Research

In this paper, we propose a novel randomization approach for randomized controlled trials (RCTs) aimed at maximizing the utilization of baseline information. Despite the widespread collection of baseline data in 90\% of recent studies published in top-tier journals, only half of them effectively incorporate this information for covariate balancing during randomization, primarily relying on stratification. However, existing methods such as stratification and pairwise matching often fail to achieve perfect balance, particularly when dealing with continuous or numerous covariates, and may limit the robustness of inference. We introduce the Cube method (Deville et al., 2004) as a solution to these challenges. The Cube method allows for the selection of perfectly balanced samples across various covariate types, ensuring consistent adherence to balance tests. Additionally, it enables the generation of precise confidence intervals and offers flexibility in determining assignment probabilities, thus addressing concerns related to sample attrition. Through comprehensive theoretical insights and simulation exercises using both simulated and real-world data, we illustrate the advantages of the Cube method in experimental design for RCTs. Our findings highlight the Cube method as a robust and versatile tool for addressing the complexities associated with covariate balancing in RCTs, providing researchers with an effective means of improving experimental design and inference.

LGBTphobia in schools is a worldwide problem, with harassment having significant detrimental effects on victims and social cohesion. Several countries have introduced school-based interventions to address this issue but have not assessed their effectiveness.  Their impact on student knowledge, attitudes, and behavior is largely unknown. This paper presents the first rigorous evaluation of school-based interventions aimed at reducing LGBTphobia. We focus on a classroom intervention that addresses the issue of LGBT harassment through perspective-taking and narrative exchange. Using a field experiment in France with more than 10,000 middle and high school students, we find robust evidence of strong positive effects, with variations across gender, age, and socio-economic status. We argue that the shift in perceptions of group norms is a key channel driving these effects. Our results suggest that school-based interventions can be a promising means of combating LGBTphobia in schools and improving students' outcomes.