RPEAVT Archive

Political trust and trust in the election process

gbain — Wed, 17 Aug 2011 13:31:37 -0700

Voter confidence in election results is of the utmost importance for the legitimacy of the chosen legislators. When the trustworthiness of the techniques and methods that are used during the elections become subject of debate, this can have a negative impact on the confidence of voters. Previous research has shown that the level of trust in the election process in the Netherlands is not determined by a ‘winner-loser’ effect and that demographic variables only have a weak influence. This study shows that the level of political trust however is an explanatory factor for trust in the election process. Also, all three components of political trust; trust in politicians, trust in institutions and trust in democracy, are of significant influence on the trust in the election process. Finally, social trust does not have a significant influence on the trust in the election process during the 2010 parliamentary elections.
The addition of political trust to the model for trust in the election process has greatly improved the ability of the model to predict the level of trust in the election process. When only the independent variable political trust is used, the explanatory level of the model is 21.7%. After adding the demographic variables, this level goes up to 24.4%. Compared to a model that only includes demographic variables with an explanatory level of 13.4%, the current model shows an improvement of 11%. However, this expanded model still does not have a very high explanatory value. This indicates that there are still other factors that influence the level of trust that a voter has in the election process. Further research will be necessary to find these factors.

Voter trust in the Netherlands between 2006 and 2010

gbain — Wed, 17 Aug 2011 13:24:17 -0700

Abstract: In this paper the trust of Dutch voters in the election process is examined. Since the Parliamentary elections of November 2006, large changes have surrounded the Dutch election process. The widely used voting machines that were introduced in the Netherlands in 1966 were decertified in 2007, causing a return to paper ballot voting. Discussions took place both in the media and in Parliament on election technologies and the trustworthiness of the election process. However, based on survey data of the last two Dutch elections, these discussions so far have not significantly influenced the trust of voters in the election process. Furthermore, more voters still prefer the use of voting machines and feel that these machines are trustworthier than paper ballots. The presence of a ‘winner-loser’ effect on trust in the election process is not found for the Netherlands. In the meantime, some demographic variables are found that influence the trust in the election process. In the Netherlands, male voters have slightly higher levels of trust than female voters. Young voters are more trusting than older voters. Higher educated voters and voters with a higher personal income level show higher levels of trust. Voters who attend religious services on a regular basis also have higher levels of trust in the election process. However, the influence of these demographic factors is not very large. Further research will be necessary to develop a more accurately prediction model for Dutch voter trust.

Disputed Elections Post Bush v. Gore: Are Federal Courts Entering the Political Contest Thicket?

gbain — Thu, 28 Apr 2011 08:45:29 -0700

Introduction

The article will first briefly provide a broad view over of our system in the United States for the resolution of disputed elections. The aspiration is that this will frame an examination of four election disputes post the Bush v. Gore decision. Three of the disputes are used because they arguably are the most widely followed and extensively contested election results since 2000: Washington Gubernatorial (2004), Minnesota Senate (2008) and Florida 13th Congressional District (2006). The final is examined because it may be the only significant decision using Bush v. Gore as precedential authority in an election contest or recount: Hamilton County, Ohio Common Pleas Court Juvenile Judge.

Some scholars have predicted that Bush v. Gore, "is not likely to be precedent for much at all." They believe Bush v. Gore would be strictly limited to the "'special instance' when there is a statewide recount ordered by a court and where it can be shown that there are diverse recounting procedures in various jurisdictions. In fact, the Court in Bush v. Gore stated that the "larger question about variability of election practice among local entities is not before the Court, so it did not render judgment on the matter."

The Cites That Counted: A Decade of Bush v. Gore Jurisprudence

gbain — Thu, 28 Apr 2011 08:39:59 -0700

Introduction

When the Supreme Court issued the opinion that resolved the 2000 presidential election in George W. Bush’s favor, the five justice coalition responsible for the decision went to great lengths to stress that the opinion should not be construed as an explication or expansion of any legal doctrine or concept. The per curium opinion, presumably authored by Chief Justice Rehnquist, specifically tried to narrow the applicability of the legal reasoning that resolved the equal protection claim by including the following passage:

The recount process, in its features here described, is inconsistent with the minimum procedures necessary to protect the fundamental right of each voter in the special instance of a statewide recount under the authority of a single state judicial officer. Our consideration is limited to the present circumstances, for the problem of equal protection in election processes generally presents many complexities (Bush v. Gore 531 US 98, 109 (2000)).

Bush v. Gore: A critical Juncture for early voting?

gbain — Wed, 27 Apr 2011 10:39:28 -0700

The title of this conference, "10 Years after Bush v. Gore," implies that the papers and discussions will focus on the impact of the 2000 election, the Court decision, and subsequent controversy on elections, election law, and election administration. But this call raised for us a basic question: what precisely is BvG? At the simplest level, BvG means George W. Bush, et al. v. Albert Gore, Jr., et al. (531 U.S. 98, 121 S. Ct. 525), a United States Supreme Court decision issued on December 8, 2000. The Court ruled 7-2 that the recounting of votes occuring in certain counties in Florida be stopped due to an inconsistent standard, and 5-4 that no constitutionally-valid scheme could be devised that would meetin Florida's statutory deadline for results certification. The decision, in essence, awarded the election to George W. Bush. But beyond this decision - explicitly limited in its judical sclpe - the "case and controversy from which it arose gave rise to a new wave of legislative and judicial attention to the mechanics of elections. A sea change in media coverage and extensive legal commentary has followed. It is hard to deny that BvG was a monumentally important moment for election administration in the United States, and one that prompted enormous change in the way that elections are administered, conducted, and scrutinized.

Absentee Ballot Regimes: Easing Costs or Adding a Step?

gbain — Wed, 27 Apr 2011 10:14:07 -0700

There has been a revolution in voting in the United States in the last 40 years. In 1972 voters in only 2 states had the option to request an absentee ballot without showing cause. In 2008, 27 states allowed voters this opportunity. In 1972 voters in 45 out of 50 states who were voting at a polling place did so on election day. In 2008, voters in 31 states could cast in-person votes on multiple days (not withstanding the suggestion of the constitution that election day is the tuesday after the first monday of November). There are some obvious poltical questions about the impact of these changes. Any time an electoral institution is changed we want to know if this will advantage one particular party or another, generally thru making it harder or easier for partisans of that party to vote. Or, by changing the incentives of parties to mobilize particular voters. In the case of these laws - the most obvious to ask question is whether or not they have aected turnout.

If we make `election day' span 2 weeks rather than 1 day, we have significantly increased the opportunities people have to vote. Will otherwise non-voters take advantage of those opportunities, or - were they non-voters by choice: they simply do not want to vote? We know that the turnout of registered voters is much higher than turnout of non-registered voters. So, if we make it easier for people to register, and increase the pool of registered voters, will we increase turnout? And if so, the turnout of whom? Young? Old? Rich? Poor? Democrats? Republicans?

In this paper we look at a particular set of institutions designed to make it easier for people to vote, those associated with absentee voting. In other work we have shown that absentee voting leads to higher levels of turnout. But as with most changes in electoral laws, the devil can be in the details. While 27 states allow for no-fault absentee voting they differ in what persons must do to request an absentee ballot. Do these differences matter? Changes in implementation of laws has had effects on turnout for a long time. The 14thth amendment gave blacks the right to vote. But poll taxes and literacy tests were quite effective at keeping black turnout low. Allowing voters to cast an absentee ballot might have minimal impact on turnout if voters can not easily get an absentee ballot. The potential upside of absentee balloting is that it removes the cost of showing up to a specific site on election day, and potentially waiting in a line that can be as long as 3 hours (or more). However, the additional hurdle imposed by absentee balloting is that the voter must acquire the ballot. If this process is suffciently costly or complex, it might outweigh the convenience of avoiding the in-person balloting experience. States vary systematically in how they allow persons to request and acquire an absentee ballot. In a project sponsored by Pew, we collected data on several aspects of this process. In particular, we gathered data on whether or not voters could request the absentee ballot by mail, and whether or not voters could request the absentee ballot electronically { either via the web or via email. And we gathered data on whether or not voters could remove this cost of requesting the ballot `in perpetuity' (or at least as long as they maintained their present address) by requesting permeanent absentee ballot status (as opposed to having to request a ballot for each election).

There is good reason to believe that all of these administrative changes could affect the use of absentee balloting to vote. For 30 years political scientists have been preaching that lowering the costs of voting could increase turnout, and that a particularly nefarious cost of the american electoral system is that people go thru a two step process to vote: first register, then vote. The need to request an absentee ballot effectively makes voting a two step process, even after a voter is registered. And it shared the characteristic of registration that has proven to be so crucial: it must happen before the election. If voters can remove this step by acquiring permanent absentee status, it should significantly enhance the impact of absentee voting. Similarly, being able to request a ballot by mail is a much smaller cost than having to request a ballot in person. But being able to request a ballot by email is a smaller cost still. We have started to gather evidence that doing tasks by email can impose much less ofa burden than requesting things by mail.

The Effect of Voting Systems on Voter Participation

gbain — Mon, 07 Feb 2011 13:19:48 -0800

Herb Asher’s 1982 conference paper, “The Effect of Voting Systems on Voter Participation,”
has a distinguished place in the literature of political science about voting technologies. Three
things stand out about this work. First, it is among the earliest research within the discipline about the performance of voting technologies and its role in influencing elections. Second, it examines as a dependent variable what we would call the “residual vote,” although here it is called “fall-off.” Third, the methodology is in the spirit of a “natural experiment,” an important feature in voting technology research that, alas, still eludes much of the field.

The occasion of the paper was the 1978 Ohio gubernatorial election, which witnessed an anomalous drop in the number of votes cast for governor, in comparison to other statewide races. As Asher explains in the paper, the research arose from the Secretary of State’s office puzzling over what was behind this anomaly. Their inability to solve the puzzle led to a phone call to the other side of Columbus, and the rest is history. The story behind the genesis of this study points out how the rare collaboration between academics interested in the machinery (pun intended) of elections and election officials can yield important knowledge both to the academy and to election administrators.

Because this paper was given as a conference paper, but never published, its virtues have been hidden under a bushel from those interested in the role of voting technologies in elections. The Voting Technology Project is delighted that Professor Asher has agreed to our publishing this classic in our series of Research Papers on Election Administration and Voting Technology.

Charles Stewart III

MIT
February 4, 2011

Making Outsiders' Votes Count: Detecting Electoral Fraud through a Natural Experiment

gbain — Thu, 19 Aug 2010 11:02:31 -0700

Weak electoral registration requirements are commonly thought to encourage electoral participation, but may also promote electoral fraud. For one, candidates and their supporters can more easily mobilize voters outside the district to register and vote for the candidates, even though these voters do not reside within the district. We statistically detect this classic type of electoral fraud for the first time, by taking advantage of a natural experimental setting in Japanese municipal elections. We argue that whether or not a municipal election was held in April 2003 can be regarded as an “as-if” randomly assigned treatment. The difference-in-difference analysis of municipality-month panel data shows that an increase in the new population just prior to April 2003 is significantly larger in treatment municipalities (with an election) than in control ones (without an election). The estimated effects are, in some cases, decisive enough to change the electoral results, especially when the election is competitive. We argue that our approach – “election timing as treatment” – can be applied to investigate not only this type of electoral fraud but also other “electoral connection[s]” (Mayhew 1974) in other countries.

A Supervised Machine Learning Procedure to Detect Electoral Fraud using Digital Analysis

gbain — Mon, 05 Apr 2010 10:07:57 -0700

This paper introduces a naive Bayes classier to detect electoral fraud using digit patterns in vote counts with authentic and synthetic data. The procedure is the following: (1) we create 10,000 simulated electoral contests between two parties using Monte Carlo methods. This training set is composed of two disjoint subsets: one containing electoral returns that follow a Benford distribution, and another where the vote counts are purposively \manipulated" by electoral tampering { a percentage of votes are taken away from one party and given to the other; (2) we calibrate membership values of the simulated elections (i.e. clean or fraudulent) using logistic regression; (3) we recover class-conditional densities using the relative frequencies from the training set; (4) we apply Bayes' rule to class-conditional probabilities and class priors to establish the membership probabilities of authentic observations. To illustrate our technique, we examine elections in the province of Buenos Aires (Argentina) between 1932 and 1942, a period with a checkered history of fraud. Our analysis allows us to successfully classify electoral contests according to their degree of fraud. More generally, our ndings indicate that Benford's Law is an eective tool for identifying fraud, even when minimal information (i.e. electoral returns) is available.

Ukraine 2010: Were Tymoshenko's Cries of Fraud Anything More than Smoke?

gbain — Tue, 30 Mar 2010 10:58:43 -0700

Abstract: This paper applies several indicators of fraud to the official precinct level returns of Ukraine‟s 2010 presidential contest – indicators developed and used previously to assess elections in Russia, Ukraine, Taiwan and Venezuela – and compares our findings to the third round of Ukraine‟s 2004 presidential vote. Overall, we conclude that Viktor Yanukovic won a free and fair contest over Yulia Tymoshenko. Indeed, our indicators yield even fewer suspicions in 2010 than they did in 2004, and thus we concur with the opinions of outside observers such as OSCE that neither candidate benefitted from any significant number of suspicious votes or final tallies. Surely there is no evidence to support Tymoshenko‟s post-election assertion that she was denied victory by upwards of a million fraudulent votes.