Improving Data Quality: Actors, Incentives, and Capabilities

Devesh Kapur Centre for Advanced Study of India, University of Pennsylvania, 3600 Market Street, Suite 560, Philadelphia, PA 19104 e-mail: dkapur{at}sas.upenn.edu e-mail: herrera{at}fas.harvard.edu (corresponding author) This paper examines the construction and use of data sets inpolitical science. We focus on three interrelated questions:How might we assess data quality? What factors shape data quality?and How can these factors be addressed to improve data quality?We first outline some problems with existing data set quality,including issues of validity, coverage, and accuracy, and wediscuss some ways of identifying problems as well as some consequencesof data quality problems. The core of the paper addresses thesecond question by analyzing the incentives and capabilitiesfacing four key actors in a data supply chain: respondents,data collection agencies (including state bureaucracies andprivate organizations), international organizations, and finally,academic scholars. We conclude by making some suggestions forimproving the use and construction of data sets. Authors' note: For generous comments at many stages in the paper,the authors would like to thank Dawn Brancati, Bear Braumoeller,Kanchan Chandra, Jorge Dominguez, Errol D'Souza, Richard Grossman,Ana Grzymala-Busse, Andrew Kydd, David Laitin, Daniel Posner,Jasjeet Sekhon, Hillel Soifer, Jessica Wallack, and Steven Wilkinsonand the Comparative Politics Research Workshop at Harvard University,and the anonymous reviewers from Political Analysis. The authorstake full responsibility for any errors. An earlier versionof this paper was presented at the American Political ScienceAssociation Annual Meetings, Boston, MA, August 2002.     

Open Data for Science,Policy, and the Public Good

Supporters of open data believe that free and complete access to research data is beneficial for science, public policy, and society. In environmental science and policy, open data systems can enable relevant research and inform evidence‐based governmental decisions. This article examines the unlikely case of Brazil's National Institute for Space Research's transition toward an open data model. Considering Brazil's young democracy, incipient practice of government transparency and accountability, and lacking a tradition of science‐policy dialogue, this case is a striking example of how open data can support public debate by making information about forest cover widely available. The case shows the benefits and challenges of developing such open data systems, and highlights the various forms of accessibility involved in making data available to the public.     

Economics, Party, and the Vote: Causality Issues and Panel Data

Conventional wisdom argues that national economic perceptions generally have an important impact on the vote choice in democracies. Recently, a revisionist view has arisen, contending that this link, regularly observed in election surveys, is mostly spurious. According to the argument, partisanship distorts economic perception, thereby substantially exaggerating the real vote connection. These causality issues have not been much investigated empirically, despite their critical importance. Utilizing primarily American, and secondarily British and Canadian, election panel surveys, we confront directly questions of the time dynamic and independent variable exogeneity. We find, after all, economics clearly matters for the vote. Indeed, once these causality concerns are properly taken into account, the impact of economic perceptions emerges as larger than previously thought. As well, the actual impact of partisanship is clearly reduced.     

数据开放、公众参与和环境治理创新

随着网络信息技术的发展与大数据的产生,公众对于政府各类数据开放共享的要求日趋强烈,同时政府也可以通过数据开放共享推动公共治理创新。研究表明,政府通过环境数据开放可以让公众更好地参与环境治理,从而推动环境治理创新;环境数据开放使公众能够获得更多的环境信息,与政府进行有效沟通并形成良性互动,从而提升环境治理效果。针对数据开放与环境治理创新问题,提出了相应的政策建议:改革政府环境数据管理体制,加强政府与市场的合作;加强环境数据开放平台建设,打造"一站式"环境数据服务;加大宣传力度,调动公众利用开放环境数据的积极性。     

A Fast, Easy, and Efficient Estimator for Multiparty Electoral Data

Katz and King have previously developed a model for predictingor explaining aggregate electoral results in multiparty democracies.Their model is, in principle, analogous to what least-squaresregression provides American political researchers in that two-partysystem. Katz and King applied their model to three-party electionsin England and revealed a variety of new features of incumbencyadvantage and sources of party support. Although the mathematicsof their statistical model covers any number of political parties,it is computationally demanding, and hence slow and numericallyimprecise, with more than three parties. In this paper we producean approximate method that works in practice with many partieswithout making too many theoretical compromises. Our approachis to treat the problem as one of missing data. This allowsus to use a modification of the fast EMis algorithm of King,Honaker, Joseph, and Scheve and to provide easy-to-use software,while retaining the attractive features of the Katz and Kingmodel, such as the t distribution and explicit models for uncontestedseats.     

The Big (Data) Bang: Policy,Prospects, and Challenges

Big data is increasingly the cornerstone on which policy making is based. However, with potential benefits and applications come challenges and dilemmas. In this set of symposium articles, authors examine the promise and problems of big data, exploring associated prospects, risks, parameters, and payoffs from a variety of perspectives. The articles address myriad challenges in the handling of big data sets, such as collection, validation, integrity, and security; ontological issues attending data analytics and conceptual transformations; the foundations of big data collection for social science research; the gap between the acquisition of data and its use to advance discovery and innovation; the costs and benefits of using big data in decision making and analysis; and, finally, related problems of privacy, security, and ethics. Issues such as these will continue to arise with increased use of big data as fundamental to policy making and governance in today's growing information society.     

大数据、权利实现与基层治理创新

治理的本质是实现权利。以权利为媒介,大数据与治理形成一种间接的因果关系,干预并驱动了基层治理的创新。依循"合作+冲突"的基本逻辑,按照权利的合作、权力的冲突、权利联合权力的三种途径,大数据从价值、技术和资源等三个维度,驱动着公民权利的实现。其中,权利合作表现为一种"差序格局",权力冲突表现为制衡与监察,"合作"权利与"冲突"权力的合作则正在成为一项基本选择。而要提升大数据驱动基层治理创新的效率,需要推动基层政务大数据的开放、共享,完善大数据治理的配套法律法规,培育一个成熟的公民群体。     

Considering Knowledge Uptake within a Cycle of Transforming Data,Information, and Knowledge

Knowledge uptake, having decision makers assimilate the ideas of experts, is recognized as an important stimulus to bringing about policy change. This is particularly true in the realm of environmental policymaking, which is characterized by knowledge intensity, complexity, and multifaceted concerns. Using examples from an innovative watershed management organization, this article presents a heuristic for understanding how knowledge uptake occurs within a cycle of organizational reasoning. This cycle is driven by activities that transform data, information, and knowledge and that link specialists with decision makers. The heuristic can be used as a diagnostic tool to identify breaks in the transformation process that impede mandate fulfillment and impair capacity building. Lack of appreciation of the dynamic relationship between data, information and knowledge leads to mistimed and ineffective policy interventions that do not result in the hoped for progress in science intended to underpin policy advances.     

Big Data for Policy Analysis: The Good,The Bad,and The Ugly

Big data holds tremendous potential for public policy analysis. At the same time, its use prompts a number of issues related to statistical bias, privacy, equity, and governance, among others. Accordingly, there is a need to formulate, evaluate, and implement policies that not only mitigate the risks, but also maximize the benefits of using big data for policy analysis. This poses a number of challenges, which are highlighted in this essay.     

Trade‐Offs,Limitations, and Promises of Big Data in Social Science Research

The promise of “big data” is a grand one. The collection and aggregation of massive datasets and the development of analytical tools by which to study these data is part of cutting‐edge efforts across scientific disciplines, with social, behavioral, and economic sciences leading the way in many of these efforts. There has been a recent rise in National Science Foundation funding for “big data” research across directorates that coincides with scholarly, public, and governmental attention to the topic. Nonetheless, there are limitations and trade‐offs to “big data” research, particularly as it corresponds to scientific modes of inquiry and the limited range of topics that fall under its umbrella, that must be recognized and incorporated into the general understanding of its long‐term promise.     

Inter-Party Competition,Intra-Party Competition,and Distributive Policy: A Model and Test Using New Deal Data

This paper presents a theoreticalmodel examining the influence of two stylized types ofvoters: ``loyal voters'' and ``swing voters''. The model showswhy both types of voters will influence thedistribution of benefits by a reelection-seekingincumbent, and it predicts how their influence willvary with the importance of the general electionrelative to that of the primary: closer competitionbetween parties in the general election, ceterisparibus, increases the influence of swing votersrelative to that of loyal voters. County-level dataon the allocation of money and jobs by New Deal reliefprograms confirm the model's predictions.     

大数据优化政府决策的机理、风险及规避

大数据对政府决策的优化有明显的过程导向,即在经验干预下,通过对数据的收集、整理、挖掘与利用形成数字化平台,利用数字化平台的丰富信息优化和纠偏决策方案,同时对决策过程进行监督,进而推动决策问题构建的民主化、决策方案制订的科学化、决策方案执行的有效性与决策效果评估的规范化。但是,受数据结构异质性、数据价值稀疏性、数据内容隐私性和数据管理人为性的影响,大数据在给政府决策带来福利的同时,也带来一定安全、道德和技术风险。为规避这些风险,明确数据使用规则、设立数据使用标准、强化数据使用监管、打破数据使用垄断和追究数据滥用责任等就显得尤为必要。     

Data Analysis and Data Visualization as Active Learning in Political Science

While data analysis and the related skills of data management and data visualization are important skills for undergraduates in the field of political science, the process of learning these skills can also be used to develop critical thinking, encourage active and collaborative learning, and to apply knowledge gained in the classroom. Drawing on our experiences using data work in upper-level courses in International Relations and American Politics, we discuss how data work and quantitative analysis can be incorporated into subject-based (i.e., nonmethods specific) courses, and how it can also enhance critical reasoning skills. An evaluation of this approach using direct and indirect assessment is included.     

Big Data,intelligence, and analyst privacy: investigating information dissemination at an NSA-funded research lab

The Laboratory for Analytic Sciences (LAS) at North Carolina State University, funded by the National Security Agency, is a collaborative, long-term research enterprise focused on improving intelligence analysis using Big Data. In its work, LAS has recently begun dealing with the trade-off between the collection, storage, and use of large unclassified data-sets and analyst privacy. We discuss particular privacy challenges at LAS, analyze privacy principles in the life cycle of LAS unclassified data-sets, what intelligence analysts themselves think about these privacy concerns, and recommend possible best practices potentially applicable to LAS, as well as future Big Data laboratories and research centers that collaborate with intelligence communities.     

Big Data and Strategic Intelligence

This article examines the intersection of Big Data and strategic intelligence from a theoretical-conceptual viewpoint. Adopting Popperian refutation as a starting point, it approaches methodological issues surrounding the incorporation of Big Data into the intelligence cycle, and argues that Big Data analytics is best used to discern long-term developments, generate intelligence hypotheses, and adduce refuting facts. The article then briefly examines the use of Big Data via social media, an increasingly fertile platform for intelligence analysis. Finally, the article argues that despite its potential in filling our epistemic gaps, Big Data should continue to complement traditional subject-matter expertise, supported by game theory, as part of a tripartite analytical framework for strategic intelligence consisting of ‘subtext’, ‘context’ and ‘metatext’. In this respect, Big Data may well become the midwife for more open modes of intelligence management and, ultimately, a more open society.     

State Agencies’ Use of Administrative Data for Improved Practice: Needs,Challenges, and Opportunities

Growing interest in the use of administrative data to answer questions around program implementation and effectiveness has led to greater discussion of how government agencies can develop the necessary internal data infrastructure, analytic capacity, and office culture. However, there is a need for more systematic research into how states find different pathways and strategies to build administrative data capacity. Drawing on interviews with almost 100 human service agency staff and their data partners, the authors examine the realities of administrative data use. They summarize the experiences of data users in order to address two main challenges: limited analytic capacity and challenges to linking or sharing data resources. The article concludes by examining a range of approaches that government agencies take to improve data quality and capacity to analyze that data.     

技术执行视域中的大数据反腐：情境、结构与绩效

大数据反腐的本质是执行大数据技术,提升权力监督的制度绩效。从技术执行视角分析,大数据反腐技术执行的要素包括情境、结构和绩效。大数据反腐技术执行的情境包括大数据技术发展的客观情势、反腐信息不对称克服的内在需求以及反腐进入攻坚期的策略选择。大数据反腐技术执行的结构包括组织结构、制度安排以及相应的主体行动介入。大数据反腐技术执行的绩效主要体现在堵塞监督漏洞、规范政府行为、精准追查腐败、有效预防腐败。优化大数据反腐技术执行,需要推进标准化设计,提升大数据反腐技术的制度嵌入性;兼顾区域和系统,打破反腐数据资源共享的技术壁垒;规范使用范围,防控大数据反腐技术应用的风险。     

Means,Motive, and Opportunity – Local Government Data Distortion in a High‐Stakes Environment

Regulatory authorities are increasingly relying upon performance data for developing public policy. However, this reliance necessarily assumes that the data are free from material distortion. This paper provides a conceptual framework for understanding the ‘means’, ‘motive’, and ‘opportunity’ for distorting data employed in high‐stakes performance‐management programmes. We present empirical evidence which suggests that the use of data drawn entirely from financial statements by no means guarantees a distortion‐free depiction of performance. In addition, we provide econometric evidence of some important determinants of performance data distortion. Taken as a whole, the following analysis provides a comprehensive picture of the salient matters which must be addressed to ensure accurate data for public policy‐making purposes.     

Evaluating the Concept of Location Affordability: Recent Data on the Relationship Between Transportation,Housing, and Urban Form

The concept of location affordability holds that housing affordability should be augmented to include transportation cost as a related and substantial household cost burden. The recent United States foreclosure crisis of 2006–2008 offers an opportunity to evaluate location affordability as a concept for policy and practice by investigating the relationship between transportation cost burdens and negative housing outcomes. This article contributes to the growing literature on location affordability and the recent crisis with an analysis of default and foreclosure data for 70 metropolitan areas. The analysis includes transportation and housing cost burdens and demographic data, as well as multidimensional measures of urban form. High rates of automobility are associated with increased foreclosure. The urban form variables yield mixed results, suggesting the relationship between urban sprawl and affordability is complex. However, across a range of specifications, high levels of development intensity are associated with increased foreclosure rates. The results have implications for both the housing and transportation sectors and lend support for the notion of location affordability.     

政治关联、地区差异与企业创新——基于上市公司的数据分析

当前已有一些文献探讨政治关联与企业创新的关系,但很少涉及政治关联所处的外部空间特征。那么,政治关联与企业创新的关系是否存在地区差异?通过对2008—2015年中国上市公司的实证分析发现,在西部地区,政治关联对于企业创新可能存在"雪中送炭"的推进作用;但在东部地区,政治关联对于企业创新可能只是"锦上添花"的作用,甚至可能产生"资源诅咒"的消极效应。具体来看,中央层级的政治关联对西部地区企业创新的积极效应比省、市、县等层级的政治关联更加显著。与民企相比,西部地区国企的政治关联对企业创新的积极效应更加显著。进一步剖析政治关联影响企业创新的作用机制,可以发现,在市场化水平较低、法制建设不健全的地区,建立和维持必要的政治关联有助于保护企业产权,为企业争取资源,进而促进企业创新。