An Empirical Evaluation of Explanations for State Repression

Abstract

T he empirical literature that examines cross-national patterns of state repression seeks to discover a set of political, economic, and social conditions that are consistently associated with government violations of human rights. Null hypothesis significance testing is the most common way of examining the relationship between repression and concepts of interest, but we argue that it is inadequate for this goal, and has produced potentially misleading results. To remedy this deficiency in the literature we use cross-validation and random forests to determine the predictive power of measures of concepts the literature identifies as important causes of repression. We find that few of these measures are able to substantially improve the predictive power of statistical models of repression. Further, the most studied concept in the literature, democratic political institutions, predicts certain kinds of repression much more accurately than others. We argue that this is due to conceptual and operational overlap between democracy and certain kinds of state repression. Finally, we argue that the impressive performance of certain features of domestic legal systems, as well as some economic and demographic factors, justifies a stronger focus on these concepts in future studies of repression.

Human Rights Texts: Converting Human Rights Primary Source Documents into Data

“…It is our hope that researchers from all different fields will use a variety of statistical analyses and machine learning algorithms to better understand the content of these documents and how this content has evolved over time and in response to political and reporting changes. More generally, these datasets, coupled with the existing human coding schemes, automated coding algorithms, and innovative new research designs [20,22,[64][65][66], will allow scholars to more thoroughly analyze reports of human rights abuse and therefore extend and contribute to the growing human rights literature.…”

“…Unsupervised statistical learning tools also exist, which are useful for revealing other patterns within the human rights document corpus [7,[13][14][15][16] without reference to the existing coded human rights variables, which we describe below. These tools are more generally part of the emergent field of computational social science or "big data" analysis [17][18][19] of which there are several recent examples in the study of human rights [1,10,[20][21][22][23] and many other examples from political science and social science more generally [11,[24][25][26][27][28].…”

“…The variables generated from these coding projects have been used in hundreds of published studies to understand why governments around the world choose to violate the civil and political rights of individuals (for reviews of the current and past states of this literature and extensions to it see [1,20,23,[41][42][43][44][45][46][47][48][49][50]. [1] provides a detailed discussion of the coding and documentation procedures of the existing datasets presented in this article as well as other sources of data used by the human rights community.…”

Maintaining ethnic dominance: Diversity, power, and violent repression

“…Tables) and oil and gas income (logged, Ross, 2013) are included because ethnically exclusive states are overrepresented among poor and natural resource-rich countries and because we expect modernized countries to have greater respect for human rights unless their economies are based on natural resources. Next, countries with large populations often have greater strain on resources and tend to use repression to keep demands down (Walker, 2007: 33 Hill and Jones, 2014;Nordås and Davenport, 2013;Poe and Tate, 1994). To minimize simultaneity bias all time-variant independent variables are lagged one year.…”

“…In comparison, as also shown in Table III, we would expect maximum changes in executive constraint and years with armed conflict to change the level of repression by 20% and 19%, respectively. Executive constraint and armed conflict have been highlighted as some of the predictors of repression with most explanatory power (Hill and Jones 2014), and the fact that ethnic exclusion comes close to such levels suggests that it should be regarded as an important predictor of violent repression.…”