We studied publication bias in the social sciences by analyzing a known population of conducted studies--221 total--in which there is full accounting what published and unpublished. leveraged Time-sharing Experiments Social Sciences (TESS), National Science Foundation-sponsored program researchers propose survey-based experiments to be run on representative samples American adults. Because TESS proposals undergo rigorous peer review, studies sample all exceed substantial quality threshold....

With widespread democratic backsliding globally, people's support for democracy-eroding leaders is receiving overdue attention. But existing studies have difficulty disentangling contextual effects (such as who in power at the time of survey) from individual differences (like which party one supports and how strongly). Moreover, we lack evidence on causal antecedents these attitudes. We propose a novel survey experimental design to strip away political context through hypothetical scenarios,...

We report the results of an intervention that targeted anti-Roma sentiment in Hungary using online perspective-taking game. evaluated impact this a randomized experiment which sample young adults played game, or unrelated Participation game markedly reduced prejudice, with effect-size equivalent to half difference between voters far-right and center-right party. The effects persisted for at least month, and, as byproduct, also antipathy toward refugees, another stigmatized group Hungary,...

Many scholars have raised concerns about the credibility of empirical findings in psychology, arguing that proportion false positives reported published literature dramatically exceeds rate implied by standard significance levels. A major contributor is practice reporting a subset potentially relevant statistical analyses pertaining to research project. This study first provide direct evidence selective underreporting psychology experiments. To overcome problem complete experimental design...

Abstract Weighting techniques are employed to generalize results from survey experiments populations of theoretical and substantive interest. Although weighting is often viewed as a second-order methodological issue, these adjustment methods invoke untestable assumptions about the nature sample selection potential heterogeneity in treatment effect. Therefore, although useful technique estimating population quantities, it can introduce bias also be used researcher degree freedom. We review...

The accuracy of published findings is compromised when researchers fail to report and adjust for multiple testing. Preregistration studies the requirement preanalysis plans publication are two proposed solutions combat this problem. Some have raised concerns that such changes in research practice may hinder inductive learning. However, without knowing extent underreporting, it difficult assess costs benefits institutional reforms. This paper examines survey experiments conducted as part...

Abstract Supporters of devolution argue that local policies better reflect citizen preferences than “one size fits all” enacted at the federal level. To test this claim, we leverage sudden abortion policy-making resulted from Dobbs decision. Using multilevel regression with poststratification, estimate latest gestational age which average resident each state believes should be permitted and compare these estimates to before after ruling. We demonstrate prior were more liberal constituent’s...

How can we reduce discrimination against ethnic minorities without the support of public institutions and state policies? In this article describe results an intervention targeted to Roma people in Hungary context rental housing. collaboration with online platform for home rental, tested impact a perspective-taking short documentary film. Our audit study revealed pronounced discrimination. Evidence from our survey experiment demonstrated that – at least run based on measures behavioral...

Abstract How well does public policy represent mass preferences in U.S. states? Current approaches provide an incomplete account of statehouse democracy because they fail to compare and policies on meaningful scales. Here, we overcome this problem by generating estimates Americans' the minimum wage them observed both within across states. Because measure same scale (U.S. dollars), can quantify association outcomes with states (responsiveness) their deviation (bias). We demonstrate that while...

Citizen support for democracy-eroding political leaders is receiving much overdue attention, but existing studies have a difficulty disentangling contextual effects (such as who in power at the time of survey) from individual differences (like which party one supports and how much). We propose novel survey experimental design to strip away context through hypothetical scenarios. This allows us gauge public’s democratic hypocrisy: citizen’s approval eroding norms institutions depends on likes...

We explore the effect of economic hardship on identification with a disadvantaged ethnic minority using longitudinal data 10,000 adolescents in Hungary. Fixed-effects and first-differenced panel models show that having Roma descent are more likely to identify as when their families experience hardship, an strongest among mixed-ethnicity parents. Adolescents who substantially less prejudiced against supportive exclusionary policies than those but not identifying Roma. These findings support...

Abstract There is a growing worry about the health of American democracy, and political scientists pundits alike are looking for possible explanations. Surveys conducted during Trump presidency showed considerable citizen support liberal democratic norm erosions, especially among Republicans. However, recent experimental research also shows that voters both parties more tolerant erosion committed by politicians party they prefer. In this note, we aim to reconcile these contradictory findings...

This multimethod project investigates discrimination against members of two populous minority groups in the European Union: Roma (numbering 6 million) and disabled 100 on a leading Hungarian carpooling platform. In field experiment, 1005 ride requests were sent to drivers, with passenger group membership (control, disabled, Roma) manipulated between participants. Widespread both was apparent significantly lower approval rates for (56%) passengers (52%) relative control (70%). Mechanisms...

What happens after issue frames shape public opinion? We offer an account of the downstream effects on candidate choice. then use three studies combining framing experiments with conjoint choice to directly assess these effects. Despite ideal setting for elite influence opinion, we find that ultimately have modest how later evaluates politicians. Our theoretical framework highlights two sources this disconnect. Frame-induced opinion change is only one component, often outweighed by other...

AbstractWith widespread democratic backsliding globally, people's support for democracy-eroding leaders is receiving overdue attention. But existing studies have a difficulty disentangling contextual effects (such as who in power at the time of survey) from individual differences (like which party one supports and how strongly). Moreover, we lack evidence on causal antecedents these attitudes. We propose novel survey experimental design to strip away political context through hypothetical...

This paper reports the findings from an analysis of ethnic identification choices adolescents in Hungary. Using a representative panel survey Hungary we test hypothesis that poverty shapes how identify and conditions transmission families. Our results indicate Roma is part dual Hungarian there substantial variation both across generations for respondents waves. We find children mixed-ethnic families are more likely themselves as poor than affluent ones. longitudinal analyses changes economic...