Search

The Canada Centre for Mapping and Earth Observation (CCMEO) uses Radarsat Constellation Mission (RCM) data for near-real time flood mapping. One of the many advantages of using SAR sensors, is that they are less affected by the cloud coverage and atmospheric conditions, compared to optical sensors. RCM has been used operationally since 2020 and employs 3 satellites, enabling lower revisit times and increased imagery coverage. The team responsible for the production of flood maps in the context of emergency response are able to produce maps within four hours from the data acquisition. Although the results from their automated system are good, there are some limitations to it, requiring manual intervention to correct the data before publication. Main limitations are located in urban and vegetated areas. Work started in 2021 to make use of deep learning algorithms, namely convolutional neural networks (CNN), to improve the performances of the automated production of flood inundation maps. The training dataset make use of the former maps created by the emergency response team and is comprised of over 80 SAR images and corresponding digital elevation model (DEM) in multiple locations in Canada. The training and test images were split in smaller tiles of 256 x 256 pixels, for a total of 22,469 training tiles and 6,821 test tiles. Current implementation uses a U-Net architecture from NRCan geo-deep-learning pipeline (https://github.com/NRCan/geo-deep-learning). To measure performance of the model, intersection over union (IoU) metric is used. The model can achieve 83% IoU for extracting water and flood from background areas over the test tiles. Next steps include increasing the number of different geographical contexts in the training set, towards the integration of the model into production.

In many developing democracies, polling often comes with elevated levels of bias and variance. We argue that electoral malpractice can be one reason why. We build a theory and test it with data from elections between 2002 and 2014 in Brazil. We find that polling errors are larger in: (a) elections with many undecided voters and large imbalances in financial resources among campaigns; (b) the poorer Northeast region of Brazil, which is more closely associated with patronage and vote buying; and (c) low-profile, low-information elections. Our analysis serves as a cautionary tale for interpreting polling in democracies like Brazil, even if/when other sources of error in the polling industry are mitigated.

Airborne LiDAR data allow the precise modeling of topography and are used in multiple contexts. To facilitate further analysis, the point cloud classification process allows the assignment of a cla...

We study the impact of compulsory voting in Brazil, where voting is mandatory from age 18 to 70 and voluntary for those aged 16, 17 and 70+. Using a survey sample of 8008 respondents, we document voter confusion about how the age criterion applies. Some people falsely believe that what matters is one's age in an election year rather than on Election Day. Next, we perform a regression discontinuity (RD) analysis of compulsory voting among young voters with register-based data from six Brazilian elections (2008–2018). We find that the effect of compulsory voting is seriously underestimated if we focus solely on the discontinuities prescribed by the law. Our findings carry important implications for studies adopting the RD design where knowledge of the cutoff is expected of the units of interest (like those about compulsory voting) and confirm that compulsory voting is a strong institutional arrangement that promotes greater electoral participation.

When the number of observed elections is low, subnational data can be used to perform electoral forecasts. Turgeon and Rennó (2012) applied this solution and proposed three forecasting models to analyze Brazilian presidential elections (1994-2006). The models, adapted from forecasting models of American and French presidential elections, considers economic and political factors. We extend their analysis to the recent presidential elections in Brazil (2010, 2014 and 2018) and find that the addition of the three recent elections does not improve the accuracy of our forecast models although it strengthens the relationship between the explanatory variables and vote for the incumbent. We also find that models based on the popularity of the incumbent outperform those based on trial-heat polls and that electoral forecast models can survive earthquake elections like the 2018 election that led to the unexpected rise of “outsider” and extremist candidate Jair Bolsonaro.

Coalition theories of presidential regimes have frequently assumed that coalition formation is a mostly post-electoral phenomenon. We challenge this view by showing that pre-electoral bargaining shapes to a substantial extent minority presidents’ disposition to cooperate with the legislature by forming a majority cabinet. Examining a dataset of pre- and post-electoral coalitions from 18 Latin American countries, we find that majority coalition cabinets are more likely to occur when elected presidents form pre-electoral coalitions (PECs), to the extent that pre-electoral agreements create stronger incentives for cooperation, by relying on a broader set of rewards than any post-electoral agreement. Moreover, we find that the likelihood of majority coalition formation increases as the share of PEC seats increases, thus reducing the need to engage in post-electoral bargaining. Our findings carry important implications for the study of cabinet formation in presidential regimes by introducing pre-electoral agreements as a key determinant of cabinet formation.

The interplay between explicit and implicit attitudes toward affirmative action (AA) policies is relevant to applied psychology. Its comprehension helps to improve our capacity to evaluate support for such policies. The purpose of this study was to determine the extent to which students’ race, political opinion of affirmative action, and prejudice against minorities influence the relationship between implicit-explicit attitudes toward affirmative action policies. 492 student participants were recruited from a large Brazilian public university about racial quotas in admissions. Implicit and explicit measures of attitude about the admission process were applied, together with measures of political opinion of affirmative action, prejudice against minorities and race. The results show that race has little effect on the difference between implicit and explicit attitudes about the admission process, but that prejudice and political position exert strong effects. Our findings suggest that implicit measures of attitudes should be used when evaluating attitudes on AA.

Relatively little is known about how voters behave under dual ballot voting systems. Do voters follow different decision rules at each ballot? Do they weigh relevant considerations differently on the second ballot, as compared to the first ballot? Does political sophistication condition the way people vote on the first and second ballots? Using survey data from five French presidential elections, we examine these questions and find that voters in France do indeed treat first and second ballots differently. More precisely, we find that both partisanship and ideology matter more on the second ballot. Demographics, socioeconomic status, and political issues, on the other hand, weigh more heavily in the first round. Political sophistication, for its part, does not condition nor reinforce these effects.

The COVID-19 pandemic has led to a vast research agenda focusing on how citizens acquire knowledge about the virus and the health expert guidelines to protect themselves and their close ones against it. While many countries and regions have been accounted for, there still remains a substantial gap with respect to public opinion about the virus in Latin America, most notably in Brazil, which currently has the second highest in number of fatalities in the world. In this article, we employ a national survey of Brazilians (n = 2,771) to measure and explain knowledge and misinformation about the coronavirus and its illness, COVID-19. Our focus concerns the role of political preferences in a context of high elite polarization with a sitting government that has systematically downplayed the risks associated with the coronavirus and its illness. Our findings are clear: political preferences play a substantial role in explaining differences in knowledge about the coronavirus and COVID-19, more than conventional determinants of learning like motivation, ability, and opportunities. Specifically, we find that supporters of President Jair Bolsonaro—an avid science and COVID-19 denier—know significantly less about the coronavirus and its illness and are more likely to believe in a conspiracy theory that claims that the coronavirus was purposefully created in a Chinese laboratory to promote China's economic power, when compared to Brazilians who are less supportive of him and his government. Our findings carry important implications for how Brazilians take informational cues from political elites in that—even in a major event like a global pandemic—supporters of the president are as likely as ever to “follow their leader” and deny expert-backed scientific evidence.

This study examines the effect of the country of origin of the vaccine on vaccination acceptance against COVID-19. More specifically, we show how the political context in Brazil has affected acceptance of vaccines produced in China, Russia, the US, and England at the University of Oxford. Our data come from a survey experiment applied to a national sample of 2771 adult Brazilians between September 23 and October 2, 2020. We find greater rejection among Brazilians for vaccines developed in China and Russia, as compared to vaccines from the US or England. We also find that rejection of the Chinese-developed vaccine is particularly strong among those who support President Jair Bolsonaro—a COVID-19 denier and strong critic of China and vaccination, in general.

As políticas de ação afirmativa implementadas nas universidades públicas no Brasil abrangem três principais grupos: alunos de escolas públicas, alunos de baixa renda e alunos negros e indígenas. Opiniões sobre essa modalidade de política de ação afirmativa mais ampla, especialmente se mensuradas por perguntas diretas, podem ser enviesadas por um efeito de desejabilidade social. Este artigo procura entender se atitudes sobre diferentes tipos de políticas de ação afirmativa são propensas a um efeito de desejabilidade social e quais são os determinantes desse efeito. Para mensurar esse efeito e seus determinantes, um experimento de lista foi aplicado a 12.490 estudantes universitários em 2013. Os resultados revelam que as opiniões sobre os três tipos de políticas de ação afirmativa são propensas a um efeito de desejabilidade social. Ademais, encontramos evidências de que tanto dois determinantes individuais (interesse próprio e preconceito em relação ao grupo-alvo) como um determinante contextual (áreas de conhecimento do curso universitário) explicam o efeito de desejabilidade social. Abstract Affirmative action policies implemented in public universities in Brazil cover three main groups: public school students, low-income students, and black and indigenous students. Attitudes about affirmative action policy, when measured conventionally in surveys, can be contaminated by the social desirability bias effect. This article seeks to evaluate if attitudes about the different types of affirmative action policies recently adopted in Brazil are subject to the social desirability bias effect, and, if so, what are the determinants of this effect. To measure the social desirability bias effect and its determinants, we conducted a list experiment among 12.490 university students in 2013. The results reveal that attitudes about the three types of affirmative action policies are prone to a social desirability bias effect. We also find evidence that two individual-level determinants (self-interest and prejudice against the target group) and one contextual-level determinant (students’ field of study) explain the social desirability bias effect. Resumen Las políticas de acción afirmativa implementadas en las universidades públicas de Brasil cubren tres grupos principales: estudiantes de escuelas públicas, estudiantes de bajo ingreso y estudiantes negros e indígenas. Las actitudes acerca de la política de acción afirmativa, cuando se miden de manera convencional en encuestas, pueden estar contaminadas por el efecto de sesgo de deseabilidad social. Este artículo busca evaluar si las actitudes sobre los diferentes tipos de políticas de acción afirmativa recientemente adoptadas en Brasil están sujetas al efecto de sesgo de deseabilidad social y, de ser así, cuáles son los determinantes de este efecto. Para medir el efecto del sesgo de la deseabilidad social y sus determinantes, llevamos a cabo un experimento de lista entre 12.490 estudiantes universitarios en 2013. Los resultados revelan que las actitudes sobre los tres tipos de políticas de acción afirmativa son propensas a un efecto de sesgo de la deseabilidad social. También encontramos evidencia de que dos determinantes a nivel individual (interés propio y prejuicio contra el grupo objetivo) y un determinante a nivel contextual (campo de estudio de los estudiantes) explican el efecto del sesgo de la deseabilidad social. Résumé Les politiques d'action positive mises en œuvre dans les universités publiques du Brésil couvrent trois groupes principaux: les étudiants des écoles publiques, les étudiants à faible revenu et les étudiants noirs et autochtones. Les attitudes à l'égard de la politique d'action positive, lorsqu'elles sont mesurées de manière conventionnelle dans les enquêtes, peuvent être contaminées par l'effet de biais de désirabilité sociale. Cet article cherche à évaluer si les attitudes à l'égard des différents types de politiques d'action positive récemment adoptées au Brésil sont sujettes à l'effet de biais de désirabilité sociale et, dans l'affirmative, quels sont les déterminants de cet effet. Pour mesurer l'effet de biais de désirabilité sociale et ses déterminants, nous avons mené une expérience de liste (list experiment) auprès de 12 490 étudiants universitaires en 2013. Les résultats révèlent que les attitudes à l'égard des trois types de politiques d'action positive sont sujettes à un effet de biais de désirabilité sociale. Nous trouvons également que deux déterminants au niveau individuel (intérêt personnel et préjugés contre le groupe cible) et un déterminant au niveau contextuel (domaine d'études des étudiants) expliquent l'effet du biais de désirabilité sociale.

Research on presidential coattails in elections has focused nearly exclusively on two-party systems like the United States. Much less is known, however, about presidential coattails in multiparty systems where electoral and governing coalitions are common currency. Under coalitional presidentialism, we argue that presidential coattails are diffused, benefiting the president’s party but also her coalition party members, especially those most strongly associated with the coalition itself. Using electoral data from Chile and Brazil, two widely studied but distinct cases of coalitional presidentialism, we find strong evidence supporting this claim. The results carry important implications for students of coalitional presidentialism, presidential coattails, and party systems, more generally.

Natural Resources Canada is in the process of defining spatial resolution requirements for a national scale optical imagery service. This study aims to evaluate the impact of spatial resolution on residential buildings extraction accuracy in an urban context. The hypothesis is that as spatial resolution gets coarser, there will be a decrease in successful identification. A 3-band Worldview-3 image was used for a study site located in Toronto, Canada. Buildings were manually extracted independently by specialized technicians for five different resolutions ranging from 30 cm to 2.5 m. Results show similar extraction accuracy between 0.3 m and 0.7 m (>90%). There is a significant decrease in accuracy for resolutions of 1 m and coarser. The study demonstrates that a spatial resolution of 0.7 m or better is required for accurate building extraction in an urban context.

Compulsory voting is known for boosting electoral turnout, even when sanctions for abstaining are small or loosely enforced. Much less is known, however, about the consequences of compulsory voting on vote choice, and, in particular, about the quality of electoral decisions. In this paper, we explore the extent to which voters meaningfully engage in the electoral process or simply vote randomly because voting is required by law. We conducted a large online survey in Brazil during the 2018 national elections to assess if voters engage in random voting. We evaluate random voting for low-profile, low-information elected offices (state and federal legislators) and others that receive greater media coverage (governor and president) and evaluate the determinants of random voting for each of them. We find that: 1) random voting does not appear to be affected by social desirability bias; 2) there is substantial random voting under compulsory voting; 3) more voters tend to engage in random voting in low-profile, low-information elections, as compared to elections that receive greater media coverage; and, 4) interest in politics, education, and disposition to vote if voting were to be voluntary reduce random voting. Our findings carry important implications for the study of citizen participation and civic competence under compulsory voting and for democratic representation, more broadly.

This paper introduces the exploitation of a convolutional neural network for the extraction of topographic features from high-resolution optical satellite imagery. A UNET based model was trained for seven feature classes of roads, buildings, waterbodies using two 3-band (RGB) images for a study site in Kingston (Canada). The trained model's accuracy was evaluated on eight tiles of 8000×8000 pixels using a confusion matrix, the overall accuracy and kappa. The results show overall accuracy varying between 90 % and 99 % and kappa varying between 0.48 and 0.98, with five of the eight tiles being over 0.85. The model generally produced accurate predictions, except for commercial and industrial buildings and for unpaved roads, which were under represented in the training data. The project provided perspective for the development of a training database for topographic feature extraction using deep learning and for expansion to the national scale.

RESUMO Introdução Décadas de pesquisas sobre a opinião pública demonstraram que a maioria das pessoas é altamente desinformada a respeito da política. Entretanto, as consequências dessa desinformação sobre o comportamento eleitoral e sobre as atitudes dos cidadãos ainda parecem pouco claras, especialmente quando se trata de jovens democracias. Métodos Neste artigo, utilizamos simulações estatísticas de um eleitorado “completamente informado” a partir de dados do Estudo Eleitoral Brasileiro (ESEB) para responder às seguintes perguntas: (1) Como os brasileiros votariam caso fossem mais informados? (2) Como seriam as opiniões políticas dos brasileiros caso fossem mais informados? Aqui, "informação" se refere a conhecimento sobre a política e o governo de maneira geral Resultados Os resultados sugerem que os votos e as opiniões políticas dos brasileiros seriam diferentes caso eles fossem mais informados. A falta de informação parece direcionar sistematicamente as preferências coletivas, sugerindo que tais erros não são aleatórios, mas sim produto de assimetrias de informação entre os cidadãos Discussão Essas assimetrias podem ser especialmente problemáticas quando se trata de quais opiniões são levadas em consideração pelos representantes dos eleitores, ao tomarem decisões importantes. Os efeitos da desinformação política têm sido vistos sob diferentes olhares pela Ciência Política. Embora alguns teóricos argumentem que a falta de informação não tem efeitos significativos sobre a democracia, outros creem que ela pode trazer consequências notáveis para as escolhas coletivas e para as atitudes políticas dos cidadãos. Nossos resultados nos colocam ao lado do segundo grupo de pesquisadores. ABSTRACT Introduction Decades of research in public opinion have showed that most people are highly uninformed about politics. However, the consequences of low information on voting and political attitudes remain unclear, especially in young democracies. Methods In this paper, we simulate a “fully informed” electorate using data from the Brazilian Electoral Study (ESEB) and ask: (1) How would Brazilians vote if they were more informed? and (2) How would public opinion look like if Brazilians were more informed? Here, information refers to knowledge about government and politics, in general. Results Our findings suggest votes and public opinion would be different in Brazil if its electorate were more knowledgeable. The lack of information seems to systematically orientate collective preferences, suggesting that such errors are not random, but rather as a product of information asymmetries among citizens. Discussion These asymmetries can be especially problematic when it comes to which views are taken into account by voter representatives when making important decisions.The effects of political disinformation have been seen under different glances by political science. Although some theorists argue that lack of information has no significant effects on democracy, others believe that it can have notable consequences for collective choices and for the political attitudes of citizens. Our results put us next to the second group of researchers.

Este artigo aborda o papel das ideologias esquerda edireita sobre o comportamento político do eleitorado brasileiro, com foco na constrição que essas ideologias devem gerar sobre suas preferências políticas. Em outras palavras, avalia se o eleitor estrutura suas atitudes políticas em consonância com a ideologia de que diz ser adepto. Usando dados de survey do Estudo Eleitoral Brasileiro (Eseb), os achados indicam que parte significativa do eleitorado do Brasil nem se autoposiciona ideologicamente. Pior ainda, a ideologia declarada por poucos eleitores não constringe suas preferências políticas. Seguramente, a grande maioria dos eleitores não conhece a complexidade envolta no antagonismo programático de cada lado do espectro ideológico. Conclui-se, portanto, que as ideologias não são importantes para o comportamento político do eleitorado brasileiro. _______________________________________________________________________________________________ ABSTRACT This article explores the role of ideology (left and right) in the Brazilian mass electorate. Specifically, we are interested in examining how ideologyconstrains the political preferences of Brazilians. Or, said differently, this article explores if Brazilians make use of ideology to structure their political preferences. Using data from the Brazilian Electoral Study Collection (Eseb), we find that many Brazilian voters do not place themselves on a left-right ideological scale when asked to. Moreover, among those voters that place themselves on a left-right ideological scale, we find that ideology does not structure their political preferences. In other words, ideology does not constrain political preferences. To be sure, the notions of left and right do not mean much for the Brazilian electorate. _______________________________________________________________________________________________ RESUMEN En este artículo se analiza el papel de las ideologías de izquierda y de derecha en el comportamiento político del electorado brasileño, centrándose en la constricción que se espera que estas ideologías generen sobre las preferencias políticas. En otras palabras, se mide si el votante estructura sus actitudes políticas en consonancia con la ideología de la que pretende ser adepto. Utilizando datos de la encuesta del Estudio Electoral Brasileño (Eseb), los resultados indican que una parte significativa del electorado de Brasil no se ubica ideológicamente. Peor aún, la ideología declarada por unos cuantos votantes no constriñe sus preferencias políticas. Sin duda, la gran mayoría de los votantes no conocen la complejidad que implica el antagonismo programático de ambos lados del espectro ideológico. Concluimos, por tanto, que las ideologías no son importantes para el comportamiento político del electorado brasileño. _______________________________________________________________________________________________ RÉSUMÉ Dans cet article, nous examinons le rôle des idéologies dedroite et de gauche sur le comportement politique de l'électorat brésilien. Plus précisément, l'article a pour objectif d'évaluer comment l'idéologie définit les préférences politiques des Brésiliens. Une analyse quantitative des données de sondage de l'Étude électorale brésilienne (Eseb) démontre que peu de Brésiliens savent se positionner sur l'échelle gauche-droite. Même parmi les électeurs qui parviennent à le faire, on remarque que leur positionnement idéologique n'explique en rien leurs préférences politiques. Assurément, une grande majorité des électeurs n´est pas consciente de la complexité comprise dans l´antagonisme programmatique de chaque côté du spectre idéologique. En d'autres mots, l'idéologie ne semble pas définir les préférences politiques des électeurs brésiliens et les notions de gauche et de droite ont peu de sens pour l'électorat de façon générale.

Fil: Nadeau, Richard. University Of Montreal; Canadá Fil: Ratto, Maria C. Instituto de Investigaciones En Diversidad Cultural y Procesos de Cambio; Argentina Fil: Ratto, Maria C. Universidad Nacional de Río Negro; Argentina Fil: Ratto, Maria C. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Patagonia Norte; Argentina Fil: Lewis Beck, Michael S. University Of Iowa; Estados Unidos Fil: Bélanger, Éric. Mc Gill University; Canadá Fil: Gelineau, Francois. Laval University; Canadá Fil: Turgeon, Matthieu. Universidade Do Brasilia; Brasil We now know a good deal about economic voting, but that knowledge comes mostly from research on high income democracies, such as those in North America and Western Europe. Of course, there are serious relevant studies on low income democracies, but they are largely of single country design, frequently employing aggregate data. Multinational, individual level survey designs are all but non existent. Here we examine a 13 nation pool of current Latin American voting surveys. From estimation of a fully specified, and vigorously challenged, voting model, we conclude there are general, and rather strong, sociotropic retrospective economic voting effects. This finding is of normative, as well as scientific importance, indicating that governments in these low income democracies cannot escape electoral sanction for the delivery of poor economic performance. En la actualidad existe un amplio conocimiento acerca del voto económico; conocimiento que, en general, proviene de democracias desarrolladas (Norteamérica y Europa Occidental). Existen trabajos relevantes sobre democracias con niveles de ingresos bajos, pero basados en estudios de casos con datos agregados. Trabajos que incorporen diferentes países y que utilicen encuestas con datos individuales son prácticamente inexistentes. A partir de encuestas electorales, en el siguiente trabajo se examinan 13 países de América Latina de manera conjunta. Mediante la estimación de un modelo de voto altamente especificado y puesto a prueba rigurosamente, se concluye que hay efectos de voto económico sociotrópico retrospectivo generales y bastante fuertes. Este hallazgo posee importancia científica y normativa, indicando que los gobiernos en esas democracias no pueden escapar a la sanción electoral por los pobres resultados económicos.

Some institutional arrangements may be undesirable for democracy by obscuring which political actors are to be held responsible for failed or successful policies and bad or good macroeconomic performances. Much of the work in the area has focused on whether institutions affect the ‘clarity of political responsibility’ and the ability of voters to punish or reward, in turn, governments and elected officials. Not much has been said, however, about the assignment of responsibility outside the electoral context, for a broad range of policy areas. This paper explores these questions in the context of French semi-presidentialism. It demonstrates that the French public is surprisingly quite responsive to the demands imposed by their political system by adjusting reasonably well their evaluations of both actors of the executive in light of major political events and changes in the economic conditions when the circumstances clearly indicate which of the two is ‘in charge’. At other times, however, this particular institutional arrangement obscures instead political responsibility.

French popularity functions are under-studied, at least when as compared to vote functions. After a first wave of studies conducted in the wake of Lafay’s pioneering work, scholarly works on French popularity functions have been few and far between in the last two decades. And this, despite the fact that public opinion data on French popularity have been collected systematically for a long time. What is more, all the popularity function work conducted to date has made use of either the IFOP measure of satisfaction or the Sofres confidence measure as its dependent variable. In this article we build a popularity function for the French executive that compares results obtained with each of these two related, but somewhat different, monthly measures of presidential popularity. We do this comparison for the time period where the two popularity measures overlap (1978–2012). The results indicate that both survey measures are indeed similar but different.

Neste artigo adotamos uma metodologia de questionamento indireto para medir as opiniões de estudantes da UnB acerca do uso da cor/raça como um critério de admissão à Universidade. Partimos da hipótese de que opiniões sobre políticas afirmativas não são obtidas com confiabilidade em surveys convencionais, pois o grupo não elegível ao benefício (no caso, os estudantes brancos) poderia ocultar uma resposta de oposição por receio de parecer preconceituoso. Esse efeito é conhecido na literatura como efeito da desejabilidade social. Portanto, adotamos uma metodologia que garante privacidade aos respondentes, denominada experimento de listas, para medir as opiniões sinceras dos estudantes acerca do sistema de cotas raciais. Porém os resultados mostraram que o viés nas opiniões ocorreu, de forma inesperada, no grupo de estudantes elegíveis às cotas (afrodescendentes e indígenas). Precisamente, eles apoiaram fortemente o sistema de cotas raciais (68,3%) quando tiveram privacidade nas respostas, mas na pergunta direta demonstraram apenas uma tímida aprovação (29,0%). Chamamos esse efeito de efeito de inibição. Paralelamente, os estudantes brancos não demonstraram nenhum efeito de desejabilidade social, contrariando alguns achados de pesquisas similares nos EUA. In this article, we adopt an indirect questioning methodology to measure attitudes toward the use of race as a criterion for admission in Brazilian higher education institutions. We hypothesize that attitudes toward such affirmative action policies cannot be measured by conventional survey questions because non-eligible students - mostly white students - may fear to appear prejudiced by showing opposition to them. This survey effect is known as the social desirability effect. Thus we adopt a list experiment to measure students' sincere attitudes toward the race-based quota admission system. We find that white students do not over-report their approval of racial quotas in university admissions. But, the results show that quota eligible students, Afro and indigenous Brazilians, tend to overwhelmingly under-report their approval of race quotas. Specifically, eligible students strongly approve of the racial quota system (68.3%) when provided privacy in their responses, but publicly voice only timid approval of it (29.0%). We label this effect as the inhibition effect. Moreover, we did not find a social desirability effect among white students, contrasting with some previous findings in the U.S.

The use of election forecasting models is common practice in the US and other established democracies like France and the UK. However, not much work has been done in the area for more recent democracies. Forecasting election results in recently (re)democratized countries poses a serious challenge, given the very few observations of the dependent variable. Thus, we ask: is it possible to make valid election forecasts when the number of elections we have is very small? In this paper, we present recommendations on how to forecast elections under such circumstances. Our strongest recommendation is to evaluate forecasting models using subnational data. We illustrate our recommendations using Brazilian presidential elections since 1994 and data from the 27 states of the Union. Our findings indicate that forecasting elections in recent democracies is neither futile nor impossible, as some of the models presented here produce reasonably accurate forecasts.

This article examines whether a public opinion survey can improve the quality of political attitudes. More specifically, we argue that simply positioning a summary attitudinal question after a balanced series of relevant items can increase people's ability to answer in a way that better reflects their underlying interests, values, and predispositions. By manipulating the location of the vote preference question in two separate national election campaign surveys, we find that there are fewer undecided respondents when the question is asked at the end of the survey rather than early on, that some people are changing their mind during the questionnaire, that a larger set of determinants is structuring late-survey vote choice, and that voting preferences based on the later question are a better predictor of the actual vote. The findings carry important lessons for students of deliberation and of citizen decision making.

Because Republican and Democratic elites have polarized in recent decades, American voters increasingly face choices between candidates who hold divergent policy positions. Such a development has potential implications for the way voters process information during campaigns and choose between candidates on election day. Drawing on research in political psychology and using a nationally representative survey-experiment, we argue and find that levels of candidate polarization—the convergence or divergence of candidates’ issue positions—affect voter information consumption, recall of campaign information, and the balance of on-line and memory-based processing employed in the vote decision. In showing that voters faced with more similar candidates rely more heavily on memory-based processing, we provide further support for hybrid models of political information processing and suggest that candidate polarization has consequences for voter attitude strength and resistance to persuasion.

The role of public opinion polls in electoral democracy is undeniable because, for good or for bad, they affect, in part, the kinds of laws and policies elected officials enact. But the voices measured in polls are not perfectly representative of their populations of interest. More precisely, polls generally sing with a more “knowledgeable” accent than those they represent because of the greater tendency of the less knowledgeable to remain silent. This distortion, however, can be palliated by providing conditions more propitious to attitude development. By relying on survey-experiments conducted in Brazil and in the U.S., I present evidence that inducing people to think more carefully before answering attitude questions reduces substantially the likelihood of the less knowledgeable, which compose most of the Brazilian and American populations, to express a nonopinion response. Thus providing people with greater opportunity to think about politics—something most of them do not do very frequently—makes for more representative measures of public opinion. But the analyses also suggest that increased thought induces greater uncertainty or ambivalence among the most knowledgeable. As a whole, this paper improves our understanding on how people come to develop political attitudes and on the conditions that lead to greater attitude uncertainty or ambivalence. It also carries important lessons and implications for survey design more generally.

Fil: Nadeau, Richard . University Of Montreal; Canadá Fil: Ratto, Maria C. Instituto de Investigaciones En Diversidad Cultural y Procesos de Cambio; Argentina Fil: Ratto, Maria C. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Patagonia Norte; Argentina Fil: Ratto, Maria C. Universidad Nacional de Río Negro; Argentina Fil: Lewis Beck, Michael S. University Of Iowa; Estados Unidos Fil: Bélanger, Éric. Mc Gill University; Canadá Fil: Gélineau, François. Laval University; Canadá Fil: Turgeon, Mathieu. Universidade Do Brasilia; Brasil Classical economic voting theory has received considerable empirical support. Voters reward the ruling party for good times, punish it for bad. But the success of this paradigm, which views the economy voting as strictly a valence issue, has crowded out testing the other theoretical dimensions (Lewis Beck and Nadeau 2011). After reviewing the literature on economic voting in Argentina, this conclusion also holds true for Argentina where most research has focused in valence dimension practically leaving out positional and patrimonial views. The aim of this paper is to provide a more complete picture of economic voting in Argentina through a deeper analysis of the three dimensions of Economic Voting Theory: valence, positional and patrimonial. Through analysis of a special battery of economic questions from the 2008, 2010 and 2012 waves of the Americas Barometer we demonstrate that the economy was important to voter on these three dimensions. Durante la última década del siglo XX la teoría clásica del voto económicoha recibido considerable apoyo empírico. Los votantes premian alpartido de gobierno por los buenos tiempos económicos y lo castigan porlos malos. Pero el éxito de este paradigma, el cual adopta la visión clásica[valence] de la economía que posee un amplio consenso, ha eclipsado laposibilidad de poner a prueba otras dimensiones teóricas. Luego de revisaren detalle la literatura sobre el voto económico en Argentina, esta conclusióntambién se vuelve cierta para el caso argentino, en donde gran partede los trabajos se han centrado en el estudio de la economía como unvalence issue dejando prácticamente vacante las dimensiones posicionalesy patrimoniales. El objetivo de este trabajo es ofrecer una visión más completay acabada de la teoría del voto económico en Argentina a partir de unanálisis en profundidad de sus tres dimensiones, a saber: la clásica [valence],la posicional y la patrimonial. A partir del análisis de una batería especialde preguntas sobre la economía de las olas de 2008-2010 y 2012 delBarómetro de las Américas, fue posible confirmar que la economía en sustres dimensiones fue importante para los ciudadanos a la hora de votar.

Este trabalho analisa de forma sistemática e abrangente os efeitos da desproporcionalidade da representação dos estados (malapportionment) sobre o viés partidário – isto é, a distribuição desigual de cadeiras legislativas entre partidos, por votos recebidos. Especificamente, examinam-se esses efeitos no âmbito da Câmara dos Deputados, de 1994 a 2010. Primeiro, utiliza-se o modelo econométrico bilogit multinominal proposto por King (1990), para analisar a responsividade eleitoral do sistema brasileiro, que se revelou elevada. Em seguida, adaptando-se a estratégia analítica de Grofman, Koetzle e Brunell (1997) ao sistema multipartidário com distritos plurinominais, e após se extrair os efeitos das coligações eleitorais, encontram-se resultados que sugerem fortes efeitos da desproporcionalidade da representação dos estados sobre o viés partidário no Brasil. Esses efeitos, entretanto, variam consideravelmente entre partidos e eleições. _______________________________________________________________________________________________ ABSTRACT This paper offers a systematic and comprehensive analysis of the effects of malapportionment on partisan bias, that is, the uneven distribution of legislative seats among parties for votes received. Specifically, it examines such effect in the Brazilian Lower Chamber from 1994 to 2010. First, we use King’s (1990) multinomial bilogit model to analyze the responsiveness of the Brazilian electoral system, which revealed to be quite high. Second, adapting the analytical strategy of Grofman, Koetzle e Brunell (1997) to multi-party system and multi-member districts, and after extracting the effects of electoral coalitions, we found results that suggest strong malapportionment effects on partisan bias in Brazil. These effects, however, vary considerably by parties and elections.

The paper examines voters’ capacity to determine which of the three main parties (the Liberals, the Conservatives, and the New Democratic Party) was weakest in their constituency in the 1988 Canadian election. We find that half of the voters correctly identified the party that would finish third in their constituency. Voters who did not identify with the third party and who were well informed were generally able to predict correctly which party would finish third. The rational expectations condition did not hold so well among other subgroups of the electorate.