Manuscripts, 2005 & 2006 & 2007
Analyzing Dependencies in Geo-Economics and Geo-Politics
Michael D. Ward & Peter D. Hoff
June 2007
Abstract
Using data over the period from 1950-2000 we estimate a model of bilateral international trade to explore the linkages between a) alliances, b) joint memberships in international institutions, c) mutual cooperation and d) conflict, e) mutual economic freedom and f) democracy and bilateral trade. We incorporate of exporter and importer effects as well as reciprocity into a gravity model and cross-validate it against annual out-of-sample data. The resulting, empirical findings show the importance of second and third order dependencies in bilateral trade data. Military alliances, membership in IGOs, international cooperation, and mutual economic freedom are shown to be strongly associated with bilateral trade. Conversely, conflict and the level of democracy do not demonstrate strong, discernable linkages to bilateral trade.
replication archive to be developed.
Social Distance in Bosnia and the North Caucasus Region of Russia
Kristin M. Bakke, Xun Cao, John O'Loughlin, & Michael D. Ward
May 2007
Abstract
We examine ethnic divisions in conflict-affected and post-conflict societies. Conventional wisdom tells us that societies that have experienced violent struggles in which individuals of different ethnic groups have (been) mobilized against each other are likely to become ossified along ethnic lines. Indeed, both policy-makers and scholars often assume that such divisions are one of the main challenges that must be overcome to restore peace after war. We comparatively assess this hypothesis by mapping dimensions of social distance among 4,000 survey respondents in Bosnia and the North Caucasus region of Russia. The surveys were carried out in December 2005. Using multidimensional scaling, we do not find evidence for clear attitudinal cleavages among members of different ethnic groups in Bosnia. Nor do we find strong evidence for ethnic divisions in the North Caucasus, although our measurements of social distance reveal a difference between Russians and ethnic minority groups.
Disputes, Democracies, & Dependencies: A Re-examination of the Kantian Peace
Michael D. Ward, Randolph Siverson, & Xun Cao
replication archive
Maximum Likelihood for the Social Sciences
Michael D. Ward
March 2007
Abstract. This is a didactic, working manuscript about maximum likelihood methods in an applied social science setting.
Draft (PDF) Version of March 23, 2007
also large;
Code; Data to follow
An Introduction to Spatial Regression Models in the Social Sciences
Michael D. Ward & Kristian Skrede Gleditsch
March 2007
Abstract. This is a didactic, working manuscript about regression analysis of, well, spatially dependent data.
Draft (PDF) Version of 15 June 2007, call it version 1.0.
Code; Data to follow
Inter-Ethnic Trust in Conflict-Affected Societies: Bosnia and Herzegovina and the North Caucasus Region of Russia .
Michael D. Ward, John V. O'Loughlin, Kristin M. Bakke & Xun Cao.
April 2007.
Abstract. Bosnia and the North Caucasus are ethnically diverse, post-communist societies, where the different ethnic groups at times have co-existed peacefully and at other times have found themselves at odds with one another or their governments. This study examines beliefs in the possibility of inter-ethnic cooperation in each society, based on survey instruments aimed at measuring attitudes and preferences towards the contemporary situation, socio-demographic characteristics, and the nature of cross-national relations in the light of experiences of conflict and continued unsettled political environment of the region. Our dependent variable, belief in the possibility of inter-ethnic cooperation, is a categorical variable based on responses the following survey stub: Among national groups, it is possible to create cooperation but never to fully trust. We measure and correct for survey response incomparability across BiH and the North Caucasus by using the anchoring vignette along with an estimation technique called chopit (short for compound hierarchical ordered probit), which allows us to incorporate anchoring vignettes.
April 2007 version --- replication archive
Diffusion and the International Context of Democratization
Kristian Skrede Gleditsch & Michael D. Ward
August 2006.
Abstract.Democracy does not evolve sui generis. The spatial clustering in democracy and transtions suggests that international factors play a prominent role in forging democracies, as well as influencing their durability. International Organization, Fall 2006, Volume 60, Number 4, pages 911- 933.
2006 uncorrected page proofs ;
Everybody Out of the Pool!
Michael D. Ward, Randolph M. Siverson, & Xun Cao.
October 2005.
Abstract. Research in international politics may have produced misleading
results because (1) typical data contain dependencies that have
been ignored, (2) populations are treated as samples, with
unwarranted reliance on misleading significance tests, and (3)
scant attention is devoted to how well the model can predict the
events of interest. Using the democratic peace research program
as an example, we show that the three elements of the Kantian
model–-mutual democracy, high trade and common membership in
IGOs--have weak effects on dampening the occurrence of militarized
international disputes within dyads. Neither do they offer
meaningful predictions about which dyads will be involved in these
disputes. A model incorporating several types of dependencies
among countries yields results with high levels of predictive
accuracy and provides new substantive insight about the prominence
of dependencies in international relations.
New, Improved Version, September 2006
2006 Draft;
Earlier Draft; Current thinking.
What Are the Neighbors Doing?
Michael D. WardSeptember 2005
Abstract. Civil conflict and civil war are clustered in time and space. I explore new ways to model spatial dependencies in the context of the complexity of civil conflict. Using methods adapted from the social relations model, it is possible to represent social dependencies and clustering among states in a way that incorporates country level as well as dyadic covariates. This clustering can recapture the often-observed clustering of societies that is observed in many empirical data sets on civil war. I describe this approach in the framework of familiar regression models and explore implications for civil war studies of adopting this approach.
Draft & Slides (PDF); Data; Code
Persistent Patterns of International Commerce (Forthcoming, Journal of Peace Research)
Michael D. Ward and Peter D. Hoff
August 2005; Revised January 2006
Abstract.
We examine a standard gravity model of international commerce augmented
to include political as well as institutional influences on bilateral trade. Using annual data
from 1980-2001, we estimate regression coefficients and residual dependencies using a hierarchy
of models in each year. Rather than gauge the generalizability of these patterns via
traditional measures of statistical significance such as p-values, we develop and employ a
strategy to evaluate their out-of-sample predictive strength. Our analysis of recent international
commerce shows that in addition to a typical gravity-model specification, political
and institutional variables are important. We show that the often reported link between
international conflict and bilateral trade is elusive, and inclusion of conflict in a model for
trade can sometimes lead to reduced out-of-sample predictive performance. Further, we
show that there are substantial, persistent residual exporter-and importer-specific effects,
and that ignoring such patterns in dyadic trade data results in an incomplete picture of
international commerce, even in the context of a well-established framework such as the
gravity model.
Revised Draft (PDF this is ~5MB); Data; Code
Diffusion and the International Context of Democratization (International Organization)
Kristian Skrede Gleditsch and Michael D. Ward
January/December 2005
Abstract. Democracy does not evolve sui generis. The spatial clustering in democracy and transitions toward democracy suggests that international factors play a prominent role in bringing about democracies as well as influence their durability. We argue that democracy comes about as result of changes in the relative power of important actors and groups as well as their evaluations of particular institutions, both of which are often influenced by forces outside the country in question. The scope and extent of connections with other democratic countries in a region can strengthen support for democratic reform and help sustain institutions in transitional democracies. Results from a transition model demonstrate that international factors can exert a strong influence on the prospects for transitions to democracy, and the spatial clustering in democracy and transitions cannot adequately be explained by the hypothesized domestic social requisites of individual countries. International Organization, 2006
Draft (PDF); Data; Code
Predicting Civil Conflicts: On the Utility of Empirical Research
Michael D. Ward & Kristin M. Bakke
June 2005
Abstract. Large-N studies of conflict produced a large number of statistically significant
results, but very little accurate guidance in terms of anticipating the onset of conflict.
There are several potential reasons for this. We provide a survey of three major efforts to
understand the onset of civil conflict. We use prediction to gauge the effectiveness of these
statistical models. We offer some lessons for the use of case control methods and forecasting
as theoretical endeavors in the study of conflict.
Draft (PDF); Data; Code
Protecting Jobs in the Age of Globalization: Examining the Relative Salience of Social Welfare and Industrial Subsidies in OECD Countries (Forthcoming, International Studies Quarterly)
Xun Cao, Aseem Prakash, & Michael D. Ward
November 2005
Abstract. The relationship between economic openness and welfare policies has become increasingly
important to policy makers. While scholars have tended to examine conditions under which budgets for
social welfare programs ebb and flow along with countries’ exposure to trade, they have overlooked how
governments may compensate domestic labor by subsidizing their employers. To address this issue we
examine the relative salience of social welfare expenditures to industrial subsidies in a panel of 16 OECD
countries from 1980 to 1995. Our results suggest that the relative budgetary salience of social welfare to
industrial subsidies is influenced by the interplay between governmental partisan gravity and changes in
imports. Unlike Right governments, Left governments tend to favor industrial subsidies in the wake of
negative, zero or moderate increases in imports. Faced with sharp increases in imports, Left governments
switch their preferences to compensating workers via social welfare.
Draft (PDF); Data; Code
Visualization and International Relations
Kristian Skrede Gleditsch & Michael D. Ward
Abstract. Geographical features exert a strong influence on the ways in which actors in
international politics interact with one another, yet our empirical analyses often
ignore such aspects. We demonstrate how visualization can help reveal
spatial patterns in political processes, which in turn can help researchers discover
new features of these processes that are not apparent from aggregate
analysis and standard representations of the data. We review how maps can
help display, identify, and highlight trends in political data, and illustrate exploratory
and descriptive spatial statistical techniques. Our example is based
on the relationship between democracy and violent conflict. We demonstrate
that although conflict and democracy may not be related at the level of individual
states, once we look at the regional level and take into account where
conflict is located, we find considerable overlap between areas where democracies
cluster and regions where stable peace prevails. Proofs. From Alex Mintz and Bruce Russett, editors, New Directions for International Relations: Confronting the Method-of-Analysis Problem, Lexington Books, Lexington, MA.
Analyzing Dependencies in International Relations
Peter D. Hoff & Michael D. Ward
February 2005
Abstract. Using data over the period from 1950-2000 we estimate a gravity model of bilateral international trade to explore the linkages between a) alliances, b) joint memberships in international institutions, c) mutual cooperation and d) conflict, e) mutual economic freedom and f) democracy and bilateral trade. Our novel contribution to the literature is the incorporation of exporter and importer effects as well as reciprocity into a gravity model of international trade. We cross-validate this model against out-of-sample data on an annual basis. The resulting, empirical findings are substantially stronger than typically obtained from gravity model estimations. Second and third order dependencies in bilateral trade are important; models which do not include these effects may lead to misleading results. Military alliances, membership in IGOs, and mutual economic freedom are shown to enhance bilateral trade. Further, speculation about the positive synergy between international cooperation and trade is supported by our results. On the other hand, conflict and the level of democracy do not appear to have strong, discernable effects on bilateral trade. We speculate about the global implications of freer trade among countries without latent tendencies to have high levels of bilateral commerce.
Draft (PDF) 3.4MB ; Code; Data