AEA RCT Registry currently lists 11821 studies with locations in 170 countries.
We conduct a multi-stage randomized experiment among university students in the UK that combines (i) randomized information about harassment prevalence drawn from prior survey evidence, (ii) randomized monetary incentives for adopting a mobile safety app, and (iii) randomized access to a real-time video-call protection feature among adopters. We study effects on crime experiences, perceived safety, beliefs about crime risk, anxiety, and mobility. We characterize selection into adoption, and study how effects vary across individuals. We additionally estimate willingness to pay for safety and develop a structural model of nighttime travel choices. The update reflects: (i) design changes in Round 2 (expanded sample, 13 universities) and (ii) an updated supplementary Prolific study with WTP...
Teaching children reading and English as a second language (ESL) skills---skills that provide lifelong returns---requires access to engaging, contextually relevant materials. For children outside of Western contexts, English-language story options often lack familiar settings, norms, problems, and characters. We aim to learn whether stories written with generative AI assistance can fill that gap by creating context-relevant stories that better engage children in India.
This study examines how exposure to policy trade-offs affects individuals’ value priorities regarding artificial intelligence (AI) governance. Many public policies involving AI require balancing multiple normative objectives, such as efficiency, safety, and privacy protection. While prior studies often measure stated preferences over these values, less is known about whether exposure to explicit policy trade-offs can change individuals’ value prioritization. We conduct a survey experiment focusing on AI-assisted traffic management systems. Respondents first report the importance they attach to three policy objectives: traffic efficiency, traffic safety, and personal data privacy. Participants are then randomly assigned to either a control group or a treatment group. The treatment group...
This study aims to investigate the impact of personalized risk feedback and information utility framing on individuals' health self-assessments, emotional responses, and willingness to engage in preventive behaviors related to Chronic Kidney Disease (CKD), and how to alleviate information avoidance in health. We will recruit participants through an online platform and present them with a CKD prevention campaign. In the first stage, we will collect baseline data on subjective health predictions, and emotional state. Participants will also provide basic health information to generate a personalized CKD risk assessment. In the second stage, participants will be randomly assigned to one of three groups: a control group that does not receive their risk assessment, a feedback group that recei...
This research project aims to understand to what extent information about a financial product, when provided either to men or women, affects spouses’ beliefs and attitudes towards the financial product, product purchase decisions, and intra-household information sharing, cooperation and conflict outcomes. We conduct a randomized controlled trial among 3100 pastoral households in Kenya and Ethiopia. We randomize, at the household-level, the gender of the spouse that receives information about an index insurance and savings financial product through video interventions. Additionally, we cross-randomize whether or not the household member that is targeted with information can directly share the information with their spouse, by requesting the intervention team to also show the video to the...
This experimental design is implemented in collaboration with the NGO Chile Transparencia. Beginning in February 2020 the Chilean Comptroller General Office rolled out 167 audits in municipalities throughout the country. The treatment consists of the audit report for each municipality in addition to a short information video that highlights the extent to which the municipality has high or low malfeasance counts. We have recruited approximately 49,000 online subjects from these municipalities (plus a control group of 177 non-audited municipalities). All 10,000 subjects will receive a pre-treatment survey measuring their pre-treatment beliefs regarding corruption in their municipality. Within each audited municipality subjects randomly assigned to the treatment will receive a v...
Understanding the mechanisms that explain altruistic actions has been subject to research for decades. To extend our knowledge of altruism in more complex situations, we investigate the role of social cohesion on altruistic transfers in network settings. To this end, we elicit a social network of friends and subsequently measure the relationship levels between them. Then, conditional on the elicited social network and relationships, we measure a series of mechanisms as potential mediational factors between social cohesion and transfer decisions with more than two players involved. Our set-up allows us to identify the degree to which individuals embedded in social networks take into account subsequent transfers between other parties, network information and individual identity.
A growing body of research, pioneered by Dana et al. (2007), has confirmed the prevalence of strategic ignorance: if acting in one’s own interest might harm the interest of someone else, many individuals prefer not to know whether interests conflict or align; moreover, if given the option to stay ignorant, subjects tend to behave more selfishly. In particular, Dana et al. (2007) found that in a binary version of the dictator game, only 26% of subjects chose the selfish alternative when payoffs were known. However, when dictators had to click a button to reveal whether there was a conflict of interest between themselves and the recipient, many abstained from revealing this information, and as much as 63% of those who actually faced conflicting interest now chose the selfish option. Alt...
This study assesses the effectiveness of an information-based intervention designed to increase maize farmer engagement with sustainable land management (SLM) training and the adoption of SLM practices in the district of Pèrèrè in North-East Benin. Farmers in the study area face declining soil fertility resulting in substantial yield losses. The intervention provides a randomly selected subset of farmers with personalized information on the status of their soil, the implications for yields as well as recommendations on how to prevent further soil degradation. The study examines whether this information can increase participation in an upcoming SLM training participation and encourage uptake of taught practices.
We evaluate the effectiveness of a sustainable land management (SLM) training program implemented by a local NGO in Pèrèrè, north-eastern Benin. The region faces severe land degradation and declining soil fertility due to reduced rainfall, intensive farming practices, and increased use of chemical pesticides and mineral fertilizers. The intervention targets 800 maize farmers across 25 villages and will take place before and during the growing season 2026. The program includes training sessions on specific SLM practices delivered before the rainy season March-April 2026, as well as practical follow-ups in the following months on a plot identified by each farmer as particularly degraded. The 25 participating villages are randomly selected from the 50 maize-producing villages in Pèrèrè. We...