Election Polls Research Articles

Taking the Next Step in Exploring the Literary Digest 1936 Poll

While many instructors are aware of the Literary Digest 1936 poll as an example of biased sampling methods, this article details potential further explorations for the Digest’s 1924–1936 quadrennial U.S. presidential election polls. Potential activities range from lessons in data acquisition, cleaning, and validation, to basic data literacy and visualization skills, to exploring one or more methods of adjustment to account for bias based on information collected at that time. Students can also compare how those methods would have performed. One option could be to give introductory students a first look at the idea of “sampling adjustment” and how this principle can be used to account for difficulties in modern polling, but the context is rich in other opportunities that can be discussed at various times in the course or in more advanced sampling courses. Supplementary materials for this article are available online.

여론조사에서 무응답 대체를 통한 선거 결과 예측

In the results of the election polls, the candidates' approval ratings are not only a major concern of the people, but also an important factor in establishing the election strategy of the party concerned. However, if there is an inaccurate or distorted result of the election poll, it can cause a big problem because of its influence, and the controversy about the poll often arises. There are many cases in which the actual election results are different from the results of the polls, and political parties and research institutes, which are related to the polls, are making efforts to improve the accuracy of the election forecasts. The purpose of this study is to improve the accuracy of election prediction by replacing non-response through machine learning algorithm using the survey data of the 20th Presidential Election, the 2021 Busan Mayor By-election, and the 8th National Simultaneous Local Election (Busan Metropolitan City, Nam-gu, Gimhae Mayor). For this purpose, first, the data excluding the non-response (I do not know/don't know) was separated into train data and the data selected as non-response (I do not know/don't know) was separated into test data. Second, Random Forest, XGBoost, and LightGBM methods were applied, and only the top three questions were used by deriving the importance of questions when predicting candidates through train data. Third, through the model learned based on the train data, the non-response of the test data was replaced with the response of one candidate, and the corresponding data and the train data were combined to calculate the percentage of votes for each candidate. As a result of comparing the actual election results with the percentage of votes for each candidate by replacing the non-response, it was confirmed that the Random Forest method is suitable for the model for improving the accuracy of election prediction through non-response substitution.

Bias and excess variance in election polling: a not-so-hidden Markov model

Abstract With historic misses in the 2016 and 2020 US Presidential elections, interest in measuring polling errors has increased. The most common method for measuring directional errors and non-sampling excess variability during a postmortem for an election is by assessing the difference between the poll result and election result for polls conducted within a few days of the day of the election. Analysing such polling error data is notoriously difficult with typical models being extremely sensitive to the time between the poll and the election. We leverage hidden Markov models traditionally used for election forecasting to flexibly capture time-varying preferences and treat the election result as a peek at the typically hidden Markovian process. Our results are much less sensitive to the choice of time window, avoid conflating shifting preferences with polling error, and are more interpretable despite a highly flexible model. We demonstrate these results with data on polls from the 2004 through 2020 US Presidential elections and 1992 through 2020 US Senate elections, concluding that previously reported estimates of bias in Presidential elections were too extreme by 10%, estimated bias in Senatorial elections was too extreme by 25%, and excess variability estimates were also too large.

Partisan Poll Watchers and Americans’ Perceptions of Electoral Fairness

Abstract Poll watchers play an increasingly visible role in American elections, yet there is little research into how their presence shapes beliefs about electoral integrity. On the one hand, poll watchers—like other election observers—may lead voters to be more trusting in elections. On the other hand, poll watchers who expressly represent political parties may contribute to polarized views of electoral integrity. We address this gap with two conjoint experiments and one vignette experiment administered on large national samples. The conjoint experiments vary several features of election polling locations, including the presence of poll watchers, voter identification requirements, and other election regulations. We find that regulations around poll watchers shape views of election fairness, but that the nature of this influence depends on the type and partisanship of poll watchers who are present. The vignette experiment shows that perceptions of poll watchers decline when these individuals act in bad faith; effects are especially large for Republican respondents. These findings help identify the potential that poll watchers have to shape Americans’ beliefs about election fairness and integrity.

Work “for” you or work “with” you: The effect of candidate's positioning on donation reception

AbstractDonations are a well‐studied area in consumer research, but our understanding of political donations is limited. Because the motivations for donating to charity and political candidates are different, we use branding perspectives to investigate this important phenomenon. In four studies, we demonstrate that candidates with a “servant” brand positioning are likely to generate fewer donations compared with those with a “partner” positioning. This is because a servant brand positioning leads individuals to perceive that the candidates require fewer resources. We also show that the relationship between brand positioning and political donation is influenced by factors such as political ideology, election polls, and the percentage of the donation relative to election expenses. Specifically, the effect is stronger among conservatives, when the candidates are leading in the polls, and when a small donation represents a significant portion of the election expenses. These findings provide valuable insights into how brand positioning affects individuals' intentions to donate to political campaigns.

Voters’ Expectations in Constituency Elections without Local Polls

Abstract How do voters form accurate expectations about the strength of political candidates in constituency elections if there are no reliable constituency polls available? We argue that voters can use national election polls and past election results to increase the accuracy of their expectations. A survey experiment during the German federal election of 2021 confirms that the provision of national election polls and past results increases the accuracy of voters’ expectations. The analysis further shows that voters leverage the information to update their beliefs. The results have relevant implications for debates about belief formation in low-information environments.

Beyond tribal politics for e-participation and development: social media influence on Nigeria's 2023 presidential general election

PurposeThe study investigates factors like media, tribal, religious and party politics' influence on Nigerias’ 2023 presidential election choice. It confirms dominant social media platforms and examines their influence on election polls, e-participation and political candidate choice. The main objectives of this study are to: investigate if tribal, religious and party politics affect the respondent’s choice of a presidential candidate, ascertain the respondent's most used social media platform for political engagement and determine how social media platforms influenced the election polls during the 2023 Nigerian presidential election.Design/methodology/approachA sample size of 384 registered voters was used to survey three states in Southeast Nigeria hinged on the technological acceptance model, the instrumentalist theory of ethnicity and the theory of reasoned action.FindingsThe study found that tribal politics did not influence political candidates during the 2023 Nigerian presidential election. However, religious and party politics influenced their choices as well as X (Twitter), found as the most used and most influential social media platform vital for enhancing participatory democracy and informing people at real-time.Research limitations/implicationsThe researchers experienced challenges such as ensuring that the respondents filled the questions appropriately to reduce the number of void questionnaires and a funding problem since they had yet to receive any grant to enhance the study.Originality/valueThe study commends improved Internet connectivity and accessibility among the citizens for increased political engagement on social media. It also recommends that the Nigerian government enforce the rule of law in politics to enable diverse tribes and religions to experience democratic e-participation and development without marginalisation or subjugation by incumbent power. The findings affirm that social media is apt in political communication during the 2023 presidential elections in Nigeria. The study is a contribution to knowledge, timely and original.

The return of non-probability sample: the electoral polls at the time of internet and social media

For the past 80 years survey researchers have used a probability sampling framework. Probability samples have a well-defined set of quality criteria that have been organized around the concept of Total Survey Error (TSE). Non-probability samples do not fit within this framework very well and some possible alternatives to TSE are explored. In recent years, electoral polls have undergone changes as a result of the dispersion of public opinion due mostly, but not only, to the development of the web and social media. From a methodological point of view, the main changes concerned sampling and data collection techniques. The aim of the article is to provide a critical contribution to the methodological debate on electoral polls with particular attention to the samples used which appear to be more similar to non-probability samples than to the traditional probability samples used for many decades in electoral polls. We will explore several new approaches that attempt to make inference possible even when a survey sample does not match the classic probability sample. We will also discuss a set of post hoc adjustments that have been suggested as ways to reduce the bias in estimates from non-probability samples; these adjustments use auxiliary data in an effort to deal with selection and other biases. Propensity score adjustment is the most well know of these techniques. The empirical section of the article analyzes a database of 1793 electoral polls conducted in Italy from January 2017 to July 2023.

Applied OHS With Empirical Ergonomics and Hazard Methods to Minimize Fatigue at KPPS Cerme Gresik

To support fair, honest, and Luber Jurdil elections, public awareness about the importance of healthy politics needs to be increased. Voters who have registered for the permanent election (DPT) must cast five ballots in the 2019 simultaneous elections. Voter participation in the 2019 elections, which reached 81.97%, and legislative participation (Pileg) in 2019, which reached 81.69%, showed a fairly good election organization in 2019. However, this is at odds with the health and safety of KPPS officers. According to the KPU RI, the number of election organizers who died in the 2019 elections was 11,239, with 527 victims dying, based on the results of the Ministry of Health's investigation in 28 provinces. Based on the investigation and analysis of the Indonesian Ministry of Health, the fatigue factor triggered the disease suffered by polling station officers. It is necessary to conduct ergonomic research on KPPS officers in the 2024 elections, which is useful for developing an election polling station model that is comfortable, safe, healthy, effective, and efficient so as to reduce factors that cause fatigue while working. The components studied include workload components, rest time, or the physical work environment at election polling stations. The data collection methods used in this research are observation, interview, direct data, and literature study.The analysis used in this study produces a list of events and classifies the incidence of fatigue of KPPS officers, the most problematic workplace. fatigue classification, Identifying Ergonomic Risk Factor (ERF), Data processing, Analysis and discussion as well as conclusions and suggestions.

Accounting for Nonresponse in Election Polls: Total Margin of Error

Accounting for Nonresponse in Election Polls: Total Margin of Error

Remote Election Polling Using Full Stack Development

Remote Election Polling Using Full Stack Development

Stop trying to predict elections only with twitter – There are other data sources and technical issues to be improved

Since the popularization of social media (SM) platforms, researchers have been trying to use their data to predict electoral results. Previous surveys point out that the most used approach is based on volume and sentiment analysis of posts on Twitter. However, they are almost unanimous in presenting that the results are not better than chance. In this context, this study aims to investigate the feasibility of predicting electoral results based only on Twitter, discover the main issues, and draw guidelines for future alternative directions. For this, we reviewed the evolution of election polling and predictions, including the “polling crises” of 1936 and 1948, and their similarities with current approaches. We also built on the official SM platforms' documentation and on our experience collecting and analyzing large-scale data from many SM platforms. Lastly, we analyzed nine reviews on predicting elections with SM data from 2013 to 2021. We observed that, contrary to initial expectations, most of the current research with Twitter has been unable to solve many of the challenges encountered since initial studies, and also shares many of the characteristics of unsuccessful straw polls performed before 1936. We illustrate that by highlighting the impracticability of polling over Twitter due to several biases and technical barriers, the need for external data, the high dependency on the arbitrary decisions of researchers, and the constant change in platforms' scenarios, that may invalidate specific models. Lastly, we indicate some of the possible future directions, such as a focus on creating repeatable processes; the use of SM data as part of statistical models, instead of polling; diversifying the input data sources, including multiple SM platforms and non-SM data such as polls and economic indicators; using machine learning for regression of the vote share, rather than for sentiment analysis; and dealing with the uncertainty of the highly divergent polling results.

The association between the interviewers’ and the respondents’ political attitudes in a telephone survey

ABSTRACT Interviewer effects in telephone surveys on political topics are likely to occur. The literature has yielded considerable evidence about the impact of basic interviewer characteristics, but research is lacking on how interviewers’ beliefs may shape responses. This study is aimed at assessing the association between the interviewers’ party affiliation and political ideology and the answers provided to the same questions by the respondents, using a large-scale telephone survey conducted in Hungary. The results show that interviewers account for a relatively large amount of variance, especially in the models predicting item-nonresponse. The interviewer’s reluctance to answer these questions strongly predicted missing data on the respondent level. On the other hand, only weak interviewer effects were found regarding the valid, substantive responses. The results demonstrate how, for instance, estimations of nonrespondents’ party preferences in electoral polls may be biased due to interviewer-related errors, and highlight the need for standardized persuasion strategies.

The 2016 US presidential election, opinion polls and the stock market

Purpose The purpose of this paper is to study the impact of the 2016 presidential election polls on the stock market. Design/methodology/approach The empirical model includes daily stock returns as the dependent variable and past asset prices, 10-year treasury rates, opinion polls and VIX (market uncertainty) as explanatory variables with a one-year lag. The model was estimated using two sets of daily polling data: from July 1, 2015, to November 8, 2016, and from June 1, 2016, to November 8, 2016. Additional descriptive statistics, such as means and standard deviations, were also calculated. Findings The estimated results did not reveal any statistically significant effects of opinion polls in favor of one candidate over another on stock returns. Simple statistical tests, however, show that the market performed better when Trump held a polling advantage over Clinton. Originality/value To the best of the authors’ knowledge, this is the only study that has examined the effects of the 2016 presidential election polls on the US stock market. This study adds value to the understanding of the relationship between election polls and the stock market in the USA.

The Perception of Public Opinion on Election Polls and the Influence on Voting Behaviour of Electors

The Perception of Public Opinion on Election Polls and the Influence on Voting Behaviour of Electors

What Kind of Election Poll Coverage Do Voters Want? : Focusing on Influencing Factors Related to Cognitive Competencies by Preferred Reporting Type

What Kind of Election Poll Coverage Do Voters Want? : Focusing on Influencing Factors Related to Cognitive Competencies by Preferred Reporting Type

Bias and Variance in Multiparty Election Polls

Abstract Recent polling failures highlight that election polls are prone to biases that the margin of error customarily reported with polls does not capture. However, such systematic errors are difficult to assess against the background noise of sampling variance. Shirani-Mehr et al. (2018) developed a hierarchical Bayesian model to disentangle random and systematic errors in poll estimates of two-party vote shares at the election level. The method can inform realistic assessments of poll accuracy. We adapt the model to multiparty elections and improve its temporal flexibility. We then estimate bias and variance in 5,240 German national election polls, 1994–2021. Our analysis suggests that the average absolute election-day bias per party was about 1.5 percentage points, ranging from 0.9 for the Greens to 3.2 for the Christian Democrats. The estimated variance is, on average, about twice as large as that implied by usual margins of error. We find little evidence of house or mode effects. Common biases indicate industry effects due to similar methodological problems. The Supplementary Material provides additional results for 1,751 regional election polls.

Should we include margins of error in public opinion polls?

AbstractPublic opinion polls have become vital and increasingly visible parts of election campaigns. Previous research has frequently demonstrated that polls can influence both citizens' voting intentions and political parties' campaign strategies. However, they are also fraught with uncertainty. Margins of error can reflect (parts of) this uncertainty. This paper investigates how citizens' voting intentions change due to whether polling estimates are presented with or without margins of error.Using a vignette experiment (N=3224), we examine this question based on a real‐world example in which different election polls were shown to nationally representative respondents ahead of the 2021 federal election in Germany. We manipulated the display of the margins of error, the interpretation of polls and the closeness of the electoral race.The results indicate that margins of error can influence citizens' voting intentions. This effect is dependent on the actual closeness of the race and additional interpretative guidance provided to voters. More concretely, the results consistently show that margins of error increase citizens' inclination to vote for one of the two largest contesting parties if the polling gap between these parties is small, and an interpretation underlines this closeness.The findings of this study are important for three reasons. First, they help to determine whether margins of error can assist citizens in making more informed (strategic) vote decisions. They shed light on whether depicting opinion‐poll uncertainty affects the key features of representative democracy, such as democratic accountability. Second, the results stress the responsibility of the media. The way polls are interpreted and contextualized influences the effect of margins of error on voting behaviour. Third, the findings of this paper underscore the significance of including methodological details when communicating scientific research findings to the broader public.

Predicting Election Results Via Social Media: A Case Study for 2018 Turkish Presidential Election

Social media platforms provide massive amounts of data that can be used to analyze social issues and forecast events in the future. However, it is a challenging task due to the biased and noisy nature of the data. In this work, we propose a method to predict election results via Twitter. In particular, we first detect the stance of social media accounts using their retweets. Subsequently, we develop four different counting methods for our prediction task. In the simple user counting (SC) method, we count labeled users without taking any further steps to reduce bias. In the city-based weighted counting (CBWC) method, we apply a weighted counting based on the number of electorate in each city. The closest-city-based prediction (CCBP) method utilizes sociological similarity between cities to predict results for cities with limited sample sizes. The using former election results (UFERs) method compares predictions for each city against former election results to detect data bias and uses them accordingly. We evaluate our proposed methods with the data collected for the presidential election of Turkey held in 2018. In our extensive evaluation, we show that utilizing domain-specific information and location-based weighted counting is effective in reducing bias. CBWC, CCBP, and UFER methods outperform tweet-counting-based baseline methods. Furthermore, UFER and CCBP outperform almost all traditional polls, suggesting that social media platforms are alternative mediums for conducting election polls.

Predicting a win by a small margin: The effect of graphic scaling in published polls on voters' predictions

AbstractThe public display of election poll results is often manipulated to influence voter predictions about the race. Narrow scaling is one such manipulation that involves truncating the chart's vertical axis such that its range extends closely around the values of the bars. This manipulation exacerbates the visual difference between bars, making the margin appear larger than an unbiased representation would suggest. The current research examines whether narrow scaling of a bar chart depicting the degree of support for political candidate affects voters' predictions about election outcomes. In three experiments, conducted during the 2022 US gubernatorial and senate elections, we displayed published polls to potential voters using a wide‐ or a narrow‐scaled bar chart. We found that when the scale is narrow, voters are more likely to predict that the leading candidate in the poll will win the election and by a larger margin. This scaling bias occurs despite voters' relative skepticism about narrow‐scaled polls. We further find that the scaling effect is attenuated when the poll margin is relatively large and enhanced when numerical value labels are removed from the graphic display.