1

Paper

Misunderstandings among Experimentalists and Observationalists about Causal Inference Imai, Kosuke King, Gary Stuart, Elizabeth Uploaded 09-16-2007 Keywords matching blocking causal inference experimental design observational studies average treatment effects covariate balance field experiments survey experiments Abstract 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. (This paper is forthcoming in the Journal of the Royal Statistical Society, but we have some time for revisions and would value any comments anyone might have. This is a revised and much more general version of an earlier paper, "The Balance Test Fallacy in Causal Inference".)

2

Paper

The Varying Role of Voter Information across Democratic Societies Sekhon, Jasjeet Uploaded 07-26-2004 Keywords Voter Information Elections Causal Inference Matching Propensity Score Matching Robust Estimation Democratization Survey Data Abstract Using new robust matching methods for making causal inferences from survey data, I demonstrate that there are profound differences between how voters behave in mature democracies versus how they behave in new ones. The problems of voter ignorance and inattentiveness are not as serious in mature democracies as many analysts have suggested but are of grave concern in new democracies. Citizens in mature democracies are able to accomplish something that citizens in fledgling democracies are not: inattentive and poorly informed citizens are able to vote like their better informed compatriots and hence need to pay little attention to political events such as election campaigns in order to vote as if they were attentive. The results from the U.S. (which rely on various National Election Studies) and Mexico (2000 Panel Study) are reported in detail. Results from other countries are briefly reported.

3

Paper

Survey Context Effects in Anchoring Vignettes Buckley, Jack Uploaded 08-22-2008 Keywords anchoring vignettes survey research differential item functioning experiment Abstract Methodologists (King et al. 2004; King and Wand 2007) have recently proposed a novel approach to adjusting for bias in interpersonal and cross- cultural comparisons in survey research. The method centers on the use of anchoring vignettes to allow the statistical correction of differential usage of ordinal response scales at the individual or group level. Using data from a randomized survey experiment I investigate whether analyses based on these vignettes may be vulnerable to the introduction of survey artifacts due to vignette ordering or the placement of the self-assessment item relative to the vignettes. I find several patterns of bias due to context effects. Researchers using anchoring vignettes should consider randomization or other methods to mitigate these problems.

4

Paper

Enhancing the Validity and Cross-cultural Comparability of Survey Research King, Gary Murray, Christopher J. L. Salomon, Joshua A. Tandon, Ajay Uploaded 07-10-2002 Keywords survey research scaling vignettes measurement Abstract We offer a new approach to writing survey questions and a new statistical model that together at least partially ameliorate two long-standing problems in survey research. The first is how to measure complicated concepts, such as freedom, health, political efficacy, pornography, etc., that researchers know how to define clearly only with reference to examples. The second problem is when different respondents interpret identical survey questions in incomparable ways, as can occur when comparing respondents in different countries speaking different languages, but it also occurs frequently with different groups in the same country. Our approach to these problems is to ask respondents for self-assessments of the concept being measured along with assessments, on the same scale, of each of several hypothetical individuals described by short vignettes. The actual (but not necessarily reported) levels for the people in the vignettes are, by the design of the survey, invariant over respondents and thus provide anchors for our statistical model to transform the self-assessments to a comparable scale. With analysis, simulations, and real surveys in several countries, we show how ignoring these problems can lead to the wrong substantive conclusions and how our approach can fix them. Our methods build on insights from application-specific research on voters and legislators in political science to produce a more general measurement device.

5

Paper

Causal Inference with Differential Measurement Error: Nonparametric Identification and Sensitivity Analyses of a Field Experiment on Democratic Deliberations Imai, Kosuke Yamamoto, Teppei Uploaded 06-30-2008 Keywords differential misclassification nonparametric bounds retrospective studies sensitivity analysis survey measurements Abstract Political scientists have long been concerned about the validity of survey measurements. Although many have studied classical measurement error in linear regression models where the error is assumed to arise completely at random, in a number of situations the error may be correlated with the outcome. We analyze the impact of differential measurement error on causal estimation. The proposed nonparametric identification analysis avoids arbitrary modeling decisions and formally characterizes the roles of additional assumptions. We show the serious consequences of differential misclassification and offer a new sensitivity analysis that allows researchers to evaluate the robustness of their conclusions. Our methods are motivated by a field experiment on democratic deliberations, in which one set of estimates potentially suffers from differential misclassification. We show that an analysis ignoring differential measurement error may considerably overestimate the causal effects. This finding contrasts with the case of classical measurement error which always yields attenuation bias.

6

Paper

Is the Road to Bad Inference Paved with Good Intentions? An Audit of Vote Intention Survey Items Arceneaux, Kevin Uploaded 07-17-2001 Keywords vote intention survey response statistical inference Abstract It is not always possible to conduct surveys near an election. This is especially true in Europe where the timing of elections varies widely from country to country. Not only do elections rarely occur at the same time across countries, election dates are not fixed within most European countries, making it nearly impossible to plan post-election surveys in Europe. Even in the United States, where elections are fixed events, off-election surveys are often used. Because there is no ongoing campaign, these surveys rely on vote intention questions as a measure of vote choice. The typical wording of a vote intention question is, “If there were a general election tomorrow, which party would you support?” Many researchers, especially users of Eurobarometer data, rely on these questions as a measure of vote choice. However, it is unclear if these questions are in fact valid indicators of voting behavior. Drawing on information processing theories, this paper develops a theoretical framework to identify the conditions under which vote intention measures accurately vote choice. Hypotheses generated by this framework are tested across multiple countries and years using Eurobarometer and American National Election Studies survey data.

7

Paper

Sampling Schemes for Generalized Linear Dirichlet Random Effects Models Kyung, Minjung Gill, Jeff Casella, George Uploaded 02-18-2009 Keywords generalized linear mixed Dirchlet model Markov chain Monte Carlo Dirichlet process priors for random effects precision parameters Scottish Social Attitudes Survey terrorism targeting Abstract We evaluate MCMC sampling schemes for a variety of link functions in generalized linear models with Dirichlet random effects. We find that models using Dirichlet process priors for the random effects tend to capture information in the data in a nonparametric fashion. In fitting the the Dirichlet process, dealing with the precision parameter has troubled model specifications in the past. Here we find that incorporating this parameter into the MCMC sampling scheme is not only computationally feasible, but also results in a more robust set of estimates, in that they are marginalized-over rather than conditioned-upon. Applications are provided with social science problems in areas where the data can be difficult to model. In all, we find that these models provide superior Bayesian posterior results in theory, simulation, and application.

8

Paper

Using Auxiliary Data to Estimate Selection Bias Models Boehmke, Frederick Uploaded 07-06-2001 Keywords selection bias two-stage estimation survey design initiative interest groups Abstract Recent work has made progress in estimating models involving selection bias of a particularly strong nature: all nonrespondents are unit nonresponders, meaning that no data is available for them. These models are reasonable successful in circumstances where the dependent variable of interest is continuous, but they are less practical empirically when it is latent and only discrete outcomes or choices are observed. I develop a method in this paper to estimate these models that is much more practical in terms of estimation. The model uses a small amount of auxiliary information to estimate the selection equation parameters which are then held fixed to estimate the equation of interest parameters in a maximum likelihood setting. After presenting monte carlo analysis to support the model, I apply the technique to a substantive problem: which interest groups are likely to to be involved in support of potential initiatives to achieve their policy goals.

9

Paper

Measuring Political Support and Issue Ownership Using Endorsement Experiments, with Application to the Militant Groups in Pakistan Bullock, Will Imai, Kosuke Shapiro, Jacob Uploaded 07-18-2010 Keywords endorsement experiment survey experiment bayesian pakistan militant groups issue ownership social desirability Abstract To measure the levels of support for political actors (e.g., candidates and parties) and the strength of their issue ownership, survey experiments are often conducted in which respondents are asked to express their opinion about a particular policy endorsed by a randomly selected political actor. These responses are contrasted with those from a control group that receives no endorsement. This survey methodology is particularly useful for studying sensitive political attitudes. We develop a Bayesian hierarchical measurement model for such endorsement experiments, demonstrate its statistical properties through simulations, and use it to measure support for Islamist militant groups in Pakistan. Our model uses item response theory to estimate support levels on the same scale as the ideal points of respondents. The model also estimates the strength of political actors' issue ownership for specic policies as well as the relationship between respondents' characteristics and support levels. Our analysis of a recent survey experiment in Pakistan reveals three key patterns. First, citizens' attitudes towards militant groups are geographically clustered. Second, once these regional differences are taken into account, respondents' characteristics have little predictive power for their support levels. Finally, militant groups tend to receive less support in the areas where they operate.

10

Paper

An Individual-Level Approach to Health Inequality: Child Survival in 50 Countries King, Gary Gakidou, Emmanuela Uploaded 11-27-2000 Keywords beta-binomial health inequality survey research Abstract BACKGROUND: Reducing health inequalities is an important part of the agenda of health policymakers globally. Studies of health inequalities have revealed large variations in average health status across social, economic, and other _groups_. However, no studies have been conducted on the distribution of the risk of ill-health across _individuals_. METHODS: We use an extended beta-binomial model to estimate the distribution the risk of death in children under the age of two in the 50 developing countries where data from a Demographic and Health Survey are available. Inequality in these distributions is measured by the WHO health inequality index. FINDINGS: At the same level of average child mortality, inequality in the risk of death across children can vary considerably across countries. Representing the entire distribution of risk with a single measure of inequality involves normative choices that we delineate and then formalise with quantitative measures. The results are not very sensitive to the choice of measure. Liberia, Mozambique and the Central African Republic have the largest inequalities in child survival, while Colombia, the Philippines and Kazakhstan have the lowest among the 50 countries measured. Exploratory analyses indicate that health inequality is predicted by low GDP, low health expenditures, and poverty, but not by income inequality or democratization. INTERPRETATION: Inequality estimates should routinely be reported alongside average levels of health, as they reveal important information about the distribution of health across individuals within populations. Measuring inequality with individual level data, rather than quantifying differences in average levels of health across social groups, enables meaningful comparisons of inequality across countries and analyses of the determinants of inequality. This approach should be extended to the measurement of inequalities in health expectancy (i.e., life expectancy discounted by expected disabilities).

11

Paper

Statistical Inference for the Item Count Technique Imai, Kosuke Uploaded 07-19-2010 Keywords list experiments sensitive questions survey experiments unmatched count technique Abstract The item count technique is a survey methodology that is designed to elicit respondents' truthful answers to sensitive questions such as racial prejudice and drug use. The method is also known as the list experiment or the unmatched count technique and is an alternative to the commonly used randomized response method. In this paper, I propose new nonlinear least squares (NLS) and maximum likelihood (ML) estimators for a multivariate analysis with the item count technique. The two-step estimation procedure and the Expectation Maximization algorithm are developed to facilitate the computation. Enabling a multivariate statistical analysis is essential because the item count technique provides respondents with privacy at the expense of statistical efficiency. As an empirical illustration, the proposed methodology is applied to the 1991 National Race and Politics survey where the investigators used the item count technique to measure the degree of racial hatred in the United States. A small-scale simulation study suggests that the ML estimator can be substantially more efficient than the NLS estimator. The software package is made available to implement the proposed methodology.

12

Paper

Binding the Frame: How Important are Frames for Survey Response? Alvarez, R. Michael Brehm, John Uploaded 08-26-2000 Keywords framing survey experiments heteroskedastic choice Abstract [not transcribed]

13

Paper

What Can We Learn with Statistical Truth Serum? Design and Analysis of the List Experiment Glynn, Adam Uploaded 07-23-2010 Keywords social desirability indirect questions list experiment item count technique privacy protection survey experiment Abstract Due to the inherent sensitivity of many survey questions, a number of researchers have adopted an indirect questioning technique known as the list experiment (or the item count technique) in order to minimize bias due to dishonest or evasive responses. However, standard practice with the list experiment requires a large sample size, is not readily adaptable to regression or multivariate modeling, and provides only limited diagnostics. This paper addresses all three of these issues. First, the paper presents design principles for the standard list experiment (and the double list experiment) to minimize bias and reduce variance as well as providing sample size formulas for the planning of studies. Additionally, this paper investigates the properties of a number of estimators and introduces an easy-to-use piecewise estimator that reduces necessary sample sizes in many cases. Second, this paper proves that standard-procedure list experiment data can be used to estimate the probability that an individual holds the socially undesirable opinion/behavior. This allows multivariate modeling. Third, this paper demonstrates that some violations of the behavioral assumptions implicit in the technique can be diagnosed with the list experiment data. The techniques in this paper are illustrated with examples from American politics.

14

Paper

Is There a Gender Gap in Fiscal Political Preferences Alvarez, R. Michael McCaffery, Edward J. Uploaded 08-12-2000 Keywords Gender gap fiscal politics taxation budget surplus multinomial logit missing data imputation framing survey experiments Abstract This paper examines the relationship between attitudes on potential uses of the budget surplus and gender. Survey results show relatively weak support overall for using a projected surplus to reduce taxes, with respondents much likelier to prefer increased social spending on education or social security. There is a significant gender gap with men being far more likely than women to support tax cuts or paying down the national debt. Given a menu of particular types of tax cuts, women are marginally more likely to favor child-care relief or working poor tax credits whereas men are marginally more likely to favor capital gains reduction or tax rate cuts. When primed that the tax laws are biased against two-worker families, men significantly change their preferences, moving from support for general tax rate cuts to support for working poor tax relief, but not to child-care relief. One of the strongest results to emerge is that women are far more likely than men not to express an opinion or to confess ignorance about fiscal matters. Both genders increase their ``no opinion'' answer in the face of priming, but men more so than women. Further research will explore this no opinion/uncertainty aspect.

15

Paper

The Influence of the Initiative Process on Interest Groups and Lobbying Techniques Boehmke, Frederick Uploaded 09-22-1999 Keywords Initiative direct democracy survey analysis interest groups lobbying selection bias Abstract I use survey data on interest groups and their activities drawn from four state populations to test hypotheses about the implications of direct democracy for the characteristics and strategic choices of interest groups. I use this data to test predictions about direct democracy's effect for group populations, confirming previous work (Boehmke 1999b) and extending it by exploring more detailed characteristics such as membership and resources. I then link these characteristics to lobbying techniques to test if the initiative process has an impact at the group level. As expected, groups involved in initiative campaigns tend to accentuate outside lobbying strategies, but even groups not currently involved in initiatives are influenced by the possibility of its use. This is because the initiative process alters the characteristics that can be effectively used when attempting to influence policy. The analysis makes use of a technique to correct for heterogeneous response rates across group types. By gathering information about a high percentage of an additional, smaller sample, I am able to correct for this response rate differential through a weighting procedure. The correction is found to have a substantial effect on the results: its absence would leave the researcher to conclude that the initiative plays little role in state interest group activities. This data will also be used to test and correct for possible sample selection bias.

16

Paper

A Mixed-Membership Approach to the Assessment of Political Ideology from Survey Responses Gross, Justin Manrique-Vallier, Daniel Uploaded 07-13-2012 Keywords latent structure model grade-of-membership mixed-membership latent variables measurement ideology beliefs core values attitudes discrete factor analysis survey response Abstract We employ mixed-membership (or grade-of-membership) techniques--of growing popularity in medical diagnostics, psychology, genetics, and machine learning--in order to identify prototypical profiles of survey respondents based on their answers to questions aimed at uncovering their basic orientations or ideological predispositions. In contrast with factor analytic techniques and IRT approaches, we treat both manifest and latent variables as categorical. A mixed membership model may be thought of as a generalization of latent class modeling, in which individuals act as members of more than one class. This notion is well-aligned with earlier theoretical work of Zaller, Feldman, Stimson, and others, who at times envision respondents to be internally complex, answering survey questions probabilistically according to what Zaller calls varying ``considerations.'' Reanalyzing data in this way, we develop new insights into the sorts of constraints that may structure mass belief systems.

17

Paper

Uncertainty and Candidate Personality Traits Alvarez, R. Michael Glasgow, Garrett Uploaded 04-16-1998 Keywords uncertainty direct measures of uncertainty survey response ordered probit candidate evaluation candidate traits presidential elections Abstract Recently, some scholars have focused attention on the role of uncertainty in elections (Alvarez 1997, Bartels 1986, Franklin 1991). They reveal that there is a great deal of uncertainty about the issue positions of candidates, and thus the costs of issue voting are burdensome for the average citizen. Further, this uncertainty affects how voters evaluate candidates in two ways. First, voters are less likely to evaluate a candidate in terms of an issue when they are uncertain about the candidate's position on that issue. Second, uncertainty about candidate issue positions has a negative impact on voter evaluations of a candidate. However, it is important to realize that for most individuals, information about the personality traits of candidates comes from the same sources as information about the issue positions of the candidates, generally media outlets. This means that information about the the personalities of candidates is passed through the same noisy channels as information about their issue positions, and is thus subject to the same types of distortions and biases that contribute to the cost of issue information. Although it is likely easier to interpret than issue information, trait information is still subject to uncertainty. In this paper we introduce direct survey measures of candidate personality trait uncertainty. Using survey data drawn from the 1995 and 1996 National Election Studies, we first establish that the direct measure of uncertainty used in this paper is a valid measure. We then examine the effect of trait opinions on candidate evaluations and test the effects that uncertainty about those opinions has on the use of traits in candidate evaluation.

18

Paper

Causal Inference in Conjoint Analysis: Understanding Multi-Dimensional Choices via Stated Preference Experiments Hainmueller, Jens Hopkins, Daniel Yamamoto, Teppei Uploaded 12-12-2012 Keywords potential outcomes average marginal component effects fractional factorial design orthogonal design randomized design survey experiments public opinion vote choice immigration Abstract For decades, market researchers have used conjoint analysis to understand how consumers make decisions when faced with multi-dimensional choices. In such analyses, respondents are asked to score or rank a set of alternatives, where each alternative is defined by multiple attributes which are varied randomly or intentionally. Political scientists are frequently interested in parallel questions about decision-making, yet to date conjoint analysis has seen little use within the field. In this manuscript, we demonstrate the potential value of conjoint analysis in political science, using examples about vote choice and immigrant admission to the United States. In doing so, we develop a set of statistical tools for drawing causal conclusions from stated preference data based on the potential outcomes framework of causal inference. We discuss the causal estimands of interest and provide a formal analysis of the assumptions required for identifying those quantities. Prior conjoint analyses have typically used designs which limit the number of unique conjoint profiles. We employ a survey experiment to compare this approach to a fully randomized approach. Both our formal analysis of the causal estimands and our empirical results highlight the potential biases of common approaches to conjoint analysis which restrict the number of profiles.

19

Paper

Attitudes, Uncertainty and Survey Responses Alvarez, R. Michael Franklin, Charles Uploaded 00-00-0000 Keywords uncertainty survey response issue voting attitude stability Abstract Theory: We assume that survey respondents are uncertain about their attitudes, and that their attitudes about political issues can be understood as probability distributions. From this perspective, we derive the ``expected value'' survey response model. We also derive a dynamic model of attitude change, based on the notion that attitudes are uncertain. Hypotheses: This perspective on political attitudes leads to two predictions. The first is that uncertain respondents will show less variance in responses than certain respondents, and that the less certain will tend to give responses towards the midpoint of issue placement scales. The second is that uncertain respondents will have less stable opinions about political issues over time. Methods: These hypotheses are tested using new survey questions we have developed to measure respondent uncertainty. These survey questions have been included in three recent national surveys, two conducted by the Letters and Sciences Survey Center at the University of Wisconsin, Madison and the other by the National Election Studies. Results: We demonstrate that uncertain respondents are more likely that certain respondents to provide issue placements at the midpoint of the scale, controlling for many factors. Also, we show that uncertain respondents have less stable political attitudes than certain respondents.

20

Paper

The Economic Sophistication of Public Opinion in the United States Sekhon, Jasjeet Uploaded 09-18-1997 Keywords Public Opinion Economic Sophistication Survey of Consumer Attitudes and Behavior (SCAB) Natural Rate of Unemployment NAIRU Unemployment Bootstrap Bootstrap Confidence Region Abstract I show that the public does indeed have coherent and sophisticated reactions to macroeconomic variables. These reactions are consistent with economic theory. Individuals form evaluations and expectations in a way which is sensitive to the complex trade-off between unemployment and inflation as determined by the nonaccelerating inflation rate of unemployment (NAIRU). The primary dataset used in this analysis has 69,680 observations and is compiled by merging 113 individual level ``Surveys of Consumer Attitudes and Behavior'' from 1976:01 to 1991:12. The data analysis makes extensive use of bootstrap methods to create confidence regions and to conduct hypothesis tests.

21

Paper

The Coalition-oriented Evolution of Vote Intentions across Regions and Levels of Political Awareness during the 1993 Canadian Election Campaign: Quotidian Markov Chain Models using Rolling Cross-section Data Wand, Jonathan Mebane, Walter R. Uploaded 08-28-1997 Keywords Markov chains rolling cross-section data macro data categorical data survey data Canadian politics strategic voting coalitions estimation Abstract We use survey data collected in Ontario and Quebec during the 1993 Canadian federal election to assess the extent to which voters were sensitive to the distribution of positions in special institutions that would possibly be created to handle negotiations between Quebec and the rest of Canada following a referendum on Quebec sovereignty expected after the election. We draw on a theory of coalition-oriented voting developed by Austin-Smith and Banks (1988) to argue that voters' anticipations regarding those institutions contributed to the catastrophic losses suffered by the Progressive Conservative party. We use a method we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in daily rolling cross-sectional survey data from 1993 Canadian Election Study. We allow each transition matrix to be updated as a function of daily vote support for either the Bloc or Reform to test for reactive coalition-oriented voting. We find significant reactive voting among Quebecois non-sovereigntists. The timing of these reactions depended on the individual's level of political awareness. In contrast, we find no evidence of reactive voting among either Quebecois sovereigntists or Ontario voters.

22

Paper

Markov Chain Models for Rolling Cross-section Data: How Campaign Events and Political Awareness Affect Vote Intentions and Partisanship in the United States and Canada Mebane, Walter R. Wand, Jonathan Uploaded 04-07-1997 Keywords Markov chains rolling cross-section data macro data measurement error categorical data ordinal data panel data survey data party identification American politics Canadian politics estimation Abstract We use a new approach we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in two collections of rolling cross-sectional survey data that were designed to support investigations of what happens to voters' information and preferences during campaigns. Using data from the 1984 American National Election Studies Continuous Monitoring Study, we show that not only did individual party identification vary substantially during the year, but the dynamics of party identification changed significantly in response to the conclusion of the Democratic party's nomination contest. Party identification appears to have measurement error only when the model misspecifies the dynamics. There are rapid oscillations among some categories of partisanship that may reflect individual stances regarding not only competition between the parties but also competition among party factions. Using data from the 1993 Canadian Election Study, we show that the critical events that shaped voting intentions in the election varied tremendously depending on an individual's level of political awareness, and that the effects of awareness varied across regions of the country.

23

Paper

Treatment Spillover Effects Across Survey Experiments Lee, Daniel Transue, John Aldrich, John Uploaded 04-05-2005 Keywords survey experiments experiments survey methods Abstract Embedding experiments within surveys has reinvigorated survey research in general and especially in political science. These designs use random assignment to create true experiments within (typically nationally) representative sample surveys. Thus, they combine the internal validity of experiments with the external validity of national surveys. We investigate whether experimental treatments spill over and effect later experiments in an unintended manner. Using the 1991 Race and Politics survey, we find evidence of experimental spillover. Specifically we find that experiments at the beginning of a survey influence later experiments. We also find (much less) evidence of adjacent experiments affecting subsequent experiments. The paper concludes with a discussion of designs for future research that could aid our understanding of experimental spillover.

24

Paper

Sensitive Questions, Truthful Answers? Modeling the List Experiment Multivariately With LISTIT Corstange, Daniel Uploaded 08-03-2006 Keywords list experiment survey sensitive questions listit Abstract Standard estimation procedures assume that empirical observations are accurate reflections of the true values of the dependent variable, but this assumption is dubious when model self-reported data on sensitive topics. List experiments can nullify incentives for respondents to lie to interviewers, but current data analysis techniques are limited to difference-in-means tests. I present a revised procedure and statistical estimator called listit to model the process multivariately. Monte Carlo simulations and a field test in Lebanon explore the behavior of this estimator.

25

Poster

A Split Population Model for Middle-Category Inflation in Ordered Survey Responses Bagozzi, Benjamin Uploaded 08-01-2011 Keywords split-population models ordered dependent Variables survey data Europe public opinion Abstract Recent research find that, for social desirability reasons, uninformed individuals disproportionately give ``neither agree nor disagree'' type responses to survey attitude questions, even when a ``don't know'' option is available (Sturgis et al. 2010). Such ``face-saving don't knows'' inflate the indifference (i.e. middle) categories of ordered attitude variables with non-ordered responses. When this inflation occurs within one's dependent variable, estimates from ordered probit/logit models are biased and inefficient. This poster develops a set of mixture models (the middle-inflated ordered probit with and without correlated errors) that estimate and account for the presence of ``face-saving'' responses in middle-categories of ordered survey response variables, and applies these models to (1) simulated data and (2) a commonly studied survey question measuring support for EU-membership among EU-candidate countries. Findings suggest that, when middle-category inflation is present in one's ordered dependent variable, the estimates obtained from middle-category mixture models are less biased than---and in some cases substantively distinct from---the estimates obtained from ``naive'' ordered probit models.