Election Forensics Research Articles

Forensic analysis of the Turkey 2023 presidential election reveals extreme vote swings in remote areas

Concerns about the integrity of Turkey’s elections have increased with the recent transition from a parliamentary democracy to an executive presidency under Recep Tayyip Erdoğan. Election forensics tools are used to identify statistical traces of certain types of electoral fraud, providing important information about the integrity and validity of democratic elections. Such analyses of the 2017 and 2018 Turkish elections revealed that malpractices such as ballot stuffing or voter manipulation may indeed have played a significant role in determining the election results. Here, we apply election forensic statistical tests for ballot stuffing and voter manipulation to the results of the 2023 presidential election in Turkey. We find that both rounds of the 2023 presidential election exhibit similar statistical irregularities to those observed in the 2018 presidential election, however the magnitude of these distortions has decreased. We estimate that 2.4% (SD 1.9%) and 1.9% (SD 1.7%) of electoral units may have been affected by ballot stuffing practices in favour of Erdoğan in the first and second rounds, respectively, compared to 8.5% (SD 3.9%) in 2018. Areas with smaller polling stations and fewer ballot boxes had significantly inflated votes and turnout, again, in favor of Erdoğan. Furthermore, electoral districts with two or fewer ballot boxes were more likely to show large swings in vote shares in favour of Erdoğan from the first to the second round. Based on a statistical model, it is estimated that these shifts account for 342,000 additional ballots (SD 4,900) or 0.64% for Erdoğan, which is lower than the 4.36% margin by which Erdoğan was victorious. Our results suggest that Turkish elections continue to be riddled with statistical irregularities, that may be indicative of electoral fraud.

Baseline comparative analysis and review of election forensics: Application to Ghana's 2012 and 2020 presidential elections

Many allegations have been levelled against the electoral process of many countries across the world by most opposition leaders, especially when they lose a presidential election e.g. Ghana in 2012 and 2020. Therefore, the need to apply election forensic techniques to the certified election results data of valid votes count to statistically verify if some suspected or possible anomalies and irregularities exist in the voting pattern. This paper seeks to provide a comprehensive review of election forensics techniques and make a comparative analysis of Benford's Second-order test of conformity (using the first two digits) and Hartigans' dip test of unimodality to examine the existence of possible anomalies and irregularities in the 2012 and 2020 presidential elections held in Ghana. The findings of the two tests suggest that the electoral process produced possible anomalous data in the 2012 presidential election results (with an overall 16.67% suspected anomalies), whilst possible non-anomalous data was produced in the 2020 presidential election results (with an overall 0% suspected anomaly) of valid votes count. Therefore, the study recommends that for better statistical data analysis on election anomaly detection, Benford's test of conformity and Hartigans' dip test of unimodality should serve as baseline tests (initial screening tools), highlighting areas that may require further investigation or more rigorous analysis and progressively dig deeper into the application of finite mixture fraud models and machine learning techniques. In spite of the promising results Benford's Law, dip test, machine learning algorithms, and network analysis have produced in detecting irregularities in election data, real-world applications remain challenging, particularly when dealing with complex and evolving forms of fraud. Therefore, there is the need for continuous research and innovation to improve the accuracy and effectiveness of these methods and promote transparency and accountability in democratic societies.

Innovative Analytical and Statistical Technology in Election Forensics

The article examines the crucial role of innovative analytical and statistical technology in electoral forensics, which are increasingly used for detecting and preventing electoral corruption and fraud. By analysing vast amounts of data and detecting anomalies, electoral forensic investigations can contribute to fair and transparent democratic processes. The research aims to explore the effectiveness of these technologies and their potential impact on improving the transparency and fairness of electoral processes, using a multi-method approach that includes analysing relevant documents, media coverage, public opinion, and recent fraud cases. The authors divide the implementation of innovative analytical and statistical technologies for combating election corruption into four groups. The first is the analysis of statistical data and research on corruption, including election processes, which can be called secondary data analysis. The second is the analysis of documentary data containing information on corrupt actions and offences, including election processes. The third is the development of mathematical methods and algorithms using cutting-edge technologies such as artificial intelligence and machine learning for detecting anomalies and hidden patterns. The fourth is experimental developments in information technologies as a means of ensuring proper governance and combating corruption. While the use of algorithms for detecting anomalies in electoral statistics data can be an important tool, it should be used with caution, and in combination with other sources of information, to avoid the consequences of delegitimising the election results.

Was there any widespread fraud in 2020 presidential election? What does Benford's Law say?

Fair elections free of any interference are integral tenets of any functioning democracy, and widespread election fraud is undoubtedly a serious threat to a free republic. While instances of electoral fraud are much more prevalent in countries with illiberal democracies, the U.S has recently faced such an accusation. Although he was unable to provide any concrete evidence, the former U.S. President Donald Trump accused his opponent, Joe Biden, now president, of electoral fraud after the presidential election. Fortunately, election forensics are often successful in investigating the validity of such fraud allegations. In this paper, I applied Benford’s law, a rule that should stand up to any large set of natural numbers, such as un-tampered electoral data. Using this law and basic statistical analysis of votes of U.S. counties for candidates of the two major parties, I completed a forensic analysis to investigate Mr. Trump’s allegation. My comprehensive investigation does not find any evidence supporting his allegation.

Analysis of Benford’s Law for No-Reference Quality Assessment of Natural, Screen-Content, and Synthetic Images

With the tremendous growth and usage of digital images, no-reference image quality assessment is becoming increasingly important. This paper presents in-depth analysis of Benford’s law inspired first digit distribution feature vectors for no-reference quality assessment of natural, screen-content, and synthetic images in various viewpoints. Benford’s law makes a prediction for the probability distribution of first digits in natural datasets. It has been applied among others for detecting fraudulent income tax returns, detecting scientific fraud, election forensics, and image forensics. In particular, our analysis is based on first digit distributions in multiple domains (wavelet coefficients, DCT coefficients, singular values, etc.) as feature vectors and the extracted features are mapped onto image quality scores. Extensive experiments have been carried out on seven large image quality benchmark databases. It has been demonstrated that first digit distributions are quality-aware features, and it is possible to reach or outperform the state-of-the-art with them.

Election forensics: Using machine learning and synthetic data for possible election anomaly detection.

Assuring election integrity is essential for the legitimacy of elected representative democratic government. Until recently, other than in-person election observation, there have been few quantitative methods for determining the integrity of a democratic election. Here we present a machine learning methodology for identifying polling places at risk of election fraud and estimating the extent of potential electoral manipulation, using synthetic training data. We apply this methodology to mesa-level data from Argentina’s 2015 national elections.

Fraud in the 2018 Egyptian presidential election?

ABSTRACT This research note scrutinizes official vote counts from the 2018 Egyptian presidential election. Drawing on data for 13,807 polling stations published by the Egyptian Electoral Authority, it uses the Election Forensics Toolkit to estimate a series of digit and distribution tests. The results point to statistical anomalies in voter turnout for polling stations located in Egypt’s major population centres. The findings align with journalistic accounts of very low turnout in those areas and demonstrate how tools from electoral forensics can be applied to the study of elections in the Middle East and North Africa.

Election Forensics: Quantitative methods for electoral fraud detection

The last decade has witnessed an explosion on the computational power and a parallel increase of the access to large sets of data – the so called Big Data paradigm – which is enabling to develop brand new quantitative strategies underpinning description, understanding and control of complex scenarios. One interesting area of application concerns fraud detection from online data, and more particularly extracting meaningful information from massive digital fingerprints of electoral activity to detect, a posteriori, evidence of fraudulent behavior. In this short article we discuss a few quantitative methodologies that have emerged in recent years on this respect, which altogether form the nascent interdisciplinary field of election forensics. Aiming to foster discussion and raise awareness on this interdisciplinary area, we hereby enumerate a few of the most relevant approaches and methods.

Using Election Forensics to Detect Election Fraud in the Philippine Elections, 2016

This paper uses the tools of election forensics to investigate electoral fraud in the most recent Philippine national elections, 2016. We first focus on digit tests, finite mixture model and its equivalents. We pay particular attention to the measurement of stolen votes and geographic allocation of election fraud across national elections. We then focus on Marcos v. Robredo court case, which helps us to validate some of our research findings for the vice-presidential election.

Elections and election fraud in Georgia and Armenia

ABSTRACTElections on unfair playing fields are common. Yet election day fraud can result in authoritarians losing office. The freer the environment, the more an authoritarian must rely on means other than election day fraud to retain office, because they are less capable of coercing the population without facing repercussions. Among those other means is cooptation through public policy. A common theme in this special issue is that public policy has been of greater import in Georgia than Armenia. This article begins to explain this phenomenon using comparative case studies of election day fraud in Armenia and Georgia over time. To do so, the article uses methods from the field of election forensics to provide a quantitative comparison of the scale of election day fraud in each country’s elections since 2007 using precinct level election results for parliamentary and presidential elections. The test results suggest, as has been widely believed, that Georgia’s elections have had less election day fraud than Armenia’s during this period. This finding provides a theoretical basis to explain why public policy has been a greater concern in Georgia than Armenia.

Worst Election Ever in Russia?

Election forensics analysis shows extensive signs of extensive frauds in the Russian 2016 Duma election. The frequency and magnitude frauds are worse than in any Russian national election since 2000.

Validating Precinct-Level Measures of Fraud: Evidence from the Russian Electoral Cycle 2011-2012

This research is designed to explore whether election fraud precinct-level estimates obtained from a finite mixture likelihood model, recently developed by Mebane(2015), can be validated against alternative, more intuitive measures of election fraud. I estimate the precinct-level probabilities from the parametric model of election fraud and compare them against measures based on election observation, different voting modes as well as several popular forensics indicators. This study also tests how well the cluster patterns of the finite mixture likelihood probabilities on different levels of aggregation are associated with the clusters of alternative election fraud measures. The data used is from the parliamentary election 2011 and presidential election 2012 in Russia. My basic findings demonstrate that finite mixture estimates if compared to alternative sources of information, seem to be effectively capturing electoral anomalies, making this new estimator very useful in election forensics research.

Linking Together Preference Falsification and Election Fraud: A Cross-National Analysis

This study sheds new light on whether public opinion polls, namely, preference falsification, can affect the level of election fraud by employing Kuran's model of preference falsification, which is empirically tested on the data collected from the most recent presidential campaign in Russia (2012) as well as a cross-national dataset. My research findings reveal the presence of a statistically significant effect of preference falsification on election frauds, thus enabling me to conclude that preference falsification is, indeed, conducive to election fraud. My findings can be generalized to a broad set of authoritarian regimes, enabling scholars to get a better understanding of the mechanism by which survey polls can incentivize officials to commit election fraud, and this, I hope, will invite more interdisciplinary research in the field of election forensics and survey methodology.

Election Fraud: A Latent Class Framework for Digit-Based Tests

Digit-based election forensics (DBEF) typically relies on null hypothesis significance testing, with undesirable effects on substantive conclusions. This article proposes an alternative free of this problem. It rests on decomposing the observed numeral distribution into the “no fraud” and “fraud” latent classes, by finding the smallest fraction of numerals that needs to be either removed or reallocated to achieve a perfect fit of the “no fraud” model. The size of this fraction can be interpreted as a measure of fraudulence. Both alternatives are special cases of measures of model fit—the π∗ mixture index of fit and the Δ dissimilarity index, respectively. Furthermore, independently of the latent class framework, the distributional assumptions of DBEF can be relaxed in some contexts. Independently or jointly, the latent class framework and the relaxed distributional assumptions allow us to dissect the observed distributions using models more flexible than those of existing DBEF. Reanalysis of Beber and Scacco's (2012) data shows that the approach can lead to new substantive conclusions.

Unifying the Concepts of Electoral Fraud and Preference Falsification: Case of Russia

This study throws new light on whether public opinion polls, namely, preference falsification, can affect the level of election fraud by employing Kuran's model of preference falsification, which is empirically tested using the data collected from the most recent presidential campaign in Russia (2012). My research findings reveal the presence of statistically significant effects of preference falsification on election fraud, thus enabling me to conclude that preference falsification is, indeed, conducive to election fraud. My findings are generalizable to a broad set of authoritarian regimes, enabling scholars to get a better understanding of the mechanism by which survey polls can incentivize officials to commit election fraud. Hopefully, these findings will also invite more interdisciplinary research within the fields of election forensics and survey methodology.

Building and Maintaining Trust in Internet Voting

Voting codes, risk analysis, and election forensics can help the public build and maintain trust in an Internet voting system by assuring that it meets important security and privacy requirements.

Election Forensics and the 2004 Venezuelan Presidential Recall Referendum as a Case Study

A referendum to recall President Hugo Ch\'{a}vez was held in Venezuela in August of 2004. In the referendum, voters were to vote YES if they wished to recall the President and NO if they wanted him to continue in office. The official results were 59% NO and 41% YES. Even though the election was monitored by various international groups including the Organization of American States and the Carter Center (both of which declared that the referendum had been conducted in a free and transparent manner), the outcome of the election was questioned by other groups both inside and outside of Venezuela. The collection of manuscripts that comprise this issue of Statistical Science discusses the general topic of election forensics but also focuses on different statistical approaches to explore, post-election, whether irregularities in the voting, vote transmission or vote counting processes could be detected in the 2004 presidential recall referendum. In this introduction to the Venezuela issue, we discuss the more recent literature on post-election auditing, describe the institutional context for the 2004 Venezuelan referendum, and briefly introduce each of the five contributions.

Forensic Analysis of the Venezuelan Recall Referendum

The best way to reconcile political actors in a controversial electoral process is a full audit. When this is not possible, statistical tools may be useful for measuring the likelihood of the results. The Venezuelan recall referendum (2004) provides a suitable dataset for thinking about this important problem. The cost of errors in examining an allegation of electoral fraud can be enormous. They can range from legitimizing an unfair election to supporting an unfounded accusation, with serious political implications. For this reason, we must be very selective about data, hypotheses and test statistics that will be used. This article offers a critical review of recent statistical literature on the Venezuelan referendum. In addition, we propose a testing methodology, based exclusively on vote counting, that is potentially useful in election forensics. The referendum is reexamined, offering new and intriguing aspects to previous analyses. The main conclusion is that there were a significant number of irregularities in the vote counting that introduced a bias in favor of the winning option. A plausible scenario in which the irregularities could overturn the results is also discussed.