Abstract: On what basis can we claim a scholarly community understands a phenomenon? Social scientists generally propagate many rival explanations for what they study. How best to discriminate between or aggregate them introduces myriad questions because we lack standard tools that synthesize discrete explanations. In this paper, we assemble and test a set of approaches to the selection and aggregation of predictive statistical models representing different social scientific explanations for a single outcome: original crowd-sourced predictive models of COVID-19 mortality. We evaluate social scientists' ability to select or discriminate between these models using an expert forecast elicitation exercise. We provide a framework for aggregating discrete explanations, including using an ensemble algorithm (model stacking). Although the best models outperform benchmark machine learning models, experts are generally unable to identify models' predictive accuracy. Findings support the use of algorithmic approaches for the aggregation of social scientific explanations over human judgement or ad-hoc processes.

[“COVID-19 Model Challenges” Shiny app] [Pre-registration]

Abstract: Who is targeted by preventive repression and why? In the Soviet Union, the KGB applied a form of low-intensity preventive policing, called profilaktika. Citizens found to be engaging in politically and socially disruptive misdemeanors were invited to discuss their behavior and to receive a warning. Using novel data from Lithuania, a former Soviet republic, in the late 1950s and the 1970s, we study the profile and behaviors of the citizens who became subjects of interest to the KGB. We use topic modeling to investigate the operational focuses of profilaktika. We find that profilaktika began as a way of managing specific threats or “known risks” that arose from the experience of postwar Sovietization. The proportion of “unknown risks” – people without risk factors in their background or personal records – increased by the 1970s. These people were targeted because of their anti-Soviet behavior, which the KGB attributed to “contagious” foreign influences and the spread of harmful values.

Abstract: We study discriminatory policies of the tsarist government against Jewish entrepreneurs. No general incorporation law existed in Imperial Russia. Every single corporate charter had to be approved by the Ministry of Finance and signed by the tsar. Since 1890, the tsarist government started to include discriminatory clauses in some corporate charters that banned Jewish entrepreneurs from share ownership and/or purchasing property. What are the causes and consequences of discrimination in the short and long run? Using newly digitized data on the universe of Russian manufacturing factories in 1890, we find higher incidence of discriminatory clauses in corporate charters issued between 1891 and 1913 in more capital-intensive industries with fewer Jewish incumbents. To evaluate economic implications of discriminatory policies, we assemble a novel dataset from corporations’ balance sheets in 1885–1894. We find that corporations with discriminatory clauses in their charters were more likely to issue debt than equity. What was the effect of discriminatory barriers on potential market entrants? To address this question, we define counterfactual entrepreneurs as the universe of factory owners in 1890. Combining ethnicity information from one million individual WWI casualty records and the merchant guilds’ records to predict entrepreneurs’ ethnicity, we find evidence that barriers for Jewish entrepreneurs led to the incorporation of relatively less productive non-Jewish entrepreneurs. Finally, we show that discriminatory policies had broader economic implications including for the corporations without restrictions in their charters: corporations with Jewish founders that publicly traded on the St. Petersburg Stock Exchange in 1865–1913 underperformed post-1890.

Abstract: Access to patronage appointments is typically believed to confer an electoral advantage. Using a newly-assembled dataset of all persons elected to all U.S. state legislatures between 1900 and 2016, we study the effects of the introduction of merit civil service reform on reelection rates. Employing multiple new statistical methods appropriate for the staggered introduction of reform legislation, our results show that reelection rates significantly increase following the abolition of patronage appointments. We then show that one reason for the increase in reelection rates comes from increases in rates of rerunning by incumbents. We interrogate this with a discussion of why it might have occurred, and advance three hypotheses that could explain observed changes in political ambition.

Abstract: Contemporary authoritarian regimes are characterized by increasing personalization of power in the hands of the leader. While studies have explored the politics of power-sharing in these regimes, less is known about personalist leaders' strategies to manage the masses. To gain traction on this question we ask whether, in the process of amassing power, personalist leaders invest in non-coercive strategies of political control. The empirical analysis leverages novel, expert-coded data on state control of the education system and of the media and covers 220 regimes in the period 1950-2010. Contrary to conventional wisdom that personalist leaders shut down non-coercive pathways of control, we show that the personalization of power drives growing investments in strategies of indoctrination and information control through schools and the media. Findings, which answer several calls to move beyond the study of repression for understanding the politics of non-democracies, have implications for research on authoritarian politics and personalism.

Abstract: Innovative and resource-demanding data collection is crucial for the advancement of social science research. Many efforts have been made to assemble original datasets and make them publicly available for the benefit of the wider research community. Unfortunately, it is common for researchers to use existing datasets without paying sufficient attention to how they were constructed. To shed light on the advantages, limitations, and implications of different data collection methodologies, and assess how (often seemingly trivial) differences in assumptions or practices influence scores, we take advantage of a unique opportunity to compare three new historical datasets. These three datasets have important similarities that facilitate comparisons, measuring similar aspects of education practices and policies across countries, but they were created using different methods and even seemingly similar measures rely on slightly different assumptions. The EPSM dataset (Education Policies and Systems across Modern History) contains information about the content of de jure school curriculum, teacher training, and other education policies and is based on hand-coding a combination of primary and secondary sources. The HEQ initiative (Historical Education Quality Database) gathers information on similar issues but relies entirely on primary sources such as education laws, regulations, and national curriculum plans. The V-Indoc dataset (Varieties of Indoctrination) relies on country expert assessments of school curriculums, teacher policies, and the presence and nature of political indoctrination. We introduce each dataset and characterize the degree of convergence/divergence between comparable variables along several relevant dimensions.

Abstract: What explains the existence of predatory criminal persecution practices? Is it rent seeking behavior and private interests of law enforcement officials, or the inefficiency of the police institution? In this paper we empirically test the relationship between indicators of economic crimes in Russia's regions, the level of economic activity, and turnover of regional elites. Our main goal is to find out whether private interests or the so-called "stick" system are responsible for the overall upward trend in economic crimes observed in 2004–2009. We use a unique ICSID database, which contains official MVD’s (Ministry of Internal Affairs) data on economic crimes (according to the articles of the Russian Criminal Code), along with biographical data for chiefs of regional police departments. Our results suggest that the “stick” system based on key performance indicators is responsible for the intensifying upward trend in the dynamics of economic crime rates in 2004–2009, which overshadows negative consequences of predatory persecution practices.