Vincent Arel-Bundock

Professor

Université de Montréal
Country flagCanada
ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (6)

Publication practices in the social sciences act as a filter that favors statistically significant results over null findings. While the problem of selection on significance (SoS) is well-known in theory, it has been difficult to measure its scope empirically, and it has been challenging to determine how selection varies across contexts. In this article, we use large language models to extract granular and validated data on about 100,000 articles published in over 150 political science journals from 2010 to 2024. We show that fewer than 2% of articles that rely on statistical methods report null-only findings in their abstracts, while over 90% of papers highlight significant results. To put these findings in perspective, we develop and calibrate a simple model of publication bias. Across a range of plausible assumptions, we find that statistically significant results are estimated to be one to two orders of magnitude more likely to enter the published record than null results. Leveraging metadata extracted from individual articles, we show that the pattern of strong SoS holds across subfields, journals, methods, and time periods. However, a few factors such as pre-registration and randomized experiments correlate with greater acceptance of null results. We conclude by discussing implications for the field and the potential of our new dataset for investigating other questions about political science.

Year:

2026

Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Year:

2025

Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Year:

2025

Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Year:

2025

Political reforms are often impeded by concentrated interest groups who lobby to block change that would benefit the majority. One under-examined policy response is to compensate the recalcitrant group in exchange for agreeing to the reform. We refer to such mass compensation schemes as public buyouts. After laying out the theoretical case for and against buyouts, we design a series of survey experiments to gauge the determinants of public support for buyouts linked to three reforms---phasing out coal energy, simplifying tax filing, and amnesty for dictators. Partisanship appears systematically related to attitudes towards buyouts, as does program design: buyouts find significantly more favour when they target individual workers, rather than companies. Yet the chief objection to buyouts is normative: individuals' "moral aversion" to compensating actors who hold up beneficial reforms dominates other salient concerns, like moral hazard. Our results also highlight a vexing credibility problem: those who support reform also support reneging on the compensation once the reform is passed. Recipients may thus be right to fear policy reversals. Buyouts appear democratically viable as a means of passing beneficial reforms that have been blocked for decades---yet their design proves decisive.

Year:

2024

How do states exercise power in complex, decentralized governance regimes? This article sheds light on this question by analyzing one of the pillars of the post-war economic order: the international tax regime. The taxation of multinational enterprises is governed by a system composed of thousands of Bilateral Tax Treaties signed between pairs of national governments. We argue that despite the bilateral nature of these treaties, their legal content is largely controlled by a small group of economically powerful governments. By drafting and promoting ``boilerplate" legal language, and by leveraging network effects, OECD countries were able to build a coherent, encompassing, quasi-multilateral regime to govern the taxation of multinationals in most of the world. We test and find support for our arguments about the role of network effects and the power of boilerplate using inferential network analysis and an automated text analysis of a corpus of 3200 treaty texts.

Year:

2024

Collaborators (2)

Jonathan Mellon

Assistant Professor

United States Military Academy

UNITED STATES
View details

Ryan Briggs

University of Guelph

CANADA
View details