Gary King Research Articles

Do people in authoritarian countries have lower standards when evaluating their governments? An anchoring vignettes approach

Why do people in authoritarian countries think more positively of their governments than people in democratic countries? Existing research suggests three explanations: (1) people in authoritarian countries lie; (2) people in authoritarian countries are indoctrinated; and (3) authoritarian governments have better performance than their democratic counterparts. In this study, I explore a fourth explanation – people in authoritarian countries apply lower standards. To test it, I apply the anchoring vignettes method developed by Gary King and others to original data from China, Vietnam, Russia, Mexico, and the United States, and from the cities of Beijing and Taipei. Adding a case study of Taiwan’s economic trajectory as a robustness check, I conclude that people in authoritarian countries tend to use lower standards when reporting political trust and government responsiveness, but the lower standards are likely to be caused by fast economic growth rather than authoritarianism.

Biodiversity, Traditional Knowledge and Patent Rights: The Case Study of Phyllomedusa bicolor

Abstract This article intends to analyze the possible structure of the appropriation of traditional knowledge through the patent rights system. The research question is whether the international legislation on patents and biodiversity may be contributing to the process of appropriating traditional knowledge and natural genetic resources associated with Brazilian biodiversity from the Amazon region. The aim is to cross-reference the empirical facts concerning the use of traditional knowledge and the effects of the legal treatment of patent rights. The theory to be constructed is that the international regulations on patent rights and biodiversity contribute to the process of appropriating traditional knowledge associated with Brazilian biodiversity. Therefore, empirical qualitative research is conducted, utilizing the rules of inference developed by Lee Epstein and Gary King and the grounded theory methodology. Finally, the case of Phyllomedusa bicolor (Kampô frog) was selected and studied to extract empirical data which can be used to enunciate the theory of appropriation of natural genetic resources and associated traditional knowledge from the South by the North.

Comments on “The Troubled Future of College and Universities”

I share Gary King and Maya Sen's view that applying the knowledge, approaches, and tools of the social and policy sciences might help us be more intelligent about shaping the future of the American system of higher education and its component parts. I have even been surprised, in turning from the scholarly field of my training and academic field to a professional preoccupation with higher education, to search in vain for sessions at APSA meetings in which my fellow political scientists might focus their analytical eyes on our own sector and institutions. Taking more systematic analytical approaches might certainly be a good alternative to riding the most recent serious source of anxiety—and there have been many in recent decades—or mining a particular strand of data in search of indicators that might serve as tea leaves or life rafts, also known in the trade as “benchmarks.”

What is R2 all about?

Which substantive meaning we can draw from the R2 statistic? The coefficient of determination is defined as the sum of squares due to the regression divided by the sum of total squares. Usually, R2 is interpreted as representing the percentage of variation in the dependent variable explained by variation in the independent variables. This definition is found by both econometrics and statistics handbooks and is widely accepted among quantitative scholars. However, this view is challenged by one of the most distinguished political science methodologists. Gary King argues that the R2 is highly misused as a measure of the influence of X on Y. This paper analyzes the R2 statistic using a non-technical approach. Our purpose is to provide an intuitive understanding of the coefficient of determination and its major shortcomings.

QnAs with Gary King

Gary King, a professor of social science at Harvard University and a member of the National Academy of Sciences, fashions tools that harness the power of statistics, machine learning, and informatics to make sense of the numbers that matter to society. From evaluating the efficacy of a Mexican health policy reform to predicting the fate of the US Social Security trust fund, King’s sophisticated number crunching has important practical implications for disciplines as diverse as social, political, and health sciences. King tells PNAS how quantitative social science research can extend from academic journals into real-world scenarios. > PNAS:You designed a health policy experiment to evaluate the Mexican health care program, Seguro Popular , which was aimed at reducing health care costs for the poor. What was new about this experiment, and what did you find? > King:Most large-scale public policy experiments fail mainly because study participants prefer to be in the treatment group rather than in the control group. Also, there is often political opposition to the random assignment of people to different groups because that’s not how government projects are typically executed. After all, the job of politicians is to try to ensure that their constituents benefit from policy measures, not to conduct scientific research. However, breaking the randomization of subject assignment …

Multiple Imputations Applied to the DREAM3 Phosphoproteomics Challenge: A Winning Strategy

DREAM is an initiative that allows researchers to assess how well their methods or approaches can describe and predict networks of interacting molecules [1]. Each year, recently acquired datasets are released to predictors ahead of publication. Researchers typically have about three months to predict the masked data or network of interactions, using any predictive method. Predictions are assessed prior to an annual conference where the best predictions are unveiled and discussed. Here we present the strategy we used to make a winning prediction for the DREAM3 phosphoproteomics challenge. We used Amelia II, a multiple imputation software method developed by Gary King, James Honaker and Matthew Blackwell[2] in the context of social sciences to predict the 476 out of 4624 measurements that had been masked for the challenge. To chose the best possible multiple imputation parameters to apply for the challenge, we evaluated how transforming the data and varying the imputation parameters affected the ability to predict additionally masked data. We discuss the accuracy of our findings and show that multiple imputations applied to this dataset is a powerful method to accurately estimate the missing data. We postulate that multiple imputations methods might become an integral part of experimental design as a mean to achieve cost savings in experimental design or to increase the quantity of samples that could be handled for a given cost.

The State of the Field: Political Science and Chinese Political Studies

The Journal of Chinese Political Science (JCPS) is publishing a series of special issues, containing retrospective and prospective papers on the state of Chinese political studies, past, present and future. We call upon China scholars worldwide to contribute to these important special issues which will have a significant impact on the intellectual development of the field. Our goal is to disseminate scholarship that can shed light on the ever changing field of Chinese political studies, provide valuable material for scholar teachers and researchers of Chinese politics, and highlight the achievements, problems and issues, and directions of this field. We have witnessed the substantial transformation of China studies, particularly Chinese political studies, in the past 30 years due to changes in China and its rising status in the world as well as changes in our ways of conducting research. As area studies specialists, we are no longer “isolated” from the larger disciplines of Political Science and International Relations (IR) but an integral part of them. A simple survey of the JCPS and other journals reveals that we have applied theories and methods from these larger disciplines to Chinese political studies and that we have sought to meet the standards of scientific research and theoretical relevance in the two disciplines. The scholarship we produce today has advanced far beyond the days of classic Sinology. However, theoretical and methodological approaches in Chinese political studies are very diverse, and substantive issues in various Chinese contexts are being studied from a broader comparative perspective. As Lynn T. White points out in his paper, “Chinese and Asian comparative research can help reverse the methodological narrowness that obscures much actual politics in the world from being studied circumspectly.” Scholars and students in our field still vividly remember texts used in graduate school and innumerable papers published by first rate comparativists such as Sydney Verba, Arendt Lijphart, David Collier, James Mahon, Giovanni Satori, Gary King, David Easton, and many others. All of them have had a great impact on how we J OF CHIN POLIT SCI (2009) 14:225–227 DOI 10.1007/s11366-009-9067-5

Seguro Popular in Mexico: is premature evaluation healthy?

In The Lancet today, Gary King and colleagues1 present an innovative cluster-randomised trial of health insurance—Seguro Popular—in Mexico. Rigorous research on health-systems issues is much needed,2 and the investigators break new ground with one of the largest randomised health-policy experiments ever. The main finding was a reduction in catastrophic expenditures on health of 1·9%. Such expenditures were defined as health-spending share of a household's spending capacity after a minimum food budget; expenditures were defined as catastrophic if a household's health spending exceeded 30% of that capacity.

Election Strategy and Ethnic Politics in Singapore

According to Minister Mentor Lee Kuan Yew, Singapore’s electoral system is essentially representative and does not suffer from significant ethnic conflict. Opposition leaders, however, denounce legislative elections as unfair and claim that Singapore's ethnic minorities disagree politically with the Chinese-dominated People's Action Party (PAP). This essay aims to test both of these hypotheses empirically, using freely available electoral and public-opinion data. Logistic regression of the 1968-2006 parliamentary election results by constituency indicates that the PAP government did create Group Representation Constituencies in 1988 so as to eliminate districts that had voted disproportionately for the opposition in 1984. Analysis using Gary King's method of ecological inference suggests that ethic polarization between Chinese and Malays was moderately high in the 1976 election, peaked in 1988, and was minimal in 2006. Indians, meanwhile, appear to have voted with the Chinese in all three elections. A parallel cross-sectional, logistic regression of the 2002 Singapore subset of the World Values Survey, however, has Indian respondents being slightly less likely to admit to dissatisfaction with the government and indicates that being Malay does not make one more willing to express such dissatisfaction. These empirical results thus cast doubt on the extent to which Singapore's elections have been truly free, fair, and devoid of ethnic tension. The findings also suggest that young, middle-class, highly educated Chinese have replaced working-class Malays as the greatest challenge to continued PAP dominance.

Using Graphs Instead of Tables in Political Science

When political scientists present empirical results, they are much more likely to use tables than graphs, despite the fact that graphs greatly increases the clarity of presentation and makes it easier for a reader to understand the data being used and to draw clear and correct inferences. Using a sample of leading journals, we document this tendency and suggest reasons why researchers prefer tables. We argue that the extra work required in producing graphs is rewarded by greatly enhanced presentation and communication of empirical results. We illustrate their benefits by turning several published tables into graphs, including tables that present descriptive data and regression results. We show that regression graphs emphasize point estimates and confidence intervals and that they can successfully present the results of regression models. A move away from tables towards graphs would improve the discipline's communicative output and make empirical findings more accessible to every type of audience.Jonathan P. Kastellec (jpk2004@columbia.edu) and Eduardo L. Leoni (eleoni@hmdc.harvard.edu) are Doctoral Candidates in Political Science at Columbia University. The authors' names appear in alphabetical order. They would like to thank Andrew Gelman, Rebecca Weitz-Shapiro, Gary King, David Epstein, Jeff Gill, Piero Stanig, and three anonymous reviewers for helpful comments and suggestions. We also thank Noah Kaplan, David Park, and Travis Ridout for generously making their data publicly available. Eduardo Leoni is grateful for support from the Harvard MIT Data Center, where he was a fellow while working on this project.We have created a web site, http://tables2graphs.com, that contains complete replication code for all the graphs that appear in this article, as well as additional graphs that we did not present due to space limitations.

Replication Data Sets and Favored-Hypothesis Bias

Jeremy Freese makes the case for data sharing as a condition of publication for quantitative research in sociology, and Gary King tells us of a Dataverse Network under construction that is designed to routinize the process of posting and storing such data sets. No matter how user-friendly that network turns out to be, it is clear that no system is entirely cost-free, either for researchers or for journal editors. It is important, then, to determine whether the benefits of mandatory data sharing (or ``data relinquishment,'' as Herrnson calls it) would outweigh the costs. In this comment, the author discusses the issue from his vantage point as a former editor and concludes that the benefits of such a requirement most likely would exceed the costs.

Introduction to the Special Section on Replication and Data Access

Sociological research proceeds to a great degree on trust—trust that investigators have truthfully reported their analyses and trust that they have actually done those analyses correctly. The question is whether this trust is warranted. The answer is that for the most part we do not know. Except in the case where analyses are based on publicly available data sets, it may be difficult if not impossible to replicate what another researcher has done. Even then, without the code that produced the results, it may be difficult to know what the original research actually did. (I spent three months attempting to replicate some results in The Bell Curve, only to be told by Charles Murray that the book inaccurately described the analyses that had been done!) Furthermore, and this is only an impression, sociology journals appear to have little interest in publishing studies that primarily carry out replications or straightforward extensions of existing analyses. Reviewers who are sociologists are too apt to demand that any report must make a theoretical contribution. The contrast with economics, for example, where hundreds if not thousands of articles have been published that examine estimates of the returns to education, is striking. The articles in this issue address the question of access to data. The issue of whether there is too little interest by sociologists and/or sociology journals in publishing articles that primarily take up replications will have to be dealt with at another time. In the first article, Jeremy Freese lays out the argument for journals requiring authors to provide far greater access than they do now. He also addresses many of the past criticisms of this recommendation. In the second article, Gary King describes a relatively simple process by which authors can archive their data and code so that it is permanently available. The last two articles are comments by Glenn Firebaugh, a former editor of the American Sociological Review, and Andrew Abbott, the current editor of the American Journal of Sociology. These journals are generally regarded as the top two journals in the

Ranking Doctoral Programs by Placement: A New Method

While many bemoan the increasingly large role rankings play in American higher education, their prominence and importance are indisputable. Such rankings have many different audiences, ranging from prospective undergraduates or graduate students, to foundations and government funders, to university administrators identifying strengths and weaknesses of their school. This diverse audience necessarily has varying hopes for what “quality” is measured in school rankings, and different uses for the rankings themselves. But although there are currently a wide variety of ways to assess graduate school quality, most existing surveys have recognized failings that compromise their usefulness to at least one of these different constituencies. The authors extend their thanks to William Bowen, Derek Bruff, Jonathan Cole, Philip Katz, Gary King, Robert Townsend, Harriet Zuckerman, and two anonymous PS reviewers for their valuable comments on and criticisms of earlier drafts of this paper.

No Easy Choices: Estimating the Effects of United Nations Peacekeeping (Response to King and Zeng)

In a study published in the American Political Science Review, we argued that United Nations peace operations have had a positive effect on peacebuilding processes after civil war and presented empirical evidence in support of our claim. Gary King and Langche Zeng have criticized our conclusions for being dependent on indefensible theoretical assumptions and unsupported by the empirical evidence. We show that their critique is based on a misinterpretation of our arguments, that our results are robust to alternative econometric assumptions, that their diagnostic tools cannot be meaningfully applied to our data and that they cannot be used to establish the validity of counterfactual inference.

Officers King and Zeng and the Case of the Unsupported Counterfactual

Let me begin by praising Gary King and Langche Zeng for their service to the field with the methods they describe in their paper (King and Zeng 2006a). It is important to know when counterfactual statements drawn from statistical estimates wander far from the data used in the estimates. I particularly like the subtlety of the observation that even when all the values of the individual variables occur in the data, it may not be the case that all combinations of those values do. There are countries that protect the civil liberties of their citizens and those that abuse them and there are democracies and autocracies, but there are not very many autocracies that protect civil liberties or democracies that abuse them. Consequently, our inferences about autocracies with civil liberties and democracies without them are not based on the same depth of data that our inferences about democracies with civil liberties and autocracies without them are. I am looking forward to using their methods to examine some of my own recent data analysis on compliance with the laws of war.' For example, U.S. military dominance characterized the Kosovo bombing and the Iraq War, but how different were these cases from the record of earlier twentieth century wars? Like Shakespeare's Marc Anthony, however, one does not get asked to contribute to such fora to praise Caesar but to bury him. While I have neither the desire, the ability, nor the intent to bury Professors King and Zeng, I do have a clod or two of dirt to toss in their general direction. My qualifications to their excellent and important article concern theory as the source of counterfactual conclusions and the range of evidence we can use to judge among competing theories. Their paper seems to assume that theory is fully captured by a well-specified regression equation, including functional form of the relationship, as judged by its ability to represent a given data set. It is as if they draw their epistemological inspiration from two fictional police officers: Captain Louis Renault of the movie Casablanca, Round up the usual suspects!, and Sergeant Joe Friday from the TV series Dragnet, 'Just the facts, ma'am. At times, I get the impression that theory plays no role in their process beyond providing a list of exogenous variables-the usual suspects-along with the appropriate specification of functional form and link function between those variables and the dependent variable, which is then applied to just the facts. But theories are more than just a list of relevant variables and a functional form; they are logical structures that provide arguments about why the variables cause one another. There are other forms of investigation than those of Captain Renault and Sergeant Friday. Theory is critical for counterfactuals because theories explain why the counterfactual should hold. Reasons why the counterfactual should be true are as

Deracialization or Racialization: The Making of a Black Mayor in Jackson, Mississippi1

In 1997, Harvey Johnson made history by becoming the first black to be elected as mayor of Jackson, Mississippi. Four years prior, Johnson failed to make the Democratic primary runoffs. The following research question is addressed: what explains Johnson’s victory in 1997, when compared to his defeat in 1993? In answering this question, I rely heavily on elite‐level interviewing and newspaper accounts. I also examine the black and white vote for Johnson, using aggregate‐level election and census data. Gary King’s Ecological Inference technique is used to examine these data. Based on the findings, I conclude that Johnson ran a deracialized campaign in 1993 and a racialized campaign in 1997. The deracialized campaign failed to attract white voters and attenuated his black support. In 1997, however, he was able to almost double his support from black voters. This increase can be attributed in part to his racialized campaign strategy. Lord have mercy! We’re going to have a black mayor in Jackson, Mississippi. —An anonymous black woman quoted in Chappell (1997, 76)

The Indigenous Vote in Ecuador's 2002 Presidential Election

The Ecuadorian Indian movement entered electoral politics in 1996 through the Pachakutik political party. For the 2002 presidential and congressional elections, Pachakutik made an alliance with ex-colonel and 2000 coup leader Lucio Gutiérrez. After a first-place finish in the first round, Gutiérrez went on to win the presidency in a run-off. This research seeks to determine whether the Indian movement, via Pachakutik, was able to transform its constituency into an effective indigenous voting bloc. The analysis of the 2002 electoral returns applies the Ecological Inference method developed by Gary King. The results show that while indigenous voters were more likely to support Gutiérrez and Pachakutik congressional candidates than non-indigenous voters, large proportions of the indigenous electorate apparently did not follow the well-organized and relatively unified movement leadership.

Decentralization: Fueling the Fire or Dampening the Flames of Ethnic Conflict and Secessionism?

Political decentralization is widely believed to reduce ethnic conflict and secessionism in the world today. Yet decentralization is more successful in reducing conflict and secessionism in some countries than in others. In this article, I explore why this difference occurs. I demonstrate using a statistical analysis of thirty democracies from 1985 to 2000 that decentralization may decrease ethnic conflict and secessionism directly by bringing the government closer to the people and increasing opportunities to participate in government, but that decentralization increases ethnic conflict and secessionism indirectly by encouraging the growth of regional parties. Regional parties increase ethnic conflict and secessionism by reinforcing ethnic and regional identities, producing legislation that favors certain groups over others, and mobilizing groups to engage in ethnic conflict and secessionism.Earlier versions of this article were presented at Harvard University and the Center for the Study of Democratic Politics at Princeton University. The author would like to thank Sandra Alfonso-Leon, James Alt, Micah Altman, Barry Friedman, Shigeo Hirano, Simon Hug, Gary King, Rose Rozaghian, Tulia Falleti, and two anonymous reviewers for their very helpful comments.

Consequences of Positivism

The aim of this article is to highlight the baleful consequences of positivism in contemporary political science. The author focuses on Gary King, Robert Keohane, and Sidney Verba's Designing Social Inquirybecause it clearly and consistently articulates positivist commitments that tacitly and so often inconsistently, animate much work in the discipline. The author first establishes the positivist commitments that animate King, Keohane, and Verba's argument and then shows that precisely because of those commitments, their theory of inquiry generates two highly regrettable consequences: It hinders our ability to make sense of successful quantitative analysis generally and of causal explanation in particular. More diffusely, but no less important given King, Keohane, andVerba's practical aims, their theory of inquiry is self-defeating in its efforts to impart intellectual unity to the discipline of political science.

Case Selection: Insights fromRethinking Social Inquiry

This book, responding to the stimulus of the volume by Gary King, Robert Keohane, and Sidney Verba, helps to carry forward King et al.'s original purpose—to bridge quantitative and qualitative perspectives and find the unity of method underlying both of them. Whether or not we accept King et al.'s own vision of the underlying unity, it is clear that their work has been a catalyst for progress in this direction.