Randolph T. Stevenson’s research while affiliated with Rice University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (29)


Figure 2. Marginal effects of cooperation score on perceived levels of cooperation, by level of habitual news reception.
What drives perceptions of partisan cooperation?
  • Article
  • Full-text available

June 2023

·

39 Reads

·

3 Citations

Political Science Research and Methods

Lie Philip Santoso

·

Randolph T. Stevenson

·

Simon Weschle

What drives voters' perceptions of partisan cooperation? In this note, we investigate whether voters have accurate beliefs about which parties regularly cooperate with one another, and whether these beliefs follow the real-time portrait of cooperation and conflict between parties that is reported in the news. We combine original survey data of voters' perceptions of party cooperation in four countries over two time periods with a measure of parties' public relationships as reported by the media. We find that voters' perceptions of cooperation and conflict among parties do reflect actual patterns of interactions. This pattern holds even after controlling for policy differences between parties as well as joint cabinet membership. Furthermore, we show that the impact of contemporary events on cooperation perceptions is most pronounced for voters who monitor the political news more carefully. Our findings have important implications for partisan cooperation and mass–elite linkages. Specifically, we find that contrary to the usual finding that voters are generally uninformed about politics, voters hold broadly accurate beliefs about the patterns of partisan cooperation, and importantly, these views track changes in relevant news. This reflects positively on the masses' capacities to infer parties' behaviors.

Download

Relative advantages of the crosswise model.
A Bias-Corrected Estimator for the Crosswise Model with Inattentive Respondents

December 2021

·

27 Reads

·

10 Citations

Political Analysis

The crosswise model is an increasingly popular survey technique to elicit candid answers from respondents on sensitive questions. Recent studies, however, point out that in the presence of inattentive respondents, the conventional estimator of the prevalence of a sensitive attribute is biased toward 0.5. To remedy this problem, we propose a simple design-based bias correction using an anchor question that has a sensitive item with known prevalence. We demonstrate that we can easily estimate and correct for the bias arising from inattentive respondents without measuring individual-level attentiveness. We also offer several useful extensions of our estimator, including a sensitivity analysis for the conventional estimator, a strategy for weighting, a framework for multivariate regressions in which a latent sensitive trait is used as an outcome or a predictor, and tools for power analysis and parameter selection. Our method can be easily implemented through our open-source software cWise .


Bias-Corrected Crosswise Estimators for Sensitive Inquiries

October 2020

·

25 Reads

The crosswise model is an increasingly popular survey technique to elicit candid answers from respondents on sensitive questions. We demonstrate, however, that the conventional crosswise estimator for the population prevalence of sensitive attributes is biased toward 0.5 in the presence of inattentive respondents who randomly choose their answers under this design. We propose a simple design-based bias correction procedure and show that our bias-corrected estimator can be easily implemented without measuring individual-level attentiveness. We also offer several useful extensions of our bias correction, including a sensitivity analysis for conventional crosswise estimates, a strategy for weighting, and a framework for multivariate regressions in which a latent sensitive trait is used as an outcome or a predictor. We illustrate our methodology by simulation studies and empirical examples and provide a practical guide for designing surveys to enable our proposed bias correction. Our method can be easily implemented through our open-source software,cWise.


Attributing Policy Influence under Coalition Governance

September 2020

·

24 Reads

·

31 Citations

American Political Science Association

Coalition governance divides policy-making influence across multiple parties, making it challenging for voters to accurately attribute responsibility for outcomes. We argue that many voters overcome this challenge by inferring parties’ policy-making influence using a simple heuristic model that integrates a number of readily available and cheaply obtained informational cues about parties (e.g., their roles in government and legislative seat shares)—while ignoring other cues that, while predictive of real-world influence, are not suitable for heuristic inference (e.g., median party status and bargaining power). Using original data from seven surveys in five countries, we show that voters’ attributions of parties’ policy-making influence are consistent with our proposed inferential strategy. Our findings suggest that while voters certainly have blind spots that cause them to misattribute policy responsibility in some situations, their attributions are generally sensible and consistent with the academic research on multiparty policy making.


Party Government and Political Information

May 2020

·

20 Reads

·

9 Citations

Legislative Studies Quarterly

We argue that the kind of political information voters should possess varies contextually in response to relevant political processes. Focusing on the partisan organization of legislatures, we derive hypotheses for what the typical American should know about politics at the national and state level and test these hypotheses in two studies. The first documents a dramatic change in American political knowledge at the national level in response to polarization—the replacement of individually oriented information with partisan information. While voters’ ability to identify the candidates running to represent them in Congress has been cut in half, their ability to rank order the parties ideologically has nearly doubled. The second study provides evidence that voters are better able to identify the majority party in their state legislature where partisan control of the legislative agenda and roll‐call voting is stronger. We conclude by discussing the implications of our findings.


Causal interaction and effect modification: same model, different concepts

April 2020

·

37 Reads

·

19 Citations

Political Science Research and Methods

Social scientists use the concept of interactions to study effect dependency. In the causal inference literature, interaction terms may be used in two distinct type of analysis. The first type of analysis focuses on causal interactions, where the analyst is interested in whether two treatments have differing effects when both are administered. The second type of analysis focuses on effect modification, where the analyst investigates whether the effect of a single treatment varies across levels of a baseline covariate. While both forms of interaction analysis are typically conducted using the same type of statistical model, the identification assumptions for these two types of analysis are very different. In this paper, we clarify the difference between these two types of interaction analysis. We demonstrate that this distinction is mostly ignored in the political science literature. We conclude with a review of several applications where we show that the form of the interaction is critical to proper interpretation of empirical results.


Figure 3. DAG structures that lead to regression coefficients that cannot be interpreted as causal effects. In panel (a) L has no effect on Y and in panel (b) the confounder, L, has an effect on Y. In both cases, one must condition on L to identify the effect of D on Y, but the effect of L on Y is not identified. (a) One structure for confounding. (b) A second structure for confounding.
The causal interpretation of estimated associations in regression models

July 2019

·

2,017 Reads

·

156 Citations

Political Science Research and Methods

A common causal identification strategy in political science is selection on observables. This strategy assumes one observes a set of covariates that is, after statistical adjustment, sufficient to make treatment status as-if random. Under adjustment methods such as matching or inverse probability weighting, coefficients for control variables are treated as nuisance parameters and are not directly estimated. This is in direct contrast to regression approaches where estimated parameters are obtained for all covariates. Analysts often find it tempting to give a causal interpretation to all the parameters in such regression models—indeed, such interpretations are often central to the proposed research design. In this paper, we ask when we can justify interpreting two or more coefficients in a regression model as causal parameters. We demonstrate that analysts must appeal to causal identification assumptions to give estimates causal interpretations. Under selection on observables, this task is complicated by the fact that more than one causal effect might be identified. We show how causal graphs provide a framework for clearly delineating which effects are presumed to be identified and thus merit a causal interpretation, and which are not. We conclude with a set of recommendations for how researchers should interpret estimates from regression models when causal inference is the goal.



Measuring Knowledge of Parties’ Legislative Seat Shares

October 2017

·

15 Reads

·

3 Citations

Political Science Research and Methods

We use two original surveys (including survey experiments) conducted the week before the 2015 elections in the Britain and Denmark to explore the best method for measuring individuals’ knowledge of the partisan distribution of legislative seats in multi-party democracies. The complete lack of such questions in the corpus of survey research on multi-party democracies is a testament to the skepticism that many survey researchers have about the feasibility of such complex questions. However, our analysis, which is the first empirical test of this skepticism, reveals little evidence of respondents’ frustration or high levels of non-cooperation with these questions. Additionally, our survey experiments, which examine the usefulness of different question formats, make it clear that such questions should be framed in terms of the numbers of seats each party holds rather than shares or percentages of seats.



Citations (24)


... In summary, theory and previous research offer only limited information about how political sophistication moderates citizens' reliance on various informational cues to estimate 7 Consider that Santoso, Stevenson, and Weschle (2023) find that citizens who report lower levels of media attention display far less accurate estimates of how cooperatively different political parties interact based on media news reports. 8 Adams, Bernardi, and Wlezien (2020) substantiate this argument with respect to governing parties, finding that citizens generally adjust their perceptions of these parties' ideologies in response to the government's actual social welfare policy outputs, but not in response to the policy promises contained in their election manifestos. ...

Reference:

Does Political Sophistication Moderate How Citizens Use Information to Infer Left-Right Distances between Parties?*
What drives perceptions of partisan cooperation?

Political Science Research and Methods

... This approach assumes that random responders answer this (sensitive) question honestly instead of randomly. Alternatively, Atsusaka and Stevenson (2023) suggested estimating the prevalence of random responders using a sensitive anchor statement with known prevalence. For instance, for sensitive statement of interest is "In order to avoid paying a traffic ticket, I would be willing to pay a bribe to a police officer", the anchor statement is "I have paid a bribe to be on the top of a waiting list for an organ transplant", with a known prevalence of zero. ...

A Bias-Corrected Estimator for the Crosswise Model with Inattentive Respondents

Political Analysis

... Müller 2024). Auf diese Weise muss sie weniger um einen Rückgang der Unterstützung bei zukünftigen Wahlen fürchten (siehe Fortunato et al. 2021;Hjermitslev 2024). Ein Beispiel für eine solche Minderheitskoalitionsregierung ist die 2022 ins Amt gekommene schwedische Regierung von Premierminister Kristersson, die aus den konservativen Moderaten, den Liberalen und den Christdemokraten besteht, jedoch parlamentarisch von den rechtspopulistischen Schwedendemokraten gestützt wird. ...

Attributing Policy Influence under Coalition Governance
  • Citing Article
  • September 2020

American Political Science Association

... Prior state-level work often focuses on the executive branch and, for example, shows that individuals are more likely to identify who their Governor is when they live closer to the state capital (Delli Carpini, Keeter, and Kennamer 1994;Hopkins 2018) or when a state is more politically competitive (Lyons, Jaeger, and Wolak 2013). Recent work also demonstrates that voters are more likely to identify the majority party in their state legislature if the legislative parties are homogenous and polarized (Fortunato and Stevenson 2021). However, research concerning Americans knowledge of their individual state legislator is often limited to single state samples and somewhat dated (e.g., Patterson, Ripley, and Quinlan 1992). 1 Fewer studies examine Americans' knowledge of "the rules of the game" or the "substance of politics" at the state level. ...

Party Government and Political Information
  • Citing Article
  • May 2020

Legislative Studies Quarterly

... I estimate causal interaction effects between different treatment factors, immigrant rights and immigrant characteristics (skill-level/ country of origin), and an effect modification in terms of the extent to which the rightsconditionality in immigration preferences varies by respondents' political predispositions to capture effect heterogeneity (cf. Keele and Stevenson, 2021). Estimating separate models for the United States and Switzerland allows for cross-country comparison to assess the role of the country-context on immigration preferences. ...

Causal interaction and effect modification: same model, different concepts
  • Citing Article
  • April 2020

Political Science Research and Methods

... The strategy I follow to identify the effect of FLFP and parties' gender labour equality positions on the gender voting gap is controlling for a set of theoretically relevant covariates able to block plausible back-door paths between my explanatory variables and the outcome (Keele, Stevenson, and Elwert 2020). 10 First, I control for the OECD's (2023) measure of the divorce rate, which is the number of divorces during a given year per 1,000 people. ...

The causal interpretation of estimated associations in regression models

Political Science Research and Methods

... Diese senden keine Vertreter in das jeweilige Kabinett, können aber in Verhandlungen inhaltliche Zugeständnisse für sich durchsetzen (vgl. hierzu Tromborg et al. 2019;Thürk 2022;Thürk und Krauss 2024). Das Stützen einer Minderheits(koalitions)regierung durch eine formelle Oppositionspartei kann für letztere elektorale Vorteile haben. ...

Voters, Responsibility Attribution and Support Parties in Parliamentary Democracies
  • Citing Article
  • August 2017

British Journal of Political Science

... It is thus widely discussed in the public which coalitions might form and how the portfolios could be allocated among potential coalition partners. Furthermore, voters also have (implicit) knowledge of 'Gamson's Law' , that is the allocation of portfolios proportional to the seat weights a party brings to a coalition (Lin et al. 2017) because this more or less one-to-one proportionality of seats brought to the coalition and a party's number of portfolios is most of the time the actual outcome for government formation processes in modern democracies (e.g. Ariotti and Golder 2018;Bäck et al. 2009;Cox 2021;Ecker and Meyer 2019;Warwick and Druckman 2006). ...

Gamson's Law and voters’ perceptions of portfolio allocation
  • Citing Article
  • June 2017

European Journal of Political Research

... While ideology and party identification are often used indistinctively, these two concepts are fundamentally different as ideology emphasizes, at least implicitly, cognitive-based instrumental modes of electoral behavior, whereas party identification stresses affect with political parties rather than instrumentality (Van der Eijk et al., 2005, p. 166). Scholars have found that the left-right orientation of citizens not only allows them to form political perceptions and policy positions (Van der Eijk et al., 2005), but also serves as a shortcut to organize party groupings which guide voters' expectations regarding the likelihood of political alliances (Fortunato et al., 2016). ...

Context and Political Knowledge: Explaining Cross-National Variation in Partisan Left-Right Knowledge
  • Citing Article
  • October 2016

The Journal of Politics

... Alternatively, theories of retrospective voting-even biased retrospective voting-suggest that voters will hold incumbent politicians accountable for important outcomes, economic and otherwise (Achen & Bartels, 2016;Fiorina, 1981;Key, 1966). Significant empirical scholarship has shown that incumbents at all levels of government are held electorally accountable for outcomes, regardless of their ultimate culpability (Kramer, 1971(Kramer, , 1983Arceneaux, 2006;Duch, 2008;Healy & Lenz, 2014;Bartels, 2014;Achen & Bartels, 2016;Sances, 2017;de Benedictis-Kessner & Warshaw, 2020). Thus, we may expect state Republicans and federal Democrats to be blamed for local closures. ...

The Economic Vote: How Political and Economic Institutions Condition Election Results
  • Citing Book
  • January 2008