Research

 

I study the political origins and consequences of our beliefs about identity. My research employs integrative designs and tools from multiple disciplines, from archived historical data, to automated text analysis, to diverse qualitative primary sources from both in and out of the archive, to elite interviews, to large-scale randomized control trials with human subjects. With these, I explore nationality in Eurasia, the politics of authoritarian regimes, racial disparities in health here in the United States, and social science pedagogy.

You can find my projects, working papers, and publications under each of these agendas below.

 
 

How does national identity form and change?


The Right to Write Badly: The Failure of Soviet State Administration in Literature in the 1930s

[ under review ]

The Soviet state endowed its writers, artists, and musicians with material privileges and institutional autonomy; in exchange, these cultural professionals were expected to manufacture a coherent, productive, and heroic “postbourgeoisie culture system.” This study draws on a range of primary sources to explain how the Writers Union became the institutional model for administrating of cultural life in the Soviet Union, even while its predecessor organizations failed. Using a body of primary sources, including diaries, public speeches, memoranda, secret police reports, and archived historical interviews, I show that significant discretion over the allocation of state resources and an erratic regulatory structure incentivized the creation of “informal cultural economy,” where writers, artists, and other creatives kept their best work unpublished, instead choosing to trade it amongst themselves in the privacy of their apartments. This informal cultural economy frustrated Soviet officials and secret police officers, but preserved the Writers Union as an institution that remained “good enough” for Soviet culture-building. This paper explains how national intelligentsias were able to emerge in a centralized authoritarian system to begin with, and is currently under review.

Portrait of Maxim Gorky, author and founding father of the Soviet Union of Writers

Portrait of Maxim Gorky, author and founding father of the Soviet Union of Writers

 

THE LONG-RUN EFFECTS OF STALIN-ERA FORCED RESETTLEMENT ON STATE FORMATION IN THE SOVIET UNION

[ Working paper ]

Between 1937 and 1949, the Soviet Union under Stalin forcibly resettled approximately 3 million Soviet citizens across more than 11 different nationalities. Regions in the Soviet interior that received these “special settlers” experienced a distinct political and economic environment, where day-to-day governance was handled entirely by the Soviet security apparatus. This paper situates this case theoretically as an expansion of state infrastructural power in the Soviet periphery. Using data on the forced relocation of ethnic Germans to south-central Russia, I show that the Soviet state here endured even after those Germans left - districts that received forced migrants of German ethnicity have a higher presence of Russian police even today.

Distribution of ethnic German settlers in western Siberia, 1937-49

 

Who Built Nations in Eurasia? How Writers, Artists, and Professors Shaped Beliefs about National Identity

[ working paper ]

Social scientists understand national identities today to be a legacy of mass native language literacy. However, this structuralist explanation fails to account for the full diversity of ways that nationality has been imagined across Eurasia. This study introduces a new class of political actor – intelligentsias – and argues that these writers, artists, and even professors changed beliefs about nationality in the Soviet Union more rapidly and dramatically than existing theories anticipate. Using original data on the first free and fair elections in Soviet history and a difference-in-differences strategy, I show that Soviet territories where intelligentsia entered formal politics were more likely to mobilize around nationalist demands during the Soviet collapse. To demonstrate the independent role of intelligentsias, I reproduce the relationship between historical mass native-language literacy and nationalist sentiment to identify cases that over- and under-performed their structural preconditions. Using a body of interviews and primary sources from in and out of the archive, I examine the deviant and least likely case of Armenia – one of the Soviet Union’s most loyal republics that became one of its most secessionist almost overnight. Here, writers, artists, and a poet popularized grievances over a disputed territory, allowing a more radical group of academics to hijack the national movement and steer it towards independence. This answer to the understudied question of the origins of national identity suggests that actors who may not hold apparent political power but do possess nationalist, religious, or other cultural credentials can change beliefs about identity in surprising ways.

Nationalist protest in Soviet provinces that did versus did not elect writers, artists, and professors to office in 1989

 

How States Achieved Mass Literacy: Evidence from the Post-Revolutionary U.S.S.R.

[ working paper ]

Classic works of sociology (Gellner, 1986; Weber, 1976) and contemporary evidence (Darden & Grzymala-Busse, 2006) tell us that national identities in Europe emerged from state efforts at mass education in the 19th century. After 1917, faced with the task of constructing a multi-national and socialist identity from the ruins of a multi-ethnic and largely illiterate Tsarist empire, Bolshevik leaders followed this conventional wisdom, prioritizing and aggressively pursuing a campaign to “liquidate illiteracy” amongst the adult Soviet population. Yet the Soviet campaign also poses tremendous within-case variation in organizational strategy. The Bolshevik campaign was led by a broad negotiated alliance of both state and non¬-state organizations, targeting different but often overlapping constituencies of illiterate and semi-literate Soviet citizens.What drove sub-national variation in the kinds of organizations tasked with liquidating illiteracy? What were the effects of organizational composition on the campaign’s success? And what were the downstream effects of these campaigns on the formation of national identities and on loyalty to the Soviet Union? This paper answers these questions using original data on the Soviet literacy campaigns collected from the State Archives of the Russian Federation (GARF) and produced by the administrators of these campaigns themselves, including data on literacy rates, literacy targets, the distribution of agitational brochures an cultural personnel, the opening of reading rooms, and the allocation of budgets to literacy education.

“Since when did you turn away from the sacred book? Since I learned to read this book here [book is labelled ‘Lenin’]”

 

How do authoritarian regimes function and change?


WHEN DO DICTATORS GO IT ALONE?: PERSONALIZATION AND OVERSIGHT IN AUTHORITARIAN REGIMES

[ wITH CHRIS CAROTHERS & ANDREW LEBER ]

Politics & Society. March 2022.

Why are some autocrats able to personalize power within their regimes while others are not? We find that often the crucial relationship is between the autocrat and retired leaders, party elders, and other elites of the outgoing generation. Authoritarian regimes are more likely to resist a personalist takeover when members of this “old guard” retain oversight capacity over their successor, restraining them from overturning norms of collective rule and maximizing individual power. We illustrate this argument with two case studies: China under President Xi Jinping and Vietnam under General Secretary Le Kha Phieu. In addition, we use an original dataset of authoritarian leadership transitions to demonstrate the generalizability of this relationship around the world, and to rule out potential confounders. This study introduces oversight capacity as a new concept that deepens our understanding of elite politics and inter-generational conflict in authoritarian regimes.

 

UNDERSTANDING MASS PURGES: TESTING HYPOTHESES IN THE HIGH STALINIST CASE

[ working paper ]

Mass purges are costly because they deprive the state of skilled and experienced personnel, yet purges have been characteristic features of some of the most durable authoritarian regimes. This paper uses an individual-level dataset of over 2.3 million political arrests made during Stalin’s “Great Terror” in 1937-38 to test two sets of questions. First: why were purges used to target some national minority groups, but not others? Second: drawing on an influential argument from the Soviet historiography (Fitzpatrick, “Stalin and the Making of a New Elite”), was the timing of mass purges constrained by the supply of qualified replacement personnel?

Distribution of political arrests by apartment, Moscow, 1937-39

 

The Democratizing Role of Undemocratic Institutions in 1830s Britain

[With Jared Abbott & Aytug Sasmaz]

[working paper]

Explaining Europe’s ‘gradual’ democratizations means answering the question: When do incumbent power-holders in autocracies choose to put their own influence at risk by extending the franchise to new voters? In this paper, our answer is that incumbents democratize when they anticipate sufficiently low costs of tolerating an expanded electorate. Based on a replication of work on the 1832 Reform Act in Britain by Aidt and Frank (2015), and using original data on political corruption in British electoral boroughs, we show that British conservatives were more willing to extend the franchise when their risk of losing office was cushioned by vote-buying networks that would survive these democratizing reforms.  

 

How does identity shape our health?


I work with a team of researchers at Boston Medical Center on two randomized control trials that seek to understand the social determinants of health amongst underserved social groups.

 

Understanding Racial disparities in Preconception Health

In the United States, severe health disparities exist on the basis of race. These disparities are especially sharp in preconception health: Black and African American women are twice as likely as white women to have children born with health issues, and four times as likely to die during childbirth. The Covid-19 pandemic has highlighted and magnified these disparities.

This randomized control trial tests whether we can improve preconception health in Black and African American women using education focused on a set of behaviors and conditions that put them at risk for adverse birth outcomes. The study tests the effects of Gabby - a Virtual Patient Advocate designed to supplement in-person care from a physician - on how women respond to 102 preconception health risks across 13 domains like access to healthcare, emotional and mental health, and social determinants of preconception health like racially based discrimination. You can read more about the Gabby project and preconception care amongst Black and African American women here.

The primary results from this work were published in The Lancet Digital Health. Subsequent published studies explored more deeply the effects of our health education intervention for nutrition-related health risks that are especially salient for Black and African Women, as well as the differential effects of Gabby amounts adolescent and young adult women. Current work unpacks data collected in the course of this project to determine whether our intervention was more likely to change health conditions that have been most affected by implicit racial bias from providers. Our subsequent studies test whether our findings can be replicated amongst women serving in the U.S. military, amongst low-income and uninsured women at “healthy start” clinics, and amongst women in Lesotho.

[ Publications based this study ]

"Improving the health of young African American women in the preconception period using health information technology: a randomised controlled trial." The Lancet Digital Health 2.9 (2020): e475-e485. (with Brian W. Jack, et al.)

"Using Health Information Technology to Engage African American Women on Nutrition and Supplement Use During the Preconception Period." Frontiers in Endocrinology 11 (2021): 1045. (with Paula Gardiner, et al.)

"Promotion of preconception care among adolescents and young adults by conversational agent." Journal of Adolescent Health 67.2 (2020): S45-S51. (with Timothy Bickmore, et al.)

"Study protocol for the implementation of the Gabby Preconception Care System-an evidence-based, health information technology intervention for Black and African American women." BMC health services research 20.1 (2020): 1-14. (with Angela Walter, et al.)

“Use of a Web-Based Preconception Health Information Technology System Designed for Black and African American Women: Results of a Hybrid Type II Implementation-Effectiveness Study.“ (with Clevanne Julce et. al.) [ Working Paper ]

“Can Culturally Sensitive Education Reduce Racial Disparities in Health? Evidence from a Randomized Control Trial.“ [ Working Paper ]

The Lancet Digital Health issue 2.9 (2020) cover featured this culturally sensitive health education system.

 

making Healthcare utilization more efficient for Patients with Depressive Symptoms

Patient rehospitalization is often used as a metric of healthcare system performance - rehospitalization proxies for the underlying health of patients, and is expensive in and of itself. Previous research found that streamlining and improving hospital discharge procedures can substantially reduce rehospitalization rates, but not in patients with mental health conditions like depression.

This randomized control trial tests whether providing more tailored information, along with therapy, can keep patients with depression out of the hospital following discharge. The study intervention provides, telephonically, patients with assistance navigating the healthcare system, tools for patient self-management, and cognitive behavioral therapy over the course of four months, and test the effects of these interventions on subsequent rehospitalization rates. You can read more about project re-engineered discharge (RED), on which this study is based, here and here.

[ Publications based this study ]

“Reducing Hospital Readmission among Hospitalized Patients with Depressive Symptoms: A Randomized Trial“ The Annals of Family Medicine. (with Suzanne Mitchell, et. al.)

A patient uses the “re-engineered discharge” (RED) system.

 

The Effects of Culturally Sensitive Diabetes Care for Black/ African American and Hispanic women

This randomized controlled non-inferiority trial, titled “Women in Control,” tests whether a culturally sensitive, online health education and diabetes-self-management tool can perform as well as in-person diabetes self-management for increasing physical activity and improving glucose control among Black/ African American and Hispanic women with uncontrolled diabetes mellitus (DM). You can read more about the Women in Control project here.

[ Publications based this study ]

“Comparative Effectiveness of In-person versus Virtual World Diabetes Self- management Education for Minority Women: The Women in Control 2.0 Randomized Controlled Trial” (with Suzanne Mitchell et. al.) [ Under Review ]

“Identifying Mediators and Moderators of Culturally Sensitive Diabetes Self-management Education for Minority Women“ (with Suzanne Mitchell et. al.) [ Working Paper ]

 

How can we teach social science better?


Setting Expectations: Rubrics as a Formative tool for Communicating in the Social Sciences

[ revise & resubmit ]

[ with Sarah James, Colin Brown, George Soroka, & Aaron watanabe ]

While many college campuses include some kind of required, institutionalized, and universal writing instruction for their students, such programs often do not cover the specific skills, conventions, styles, and writing goals demanded by individual disciplines. This series of projects seeks to test whether a bundle of low-cost pedagogical tools found to be effective in the education literature - including rubrics, low-stakes assignments, peer review, and writing across the curriculum - can improve performance in writing specifically for the social sciences.

The most current piece of this long-term research project randomly assigns a writing rubric and a series of short, in-class instructional sessions focused on teaching and practicing individual rubrics rows, and compares post-intervention quality of student writing with a set of controls receiving usual instruction, within two large gateway political science classes. This randomized control trial follows up on a study by the same group of authors, which found that a similar rubric-based intervention can improve the specific writing skills taught in the intervention - the full paper for this earlier study can be found here.

A second follow-up study currently underway evaluates the same rubric-based intervention across three campuses - one large elite research institution, one mid-sized private research university focused on professional job experience, and one small public university - to test external validity across student cohorts that differ in demographic makeup and educational goals.

Future iterations of this research - currently submitted for grant funding - seeks to expand the intervention by testing other low-cost writing interventions across a greater diversity of student bodies - including in community college settings.