Although a growing body of research suggests that the constabularization of the military for domestic policing is counterproductive, this increasingly prevalent policy has nonetheless enjoyed widespread support in the developing world. This study advances our understanding of the consequences of militarization for perceptions of law enforcement: whether visual features shape perceptions of effectiveness, respect for civil liberties, proclivity for corruption and acceptance of militarization in one's own neighborhood. Based on a nationally representative, image-based, conjoint experiment conducted in Mexico, the authors find that military weapons and uniforms enhance perceptions of effectiveness and respect for civil liberties, and that the effect of military uniform becomes greater with increased military presence. The study also finds that gender shapes perceptions of civil liberties and corruption, but detects no effect for skin color. The findings suggest that a central feature of militarization linked to greater violence – military weapons – is paradoxically a key factor explaining favorable attitudes, and that women can play a crucial role in improving perceptions of law enforcement.

Knowledge about political representatives' behavior is crucial for a deeper understanding of politics and policy-making processes. Yet resources on legislative elites are scattered, often specialized, limited in scope or not always accessible. This article introduces the Comparative Legislators Database (CLD), which joins micro-data collection efforts on open-collaboration platforms and other sources, and integrates with renowned political science datasets. The CLD includes political, sociodemographic, career, online presence, public attention, and visual information for over 45,000 contemporary and historical politicians from ten countries. The authors provide a straightforward and open-source interface to the database through an R package, offering targeted, fast and analysis-ready access in formats familiar to social scientists and standardized across time and space. The data is verified against human-coded datasets, and its use for investigating legislator prominence and turnover is illustrated. The CLD contributes to a central hub for versatile information about legislators and their behavior, supporting individual-level comparative research over long periods.

Can referendums help increase perceived legitimacy among citizens with populist attitudes? Indeed, public opinion surveys show that populist citizens are especially in favour of referendums. However, we do not know whether this support reflects a principled desire for different decision-making procedures or an instrumental one (that is, because they expect referendums to yield favourable outcomes). We study this question on a real-life case: the Dutch 2018 referendum on the Intelligence and Security Services Act 2017. Using high-quality survey data from both before and after the referendum, we find that, counter to conventional wisdom and our hypotheses, populists' support for referendums is less driven by instrumental motives compared to that of non-populists, and that populists are more likely than non-populists to accept the outcome of a referendum, even when this outcome is unfavourable.

By many measures, the public knows little about politics. But just how little people seem to know depends on the questions that are put to them. In particular, knowledge levels seem higher when people are asked closed- rather than open-ended questions. In turn, differences between estimated knowledge levels are sometimes attributed to fundamental differences between these types of questions. Building on this previous research, the present study uses a pre-registered experiment conducted with a representative national sample to shed new light on the relationship between question form and knowledge measurement. The authors find that inferences about political knowledge depend less on fundamental differences between open- and closed-ended questions than on two little-appreciated aspects of survey design: the number and difficulty of the response options that accompany closed-ended questions. These aspects of survey design have large effects. Scholars who use the same questions with different response options may reach substantively different conclusions about the public's levels of knowledge.

Prior research suggests that partisanship can influence how legislators learn from each other. However, same-party governments are also more likely to share similar issues, ideological preferences and constituency demands. Establishing a causal link between partisanship and policy learning is difficult. In collaboration with a non-profit organization, this study isolates the role of partisanship in a real policy learning context. As part of a campaign promoting a new policy among local representatives in the United States, the study randomized whether the initiative was endorsed by co-partisans, out-partisans or both parties. The results show that representatives are systematically more interested in the same policy when it is endorsed by co-partisans. Bipartisan initiatives also attract less interest than co-partisan policies, and no more interest than out-partisan policies, even in more competitive districts. Together, the results suggest that ideological considerations cannot fully explain partisan-based learning. The study contributes to scholarship on policy diffusion, legislative signaling and interest group access.

How does an individual's criminal record shape interactions with the state and society? This article presents evidence from a nationwide field experiment in the United States, which shows that prospective applicants with criminal records are about 5 percentage points less likely to receive information from college admission offices. However, this bias does not extend to race: there is no difference in response rates to Black and White applicants. The authors further show that bias is all but absent in public bureaucracies, as discrimination against formerly incarcerated applicants is driven by private schools. Examining why bias is stronger for private colleges, the study demonstrates that the private–public difference persists even after accounting for college selectivity, socio-economic composition and school finances. Moving beyond the measurement of bias, an intervention designed to reduce discrimination is evaluated: whether an email from an advocate mitigates bias associated with a criminal record. No evidence is found that advocate endorsements decrease bureaucratic bias.

The interest group literature has long struggled with how to empirically approach the normative idea of a non-biased group system. While most previous attempts have focused on the descriptive representation of different types of groups, this article argues that substantive representation of citizens' attitudes is closer to the democratic principle of equal effective participation. It develops a methodological approach that captures substantive representation with respect to agenda priorities and policies by surveying interest groups on how much time they spend on lobbying in different policy areas, and in which direction they lobby on salient policy issues. The responses are compared with opinion data to estimate the level of political (in)equality. The findings from the case of Sweden – where relatively high levels of equality would be expected, but striking levels of inequality based on socio-economic status are instead found – highlight the perseverance of what Schattschneider once called the upper-class bias of the pressure system.

How can one assess which countries select more experienced leaders for the highest office? There is wide variation in prior career paths of national leaders within, and even more so between, regime types. It is therefore challenging to obtain a truly comparative measure of political experience; empirical studies have to rely on proxies instead. This article proposes PolEx, a measure of political experience that abstracts away from the details of career paths and generalizes based on the duration, quality and breadth of an individual's experience in politics. The analysis draws on a novel data set of around 2,000 leaders from 1950 to 2017 and uses a Bayesian latent variable model to estimate PolEx. The article illustrates how the new measure can be used comparatively to assess whether democracies select more experienced leaders. The authors find that while on average they do, the difference with non-democracies has declined dramatically since the early 2000s. Future research may leverage PolEx to investigate the role of prior political experience in, for example, policy making and crisis management.

When authoritarian regimes liberalize, are there observable patterns in the ordering of reforms, and are these patterns distinct for cases that transition to democracy? While the prevailing literature tends to focus on exogenous ‘determinants’ of democracy, this letter describes the endogenous dynamics of liberalization itself. Using pairwise domination analysis, it assesses the institutional order of reforms during 371 episodes of liberalization in autocracies between 1900 and 2019. Based on twenty-four indicators of democratic institutions and practices, our findings reveal (1) a clear pattern of reform during liberalization episodes, (2) with strong similarities across outcomes, but also that (3) reforms to the administration of elections tend to develop comparatively earlier in episodes of liberalization that produce a democratic transition.

This letter provides firm-level evidence that policy makers tailor trade policy to suit selected firms. It argues that firms with higher levels of specific assets find it more costly to reorganize production, and are hurt more by international competition. In response, policy makers grant more trade protection to firms with fixed assets. Since protectionism is costly, firms compete for it, which creates diffusion dynamics in which the protection granted to one firm affects the protection granted to others. This claim is tested utilizing the special role antidumping duties (ADDs) play in international trade, and combining petitions for ADDs with financial data on the firms filing them in a unique dataset. Using spatial autoregressive models, the authors find that firms with specific assets are granted more protection. However, diffusion dynamics differ within and between groups of firms producing the same good. This suggests that firms can partly shape their own level of trade protection.

This comment discusses the impact of social media rule enforcement protocols on research on online data sources. It argues that the conclusions of the article ‘Do Islamic State's Deadly Attacks Disengage, Deter, or Mobilize Supporters?’ concerning the recruitment effects of deadly attacks cannot be assumed to hold when considering the timing of Twitter account suspensions. It highlights four ways in which suspensions can confound evidence of demobilization despite the introduction of control variables and fixed-effects model specifications. All change the composition of the sample in four non-random ways. First, suspending connected Islamic State accounts may result in follower loss. Secondly, Twitter suspension procedures may be tied to account characteristics, such as follower accrual rates. Thirdly, suspended accounts that re-emerge introduce replication bias. Fourthly, account closure may reflect user movement to other platforms in response to changing security environments following deadly attacks. In conclusion, caution is advised when platform-introduced variation risks altering the sample composition in non-random ways.

This article responds to Hansen's (2022) comment on the use of social media data to evaluate the effects of terrorist attacks on related online behavior. Hansen casts doubts on our previous finding that terrorist attacks disengage supporters of terrorist groups. The author speculates that this result might not hold when considering the bias introduced by the timing of Twitter account suspensions. We contend that this critique is deficient in several respects. First, Hansen speculates about a possible bias, yet no empirical evidence is offered to support this claim. Second, the author largely ignores our discussion of the role of Twitter account suspensions and our measures of Twitter suspension efforts and Anonymous reporting activity to alleviate this concern. In this article, rather than engaging in a qualitative debate on the amount of bias, we make two contributions. First, we offer an empirical investigation of the fragility of our previous finding to the presence of omitted variable bias. Once we account for Twitter suspension activities, we find that an extreme, unlikely amount of confounding is required to alter the estimated effect of terrorist attacks on disengagement. Second, we employ sequential g-estimation to calculate the average controlled direct effect of terrorist attacks on disengagement after controlling for two intermediate confounders: Twitter suspension behavior and Anonymous reporting activities. The estimated average controlled direct effect indicates that Islamic State's terrorist attacks significantly reduced followers in Islamic State-related Twitter accounts after de-mediating this effect from Twitter suspension efforts and Anonymous reporting activities. Further, we show that this average controlled direct effect is robust to a massive, and implausible, violation of the sequential unconfoundedness assumption. Overall, these analyses show that the timing of Twitter account suspensions, as well as any other confounder, is extremely unlikely to alter our conclusion: Islamic State's terrorist attacks disengage their supporters. We conclude this article by offering guidance on how to address practical challenges in political science research using social media data.