and education in health behavior, and the role of subsidies in increasing adoption of health technologies. She is afilliated with the National Bureau of Economic Research and the Abdul Latif Jameel Poverty Action Lab.

December 7, 2017 • Kellogg Global Hub, Room 1410, 2211 Campus Drive • IPR