We develop a computer-assisted method for the discovery of insightful conceptualizations, in the form of clusterings (i.e., partitions) of input objects. Each of the numerous fully automated methods of cluster analysis proposed in statistics, computer science, and biology optimize a different objective function. Almost all are well defined, but how to determine before the fact which one, if any, will partition a given set of objects in an "insightful" or "useful" way for a given user is unknown and difficult, if not logically impossible. We develop a metric space of partitions from all existing cluster analysis methods applied to a given data set (along with millions of other solutions we add based on combinations of existing clusterings), and enable a user to explore and interact with it, and quickly reveal or prompt useful or insightful conceptualizations. In addition, although uncommon in unsupervised learning problems, we offer and implement evaluation designs that make our computer-assisted approach vulnerable to being proven suboptimal in specific data types. We demonstrate that our approach facilitates more efficient and insightful discovery of useful information than either expert human coders or many existing fully automated methods.
Existing research on the extensive Chinese censorship organization uses observational methods with well-known limitations. We conducted the first large-scale experimental study of censorship by creating accounts on numerous social media sites, randomly submitting different texts, and observing from a worldwide network of computers which texts were censored and which were not. We also supplemented interviews with confidential sources by creating our own social media site, contracting with Chinese firms to install the same censoring technologies as existing sites, and—with their software, documentation, and even customer support—reverse-engineering how it all works. Our results offer rigorous support for the recent hypothesis that criticisms of the state, its leaders, and their policies are published, whereas posts about real-world events with collective action potential are censored.
The increasing availability of digitized text presents enormous opportunities for social scientists. Yet hand coding many blogs, speeches, government records, newspapers, or other sources of unstructured text is infeasible. Although computer scientists have methods for automated content analysis, most are optimized to classify individual documents, whereas social scientists instead want generalizations about the population of documents, such as the proportion in a given category. Unfortunately, even a method with a high percent of individual documents correctly classified can be hugely biased when estimating category proportions. By directly optimizing for this social science goal, we develop a method that gives approximately unbiased estimates of category proportions even when the optimal classifier performs poorly. We illustrate with diverse data sets, including the daily expressed opinions of thousands of people about the U.S. presidency. We also make available software that implements our methods and large corpora of text for further analysis.
This article led to the formation of Crimson Hexagon
This is a poster that describes our analysis of "partisan taunting," the explicit, public, and negative attacks on another political party or its members, usually using vitriolic and derogatory language. We first demonstrate that most projects that hand code text in the social sciences optimize with respect to the wrong criterion, resulting in large, unnecessary biases. We show how to fix this problem and then apply it to taunting. We find empirically that, unlike most claims in the press and the literature, taunting is not inexorably increasing; it appears instead to be a rational political strategy, most often used by those least likely to win by traditional means -- ideological extremists, out-party members when the president is unpopular, and minority party members. However, although taunting appears to be individually rational, it is collectively irrational: Constituents may resonate with one cutting taunt by their Senator, but they might not approve if he or she were devoting large amounts of time to this behavior rather than say trying to solve important national problems. We hope to partially rectify this situation by posting public rankings of Senatorial taunting behavior.
We present new statistical methods for evaluating information extraction systems. The methods were developed to evaluate a system used by political scientists to extract event information from news leads about international politics. The nature of this data presents two problems for evaluators: 1) the frequency distribution of event types in international event data is strongly skewed, so a random sample of newsleads will typically fail to contain any low frequency events. 2) Manual information extraction necessary to create evaluation sets is costly, and most effort is wasted coding high frequency categories . We present an evaluation scheme that overcomes these problems with considerably less manual effort than traditional methods, and also allows us to interpret an information extraction system as an estimator (in the statistical sense) and to estimate its bias.
We offer the first large scale, multiple source analysis of the outcome of what may be the most extensive effort to selectively censor human expression ever implemented. To do this, we have devised a system to locate, download, and analyze the content of millions of social media posts originating from nearly 1,400 different social media services all over China before the Chinese government is able to find, evaluate, and censor (i.e., remove from the Internet) the large subset they deem objectionable. Using modern computer-assisted text analytic methods that we adapt to and validate in the Chinese language, we compare the substantive content of posts censored to those not censored over time in each of 85 topic areas. Contrary to previous understandings, posts with negative, even vitriolic, criticism of the state, its leaders, and its policies are not more likely to be censored. Instead, we show that the censorship program is aimed at curtailing collective action by silencing comments that represent, reinforce, or spur social mobilization, regardless of content. Censorship is oriented toward attempting to forestall collective activities that are occurring now or may occur in the future --- and, as such, seem to clearly expose government intent.
A method of computerized content analysis that gives “approximately unbiased and statistically consistent estimates” of a distribution of elements of structured, unstructured, and partially structured source data among a set of categories. In one embodiment, this is done by analyzing a distribution of small set of individually-classified elements in a plurality of categories and then using the information determined from the analysis to extrapolate a distribution in a larger population set. This extrapolation is performed without constraining the distribution of the unlabeled elements to be equal to the distribution of labeled elements, nor constraining a content distribution of content of elements in the labeled set (e.g., a distribution of words used by elements in the labeled set) to be equal to a content distribution of elements in the unlabeled set. Not being constrained in these ways allows the estimation techniques described herein to provide distinct advantages over conventional aggregation techniques.
The (unheralded) first step in many applications of automated text analysis involves selecting keywords to choose documents from a large text corpus for further study. Although all substantive results depend on this choice, researchers usually pick keywords in ad hoc ways that are far from optimal and usually biased. Paradoxically, this often means that the validity of the most sophisticated text analysis methods depend in practice on the inadequate keyword counting or matching methods they are designed to replace. Improved methods of keyword selection would also be valuable in many other areas, such as following conversations that rapidly innovate language to evade authorities, seek political advantage, or express creativity; generic web searching; eDiscovery; look-alike modeling; intelligence analysis; and sentiment and topic analysis. We develop a computer-assisted (as opposed to fully automated) statistical approach that suggests keywords from available text without needing structured data as inputs. This framing poses the statistical problem in a new way, which leads to a widely applicable algorithm. Our specific approach is based on training classifiers, extracting information from (rather than correcting) their mistakes, and summarizing results with Boolean search strings. We illustrate how the technique works with analyses of English texts about the Boston Marathon Bombings, Chinese social media posts designed to evade censorship, among others.
Representative embodiments of a method for grouping participants in an activity include the steps of: (i) defining a grouping policy; (ii) storing, in a database, participant records that include a participant identifer, a characteristic associated With the participant, and/or an identifier for a participant’s handheld device; (iii) defining groupings based on the policy and characteristics of the participants relating to the policy and to the activity; and (iv) communicating the groupings to the handheld devices to establish the groups.
A method for selecting clusterings to classify a predetermined data set of numerical data comprises five steps. First, a plurality of known clustering methods are applied, one at a time, to the data set to generate clusterings for each method. Second, a metric space of clusterings is generated using a metric that measures the similarity between two clusterings. Third, the metric space is projected to a lower dimensional representation useful for visualization. Fourth, a “local cluster ensemble” method generates a clustering for each point in the lower dimensional space. Fifth, an animated visualization method uses the output of the local cluster ensemble method to display the lower dimensional space and to allow a user to move around and explore the space of clustering.
A vast literature demonstrates that voters around the world who benefit from their governments' discretionary spending cast ballots for the incumbent party in larger proportions than those not receiving funds. But surprisingly, and contrary to most theories of political accountability, the evidence seems to indicate that voters also reward incumbent parties for implementing ``programmatic'' spending legislation, passed with support from all major parties, and over which incumbents have no discretion. Why voters would attribute responsibility when none exists is unclear, as is why minority party legislators would approve of legislation that will cost them votes. We address this puzzle with one of the largest randomized social experiments ever, resulting in clear rejection of the claim that programmatic policies greatly increase voter support for incumbents. We also reanalyze the study cited as claiming the strongest support for the electoral effects of programmatic policies, which is also a very large scale randomized experiment. We show that its key results vanish after correcting either a simple coding error affecting only two observations or highly unconventional data analysis procedures (or both). We also discuss how these consistent empirical results from the only two probative experiments on this question may be reconciled with several observational and theoretical studies touching on similar questions in other contexts.
Methods for Observational Data
Evaluating Model Dependence
We discuss a method for improving causal inferences called "Coarsened Exact Matching'' (CEM), and the new "Monotonic Imbalance Bounding'' (MIB) class of matching methods from which CEM is derived. We summarize what is known about CEM and MIB, derive and illustrate several new desirable statistical properties of CEM, and then propose a variety of useful extensions. We show that CEM possesses a wide range of desirable statistical properties not available in most other matching methods, but is at the same time exceptionally easy to comprehend and use. We focus on the connection between theoretical properties and practical applications. We also make available easy-to-use open source software for R and Stata which implement all our suggestions.
This program is designed to improve causal inference via a method of matching that is widely applicable in observational data and easy to understand and use (if you understand how to draw a histogram, you will understand this method). The program implements the coarsened exact matching (CEM) algorithm, described below. CEM may be used alone or in combination with any existing matching method. This algorithm, and its statistical properties, are described in Iacus, King, and Porro (2008).
We introduce a new "Monotonic Imbalance Bounding" (MIB) class of matching methods for causal inference with a surprisingly large number of attractive statistical properties. MIB generalizes and extends in several new directions the only existing class, "Equal Percent Bias Reducing" (EPBR), which is designed to satisfy weaker properties and only in expectation. We also offer strategies to obtain specific members of the MIB class, and analyze in more detail a member of this class, called Coarsened Exact Matching, whose properties we analyze from this new perspective. We offer a variety of analytical results and numerical simulations that demonstrate how members of the MIB class can dramatically improve inferences relative to EPBR-based matching methods.
Although published works rarely include causal estimates from more than a few model specifications, authors usually choose the presented estimates from numerous trial runs readers never see. Given the often large variation in estimates across choices of control variables, functional forms, and other modeling assumptions, how can researchers ensure that the few estimates presented are accurate or representative? How do readers know that publications are not merely demonstrations that it is possible to find a specification that fits the author’s favorite hypothesis? And how do we evaluate or even define statistical properties like unbiasedness or mean squared error when no unique model or estimator even exists? Matching methods, which offer the promise of causal inference with fewer assumptions, constitute one possible way forward, but crucial results in this fast-growing methodological literature are often grossly misinterpreted. We explain how to avoid these misinterpretations and propose a unified approach that makes it possible for researchers to preprocess data with matching (such as with the easy-to-use software we offer) and then to apply the best parametric techniques they would have used anyway. This procedure makes parametric models produce more accurate and considerably less model-dependent causal inferences.
Matching is an increasingly popular method of causal inference in observational data, but following methodological best practices has proven difficult for applied researchers. We address this problem by providing a simple graphical approach for choosing among the numerous possible matching solutions generated by three methods: the venerable ``Mahalanobis Distance Matching'' (MDM), the commonly used ``Propensity Score Matching'' (PSM), and a newer approach called ``Coarsened Exact Matching'' (CEM). In the process of using our approach, we also discover that PSM often approximates random matching, both in many real applications and in data simulated by the processes that fit PSM theory. Moreover, contrary to conventional wisdom, random matching is not benign: it (and thus PSM) can often degrade inferences relative to not matching at all. We find that MDM and CEM do not have this problem, and in practice CEM usually outperforms the other two approaches. However, with our comparative graphical approach and easy-to-follow procedures, focus can be on choosing a matching solution for a particular application, which is what may improve inferences, rather than the particular method used to generate it.
To reduce model dependence and bias in causal inference, researchers usually use matching as a data preprocessing step, after which they apply whatever statistical model and uncertainty estimators they would have without matching. Unfortunately, this approach is appropriate in finite samples only under exact matching, which is usually infeasible, or approximate matching only under asymptotic theory if large enough sample sizes are available, but even then requires unfamiliar specialized point and variance estimators. Instead of attempting to change common practices, we show how those analyzing certain specific (but extremely common) types of data can instead appeal to a much easier version of existing theory. This alternative theory is substantively plausible, requires no asymptotic theory, and is simple to understand. Its core conceptualizes continuous variables as having natural breakpoints, which are common in applications (e.g., high school or college degrees in years of education, a governmental poverty level in income, or phase transitions in temperature). The theory allows binary, multicategory, and continuous treatment variables from the outset and straightforward extensions for imperfect treatment assignment and different versions of treatments.
We show that propensity score matching (PSM), an enormously popular method of preprocessing data for causal inference, often accomplishes the opposite of its intended goal -- increasing imbalance, inefficiency, model dependence, and bias. PSM supposedly makes it easier to find matches by projecting a large number of covariates to a scalar propensity score and applying a single model to produce an unbiased estimate. However, in observational analysis the data generation process is rarely known and so users typically try many models before choosing one to present. The weakness of PSM comes from its attempts to approximate a completely randomized experiment, rather than, as with other matching methods, a more efficient fully blocked randomized experiment. PSM is thus uniquely blind to the often large portion of imbalance that can be eliminated by approximating full blocking with other matching methods. Moreover, in data balanced enough to approximate complete randomization, either to begin with or after pruning some observations, PSM approximates random matching which, we show, increases imbalance even relative to the original data. Although these results suggest that researchers replace PSM with one of the other available methods when performing matching, propensity scores have many other productive uses.
We propose a simplified approach to matching for causal inference that simultaneously optimizes balance (similarity between the treated and control groups) and matched sample size. Existing approaches either fix the matched sample size and maximize balance or fix balance and maximize sample size, leaving analysts to settle for suboptimal solutions or attempt manual optimization by iteratively tweaking their matching method and rechecking balance. To jointly maximize balance and sample size, we introduce the matching frontier, the set of matching solutions with maximum possible balance for each sample size. Rather than iterating, researchers can choose matching solutions from the frontier for analysis in one step. We derive fast algorithms (about one million times faster than the best existing approach) that calculate the matching frontier, for several commonly used balance metrics. We demonstrate with analyses of the effect of sex on judging and job training programs that show how the methods we introduce can extract new knowledge from existing data sets.
Software that implements all our ideas is available at http://projects.iq.harvard.edu/frontier.
This is a poster presentation describing (1) the largest ever experimental study of media effects, with more than 50 cooperating traditional media sites, normally unavailable web site analytics, the text of hundreds of thousands of news articles, and tens of millions of social media posts, and (2) a design we used in preparation that attempts to anticipate experimental outcomes
We highlight common problems in the application of random treatment assignment in large scale program evaluation. Random assignment is the defining feature of modern experimental design. Yet, errors in design, implementation, and analysis often result in real world applications not benefiting from the advantages of randomization. The errors we highlight cover the control of variability, levels of randomization, size of treatment arms, and power to detect causal effects, as well as the many problems that commonly lead to post-treatment bias. We illustrate with an application to the Medicare Health Support evaluation, including recommendations for improving the design and analysis of this and other large scale randomized experiments.
We attempt to clarify, and suggest how to avoid, several serious misunderstandings about and fallacies of causal inference in experimental and observational research. These issues concern some of the most basic advantages and disadvantages of each basic research design. Problems include improper use of hypothesis tests for covariate balance between the treated and control groups, and the consequences of using randomization, blocking before randomization, and matching after treatment assignment to achieve covariate balance. Applied researchers in a wide range of scientific disciplines seem to fall prey to one or more of these fallacies, and as a result make suboptimal design or analysis choices. To clarify these points, we derive a new four-part decomposition of the key estimation errors in making causal inferences. We then show how this decomposition can help scholars from different experimental and observational research traditions better understand each other’s inferential problems and attempted solutions.
MatchingFrontier is an easy-to-use R Package for making optimal causal inferences from observational data. Despite their popularity, existing matching approaches leave researchers with two fundamental tensions. First, they are designed to maximize one metric (such as propensity score or Mahalanobis distance) but are judged against another for which they were not designed (such as L1 or differences in means). Second, they lack a principled solution to revealing the implicit bias-variance trade off: matching methods need to optimize with respect to both imbalance (between the treated and control groups) and the number of observations pruned, but existing approaches optimize with respect to only one; users then either ignore the other, or tweak it, usually suboptimally, by hand.
MatchingFrontier resolves both tensions by consolidating previous techniques into a single, optimal, and flexible approach. It calculates the matching solution with maximum balance for each possible sample size (N, N-1, N-2,...). It thus directly calculates the entire balance-sample size frontier, from which the user can easily choose one, several, or all subsamples from which to conduct their final analysis, given their own choice of imbalance metric and quantity of interest. MatchingFrontier solves the joint optimization problem in one run, automatically, without manual tweaking, and without iteration. Although for each subset size k, there exist a huge (N choose k) number of unique subsets, MatchingFrontier includes specially designed fast algorithms that give the optimal answer, usually in a few minutes.
MatchingFrontier implements the methods in this paper:
Does the U.S. Supreme Court curtail rights and liberties when the nation’s security is under threat? In hundreds of articles and books, and with renewed fervor since September 11, 2001, members of the legal community have warred over this question. Yet, not a single large-scale, quantitative study exists on the subject. Using the best data available on the causes and outcomes of every civil rights and liberties case decided by the Supreme Court over the past six decades and employing methods chosen and tuned especially for this problem, our analyses demonstrate that when crises threaten the nation’s security, the justices are substantially more likely to curtail rights and liberties than when peace prevails. Yet paradoxically, and in contradiction to virtually every theory of crisis jurisprudence, war appears to affect only cases that are unrelated to the war. For these cases, the effect of war and other international crises is so substantial, persistent, and consistent that it may surprise even those commentators who long have argued that the Court rallies around the flag in times of crisis. On the other hand, we find no evidence that cases most directly related to the war are affected. We attempt to explain this seemingly paradoxical evidence with one unifying conjecture: Instead of balancing rights and security in high stakes cases directly related to the war, the Justices retreat to ensuring the institutional checks of the democratic branches. Since rights-oriented and process-oriented dimensions seem to operate in different domains and at different times, and often suggest different outcomes, the predictive factors that work for cases unrelated to the war fail for cases related to the war. If this conjecture is correct, federal judges should consider giving less weight to legal principles outside of wartime but established during wartime, and attorneys should see it as their responsibility to distinguish cases along these lines.
Event Counts and Durations
We introduce a new framework for forecasting age-sex-country-cause-specific mortality rates that incorporates considerably more information, and thus has the potential to forecast much better, than any existing approach. Mortality forecasts are used in a wide variety of academic fields, and for global and national health policy making, medical and pharmaceutical research, and social security and retirement planning.
As it turns out, the tools we developed in pursuit of this goal also have broader statistical implications, in addition to their use for forecasting mortality or other variables with similar statistical properties. First, our methods make it possible to include different explanatory variables in a time series regression for each cross-section, while still borrowing strength from one regression to improve the estimation of all. Second, we show that many existing Bayesian (hierarchical and spatial) models with explanatory variables use prior densities that incorrectly formalize prior knowledge. Many demographers and public health researchers have fortuitously avoided this problem so prevalent in other fields by using prior knowledge only as an ex post check on empirical results, but this approach excludes considerable information from their models. We show how to incorporate this demographic knowledge into a model in a statistically appropriate way. Finally, we develop a set of tools useful for developing models with Bayesian priors in the presence of partial prior ignorance. This approach also provides many of the attractive features claimed by the empirical Bayes approach, but fully within the standard Bayesian theory of inference.
Duration of Parliamentary Governments
Although not widely known until much later, Al Gore received 202 more votes than George W. Bush on election day in Florida. George W. Bush is president because he overcame his election day deficit with overseas absentee ballots that arrived and were counted after election day. In the final official tally, Bush received 537 more votes than Gore. These numbers are taken from the official results released by the Florida Secretary of State's office and so do not reflect overvotes, undervotes, unsuccessful litigation, butterfly ballot problems, recounts that might have been allowed but were not, or any other hypothetical divergence between voter preferences and counted votes. After the election, the New York Times conducted a six month long investigation and found that 680 of the overseas absentee ballots were illegally counted, and no partisan, pundit, or academic has publicly disagreed with their assessment. In this paper, we describe the statistical procedures we developed and implemented for the Times to ascertain whether disqualifying these 680 ballots would have changed the outcome of the election. The methods involve adding formal Bayesian model averaging procedures to King's (1997) ecological inference model. Formal Bayesian model averaging has not been used in political science but is especially useful when substantive conclusions depend heavily on apparently minor but indefensible model choices, when model generalization is not feasible, and when potential critics are more partisan than academic. We show how we derived the results for the Times so that other scholars can use these methods to make ecological inferences for other purposes. We also present a variety of new empirical results that delineate the precise conditions under which Al Gore would have been elected president, and offer new evidence of the striking effectiveness of the Republican effort to convince local election officials to count invalid ballots in Bush counties and not count them in Gore counties.
Discussions and Extensions
We propose a remedy for the discrepancy between the way political scientists analyze data with missing values and the recommendations of the statistics community. Methodologists and statisticians agree that "multiple imputation" is a superior approach to the problem of missing data scattered through one’s explanatory and dependent variables than the methods currently used in applied data analysis. The discrepancy occurs because the computational algorithms used to apply the best multiple imputation models have been slow, difficult to implement, impossible to run with existing commercial statistical packages, and have demanded considerable expertise. We adapt an algorithm and use it to implement a general-purpose, multiple imputation model for missing data. This algorithm is considerably easier to use than the leading method recommended in statistics literature. We also quantify the risks of current missing data practices, illustrate how to use the new procedure, and evaluate this alternative through simulated data as well as actual empirical examples. Finally, we offer easy-to-use that implements our suggested methods. (Software: AMELIA)
Applications of modern methods for analyzing data with missing values, based primarily on multiple imputation, have in the last half-decade become common in American politics and political behavior. Scholars in these fields have thus increasingly avoided the biases and inefficiencies caused by ad hoc methods like listwise deletion and best guess imputation. However, researchers in much of comparative politics and international relations, and others with similar data, have been unable to do the same because the best available imputation methods work poorly with the time-series cross-section data structures common in these fields. We attempt to rectify this situation. First, we build a multiple imputation model that allows smooth time trends, shifts across cross-sectional units, and correlations over time and space, resulting in far more accurate imputations. Second, we build nonignorable missingness models by enabling analysts to incorporate knowledge from area studies experts via priors on individual missing cell values, rather than on difficult-to-interpret model parameters. Third, since these tasks could not be accomplished within existing imputation algorithms, in that they cannot handle as many variables as needed even in the simpler cross-sectional data for which they were designed, we also develop a new algorithm that substantially expands the range of computationally feasible data types and sizes for which multiple imputation can be used. These developments also made it possible to implement the methods introduced here in freely available open source software that is considerably more reliable than existing strategies.
We extend a unified and easy-to-use approach to measurement error and missing data. In our companion article, Blackwell, Honaker, and King give an intuitive overview of the new technique, along with practical suggestions and empirical applications. Here, we offer more precise technical details, more sophisticated measurement error model specifications and estimation procedures, and analyses to assess the approach’s robustness to correlated measurement errors and to errors in categorical variables. These results support using the technique to reduce bias and increase efficiency in a wide variety of empirical research.
Although social scientists devote considerable effort to mitigating measurement error during data collection, they often ignore the issue during data analysis. And although many statistical methods have been proposed for reducing measurement error-induced biases, few have been widely used because of implausible assumptions, high levels of model dependence, difficult computation, or inapplicability with multiple mismeasured variables. We develop an easy-to-use alternative without these problems; it generalizes the popular multiple imputation (MI) framework by treating missing data problems as a limiting special case of extreme measurement error, and corrects for both. Like MI, the proposed framework is a simple two-step procedure, so that in the second step researchers can use whatever statistical method they would have if there had been no problem in the first place. We also offer empirical illustrations, open source software that implements all the methods described herein, and a companion paper with technical details and extensions (Blackwell, Honaker, and King, 2014b).
Amelia II is a complete R package for multiple imputation of missing data. The package implements a new expectation-maximization with bootstrapping algorithm that works faster, with larger numbers of variables, and is far easier to use, than various Markov chain Monte Carlo approaches, but gives essentially the same answers. The program also improves imputation models by allowing researchers to put Bayesian priors on individual cell values, thereby including a great deal of potentially valuable and extensive information. It also includes features to accurately impute cross-sectional datasets, individual time series, or sets of time series for diﬀerent cross-sections. A full set of graphical diagnostics are also available. The program is easy to use, and the simplicity of the algorithm makes it far more robust; both a simple command line and extensive graphical user interface are included.
How Surveys Work
Scientific Inference in Qualitative Research
In Legal Research
Case Control and Rare Events Bias Corrections
Hidden Region 1
We offer the first independent scholarly evaluation of the claims, forecasts, and causal inferences of the State Failure Task Force and their efforts to forecast when states will fail. State failure refers to the collapse of the authority of the central government to impose order, as in civil wars, revolutionary wars, genocides, politicides, and adverse or disruptive regime transitions. This task force, set up at the behest of Vice President Gore in 1994, has been led by a group of distinguished academics working as consultants to the U.S. Central Intelligence Agency. State Failure Task Force reports and publications have received attention in the media, in academia, and from public policy decision-makers. In this article, we identify several methodological errors in the task force work that cause their reported forecast probabilities of conflict to be too large, their causal inferences to be biased in unpredictable directions, and their claims of forecasting performance to be exaggerated. However, we also find that the task force has amassed the best and most carefully collected data on state failure in existence, and the required corrections which we provide, although very large in effect, are easy to implement. We also reanalyze their data with better statistical procedures and demonstrate how to improve forecasting performance to levels significantly greater than even corrected versions of their models. Although still a highly uncertain endeavor, we are as a consequence able to offer the first accurate forecasts of state failure, along with procedures and results that may be of practical use in informing foreign policy decision making. We also describe a number of strong empirical regularities that may help in ascertaining the causes of state failure.
Estimating Base Probabilities
Classic (or "cumulative") case-control sampling designs do not admit inferences about quantities of interest other than risk ratios, and then only by making the rare events assumption. Probabilities, risk differences, and other quantities cannot be computed without knowledge of the population incidence fraction. Similarly, density (or "risk set") case-control sampling designs do not allow inferences about quantities other than the rate ratio. Rates, rate differences, cumulative rates, risks, and other quantities cannot be estimated unless auxiliary information about the underlying cohort such as the number of controls in each full risk set is available. Most scholars who have considered the issue recommend reporting more than just the relative risks and rates, but auxiliary population information needed to do this is not usually available. We address this problem by developing methods that allow valid inferences about all relevant quantities of interest from either type of case-control study when completely ignorant of or only partially knowledgeable about relevant auxiliary population information.
Hidden Region 2
Automatic Coding of Rare Events
We report the results of several randomized survey experiments designed to evaluate two intended improvements to anchoring vignettes, an increasingly common technique used to achieve interpersonal comparability in survey research. This technique asks for respondent self-assessments followed by assessments of hypothetical people described in vignettes. Variation in assessments of the vignettes across respondents reveals interpersonal incomparability and allows researchers to make responses more comparable by rescaling them. Our experiments show, first, that switching the question order so that self-assessments follow the vignettes primes respondents to define the response scale in a common way. In this case, priming is not a bias to avoid but a means of better communicating the question’s meaning. We then demonstrate that combining vignettes and self-assessments in a single direct comparison induces inconsistent and less informative responses. Since similar combined strategies are widely employed for related purposes, our results indicate that anchoring vignettes could reduce measurement error in many applications where they are not currently used. Data for our experiments come from a national telephone survey and a separate on-line survey.
How Surveys Work
When respondents use the ordinal response categories of standard survey questions in different ways, the validity of analyses based on the resulting data can be biased. Anchoring vignettes is a survey design technique intended to correct for some of these problems. The anchors package in R includes methods for evaluating and choosing anchoring vignettes, and for analyzing the resulting data.
Unifying Statistical Analysis
Unifying Approaches to Statistical Analysis
"Robust standard errors" are used in a vast array of scholarship to correct standard errors for model misspecification. However, when misspecification is bad enough to make classical and robust standard errors diverge, assuming that it is nevertheless not so bad as to bias everything else requires considerable optimism. And even if the optimism is warranted, settling for a misspecified model, with or without robust standard errors, will still bias estimators of all but a few quantities of interest. The resulting cavernous gap between theory and practice suggests that considerable gains in applied statistics may be possible. We seek to help researchers realize these gains via a more productive way to understand and use robust standard errors; a new general and easier-to-use "generalized information matrix test" statistic that can formally assess misspecification (based on differences between robust and classical variance estimates); and practical illustrations via simulations and real examples from published research. How robust standard errors are used needs to change, but instead of jettisoning this popular tool we show how to use it to provide effective clues about model misspecification, likely biases, and a guide to considerably more reliable, and defensible, inferences. Accompanying this article [soon!] is software that implements the methods we describe.
This (two-page) article argues that the evidence base of political science and the related social sciences are beginning an underappreciated but historic change.