Paley Anna, van de Ven Niels
Department of Marketing, Tilburg University, 5037AB Tilburg, the Netherlands.
Proc Natl Acad Sci U S A. 2023 Oct 31;120(44):e2308129120. doi: 10.1073/pnas.2308129120. Epub 2023 Oct 23.
Creating effective nudges, or interventions that encourage people to make choices that increase their welfare, is difficult to execute well. Recent work on megastudies, massive field experiments that test many interventions simultaneously, reveals that nudge effectiveness both varies widely and is difficult for experts to predict. We propose an Iterative Crowdsourcing Procedure, which uses insights from members of the target population to generate and preselect nudges prior to testing them in a field experiment. This technique can supplement existing methods or stand alone as a way to generate conditions for testing in a high-quality field experiment. We test the effectiveness of this method in addressing a challenge to effective financial management: consumer oversubscription. We first document that people have more subscriptions than they think they have and that enhancing subscription awareness makes people want to cancel some subscriptions. We then use our crowdsourcing procedure to motivate people toward subscription awareness in a field experiment (N = 4,412,113) with a large bank. We find that the crowdsourced nudges outperform those generated by the bank, demonstrating that the Iterative Crowdsourcing Procedure is a useful way to generate effective nudges.
创造有效的助推手段,即鼓励人们做出能提升自身福祉的选择的干预措施,要出色地实施并非易事。近期关于大规模研究(同时测试多种干预措施的大规模实地实验)的工作表明,助推效果差异极大,且专家难以预测。我们提出了一种迭代众包程序,该程序利用目标人群成员的见解,在实地实验测试之前生成并预选助推手段。这种技术可以补充现有方法,或者独立作为一种为高质量实地实验创造测试条件的方式。我们测试了这种方法在应对有效财务管理挑战(消费者过度订阅)方面的有效性。我们首先记录到人们实际拥有的订阅服务比他们自己认为的要多,而且提高订阅服务意识会让人们想要取消一些订阅。然后我们在与一家大型银行合作的实地实验(N = 4,412,113)中,使用我们的众包程序促使人们提高订阅服务意识。我们发现众包生成的助推手段比银行生成的更有效,这表明迭代众包程序是生成有效助推手段的一种有用方式。