2024

1. PC

   Axiomatization of Plurality Refinements

   Ozkes, Ali and Sanver, Remzi

   Public Choice, Volume 200, pages 285–292 (2024)

   Abs HTML

   Plurality rule uniquely satisfies anonymity, monotonicity, neutrality, and tops-onlyness. However, it is not always able to produce resolute outcomes. We study singleton-valued refinements of plurality rule that satisfy all but one of these four axioms. Monotonicity is preserved by all refinements of plurality, whereas no refinement satisfies the remaining three except for a very limited case. We explore what dropping one of the three remaining axioms brings about towards singleton-valued refinements.

2. ManSci

   Reproducibility in Management Science

   Fišar, Miloš, Greiner, Ben, Huber, Christoph, Katok, Elena, and Ozkes, Ali

   Management Science, Volume 70, No 3, pages 1343–1356, (2024)

   Abs OSF HTML

   With the help of more than 700 reviewers we assess the reproducibility of nearly 500 articles published in the journal Management Science before and after the introduction of a new Data and Code Disclosure policy in 2019. When considering only articles for which data accessibility and hard- and software requirements were not an obstacle for reviewers, the results of more than 95% of articles under the new disclosure policy could be fully or largely computationally reproduced. However, for almost 29% of articles at least part of the dataset was not accessible for the reviewer. Considering all articles in our sample reduces the share of reproduced articles to 68%. The introduction of the disclosure policy increased reproducibility significantly, since only 12% of articles accepted before the introduction of the disclosure policy voluntarily provided replication materials, out of which 55% could be (largely) reproduced. Substantial heterogeneity in reproducibility rates across different fields is mainly driven by differences in dataset accessibility. Other reasons for unsuccessful reproduction attempts include missing code, unresolvable code errors, weak or missing documentation, but also soft- and hardware requirements and code complexity. Our findings highlight the importance of journal code and data disclosure policies, and suggest potential avenues for enhancing their effectiveness.

3. SCWE

   Voting Behavior in One-Shot and Iterative Multiple Referenda

   Grandi, Umberto, Lang, Jérôme, Ozkes, Ali, and Airiau, Stéphane

   Social Choice and Welfare, Forthcoming

   Abs HTML SocArXiv SSRN

   We consider a set of voters making a collective decision via simultaneous vote on two binary issues. Voters' preferences are captured by payoffs assigned to combinations of outcomes for each issue and they can be nonseparable: a voter's preference over an issue might be dependent on the other issue. When the collective decision in this context is reached by voting on both issues at the same time, multiple election paradoxes may arise, as studied extensively in the theoretical literature. In this paper we pursue an experimental approach and investigate the impact of iterative voting, in which groups deliberate by repeating the voting process until a final outcome is reached. Our results from experiments run in the lab show that voters tend to have an optimistic rather than a pessimistic behaviour when casting a vote on a non-separable issue and that iterated voting may in fact improve the social outcome. We provide the first comprehensive empirical analysis of individual and collective behavior in the multiple referendum setting.

4. EcoLet

   Ultimatum bargaining: Algorithms vs. Humans

   Ozkes, Ali, Hanaki, Nobuyuki, Vanderelst, Dieter, and Willems, Jurgen

   Economics Letters, Volume 244, 111979 (2024)

   Abs HTML

   We study human behavior in ultimatum game when interacting with either human or algorithmic opponents. We examine how the type of the AI algorithm (mimicking human behavior, optimising gains, or providing no explanation) and the presence of a human beneficiary affect sending and accepting behaviors. Our experimental data reveal that subjects generally do not differentiate between human and algorithmic opponents, between different algorithms, and between an explained and unexplained algorithm. However, they are more willing to forgo higher payoffs when the algorithm’s earnings benefit a human.

5. MSS

   Classification Aggregation without Unanimity

   Cailloux, Olivier, Hervouin, Matthieu, Ozkes, Ali, and Sanver, Remzi

   Mathematical Social Sciences, 2024

   Abs HTML arXiv

   A classification is a surjective mapping from a set of objects to a set of categories. A classification aggregation function aggregates every vector of classifications into a single one. We show that every citizen sovereign and independent classification aggregation function is essentially a dictatorship. This impossibility implies an earlier result of Maniquet and Mongin (2016), who show that every unanimous and independent classification aggregation function is a dictatorship. The relationship between the two impossibilities is reminiscent to the relationship between Wilson's and Arrow's impossibilities in preference aggregation. Moreover, while the Maniquet-Mongin impossibility rests on the existence of at least three categories, we propose an alternative proof technique that covers the case of two categories, except when the number of objects is also two. We also identify all independent and unanimous classification aggregation functions for the case of two categories and two objects.

2023

1. ExpEcon

   Strategic Environment Effect and Communication

   Hanaki, Nobuyuki and Ozkes, Ali

   Experimental Economics, 2023

   Abs SSRN HTML

   We study the interaction of the effects of the strategic environment and communication on the observed levels of cooperation in two-person finitely repeated games with a Pareto-inefficient Nash equilibrium. We replicate previous findings that point to higher levels of tacit cooperation under strategic complementarity compared to strategic substitution. We find that this is not due to differences in levels of reciprocity as previously suggested. Instead, we find that slow learning coupled with noisy choices might drive this effect. When subjects are allowed to communicate in free-form online chat before making choices, cooperation levels increase significantly to the extent that the difference between strategic complements and substitutes disappears. A machine-assisted natural language processing approach shows how the content of communication is dependent on the strategic environment and cooperative behavior. In particular, we find that subjects in complementarity games reach full cooperation by agreeing on gradual moves towards it.

2. EcoLet

   Uniform Random Dictatorship: A characterization without strategy-proofness

   Ozkes, Ali and Sanver, Remzi

   Economics Letters, Volume 227, June 2023, 111127

   ABS HTML

   Although they exclude any possibility of a compromise ex-post, uniform random dictatorship methods provide a solution to conflicts of interest that guarantees an ex-ante fairness. Axiomatic characterizations of random dictatorships in the classical literature of social choice theory use strategy-proofness. In a probabilistic framework that embeds tops-onlyness and anonymity, for three or more alternatives, we provide a characterization that uses an independence condition instead: uniform random dictatorship is the only social choice rule that is efficient and independent. This characterization also establishes that under efficiency and anonymity, independence and strategy-proofness are equivalent. In the particular case of two alternatives, independence becomes vacuous and we propose a characterization without independence.

3. IJSR

   Are robots to be created in our own image? Testing the ethical equivalence of robots and humans

   Vanderelst, Dieter, Jorgensen, Corinne, Ozkes, Ali, and Willems, Jurgen,

   International Journal of Social Robotics, 15, pages 85–99 (2023)

   Abs HTML

   Service robots need to adhere to the ethical expectations of the people with which they interact. Several research groups have developed methods for implementing artificial morality for robots. The hidden assumption underlying this work is that people hold artificial systems to the same ethical standards as humans. In this sense, humans and robots are considered ethically equivalent. However, this assumption remains untested. This paper presents a series of survey-based studies to measure people’s opinions about acceptable and ethical behavior for robots and how this compares to proper and ethical conduct for human caregivers. As such, we assess the assumption that behavior acceptable for humans is also acceptable for robots. In a series of surveys with different samples and methodologies, we find little evidence for rejecting the assumption that people hold artificial systems to the same ethical standards as humans. In the absence of evidence against this widely held assumption, we conclude that ethical norms governing human-robot interaction can be modeled on existing moral norms regulating human-human interaction.

2022

1. JMathE

   Polarization in Networks: Identification-alienation Framework

   Huremović, Kenan and Ozkes, Ali

   Journal of Mathematical Economics, 2022

   Abs HTML arxiv SSRN

   We introduce a model of polarization in networks as a unifying setting for the measurement of polarization that covers a wide range of applications. We consider a substantially general setup for this purpose: node- and edge-weighted, undirected, and connected networks. We generalize the axiomatic characterization of Esteban and Ray (1994) and show that only a particular instance within this class can be used justifiably to measure polarization in networks.

2021

1. JEBO

   Inclusive Cognitive Hierarchy

   Koriyama, Yukio and Ozkes, Ali

   Journal of Economic Behavior and Organization, 2021

   Abs HTML SSRN SUPP

   Cognitive hierarchy theory, a collection of structural models of non-equilibrium thinking, in which players’ best responses rely on heterogeneous beliefs on others’ strategies including naïve behavior, proved powerful in explaining observations from a wide range of games. We propose an inclusive cognitive hierarchy model, in which players do not rule out the possibility of facing opponents at their own thinking level. Our theoretical results show that inclusiveness is crucial for asymptotic properties of deviations from equilibrium behavior in expansive games. We show that the limiting behaviors are categorized in three distinct types: naïve, Savage rational with inconsistent beliefs, and sophisticated. We test the model in a laboratory experiment of collective decision-making. The data suggests that inclusiveness is indispensable with regard to explanatory power of the models of hierarchical thinking.

2. SCWE

   Anonymous, neutral, and resolute social choice revisited

   Ozkes, Ali, and Sanver, Remzi

   Social Choice and Welfare, 2021

   Abs HTML SSRN

   We revisit the incompatibility of anonymity and neutrality in single-valued social choice. We first analyze the irresoluteness outlook these two axioms together with Pareto efficiency impose on social choice rules and deliver a method to refine irresolute rules without violating anonymity, neutrality, and efficiency. Next, we propose a weakening of neutrality called consequential neutrality that requires resolute social choice rules to assign each alternative to the same number of profiles. We explore social choice problems in which consequential neutrality resolves impossibilities that stem from the fundamental tension between anonymity, neutrality, and resoluteness.

3. WP

   Ability or motivation? Voter registration and turnout in Burkina Faso

   Ozdemir, Ugur, Ozkes, Ali, and Sanver, Remzi

   Working Paper, 2021

   Abstract OSF

   2015 elections in Burkina Faso marked a regime change driven by the public uprisings of the year before. The elections were widely taken transparently and freely run, in contrast to past experiences. However, only around two-thirds of the registered voters, who constitute around two-thirds of the voting-age population, cast their vote in the elections. Using data from an original survey experiment, we investigate the determinants of participation in these elections, focusing on the process --- looking at both the registration and voting stages. Using a two-stage sample selection model, we find that socio-demographic and economic indicators are important only in the registration stage, whereas ethnicity becomes the prominent factor in the turnout decisions.

2017

1. EcoLet

   Absolute qualified majoritarianism: How does the threshold matter?

   Ozkes, Ali and Sanver, Remzi

   Economics Letters, 2017

   Abs HTML

   We study absolute qualified majority rules in a setting with more than two alternatives. We show that given two qualified majority rules, if transitivity is desired for the societal outcome and if the thresholds of one of these rules are at least as high as the other’s for any pair of alternatives, then at each preference profile the rule with higher thresholds results in a coarser social ranking. Hence all absolute qualified majority rules can be expressed as specific coarsenings of the simple majority rule.

2. WP

   Generalized Measures of Polarization in Preferences

   Can, Burak, Ozkes, Ali, and Storcken, Ton

   Working Paper, 2017

   Abs RePEc

   We provide axiomatic characterizations for measures of polarization in profiles of preferences that are represented as rankings of alternatives. Polarization is seen as the extent to which opinions are opposed. We provide characterizations for an extension of this simple intuition on the pairs of alternatives to the cases with more than two alternatives. Our primary generalization allows for different treatment among issues, i.e., pairs of alternatives. Secondly, we show that the characterization result continues to hold when preferences are allowed to attain indifferences. Finally, we show that we can also impose a domain restriction that only allows for single-peaked preferences and retain our characterization. Our results point to a fundamental feature of measures on profile of preferences that are based on pairwise comparisons of alternatives.

2016

1. SCWE

   Hyper-stable Social Welfare Functions

   Lainé, Jean, Ozkes, Ali, and Sanver, Remzi

   Social Choice and Welfare, 2016

   Abs HTML

   We define a new consistency condition for neutral social welfare functions, called hyper-stability. A social welfare function (SWF) selects a weak order from a profile of linear orders over any finite set of alternatives. Each profile induces a profile of hyper-preferences, defined as linear orders over linear orders, in accordance with the betweenness criterion: the hyper-preference of some order P ranks order Q above order Q’ if the set of alternative pairs P and Q agree on contains the one P and Q’ agree on. A special sub-class of hyper-preferences satisfying betweenness is defined by using the Kemeny distance criterion. A neutral SWF is hyper-stable (resp. Kemeny-stable) if given any profile leading to the weak order R, at least one linear extension of R is ranked first when the SWF is applied to any hyper-preference profile induced by means of the betweenness (resp. Kemeny) criterion. We show that no scoring rule is hyper-stable, unless we restrict attention to the case of three alternatives. Moreover, no unanimous scoring rule is Kemeny-stable, while the transitive closure of the majority relation is hyper-stable.

2015

1. MSS

   Measuring Polarization in Preferences

   Can, Burak, Ozkes, Ali, and Storcken, Ton

   Mathematical Social Sciences, 2015

   Abs HTML

   In this paper, we study the measurement of polarization in collective decision making problems with ordinal preferences over alternatives. We argue that polarization can be measured as an aggregation of antagonisms over pairs of alternatives in the society. We propose a measure of this sort and show that it is the only measure satisfying some normatively appealing conditions.

2014 BC

1. WP

   Measuring Public Preferential Polarization

   Ozdemir, Ugur and Ozkes, Ali

   Working paper, 2014

   Abs RePEc

   We adapt an axiomatically derived measure of polarization due to Esteban and Ray (1994) to measure polarization of political preferences. Previous work used different measures such as variance, kurtosis, Cronbach's alpha, median distance to median and the mean distance between groups. Yet, none of these measures are theoretically connected to a notion of polarization. Although the initiation of the current one is in the lieu of income inequality measurement, it is conceptually suitable for preferential polarization as well. This paper offers a methodology for that purpose. The second contribution of the paper is that we use the Aldrich-McKelvey Scaling to correct for differential-item functioning in estimating ideal points of the individuals. We use the American National Election Survey Data for years between 1984-2008 to implement the theory offered in the paper. Our findings suggest that there is not a statistically significant increasing trend in polarization in this time period in many issue dimensions but there is an upward trend in the latent ideology dimension which is significant during the 1990s.