Christopher J. Fariss

Military spending data measure key international relations concepts such as balancing, arms races, the distribution of power, and the severity of military burdens. Unfortunately, missing values and measurement error threaten the validity of existing findings. Addressing this challenge, we introduce the Global Military Spending Dataset (GMSD). GMSD collates new and existing expenditure variables from a comprehensive collection of sources, expands data coverage, and employs a latent variable model to estimate missing values and quantify measurement error. We validate the data and present new findings. First, correlations between economic surplus and military spending are currently higher than at any point in the last two-hundred years. Second, updating DiGiuseppe and Poast’s (2018) analysis, we find larger substantive effects. Specifically, we find that the (negative) effect of a democratic ally on military spending is three times larger, and the (positive) effect of an increase in GDP is five times larger than previously estimated.

What information about climate change related extreme weather events or experiences with these disasters reduce the partisan gap? Despite the scientific consensus that human activity is the primary cause of climate change, individual beliefs and policy preferences about climate change remain divided along partisan lines. However, extreme climate events and disaster experiences force vividly tangible information about the proximity and severity of climate change upon exposed individuals, potentially prompting belief-updating and preference-shifting toward pro-climate policies. As climate change events increase in frequency and severity, more individuals are directly exposed to climate change on a personal level. A two-wave survey corresponding to multiple extreme weather events in Texas and a difference-in-differences analysis using power outage data in a natural experiment from the 2021 North American Winter Storms show that the partisan divide is smaller among individuals who had direct personal experiences with extreme weather events.

Where is the human rights discourse most resonant? We use aggregated cross-national Google search data to test two divergent accounts of why human rights appeal to some populations but not others. The top-down model predicts that nationwide interest in human rights is attributable mainly to external factors such as foreign direct investment, transnational NGO campaigns, or international legalization, whereas the bottom-up model highlights the importance of internal factors such as economic growth and persistent repression. We find more evidence for the latter model: not only is interest in human rights more concentrated in the Global South, but the discourse is also most resonant where people face regular state violence. In drawing these inferences, this article confronts high-level debates over whether human rights will remain relevant in the future, and whether the discourse still animates counter-hegemonic modes of resistance. The answer to both questions, our research suggests, is “yes.”

Most cross-national human rights datasets rely on human coding to produce yearly, country-level indicators of state human rights practices. Hand-coding the documents that contain the information on which these scores are based is tedious and time-consuming, but has been viewed as necessary given the complexity and detail of the information contained in the text. However, advances in automated text analysis have the potential to streamline this process without sacrificing accuracy. In this research note, we take the first step in creating this streamlined process by employing a supervised machine learning automated coding method that extracts specific allegations of physical integrity rights violations from the original text of country reports on human rights. This method produces a dataset including 163,512 unique abuse allegations in 196 countries between 1999 and 2016. This dataset and method will assist researchers of physical integrity rights abuse because it will allow them to produce allegation-level human rights measures that have previously not existed and provide a jumping-off point for future projects aimed at using supervised machine learning to create global human rights metrics.

Gross domestic product (GDP), GDP per capita, and population are central to the study of politics and economics broadly, and conflict processes in particular. Despite the prominence of these variables in empirical research, existing data lack historical coverage and are assumed to be measured without error. We develop a latent variable modeling framework that expands data coverage (1500 AD–2018 AD) and, by making use of multiple indicators for each variable, provides a principled framework to estimate uncertainty for values for all country-year variables relative to one another. Expanded temporal coverage of estimates provides new insights about the relationship between development and democracy, conflict, repression, and health. We also demonstrate how to incorporate uncertainty in observational models. Results show that the relationship between repression and development is weaker than models that do not incorporate uncertainty suggest. Future extensions of the latent variable model can address other forms of systematic measurement error with new data, new measurement theory, or both.

I present two key components from a course designed to introduce undergraduate students to human rights: a set of group-based active learning tasks and an individual-based sequential research project. In the classroom, active learning opportunities allow students to creatively and collectively engage with course material. The sequential research project is a step-by-step guide for creating an original research paper. For the two components, the students draw from a set of primary source documents combined with additional readings to build knowledge in the classroom. With this new knowledge, the students generate ideas and content that they use to write a sequence of research essays about that course topic outside the classroom. In this manuscript, I describe the shared structure of the two learning components, discuss details about each of the sequential essays, present assessment data, and provide suggestions about how to adapt the course to other social science topics.

Gross Domestic Product (GDP), GDP per capita, and population are central to the study of politics and economics broadly, and conflict processes in particular. Despite the prominence of these variables in empirical research, existing data lack historical coverage and are assumed to be measured without error. We develop a latent variable modeling framework that expands data coverage (1500 A.D--2018 A.D) and, by making use of multiple indicators for each variable, provides a principled framework to estimate uncertainty for values for all country-year variables relative to one another. Expanded temporal coverage of estimates provides new insights about the relationship between development and democracy, conflict, repression, and health. We also demonstrate how to incorporate uncertainty in observational models. Results show that the relationship between repression and development is weaker than models that do not incorporate uncertainty suggest. Future extensions of the latent variable model can address other forms of systematic measurement error with new data, new measurement theory, or both.

The legitimacy of the state rests on individuals' perceptions of fairness when interacting with state institutions and state agents. The police as an institution and as individual agents have wide latitude to detain and use force against individuals. We argue that encounters with state bureaucracy and civil servants, specifically the police, can generate individual-level grievances against the state, and that these grievances make it more likely an individual participates in protest against the state. We study support for and the legitimacy of policing in the context of the anti-police protests in Baltimore, MD following the death of Freddie Gray in April, 2015. Using data from police records and social media, we show that individuals with higher exposure to discretionary arrests --- arrests that are potentially viewed as illegitimate or arbitrary --- are more likely to support protests against the police. In contrast, we demonstrate that exposure to arrests for major crimes such as murder does not follow the same pattern. Thus, support for the police as an institution varies systematically with exposure to arbitrary and capricious encounters with police agents. As these grievance generating encounters become more widespread, we expect to see increased protests against the police and further erosion in support of the police as an institution. Alternatively, shifting institutional resources to focus on major crimes and limiting the discretionary authority of police agents when interacting with the public may help to repair the legitimacy of policing institutions over the long term.

The observation, measurement, and analysis of violent and contentious processes are essential parts of the scientific study of peace and conflict. However, concepts such as the level of repression, the number of individuals killed during a civil war, or the perception of members of an out-group, are often by definition difficult to observe directly. This is because governments, non-state groups, NGOs, international organizations, monitoring organizations, and other actors are not incentivized to make information about their actions systematically observable to analysts. In this context, latent variable models can play a valuable role by aggregating various behavioral indicators and signals together to help measure latent concepts of interest that would not otherwise be directly observable. Each of the articles in this special issue uses some form of a latent variable model or related measurement model to bring together observable pieces of information and estimate a set of values for the underlying theoretical concept of interest. Each of the articles pays special attention to the processes that make the observation of peace and conflict processes so challenging. As we highlight throughout this introductory article, the unifying framework we present in this special issue is validation. Though the substantive content of each of the articles in this special issue varies, they represent the diversity of substantive interests that span the study of peace and conflict, broadly conceived. Overall, we hope that the special issue becomes a standard reference for scholars interested in developing and validating new measurement models for the study of peace and conflict.

Counting repressive events is difficult because state leaders have an incentive to conceal actions of their subordinates and destroy evidence of abuse. In this article, we extend existing latent variable modeling techniques in the study of repression to account for the uncertainty inherent in count data generated for this type of difficult-to-observe event. We demonstrate the utility of the model by focusing on a dataset that defines ‘one-sided-killing’ as government-caused deaths of non-combatants. In addition to generating more precise estimates of latent repression levels, the model also estimates the probability that a state engaged in one-sided-killing and the predictive distribution of deaths for each country-year in the dataset. These new event-based, count estimates will be useful for researchers interested in this type of data but skeptical of the comparability of such events across countries and over time. Our modeling framework also provides a principled method for inferring unobserved count variables based on conceptually related categorical information.